Category: Medicine

Economic arguments I have never understood

Have you ever heard the claim that U.S. medical care is in trouble because we subsidize third-party insurance through the tax system? Glenn Hubbard presented this view in the Wall Street Journal this Tuesday. Hubbard writes:

Reform the tax treatment of health-care expenses. The most far-reaching and misguided government policy, established more than 60 years ago, allowed employer-provided health care to be exempt from taxation. Under this policy, medical care purchased through an employer’s insurance plan is tax-free, while direct medical-care purchases by patients must be made with after-tax income. The tax preference for employer health insurance has been instrumental in creating today’s third-party payment system. In this perverse world, true insurance, in the form of coverage for catastrophic health events, is the exception; prepaid health care, in the form of coverage with low deductibles and copayments, is the rule. The tax preference for insurance is the primary reason five out of every six dollars of health-care spending are paid by third parties…

Low copayments and deductibles fuel excessive cost growth and breed wasteful medical practice…consumers have little incentive to limit their use of unnecessary medical-care services, little incentive to shop for the health plan that best suits their needs in a cost effective way, and little incentive to evaluate their care on the basis of value.

But I’m stumped. If the argument is that tax deductibility leads to too much health care, I can see the logic. But then the problem is in the pretzels and beer markets; health care should be doing fine, albeit in bloated form.

Alternatively, it might be argued that buying health insurance involves a negative externality on others. Maybe insurance companies are intrinsically bad monitors, and more insurance corrupts the system as a whole. Grant this premise, but where do we end up?

1. We would have a good argument for taxing insurance purchases. Yet the insurance point is rarely raised with this conclusion in mind. We might have (yikes!) an argument for greater government involvement in health care.

2. If insurance companies are such poor cost monitors, why doesn’t this raise premia accordingly? The poor monitoring of the company would be reflected in policy price and thus would be internalized by the people or institutions who buy the policies. The externality should vanish or at least significantly diminish.

3. Why should insurance subsidies lead to “low copayments and deductibles”? Insurance with high copayments and deductibles is favored by the tax system as well. That being the case, why do we blame the tax system for how insurance is (perhaps) poorly structured?

All these points collapse into a more simple query: how can a simple relative price, whether a distortion or not, corrupt the cost control practices of an entire industry?

And if government provision of health care is ineffective and costly, isn’t there a positive externality from the purchase of private health insurance?

Many of the people who cite this argument about health insurance are smarter and more accomplished than I am. I will grant their greater wisdom and authority. But at the end of the day, I still don’t get it.

Becker on the FDA

In the latest Milken Institute Review, Nobel laureaute Gary Becker argues (sign up required) that the FDA should permit drugs to be sold once they have passed a safety standard, i.e. a return to the pre-1962 system. He writes:

…a return to a safety standard alone would lower costs and raise the number of therapeutic compounds available. In particular, this would include more drugs from small biotech firms that do not have the deep pockets to invest in extended efficacy trials. And the resulting increase in competition would mean lower prices – without the bureaucratic burden of price controls…

Elimination of the efficacy requirement would give patients, rather than the FDA, the ultimate responsibility of deciding which drugs to try…To be sure, some sick individuals would try ineffective treatments that would otherwise have been prevented from reaching market under present FDA regulations. But the quantity of reliable health information now available with only a little initiative is many times greater than when the efficacy standard was introduced four decades ago.

Dan Klein and I have written extensively on this issue at our web site,, and in our latest paper Do Off Label Drug Practices Argue Against FDA Efficacy Requirements?

Is American health care more productive?

Advocates of national health insurance point out that the U.S. spends more on health care, per capita, than any other country in the world. At the same time, Americans rank only in the middle when it comes to actual health and longevity. So you might believe that we could nationalize the industry, save money, and improve our health. Think again:

The proper way to measure the performance of health care is to measure the difference it makes in the quality of life of people who come for help…What we need to know is whether the higher level of spending means the United States is much less productive in health care than other countries.

In an attempt to test the limits of knowledge here, we studied the treatment of four diseases — diabetes, cholelithiasis (gallstones), breast cancer, and lung cancer — in three countries: Germany, the United Kingdom, and the United States. These three countries were the only countries for which comparable data existed for these diseases, either nationwide or for large regions.

The relevant measures were either life expectancy after treatment or measures of the quality of life. And how about the results:

The United States is more productive in all these diseases except for diabetes in the United Kingdom. [emphasis added] The reasons for this result can be traced directly to the huge differences in the way the health care sector is organized and governed across these three countries. The UK health care system is almost entirely government owned and run…The result has been that the United Kingdom has no invested as quickly in technologies that have dramatically improved the diagnostic capabilities of medicine and significantly reduced recovery time…Germany, on the other hand, has a system more like the United States had twenty years ago. In Germany, medical expenses are paid for on a task-by-task basis for services of doctors and hospitals. As a result, hospitals in Germany have no financial incentive to reduce length of stay.

In other words, Americans pay more but get better health care in return. We die sooner because we eat too much and exercise too little, among other facts. For similar results, see this comparison of the U.S. and Japan.

The quotations are from William Lewis’s interesting The Power of Productivity, see p.97. Lewis is a partner at McKinsey, an economics and management consulting firm. Here are other McKinsey writings on health care ($$), including the comparison with the UK and Germany.

By the way, this essay suggests that most of the productivity benefits of health care spring from pharmaceutical consumption. Of course we lead pharmaceutical production but also pay the highest prices. It would be a disaster for the world as a whole if we tried to save money on this front with tight price controls.

The bottom line: National health insurance is unlikely to save on medical costs, unless it cuts back on treatment drastically.


Mainstream media is finally getting the message about DDT. Tina Rosenberg, writing in the New York Times Magazine, puts it well:

DDT killed bald eagles because of its persistence in the environment. ‘Silent Spring’ is now killing African children because of its persistence in the public mind.

I’m delighted to see the Rosenberg piece, even if everyone I know has understood this issue for years.


For two centuries, the American man stood tall in the world. Literally. But today the average Dutch man is six foot one and the average American man is much shorter. Even as little as fifty years ago, American men were considerably taller than Dutch or other European men but since the mid 1950s the Northern Europeans have shot up while Americans have grown wider but not taller. No, it’s not a composition effect due to immigration. Native born, English-speaking American men are only five feet nine and a half and this has not changed much in more than a century. Why then the difference?

One possible clue is the enormous impact that nutrition can have on height. In Guatemala the native Mayan man averages only five foot two, so short compared to the Spanish-descended Ladinos that most people assumed the difference was genetic. But lo and behold when a million Guatemalans fled to the United States a natural experiment began – the children of the American Maya are four inches taller than the Guatemalan Maya of the same age and about as tall as the Ladinos. Good nutrition, especially in the growth years of infancy, 6-8 years and adolescence can increase height in remarkable fashion.

But if the problem is poor nutrition then surely the figures for the average American ought to be masking a growing drift in height between the well-fed rich and the poorly-fed poor? And yet that appears not to be the case. The mystery remains.

The discussion is based on Burkard Bilger’s excellent piece in the New Yorker. See also this interview with Bilger.

Adam Smith on the morality of financial compensation

I have a new article on the shortage of human organs. Regular readers will be familiar with my views so I won’t go into the details but I was especially pleased with the conclusion examining moral objections to organ selling.

Adam Smith noted that in his time there were “some very agreeable and beautiful talents of which the possession commands a certain sort of admiration; but of which the exercise for the sake of gain is considered, whether from reason or prejudice, as a sort of public prostitution.” As examples, Smith lists acting, opera singing and dancing. Today the list strikes us as peculiar, perhaps even foolish. What could possibly make opera singing admirable when done for free but despicable when done for pay?

And yet, however peculiar the views of Adam Smith’s contemporaries were on financial compensation for opera singers we can an at least say this in their favor–no one ever died because of a shortage of singers.

Addendum: I learned of the Smith quote from a wonderful paper on prostitution by Martha Nussbaum, Whether from Reason or Prejudice: Taking Money for Bodily Services. I can’t find the paper online but it is reprinted in Liberty for Women which has a number of other excellent papers.

Was Nietzsche right?

In recent weeks there’s been a furor in the Washington D.C. area over lead in the District’s water supply. Today, the Washington Post (registration required) looks at why lead is bad for you and covers some of the science and public policy. That lead is bad for you is open and shut. Too much lead kills you and for kids, too much is not that much. But I am skeptical of recent studies that find that the worst effects of lead happen at the lowest levels of exposure.

Here‘s a typical newspaper account of one of those studies and a quote from a leading researcher on the topic:

“There is no safe level of blood lead,” said Dr. Bruce Lanphear, lead author of the lead study presented Monday at the Pediatric Academic Societies annual meeting.

Edward Calabrese would not agree. Calabrese is a toxicologist at UMass-Amherst and a leading scholar of hormesis, the phenomenon that most if not all toxins are actually good for you at sufficiently low doses. This does not imply that you should start adding mercury to your eggs or lead back into your pots. But the impact of toxins appears to be U-shaped–good for you at sufficiently low levels then bad as exposure increases. Whatever doesn’t kill you makes you stronger.

Hormesis also implies that linear models or threshold models of toxic impact are misspecified and understate the impact of toxins over some ranges of exposure.

Here’s a Scientific American article on Calabrese and hormesis.

Here’s my take on the economics and policy implications of hormesis.

“Curing” Obesity

Researchers claim to have discovered one of the key causes of obese America. The AP reports:

Researchers say they’ve found more evidence of a link between a rapid rise in obesity and a corn product used to sweeten soft drinks and food since the 1970s.

The researchers examined consumption records from the U.S. Department of Agriculture for 1967-2000 and combined it with previous research and their own analyses.

The data showed an increase in the use of high-fructose corn sweeteners in the late 1970s and 1980s “coincidental with the epidemic of obesity,” said one of the researchers, Dr. George A. Bray, a longtime obesity scientist with Louisiana State University System’s Pennington Biomedical Research Center. He noted the research didn’t prove a definitive link.

I like the hedging–the link isn’t “definitive.” No, I guess it wouldn’t be. Obesity is surely also “linked” to the Iran hostage crisis and the stagflation of the late ’70s and early 80s. Maybe I shouldn’t be so skeptical. The study may be a little more scientific than merely looking at correlation rather than causation. But if the research is right, it will be easy to make America thin again. Just ban those high fructose sweeteners. One problem with this will be explaining why the advent of low calories sweeteners didn’t stem the tide of fat that allegedly threatens to overwhelm us.

My theory is that we’re fat because we enjoy it. We like food. It gives us pleasure. We’re wealthy and food’s cheap so we’re taking on a few pounds. Alex points out that the entire increase in weight over the past several decades can be explained by an extra Three Oreo Cookies a day! Here is a paper by Glaeser, Cutler and Shapiro that takes the economists’ approach to weight gain.

Here’s my take on the claim that we should tax fatty foods because of the externalities.

Off-label drugs

Alex has written much about the importance of off-label uses for drugs, which are not generally restricted by the FDA. I came across the following in the 17 March 2004 Wall Street Journal, Marketplace section:

A high-price biotech drug, developed in the 1980s to treat a rare form of hemophilia, is fast becoming a blockbuster, with physicians around the world using it to stanch severe bleeding from car accidents, gunshot wounds and postsurgery hemorrhaging.

But here’s the rub: The drug, a human bloodclotting protein called NovoSeven that costs $5,000 a dose, hasn’t been approved by the Food and Drug Administration for such uses…Some doctors are hailing NovoSeven as a lifesaver, with word spreading about near-miraculous cures of dying patients. It’s the new wonder drug,” says Thomas Scalea, director of the shock trauma center at the University of Maryland Medical Center. He says the center has used the drug 80 times in three years, saving about 35 lives.

Not all doctors agree about the merits of the drug, and insurance companies will not reimburse drug usage for this purpose.

Question: Should the FDA have regulatory power over the off-label use of this drug to stop traumatic bleeding? If it had had such power, would any of these lives have been saved at all? In his paper, Alex asks the requisite follow-up question. If it is bad idea to give the FDA regulatory power over off-label uses why give it so much power over initial uses?

As things stand now, the FDA will not allow “human trials” of the drug in the laboratory, although real life trials occur in hospitals on a regular basis. Furthermore the company’s sales force cannot promote the drug for non-hemophilia uses and must wait for doctors to ask, noting that it employs 25 full-time people simply to handle the flood of inquiries.

Further comment by Alex is welcome. This is his bailiwick not mine, I just thought the example was especially striking. Here is some basic information about the drug, here is the company’s web site.

Further thoughts on artificial hearts

A number of people emailed me or blogged (eg. here and here) on my post, Artificial heart won’t save lives. The number of transplants is constrained by the number of donated organs thus the main effect of the artificial heart, which is just a temporary stop-gap, is to redistribute organs. The artificial heart makes some people better off at the expense of other people who are made worse off. No one challenged this conclusion but it seemed to make some people uncomfortable. Two arguments were raised in opposition, both of which are weak.

First, the heart does allow some people to live a little bit longer – this is a benefit, but a few weeks of life while chained to a big machine doesn’t seem like a big breakthrough to me. Second, the artificial heart could allow for better matching. Theoretically true, but there are already many more patients on the waiting list than there are hearts available so the opportunity for better matching is negligible. Consider, that for a given heart there are now 3500 people on the waiting list to choose from – how much better is the match going to be if we add a few more people to this list?

I am not against artificial hearts (some people say I have one!) perhaps one day the technology will improve enough so that someone on an artificial heart can be taken off the list, but the issue is comparative. Suppose that we put the funds gong into artificial hearts into programs to increase organ donation. One donated organ is say good for 10 years of extra life. Average time on the artificial heat is 77 days and it is not clear how many of these days represent extra days of live. Let’s say very charitably that 50 days are extra then this means that one real heart is worth 73 times as much as an artificial heart (10*365/50) and that is before adjusting for quality of life.

Artificial heart won’t save lives

An FDA panel announced today that they would support approval of a new artificial heart. NPR and other media suggested that the new heart, which is designed only for temporary use and is not portable, would save lives by extending survival time until a transplant became available. But even if the artificial heart performs exactly as designed and even if it prolongs the lives of those who receive it, it won’t save lives overall.

The mathematics is simple; there are approximately 2200 hearts donated for transplant every year (data here). That means we can save 2200 lives a year and no more. All the artificial heart can do, therefore, is change who gets saved. Some people who previously died will live long enough to receive a transplant but this means there will be one less heart available for someone else on the waiting list. The artificial heart will make the waiting list longer but it will not save lives.

The only way we can truly save lives is to increase the number of organ donors. As readers of Marginal Revolution will know I have suggested financial compensation and organ donor clubs as the only realistic solutions.

Why has Russian health collapsed?

The health care collapse in the former Soviet Union is old news, less well known is the contrast between Russia and Poland:

…another Slavic nation with a traditional affection for vodka — Poland — is experiencing one of the greatest improvements in health ever known. The difference tells a story of how democracy has transformed the center of Europe in the past 15 years — and how it has failed in Russia.

Start with the figures. In the early 1980s life expectancies in Soviet Russia and Communist Poland were roughly similar, and both were starting to get worse. Cancer and cardiovascular disease were beginning a rapid rise, in lock step with their prime causes: smoking and alcoholism.

Two decades later, Poland’s life expectancy for men, at 70, has risen by four years since the collapse of communism and now is more than 10 years longer than that of Russian men. In Poland, cardiovascular disease has fallen by 20 percent in a decade, while in Russia, it has risen by 25 percent. Sudden deaths from accidents and other external causes have fallen 19 percent in Poland, while in Russia the rate has soared to an unprecedented level. Poland’s rate of HIV infection is one of the lowest in Europe; Russia has one of the world’s highest rates of new infection.

Polish researcher Witold Zatonski has an explanation for this difference:

…he has boiled his answer down to a simple slogan: “Democracy is healthier.” “It’s the only way to explain what has happened,” he said during a recent visit to Washington. “It turns out that the free-market economy and a free political debate correlate directly with good health in Eastern and Central Europe.”

I was incredulous when I first read this, apparently I was not the only one:

That conclusion used to be doubted by some of Zatonski’s colleagues, both in Poland and in the West. After all, democracy brought Poland freedom for cigarette and alcohol advertising, Western brands, and a parliament presumably susceptible to special interests. Tobacco companies spent $100 million a year on marketing to Poles in the 1990s.

Remarkably, though, all that money and influence have been outweighed by the other products of a free society, especially independent civic organizations and media that promote knowledge and open debate about health issues.

It is well known from happiness surveys (see the work of Bruno Frey) that people are happier in democracies, so maybe there is a link to health as well, even after adjusting for income.

What about the comparative statics?

In addition to Poland, the Czech Republic, Hungary and other newly democratic states have recorded dramatic gains in health. But Ukraine and Belarus, which have followed Russia’s political course of far more restricted freedom, have seen their health measures decline. The Baltic states of Latvia, Lithuania and Estonia, which were once republics of the Soviet Union, at first shared Russia’s downward spiral; but since 1995, as they have built Western-style democracy, they have reversed the trend and now follow Poland’s path.

Here is a comprehensive paper with data and cross-country comparisons. From this excellent piece I learned the following:

1. Health in Belarus has continued to decline. Since Belarus has stayed largely communist, the country may serve as a possible control for where Soviet health was headed before communism fell.

2. In Russia, many of the biggest negative health changes have come for the 18-34 group, not for the elderly.

3. In the CIS countries, injuries and violence account for a quarter of all deaths about men aged 25-64; this is six times higher than the death rate from Western Europe.

4. Homicide and suicide rates in CIS countries exceed those for the West by about 20 times.

5. In Russia, deaths from all external causes correlate closely with deaths from alcohol poisoning.

6. Men of low educational background have been to most vulnerable to bad health.

7. Russian life expectancy has declined but it actually improved during the 1994-8 period and has moved with economic crises.

Here is some additional information on mortality trends in Eastern Europe, here is another related link. Zatonski has a longer essay, which I do not currently have access to.

The bottom line: Here is what I used to think: “I blamed the Russian health collapse on the loss of relative status for the elderly. While market reforms have increased aggregate wealth, this has been mostly for the young. Older people have lost their grip on power, and suffered psychologically through their loss of international relative status as well. They lost the will to live and died early.”

Here is what I now think: “Russian young and middle-aged men have found few useful institutional supports during the transition. They’ve gone crazy with drinking and violence.” That being said, I don’t think we have sorted out the relative importance of economic and political factors.

Teenage pregnancy

Since 1991 the teenage pregnancy rate has fallen by about 22 percent, reversing a 40 year trend. In a lengthy story, the NYTimes suggests that learning from the hard experience of others is the explanation for the drop without explaining why it should take 40 years for this learning to take effect. They do note “teenage pregnancy had already begun its decline in 1991, well before welfare changes and the economic boom, and well after the first round of sex education programs.” The Times, however, does not examine the most controversial but well-supported explanation, the introduction of legalized abortion in the 1970s.

If this explanation rings familiar it should. In a very controversial paper, Steve Levitt and John Donohue provided evidence that legalized abortion in the 1970s reduced crime some 18 years later. The theory is simple. Abortion rates are higher among the poor, the unmarried, teenagers, and African Americans than among other groups and children born to mothers with several of the preceeding characteristics are at increased risk for becoming involved in crime. Legalized abortion gave these mothers an option and thus reduced the number of at-risk children who might otherwise have grown up to become criminals (note that abortion doesn’t mean fewer children per-se, it may simply delay childbearing to when the mother is not poor, a teenager or unmarried which works just as well.)

In brief, the evidence for the Levitt-Donohue theory is a) the timing is consistent, b) states that legalized earlier had earlier drops in crimes, c) there is a dose-response effect i.e. states that had more abortions had bigger drops in crime, d) the drop in crime in the 1990s occured among those cohorts who were potentially affected by abortion policy in the 1970s (and not among say 40 years olds.)

Joined by co-author Jeff Grogger, Levitt and Donohue apply the same idea to teenage pregnancy and find very similar results – thus reinforcing their earlier story. They write:

Parents who are least able or willing to begin caring for a newborn are most likely to make use of abortion. The abortion rates for teens, the unmarried, and the poor are substantially higher than for the general population. Children who are born unwanted are subjected to poorer care both during pregnancy and the early years of life. With the legalization of abortion, mothers with unwanted pregnancies suddenly had a new recourse. Consequently, the number of children raised in adverse environments dropped substantially. Donohue and Levitt [2001] showed how this change reduced crime among the subsequent generation by 15-25 percent. As teen childbearing is a closely associate social pathogen, the magnitude of the drop should be similar.

Our empirical evidence suggests that birth rates as teens are strongly negatively associated with being born in a state and time period in which abortion rates were high. Our results suggest that teen birth rates today may be 20 percent lower as a consequence of legalized abortion in the 1970’s.

Let’s spend more on health care?

That’s what David Cutler argues in his recent book Your Money or Your Life.

Cutler believes that our expenditures on health care have more than justified their cost. He therefore opposes the traditional recipe of “cut costs and use the savings to finance greater access.” His attitude is closer to “expand care now and improve the quality of outcomes.” If you think that more discretionary spending doesn’t make many people much happier, why not make them healthier and longer-lived instead?

As I read the book, Cutler is pushing two major ideas:

1. Subsidize insurance to ease the problems of the forty million uninsured. But he repeats the usual numbers, without convincing me that the problem is as bad as it sounds.

2. Pay for health care results, rather than rewarding expenditures per se. In other words, give doctors and hospitals bonuses for actually making patients better.

A loyal reader of MR should not be surprised to read that Robin Hanson had the idea first. Read his intriguing essay at the link. In Robin’s vision you buy your medical care from an institution that contracts with a third party to pay penalties, or receive bonuses, depending on your longevity, disability, et.c — whatever can be measured. I can imagine such incentive schemes working in the decentralized private sector, especially after much trial and error experimentation. (Note the potential adverse selection problem: you don’t want providers to have an incentive to shun hard-to-improve cases.) It is much harder to see federal or even state governments getting the incentives right, and having the political capital to see the correct decisions through.

The bottom line? Cutler is obviously a smart guy but overall I found the book underargued. I like his optimistic, can-do attitude, but I don’t trust it in the hands of politicians.