the expense of overlooking possible over-all improvements in medical and pharmaceutical services. The view also was expressed that while the formulary system may help reduce costs, it should not be so restrictive that an individual patient is penalized by paying more for a drug not included in the formulary. In the panel on the cost and impact of third-party payment, the first speaker reviewed the advantages and disadvantages of the third-party system of payment in this country and suggested that there may be a strong case for the need for individuals to participate with their own funds under health insurance that is, to pay part of the cost. This can be an effective means of checking and controlling the demand for health services, he said, adding that there may be much to be said for reimbursement of hospitals and physicians on a charge basis.

The present method of third party payment embodies serious misconceptions of the nature of medical care and its organization, another speaker insisted. He emphasized that you can't talk about cost of care without talking of the quality of care and defining care. The central issues of cost in health care are issues having to do with the nature of organization of medical practice.

We started out in health insurance with a "casualty insurance" approach, dealing with medical catastrophe, providing coverage of hospital care and surgical procedures, which tend to be well defined, "one shot" services. In this view, comprehensive care has to do with comforting and strengthening the patient, and the whole present system of third party payments is inconsistent with this type of care. Physicians can be much better remunerated for comprehensive care on a retainer or capitation basis than by separate fees for each service, it was proposed. In the discussion, a panelist from industry thought the catastrophe element in surgical expense had been over-emphasized. The system grew up to meet health care expenses, and hospital and surgical costs were the most important ones, he pointed out. He liked the idea of a yearly fee or retainer which would be paid by the individual. We need control in prepayment, which is why we have fee schedules and have limited insurance in large part to hospital care and surgery— services that people generally don't like to have. One fault with our prepayment system is that it is organized around employment and leaves out those not regularly employed.

Unions don't like deductibles and co-insurance, it developed, and they have been mainly responsible for getting health insurance on a "first dollar" basis. Several speakers made the point that it would not be possible or desirable to lower the total cost of or expenditures for health care, and that possibly we aren't spending enough. The job is to get better value for what we do spend, it was suggested.

The panel generally agreed on the desirability of alternative approaches and flexibility in prepayment programs. The methods of compensating doctors and hospitals might have to be radically altered if we want emphasis on continued. comprehensive, nonepisodic medical care, it was pointed out.

One speaker deplored the sale of insurance policies that don't give adequate coverage of hospital costs. This was explained by another panelist who said that some people didn't have the means to buy adequate coverage. Still another panelist proposed that government should help to provide good coverage for everybody.

A number of speakers expressed the need for more effective exercise of cost control responsibility by the prepayment insurance mechanism, and the need for controls on the insured consumer, including such controls as co-payment or co-insurance and deductible insurance. Others insisted that deductibles and co-pay features "put the patient back in the practice of medicine" and raise economic barriers to health care that will work hardships on many families. Discussion was far-ranging. Among the points made for specific action were the following:

1. Half the states have legislation which prevent the development of group practice prepayment plans. The States should eliminate these restrictions.

2. Title 18 should be changed so that group practice plans could be paid on a combined basis for hospital and medical benefits. Under Title 19 the States should be encouraged to provide comprehensive benefits through group practice. 3. Either the legislation for Titles 18 or 19 should be changed or the administrative agencies should take action to permit more flexibility in carrying out these programs. Several panel members urged that the proposal in the Gorham report for development of a center for health services research and development should be implemented immediately.

4. The medical profession should be drawn into management and labor groups to consider problems at the local and national levels.

5. The Government should set minimum standards for what carriers do; there should be more Government regulation of carriers and more visibility of their actions to the public, perhaps through an SEC-type agency.

6. A proposal was made that the Federal Government should establish a coordinating advisory group or council with representation of all parties to help plan further action as a follow-up to the Conference and that the President should recommend to the governors of each State that they should establish similar groups.

7. One speaker said we accept the role of Government in providing health care for the aged but have given little attention to the role of Government in providing care to the whole population. He thought the present prepayment arrangements are completely inadequate to provide comprehensive health care for the whole population and we must think of a much wider role for the Government in providing such care.

8. There seemed to be a concensus on the urgent need for the encouragement of flexibility and experimentation. This requires action aimed toward this objective both by Government at all levels and private parties.

Now it is not strictly within the terms of my assignment to add any gratuitous observations to what has been reported from the panelists, but since there is ample precedent today-here and elsewhere for the violation of boundaries, I am going to add an observation of my own anyway.

Unlike scientific medicine, organization of health service is a field that more than many others has been dedicated unflinchingly to the rediscovery of the wheel. Many of the concepts that have been discussed here such as incentives, and productivity, and group practice, and community health systems, and utilization controls-these and other organizational and procedural goals and methods have been lying around in plain view, some of them for 20 years or more, now and again being re-invented, and re-explained, and re-exclaimed-over as succeeding generations of physicians, administrators, economists, social scientists and journalists have entered or turned their attention to the health field.

As one who has observed and on occasion taken part in these periodic rites, I feel constrained to report two things that have seemed to me to distinguish this gathering: First, of course, it is taking place in a setting characterized by rising public expectations for health care and rising public demands for efficiency that are obviously not going to be put down by conversation, promises and timid experimentation, at a time when innovative legislation has provided at least some of the means for grasping the opportunities for improvement that have been examined here without being overwhelmed by the risks and difficulties that are unquestionably involved.

Second, this conference has concerned itself not just with restatements of problems and theoretical constructions of goals but notably with many recommendations for method, action and in some cases assignment of responsibility. That's what you came to do.

G-"THE BASIC FORCES INFLUENCING COSTS OF MEDICAL CARE"

A STATEMENT BY VICTOR R. FUCHS, PH. D., ASSOCIATE DIRECTOR OF RESEARCH, NATIONAL BUREAU OF ECONOMIC RESEARCH, AT NATIONAL CONFERENCE ON MEDICAL COSTS, JUNE 27-28, 1967, WASHINGTON HILTON HOTEL, WASHINGTON, D.C.

It is both gratifying and challenging to be invited to address such a distinguished assembly of health experts. Until quite recently, an economist was rarely to be found in the company of the nation's leading physicians, and on those few occasions, he was likely to be flat on his back with one or more of his vital organs exposed to public view.

It is my intention this morning to provide exposure of a different sort. The assigned question-"The Basic Forces Influencing Costs of Medical Care"-is one which almost every man and woman in this room would be prepared to tackle. My aim is to indicate how an economist goes about answering it. This is not just another way of saying that I will give you my opinions-I will do that also but it is an attempt to take explicit note that economics is, above all else. a way of looking at questions. In Lord Keynes' words, "The Theory of Economics does not furnish a body of settled conclusions immediately applicable to policy.

It is a method rather than a doctrine, an apparatus of the mind, a technique of thinking, which helps its possessor to draw correct conclusions."

To be sure, even among economists there is not always just one way of looking at things. Winston Churchill used to complain that whenever he asked Britain's three leading economists a question, he received four different answers-two from John Maynard Keynes. Nevertheless, there is a common fund of concepts, a common core of analysis, that nearly all economists use. When, in the course of applying these concepts, my own value judgments or empirical estimates appear, they will be appropriately labeled.

The basic analytical approach is a consideration of those factors affecting the demand for medical care, and those affecting the supply. Demand and supply, the two magic words. Some of us, when visiting hospitals, have discovered that by putting on a white coat and talking rudely to nurses, it is easy to pass as a physician. To be mistaken for an economist is often even simpler. All one needs to do is nod gravely and say “demand and supply.”

Definition of terms

Demand for and supply of what? I shall assume that medical care refers to the services rendered by physicians, dentists, and other health professionals, plus all the goods and services consumed in connection with their work, or upon their direction. Thus, the costs of medical care include the costs of hospitals, drugs, and the like. This lumping of diverse health services is a concession to convention and to the limitation of time. Ideally one should apply the demandsupply analysis separately to hospitals, dentists, drugs, and so on because the forces that influence the cost of one type of health service are often different from those that influence another.

What is meant by costs? At least three possible meanings can be distinguished. It could mean price, or cost of production, or expenditures. When people speak of the rising costs of medical care, they frequently are referring to rising expenditures, and this is the way I shall use the term this morning. Expenditure trends

We all know that these expenditures have been growing rapidly. In round numbers, expenditures for medical care have risen from under $4 billion in 1929 to over $40 billion in 1965 and probably close to $50 billion in 1967. Even as recently as twenty years ago, expenditures were only $10 billion. Of course, expenditures for most other goods and services have also risen; it is therefore more meaningful for some purposes to look at the share of total spending allocated to medical care. This too has risen, from under 4 per cent in 1929 to about 6 per cent in recent years. Nearly all of this relative increase has occurred since 1947. Before examining the factors responsible for this trend, it is worth noting that there is nothing wrong a priori with changes in industry and sector shares of gross national product. Indeed, such changes seem to be a natural concomitant of economic growth. For instance, the relative importance of agriculture has declined precipitously in most western countries. During the last half of the nineteenth, and the first half of the twentieth century, there was a significant rise in the relative importance of manufacturing. Now we are witnessing in this country the growth of what I have described elsewhere as the "first service economy." If agriculture's share of GNP falls from over 9 per cent to under 4 per cent, as it did in the United States between 1947 and 1965, some other industries must show increases. There is no magic in the 4 per cent figure for medical care; it is now 6 and it could be 8 or 10.

Reasons for concern about costs

Why then should there be a national conference on the costs of medical care? Let me suggest three reasons for concern.

First, questions arise in my mind concerning the contribution that these increased expenditures make to health. Although we spend much more per person for medical care than any other country, the blunt truth is that we do not enjoy the highest health levels. On the contrary, many European countries have agespecific death rates considerably below our own. The relatively high infant mortality rate in this country is disturbing, and difficult to explain. The disparity in death rates for middle-aged males is even more shocking, and has more serious economic implications. In the U.S. of every 100 males who reach the age of

Victor R. Fuchs, The Growing Importance of the Service Industries, Occasional Paper 96. National Bureau of Economic Research, Columbia University Press, New York, 1965. 83-481 0-67-pt. 1-14

45, only 90 will reach 55. In Sweden the comparable figure is 95. During this critical decade when most men are at the peak of their earning power, the U.S. death rate is double the Swedish rate, and higher than that of almost every western nation. It certainly seems legitimate to ask why. This is not necessarily with a view to spending less for medical care-I doubt if anyone can foresee a decline but with a view to developing more effective use of the resources that we are now devoting to health.

A second reason why we should be concerned about medical care costs is the peculiar structure of the medical care industry. Most industries in the United States consist of profit-seeking firms actively engaged in competition with one another. The fundamental rationale of the American economic system is that the ope of profit (and the fear of loss) under conditions of open competition are the best guarantees of efficiency, an appropriate price and rate of output, and a fair return to the various factors of production.

The medical care industry is organized along radically different lines. Nonprofit operations are the rule in the hospital field; there are severe restrictions on entry and competition in medical practice, and advertising and patent control dominate the market for drugs. Thus, there is no a priori basis for believing that the prices and quantities of medical care approach those that would be socially optimal.

A third reason, it seems to me, is that a large and increasing portion of the cost of medical care is paid by third parties. In particular, the taxpayer is being called upon to pick up a substantial share of the bill. Because payment for medical care is increasingly regarded as a collective responsibility, it is natural and appropriate that there should be collective expressions of concern, such as this conference reflects, about the quantity and quality of medical care, and about its price.

These quantities and prices are determined by demand and supply. Let us consider each side of the equation in turn.

DEMAND FOR MEDICAL CARE

Economists say that the demand for any good or service depends upon relative prices, income, and tastes.

Price

How does price affect expenditures? Perhaps the most firmly established proposition about the demand for medical care is that it is relatively inelastic with respect to price. If the price rises relative to other prices, the decline in the quantity demanded will be proportionately less than the increase in price. The result is an increase in medical care expenditures. If, other things remaining unchanged, price rises by 10 percent and quantity demanded falls by only 5 per cent, expenditures will rise by approximately 5 per cent. Some studies suggest that the price elasticity of demand for medical care may be as low as .2, i.e., quantity demanded declines by only 2 per cent when price rises by 10 per cent. But present knowledge does not permit fixing a specific value other than to say that the elasticity is surely below unity.

An aspect of the price of medical care that is not widely recognized, is that it really has two components. One is the nominal price charged by the physician or hospital; the other is the value of the patient's time. For instance, the nominal price of a visit to a physician might be ten dollars, but the trip to and from his office, the wait, and the actual examination will probably take an hour or more. This time might be worth more or less than ten dollars depending upon the alternatives available to the patient.

Once it is understood that the price of medical care includes both components, a number of interesting implications become apparent. Even when a sliding fee scale is not used, the total price of medical care tends to vary with earning power. The price is lower for retired people and the unemployed than for those with jobs, is generally lower for women than for men, and so on. Also, even when the nominal price is reduced to zero, as under prepayment plans or socialized medicine, the true price is not zero.

Income

One of the factors to be considered in any demand study is real per capita income. During the past twenty years this has risen by over 50 percent, and

Gary S. Becker, "A Theory of the Allocation of Time," Economic Journal, 75, No. 299 (September, 1965), pp. 493-517.

there is no doubt that the demand for medical care increases with income. What is less clear is whether the demand for medical care is elastic or inelastic with respect to income, i.e., does a given percentage increase in income lead to more than, or less than, the same percentage increase in medical care expenditures, other things remaining the same. The question is only gradually yielding to attack as more and better data become available and analytical techniques are sharpened. Some recent studies suggest that the elasticity may be significantly below unity, and few investigators believe that it is greater than unity. At most, the demand for medical care seems to increase approximately in proportion to income. If this is true, we cannot attribute any of the increase in the share of total expenditures accounted for by medical care to rising income.

Insurance

A special factor that complicates the analysis of the demand for medical care is the growth of insurance and prepayment plans. Once a person is covered by such a plan, the effective price to him of additional units of medical care depends only upon the value of his time. It seems to me that this may explain a large part of the increase in the quantity of medical care demanded, and may also help explain the apparent insensitivity of insured consumers to increases in the nominal price of medical care. It is worth noting that hospital care has shown the most rapid rate of increase in expenditures, and it is hospital care that has been most thoroughly covered by insurance and prepayment.

The curious behavior of dental expenditures also offers support for this hypothesis. All the available evidence suggests that at any point in the time the demand for dental care is more elastic with respect to income than is the demand for physicians' services. Nevertheless, during these recent decades of sharply rising real income, expenditures for dental care have increased less than have expenditures for physicians' services. One possible explanation is the very small role played by insurance and prepayment plans in the dental field. Expenditures for eye glasses and appliances, and for drugs, two other components of medical care that are typically paid for directly by the consumer, have also risen much less rapidly than have expenditures for hospitals or physicians.

This should not come as a surprise. The advocates of insurance and prepayment had something like that in mind. They wanted to remove any financial barriers to obtaining medical care. But it is a basic law of economics that if you lower the price, the quantity demanded will increase. A critic of the British National Health Service put the matter cogently, albeit a bit strongly, in a recent issue of The Lancet. He wrote, "if taxi fares and meters were abolished, and a free National Taxi Service were financed by taxation, who would go by car, or bus, or walk... the shortage of taxis would be endemic, rationing by rushing would go to the physically strong, and be more arbitrary than price, and 'the taxi crisis' a subject of periodic public agitation and political debate.""

I am not suggesting that insurance and prepayment should be abandoned. But we do need to discover techniques, possibly such as coinsurance, deductibles, or experience rating, to check prices and expenditures without interfering with essential health services.

Tastes

All factors other than income or price that affect demand are put by economists in a catch-all category called taste. In the case of medical care, these would be the factors that affect the health levels of the population, and those that affect attitudes toward seeking medical care at any given level of health. Taste for medical care, therefore, would be related to: (1) demographic variables, (2) education, (3) envirnoment, (4) ways of living, and (5) the genetic stock of the population. Research on these matters is only in its infancy, and there are few reliable findings to report. We know that an increase in the proportion of elderly people in the population tends to increase the demand for medical care, other things remaining the same. The effect of increased education is unclear. It probably leads to improved health levels, and thus less need for medical care, but may also lead to a greater demand for medical care at any given level of health.

Most observers believe that recent environmental changes, particularly the increase in real income per capita, have contributed to better health status. I think that this inference is incorrect. Some tentative findings from my research

* Arthur Seldon, "National or Personal Health Service," The Lancet, March 25, 1967, No. 7491, Volume 1, page 675.