ing individuals to supplement that privately if they want but not subsidizing them through the tax system to buy health insurance which is really not worth the cost to them.

Mr. ROSTENKOWSKI. Professor Klarman?
Mr. KLARMAN. May I make several points?

One, Professor Feldstein has raised a basic question, namely, how do you go about determining the appropriate level of health care expenditures. He tells us that we ought to look to the consumer for guidance.

But he himself has taught us that often it is not the consumer who makes the decision; it is the physician who makes it, presumably acting on behalf of the consumer.

Two, it is enlightening to ask why and how we got into health insurance. Whether or not it is properly prepayment, rather than insurance, we got into it for a very good reason: The unexpected and high costs of medical care. These can be very threatening to an individual or a family and, indeed, there is no reason to believe that an individual or a family can always accumulate enough funds to pay for a particular episode of illness.

It turns out that as prepayment or insurance relieved some of the financial burden of illness, it has also produced other consequences that are not so positive. Thus, there is the tendency for people to use more health care services than they would use in the absence of insurance. In particular, we have observed the phenomenon characterized as Roemer's law (named after Professor Milton Roemer of UCLA): under conditions of prepayment, if a hospital bed is built, it will be used. Clearly this tendency is linked with the presence of insurance or prepayment, and I can see no way to avoid it.

Three, still on the point of what would be the appropriate level of expenditures, I am not in a position to report to this committee that economists are now able to prepare usable cost-benefit analyses, in which the marginal benefits and marginal costs of health care are equalized.

Given these considerations, what is the next best strategy to adopt? It seems to me that the next best strategy is to curtail expenditures where we can without doing harm. That is why I have recommended the policy of limiting and curtailing the number of short-term hospital beds.

I have also suggested that we move promptly toward changing the reimbursement of hospitals from retroactive costs to some other prospective basis and linked to a hospital's own budget.

May I add that this country's experience with respect to health care expenditures has not been as bad as we thought it was going to be. The figure on the percentage of such expenditures to the GNP remained stable at 7.7 percent for a period of 4 years, 1971 to 1974. In fiscal year 1975 it has gone up some, to approximately 8 percent.

I would be less than candid if I did not add that if some of the things I suggested in long-term care were adopted, some expenditures would go up. Presumably, for this purpose we should be willing to pay the bill.

In summary, several other factors are at play, in addition to consumer preferences, and there are some instruments available to us that we can employ to reduce expenditures where little harm would be done.

Mr. SIEGFRIED. I may have misunderstood some of the statements but I was pleasantly struck by the large areas of agreement among the various panelists this morning. I think we have come a long way and we see many of the problems confronting us in the same way. I was shook up a little more by the statement of Professor Feldstein, but it brought back to mind the problems we had in the early days of developing insurance plans where those of us who favored deductibles and coinsurance were treated scornfully and the emphasis seemed to be No. 1 on hospital care and paying hospital care in full from the first day. Those of us who favored emphasis on catastrophic coverage were troubled by this. For reasons I understand in part and don't understand in part, the consumer, when he had a chance to express a choice, seemed always to prefer first dollar coverage to catastrophic coverage and it has been an uphill struggle to turn that around. I think it is worth commenting on at this time because I think it is still one of the major issues that confront us as we design a national health insurance plan because many of the proposals, I think, include a greater degree of emphasis on first dollar coverage than would be desirable, so I think some further exploration of the ideas that were advanced by Professor Feldstein are appropriate and in order. Mr. ROSTENKOWSKI. Mr. Cohen.

Mr. COHEN. I think eventually you are going to cover all first dollar costs in some national health insurance system at some time. That is the consumer and voter attitude and ultimately in the American system the consumer and the voter decides, not the economists and not anybody else.

But I don't believe we ought to start that way. I would start with Mr. Waggonner's bill on catastrophic costs and proceed by the basis of experience. This whole issue of deductibles and coinsurance is a highly theoretical issue in which people who are insurance-minded and economists have a lot of considerations which the average man in the street does not believe in and does not adhere to.

I would think that, therefore, the best way to start is to have coinsurance and deductibles. Why? Because then you get experience and if you want to lower them, you can lower them; you can get to first dollar costs if that is what the taxpayer wants to pay.

If you start with first-dollar costs, putting coinsurance and deductibles in is a tremendous political difficulty. Anybody who has tried to put a deductible or coinsurance feature in after you have had first-dollar costs as they had in the previous system will have difficulties. Therefore, start the other way around. Start with the coinsurance and deductible along the lines of Mr. Waggonner's bill and then when you see what happens and what the taxpayer and consumer want, you can always make an adjustment.

I think that there is no real good answer to this deductible coinsurance issue. In the first place, taking the points Dr. Feldstein mentioned, the rational use of resources-nobody but the economists really care about that issue. The consumer does not care about rational use of the resources. When a baby is sick at 2 o'clock in the morning and the mother wants a doctor, does she ask, is this a rational use of available resources? Of course not. It does not do much for the baby when she calls the doctor and he says take two St. Joseph's aspirins, which the mother has probably already done. It is for the mental health of

the mother. Do you put that into quality of care? It is virtually impossible. You know there is going to be inefficient use of resources. When a mother calls a doctor at 2 o'clock in the morning or any time she calls him, is that a valid component of quality of care? From the mother's standpoint that is what she sees, just like Dr. Donabedian

said.

What does she want? She wants a psychological personal relationship of a doctor reaffirming what she knows what he is going to do anyway. It is very important and she wants to pay for it and the husband wants to pay for it and the union he belongs to wants to pay for it.

Mr. FELDSTEIN. It is a good use of resources.

Mr. COHEN. That is right, but she determines that, not the economists. There is no way to determine that except in the free marketplace by mama.

My view on this matter is as Mr. Siegfried said. You have to approach this in an evolutionary way. You can't get to the millenium by enacting it on one given date. You have to go through a system of difficulties and torture and distress in the evolution of whatever this committee is going to enact. But out of that will come experience. Out of that will come the taxpayer indicating what he wants to spend his money for. Out of that will come whether various people think it is a good system.

Another reason why I think this matter of the deductible coinsurance is not really an important issue is no matter what this committee does, if it includes a deductible and co-insurance in the public plan, the private insurance companies will sell a policy to take up the difference. So, you will end up with a first-dollar system anyway if you put the public system and private system together. If you look at the total use of the resource of the economy, you have to look at both of them. Therefore, I come to the conclusion by a different route with the same conclusion of Dr. Felstein, to start with the co-insurance in the public sector and then see what happens by the combination of public and private and then you can always change it later on if you. want to or have to.

Mr. ROSTENKOWSKI. Dr. Donabedian.

Dr. DONABEDIAN. I want to present my biases and prejudices. I think a problem we ought to remember as we think about co-insurance, deductible, and so on, is this: When a person is ill or a child in the family is ill, as Dean Cohen mentioned, we often don't know what is wrong and we don't know what is needed and we don't know how much is needed and, therefore, we can't really make good decisions partly because we don't know and partly because we are so upset and worried. So, I would not like to see anything in a health insurance scheme that prevents or discourages people from seeking advice about what they need to do, going to a doctor or getting some medical advice.

The problem is what happens once you go to the doctor and what does he do for you when you are ill. That is where the issue of quality comes which is partly quality, partly quantity, and partly, of course, cost. Does the physician do the things that are necessary and avoid doing unnecessary things? Does he do harmful thinos? Is he careful about the relationship between costs and benefits and harm due to the procedures he recommends?

I would like to see a great deal of our attention in terms of cost control focused at this point, at what is the kind of good care which is not excessively costly, rather than at encouraging a person who does not know what is wrong with him make the decision on the basis of how much it is going to cost him if he goes to the doctor.

Mr. ROSTENKOWSKI. Professor Feldstein?

Mr. FELDSTEIN. I think Wilbur and I actually agree on a lot. We agree that what is a good use of resources is determined by the consumer. There are some areas like military spending or medical research where the consumer really cannot get into that action. For personal health services, Wilbur and I agree that that mother at 2 in the morning is the right person to determine it.

Of course, if you make the call to the doctor free, then she does not have to think about the cost of the resources at all but if, as he said, it depends on a bit out of pocket and what the labor union decides, basically the costs are brought home to roost. Then, what economists call the right allocation of resources occur. We really agree and I am delighted we do. I am also delighted we agree about co-insurance and deductibles.

I think that co-insurance and deductibles should be an important part of any national health insurance plan that you enact. I think it is not a small issue as Wilbur suggests. I think as you know from some of the cost estimates you have seen high co-insurance and deductibles can make the difference between a $70 billion tax bill for that program and a much cheaper tax bill. Whether you do it through the tax route or mandated coverage of the employers, a tax is a tax is a tax. I don't share his feeling if you have substantial co-insurance and deductibles, private insurance companies will sell the fill-in. I think we have to ask why it is consumers have preferred shallow coverage in the past.

I think one reason why consumers have preferred shallow coverage in the past is the tax subsidy I have been talking about. If you give a 30-percent tax subsidy or $4 billion tax subsidy to the individual. he will prefer to buy shallow coverage rather than taking it in cash and buying deductibles and co-insurance.

At the other end, doctors and hospitals have forgiven large bills so the patient shies away from the catastrophic insurance because this "insurance" was provided by the providers themselves. If you provide for a family with $10 or $12 thousand income a co-insurance rate that limits their expenditure per year to, say, a thousand dollars, if they went out to buy private insurance to fill in that thousand dollar gap it would cost them something like $600. I think most families would prefer to take the chance knowing most of the time they will not run up a large bill rather than buying the insurance that costs them almost as much as the amount they are protecting themselves against. I think if you limit each family's expenditures under a new national health insurance plan to a fraction of their income, the demand for additional insurance will not exist unless you subsidize them either through the tax system or failing to have corridor deductibles in the program.

Similarly, the taxpayer will not want to pay that bill to get the fill-in. If you start with high deductibles and co-insurance it is likely to stay that wav, as Wilbur suggested.

Mr. ROSTENKOWSKI. Professor Klarman.

Mr. KLARMAN. Obviously, there is much agreement, but I am not so sure that we agree on the policy implications.

With respect to deductibles and co-insurance, we don't know all that much. There have been a couple of studies, but we really don't know about long-term effects. The main basis for judgment at this time, it seems to me, would be a political basis; namely, that if you start with co-insurance and deductibles and you don't like the consequences, you can make changes. It is difficult to move in the opposite direction.

In this connection let me point out that in the case of short-term hospital care a very high proportion of the bill, perhaps 90 percent, is now paid for, one way or another, by third parties. If we were to move toward deductibles or co-insurance for hospital care, we would be taking something away from patients. I submit that we would also be taking something away from the providers, who, rightly.or wrongly, have come to rely on a steady stream of revenues.

Finally, I believe that if we were to take co-insurance and deductibles seriously as a means to deter the utilization of services, then Congress would have to enact a law that would prohibit private insurance plans from filling the gap. I see no way to escape that conclusion. Mr. COHEN. I agree. I would like to expound on that, if I could.

I go along with Dr. Feldstein up to a point but unless you are going to prohibit unions from bargaining with employers on the supplementation, they are going to press for first-dollar costs.

Mr. ROSTENKOWSKI. Under the current tax law?

Mr. COHEN. In my opinion because they will assume it is wages. They assume it is a fringe benefit.

Mr. FELDSTEIN. They could have wages instead and now there is a substantial subsidy for taking it in the form of health insurance.

Mr. COHEN. I don't think you give enough weight to the consumer attitude on this question. You are treating it like an economist does as what are the economic incentives. I go along with Mr. Siegfried. He said he has been trying over the last several years in the selling of insurance to try to get over the idea that co-insurance is a much more sensible policy, but my experience has been, and the best indication is the extent to which people want maternity insurance covered under it. I always thought that that maternity was a voluntary act. Yet, the whole point of people is not that they want to get some tax benefit, they want their medical expenses spread over a long period of time, in my opinion, and that is why the consumer has gone to first dollar costs.

I am for starting with some coinsurance and deductible but I don't do it on the theory that that is going to prevent insuring substantially the total medical bill because I think it is going to happen one way or the other.

Unless you prohibit the collective bargaining, I don't see that you can do it.

Mr. FELDSTEIN. You say people want maternity care insurance but that is because it is tax subsidized.

Mr. COHEN. No, you have the wrong attitude of consumers on this matter as many economists do.

Mr. FELDSTEIN. He can put another $50 or $70 into a pay package. Mr. COHEN. I am talking about momma and pappa. When they consider they are going to buy their insurance