Interpreting life expectancy statistics and other health care issues

Matt Yglesias and Paul Krugman weigh in on interpreting life expectancy statistics across the U.S. and the Netherlands.  The fact under consideration, from a few days ago, is that the U.S. has low life expectancy overall but superior life expectancy after you reach the age of 65.

One way to interpret this data (re: Yglesias and Krugman) is to think that the U.S. should spread Medicare to its entire population.

Another interpretation is that spreading Medicare to the entire population would lead to higher expenditures on the health of the young and lower expenditures on the health of the old, for better or worse.  "Medicare for everyone" doesn't simply replicate current Medicare outcomes across a broader swathe of the population.  Medicare works as well as it does, in part, because not everyone is on Medicare or something comparable.  The U.S. split system makes Medicare, at the same time, both more effective in terms of outcomes and more costly in dollar price terms.

In this country the old don't seem willing to accept losing their privileged "first in line" position, namely Medicare for them and few others.  And Congress won't let some egghead committee come along and cut the waste out of Medicare.  The immediate results of the current proposed plan would thus be greater health care expenditures overall, pressures on doctor supply a' la Massachusetts, an even more severe long-term insolvency for the whole system, combined with an unclear resolution for all these escalating pressures.  I don't see many people on the pro-Obama side simply coming out and admitting these increasingly obvious truths, although Andrew Sullivan deserves credit on this score.

You may or may not think that's a good deal overall and perhaps you still think it's a good deal if you assign a high enough priority to covering more of the uninsured.  Or maybe (Kevin Drum has made this argument) you think it's the only path toward long-run cost control.  But if you think it's a bad deal overall, it doesn't mean you are in denial about the fundamental facts of U.S. health care supply or for that matter in denial about the cross-sectional comparisons with Europe.  Choosing French health care institutions for the United States has never been on the table, not even in evolutionary terms.

Sullivan put it very well a few days ago.  He noted that Obama — a master communicator — can't convince most people that the proposed reform is a good deal for them because…it isn't a very good deal for most people. That includes some of the people receiving new coverage, such as those receiving the new forced employer mandates.  (NB: Their wages will go down and they really need the money!  In the shorter run their wages won't go down and some of them will lose their jobs, even with phase-in a few years from now.  I'm still waiting for good Democratic economists to condemn this idea but I fear there is so much fixation on a "victory vs. defeat" framing of the struggle, and desire to skirt the CBO, that this isn't receiving the critical analysis it ought to.)

This desire to claim and promote a more universal distribution of benefits is one reason why you see so much attention paid to the public plan option.  The competing public plan at least offers the promise that some part of the proposed health care reforms will benefit virtually everyone.  My view is that a public plan would soak up many high-risk cases, benefit those cases and few other people, and that overall a public plan is superior to mandates, not Satan incarnate, but not a cure-all for the system as a whole by any means.  Advocates remain oddly silent as to what in concrete terms the public insurer will be instructed to maximize and how that fits in with pressures to extend coverage to more people.

Plan supporters are quite willing to admit "it's not nearly as good as what we wanted," but they're in denial about how truly bad the proposed reforms are in absolute terms or as a matter of economic logic and by that term I mean the economic logic of good Democratic economics, not extreme libertarianism.

In the meantime, repeat this sentence after me: if we don't solve the costs problem, in egalitarian terms things will only get worse, no matter how many people we cover.

The Republicans on this issue are (mostly) very bad and hypocritical but that doesn't give the Democrats license to proceed without a solution.


Life expectancy differs radically across racial and ethnic groups in the US. So it doesn't make much sense to compare the US average to a very homogeneous European country. In general, Americans of European descent do as well or better than their counterparts across the Atlantic on almost any health measure.
Ditto for Americans of northern European descent versus northern Europeans.

This is explained in "Do Other Countries Have the Answers" at my web site here:

What I would most like is a state administered 'barebones' healthcare system. A system that would provide cheap no thrills healthcare for relatively poor people and people who do not really want to spend a lot of money on health care. It probably wouldn't be much worse in terms of actual outcomes for most people but if the idea of money being an object or being treated like cattle you could always buy into a grander private plan. It seems that there are currently plenty of incentives to develop better procedures but very few incentives to develop cheaper procedures. I think a thoughtfully crafted state institution might be able to create such an incentive.

Could the lower "at birth" numbers for the US reflect the point that's been made about US-Cuba comparison, namely that because our superior technology for premature births allows more babies to be born who then don't survive beyond a short period, but who never would been born alive at all in other countries, it "drags down" our overall expectancy?

That argument is consistent with both the data and with the claim that US healthcare is still superior across the board.

Do Krugman and his lot ever address a) the difference in definitions and accounting for infant mortality and b) the studies that demonstrate that, absent far higher traffic mortality and homicide rates, the United States is actually first in life expectancy? Either way, given heterogenous nature of the U.S. population, it still seems bizarre to me to seize on life expectancy differences which are so small.

"can't convince most people that the proposed reform is a good deal for them isn't a very good deal for most people"

This is really a time horizon problem, with people making the incorrect comparison. It may not be a very good deal now, but infinitely better than what it will be in the future without change. They assume they will continue to have healthcare, will continue to be able to afford it, and all the evils of the existing system will be visited on the unfortunate, failing to recognize themselves among them.

I think this is a fallacious comparison. According to the links in the original MR comment, Americans have about an 83% chance of living to 65, while the Dutch probability looks to be around 90%. (There is no table, just a graph, so that's approximate.)

So the relevant question is whether the longest lived 92% (83/90) of Dutch who live to 65 have a longer life expectancy than Americans age 65. IOW, some of the people who drive the Dutch age-65 life expectancy down relative to the US wouldn't live to 65 in the US. Hence the discrepancy in life expectancy at 65.

So I'm not sure what conclusions we can draw. Maybe higher US death rates pre-65 from violence or car accidents make up the difference, but then you might expect to see the same life expectancy at 65. On the other hand, it might just be that the Dutch do a better job of staying alive to 65, which would be an argument for their system.

Much of this debate appears to be made of a lot of very arrogant statements from both sides. A bunch of dilettantes (Yglesias and Krugman) trotting out statistics are in no position to design a health care system for everyone. The healthcare "market", a mixture of free market upstarts, government subsidies, and perverse incentives defies easy and glib comparisons, particularly with other countries...comparing something so simple as infant mortality is enough to consume a whole book. What truly surprises me, for a bunch of economists especially, is the lack of humility to what the "market" knows, or could be empowered/allowed to know. Most other aspects of life in the US that work well are driven by market forces. We routinely entrust our very lives and well being to markets in other areas...yet here are debating entrusting this vast, complex system to a government we wouldn't entrust with making pencils, much less the laptops the know-it-alls blog on.

The few areas where medical care is truly market-driven, say Lasik, embody much of what we'd like to see in other parts of health care: innovation, falling prices, price transparency, and generally robust competition. It seems quite silly to waste time arguing over which top-down solution is best....other than one that creates a truly free market in insurance and care.

"The U.S. split system makes Medicare, at the same time, both more effective in terms of outcomes and more costly in dollar price terms."

What is the basis of this claim?

The notion that we can widen coverage AND save costs AND keep high levels of health all at the same time would seem stupid, if it didn't occur in other countries already.

The evidence indicates that differences in health care, or health care "coverage," have little to do with differences in "levels of health" between different countries. Other variables swamp the effects of health care - things like diet, exercise, smoking, drug use, accidents, crime, stress, etc. All the heavy lifting from "health care" in terms of increasing life expectancy and improving the overall health of the population was done by basic public health measures like sanitation and clean water, and by cheap medicines like vaccines and antibiotics that are available to essentially the entire population in all developed countries, including the United States. Even attempts to compare countries in terms of more direct and specific measures of health care "outcomes," e.g. cancer survival rates, are fraught with complications and difficulties.

I have to say that many of the stories offered by those on the right seem like labored attempts to rationalize how our healthcare system is actually secretly the best somehow. They don't seem intellectually honest, like you walked into the game with the answer that the less the state is involved the better and then even though it is definitely more expensive, and definitely provides coverage to fewer people, and it certainly doesn't seem as though it's any better as far as providing good outcomes all of this must be an artifact of an elaborate series of accounting errors?

There's something to be said for Occam's Razor here.

I know this is a side issue, but I don't know what Republicans in congress are realistically expected to do at this point. The house progressive caucus threw a fit over the extremely minor concessions made to the blue dogs. This suggest that it's unlikely Democrats would consider serious free market reforms. Perhaps if the Democrats become more desperate in the coming months, the Republicans could play a role in crafting the final bill.

The typical liberal perspective on healthcare in the US is that somehow, its quality is intrinsically inferior to outcomes produced state-sponsored medical systems. They cite statistics such as life expectancy as evidence for this. In reality, life expectancy numbers in the US are inferior to many countries in Europe due to factors unrelated to healthcare systems, but are attributable directly to lifestyle issues (Americans in many parts of the country eat a diet rich in fried foods, and live a much more sedentary lifestyle compared to their European counterparts. The US has one of the highest rates of obesity in the world, and this can contribute to a wide variety of health conditions -- totally unrelated to quality of care received. Higher rates of violent crime do not help US longevity statistics either. The main issue for most people in the US is the cost of care, not its quality, and comparatively few people care about equity of access to healthcare. The US has some of the finest medical institutions in the world, with a global reputation - Mayo Clinic, Sloan-Kettering, among many others. Yes, these institutions are extremely expensive, but if I ever have a life-threatening illness, this is where I want to be, not in some state-subsidized care facility. I have lived in two countries (Australia, Singapore) over the past 7 years, which have state-subsidized healthcare -- while adequate for routine medical checkups and basic care, I would in no sense say that they are better in quality that what the US presently offers in most significant population centers -- and they certainly are not necessarily cheaper in local currency terms.

LOL, i love the racist logic. "oh but we are not homogenus, so we cant do it" translate: Blacks are stupid, thats why they die earlier and i dont care. Bullshit. The underlying cause is that blacks etc are poorer (which is again largely explained by discrimiantion) , and poor people get screwed by the social darwinist us system.

--The Republicans on this issue are (mostly) very bad and hypocritical

Can you cite something re: the Republicans in this case? Which are very bad? Which are hypocritical? This swipe seemed unnecessary without any actual evidence.

The most important point in this post is that guaranteeing someone a doctor visit DOES NOT guarantee that they receive the care they actually need. Those of us who have experienced real socialized medicine know this issue full well. Fans of socialist medicine seem to believe that all you need to do is get someone into the system and that's the end of the story. Well, that's not the end of the story. What Republicans call "rationing" takes many forms: waiting lists are just one form. A complex network of referrals from M.D.s to blood testing centers, MDs to x-ray centers, MDs to specialists, etc. mean that just seeing a doctor isn't enough for many patients.

What this means is that the only people who get the care they need are people who can afford to take multiple days off work. Guess who that is - the rich and the middle class! The poor still don't get the care they need, even though they are legally guaranteed a visit to the doctor!

Failure to understand this is the single most disappointing aspect of healthcare "reform" "debate". There is a shocking and disgusting amount of ignorance among Americans when it comes to this issue. The American public seemingly wants to shut its eyes to what socialist medicine actually looks like. Those who tell there stories are dismissed as conservative pundits. You can't just look at life expectancy as a lone indicator of quality of healthcare. There are so many issues to deal with: Rationing, drug prices, wages, risk pools, etc. The biggest issue I have seen in the Canadian healthcare system is the fact that new and expensive treatments are strongly desired, but completely unaffordable. You can't just ignore these things and hope they go away.

I really hope people take the time to consider what socialist medicine is like before they sign their liberties away. Alas, I think it's hopeless.

"if we don't solve the costs problem, in egalitarian terms things will only get worse, no matter how many people we cover."

Having recently finished the Fuchs book (Who Shall Live), I wonder if anyone has more recent data on the UK? Fuchs points out that the class differences in the UK are no different than those in the US, i.e. those with lower incomes have lower life expectancies, and the comparison was fairly close.

Also, I keep reading stats that imply that most spending occurs in the last 6 months of life. Given that those are probably Medicare dollars, I think that Medicare expanders are basically saying "Let's spend money like there's no tomorrow." Either they forget that they used to claim that single payer systems like those in Europe are less expensive, or they know fully well that they expect Medicare to spend a lot less on seniors as it expands to the rest of the country all the while claiming that everyone is going to have everything they have now, only more and better.

@John Goodman: "So it doesn't make much sense to compare the US average to a very homogeneous European country."

As anyone who has visited the The Netherlands would know, it is not "a very homogeneous European country." It has very large immigrant groups from Turkey, Morocco, Suriname, the Dutch Antilles and Indonesia.

From the total population of 16.5m, 3.2m of 20% are from non-Dutch ("allochtoon") backgrounds. Non-Western allochtones make up 1.8m of these. Some cities are close to being majority non-Dutch. Rotterdam for example, is 47% non-Dutch, and 78% of these are from non-Western backgrounds. (All data is 1 Jan 2009, from

The Dutch 20% allochtoon is not that far below the equivalent US numbers. A quick check of Wikipedia shows that Hispanics/Latino and Black/African Americans made up 12.5% and 12.1% of the USA population (2000 census).

urgs is descrimination the reason that people of west African descent have more multiple births?

Ditto on John Goodman's post; in addition, I wonder which "good outcomes of Medicare" Tyler is referring to. Is there any evidence that life expectancy for old people improved faster after the passage of Medicare in the 1960s than it had before? More generally, most of the improvements in life expectancy in the U.S. occurred before the interventions of government into the medical insurance scene around mid-century. So conceding to the Paul Krugmans at the outset the "success" of Medicare seems like an excellent way of needlessly losing the argument.

The bottom line is that real people, when they screw up, expect to have to begin the process of recovery by undoing their past mistakes. No reason that logic shouldn't apply to governments as well. On that principle, let's begin by: 1) undoing the employer-provided tax exemption from the WWII era, making medical insurance both individual and portable; 2) undoing the prohibition on interstate shopping for medical insurance; 3) undoing the maximization of minimal state standards for health insurance policies, which makes them so much more difficult to afford than almost any other kind of insurance.

Give these 3 simple corrections 5 years or so to play out and see where we are. THEN it will be time to take stock and decide what new government interventions we might like to contemplate. To pile new sausage-like errors on top of the existing ones is a good example of Einstein's famous definition of insanity.

Kel is making the same mistake I see repeatedly- confusing spending on marketing and advertising with money spent on administration.

"Remember, the "free market" healthcare system doesn't make money curing you. It makes money by creating profitable health care products. If there are less expensive but unprofitable ways to achieve a better outcome, well, too bad."

That's a child's view of the profit incentive. Remember that the "free market" healthcare "system" has many actors trying to optimize along different lines. If one provider has a less expensive means than another to cure you, your insurance company should favor the less expensive means. These observations have their analogies in gov't care: no office wants their budget to be cut, so they have no incentive to choose less expensive alternatives and leave money unspent. But that's just the closest analogy and not the best one to explain why gov't healthcare doesn't have all of the advantages its cheerleaders think it does (it has some, but not as many nor as strong as they think). Something like 50% of all new therapies (drug and surgical) have their start in the US and not in the EU, and why generics cost less in the US than in the EU or Canada - that can't be explained with this "they want to screw you, nothing more and nothing less" theory.

The real barrier to change though is neither of these issues but the reluctance of the health care professionals and patients to adopt more rational treatment approaches. Got a boo-boo? No more "take two aspirin and call me in the morning"; today, it's "let's do a complete blood work-up and an MRI just to make sure it isn't flesh-eating bacteria or some unknown cancer." None of which has any significant outcome on life expectancy stats, but does increase the cost.

Excellent point, Yancey. Isn't this something that could be brought to the WTO? It reminds me of my time in China, where you could find any US movie, software, music, or technology for 10-25% of the US price. Sure, it doesn't cost much to burn a DVD and stuff it in an envelope, but they're ignoring the true costs of production. The US made a stink about China to the WTO and rightfully won, but we are sitting idly by while Europe does the equivalent with medical/bio-technology. No question though - this is a political powder keg, and I doubt anyone wants to make a fuss.

The Republicans on this issue are (mostly) very bad and hypocritical but that doesn't give the Democrats license to proceed without a solution.

Let's see, Democrats included a provision that requires that Medicare pay for a patient-doctor end of life consultation every five years and the Democrats are then attacked with ads claiming the Democrats are forcing Medicare patients to sign documents authorizing their doctor end their life.

And the Democratic proposal in response to the high costs of Medicare paying for lots of medical services with no apparent benefit results in a proposal to study medical services and recommend the cost effective ones is called in ads a rationing of health care that will result in death.

And the Republicans are running ads that claim that the Canadian government was forcing a woman to die until she came to the US and got care after mortgaging her house, and that is what the Democrats are going to force on everyone in the US. Of course she wasn't going to die and would have gotten the care for less in Canada a few months after she got the care in Canada, but hey, that just proves that limiting health care spending is an evil Obama plot to kill Americans.

Now a much cheaper model for delivering health care has been demonstrated by the private sector, but no for-profit private sector group is rushing to implement the system of Rural AMerica Program No Cost Health care. Why, we don't even see hundreds of private sector not-for-profits implementing this model. This No Cost Health care model proves that the government does not force medical costs in the US to be extremely costly. See for details.

Now if the free market can't deliver sufficient no cost health care to everyone who can't aford anything but no cost health, then who can, other than the government.

But even with a government program of that sort, I'm sure Republicans would attack the Democrats for forcing a hundred million Americans to get their care in cattle barns, arguing they would be better off without any health care at all.

Now, look at the difference between 1-year death probabilities:

Age USA Netherlands Difference
0 0.680% 0.480% 0.200%
10 0.013% 0.014% -0.001%
20 0.094% 0.039% 0.055%
30 0.104% 0.052% 0.052%
40 0.195% 0.104% 0.091%
50 0.453% 0.313% 0.140%
60 0.945% 0.772% 0.173%
70 2.155% 1.849% 0.306%
80 5.301% 5.676% -0.375%

Notice that even at age 70, several years after becoming eligible for medicare, American death probabilities are higher than dutch death probabilities. It is only after age 76 that the US one-year death probabilities become better than the dutch.
Also, look at the values at ages 20, 30 and 40. At ages 30 and 40, an American is about twice as likely to die within the next year as a dutch; at age 20 he is 2.4 times more likely to do so. Can such a huge difference be explained solely by the availability or unavailability of health care? Or is it not more likely than various things such as obesity, crime and traffic accidents (all of which are more common in the US than in the Netherlands) are also important drivers? If the availability of Medicare where the most important factor, you would see a sharp reduction in the difference between US and Dutch mortality around age 60-65... and you don't; you actually see a sharp increase in the difference, which is much more consistent with the hypothesis that bad-risks are dying earlier in the US due non-healthcare factors.

@ Kel - Pharma companies DO NOT spend more on advertising than on R&D. As an example, Pfizer's FY08 10-K report (that's audited financials filed with the SEC) lists an R&D expense of about $7.9 billion, and advertising expense of about $2.6 billion. I'm not going to bother checking the other major companies, but given the disparity between those expenses in this example, it's unreasonable to posit that a similarly situated (i.e. large, established) pharmaceutical company would spend "much more" on advertising than on R&D.


The US healthcare system has always had very different compensation rates for procedures and treatments vs. cognitive and preventive healthcare services. One interpretation of the netherlands/US differential is that our system is build to treat illness -- not keep people healthy. I think that is an under-appreciated difference in philosophy.

The young are disproportionately benefited by a preventive system. By the time you are old much of the bang for prevention has been diminished and the most important feature becomes access to therapies.

Put another way, the US healthcare system is dominated by medicare and provides excellent services for sick people -- just what a medicare beneficiary needs. The Netherlands is perhaps geared towards a larger client base with more general healthcare needs including prevention.

Regardless of the validity of the comparison, the fact remains an anthropologist/economist from another planet looking at the US healthcare system would easily conclude that the system is trying to optimize around large numbers of chronically ill patients needing frequent and costly procedures. Afterall, that is how the decision makers, doctors, get paid. And that seems to be exactly what we have. It's not rocket science, it's incentives. Change how US doctors are reimbursed and you might find very different utilization patterns.

I recently came across your blog and have been reading along. I thought I would leave my first comment. I don't know what to say except that I have enjoyed reading. Nice blog. I will keep visiting this blog very often.


I imagine that Republicans could do far better by looking around and picking up a plan to stand on

Really? Just saying no worked better for the Democrats on Social Security reform.

Medical companies spend more on advertising and marketing than on research. A lot more.

So? Advertising and marketing is supposed to increase profit, not cost money. If they didn't market, they wouldn't have more research money; likely less. Especially if it's for "lifestyle drugs," since those are exactly the kind of drugs that people won't buy and will live with the condition instead. No one advertises for something that you have to take or die, because you'll try to get that without advertising.

And as people (and Arnold Kling) have noted, the difference here is mostly accident and homicide statistics. Those mostly affect young Americans, which is why Americans who escape their twenties keep catching up.

Let's talk about about health care. I know! I'll drop Paul Krugman's name, further ingratiating myself. I only dropped it twelve times last week, so this is a good time for a makeup name drop. After all, nothing will cement my credentials as a serious thinker [and book seller] than to cite The Krugman.

Now, I wonder if better health care will reverse the trend of autism, saving us from the influence of autistic economics...

Comments for this post are closed