conceivably, I don't know what that would be, $500, $800, $1,000 a month.

Did you look at that question?

Mr. AARON. We did go into the treatment of spouses and recommended some modest changes in the way in which nonworking spouses are treated under the social security system. But I will assert that the most devilishly complex and continuously confusing aspect of the Advisory Council's work concerned the treatment of the family under social security.

There was no issue that caused us so much difficulty as that one. It might be worth saying why. There were a number of suggestions that came before us that on the surface seemed to make sense, which by themselves were clearly desirable. Indeed, a number of them we felt were highly promising for the modifications in the social security system.

But in almost every case, the longer we looked, the more certain one could become that we would come up against some unforeseen implication, some twist, some consequence of what seemed to be a desirable principle that we really didn't want to accept, that didn't seem to be desirable.

In the end, we proposed two changes in the treatment of the family. We suggested that on divorce after 10 or more years of marriage the earnings records acquired during marriage should be divided between the spouses for purposes of retirement benefits only.

We recommended that when one spouse dies, the other spouse should be entitled to inherit the earnings record of the decedent and that benefits would then be computed on the basis of the combined record so generated.

Senator NELSON. I don't follow that. The spouse now who has no earnings record gets half of whatever the spouse who has an earnings record gets. If the spouse with the earnings record dies, doesn't the surviving spouse without the earnings record get the full 100 percent of what the deceased spouse would get as a single person?

Mr. AARON. That is correct.

Senator NELSON. What are you saying?

Mr. AARON. In the event the surviving spouse had not worked at all during his or her, usually her, lifetime, the benefit would be unchanged. In the event, however, that the surviving spouse had earnings credits-

Senator NELSON. Oh, you would add those on top.

Mr. AARON. That is right. The purpose would be to make sure that widows derive something extra after retirement from having worked.

Senator NELSON. I am sure you see the kind of trap involved there, however. At the time social security was adopted, a very small percentage of married women worked. So the system decided to make a 50-percent contribution to a spouse who didn't work. So you make a 50-percent contribution to the spouse who doesn't work, but the system also provided that the spouse who did work could take the 50 percent or the benefits that spouse earned, whichever one was greater. So you start doubling up on the cost of the system if you do that.

Mr. AARON. The point you are raising about cost is also one that we worried about a good deal. We were careful in our recommendations to make sure that we presented a plan that did not increase the overall cost of social security. So although I have described a couple of the benefit liberalizations the Council recommended, the overall package did not increase program costs at all.

Senator NELSON. Thank you. Go ahead.

Mr. AARON. I think there are a couple of other points on the longer term modifications in financing that might be worth mentioning. One is to reemphasize the point that the modifications in the use of payroll taxes in the financing of medicare and in the financing of cash benefits programs would result in both short- and long-term actuarial balance for the cash benefits programs, something which present law does not at present provide.

An additional recommendation of the Council was that it would be desirable not to have the ad hoc base increases scheduled to take place and go into effect. It would not be possible to modify those ad hoc base increases at all, of course, unless additional financing was provided because, as Mr. Driver pointed out, under current financing arrangements, every dollar now flowing into the system is required for the foreseeable future.

But if the Council's recommendation for moving a portion of the medicare tax over to the cash benefits programs was adopted, then it would be possible to also begin to move the wage base back to the same relationship to covered earnings as obtained in 1979. Senator NELSON. The same base?

Mr. AARON. The same relationship to earnings. The base goes up each year by the same rate as earnings.

Senator NELSON. You mean the so-called ad hoc base increase you could drop. Is that what you are saying?

Mr. AARON. But you would retain the automatic increase in each year.

Senator NELSON. All right.

Mr. AARON. The base would rise with wages.

I think it is worth stressing that what I have described so far are principles that the Council recommends, and they are far more important than the specific numbers or details of the proposal that I have set forth.

For example, the Congress might feel it was possible to finance some part of medicare with general revenues and thereby to reduce somewhat the aggregate payroll tax from 6.65 percent scheduled in 1981. But it might not want to go as far as the Advisory Council did.

The Council considered partway steps, for example, and endorsed one which would have resulted in a smaller infusion of general revenues into the medicare system.

Finally on the longer run issues, the Council explicitly considered and unanimously rejected the value-added tax as one of the sources of revenue for social security. The members felt, again unanimously, that introducing such a tax during the present inflation would make no sense, and that at any time, the VAT would be less equitable than the income tax because it provides no exemptions and deductions and it lacks the close relationshp to earnings

which makes the payroll tax attractive as a support for the earnings-related cash social security benefits.

I would like to turn to the state of the trust funds, the shorter run issues and the ones dealt with by interfund borrowing.

Since 1940, the social security trust funds have served as contingency reserves to prevent the need for raising taxes in the event of recession. The Council considered and rejected arguments that this longstanding policy be scrapped in favor of an effort to build up a large social security trust fund in order to increase national savings.

The social security trust funds have served admirably as contingency reserves during all of the recessions following World War II through what I call the great recession of the mid-1970's. That recession seriously depleted the social security trust funds, necessitating the substantial increase in tax that was enacted in 1977. Furthermore, it is now apparent that the American and world economies are destined to experience more serious and possibly more protracted instabilities for the foreseeable future than were experienced before 1974.

It is important that the financial health of social security not be held hostage to such economic events as a doubling of petroleum prices which can trigger recession in the United States and the rest of the world.

To deal with this heightened instability, the Council put forward a number of important recommendations. Together with the plan to shift some of the medicare payroll tax to OASDI, which would assure an adequate flow of revenues during the period with economic fluctuations no greater than those experienced before 1974, these other recommendations would protect the social security system against even more severe fluctuations.

First, the Council unanimously recommended merging the old age and survivors insurance trust fund and the disability insurance trust fund, although separate cost analyses of those programs should be continued and published.

When you spoke to Mr. Driver about that possibility, it was clear that you had reached a similar conclusion to the one that the Advisory Council proposed. Such a merger would obviate the need for borrowing between these two funds, which would otherwise be necessary and desirable, during periods such as the present when one fund is insufficient and the others are more than adequate. Second, the majority of the Council recommended general revenue payments to the social security trust fund to compensate the trust funds for revenues lost during periods of excessively high unemployment. This recommendation is similar to the one advanced in 1977 by the Carter administration.

Third, a majority of the Council also recommended that the trust funds be authorized to borrow from Treasury if reserves fall below about 3 months' payments of benefits. But in the event of such borrowing, the repayment of the loan should begin automatically when reserves reach about 5 months payments, and there are some other conditions that I describe in the testimony.

I believe that the bad economic news of the past several months increases the attractiveness of these proposals. The administration now projects that the OASDI trust funds will lose revenues

through 1984 based on economic assumptions underlying the budget. These assumptions were not chosen for their realism or even their plausibility, but rather to express progress toward the goals of the Humphrey-Hawkins Act.

Other economic forecasts of what is most likely to happen paint an even bleaker picture, with higher unemployment, suggesting larger resulting trust fund deficits. In a footnote I describe some of the alternative forecasts floating around. Just for evidence on the fragility of those proposals, the DRI projection, which you cite in your blue book, which called for 7.8 percent unemployment in 1981 and 7.3 percent in 1982, that projection being made in January, was revised in February to project 7.5 percent in 1981 and 6.9 percent in 1982.

That is good news. I hope that the revision doesn't come in March that carries bad news. But based on the fluctuations of economic events well beyond the control of economic forecasters, I am afraid we can't rule that out.

Senator NELSON. It is always easy to project good news. It is when reality catches up with one that the projection doesn't look so good.

Mr. AARON. I must confess the forecasters have been much more honest in recent years. They do publish post mortems on the relationship between their forecasts and what has happened. And my purpose is not to downgrade the abilities of economic forecasters, which I think are getting better and better, but rather to underscore the increasing importance of events beyond their control, external to the United States, that affect economic conditions.

I think that the interfund borrowing authority is desirable as a short-term measure and may even be sufficient to keep everything balanced through the immediate future if the administration's assumptions are not unduly optimistic. But as frequent revisions of forecasts and a depressing past record of inaccuracy by most forecasters make clear, we can't be sure.

For that reason, I think structural reform along the lines proposed by the Advisory Council is in order.

Let me conclude by saying that I don't believe that any of these recommendations for reforming social security financing should shield the entire system from a thorough, ongoing reexamination. We may well conclude that some benefits are overly generous and should be scaled back, just as we may conclude that some benefits remain inadequate and must be increased.

The Advisory Council undertook such an examination and made some recommendations of both kinds within an overall self-imposed constraint that its recommendations in combination should not increase program costs.

Debate about the desirability and timing of such modifications in social security should and will continue, but while that debate proceeds, I think no good purpose is served by causing needless worry for the millions of aged and disabled for whom social security is the most important and, in some cases, the only financial support.

I also put it another way, which is I don't think that social security should experience the perils of Pauline whenever economic

instability or cartel collusion boosts petroleum prices and plunges the United States and the world into recession.

That is the conclusion of my formal remarks.

Senator NELSON. Thank you very much for your very thoughtful testimony. We appreciate your taking the time to come over and present it today. I will at my leisure examine closely the recommendations of the Advisory Council, which I have not yet had the opportunity to do.

Thank you very much.

Mr. AARON. Thank you.

[The prepared statement of Mr. Aaron follows:]

STATEMENT OF HENRY AARON*

Mr. Chairman, thank you for the opportunity to present to the Subcommittee on Social Security of the Committee on Finance a summary of the recommendations of the Advisory Council on Social Security concerning the financing of social security. The Council consisted of thirteen members representing business, labor and the public. It met over a period of 18 months, held public hearings, and submitted its report in December 1979. That report dealt comprehensively with the social security system. Unless the Committee has specific questions for me on other matters, however, I shall confine my testimony to issues relating to the financing of social security benefits.

Before turning to those recommendations, I should point out that the current forecasts of social security revenues and expenditures differ in important respects from those available to the Advisory Council. In my opinion, the revisions of these forecasts make a number of the Council's recommendations even more attractiveand urgent-than they were before. Because of these revisions, the specific form of the steps necessary that I set forth in this testimony to implement some of the principles that the Council recommended differ somewhat from illustrative examples contained in our report.

My testimony is divided into two main parts. The first presents the Council's recommendation for paying for hospital insurance with earmarked general revenues, thereby ending exclusive reliance on the payroll tax to pay for social security. The second describes a number of recommendations that would make the social security system less vulnerable to economic fluctuations than it is at present. Throughout its deliberations the Advisory Council was aware of, and deeply concerned by, the growing popular concern about the financial security of social security benefits. It was the Council's unanimous view that benefit obligations to all who are now retired or soon will retire will be and should be met in full. Council members differed about how and when the benefit structure should be modified for those who will retire many years hence. But, we can see no value to causing the disabled or those past or near retirement age needless worry about the safety of their benefits. These benefits will be paid, and it should be made obvious to those who now are or soon will be beneficiaries, that benefits will be paid.

Confidence in the safety of social security benefits is fragile and cannot withstand announcements every couple of years that the trust funds will run dry unless some new action is not immediately taken to deal with some new and unforeseen contingency. It is imperative for the peace of mind of beneficiaries now and in the future that the financing of social security be restructured so that it can withstand the kinds of economic fluctuations that to everyone's regret the nation now faces. With that principle in mind, the Advisory Council structured its recommendations for reforming the financing of social security to accomplish the following objectives. It adopted, the Council's recommendations would:

Provide financing for hospital insurance from general revenues,

Permit payroll taxes to be cut from 6.65 percent in 1981 to 6.0 percent and held at that level for the next twenty-five years,

Prevent the problems of inadequate revenues during the course of a recession and the attendant and needless concern that such shortfalls cause to beneficiaries,

*Henry Aaron is Chairman of the Advisory Council on Social Security and a Senior Fellow at the Brookings Institution.

The views expressed in this statement do not necessarily reflect those of Brookings staff members or the officers and trustees of the Brookings Institution.