Why American Health Care Is So Expensive

Slate.com reports that it isn't the fault of the insurance companies or health care providersd; rather, it's our fault:

The debate about health care tends to be informed by three notions about health insurance:

The profits of private insurers are so big that cutting them out would meaningfully lower costs.
Private insurance clearly costs more than a government-run system such as Medicare.
Mergers that have created a small number of huge and powerful insurers increase health care costs.
None of these is true.

Myth No. 1: Insurers' profits are responsible for our health care costs.

This is the most pervasive and most crowd-pleasing of the health care myths. The profits of the big health insurance companies are central to the rhetoric of the health care debate, figuring heavily in the Democratic primary campaign. Barack Obama's platform includes a promise to force insurers to spend enough on care "instead of keeping exorbitant amounts for profits and administration." Michael Moore, the director of Sicko, has hammered the point repeatedly, thundering about how insurers maximize profits by "providing as little care as possible."

The problem here is that between them the five biggest health insurers—UnitedHealthCare, Wellpoint, Aetna, Humana, and Cigna—which cover 105 million members, last year had profits between them of $11.8 billion. This is not a small number; these are very profitable companies. But total U.S. health care costs last year were in the area of $2.3 trillion.

So, with a membership that included a little more than half of the Americans covered by private insurance, these five insurers' profits came to 0.5 percent of total health care costs. (One interesting point of comparison: In 2006, the income earned by the 50 biggest nonprofit hospitals alone came out at $4 billion.)

Critics also argue that insurance companies pass along excessive administrative costs to their customers. Wellpoint, for instance, spends 18 percent of the premiums it takes in on sales and administrative costs. That represents a real concern but merely raises the next question: Can a government-run program that cuts out insurers do it for less?

...

Diagnosis

Patient, heal thyself. It's not insurers that push expensive drugs, long-shot end-of-life treatments, and redundant procedures. It's customers who ask for them. And mainly doctors and hospitals who profit. How to deal with those issues is a question that will affect the health care bottom line more than whether it's the government or private companies that provide insurance. Too bad it's one we have hardly even started to answer.

I wonder if Michael Moore is ready to lose his (free) lunch?

Health Care Wonkiness

Matthew Yglesias muses on the relative demerits of universal health care coverage, and kinda-sorta offers a compromise:

One way or another, in other words, if we aren't going to try to have a free market in health insurance then we're going to be relying on the idea that we can have an adequate level of accountability through the political system. There's reason to be skeptical about politics' ability to deliver in the right ways, but the whole case for universal health care programs is founded on solid evidence that market accountability doesn't deliver socially desirable outcomes on the health care front. Having politicians create a pseudo-market that's highly distorted by mandates, regulations, and subsidies is counting on politics to deliver the good every bit as much as a single-payer system would. ...

The other is that you can hybridize single-payer and market elements. Make preventive medicine free. Make it less than free -- have nurse practitioners kicking down doors and immunizing children. But for other things, you can implement cost-sharing that's scaled to the recipient's income. You guarantee that lack of funds never forces anyone to go without care, but you also ensure that everyone has an incentive to at least think a little about whether or not he really needs treatment. See Jason Furman's paper on cost-sharing and then imagine that all costs not borne by the individual would be picked up by the government. That gives you some market pressure, but also substantial equity because price-to-consumer is determined by ability to pay in a way that ensures that struggling families aren't left out in the cold.

I'm not really sure how we can institute a system that can "guarantee that lack of funds never forces anyone to go without care, but you also ensure that everyone has an incentive to at least think a little about whether or not he really needs treatment" since the only way I know how to do that is through the price mechanism (or rationing-by-government-dictate, but in that case there's no choice at all for the individual). But ignoring that, his preferred plan sounds like something close to the "Jane Galt Health Care Plan," to which I (basically) subscribe:

Have the government pay for all health care expenditures above 15% of adjusted gross income, and cover 100% of health care expenditures by people living under 200% of the poverty line.

I'd like to see the two of them discuss this (Bloggingheads?!?) in a bit of depth, and see if they can reach some sort of resolution.

To Spend or Not to Spend

Matthew Yglesias says:

Or to look at it another way, if Hillary Clinton's entire agenda were enacted, her climate change proposals would wind up doing more to improve public health than would her health care proposals.

Why? Well, for one reason, increasing spending on health care has almost no effect on health outcomes. So says Robin Hanson, and he advocates cutting health care spending in half to bring spending and outcomes in line.

Lessons from Singapore

Bryan Caplan has evidently been reading up on Singaporean government policies. He has two posts in favor of two Singaporean policies: health care and unemployment. First, health care:

If the following is true, all the comparisons showing that the U.S. greatly outspends Europe without getting better health are beside the point, because Singapore makes Europe look like the U.S.:

The Singapore government spent only 1.3 percent of GDP on healthcare in 2002, whereas the combined public and private expenditure on healthcare amounted to a low 4.3 percent of GDP. By contrast, the United States spent 14.6 percent of its GDP on healthcare that year, up from 7 percent in 1970... Yet, indicators such as infant mortality rates or years of average healthy life expectancy are slightly more favorable in Singapore than in the United States... It is true that such indicators are also related to the overall living environment and not only to healthcare spending. Nonetheless, international experts rank Singapore's healthcare system among the most successful in the world in terms of cost-effectiveness and community health results.

How does Singapore do it? ...

More details on how Singapore's system works:

• There are mandatory health savings accounts: "Individuals pre-save for medical expenses through mandatory deductions from their paychecks and employer contributions... Only approved categories of medical treatment can be paid for by deducting one's Medisave account, for oneself, grandparents, parents, spouse or children: consultations with private practitioners for minor ailments must be paid from out-of-pocket cash..."
  
• "The private healthcare system competes with the public healthcare, which helps contain prices in both directions. Private medical insurance is also available."
  
• Private healthcare providers are required to publish price lists to encourage comparison shopping.
  
• The government pays for "basic healthcare services... subject to tight expenditure control." Bottom line: The government pays 80% of "basic public healthcare services."
  
• Government plays a big role with contagious disease, and adds some paternalism on top: "Preventing diseases such as HIV/AIDS, malaria, and tobacco-related illnesses by ensuring good health conditions takes a high priority."
  
• The government provides optional low-cost catatrophic health insurance, plus a safety net "subject to stringent means-testing."

This doesn't sound too far off from the "Jane Galt" health care plan, which I generally support, although Singapore's plan is more detailed. But Caplan also likes Singapore's mechanism for fighting unemployment:

Imagine my surprise, then, when I discovered that Singapore has figured out a stunningly clever way to use tax cuts to reduce unemployment. Instead of focusing on stimulating demand, Singaporean tax policy hits the margin that matters: labor costs. When there is a surplus of labor, they cut employers' share of the payroll tax (known in Singapore as the CPF). Details appear in Henri Ghesquirre, Singapore's Success:

The government directly intervened to temporarily lower the cost of business in Singapore through... its power to lower the CPF contribution rate of employers...

Elsewhere, substantial nominal currency devaluation is often the last and only resort in the face of downwardly sticky nominal wages, often with higher inflation as an undesirable side effect. In contrast, Singapore uses the direct intervention methods at its disposal. In addition, there is built-in wage flexibility, because an important portion of workers' remuneration is automatically lowered if GDP falls short of target.

With flexible wages, of course, it doesn't matter who legally pays the a tax. But the whole problem with recessions is that wages are somewhat sticky - you can have surplus labor for years before wages fall enough to restore full employment. By cutting employers' share of the tax, the Singaporeans greatly speed up the wage adjustment process.

Caplan asks whether such a system for correcting unemployment would be successful in the United States. I don't think it would be, for the simple reason that our government is much more democratic that Singapore's. In the U.S., populists would likely oppose lower corporate taxes in lean periods even if the positive effects were strongly felt by workers. Economic conservatives would oppose raising corporate tax rates in boom periods. The advantage of this plan is its flexibility, which can be taken advantage of by an authoritarian government but not by a democratic one. The proper policies would likely not be enacted at the appropriate times, so the system would break down.

This is a shame.

The Economic Future of America

In the newest issue of The Atlantic, Megan McArdle looks into her crystal ball to examine what will happen when the baby boomers have all retired. Specifically, she is concerned with how we are going to pay for our Social Security and Medicare obligations to them. She is cautiously optimistic. Well recommended.

As a companion, McArdle participated in a roundtable discussion with Clive Crook and Philip Longman which may be found here. On her blog, McArdle responds to some general criticisms of her piece here.

I am taking some flak, as expected, for the optimism of my piece on the baby boomers. How can I be so darn cheery when we've got $55 trillion in unfunded obligations coming down the pike?

Well, the point of the piece was to stop focusing so much on cash figures. All government spending numbers are impressively large--that's what happens when you're dealing with a very rich, very populous nation.

But our entitlement problem is not that our unfunded liabilities have a net present value of $55 trillion. That is a very big number, but we're not going to get handed a bill for it all next week. It's a slice of a much, much bigger figure: the NPV of our future income.

Prices are very, very useful things. But they can often obscure as much as they hide in economic debate: a $20,000 Hermes handbag does not consume 100 times as many economic resources as a $20 Target handbag, and the illusion that it does fuels much of the current debate over redistributing wealth.

The ultimate price tag of Medicare is hazily unknowable, and really, the number does not much matter. And the argument that this spending is "unsustainable" is not particularly interesting; as Herb Stein famously said, "If something cannot go on forever, it will stop." If the spending is really, actually, unsustainable then the government will eventually stop trying to sustain it. These are not the interesting questions.

My (partial) list of the interesting questions:

1. Is Medicare well structured to provide good health care to the elderly?
2. Is Medicare encouraging today's workers to save less than they will ultimately need? In other words, is the program likely to be severely curtailed in the future, leaving workers who counted on it worse off than they would have been had it never existed?
3. Is Medicare encouraging, or discouraging, the medical innovation that could make future generations better off?
4. What is the deadweight loss of the taxation required to pay for Medicare, and does this represent a good use of our money?
5. Is this massive transfer from old to young just? Does this answer change if it turns out that future generations are likely to massively trim the program?

My answer is that Medicare is a bad program on many fronts: badly structured, economically and medically destructive, and a fairly injust transfer of resources. But none of those things have anything to do with eye-popping NPV figures. A bad program doesn't become a good program just because it's cheap.

The End of Fiscal Politics

Much of American politics today is centered on fiscal policy. The right wishes to shrink the size of the government and rebate many taxes. The left wishes to expand the size of the government and pay for it by raising many taxes. But in the near future, neither one may be possible, says Megan McArdle:

The budget problem isn't in 2041; the budget problem is now. Sometime after next year, the Social Security surplus will shrink, starting to put pressure on the budget. Democrats trying to implement spending plans will start to find their tax increases eaten, not by national health care, but by seniors. By 2011, the problem will be large. By 2017, money will be flowing from the general fund to social security. By 2025, the hole will be about as big as it's going to get. Around about 2015, social progressive plans will be DOA. Republican tax schemes will be DOA. The only thing we will talk about for the following 15 years is where to find the money to pay for Social Security and Medicare.

Elsewhere, and not in response, Bryan Caplan says, while discussing health care reform:

What is odd is that the biggest obstacles to redistribution are the Democrats' favorite programs, Medicare and Social Security. Those programs are going to soak up all of the tax increases the Democrats can dream up, and then some. Having spent the last six years denying that there is a problem, they are going to spend Hillary Clinton's first term watching these non-problems drain tens of billions of dollars from the Treasury, leaving them very little room to maneuver on redistributionary spending.

They are both right, but the bigger problem is surely Medicare, and there is no easy answer. Social Security will work itself out in time, as the Boomers die off and/or the economy expands enough to compensate them. At worst, it's a one-generation problem, and not an especially crippling one. If we must, we can simply raise the eligibility age by a couple of years, and the problem disappears.

Medicare/caid, however, is the bastard gift that keeps on giving. If health care costs continue to inflate at something close to current rates, we will be forced to start rationing care along a steeper health-to-income gradient. In other words, the inequality which already exists in access to care between poor and rich will broaden. Those who can pay for care will get it; those who can't, won't. The only other answer is to ration care by government mandate, perhaps coupled with price controls, and nobody wants that.

This problem isn't unique to America. Japan has had a fiscal crunch for some time now, mostly due to the burden that the elderly there are placing on the rest of society. Much of western Europe has some sort of similar problem. Basically, social democracies are largely susceptible to demographic and economic changes. If the demand for social services increases (due to an aging population, say) or price inflation is high, then the system crumbles on top of itself.

Economic realities may reduce political bickering in the next decade, but surely not for positive reasons. The future will be a test for politicians: play the game honestly, and accept the economic hardships and inflexibility which are coming, or try to cheat the system (by inflating the current, for example, or instituting price controls) and suffer extremely dire consequences.

Fact of the Day

- % of Walmart employees who get health insurance through Walmart: 47.4

- % of Walmart employees who get health insurance from some source: 90*

- % of Starbucks employees who get health insurance through Starbucks: 42

as a side note, Starbucks spends more on employee health insurance than it does on coffee.

* ~ 43% of Walmart employees receive health insurance from their spouse's employer or from the government.

A Data Point Ignored in SiCKO

Andrew Sullivan's got the goods, quoting the Telegraph:

Tony Blair poured millions into Britain's socialized healthcare system, pumping unprecedented resources into a healthcare system that Michael Moore admires and the American left loves. This is the result:

More than 70,000 Britons will have treatment abroad this year – a figure that is forecast to rise to almost 200,000 by the end of the decade. Patients needing major heart surgery, hip operations and cataracts are using the internet to book operations to be carried out thousands of miles away.

India is the most popular destination for surgery, followed by Hungary, Turkey, Germany, Malaysia, Poland and Spain. But dozens more countries are attracting custom. Research by the Treatment Abroad website shows that Britons have travelled to 112 foreign hospitals, based in 48 countries, to find safe, affordable treatment.

Wait just a second, says Ezra Klein:

Indeed, there are more Americans -- 100,000 -- traveling abroad for cosmetic surgery alone than there are Britons seeking any type of services in foreign lands. America is actually driving the medical tourism industry that some Britons are taking advantage of. The growth of foreign treatment centers aren't a result of the failings of the British health care system (of which there are many). They're a result of the cost of American health care, and the huge numbers of sick individuals we price out.

Well, first things first. There are five times as many Americans as Britons, so comparing raw numbers isn't appropriate. Also, Americans are going for plastic surgery while Britons are seeking treatment abroad in order to stay alive. Obviously they aren't the same. In addition, the American model is a free trade model, so "medical tourism" isn't exactly a critique of the system; it's part of the system. Seeking out lower prices for procedures is what you're supposed to do. In Britain, however, everything is intended to be provided by the government (and at much lower cost, as Klein goes on to remind us). But the fact is that the service is so unsatisfactory that people are choosing to opt out of Britain's system in favor of the free trade model. The fact that citizens living in countries which provide UHC are choosing to forego the free coverage (for which they've already paid, through taxes) in favor of a free exchange is not supportive of arguments in favor of UHC.

So, tell me: how is this supposed to support Klein's position that we should move to a European-style UHC system post-haste?

WHO Health Care Measures Are Wrong

Eat your heart out, Michael Moore:

So what's wrong with the WHO and Commonwealth Fund studies? Let me count the ways.

The WHO judged a country's quality of health on life expectancy. But that's a lousy measure of a health-care system. Many things that cause premature death have nothing do with medical care. We have far more fatal transportation accidents than other countries. That's not a health-care problem.

Similarly, our homicide rate is 10 times higher than in the U.K., eight times higher than in France, and five times greater than in Canada.

When you adjust for these "fatal injury" rates, U.S. life expectancy is actually higher than in nearly every other industrialized nation.

Diet and lack of exercise also bring down average life expectancy.

Another reason the U.S. didn't score high in the WHO rankings is that we are less socialistic than other nations. What has that got to do with the quality of health care? For the authors of the study, it's crucial. The WHO judged countries not on the absolute quality of health care, but on how "fairly" health care of any quality is "distributed." The problem here is obvious. By that criterion, a country with high-quality care overall but "unequal distribution" would rank below a country with lower quality care but equal distribution.

It's when this so-called "fairness," a highly subjective standard, is factored in that the U.S. scores go south.

The U.S. ranking is influenced heavily by the number of people -- 45 million -- without medical insurance. As I reported in previous columns, our government aggravates that problem by making insurance artificially expensive with, for example, mandates for coverage that many people would not choose and forbidding us to buy policies from companies in another state.

Even with these interventions, the 45 million figure is misleading. Thirty-seven percent of that group live in households making more than $50,000 a year, says the U.S. Census Bureau. Nineteen percent are in households making more than $75,000 a year; 20 percent are not citizens, and 33 percent are eligible for existing government programs but are not enrolled.

For all its problems, the U.S. ranks at the top for quality of care and innovation, including development of life-saving drugs. It "falters" only when the criterion is proximity to socialized medicine.

Fighting Terrorism By Legalizing Drugs

Daniel Drezner is pointed by Mark Thoma to the Financial Times, where Willem Buiter explains how we can fight terrorism by legalizing drugs:

A pragmatic argument against criminalising drugs is that criminalisation creates vast rents and encourages criminal entrepreneurs to use violence, intimidation, bribery, extortion and corruption to extract these rents. Another pragmatic argument is that it is pointless to waste resources fighting a war that cannot be won. The losing war on drugs wastes resources that could be used to fight terrorism and other crimes.

Another important argument for legalising, in particular, all cultivation of poppy and of coca (and their illegal derivatives) is that this would take away a vital source of income and political support for terrorist move- ments, including the Taliban and al-Qaeda in Afghanistan, and Colombia’s Revolutionary Armed Forces (Farc) and various paramilitary groups.

The United Nations estimates that opium production in Afghanistan grew to more than 6,000 metric tonnes last year with a value exceeding $3bn. It is the origin of more than 90 per cent of the world’s illegally consumed opiates.

A significant portion of the profits flows to the Taliban, who act as middlemen in the opium business. They combine extortion and threats of violence towards the poppy farmers with the sale of protection to these same farmers against those who would destroy their livelihood, mainly the Nato allies and the Afghan central government.

Following legalisation, the allies in Afghanistan could further undermine the financial strength of the Taliban and al-Qaeda by buying up the entire poppy harvest. If a sufficient premium over the prevailing market price were offered, the Taliban/al-Qaeda middle-man could be cut out altogether, and thus would lose his tax base. Winning the hearts and minds of poppy growers and coca growers is a lot easier when you are not seen as intent on destroying their livelihood.

Of course, there are plenty of other reasons to legalize drugs, and Buiter acknowledges some of those as well:

The principle-based argument for legalisation is that behaviour that harms others ought to be criminalised, not behaviour that hurts only the person engaged in it. It is not the government’s job to protect adults of sound mind from the predictable consequences of their actions.

If the public is ill-informed about the consequences of drug taking, there is an educational role for the state. Children should be protected from drugs, as they are from tobacco and alcohol. So should the mentally ill and mentally incapacitated. Parents should be paternalistic, but when it comes to mentally competent grown-ups the state should not be. It is not the responsibility of the state to ensure our “happiness” – whatever that is. That is the road to a Brave New World.

He could have mentioned, but didn't, the extreme social ills of drug-centered gang and mob violence, the burden of imprisoning so many people for committing victimless "crimes," and the resources saved by ending the imprisonment of those people, as well as slashing the government bureaucracy dedicated to hunting and arresting those who illegally peddle drugs. Plus the revenue which could be had by legalizing and taxing the drugs. Plus the social health benefits caused by government testing and regulation, thus preventing unsafe drugs from hitting the street. Etc.

It's a great article. Obviously, the case for legalization is much stronger than simply as a mechanism of combating terrorism, but that argument is persuasive enough on its face to open up the debate in broader contexts than has been previously done.

Big Business for Universal Health Care?

Matthew Yglesias links to a BusinessWeek article which highlights wait times for American health care consumers. Of course, this argument is often used against universal health care systems in Canada and the U.K. by those who support privatized health care systems. To me, this isn't really a substantive point: in Canada and the U.K., wait times are common for specialized and elective care, while basic doctor visits are quicker. It's commonly said that hip surgeries in Canada require about a 9 month wait time, and if you are old enough in the U.K. you are simply ineligible for one. In the U.S., you might have to book a doctor's visit a few weeks in advance, but you get emergency care and elective procedures much more quickly. There is a trade-off there, and the U.K./Canadian systems are weighted to one end while the U.S. system is weighted to the other.

But the larger point was missed by both BusinessWeek and Yglesias: those in favor of a universal health care system could find an ally in Big Business. After all, a universal system would remove the high costs of health care provision currently being paid for by businesses. It requires a huge, expensive human resources bureaucracy, as well as opportunity costs of negotiating with insurance companies, form-filling, etc. A case might be made that the efficiency gains in the business sector from removing the burden of providing health insurance for their employees might mitigate some of the costs of nationalizing the insurance industry. I've never seen an empirical approximation of this effect; it may be small. But it's certainly a political tool which could be used by those in favor of universal insurance. If Big Business gets on board with universal coverage, then it's hard to imagine that it won't succeed politically. Why proponents of universal coverage haven't latched on to this is beyond me; perhaps those on the left are simply uncomfortable with building a coalition with Big Business, even though it is a rational alliance.

Medicine Blogging

Of course, a lot of attention is being paid Michael Moore's new take on the health care industry; this is to be expected. A lot of bloggers have responded to the film, and to each other. This is also to be expected. Here's some of the best stuff:

Austin Goolsbee, Barack Obama's economic advisor, on the (un)feasibility of Sicko's policy prescriptions:

So, to do as Moore wants in the United States, you would need to do more than just overcome the insurance industry. You would need to cut the salaries of doctors, reform the legal system, enrage our allies by causing their prescription drug costs to escalate, and accustom patients to a central decision-maker authorized to determine what procedures they are and are not allowed to get. Unless every one of these changes comes together, Moore's new system would end up costing an enormous amount of money.

But Arnold Kling isn't buying what Goolsbee is selling.

Plus:

Jane Galt (er, Megan McArdle) on Matthew Yglesias on CNN:

Aside from that, however, most of my ideas are simple, elegant, and doomed to die an agonising death in committee--like bringing back open wards, slashing the salaries of doctors and nurses, or denying expensive treatments to the elderly, disabled, and other severely ill people. If Matt has better ones for trimming down that 7.7% to a level where we might feasibly cover 200 million other people with what remains from France's spending, I am very interested to hear it.

Jane Galt (er, Megan McArdle) on adverse selection and moral hazard:

For those who are not familiar with the concept, adverse selection is what happens in markets like those for insurance, when one side has much more information than the other. In the case of health insurance, it means that only those who think they are likely to be sick will buy insurance; which means that the average cost of covering health care for those people will go up; which means that the health insurance company will raise the premiums; which means that those who aren't that sickly will stop buying it; which means that the average cost of covering health care for those people will go up . . . . and presto, suddenly there's no market.

And the Boston Globe on "medical self-defense"

I tend to be on Ms. Galt's (er, Ms. McArdle's) side of this particular debate. If we really decide that we want to spend a whole shit-ton more money on health care, and if we are willing to accept the tradeoffs that go with it, then let's do it. But let's not pretend that there is a magic bullet that can everyone free health at no societal cost. Using unrealistic hypotheticals (i.e. lies) to sell a policy that is based on normative priors and not positive analysis is not only bad politics; it's unethical.

Cuban Health Care

Michael Moore's latest travesty of truth -- Sicko -- premiered at Cannes and will receive a full release soon. After watching it, many people will be convinced that the Cuban health care system is better than America's. The NY Times says that it ain't so, especially if you are a commoner:

“Actually there are three systems,” Dr. Cordova said, because Cuba has two: one is for party officials and foreigners like those Mr. Moore brought to Havana. “It is as good as this one here, with all the resources, the best doctors, the best medicines, and nobody pays a cent,” he said.

But for the 11 million ordinary Cubans, hospitals are often ill equipped and patients “have to bring their own food, soap, sheets — they have to bring everything.” And up to 20,000 Cuban doctors may be working in Venezuela, creating a shortage in Cuba.

Elsewhere, British magazine The Prospect has an eyewitness account:

Healthcare and education are supposed to be the redeeming graces of the regime, but this is questionable. There are a large number of doctors, but, according to most Cubans I know, many have left the country and the health system is in a ragged state—apart from those hospitals reserved for foreigners—and people often have to pay a bribe to get treated. Michael Moore, the American film director, who has recently been praising the system should take note of the real life stories beneath the statistics. I went into a couple of hospitals for locals on my latest visit. In the first, my friend told me not to say a word in case my accent was noticed, as foreigners are not allowed in these places. I was appalled by the hygiene and amazed at the antiquity of the building and some of the equipment. I was told that the vast majority of Cuban hospitals, apart from two in Havana, were built before the revolution. Which revolution, I wondered; this one seemed to date from the 1900s.

And Jane Galt disputes the notion that Cuban infant mortality rates are lower than those in the U.S. as well, and reminds us that there are three kinds of lies: lies, damn lies, and statistics.

Very interesting article on Cuban v. American infant mortality shows one of the trickiest aspects of using statistics: making sure you're not comparing apples to oranges.

The reason this is so difficult is that often statistics which sound like they're the same thing actually aren't. For example, comparing rape statistics. Most of us think of rape as being forcible intercourse with an unwilling victim. However, there are studies that have included such things as statutory rape (which may be appalling, but isn't what we're trying to get at, which is the rate at which people are sexually assaulted), "rapes" in which the victim never struggled or said no, but merely reported later that she was uncomfortable, or other definitional expansions that make it hard to establish valid comparisons with other studies that focused on forcible intercourse.

In this case, the article points out that while Cuba seems to have a lower infant mortality rate than America, this appears to be because extremely low-weight births for which American doctors perform heroic intervention (and thus get recorded as a live birth in America, followed by a death a few hours later), get reported as stillbirths in Cuba. So it's very important, when you see a statistic, to ask yourself if the two numbers are really comparing the same thing.