“The Sad Statistic that Trumps the Others”

Here is my latest column, some of it will sound familiar to readers of TGS:

One bias in the economic statistics — which never shows up in published revisions — is embedded in the health care sector, where third-party payments, subsidies and care quality are hard to monitor and measure. A result is that a dollar spent on health care does not necessarily mean a true dollar’s worth of value added. The United States spends more per capita on health care than any other country, yet without producing measurably superior results. To the extent that some of these expenditures are wasteful, the gross domestic product and productivity numbers overstate economic growth.

Here’s another problem: Expenditures on the military and domestic security have risen since 9/11, but those investments are intended to neutralize external threats. Even if you agree with this spending, it generally doesn’t produce useful goods and services that raise our standard of living.

One of the most commonly cited productivity numbers describes per-hour labor productivity, but this, too, has intrinsic flaws. Labor force participation has been falling for more than a decade, and low-skilled workers are leaving the work force in disproportionate numbers. Taking some lower-paying jobs out of the mix will raise the measure for average productivity, which is hardly the same as increasing the economic gains from a given set of workers or, for that matter, from putting more people to work by making them more productive.

There is more at the link, including points which do not appear in TGS.


Really, this sentence shows one of the fundamental problems when Americans look at their health care system -
'The United States spends more per capita on health care than any other country, yet without producing measurably superior results.'

No, it produces measurably inferior results, both when looking at expenditures (essentially 50% higher in per capita terms than the next two most expensive health care systems, that of Germany and Switzerland) and at a wide range of quantitative measurements measured over populations.

This is one reason the rest of the world is unable to understand why Americans start with such a flawed assumption about health care - the 'system' is measurably inferior. Individual clinics, doctors, companies, technologies, and so on may be superior, but this means nothing in terms of the U.S.'s actual infant mortality rate. The U.S. may do a fantastic job handling individual cases of premature birth in terms of highly expensive specialists and technology being deployed piecemeal fashion, generally for those that can afford it, but it doesn't change the trend -

'The United States ranked 28th in the world in infant mortality in 1998' - http://www.cdc.gov/omhd/amh/factsheets/infant.htm

Currently, the U.S. either at place 34 (United Nations) or place 46 (CIA World Factbook, 2009 estimate) - http://en.wikipedia.org/wiki/List_of_countries_by_infant_mortality_rate

Even as our healtch care expenditures increase, we are unable to keep even the position we had in 1998 - thze American system is measurably inferior, and not improving.

The infant mortality rate is a poor indicator of the overall healthcare system, that is not to say the system is not broken. The methods of measuring infant mortality vary greatly across countries with the U.S.using one of the broadest measures. For example: Some countries do not count premature births or infants born under 500 grams while some don't count those under a 1000 grams. Others do not count those who don't initially breath at birth, they are counted as fetal deaths not infant deaths. The U.S. also has a high rate of IVF which is more likely to lead to premature births, a growing number of older mothers, high risk pregnancy care that allows more women to deliver at risk infants. The point being when we compare world infant mortality rate we are talking apples and oranges.

That said, the system's cost exceeds its benefit. We can reduce its cost by reducing the number of third party payers, expanding access to low cost alternatives such a Physician Assistants and Nurse Practitioners, reducing the cost of bring new drugs to markets, expanding the supply of doctors.

We should also bare in mind that healthcare is a normal good as we get wealthier we spend more on our health.

Do you have any evidence that the sources cited (CIA factbook, etc) have not taken that into account? Sounds like a just so story to me.

The burden would be on critics to provide evidence they have taken that into account. Do the sources state that they do have? If not, it's a comparison of a self-reported statistic using definitions that vary wildly from country to country. Given those differences, it's unlikely that data even exists, and comparisons are thus meaningless.

Patients and insurance providers in the US also pay most of the development costs for new medical treatments, drugs in particular. That's not really fair, but it is a superior outcome.

"Even if you agree with this spending, it generally doesn’t produce useful goods and services that raise our standard of living"

Huh? If we reduced military spending to a level that purely defended the conus without the ability to project power, we'd lose access to almost half the oil we use about 20 minutes later; maybe more than that if Canada and Mexico were "persuaded" to cut us off by whoever replaced us. Life as you know it, and the standard of living you enjoy would disappear with it.

So you're saying that the people who have the oil would refuse to sell to their biggest customer?

Yes. If the Saudi royal family was advised they'd be killed if they sold another drop of oil to the United States, by an entity that could (without our protection there are probably several) or they believed could make good on such a threat, they would not sell another drop of oil to the United States.

Oil is not sold "to the United States" apart from a small fractioin sold to the USD govermment. It is sold to multinational corporations on the open market and the seller has no control over where it ends up after that. Even if there were some way for a specific country to prevent its oil from coming to the US, that oil would be sold elsewhere freeing up oil from other sources to reach US markets instead. The fungibility of a natural resource like oil is very high.

No, the burden of proof is on he that makes the claim; one RonB. I'd like to see Ron's evidence. Why should I accept his premise sight unseen?

JonF - I admire your optimism, but I'm skeptical multinational corporations would be significantly more resistant to the type of threat I mentioned than governments would. If the power that replaced us was interested primarily in getting their cut, you'd probably be right. If the power that replaced us fell more into the category of wanting trade stopped no matter how much damage it caused, trade would be stopped. The fungibility of a natural resource like oil is very high because the forces that favor trade in it are more powerful than those that don't. That doesn't mean that situation can't change.

Benny - We'll have to agree to disagree there; the definition issue is sufficiently important that I'd expect any study to prominently note that the data had been adjusted to correct for it. The online CIA Factbook doesn't mention any such correction (https://www.cia.gov/library/publications/the-world-factbook/rankorder/2091rank.html). Ron will have to check the hard copy for you.

Benny, it's been long documented. It's just that those who want to bash the US system don't want to listen.

I do wish Tyler would change his opening to "though it produces measurably superior results", because it does, and then would follow with his criticisms, which are just. The system is not perfect. It is too expensive, but it is the best.

As for the longevity, Japense-American women live longer than Japanese women, Dutch-Americans live longer than there European counterparts, and the same goes for African-Americans. It's just that we have a huge variety of people in the US with varying genetic starting points longevity-wise.

I just spent (wasted) an hour trying to verify your claim that Japanese-American women live longer than Japanese women. First of all, life expectancy is a slippery measure. Secondly, the US data does not regularly report life expectancy for subgroups such as Japanese-American women. Thirdly, even if they did, there is some ambiguity about whether "Japanese-American" means born in Japan and emigrated to the US or born in the US from Japanese parents. In any case, the most reliable US estimate I could find was 84.5 years at http://minorityhealth.hhs.gov/templates/browse.aspx?lvl=2&lvlid=53. Notably, they do not provide a date for this estimate, though it appears to be around 2007. I have found a number of reports that the figure for Japanese women circa 2007 is around 86 years. So, I'd like to see your evidence for your claim.

Silly, don't you know, evidence doesn't matter. Just because it is believed and fits a narrative is all that matters. George Orwell didn't know the half of it.

Not dispositive, but here's what I found at a glance:


The Honolulu Heart Program studies began in 1965 with a cohort of 8006 Japanese American men and is still continuing. Much of what we know about the health and aging of Japanese Americans is based on the several hundred publications that have come out of the studies of these men, and now some women, as they age. The cohort of Japanese men in the Honolulu Heart Program studies has a life expectancy that is longer than their counterparts in Japan,

If you really want to be confused, look at LE by state and then compare to other countries. Those in the Plains states live about as long as those in Japan, and the differences between U.S. states are actually wider than those between all OECD countries.

At any rate... it's pretty apparent that LE is a very bad proxy for the effectiveness of a national health care system. The counfounding factors are much larger than the differences between OECD health care systems.

If it is long documented then where is it? Why can you and the other not provide one single link of evidence? Again this sounds like a just so story.

I’m also not persuaded that infant mortality rates are a valid way for comparing health care systems among developed nations such as the United States:

1) The United States reports an infant mortality rate of about 0.6 percent (six-tenths of a percent) compared Monaco which reports an infant mortality rate of just under two-tenths of percent. When you have differences in the numbers among developed nations are so small (less than half a percent) that they barely qualify as a rounding error, it’s hard to make any meaningful comparisons. Which is probably why people who cite infant mortality rates in these discussions invariably talk about “ranking” rather than actual “rates.” If you tell someone we’re “34th in the world” you get one reaction. If you tell them that the difference between the United States and the “best” country (in terms of infant mortality rates) is less than one half of one percent, it’s less likely to get much of a reaction at all.

2) As others have pointed out (and this is usually buried in the footnotes of most studies that do comparisons of infant mortality rates between nations) there are differences in how nations report infant mortality statistics. One of the challenges in gathering international health data is the reliance on local reporting. If different countries are recording and reporting events in different ways, it makes it difficult to do valid comparisons.

3) After you get to a certain point in your health care system (hence the large disparities between developed and developing nations) infant mortality is driven by other factors which have little do with the availability or quality of health care.

There is an interesting report on infant mortality from the Centres for Disease Control here:

It indicates that:

Differences in definitions and reporting standards do not significantly affect the comparison between the US and Europe.

The US does better than Europe at ensuring the survival of premature infants, but worse for full-term infants.

The US has substantially more premature births than Europe, and this is a major contribution to the fact that US mortality is worse overall. I do not know why this is, and the report does not tell me (except that it is not an artifact of definitions or reporting standards).

A contributing factor is fertility treatments which are more common in the USA. Interestingly Hispanics in the USA have lower infant mortality than non-Hispanics (that may be due to having a lower rate of multiple births) they also live longer.

Average life expectancy and infant mortality rate depend so strongly on socioeconomic variables that have little or nothing to do with the health care system that they cannot meaningfully be called "results of the health care system." If you want to compare the performance of health care systems you have to look at statistics that measure the things the health care system actually does. And those statistics seem to show that the U.S. health care system generally compares favorably with those of other countries. Here is a paper that documents some ways in which the U.S. health care system produces "measurably superior results" to the health care systems of other countries. From the abstract:

We find that, by standards of OECD countries, the US does well in terms of screening for cancer, survival rates from cancer, survival rates after heart attacks and strokes, and medication of individuals with high levels of blood pressure or cholesterol. We consider in greater depth mortality from prostate cancer and breast cancer, diseases for which effective methods of identification and treatment have been developed and where behavioral factors do not play a dominant role. We show that the US has had significantly faster declines in mortality from these two diseases than comparison countries. We conclude that the low longevity ranking of the United States is not likely to be a result of a poorly functioning health care system.

I can't speak for what the rest of the world does or doesn't understand, but here is what I think is flawed with the premise. All expensive healthcare means is that it is expensive healthcare. And since an American life is more expensive (I didn't say more important) and part of this fixed cost is linked to employment, it is natural that low-skilled jobs are being underutilized.

'All expensive healthcare means is that it is expensive healthcare.'
No - it can be measured in the deaths of infants, for example. Many more Americans die as infants, while the costs of health care in the U.S. are significantly higher - 50% per capita, as the starting point to the next most expensive systems.

It is not only that 'health care' in the U.S. is expensive - it is also measurably inferior, which is a factually supportable starting point when talking about measuring U.S. health care expenditures. And yet. Americans continue to imagine that their system is somehow world leading in anything but being expensive.

The more correct formulation of 'yet without producing measurably superior results' is 'while producing measurably inferior results' - a formulation of reality which simply seems impossible for Americans to state, much less understand, as they continue to concentrate on Hollywood stories as the source of their belief.

To put it somewhat more obliquely in these days of American nostalgism (where the U.S. is certainly not exceptional) - before Reagan played the role of president, he was also hired for this gig -
'Michael Moore's Sicko turned me on to another great moment in Ronald Reagan. In 1961 The American Medical Association hired the Gipper for a viral marketing campaign dubbed I kid you not, 'Operation Coffeecup.' Doing his part to scuttle the arrival of Medicare Reagan lays down an 11 minute rap explaining how Socialized Medicine can only lead to an America where men are not free. This record was then mailed out to the 'ladies auxiliary' (doctors wives) of the AMA in each county.

Was it this little record that kept Medicare from being signed into law until July 1965? Is this why we still don't have Universal Health Care? Give a listen: Reagan Speaks Out Against Socialized Medicine (mp3)

For more on Operation Coffeecup here is a history lesson from the Huffington post (link) and some more archival material from the TexasBestGrok site (link)'

Best system money can buy - though for those getting the money, not the ones who are sick.

See, he was right even back in 1961.

"Many more Americans die as infants,"

This is not true. In fact, fewer Americans die as infants. Our IM number is higher because other countries tend to 1) abort more marginal fetuses and 2) tend to record low-weight infants as "stillbirths" where we call them live births. There are several studies on this, including one in Japan that found female infants tend to be allowed to die at a rate 5x that of males.

In fact, #2 is especially perverse, because they often use the same guidelines to determine what measures will be taken to attempt to save the infant. So they get a higher IM number by letting marginal infants die.

This was discussed at Megan's in great detail during the PPACA debate.

Here's an example of this in action: http://www.dailymail.co.uk/news/article-1211950/Premature-baby-left-die-doctors-mother-gives-birth-just-days-22-week-care-limit.html

Maybe I'll dig up some the studies if anyone's interested.


The article that you linked to says the following:

"Amillia Taylor was born in Florida on October 24, 2006, after just 21 weeks and six days in the womb .... Doctors believed she was a week older and so gave her intensive care, but later admitted she would not have received treatment if they had known her true age."

That indicates that there is at most a couple of days difference (and quite possibly none at all) between when the doctors in Florida would have withheld care, and when the doctors in Norfolk did.

True, but we still would have counted it as a live birth. Most countries would not..

Nowhere in this article does it state how the dead baby would be counted statistically, though you claim that it is evidence of this great underreporting of infant mortality rate. I'm still waiting for actual evidence of this just so story.

Good points, and the explanation lies with American non-competitiveness in a world, tradeable, economy: both defense and healthcare are industries which can, as of now, be only supplied by American domestic producers. Both are supported by an uninformed and indifferent purchaser, the federal government, which comes under pressure from doctors and hospitals and defense contractors and military personnel to continue with the same system, and ask for more.

I disagree, though, that it is difficult to measure output and performance in healthcare. Its just that when we do, we do not like what we see.

And, as for the military, what can I say: just one big self sustaining behemoth.

We are number 1.

Well thought out and well written column today. Kudos!

The United States spends more per capita on health care than any other country, yet without producing measurably superior results

This is mostly but not quite entirely true. I agree the marginal contributions of each additional dollar are lower, but we actually do have the best health care in the world by many measures. These don't show up in LE, IM, etc because the marginal contributions are very small and counfounding factors are large.

Of course the most efficient healthcare system would be one that did nothing but provide antibiotics and vaccines, because those two things together comprise about 1% of hc spending, but provide something like 90% of the marginal benefits. But I'm not sure you'd want to live there!

Very good points Dave. By the way, we also spend more per capita in education and there you don´t have all the supposedly inneficiencies associated with our health care system. I think it is obvious that the wealthiest country in the world will spend more per capita in health care - it is actually easy to prove this, we just need to compare what poor countries spend to rivher countries (per capita).

Like Dave said, the improvements become very marginal after a certain point but that shouldn´t (and it won´t) stop people from spending that money anyway.

Tyler, if I as an individual wanted to help ease TGS, where would my efforts be best placed? Should I seek graduate education in a particular field (what field?) and/or become an entrepreneur? Should I take a macro//indirect approach like lobbying for political change or becoming an economist to shed light on the issues? Or something totally different?


One of the biggest takeaways from the recent "grand GDP revisions": the "producivity miracle" of the Great Recession never really happened.


Not too different from the argument Krugman was making back in early 2009 - I disagree with Krugman a lot these days, but he did call this one 100% correctly.


We lead nine other major countries in lower extremity amputations from diabetes --

See also --

Why is that people leaving low-paying jobs are expected to *raise* apparent productivity? Are you saying that those where low or zero marginial product jobs?

Even if you agree with this spending, it generally doesn’t produce useful goods and services that raise our standard of living.

If military/security spending prevents the destruction of life and wealth from terrorist attacks (and those attacks would have destroyed more value than the spending does), then our standard of living is higher with the spending than it would have been without it. Preventing loss of wealth raises our standard of living just the same as adding more wealth does. You may think the spending destroys more value than the terrorist attacks would, but that's not what the quoted passage says.

Your argument falls short since you need to factor in how much of the terrorist activity whether real or just threatened is a function of US military incursions in areas where they really don't want us.

Surely, by now, we must realize that we have given our jobs to Asia. They in turn are doing well, their diets are so much better that they are impacting food prices around the world. Sure, the liberals will say that this is good. But in the meantime, we in the West cant make ends meet, are starting to have trouble feeding our families, cant get jobs, worry about debt etc. this will result in Western life expectations going down. We must worry about our own society. We must stop family based immigration which is resulting in unlimited and unlimitable numbers. These people dont want us in their countries, and their countries with their slum standard of living and corruption, are happy to export their people to the West. Why not, they get rid of issues with educated people with no jobs, they get rid of criminals, and they create a foreign exchange income from money sent home. We must be crazy to destroy what we have, including health.

You could add schooling as another area where people in the USA spend much more but do not get better results.
Never the less it seems to me that the standard of living is better in the USA.

Tyler-from your article with attached comments.

“The United States spends more per capita on health care than any other country, yet without producing measurably superior results."

Here are two articles (from 2004 and 2007) dealing with measurable comparative survival rates from cancer and hear attacks refuting your statement. Major is correct in his appraisal above. I don't have time to gather additonal articles because life in Helmand Province is a little busy right now.


“Expenditures on the military and domestic security have risen since 9/11, but those investments are intended to neutralize external threats. Even if you agree with this spending, it generally doesn’t produce useful goods and services that raise our standard of living. “

As a thought experiment, think of conducting sequential life satisfaction surveys in Kuwait in July and September 1990. Would a Kuwaiti in September think his standard of living had fallen? Would he think that having invested more to neutralize external threats in the 1980s might have left him better off in Septemeber 1990?

One of the biggest cognitive mistakes we can make is to assume that once we've ascended to a certain level on Maslow's pyramid, that our perch there is somehow guaranteed and that we can afford to ignore the lower levels. It's a long way down.

Other than that, it was a good article. More productivity is generally a good thing. Insight and perspective are good too.

Survival rates are not the same as mortality rates

Right. Survival rates are a much better measure of "the results of the health care system" than mortality rates.

Precisely. The constant harping about infant mortality and average life expectancy really misses the point about what physicians, in general, do.

In over 20 years of psychiatric practice (mainly military but also numerous civilian moonlighting jobs in correctional facilities, psychiatric emergency rooms, and private offices) I conservatively estimate that the impact I've had on the infant mortality rate for the offspring of my female patients' has been bumpkus. As for life expectancy, suicide prevention appears to be the most direct manner in which psychiatrists could possibly make a direct impact (whether we actually do so and to what extent are matters of controversy).

Doing some back of the envelope calculations, even if psychiatrists could somehow prevent every suicide (positing a general population rate of 12/100,000 per year leaving a slightly less than 1% lifetime suicide risk over a 75-80 year average lifespan; generously assuming that those who didn't kill themselves lived an otherwise normal life span; and assuming that given the age distribution for suicides that this saved an average of 35 years of life per suicide avoided), we would be adding at most .3 years to the average lifespan. We clearly won't have anywhere near that impact. Even adding a fudge factor for indirect effects (for example, treating depression and anxiety may improve survival after a heart attack), or substance abuse treatment, it's hard to see that we have much impact on how long the general population lives, or (talking about life expectancy at birth) how long an infant born today could expect to live.

Thus if infant mortality and life expectancy are the major outcome measure by which to judge a health care system, one would consider treating mental illness to be largely a waste of time and money. It's not of course. Most of my patients seem to appreciate the relief in suffering that they obtain. That’s a harder thing to measure.

The expense of the health care system has become such a loaded issue because Medicare is funded with tax dollars and because many workers receive health care insurance as a tax free benefit of employment. These interventions distort the marketplace. Government funding of healthcare makes sense in certain areas involving government service (e.g., the military and Veterans Administration), barebones care for the absolutely impoverished, and those who for whatever reason cannot access the marketplace (e.g., prisoners, the severely mentally ill or mentally retarded in state hospitals). Other than that, government funding and subsidization of health care makes little sense to me.

Consider automobiles, Ipads, big screen televisions and other things that people buy. No one really cares how much the public spends on these privately purchased items and they, in general, do not rise to the level of public controversy. Would that health care were like that for the bulk of our citizenry.

If you want your unborn child to have a better chance at survival, go to prenatal care and follow the obstetrician’s very basic instructions about nutrition, avoiding hazardous substances, etc. If you want to live longer, don’t smoke, drink moderately, drive carefully, watch your weight, exercise, and seek relatively inexpensive primary care for any chronic conditions you develop, such as hypertension or hyperlipidemia.

If you’re depressed or anxious, come and see me. You probably won’t live longer, but you might feel better.

Here’s another problem: Expenditures on the military and domestic security have risen since 9/11, but those investments are intended to neutralize external threats. Even if you agree with this spending, it generally doesn’t produce useful goods and services that raise our standard of living.

Isn't that statement the essence of "The Rise and Fall of the Great Powers" by Paul Kennedy from about 20 years ago.

the material is crushed and ground, thus the mobile crushing plant will be done. When the moving jaw is down

Comments for this post are closed