Page images

NATIONAL HEALTH INSURANCE (Problems and Issues in Health Care Organization, Delivery,

and Financing)



Washington, D.C. The subcommittee met at 10 a.m., pursuant to notice, in the committee hearing room, Longworth House Office Building, Hon. Dan Rostenkowski (chairman of the subcommittee) presiding.

Mr. ROSTENKOWSKI. Good morning, gentlemen. If our witnesses would kindly take their chairs we will get underway. We welcome you to the Subcommittee on Health of the Ways and Means Committee. I would like to explain what we expect to do this morning.

It is our intention to hold this panel discussion at least until 12:30 when the full committee will have to meet on another matter. If we have not concluded the discussion and the questions on the part of the members of the subcommittee, we will return at 2 o'clock.

I would like to explain that you have the privilege of submitting your whole statement for the record if you so desire, and now you can summarize it. Immediately after all the panelists have concluded their statement we expect an interchange in various areas where there might be some disagreement.

We are looking forward, gentlemen, to a lively discussion. We hope that you will gain as much as we have gained from the other panelists who have participated in these discussions.

I would ask you each, when you make your statement beginning with Professor Feldstein, to introduce yourselves.



Mr. FELDSTEIN. I am Martin Feldstein, professor of economics at Harvard University. I am very pleased to have this opportunity to talk with all of you this morning. I understand that the subcommittee wants to concentrate today's discussion on the basic problems of our health care system and not get into details about alternative national

health insurance proposals. I would like to focus, therefore, on the single problem that I believe is responsible for the widespread interest in national health insurance: The rapid explosion of health care costs.

This inflation of health costs is important because it reflects a serious misallocation of resources and a failure of the health-care system to reflect individual preferences. Moreover, high medical care costs are the primary source of inequity in our health system, creating financial hardship and imposing a barrier to adequate medical care.

I think the basic facts of health cost inflation are well known to the committee. I summarized them on the sheet I handed out this morning. They indicate in the period from 1950 to 1974 prices as a whole in the overall Consumer Price Index went up about 114 percent. During the same period medical care costs rose more rapidly, 193 percent. But the cost of hospital care was much more dramatic in its upsurge. Average cost per patient-day was only $16 in 1950. It was about $125 in 1974, an increase of more than 650 percent.

While prices in general have doubled, health care costs have gone up triple and hospital costs seven and a half times. These increases in hospital costs have really dominated the rise in health care costs. By comparison the prescription and drug charges hardly changed over that period and physician fees have only gone up at the same rate as all other services in our economy. I will, therefore, limit my attention to the problem of hospital costs.

I think everyone is aware that hospital costs have risen rapidly, but there is little understanding of why this happened or why hospital cost inflation is very different from the other types of inflation that currently trouble our economy. Let me begin by giving you my own view of the nature and cause of this problem.

The two key ideas in my explanation are: First, the changing nature of the hospital product; and, second, the impact of insurance.

Consider first the importance of product change. The most obvious thing about hospital care today is that it is very different from the care given in hospitals 25 years ago. Today's care is more complex, more sophisticated and, hopefully, more effective.

The rapid rise in hospital costs really reflects a rapidly changing product, a rapid change in quality or style of care. It is, therefore, unwise to compare the rate of hospital cost inflation to the rate of inflation of most other goods in the Consumer Price Index. The cost of inflation of hospital care is not that consumers are paying so much more for the same old product that they bought before, but that they are buying a different and much more expensive product today.

Therefore, hospital cost inflation is quite different from other types of inflation in our economy. It is nevertheless a serious problem. To understand the nature of this problem, we have to ask why hospital cost inilation has occurred, why the style of care has become much more sophisticated and, therefore, much more expensive.

Higher incomes and greater education have, no doubt, played some role in increasing the demand for sophisticated hospital care and scientific discoveries have obviously changed the technological possibilities in hospitals. But the major reason I believe, for the hospital cost inflation has been the very rapid growth in insurance.

Let me explain why. In addition to providing protection against unpredictable medical expenses, health j asurance substantially lowers


the net price of care that the patient pays out of pocket at the time he consumer services.

There is now substantial evidence that consumers are responsive to price differences, that is, that patients, guided by their doctors, demand more services and more expensive services when a large part of their costs are offset by insurance.

Some simple but striking numbers, which I have noted on the sheet I have passed out, will illustrate my point. In 1950 when average cost per patient-day was $16, private insurance paid 37 percent of hospital bills. That means on the average the net cost to a private patient was $10. By 1974 average cost per patient-day had jumped to about $125, but private insurance was paying 77 percent of the private hospital bill, leaving a net cost to the patient of $28.50. Thus, cost per day was up from $10 to $28.50, but $28.50 in 1974 really only brought $13 worth of goods and services in 1950 prices.

So in real terms the net cost to the patient at the time he consumes care had hardly changed at all during the 25-year period, having risen only $3. Consumer demand has, therefore, encouraged and supported the growth of more sophisticated and expensive care and consumer demand has done this primarily because of the growth of insur

Looked at somewhat differently, with 77 percent of private hospital bills now paid by insurance, an extra $10 of expensive care only costs the patient $2 out-of-pocket. It is not surprising, therefore, that patients and their doctors continue to encourage the growing sophistication and expense of hospital care.

The same process has been occurring during the 8 years since medicare was introduced. In 1966 hospital cost per patient-day was only $18. The 1974 figure of $125 thus represents a 160-percent increase in only 8 years. But because private insurance also grew rapidly, the net cost to the patient in 1966 prices remained essentially unchanged.

I think this is the essence of the hospital cost inflation problem: Increased insurance has induced hospitals to change their product and provide much more expensive and sophisticated care.

Before I talk about the implications of this explanation, let me contrast this with the usual reasons offered for the rise in hospital costs. These traditionally boil down to four ideas: (1) Hospitals are ineflicient; (2) labor costs have risen particularly rapidly; (5) hospitals had a low rate of technical progress; and (1) supply has not kept up with increasing demand.

I think each of these ideas is basically incorrect as a diagnosis of hospital cost inflation and misleading as a basis for policy in this area. Let me explain why.

Perhaps the most frequently heard explanation of rising hospital costs is that hospitals are technologically and mangerially inefficient, that they get less output for input than ordinary business firms. Even if there are reasons for criticizing the inefficiency of hospitals, there is no reason to believe that this ineificiency has been rapidly increasing. Inefficiency could not account for a 650-percent increase in hospital costs. If hospitals are less efficient, their costs are higher than they should be, but not necessarily rising, let alone at such rapid rates.

Rising labor costs are also cited as the primary cause of hospital inflation. It is true that wages and salaries constitute a large share of

hospital costs and wages have risen there more rapidly than wages in the general economy. Nevertheless, this does not begin to account for the rise in hospital costs. From 1955 to 1973 labor costs rose 350 percent. But as a fraction of the total hospital bill labor costs actually decreased from 62 percent of total costs in 1955 to 56 percent in 1974. In other words, nonlabor costs have risen faster than labor costs.

Moreover, about a fifth of the increase in labor costs reflects a rise in the number of personnel per patient today. Only a third of the increase in total costs can be attributed to wage increases per se. Moreover, since hospital wages rose 188 percent while wages in the economy rose about 130 percent during this period, the excess rise in hospital wages—the difference between the 188 percent and 130 percent increase—can only account for a small part of the total increase in hospital costs, probably on the order of 14 percent.

The third explanation is that hospital cost inflation is due to a low rate of technical progress. I think this is clearly and obviously false. Hospitals have been the scene of extremely rapid technical change. But the character of these changes has been different from that in other industries. It has not been cost-reducing. Technological progress in hospitals does not involve making the old product more cheaply, but making a new range of products that are more expensive.

Why have hospitals moved toward increasingly expensive ways of doing things rather than providing old products more cheaply? Although some of this reflects the path of basic scientific progress, it is our method of financing health services that primarily determines the pattern of technical change.

The final traditional explanation is that hospital costs have risen because supply has not kept up with demand. I have already pointed to the reasons for increasing demand, higher incomes, greater education, and especially a rise in insurance. Usually economic analysis of ordinary markets shows that prices rise because supply does not increase as rapidly as demand. But in the case of hospitals, I think the opposite is true.

It is precisely because supply has kept pace with demand that hospital costs have gone up. Hospitals have responded to the increased demand and willingness to pay for sophisticated services by providing those services and costs have gone up accordingly. The increase in demand has induced a rapid increase in the supply of a more expensive type of hospital care.

This brings me back to my original contention that the rise in hospital costs reflects a change in product induced largely by the growth in insurance. But this explanation of the rise in hospital costs raises an awkward question. Implicit in every discussion of hospital cost inflation is the assumption that the rise in cost has been excessive and should not be allowed to continue at the same rate in the future. But if this rise reflects a change in product rather than an increase in inefficiency or a low rate of technological progress, why is it really a problem?

The answer in brief is that the current type of costly medical care does not really correspond to what consumers or their physicians would regard as being appropriate if their choices were not distorted by insurance. The effect of prepaying health care through insurance, both private and public, is to encourage hospitals to provide a more expensive product than the consumers actually wish to purchase.

Although the consumer pays for the expensive care through higher insurance premiums, at the time of illness the patient's demand for service reflects the net cost of the care. Because this net out-of-pocket cost appears so modest, the patient is willing to buy more expensive care than he would if he were not insured. In this way our current method of financing hospital care denies patients the opportunity to choose effectively between higher cost and lower cost hospital care.

If insurance is responsible for such an inappropriate expansion in the demand for expensive care, why has insurance grown so rapidly? In part the growth of insurance reflects a family's rational demand for protection against unexpected illness. It is unfortunate, but inevitable, that this process is self-reinforcing. The high cost of care induces the patient to buy more complete insurance and the growth of insurance induces the hospital to produce more expensive care.

But this demand for protection cannot explain the comprehensive first dollar insurance that now exists. Current insurance is often inadequate in protecting the family against the substantial bills that can cause real hardship. Why then have American families bought such complete coverage for relatively small bills? Why have they been willing to pay for insurance that often provides little real protection, but induces them to buy more expensive and sophisticated care than they really want?

As you know, most insurance is now group insurance and, more specifically, insurance bought for employee groups. The decisions on the scope of coverage on co-insurance rates and deductibles are generally made in collective bargaining agreements by expert representatives of labor and management.

Why should such experts forego higher wages in order to obtain excessive, shallow insurance? The answer, I believe, lies in the tax treatment of premiums. Federal policies encourage insurance by a tax deduction and exclusion that now costs the Treasury more than $4 billion a year. As

you all know, individuals can deduct about half of the premiums they pay for health insurance. More important, employee payments for insurance are excluded from the taxable income of the employee as well as the employer. These premiums are not subject to social security taxes or State income taxes.

Even for a relatively low-income family the inducement to buy insurance can be quite substantial. Because of the income and payroll taxes, a married man with two children earning $8,000 a year will take home an additional $70 for each $100 the employer adds to his income. If the employer buys health insurance instead, the full $100 can be applied against the premium and there is no tax to be paid.

In this case the dollar buys nearly 50 percent more health care services if it is paid through an insurance premium than if paid in wages and left to the individual to buy the care directly. For workers in higher tax brackets, the incentive is stronger. In the aggregate the Government tax subsidy exceeds the total profits and administrative costs of the insurance industry by a substantial margin.

I believe the subsidy is strong enough to induce employees and unions to opt for higher insurance instead of higher wages. The primary effect of this insurance is to distort the pattern of care and exacerbate the rising cost of hospital care. This tax subsidy costs the Government

« PreviousContinue »