able to them at actuarially fair premia. (Actuarially fair premia means that one pays roughly what the average expenditure on a family would be.) Clearly, the fact that the income distribution is such that families cannot afford health insurance does not indicate a failure on the part of the health insurance industry, and the problem could certainly be solved through income redistribution.

There is another problem that can be more properly traced to the health insurance industry in this country. I propose on page 3 of my formal statement that a useful exercise for this committee might be to go through a typical health insurance policy of the sort now being marketed in this country, to imagine a variety of different illness scenarios and then to determine, without expert advice, precisely what the coverage is under the policy. I would suggest that you will find this exercise to be excruciating, as do many American citizens today. The health insurance industry will undobutedly defend itself on the argument that the embroidering of its policies with exclusions and provisos is designed to accommodate a desire for efficiency, and I accept that argument. I do, however, also argue that whatever efficiencies you purchase in this manner are likely to be purchased at a very high price. First, there is the real possibility that many Americans purchase totally inappropriate health insurance coverage on highly mistaken beliefs about the maximum risks to which they are exposed. You will, without any doubt, be exposed to a famous experiment at the University of Pennsylvania where a good health insurance package had been designed that provided for coinsurance and deductibles, and yet presumably highly educated faculty members bought first dollar coverage. This is often cited as evidence of irrationality. I believe that it was not evidence of that. It was evidence of ignorance, I suspect, and ignorance is quite an excusable trait even among professors. I suggest that these faculty members did not know the maximum risks to which they were actually exposed and were motivated to insure themselves for much higher imagined risks than those to which they were truly exposed.

Therefore, I have, in the fattest letters I could find on the typewriter, recommended in my statement that:

Whatever Congress does after its deliberations in the area of health insurance, it is to be hoped that it will present to the American people health insurance options or policies that are readily understood by the average person, that explains to the average person the maximum liability to which he or she is exposed, even if such an approach involves some inefficiencies.

Many experts to appear before this committee will argue for coinsurance and deductibles, or what is generically referred to as "cost sharing" on the part of patients. The hypothesis here is that it is the patient who is central to the health care consumption decision and that, if the patient shares in the costs of these decisions, these decisions will be made rationally and efficiently.

I cannot, at this stage, launch into this controversial subject matter. I would not agree with some of my colleagues-and clearly an economist should not agree with the proposition-that the decisions in health care consumption are made only by physicians. I do believe that coinsurance would reduce the consumption of health care in this Nation. However, in deciding whether or not to introduce coinsurance

features in this Nation's health insurance legislation, one must be aware of a number of points that really don't pop out of our microeconomic textbooks.

The first point is this. The objectives being pursued with cost sharing are not likely to be reached in this country, if only a narrow stratum of the population-presumably lower income families-are exposed to cost sharing, while upper income and middle income families and particularly the members of trade unions, manage to avail themselves of first dollar coverage. In this case, perhaps only 10 percent of the American people face coinsurance, and the effort to economize over the whole system is focused on that narrow stratum of the population. Such a system would not only be unfair, but it would not work. Congress might toy with the idea of prohibiting first dollar coverage altogether. I can't imagine how you could do that in the face of the strong objections from the interested parties you are likely to encounter, so I won't dwell on this point.

Second, a system of coinsurance and deductible is likely to be worth its cost only if it can be relatively easily administered. One could certainly design a system of coinsurance and deductibles that would eliminate unemployment in this country altogether simply by creating a large bureaucracy to administer the system. In fact, such system may not be worth its cost.

Third, as already mentioned, a system of coinsurance and deductibles is likely to be self-defeating if it is so complex as to generate enormous psychic costs of uncertainty on the part of the insured. It is generally assumed by economists and everyone else that people suffer from uncertainty, and those psychic costs are very real even if they are not expressed in monetary terms. So, I would like to suggest some caution on a system of coinsurance and deductibles, although as an economist I certainly believe they could be used to reduce health care consumption and hence health care costs.

I would finally like to offer some observations on the assertion that there is nothing amiss in the American health care system.

Later on in my formal remarks, on pages 11 through 13, I list a veritable catalog of complaints that have been lodged against the American health system at one point or another. The list of these complaints is long and varied. On closer examination, many of these complaints are based on value judgments and peculiar interpretation of available data. This being so, some observers of the American health system have sought to reject the current criticism of that system altogether by insisting that ours is the finest health care system in the world.

One point to note here, of course, is that, even if one has the finest health care system in the world, one cannot necessarily argue that one ought not to make such improvements which are feasible. It is also within the domain of public policy to speed those improvements.

But the people who defend the American system may indeed have a point, and I wish to speak to that, because there will again be many people who will come before you who consider ours to be the worst health-care system in the world. We have humbled ourselves considerably before the rest of the world by pointing to infant mortality rates and maternal death rates, on which statistics we rank roughly

11th or 14th. It used to be 18th. It is often thought that the health care system we have in the absence of health insurance is to blame for this. An interesting case in point is the health system of West Germany. I have left with the committee staff a paper describing that system in detail. The residents of West Germany have had universal comprehensive health insurance without coinsurance or cost sharing of any sort for many, many decades; in fact it was introduced in its early phases in 1887. The health care delivery system in West Germany is roughly similar to ours, although physicians in hospitals receive salaries, a method of payment some would like to have in this country. Remarkably, West Germany fares considerably worse on such widely used health status indicators as infant mortality rates and infant death rates.

As a matter of fact, there are some tables before you, I believe, where you will see that the maternal death rates and infant mortality rates in West Germany, which is a fairly homogeneous country, are as shockingly high as those rates now found among nonwhites in this country. I raise this point not to assert that the West German health care system has failed society or to imply that no improvements are available in our system because we are already very much at the top. I wish simply to raise the caveat that health services are not the only input into the production of a nation's health status, and that the introduction of health insurance in this country, a redistribution of access to health services, or even a drastic reorganization of the American health care system may not trigger the improvements we expect from those reforms or may induce changes which we had never anticipated and which we may not like in the end.

In my formal statement I go at length into the second objective, but Professor Fein has already commented on it, and I won't dwell on it. I would just like to offer briefly some remarks on the possibility of using the National Health Insurance System as a lever to alter the health delivery system, that is, on the possible objective.

Two changes are often proposed for the American system. First, one source of evils is said to be the fee for services system. A second source of problems is said to be the fragmentation of the system. For that reason it is often supposed that we ought to move from fee for service to prepayment, and on the delivery side from small solo practices to large group practices.

I would assert that no social scientist with integrity could now come before you and tell you unambiguously

Mr. ROSTENKOWSKI. Professor Reinhardt, we are going to have to suspend. The committee will recess for 10 minutes. We will return to hear your conclusion, and then the interchange among panel members, and then participation by the members as well.

It is the intention of the Chair to work from the next recess until 12:45 and to recess then from 12:45 until 12 o'clock for lunch and to return at 2 o'clock for further discussions.

[Short recess.]

Mr. ROSTENKOWSKI. The committee will resume its sitting.

Professor Reinhardt, if you will make your conclusion, we will open the panel for comments by other panel members.

Mr. REINHARDT. Thank you, Mr. Chairman. I would just like to summarize briefly the remarks I made.

The first point is, as I mentioned, that the health insurance system. in the United States should be designed to be primarily just that, and one of its goals should be to free the American citizen from the psychic costs of uncertainty.

I could not stress that enough because I see that point so rarely made and yet I could not think of anything more important than this feature. In fact, the Canadians have simply and quite boldly eliminated this type of uncertainty. They have introduced a comprehensive health insurance system that leaves the delivery system more or less alone, but gives the Canadian citizen complete freedom from worry about the financial impact of illness. The Canadian innovations certainly deserve consideration by reformers in this country.

In creating such a system in the United States, however, one might I admit-foster a mushrooming medical sector that eats up ever-increasing proportions of the gross national product.

I would endorse Dr. Fein's comment that there has to be in such a system a bottom line, there has to be an overall budget within which one must make tradeoffs. It is my opinion, and I would advertise it as such, that in this country we will not get away from fee-for-service reimbursement of physicians.

This being so, I would deem extremely necessary in the design of a national health insurance system renewed control by the public sector over the fee schedules that will be used. I think nothing could be more disastrous to the evolution of the American health care system than to perpetuate the system of customary local fees that we now, by necessity, have to adopt in medicare, part B.

I say by necessity because medicare is only a small part of the overall health delivery system. Once the entire system comes under the control of a more unified insurance mechanism, however, the public sector can gain control over the fee schedule negotiations with the American profession as a whole.

A publicly controlled fee schedule could be used to attack one of the major problems the American health system does have, and that is the maldistribution of medical manpower over specialties and over regions of this country. We now pay physicians who prefer-for whatever reasons-Cambridge, Mass., to other parts of this country, two to three times as much to practice in Cambridge, Mass., as we pay them to practice in other parts of the country. It does not surprise me, perhaps because I am an economist, that this generates an unequal distribution of manpower or at least that it ratifies it.

I would suggest that fee schedules could be used to redress these incentives, to make them, at the very least, uniform across the country, as the Canadians have done.

You might be bolder still. You might attempt to establish a system like that in West Germany wherein exchange for gaining a monopoly over the delivery of primary care, the medical profession must assume responsibility for making health care available where and when needed.

We have not done that in this country. I don't believe organized medicine in this country views it as its mandate to make sure that health services are available to all Americans when and where needed. Other countries require their legally monopolistic medical professions to foster an adequate distribution of medical services. In West

Germany fee schedules are used to entice physicians into the so-called cultural hinterlands, into the rural areas. Doctors are paid more for a given service in a rural area than in a city. I think this is an interesting idea. I think we ought to study the effects of that system. The third point I wanted to make, is that we shy away at this time from attempts to regulate and to reshape through forceful direct regulation the American health care system, for the following reasons:

First, I know of few instances where direct regulation of economic activity really has worked directly and ultimately to the advantage of the American consumer. I need to mention here only transportation which should trigger in your mind an entire catalog of side effects we never intended.

Second, I don't believe that we know precisely where we wish to go in connection with the health care sector. One interesting exercise that you might want to undertake is the following: When somebody asserts that there is a maldistribution of medical services in this country as we all agree ask that person to tell you what he or she would deem an optimal distribution to be? In other words, "Can you tell me how many physicians I should have in each specialty in each country in the United States?" Such detailed instructions would be required if the Government were indeed to attempt to regulate directly, as it tried to do in some legislation 2 years ago.

I could go on and analyze changes in the delivery system that are felt to be panaceas; however, when you question social scientists who have evaluated such proposed changes they cannot in all honesty tell you that these proposed changes are unambiguously advantageous. Finally, I make a plea that in designing National Health Insurance for the United States we look abroad to other nations who have led in that respect: Canada, France, Sweden, England, and West Germany, who have tried many of the changes that are being proposed as panaceas for this country who have sometimes succeeded but who have often failed.

One of the remarkable features which one observes in international comparisons of health systems is that, regardless of what proportion of the GNP is absorbed, the medical profession invariably controls the health care system, and shapes it according to its preferences. We have a faith in this country that if only we tinker a little bit more with the system, if only, for example, we pay salaries to our physicians in this country, we will solve the major problems facing us. I would urge you to look abroad, to invite perhaps foreign experts to come and deliberate with you, and to learn some of the bitter lessons which foreign nations have learned in this respect.

Thank you very much.

[The prepared statement follows:]

STATEMENT OF U. E. REINHARDT, ASSOCIATE PROFESSOR OF ECONOMICS AND PUBLIC AFFAIRS, WOODROW WILSON SCHOOL OF PUBLIC AND INTERNATIONAL AFFAIRS AND DEPARTMENT OF ECONOMICS, PRINCETON UNIVERSITY, PRINCETON, N.J. Mr. Chairman, it is a privilege and a pleasure to join with you, the members of your Committee, and this panel in a discussion of issues surrounding the introduction of national health insurance in the United States. By its very nature, such a system will alter the financial flows accompanying the delivery of health services and thereby the economic incentives and constraints confronting the consumers and the providers of health services. One need not be a card-carrying