accounts. If employers can be required to provide health insurance as a condition of doing business, why not a given level of private pension protection or day care for working mothers, or whatever other benefits might seem desirable?

This concept of Congress voting social benefits but not having to face the cost seems to me a rather strange approach to be recommended by the relatively conservative administrations of President Nixon and President Ford. I suppose it must be that the ideological appeal of keeping the administration of programs in private hands to the greatest extent possible outweighs for them the lack of financial discipline inherent in the approach.

Let me say briefly why it is that the supporters of a Government plan are so strongly opposed to a mandated plan. Leaving this precedent question aside, what are the different results of selecting one plan rather than the other? I am going to say this in very summary form and then will be glad to respond in more detail on what I mean by the individual points.

(1) I do not think there are solutions leading to universal coverage in employer-based coverage, where the coverage is tied to a particular employer. Even though Mr. McNerney has made some ingenious suggestings for reducing the size of the problem, he has by no means eliminated it. The coverage cannot follow everybody if the main idea is to tie the protection to an individual employer. Thus, it is necessary to accompany the employment-based plan with a low-income plan to fill in the gaps.

You might say, "What is so bad about that?" I think what is bad about it deserves extended discussion. It may be necessary under any approach, at first. I am not saying all income-tested programs can be done away with immediately, but it seems to me it is very important that we start on an approach which, over time, will cover everyone in the same system. It is very difficult to prevent plans that are aimed specifically at the poor from becoming second-class plans. Most of us have strong charitable feelings from time to time, but for sustained year in and year out interest, we are most likely to give our attention to the systems that affect us directly.

Everyone will be interested in seeing that a universal plan is a good one and well administered. Fewer will be interested in seeing that a plan for the poor meets high standards. In other words, low-income people are likely to receive equal treatment in medicine only when they are free to go to the same hospital and clinic and the same physician, under the same conditions, as the worker whose costs are paid for by an insurance plan. I wish to make the following points:

One: The inability to provide universal coverage of everyone in the same plan through an employer-based approach.

Two: Another difference, it seems to me, is that since private group plans, except very small ones, tend to be experience rated, the approach coverage to the individual employer makes the cost to that individual employer dependent on the health experience of his own labor force. This adds to the incentives that are already there for employers not to hire older workers, the handicapped and all who have above-average health costs.

Three: If you have a contributory plan in the mandated approach, which was true of the Nixon-Ford proposals, then the costs fall more heavily on the lower paid worker because the dollar premium does not vary under such a plan by the amount of wages of the worker, but rather vary by family composition, that is, single or family coverage. The low-wage earner is paying the same dollars as the high-wage earner, and, of course, that is a higher proportion of his earnings. If you use a percentage of earnings approach, as in social security, the lower paid pay less in dollars.

Four: There has already been some discussion earlier today of the extent to which private insurance would have larger costs by reason of the selling cost involved. I think it is indisputable there will be some difference. How much can be argued.

Five: The final point, which to me is perhaps the most important one of all, is that others who, like me, believe in the Government approach, believe that over time a single plan which brings the costs together as an expenditure by Government will be under greater pressure to control costs, and to take steps to see that fair value is obtained for what is spent. I know it is popular today to deride the ability of the Federal Government to run programs. This is part of the mood of the times, but many of them do work and work well-social security and hospital insurance under medi

uare, to name two.

I might say in an aside, Mr. Chairman, since the issue has been raised by the previous witnesses, that there was a mixed ability shown by private carriers in the early days of medicare. I would say that there was a lot of handholding on the part of Government in order to make that program work. It was not as if we had made contracts with highly skilled people competent to pay only the claims that were valid and operating highly efficient systems. This was true in only a very minority of cases.

Social security has developed a good partnership by now with most of the contractors, and that it does work reasonably well. My own suggestion would be that a general Government-operated program should use intermediaries along the same lines as medicare. I would not propose that Government do everything. But the idea that there is this great difference in capacity between Government to operate directly and those private plans that I am familiar with to operate, in my opinion, is not so.

The main point I want to make here, however, is that although the Federal Government may make mistakes, and it certainly does, there are pressures that drive program administrators strongly in the direction of being responsive to program beneficiaries and payers. If things don't go well, hearings are held. For one thing, mistakes are brought out in the open, corrective measures are taken and over time the right questions of cost, quality and access to care are directly addressed. Under the more diffuse mandated approach, there is bound to be, in my opinion, less accountability for performance, and in the medical care market competition just does not work as a substitute for accountability. It is not that kind of market.

I believe that a Government plan in particular would be expected over time to define much more carefully in the future the medical care product that it is willing to pay for. At a minimum, the plan at some point will have to be concerned about whether it is paying for procedures that are likely to do more harm than good. Increasingly, it will be expected that procedures paid for be scientifically established as efficacious.

I have considerably more, Mr. Chairman, in my prepared statement on this same general point about the drive that I believe will result from a Government plan to both control costs and to take steps to see that quality care is given as compared with the more diffused approach of the mandated plan. I don't mean to make this contrast on a 100-percent basis. It is my conviction that either approach will be required by the public and the public's representatives to do much more in the future than has been done in the past about both cost control-by which I mean as much utilization control as I do unit cost control-and also about the quality of the product that is being paid for. But I believe the pressure will be greater if it is focused on one accountable in a Government plan. Now there are many other important questions in the design of a national health insurance system. What benefits are to be covered, whether there are deductibles and coinsurance and so on. I think those matters can be changed over time with experience. What is very difficult to change is the basic institutional setup-a socialsecurity type health insurance plan, on the one hand, and mandated coverage under private insurance, on the other. Nothing is more persistent in Government than a basic decision or financing and administration. Vested interests are created, once a program is in operation, that make fundamental changes later very difficult to accomplish. Workmen's compensation follows a pattern adopted before World War I. Unemployment insurance is still a Stateoperated program plagued with major inefficiencies and inadequacies, in past, because of a 1935 notion that differing unemployment insurance tax rates would create incentives for employers to stabilize employment.

The way we go on the fundamental issue of Government responsibility could shape the national health insurance program for a generation to come, at least.

Thank you, Mr. Chairman.

[Mr. Ball's prepared statement follows:]

STATEMENT OF ROBERT M. BALL, SENIOR SCHOLAR, INSTITUTE OF MEDICINE,

NATIONAL ACADEMY OF SCIENCES

Mr. Chairman and Members of the Committee: My name is Robert Ball and I am now a Senior Scholar at the Institute of Medicine of the National Academy of Sciences. From April 1962 until March 1973 I was Commissioner of Social Security and prior to that served for approximately 20 years in various positions in the Social Security Administration and its predecessor organization, the Social Security Board. I am testifying today as an individual, and my opinions do not necessarily represent those of any organization with which I am associated.

The central issue in the national health insurance debate-the one that will have more importance than any other in the long run-is whether to set the program up like social security, federally administered (although perhaps as with Medicare, contracting out part of the administration to private carriers)

and federally financed, or whether the government's role should simply be to require employers to take out private health insurance as in most state workmens' compensation programs.

On this fundamental issue those with a special interest in health insurance have divided into two camps, supporting approaches that are basically so different that compromise is next to impossible. On the one hand, there are the physicians, the insurance organizations, and the hospitals supporting the use of government to assure widespread coverage under private health insurance-an approach which has come to be called the "mandated approach." On the other hand, we have labor and a variety of social welfare and consumer groups supporting a direct government program financed through government and with overall administrative responsibility lodged in the government.

In some ways the mandated approach has great political appeal. It is very, very appealing to vote for the benefits-that is, pass a bill requiring employers under defined conditions to provide health insurance protection-and yet, at the same time not have to vote on how to raise the money. Under the mandated approach, the problem of cost is hidden in the private insurance premiums paid by employers and employees. The same benefits may cost the same, probably more in a mandated plan, but the cost doesn't show up in the government's accounts.

At the same time, it is this political appeal that could make this approach a precedent for quite radically enlarging the scope of social benefits required by law, but paid for outside of government accounts. If employers can be required to provide health insurance as a condition of doing business, why not a given level of pension protection, day care for working mothers, a minimum amount of vacation, or whatever other benefits might seem desirable? This proposal to have Congress vote social benefits but not have to face the cost seems to me an unusual one to come from the relatively conservative Administrations of President Nixon and President Ford. I can only speculate that the ideological appeal of keeping the administration of programs in private hands to the greatest extent possible outweighs for them the lack of financial discipline.

But why is it that the supporters of a government plan are so strongly opposed to a mandated plan? Leaving precedent aside, what are the differing results of selecting one plan rather than the other? I have indicated below the differences that seem to me the most important:

1. Under a social security-like approach, coverage can be made universal and follow the worker from job to job. If, on the other hand, coverage is tied to employment with a particular employer, as it is under the "mandated approach," major gaps are unavoidable as employees become unemployed or change employers. The necessary definitions of who is covered (for example, full-time employees versus part-time employees), the length of time a worker has to be employed before the employer is required to pay for his coverage, how long coverage lasts after a worker loses his job, etc., all result in some people not having the protection when they need it.

2. Since private group plans, except very small ones, are experience rated, the approach of tying coverage to the individual employer makes the cost to each individual employer dependent on the health experience of his own labor force. Thus the approach of mandatory private insurance coverage sets un additional economic motives for not hiring older workers, the handicapped, and all who have above-average health costs.

3. Under a contributory plan of mandating private insurance coverage, as in the Nixon-Ford proposal, costs fall more heavily on the low-wage worker than on those who are higher paid. The private insurance premium is the same dollar amount for the low-wage worker and the high-wage worker, and thus, of course, a higher percentage of earnings for the low-wage worker. Under social security, the percentage of earnings is the same so that the lower paid pay less.

4. Requiring employers to buy private insurance increases costs by the amount of the selling costs involved in the use of private insurance.

5. Under a social security-like approach, coverage could be universal, ninetyfive percent of the population would be eligible either as insured workers or as the dependents of insured workers. The other five percent of the population could be covered through the same plan by providing for general revenue contributions on their behalf. Under the mandated approach, it is impossible to 70-411-76-vol. 3-36

get everyone covered. The mandated employer plan needs to be supplemented by a plan for those low-income people who do not earn eligibility through a place of employment. Thus the Nixon proposal had the two plans-a mandated employer plan and a low-income plan-plus a continuation of Medicare for elderly people.

If at all possible, it is best, in my opinion, to adopt an approach which— even if not immediately-over the time would cover everyone in the same system. It is very difficult to prevent plans aimed specifically at the poor from becoming second-class plans. Most of us have strong charitable feelings from time to time, but for sustained year in and year out interest we are most likely to give our attention to the systems that affect us directly. Everyone will be interested in seeing that a universal plan is a good one and well administered; fewer will be interested in seeing that a plan for the poor meets high standards. In othr words, low-income people are likely to receive equal treatment in medicine only when they are free to go to the same hospital or clinic and the same physician under the same conditions as the average worker whose costs are paid for by an insurance plan.

6. The supporters of the government approach believe that, over time, a single plan which brings the costs all together as an expenditure by government will be under greater pressure to control costs and to take steps to see that fair value is obtained for what is spent. I know it is popular today to deride the ability of the Federal Government to run programs. This is part of the mood of the times. But many of them do work and work well-social security and hospital insurance under Medicare, to name two. The point here is that although the Federal Government may make mistakes there are pressures that drive program administrators strongly in the direction of being responsive to program beneficiaries and payors. If things don't go well hearings are held, mistakes are brought out in the open, corrective measures are taken, and over time the right questions of cost, quality, and access to care will be directly addressed. Under the more difuse mandated approach there is bound to be less accountability for performance. And in the medical care market. competition just does not work as a substitute for institutional accountability. In the medical care market, where only the seller of the services and other physicians are competent to decide what the buyer needs, th government as a third party will be expected to be a sophisticated stand-in for the individual consumer. The government would be expected to define much more carefully in the future the medical care product it is willing to pay for. At a minimum, the plan at some point will have to be concerned about whether it is paving for procedures that are likely to do more harm than good: increasingly. it will be expected that procedures paid for be scientifically established as efficacious.

A government plan as compared to a mandated plan will have tougher controls on the growth of hospitals and other institutional facilities tied to reimbursement, and there will be a push for measures to prevent illness. As the high cost of half-way technologies that treat the symptoms of the disease are highlighted by the national plan. I would expect the plan to be directly involved in pressing for, perhaps helping to pay for. research to prevent or cure the disease. A concentration of cost will push the plan into supporting the greater use of paramedicals where they can give less expensive but adequate care, group practice prepayment plans, and other organizational changes that may improve the delivery of medical care and help to control costs.

The national medical plan will need to be concerned about how its methods of payment affect the behavior of medical personnel in institutions-it will need to be concerned about its influence on the production and geographic distribution of specialists and primary care physicians. The plan will be concerned not only with payment but with how best to organize care in under-served rural and central city areas. It will have to be a health plan, not just an insurance plan.

The general point is that the extent to which national health insurance takes responsibility that goes beyond bill paying will be greater under a direct gov ernment plan than it will be in a mandated plan. If the desired role for the insurer is to go beyond protecting the beneficiary against medical costs to taking responsibility for the safety, efficacy, and quality of services, their cost, and the actual availability of covered services in a particular geographic area for all covered people regardless of income or race, it is more likely to come about under government.