TCS Daily

Hating the Producers

By Arnold Kling - September 9, 2004 12:00 AM

"the United States spends more on health care than any of the other OECD countries spend, without providing more services than the other countries do. This suggests that the difference in spending is mostly attributable to higher prices of goods and services."
-- Gerard F. Anderson, Uwe E. Reinhardt, Peter S. Hussey, and Varduhi Petrosyan

Some people believe that the United States mis-allocates health care spending, meaning that we make poor choices of where to spend our health care dollars. In an earlier article, Winning The Health Care Olympics, I pointed out what is unpersuasive about that argument.

In this essay, I look at another complaint about U.S. health care, which is that we pay too much for the services that we get. Princeton economist Uwe Reinhardt, the marquee co-author of the paper quoted above, is a prominent advocate of the thesis that suppliers are overpaid. His views on the nature of the health care problem influenced the infamous Hillarycare plan, and his work continues to be the basis for Democratic Party health care policy proposals.

Left and Right

I think that the following generalization holds true for left-wing vs. right-wing economists:

Right-wing economists tend to emphasize the benefits of private producers and the harms of government intervention. Left-wing economists do the opposite.

For example, recently left-wing economist Jeff Madrick argued that Wal-Mart causes harm by hiring workers at low wages. He suggested government solutions, such as raising the minimum wage and changing laws to make it easier for labor unions to organize.

For non-economists, hating producers like Wal-Mart is easy. However, it gets a lot trickier once you understand some economics. It is difficult, although not impossible, to use economic analysis to blame Wal-Mart for low wages.

My own thinking is that we should be happy with Wal-Mart, not only for lowering prices for consumers, but for finding employment for low-skilled workers. If those workers are being paid according to the value of their output, then artificially raising their wages will cause them to lose their jobs. On the other hand, if they are not being paid as much as the value of their output, then what they need are other employers willing to hire them. I would say that what they need are more Wal-Marts.

The Mongoose and the Snake

Uwe Reinhardt's view is that unspecified health care suppliers are charging too much for medical services. He believes that prices in the medical industry are determined by negotiating power. He says that in the United States, consumers and insurance companies are too fragmented relative to service providers. As a result, health care supplier incomes are too high, and services cost too much.

If fees are too high, then why not use price controls? In fact, government does set many prices, particularly for services covered by Medicare. However, Reinhardt and other economists are loathe to suggest that government knows the right price for each medical service.

Instead, Reinhardt would advocate shifting the balance of power in the health care industry. Think of overpayments to monopolistic doctors as a snake. Reinhardt says that we need to bring in a mongoose in order to kill the snake. Reinhardt tends to shy away from describing the mongoose. Under Hillarycare, a byzantine scheme of health care purchasing co-operatives was going to be the mongoose. Today, left-wing health care pundits seem to favor a single-payer system, meaning nationalized health insurance, as the mongoose.

Suppose that the problem in health care costs is that hospital workers are paid a premium over what they could earn in other industries. This might be because of strong unions among hospital workers. Perhaps hospital costs would be lower if hospitals behaved more like Wal-Mart. But something tells me that Wal-Mart is not the mongoose that Reinhardt and his friends in the Democratic Party had in mind.

In my view, there is reason for skepticism that government can curb excess payments to suppliers. Government has not exactly brought the teachers' unions to heel, for example. A government regulator winds up being more like a lapdog than a mongoose.

Indirect Evidence

After reviewing Reinhardt's research, I am not persuaded that price gouging by health care suppliers is the reason that total health care spending is high in the United States. The evidence that Reinhardt provides is indirect and questionable.

The paper quoted above reports that in 2000, we in the United States spent 13.0 percent of our GDP on health care, the next highest OECD country spent only 10.7 percent, and the median OECD country spent 8.0 percent. Of the 5 percentage point difference (13 minus 8) between the share of GDP that the United States spends on health care and the share that the median OECD country spends, how much represents a difference in the quantity of services and how much represents a difference in the cost of services? (Note: more recent data appear to be broadly consistent with the data in the paper.)

Reinhardt and his colleagues arrive at a striking conclusion: none of the higher spending in the United States reflects more services! The basis of this claim is somewhat weak, however. In health care, it is difficult to measure output. In theory, you want to measure the impact that health care has on people's lives, but in practice this is difficult to quantify. The things that are quantifiable -- such as the number of x-rays taken, the number of prescriptions written, or the number of surgeries performed -- are only weak approximations of the "true" output of the health care industry.

The Reinhardt paper tries to use different sorts of approximations to measure health care output. For example, they compare across countries the number of physician visits and the number of hospital admissions per capita, and they find that the United States does not rank especially high on these indicators. However, I would say that those are measures of health care inputs, not outputs. An input is what you use to make something, and an output is what you make. In the automobile industry, you would measure output as the number of cars, not as the number of assembly-line workers.

Using inputs as a proxy for outputs is quite problematic, because of productivity differences. If you measured agricultural output by counting the number of farmers, then you would never guess that the United States is such a large food producer relative to other countries.

More generally, I am concerned that Reinhardt's approach risks running afoul of the fallacy of indirect inference. Indirect inference means drawing a conclusion about something you have not observed based on what you have observed. For example, if you and your roommate are the only ones home, the cookie jar is empty, and you did not eat the last cookie, then you would infer that your roommate took the last cookie.

In statistical economic analysis, indirect inference is much more dangerous. For example, suppose that you find that differences in years of schooling explain only 10 percent of differences in income. You might infer that the all of the other differences -- 90 percent -- must be due to innate ability. That would be an indirect inference, and it would be wrong. To give just one example of how you could be mistaken, it might be that a better measure of schooling -- which takes into account quality, for example -- explains 80 percent of the differences in income. (I'm not suggesting that 80 percent is the true figure, only that it is theoretically possible.)

Diagnosing high prices as the cause of high spending on the basis of imperfect measures of the quantity of services is an example of indirect inference. Such an inference is quite tenuous. In this context, it might be considered statistical malpractice.

The Direct Approach

The direct approach to answering the question about whether there is price gouging in medical care would be to look at some actual prices. The theory is that suppliers are earning economic "rents," meaning wages or profits higher than what would be found in a reasonably competitive market. Why not look for those rents?

For example, many people complain that profits are "too high" at pharmaceutical companies, and they point out that other countries have negotiated lower drug prices. Suppose that Marcia Angell, the author of a recent book attacking the pharmaceutical industry, got her wish, and pharmaceutical companies had their profits regulated. Could we close the 5 percentage point difference with other countries in our share of GDP devoted to health care? In her recent article, Angell reports that in 2002 the top ten pharmaceutical firms earned $35.9 billion in profits. That represents less than four-tenths of one percent of that year's GDP of $10 trillion. If you eliminated all profits in the pharmaceutical industry and nothing else changed, then our share of health care spending in GDP would only decline from 13.0 percent to 12.6 percent. Eliminating only "excess" profits would make an even smaller difference.

The same small-potatoes argument applies to putting a ceiling on settlements for medical liability lawsuits. Again, the numbers are low relative to GDP. I am strongly in favor of getting rid of medical liability altogether. Instead, I would like to see a random sample of each physician's patient interactions peer-reviewed by an independent board. This sort of audit-based system of physician accountability would be more effective in the sense of picking the right doctors to punish, but I doubt that it would bring overall medical costs down. So for the purposes here, I will give the trial lawyers a pass.

You could make a much bigger dent in the 5 percent gap by going after doctors. On average, physicians earn higher salaries relative to the median income in the United States than is the case in other countries. Perhaps we ought to think about lowering doctor incomes here to the same percentage of the median that doctors earn in the typical OECD country.

In their defense, however, the doctors might want us to make some adjustments. We might want to control for the type of specialty -- it is not fair to lower the salary of a brain surgeon in the U.S. because of how it compares to that of a general practitioner in Finland. Also, we would want to make adjustments for differences in hours worked per year. Finally, we might want to adjust for differences in quality of training -- perhaps you would be willing to pay more to see a doctor trained at Harvard than one trained in Tijuana.

Once we find the "just wage" for an American doctor and set that as a ceiling by regulatory fiat, what will happen to the supply of doctors? Reinhardt believes that it will remain unchanged, because he is convinced that high physician salaries here represent "excess rents." In technical terms, he believes that the supply of physicians is perfectly inelastic with respect to their salaries, meaning that they will work just as hard no matter how much they get paid.

I would like to see evidence that doctors will continue to work at lower incomes. If that were true, then it seems to me that malpractice insurance cost increases would not drive physicians out of business. But anecdotal evidence suggests otherwise.

Beggar Thy Neighbor?

Finally, there are the insurance companies to go after. For many on the left, single-payer health insurance (meaning government as a health insurance monopoly) is the solution to every problem -- what I referred to in Winning The Health Care Olympics as the hammer in search of a nail.

There are three arguments for putting government in charge of health insurance. The first argument, given by Reinhardt, is that the insurance companies are too fragmented to negotiate effectively with health care suppliers. In this view, it is not that private health insurance per se is inefficient, but that it fails to act as a mongoose to kill the snake of excess fees from health care service providers.

The second argument for single-payer health insurance is that it would reduce overhead expense. However, in every other industry we think that efficiency is promoted by competition rather than monopoly. It is not clear why competition should lead to excess overhead in insurance.

To explain why health insurance is exceptional, Paul Krugman offers,

"John Kerry's economic advisers have a very different analysis: they believe that health costs are too high because private insurance companies have excessive overhead, mainly because they are trying to avoid covering high-risk patients."

If one insurance company succeeds in weeding out cancer and diabetes risks from its portfolio, that only means that someone else has to carry those risks. That is the "beggar-thy-neighbor" aspect of risk screening. Risk screening saves costs for the individual insurance company, but not for the economy as a whole.

Arguing that excessive risk screening accounts for the high cost of health insurance suggests to me that insurance companies are hiring huge armies of economists and actuaries to build statistical classification systems to defend themselves from adverse selection by their competitors. However, if Kerry's economic advisers and Krugman would go out into the real world and look at actual staffing and resource allocation within health insurance companies, I am fairly sure that they would find that the amount of effort that health insurance companies put into risk screening is small relative to that which they put into claims processing and other forms of administration.

The final argument against a "fragmented" (i.e., private) health insurance market is that it helps keep the health care sector frustratingly paper-intensive and inefficient. However, even fragmented industries can become efficient through standardization, as bar codes and ATM networks illustrate. You can hang that picture without using the single-payer health insurance hammer.

What is the Problem?

Because Reinhardt keeps coming back with the same definition of the problem, Paul Krugman and other Democratic policy mavens keep looking to forced reductions in provider payments as the solution. But if a problem does not exist in the first place, then a "solution" is going to make things worse. And the evidence for price gouging by health care producers is not compelling. Hating the producers will not solve the health care problem.

The right-wing economists' view of the health care problem tends to focus more on the harms caused by government. That will be the subject of a subsequent essay.


TCS Daily Archives