Social Security/Medicare Trustees

History of the Boards of Trustees and the Public Trustee Positions of the Social Security & Medicare Trust Funds
Transcript of Policy Symposium


MS. JAFFE: If I can entice everybody to sit down, we can get the program going please. If we can get the people in the back to sit down, we can get going here.

Good morning. I'm Susan Jaffe, one of the two public trustees of the Social Security and Medicare Trust Fund Boards. On behalf of Mary Falvey Fuller and myself, I would like to welcome you here today to what I'm sure it's going to be an interesting, productive and lively dialogue.

Before we get into the public policy implications of the projected build-up of the trust funds, I would like to welcome Dorcas Hardy. Dorcas became the Commissioner of Social Security in the summer of 1986, and before that she had served as assistant secretary for Human Development Services at HHS. She has a long and extensive background in health care policy issues.

I could give you a long list, but you probably know them better than I do. One of the things I have been impressed about Dorcas with is the fact that she spent an enormous number of days and hours traveling, all around the country, meeting with various constituency groups and beneficiaries, trying to explain to them what actually is the Social Security system and what their benefits are.

To that end, she has introduced a new product. I hate to call it a product Dorcas, but it's called PEBS, the Personnel Earnings Benefit Statement, an opportunity for each individual to learn what their anticipated monthly benefits will look like from Social Security, based on their own earnings. The interesting part about this is that I've found a number of people in the private sector who said, "How come my corporation can't do this for us?" So here's a good example of something that started in the public sector. I suspect it's going to get picked up in the private sector.

I can't let this moment pass without publicly thanking Dorcas and her really fine staff, for all the help and support that they have given to the public trustees over the last few years. With that, I would like to introduce Dorcas.


MS. HARDY: Thank you, Susan. Nice to be here. I'm please to have the opportunity to at least welcome you all and start off the second Public Trustees Symposium. The first one was extremely successful on the projected build-up of the trust fund reserves. Before I share some of my thoughts on build-up and reserves and what everybody else seems to call surpluses, I wanted to also publicly thank Mary Falvey Fuller and Susan Dembo Jaffe for their excellent leadership.

They have initiated this research study. It took a while. We got it off the ground and I believe that they have performed a great public service by bringing together a variety of individuals, very distinguished, who are very interested in the kinds of implications of the trust fund build-up. So thank you to both of them because they have had that vision and foresight to put this kind of thing together.

They are also, I think, making a vital contribution in what Susan referred to as education of the public. We need to broaden the public debate and the understanding of the trust funds issue. One of my goals, as Susan pointed out, since I began as Commissioner has been to really have -- to spur, I guess, more comprehensive discussions, public discussions about the nature and the future of Social Security.

Since modern science has not perfected a crystal ball with which we can foretell the future, I continue to believe that we should move in I guess we would say the old-fashioned way of discussing key issues, looking at policy options, reaching national consensus and moving forward.

The public trustees have carried out what I think has been ground-breaking work in public education, both -- not only public at large but also specialists throughout the field and talking about alternative economic and fiscal scenarios that will face the Social Security program in the next century.

So we have been able to provide not only but some budgetary support to the public trustees and I think we can continue to do so by publishing proceedings and assisting as the symposia go forward. I think the publication of the document that comes out of these will really be a benchmark in the beginning of future discussions about Social Security financing.

I think there is also, as I've said often before, that there's no better time to start to think about the long-range future, because the vibrant fiscal health of the Social Security system today has opened that window of opportunity for all of us to look at the future strategies and to assure that there is economic security for future generations of retirees.

Clearly, most of us in this room remember and some more vividly than others that that wasn't quite the case a few years ago. So we do have that opportunity now to think about a blueprint for the future. We have clearly changed the debate from minus $10,000 a minute to a plus $76,000 a minute in the trust fund's fiscal economic growth. I think that that has clearly been very positive.

The increases in the economy as well have served the trust funds extremely well and I think that because it's in such good health it's an opportunity for all of us to look at what is the future and how do we make sure that it's what we want it to be.

I think we have looked at several critical questions and you should be continuing to raise more. I don't think any of us have all of the answers, but what level of resources will we as a nation be able to devote to 21st century retirees whose numbers will grow dramatically in relationship to the working population. How do we equitably divide resources between generations?

Do we ask working Americans to give up more of their income in order to ensure adequate support for retirees. Will the baby boom generation be willing to pay additional amounts into the trust funds, if necessary, to ensure their own retirement as well as that of the current retirees instead of relying entirely on the future generation of workers?

It may surprise you to know and I'm sure only some of you here to know that already current workers are seeing about 13 percent or so of their current Social Security tax bill being devoted to their own future retirement. So clearly the system has changed somewhat.

They're complex questions and I believe that we need to think about them and we need to begin to answer them now or else we'll be faced with the proverbial rock or hard place, if you will, option to re-think our commitment, which is not acceptable to future retirees or to collect, if it were necessary, substantially higher payroll taxes for Social Security. I think we should be -- we should avoid being caught in that trap.

Susan talked about my travels across the nation and I clearly have spoken with a lot of younger workers as well as current retirees. They have convinced me that we are all becoming extremely sensitive and I guess positive, to discussions about the future build-up of the trust funds, how they will be invested, the raising of revenues to finance our future retirement benefits and their concerns should not be overlooked. They need to be discussed and I think it's all very healthy.

Making our task, I think, more difficult are the assumptions and the variables that we all use that if they are ever dramatically changed, they can clearly have a dramatic impact on the future shape of the Social Security program. So any changes in any of our excellent assumptions clearly are something that we all are also concerned about.

So I think in examining the trust fund build-up, we need to also place Social Security in a broader context of a national retirement income security system, if you will. That is difficult because our national retirement income policy is really very ad hoc. There is not one at the moment, and that clearly is to preserve Social Security, provide for employer pensions, encourage some form of individual savings and retirement planning. But how specifically that happens is not as clear, I think, as some of us would like.

So I also believe that because of our success of the personal earnings and benefit estimate statement, clearly we have seen that the American public is both ready, willing and able to think about how the nation will provide for a greater number of retirees in the future.

Just since the 4th of August we have had nearly three and a half million people in this country ask us for their personal statement. They have -- and we have responded to almost one-third of them in a very timely fashion, and that system is continuing to operate quite well. But as we -- I don't think that number will decrease rapidly and I am expecting that there will be a lot more people who will continue to ask us for "what about me, what can I estimate, what might I expect from Social Security."

So I think it's a demonstration that people are willing. They are eager to plan for their own retirement. They need the tools that we can give them, not only on the fiscal economic side and the big picture, but also on the personal side. I applaud you all for not having a hand-wringing or desperation attitude about all of this issue because there is plenty of time. There's lots of things to talk about and I believe that we do currently have that window of opportunity.

The symposium, the first one, today's symposium is a major step in beginning all of that discussion and creating a vision about what that future retirement income policy program planning should be. I expect as you all have done before, that you will continue to have views about all of this, and that you will share them with each other and that you will have a very successful opportunity here.

Especially thanks to both Mary and to Susan for their vision and leadership in this area. With that, I wish you well and I expect that you all will have the answers to exactly how much we'll build and what we all will do with it. I would like to -- my personal bias is they are reserves. They are not surpluses. Thank you.


MS. JAFFE: Thank you Dorcas for your thought-provoking ideas and for the terrific job you've done. As Dorcas mentioned, back in June we convened a symposium on the technical aspects of the build-up of the Social Security trust funds. This was based on the Brookings and Lewin-ICF studies.

We'd intended to keep that discussion solely focused on the technical aspects. But as many of you who were there know, it was impossible to exclude completely any discussion of the public policy implications. Issues ranging from the use of the surplus to the impact on national savings were raised. So today the entire program will focus on the public policy issues.

Before get into the meat of the symposium, I want to give you some brief history about our role as public trustees and about how we came to focus on the subject of the unprecedented projected build-up of the trust funds. My apologies to those of you who have heard this before, but the staff has convinced me that it will be new to some of you.

Back in the fall of '84, when Mary and I became the first public trustees, the papers were filled with headlines with proclaimed the looming bankruptcy of Social Security. Most people, both retirees as well as active employees, did not expect to collect their Social Security checks. There was a crisis of confidence in the system.

As the first two public trustees, Mary and I wanted to help restore that confidence. What could we do and what was expected of us?

The National Commission on Social Security in '83 had recommended that two members of the public be added to the board, for the dual purpose of inspiring more confidence in the investment procedure and helping to assure that the demographic and economic assumptions of the program would be developed in an objective manner.

Turning to the Social Security Act, we learned our duties, to develop and issue the Trust Fund report, and to review and, if appropriate, recommend changes in trust fund investment policies. Mary and I were new at this and we began asking a lot of questions about current investment policy and practices.

Then we became involved in the preparation of the '85 Trustees' Report to Congress. That report in '85 projected that the trust funds would accumulate substantial reserves, as we move toward the next century. We recognize that the size of the accumulation was unprecedented, and almost no one knew about it.

Surely the general public would become a lot more confident in the system if they believed that there would be money there when they retired. So in March of '85, we proposed to the board that we undertake a project to study investment policies. Our project included the following: to evaluate current policy, to review the management of the trust funds from both the statutory and administrative point of view, to consider alternative policies, and to recommend to the board what future policies should be.

A great deal of the study could be done in-house by the SSA and Treasury experts, including the history of investments, current operations and practices and principles that had been followed. However, we recognized that we needed outside consultants to examine the economic and fiscal implications of the projected build-up.

Well, that process which we started back in early '85 finally resulted in contracts being let to Brookings and Lewin-ICF in early 1987. Those studies were completed this spring, and presented to the board this past April.

Finally, I feel compelled to reflect on just how far we've come. In '84 we had a crisis of confidence in the system. Social Security was going broke. No one under 40 believed that he or she would ever see one dollar's worth of their benefits. Now I pick up the papers and I'm bombarded with good news. We currently have surpluses which are expected to continue well into the 21st century. As an aside, I might add that the build-up will occur only if Congress keeps their hands off the money. It's a very tempting pot.

Now I'd like to turn the program over to Mary, who will provide you with some details about the current investment practices.

MS. FULLER: Good morning. It is a real pleasure to see all of you here today, because I think we really have an opportunity here to explore some issues that are complex, subtle and very sensitive in terms of how the variables interact with one another, and at the same time have far- reaching implications in terms of how our economy will be effected by the various policies and acted not only for Social Security, but more broadly in the fiscal management of the tremendous deficit that we have.

Having been on the National Commission, I had the point in experience of the characteristic we all know of democracy, that we aren't very good at using a crystal ball and taking our windows of opportunities and getting everything all laid out before the crisis comes. But nonetheless, as a believer in the ideal and as someone who has really tried to work toward it, we're hoping that today's discussion will spark dialogue that will continue and that we will be able to get some issues on the table to feed the policymaking process, so that we will have as much substantive input as soon as possible into the dialogue, in the hopes of making really sound choices as we go forward.

My purpose in outlining the current investment policies is not to go into details or to evaluate or analyze them in any way at this stage. We do have the internal work done by the SSA and Treasury staff, to which Susan referred, and which we can make available to you in due course if any of you want, a more detailed explanation. Please let Dalmar know. The information is in almost final form and we created it for the benefit of people so we'd be happy to make that available to you.

What I just want to do is to make sure that for those of you who don't know how the trust funds are invested, that I'll just briefly tell you what their invested in, and what some of the principles are so that you'll have it as a context for listening to and debating and dialoguing about the possible scenarios and alternatives that you will be dealing with today.

First of all, the trust funds are required by law to be invested in issues either issued by or guaranteed by the United States Government, and that's the Federal Government. Up to 1960, there was heavy investment in the marketable government securities. Starting in 1960's there started to be a preference for special obligations, which were securities created for the sole purpose of being invested in by the trust funds.

Since 1980, no new investments have been made in marketable securities, but only in the special obligation securities. Now what were these? These were designed to provide security to the trust funds. They have the unique characteristic of being redeemable at par prior to maturity, so that there is no risk of lost of principal if in fact the trust -- this money is needed to pay benefits ahead of the maturity date.

The interest rate. That is a formula now. It is calculated as the average of all government securities, marketable securities with a call -- due or callable date of four years or later. So again this was an attempt to recognize that these trust funds are essentially long-term funds, and to create a rate that approximates the market rate, but that is a totally predictable rate and not one that is subject to the judgement of anybody making these investment decisions.

The maturities were by administrative practice set to be distributed over 15 years into the future. Again the principle here is to design the maturities to reflect the anticipated needs of the trust funds, so that when the money comes in throughout the year, it's invested first in short-term certificates of indebtedness. All of these have a due date of June 30th of the coming year.

So that everything that comes in every month now during 1988 through June 30, 1989 will be invested in a short-term certificate of indebtedness, with a due date, maturity date of June 30, 1989. Then on June 30, 1989 these are rolled over into a long-term note, special obligation note, calculated with the rate that I just described and given a maturity that spreads out over 15 years, the various maturities of that investment, coupled with whatever else is within the portfolio at that time.

Now when we are in a period where the outgo is exceeding the income and securities have to be redeemed, these are redeemed at par as I mentioned, and they take the earliest maturity date first and then for all those securities maturing on the same date, they take the lowest interest rate first. So that's how that's organized, to actually make the funds available to pay benefits.

So that's basically how they're invested. When we undertook our work, first to figure out what was being done now and why and to look at possibilities for the future, we articulated, we went back into the documentation and made an attempt to articulate the principles which underlie these selections of current investment policies and which would underlie any choices that might be made out into the future to change those.

We identified four principles, one of which is non-intervention in the private economy. This is why we are confined to the issues either issued by or guaranteed by the Federal Government. So no state, no local, and no direct intervention in the private economy. Another was no active management by the managing trustee.

The Secretary of the Treasury is the managing trustee. He has the sole discretion to manage the trust funds however it's done throughout the organization. So it's been always very important to the Secretaries of Treasury that they be in a position where they're not actively managing these funds or they're not subject to the debate, controversy and liability involved in doing that. So minimum discretion as been one of the principles as well.

The other -- the third one is investment only financially secure instruments, and this is why the backing of the U.S. Government has been an intrinsic and essential part of all the investments. The other is a concept called neutrality, and this comes from the fact that the managing trustee is both managing the general fund and managing the trust funds. Since the trust funds are invested in obligations of the general funds, it's very important that as policies are developed, there isn't undue advantage or discretionary advantage given either to the Social Security trust funds or to the general funds.

So that in setting interest the interest rate as a formula that was all part of this -- implementing this principle, instituting the maturity spread out over 15 years. You know, you're not going to create maturities based on what you think the interest rates are going to do, so you maximize the earnings of a trust fund because by doing that, you'll be disadvantaging the general funds.

Then the whole process of redemption, where you take the earliest date and within that the lowest interest rate. The redemptions are only to be made prior to maturity for the purpose of supporting the Social Security program. There are no redemptions to be made for redeeming a lower rate interest rate security, in order to get a higher rate because market rates today are higher than they were when the securities were issued.

So these are the four principles that have guided the investment policies up to now and that we expect would guide any future changes to those that may come out of this dialogue or anything coming to the future.

So given that context, let's proceed and I'll turn the podium back to Susan to introduce our first panel chairman.

MS. JAFFE: I just hope everybody can hear me down here. I have a couple of housekeeping things I want to mention. I apologize for the room. Washington seems to be sold out this week, but I hope everybody can hear us and see us.

Secondly, if you notice in the program that we are scheduled for a break at 10:00. I'm not sure whether we're going to be able to have the break at 10:00 or not. We have a problem with one of our panelists in the second panel, a scheduling problem, and we'll let you know as soon as this panel is over. But don't everybody just jump and walk out of the room, because we'd hate to have you miss the second panel.

Thirdly, I want to just mention that lunch will be served promptly at 12:30 down the hallway here. I think it's called the Mount Vernon Room. It's going to be a buffet and all the panelists will be divided up around the room and it's open seating for everybody else. If you want to know who is seated where, there will be somebody at the door to tell you.

Finally, in listening to some of the dialogue this morning, before the symposium started, I have heard a number of people say this is getting to be boring, and we really want to get on with it. So I would like to encourage both the panelists and the presenters and the participants to please be as thought-provoking and as outrageous as you'd like to be, because we think this is a good forum to move the dialogue another step.


MR. ENOFF: If you could go to one of the microphones please and we'll ask you to identify yourself. Keep your comment as brief as you can and allow us to continue. Right here is our first --

MR. HUSTEAD: I'm Ed Hustead. Dr. Nordhaus, in his comments on the two papers, said he was not surprised about the fact that they got the same results, because if you use the same assumptions you get the same results. The Brookings paper and your comments noted that you were constrained, I guess, to use 2B assumptions, and you pointed out that there were two particular problems you had with the 2B assumptions, the real earnings rate -- the real interest rate and the real salary growth.

But you also said that if you had used another set, you'd get about the same conclusions, the same general conclusions. I think the Lewin-ICF report didn't really deal with this question, if I read it, as to the validity of the 2B assumptions. So let me throw it to Mr. Anderson and say do you agree that 2B was most appropriate to use and second, if not, would another set of assumptions have produced about the same answers?

MR. ANDERSON: To answer your first -- well, to give you some background, the base case simulation that we would develop with the model would have departed from the 2B assumptions, but not dramatically, and we would endorse the detailed criticisms in the Brookings report of the 2B assumptions. The more recent trustees' assumptions, by the way, are somewhat more reasonable, I think, in both Brookings' and our estimation, than the ones that we addressed.

To answer your first question, were they appropriate, I think the answer is yes, because the issue that we were to discuss is what are the effects of this trust fund accumulation which is being projected in the 2B assumptions on the economy, slightly different assumptions that would be more realistic, which still show us a large trust fund accumulation and the qualitative conclusions that we came to would have been the same.

MR. ENOFF: Okay thank you. While the next question is being asked, I might ask the panel to begin assembling, the next panel to be starting to assemble in the next -- over this direction. Is that good? Okay. The next question?

MR. FRIEDMAN: Yes. I'm Ben Friedman from Harvard. This is a question for Joe Anderson. I was quite struck by your Table 2 on interest rates, and in particular the notion that the difference between these two simulations that you present results in an approximately six percentage point difference, if I understand correctly, in the real interest rate on Treasury bills. Did I understand that right?

MR. ANDERSON: To be sure, this is a -- in the year 2020 --


MR. ANDERSON: --to focus on a 5.7 percent reduction, not percentage point.

MR. FRIEDMAN: Oh. 5.7 percent. So the difference between six percent and 5.70, some such thing?

MR. ANDERSON: That's right.


MR. ANDERSON: That's correct.

MR. FRIEDMAN: That answers the question. Thank you.

MR. ENOFF: Very good. We have another question right here, I think?

VOICE: This is addressed to Mr. Anderson. First, following up on the previous speaker, I only glanced hardly at your report, but one thing that struck me is that I got the impression that the condition of the trust funds would be very, very sensitive to relatively small changes in the interest rate, and I wondered if you would say something about that.

The second question I'd like to put to you is whether you agree with Gary Burtless and the Brookings people, that a

-- I'm from the labor movement, so any increase in payroll rates is something serious for us. But nevertheless, a relatively moderate increase in the payroll tax, whenever the surplus -- whenever the long-term condition is deficit, would actually return the system to balance.

MR. ANDERSON: I'll answer the two questions. The system is reasonably -- the first question was the sensitivity of the trust fund balance to relatively small changes in the real interest rate, and in these projections, in these simulations it is relatively sensitive because the interest rate changes the most at the time when the trust fund balance is the largest, and where interest earnings constitute the largest part of total Social Security revenues.

To give you a more detailed answer, we did simulations of the model holding the interest rate constant at the trustees' projections, and in one simulation and in another simulation with everything else the same except permitting the interest rate to be solved by the model, so we can actually give you the figures as to the effects on the trust fund. That's in the complete report.

For your -- your second question was would increased payroll taxes improve the trust fund balance; is that correct?


MR. ANDERSON: The answer in a mechanical sense is yes. In the simulations that we did this -- with the model, we did not have any feedback effects of payroll taxes on labor supply. So that an increase in payroll tax rates is being applied to the same total earnings tax base. So it results in a proportional increase in tax collections.

MR. ENOFF: Thank you. I'm going to apologize to the other questioners. We are just out of time, but Mr. Burtless and Mr. Anderson will be here throughout the day and please ask your questions at one of the other sessions or during the lunch time, and I'm going to ask the second panel to come forward. Let's show our appreciation to these two presenters.


MS. FULLER: We had discussion in the last panel about the questioning of our choices when we commissioned these studies, of requiring the two contractors to stick with the 2B assumptions, and to stick with the benefits and taxes in current law. The reason we did that was to try to set an analytical context that would be common, so that we could really see the two respective contractors' views of how the different variables interact under a common set of assumptions.

The reason we're trying to do is to try to set, to the extent one can, an orderly analytical framework to surface issues, knowing that this is a very wide ranging and complex subject. So this will be -- this panel coming up now is our first chance to have a dialogue that is against this backdrop, unconstrained with respect to the opinions and assertions that our panelists and people in the audience may want to make, concerning the effect of this trust fund build-up on the fiscal issues related to our economy and the Social Security system and its issues in the broader context, in the overall management of our economy.

So we welcome that opportunity. What we're going to do, acknowledging that we had an overly short question period that last time, we are trying to move the program along and at the same time may feel free, where we have for example, half an hour break followed by an hour and a half lunch, to convene earlier or to somehow make room for as much discussion as possible as we move through the day.

The objective is to have a lot of input from you, and we will do everything we can to extend the discussion periods as we go through.


Thank you. That gets me off to a good start, and I will tell you I will be quite unconstrained, as it was suggested.

The first thing I'll remind you of, and Joe Anderson did suggest it, what you have in the Social Security trust funds, all of these surpluses, has virtually nothing directly to do with the benefits that will be available to the baby boomers in the year 2020, or to any other retirees.

The amount that retirees can get, the earnings that they can actually receive, is based entirely upon the amount that is produced by the people working when they are retired, plus what can be earned on that fine investment. The amount that people can, who are working when they are retirees, the amount that they produce depends on the number of them and depends upon the capital with which they are endowed.

That last really contributes a major amount to the focus of much of the discussion. We keep saying that it is very important to save, to have wealth, to have capital. Then they can be used to provide benefits for people in the future. The problem with the discussion is that -- and this is my central theme as perhaps was suggested to you -- that somehow we got caught up in the terminology, and we forget that the words we're using have almost nothing to do or very little to do with the economic concepts with which we're concerned.

I fished out some figures some of you may be familiar with that I developed in the past and I'll be repeating. The total capital of the United States economy I estimated as 1981, was something over real capital, something over 20 trillion dollars. Of this, the amount that we usually talk about, relating to saving, relating to gross private domestic investment, would come to well under a quarter.

Over 75 percent of the capital is Government capital, it is household capital, it is capital in terms of what is invested in research and development, and most important, the human capital, the education, the training which makes workers productive. I think more and more of us are being exposed to that idea. We see it as a big issue in Business Week, which some of you may have seen last week, on the huge importance of human capital.

You hear how we can't compete presumably with the Japanese and the South Koreans and other people because our kids don't know any math and our kids can't pass the exams to be a teller in a bank, let alone know how to operate computers or do any kind of sophisticated work.

There's the issue, there's what's a matter of major concern in terms of providing productivity for the future. Now the -- the studies have you have -- I hate to sound nasty and while they're well done technically, I have some objections I can offer. But it's curious that despite all this awareness that most of us have, and people probably say "well, we agree. Of course human capital is important. Saving is much more than the saving that is officially measured." There's virtually no recognition of that.

There's reference over and over again to national saving. You know what national saving is defined as? Most of you do. I'll remind you. It's personal saving plus corporate saving plus Government saving. How each of these are defined and what in the world do they have to do with the saving or with the investment which economic theory talks about? I'll tell you. Very damn little.

I mean take this Government saving. Government saving is simply the difference between what the Government takes in as revenues and what it pays out. But saving is equal to investment. What about all that the Government spends on infrastructure, on roads, on bridges, on education, on research? None of that is counted as Government saving. It's presumably Government consumption. Everything the Government spends is consumption.

I can go on with all that's wrong with the personal saving measure, all that's wrong with a corporate saving measure, and some of that is outlined in my prepared remarks. So the first problem is that when we talk about saving, increasing saving, validating the Social Security surpluses presumably in order to have saving, we literally don't know what we're talking about. We're not talking about the saving that is meaningful in an essential way, meaningful to the economy and meaningful to providing for the future.

Now the next quarrel I have is with the notion somehow implicit, if not explicit in much of the discussion, that since saving as is being conventionally defined -- and I remind you it's corporate saving plus personal saving plus Government saving -- somehow that's an identity. So if you raise Government saving, which means reduce the budget deficit or validate the Social Security surplus, you're raising national saving.

Well, you know, it just doesn't follow. That depends. It depends upon what you think is going to happen to the economy, what the effect will be of taking away purchasing power from the private sector. Is that going to increase total saving? Well, if it does -- in order for it to do it, it has to increase, even in the narrow measure, then it has to increase gross private domestic investment or net foreign investment, because the rest of the identity is that gross saving equals gross investment, and gross investment is gross private domestic investment plus net foreign investment.

Well, you know, I hate to sound old. I think it was one of the -- LaSharon (?) up here. I met him last night. He says you show your generation, because I had a reference to Oscar Longer in the prepared remarks.


Well, what he wrote, that beautiful article in the [name of journal], I think is still perfectly valid. Reducing consumption does not necessarily increase investment. It depends. As you reduce consumption, it is true you free resources for investment. But whether those resources will be used for investment depends upon the balance of two effects: reducing consumption, as I said in other places.

You tell people not to buy cars, is that going to make Mr. Iacocca tell Chrysler to invest mope or invest less? Well, the old pre-Keynesians would say, and some of the new, the new macroeconomists will say well, of course. They'll go through some scenario, perhaps, of interest rates going down or about real balance effects, and they'll say therefore, you get more investment.

Well, it's not at all clear, because on the one hand the loss of consumption demand will depress investment. On the other hand, the freeing of resources and the interest rate or other mechanism will induce more investment. The general notion is that usually, unless you're close to full employment, you're more likely to depress investment by reducing consumer purchases, which you reduce by building up a Social Security surplus. You're more likely to reduce investment than increase it.

So, you know, where do we get all this, by the way, is because as I suggest in my paper, we somehow think that well, all that I'm saying may be true in the short run. It's true, of course, if you suddenly raise payroll taxes or suddenly reduce the deficit you may bring on a recession. But we're talking the long run and the long run, as I suggest, is the habitat of the natural rate of unemployment, a horrible concept to me. There ain't nothing natural about unemployment.

But somehow that's it and the fact that our notion of national rate has changed a good bit. It used to be six and a half percent or six percent to many of my colleagues and when unemployment got down to 5.2 percent, I guess they were left with something hanging out, but they figured that maybe that's temporary. That's what told us the economy was being overheated, despite any lack of particular evidence of that.

But I'm one of those guys that insist that that old four percent target was right, and that we have three and half percent unemployment in twenty years ago, I'll remind you, and I see no reason why we shouldn't aim at it again. The long run, actually, is a succession of short runs. The economy is never at any particular equilibrium and if you keep banging it over their head with Social Security surpluses and taking taxes, taxing people away, you may well run yourself into a chronically lower level of employment, and therefore less total saving.

Now let's get back close to home, then, as to what we should be doing with the Social Security surplus and what the various impacts will be on the economy. It is frequently argued that well, if we take the Social Security surpluses, build them up, validate them in the sense of having a total Government surplus in the unified budget, we will lower interest rates by stopping the drain on the capital markets in one way or the other.

Well, the first thing I'll tell you if you want to lower interest rates, there's a very simple way which most of us teach in our courses, I hope still, and that is have the Fed buy securities in the open market. You'll lower interest rates and you'll have none of the consequences of trying to drain purchasing power away from the public as a means of doing it.

That, I might add, would have the great advantage of increasing investment, of technically lowering the budget deficit because we count these nominal interest payments that we're making in the budget deficit, and it would also allow exchange rates to lower, which would tend to improve our net foreign investment position. If I had time, I would go into what's wrong with that measure, which is a really absurd one because it doesn't take into account the market value of our direct investment in other countries.

I could spend a bit of time reminding you about what's wrong. Some of you have seen some of my writings about this with our measures of the deficit. I might suggest in the growing economy it would be much more plausible to have a target not of a zero budget deficit as we measure it, but of a balanced debt to GNP ratio so as with private individuals, with business, debt simply does not grow any larger or faster than GNP.

That would imply, by the way, a deficit of some $175 billion currently, or 3.7 percent of GNP, which is somewhat larger than what we have. In fact, the current deficit, I think, is going to be running closer to 3 percent than the 3.6 percent we just heard.

But now what, you know, can we be doing to have a -- make sense of the Social Security surpluses, which are being built up, and which I perhaps would not have advocated be built up but they are being built up. Well, recognizing that to provide for the future, we have to have as much capital as we can, much ability to produce -- we're concerned about that -- and recognizing as well that a very tight fiscal policy runs you serious risks of recession, of lowering income and employment, which is the most destructive thing you can do to any kind of saving, I think I have a reasonable solution for you people, and also recognizing what I told you about what capital is.

What we should be doing is validating the Social Security surpluses and the one way that we can, reasonably surely, and that is validate them by using them to invest in human capital and Government capital and all the things that we cannot rely upon the free market economy, and which I warmly believe, to operate.

If you try to use these funds to go ahead and buy up private securities -- I've been facetious about making the Social Security trust funds the agent of socialism in the economy by buying up private assets -- if you do that, it's not really guaranteed you're going to get any more investment. That goes back to the old problem, you know, of how much trying to drive down interest rates will get you anywhere.

It also raises a problem which I think is sort of skirted, and I've had long battles on this if you go back into the professional literature. There is somehow an assumption that's in both of the models you work, that we have nice, well-behaved production functions, for those of you who want to get technical. We either conveniently assume a Cobb-Douglas ? function, which is very nice, or even a constant [name] function where the elasticity of substitution is not too low.

That generates what kind of a result? Well, it generates the result that you can keep accumulating saving, wealth if you wish, and it's true you'll drive down the real rate of return somewhat and how much you drive it down depends upon the particular parameters of the production function you assume. But there's always room to have more capital. In one of Sanderson's early editions, which I think I managed to persuade him to eliminate the phrase, you know, "there's after all always some rule that can be -- railroad track to be straightened out, and as long as anything, everything has a positive return, you can always accumulate more capital."

Well, God did not necessarily make the world Cobb-Douglas, and there's no particular reason to assume that you can always keep accumulating private capital with a given technology, plant and equipment and always have a positive return from it. If you can't, as you try to accumulate it what's going to happen is in a free economy, since businesses are not going to want to acquire plant and equipment that is not productive net, they simply won't acquire it. You have a recession.

If you force them to require it by giving them nice investment tax credits, which I'm happy to see we've gotten rid of, all you're doing is encouraging them, as I put it, to spend $100 for a piece of equipment that's worth $95 in terms of what it will return them, and yet the Government gives them a $10 payout and a credit, so it's worth $105. That's not a way to economic growth or to providing for the future.

That leads me back, then, to my central theme, that what you should be using these trust funds for and indeed healthy budget deficits approaching -- keeping, in the long run, approaching keeping the Government debt to GNP ratio constant, what you should be using them for is a huge amount of investment that the private economy does not provide for, which gives you the bulk of the capital which is related to productivity, which sober analysis tells us it's the kind of capital we really need.

All these bridges, end roads and airports and investment in research and education and training, a situation where we can get rid of one where 10, 20 percent of the population is functionally illiterate, where most of the population is, as I say, falling behind and ill-equipped for the production in a modern age.

I could, I guess, close off of one other note, which is on therefore what you should be investing in, in terms of the Social Security trust funds. One suggestion is that, you know, get away from this restraint on not only the special non-marketable Government securities, but get into private securities. I would suggest that you should -- you'd do well to have the trust funds invest in whatever is most profitable.

There are some distribution effects there. You're financing it by a payroll tax, which is largely regressive. If you were to invest in more profitable securities, higher return, you would have less need to rely on the payroll tax and more you could rely upon the earnings from the securities and that, I think, would be of some advantage distribution-wise.

But my preferred way to go is to have the Social Security trust funds, reassure the public they've got something there. They'll have real securities that they can sell, from the Government, and these real securities they're getting from the Government would be used, in turn, to finance all of the major public investment, investment in human capital, which is one way to provide for the future. Thank you.



Well, thank you and I'm pleased to have the opportunity to talk about these issues. I was sent a note when I was preparing for this session. The note was supposed to tell me that I should be thought-provoking. Unfortunately, it was garbled in transmission and it came through to me as just be provoking. So I'm prepared to do that, and I hope I won't be out of order.

My message this morning is simple. Ask not what the budget deficit will do to Social Security; ask what Social Security will do to the budget deficit. In order to get the answer right, you have to get the question right. This is illustrated by the history of man's investigation of the law of gravity.

2,500 years ago we got off to a bad start in trying to figure that out, because we asked the question "how do the planets move around the earth," which they do not do. That problem successfully resisted the best minds in history for 2,000 years, until Newton realized that the question was "how are bodies attracted to each other through gravity." Then in less than a generation, we got it worked out.

It looks to me like we are in danger of having the wrong question be the only question that we look at. We've looked at a lot of questions this morning like what influence will -- conditional on the budget deficit, what influence will the way we invest Social Security reserves have on the economy? The answers are if you hold everything fixed, nothing changes.

A first order question, I want to suggest, is what impact will how we invest the Social Security reserves have on how we think about how much money we have available to spend in the public sector? People talk about the long-term in Social Security. They talk about the effects on wages and productivity. I think that's well-advised and important for us to look at and I think we've got excellent work on looking at that.

I want to suggest that the big game is already in progress. It will be in progress for 30 years, and that we've gotten off, generally speaking, to a bad start. The first order game is will the build-up in Social Security reserves be validated, or will it in fact be offset through other spending. Everyone seems to agree. Don't spend the reserves, or at least don't spend the reserves frivolously. But it's not quite that easy. No one is suggesting that the reserves are going to be spent directly, that the payroll tax revenues going into Social Security are going to be diverted directly to other spending. It works through a more complicated mechanism. Congress first borrows the money. We know that those debts will be redeemed, that that redemption will result in the payment of Social Security benefits. I take those things as not being in doubt.

The question is what else is going to happen, and in particular, what else will -- how does what else will happen depend on how we account for and how we invest Social Security reserves. Now one track on that question is the Brookings and ICF track. Hold fixed the non-OASDI budget deficit, and ask what happens when you vary the investment policies? The answers are some things, but relatively minor things.

In effect, that says examine the feedback of the Social Security system on itself, holding other things constant. The other track, which is the track I'd like to suggest that we think a little bit about, is what if the non-OASDI budget deficit is itself influenced by investment policies?

Well, that would be first order. That would be a problem which would have an impact on the scale of the reserves themselves. Now as Gary Burtless and his colleagues point out in a chapter of a book related to the work that they've done that's reported here, it is not scientifically predictable how Congress will respond to having a budget surplus on the Social Security side and what difference that will make elsewhere.

Unfortunately, the economist's principle of free disposal doesn't apply to policy problems. Okay, the fact that it's hard and it doesn't fit into our models and we can't make scientific predictions about it, doesn't get us off the hook. We've still got to think through the question. What is likely to happen. Move 1 in this game is the 1983 changes in Social Security. Strong political support for Social Security plus a desire to have the payroll tax rate be level, plus the demographics of the baby boom, lead to a now-ongoing accumulation of Social Security reserves.

Move 2. In the law, Congress borrows the money. The question is, what is Move 3? How does Congress respond to that? I agree it's impossible to predict and it's impossible even quite to know after the fact what they did. I read it this way. There's strong pressure to keep up services and spending. There's strong pressure not to raise taxes in that position. I think it's very, very hard not to feel quite relieved that someone lent you automatically $50 billion.


Put it the other way for a minute. What do you suppose would have happened last year if there had not been a surplus? That is, how would Congress' actions have been different? I think it's hard to argue that they would have been less vigilant about taxes, or that they would have been less likely to cut spending.

The direction is probably right; that is to say, the existing build-up of the Social Security reserve probably provides at least some relief in the Congress, if nothing else. The magnitude is quite unclear. To cite -- to quote Alice Rivlin on this issue, partly right part of the time is about as close as economists ever get. So knowing the sign of this seems like at least a start on this problem.

It appears to me, though, it's a counter-factual, because it's hard to know what Congress would have done differently if other things had been different. They were off to a bad start. That Congress, in the thing that it took apparently most seriously, Gramm-Rudman-Hollings, it included the Social Security surplus as part of on its side of the ledger. That is, we get that money and now the question is how much we spend.

Suppose for a moment that that's right, that this does have some influence on the way Congress thinks about how much money is available to be spent. What's wrong with that? What's going on. Well, what's going on is that Social Security as a political logic is being used to produce funding for other spending. Now that may be bad or good. We don't know unless we can figure out, and I submit it's hard to figure out, what that other spending turns out to be and your view of whether it's bad or good depends on how you like the spending that Congress does.

But whether you like it or not, it's an important issue, because the political logic of Social Security is the strongest political logic of any system invented in modern history. It's a logic which you could refer to as righteous entitlement. I paid; therefore, I deserve. Millions of taxpayers are out there taking that pledge every day. The pledge of allegiance of the Social Security system goes like this:

I pledge allegiance to the Social Security system of the United States of America, and to the insurance scheme for which I suppose it stands. One contributory trust system, indivisible, with security and benefits for all.

Now we may view that as a wrong view of what Social Security is, because we think of it as a pay-as-you-go system. Even with partial funding it's a largely pay-as-you-go system. But their perception, the millions of taxpayers who are taking that pledge of allegiance every day. Their perception is the political reality of this system. That's what generates the politics that allows us to get a payroll tax rate that produces these revenues.

In my view, this brilliant political logic is on the verge of being abused. That is, being used to produce funds which when borrowed back are then used to be spent on other purposes. Not -- I don't regard that as a fundamental miscarriage. I don't think that was a plot against the future of the society. But it's a very natural kind of outcome in a system which otherwise doesn't have much protection against it.

If we don't validate, that is, if we don't insulate the surplus, then we are at least on the verge of committing something that looks to me like a political fraud. Again, you can be for or against that kind of spending and I'm not trying to take a position on that. I want to be quite clear about that. I'm not arguing that the non-OASDI budget deficit should be zero. It doesn't have to be zero, by the way, to validate. I think that's an incorrect use of the word "validation" in the context that we're talking about it.

What the budget deficit has to be in order to validate the build-up is whatever it would have been in the absence of the surplus, which by the way, is an unknowable. It's a political economy counter-factual. Slippery ground indeed. We can't know what that would have been, and so it's hard to find a standard against that. My point is simply that when we start -- when the first move in the game is lending the Social Security surplus to the Treasury, the Social Security system and the changes in the system at least partially obscure what's going on in the rest of the enterprise.

Only partially because it's actually a fairly small fraction of the total enterprise, about five to ten percent annually. The build-up of the reserves, as we now project it, is on the order of magnitude of one year's Federal budget over the course of 20 years. So it's not a huge fraction of Federal spending, and it would be easy to lose in the shuffle whether or not we had validated in the sense of holding fixed our other behavior as the Social Security surplus was built up.

The question then becomes how do you structure Social Security so as to enhance the prospects of validation? That is not a trivial problem. The question of how, since it's already -- we've already observed it's unknowable how much validation there was -- how do you structure the political economy of that system so as to have some impact?

I think the answer to that is straightforward. The answer starts with lend the money to someone else. Almost anyone else. Now there are other effects of that, and I take that seriously. But the first order idea is make the Congress, if it chooses to spend that money, for whatever money it chooses to spend, either raise taxes or borrow that money from some non-captive source.

Now it's true all the cognoscenti will be unchanged by that. They'll all know. We all know already. We can already do the arithmetic of how much that looks like it is. But when you force the Congress to explicitly borrow money for what it's spending, or to raise taxes for it, it seems to me you're at least being consistent with the constitutional logic that money will not be spent except when directly raised in that sort of form.

Now usually we debate investment policies in terms of the risk and return, looking at the impact on the lender. What I'm trying to suggest is that an important issue here is to look at the impact on the borrower. If you hold everything fixed, nothing will change, but it's possible that if you change the way we account for and examine and in particular hold apart from the rest of the budget the Social Security system, that you may have some impact on that.

Now Social Security, I'm not here to suggest, that the way you invest Social Security reserves is going to have a big impact on the way Congress feels about how much money it has and how important it thinks different service demands are, because I don't think it's going to have a big influence. But not a lot of other things have a big influence either, and it's at least trying to get this one, it seems to me, straight.

Social Security is a brilliantly designed political system. That it will pay benefits into the foreseeable future I think no one doubts. That it will temporarily have significant reserves now seems reasonably clear, depending on how you define reserves. How those reserves are invested is a matter of some consequence, but not principally because of their impact on Social Security.

The politics of Social Security will inevitably lead us to pay Social Security benefits. That's not the question. The question is what else we will do and what else the existence of Social Security and its reserves will have permitted us to do consciously or unconsciously. It just seems to me the argument is it's better to be explicit about that than inexplicit.

The investment policy's first order of facts will be to either facilitate or make more difficult the unconscious offsetting for the build-up of Social Security reserves. The Social Security system can, arguably, help us keep from spending the reserves twice by finding a borrower other than the Treasury as the borrower of first resort. This is going to exert only a small amount of leverage, but some amount of leverage might be of use, and it's at least worth trying to think about whether that's not the most -- the first and most important question that Social Security really puts before us.

The particular point of that is to say we can argue about Social Security as if it were something that were going to happen, starting when the build-up gets to be a really significant amount of money. What I'm trying to argue, this game is already in progress. It's a game in which it's difficult to tell where you are in, and it's worth trying to think of any way we can make clear to ourselves what it is we are actually doing.



I will duck both of those questions, but any good citizen would want to see the same good policies whoever is elected president. But any reasonable good citizen will think you're more likely to get good policies with one presidential candidate than with another. Said that far, it's hard to see how anyone could disagree.

I do not yield to anyone, not even Bob Eisner, in my enthusiasm for more education, for more technology, and for more infrastructure investment in the United States. Those are all critical objectives. We will not succeed in getting from here to sound economic future without investing more heavily as a nation in all of those things.

Economists tend to focus very heavily on fiscal and monetary policy, to ignore those kinds of investments. It is constructive to remind one's self from time to time that the reason the United States is twenty times as wealthy as India is not primarily that we've had better fiscal and monetary policies over the last 100 years. The reason why economies like Massachusetts and New Jersey -- you see I am non-partisan -- have outperformed substantially the economies of the rest of the nation, is not because they have had better fiscal and monetary policies either.

All of that said, I believe that that has almost nothing to do with the appropriate policy we should pursue with respect to the Social Security trust funds, or with the urgent need for budget deficit reduction in the United States. I believe that the the unholy alliance of the far left and far right that forms to take the position that budget deficits are not a serious problem, that you can have it all, that we can have tax cuts, more infrastructure, print more money, that we can simply have more of everything and in that way solve our economic problems is simply irresponsible, simply a prescription for the Argentinization -- this is an Al Haagism

-- of the United States.

Down that road, lies disaster. We cannot responsibly cut taxes, invest much more heavily in infrastructure, education, accept $175 billion deficits, solve the problem of the high interest rates that they create by expansionary monetary policies. That is simply not a sound basis for getting from here to the future.

What is? The statistic that I focus on more than any other, diagnosing the current economic situation, is that measure of national saving rate that Bob Eisner spoke of. Of course, economists can argue about a variety of measurement issues, but these things are true.

Consumption relative to our GNP is higher than it has been since World War II. We are consuming, meaning we are not putting aside, meaning we are consuming 98 cents out of every dollar. Or to put it differently, we are investing only about 5.5 percent of our GNP, close to the lowest in the OECD in plant equipment, and we are financing -- in plant and equipment and new housing, net investment and we are financing the lion's share of that investment by borrowing from abroad.

If you do not like talking about the measurement of saving, think of it as the amount of new investment and physical capital that we're making, less the amount that we are borrowing from abroad. Unless you believe that the United States either can continue borrowing on the current scale from abroad, or that the United States should reduce that already low level of investment, unless you believe one of those two things, you must believe that it is an important priority to get that two percent national saving rate, as we measure it up, because it's an arithmetic identity.

The amount of plant equipment investment that we do comes either from our national saving or it comes from borrowing from abroad. If you believe you want to borrow less from abroad, and have more plant and equipment investment, you believe we must get our national saving rate up.

I think that while economists -- and let me just say, without developing the point, that the prospects in the current economic environment for simply expanding the economy, by virtue of having had more expansionary policies, have more income and therefore more saving do not strike me as being there. There is simply not room to create substantial saving simply by expanding our demand policies. That means that we need to increase our national saving and alter through time the composition of output.

Economists differ on what incentives of various kinds can do to influence the level of private saving. I am more optimistic than most about what they can do. But it seems to me that any reasonable reading of the evidence suggests the most potent and reliable way to increase national savings is to increase the public sector's contribution to national saving. Or, more accurately, to reduce the public sector's drain on national saving. That means taking policy measures that reduce the level of Government budget deficits.

From the point of view of national saving, it is -- and this is the part what Dutch Leonard was saying -- it is irrelevant whether those savings are called the Social Security trust fund or whether those savings are called budget deficit. The crucial priority is to move as far as we can to move the consolidated budget situation towards surplus and thereby raise our national saving rate.

To put that two percent national saving rate in perspective, it's useful useful to compare it with those in other countries. Two percent is about a third of what the British have been saving. It is about a fifth of what the average of our industrial competitors have been saving, and it is about an eighth of what the Japanese have been saving. It seems to me that the risk that we will have too much saving, because we will move too far to promote saving, is a rather remote one.

How does the Social Security trust fund fit in to that process? Well, I think it is very clear that anything that does not insulate it from the rest of the budget, anything that enables the Social Security trust funds to be used as a reason for doing less with respect to the rest to the budget than would otherwise be done, is something causes the budget deficit to be larger than it would otherwise be on a consolidated basis, and therefore reduces the potential public contribution to increased national saving.

For that reason, provision for the integrity of the Social Security trust funds in the budget process, focusing attention over time on balancing the budget, on an exclusive of Social Security basis, seems to me to be a way to provide most adequately for the future.

Now I am not making the recommendation. It would be a naive one, that we move over the very short term to balance the budget exclusive of Social Security. There are certainly potential dislocations with moving too rapidly towards budget balance, although frankly the prospects that the political constraints on movements to budget balance will come after the economic constraints on movement to budget balance, strike me as being pretty minimal. I think that the prospects of political process generating something that is a movement too far does not strike me as being a very real one.

But the critical thing is that over time, I think we need to be moving towards protecting the Social Security trust funds, and that means running and using them as a device for running small budget surpluses in the Government account. Now there is all this question whether the Social Security trust funds are real, and whether it will really be there in the year 2020, or whether all the stuff with the trust fund is just an accounting fiction. I think that misses the point.

Ultimately, how we as a society care for people like me, 32 or 35 years from now when we retire, will depend upon how the society has fared economically over the intervening 30 years. Whatever one says about what we write into some law right now, if this economy grows rapidly and healthfully over the next 30 years, you are much more likely to see adequate and generous provision for security, than if this economy grows slowly and follows the kind of path that the Great Britain has followed over the last 40 years.

The thing that is the most important for preserving the objectives of the Social Security system is doing the things necessary to keep the economy growing at a healthy rate. I believe that there is no realistic prospect that we will enjoy substantial economic growth, at rates which we have been accustomed to, as long as our national saving rate stays anywhere near two percent.

That makes it a critical priority, from the point of view of overall economic policy, to get that national saving rate up, and from the point of view of preserving the integrity of the Social Security system and its objective of providing security. We need to get the national saving rate up.

I think that is best done by respecting the integrity of the Social Security trust funds in the budget process over time. Let me just stop there. Thank you very much.

PROFESSOR EISNER: Well, I agree with Larry's beginning and with his ending, but almost nothing in between.


I would just ask Larry if he would please respond to my criticism of the national saving rate. He knows well what saving should involve in his terms of Government capital, of human capital. Yet he keeps talking about this national saving rate of two percent. If he'd read my paper or others, he could see how little sense that makes, including in terms of foreign investment.

In those terms I ask him, how is he going to reduce the budget deficit? I should not embarrass him, but he cannot go on the record for raising taxes now, and I think that's a political no-no. So the only way he's going to reduce that budget deficit is by cutting Government expenditures. What expenditures is he going to cut that's going to exclude education, research, infrastructure and Government capital investment generally?


PROFESSOR SUMMERS: Gee, you know, I really --

VOICE: Tell him you're not getting an honorarium.

PROFESSOR SUMMERS: I don't think I am. I sort of think I'm earning one.


Bob, you know, we can argue about the words. Let me call it the amount of housing and plant and equipment investment financed domestically. Let me call that only two percent and tell you that I regard that as completely unsatisfactory. It is a fact that whatever you can argue about, whether -- I think about some of the things that my parents' tuition money in college bought. I think they cringe at the use of the word saving to describe some of the activities that were financed in that way.

Look. I said in as strong terms as I know how, that I think that stuff is critical. The issue really is can we do with so little of these other categories of investment, and can we continue to finance those other categories of investment so heavily from abroad. I think the answer is we need less financing from abroad and more of those other categories of investment. The only way we get that is by increasing saving as we measure it in the national income accounts.

PROFESSOR LEONARD: May I ask you a question?

PROFESSOR SUMMERS: Yes, because I was coming to stuff about what the next president should do. So I'd love it if you ask me a question.




I think that the Social Security system is part of the package of income that people receive when they retire. I think that retirement is one important motive for saving. So Social Security no doubt is part of the calculus that people engage in when they make saving decisions.

But there again, it may be the case that if one slashed Social Security, people would feel more insecure and would therefore -- about their futures, and would therefore decide to save more. No doubt the same would be true if we outlawed life insurance or if we announced that there would be compulsory participation in a national lottery on a large scale.

There are a lot of ways of making people feel insecure and creating needs for them to save. I do not think that abdicating on a commitment that we have made, for good or for ill, would be an effective way of seeking to increase national saving. I think a much more effective way of increasing national saving would be reducing the public sector's drain on national saving.

PROFESSOR LEONARD: It's also not a possible way to

-- a politically possible way to influence. So it's not clear by the question it's a direct consequence immediately.

But Larry, you spoke about the prospects of insulating Social Security from the budget and trying to have it have as little influence as possible, at least in terms of having the Congress feel like there was a lot of money available. What are the devices you would propose for doing it?

PROFESSOR SUMMERS: I have a tough enough time with economic science without trying to venture into political science. I think the terms in which the discussion is normally framed by opinion leaders, by people who are involved in issues, make a difference. Every time you see a budget, a table describing the budget deficit situation in which there's no mention of the part of the improvement that is coming from accumulation of the Social Security trust funds, a blow has just been struck for not insulating the trust funds.

Every time you see a set of projections that highlight the contribution of Social Security trust funds, one way of putting it might be that their deficits are being balanced with our trust funds. Every time you see such a projection, you are highlighting what Social Security is doing. So I think it is probably the terms of discussion which in turn influence the terms in which legislation is crafted and so forth, more than it is any magic one-shot bit of legislation, that is influential.

MR. DAHLMAN: Anybody in the audience?

MR. PODOFF: David Podoff with the Joint Economic Committee. Isn't there a middle ground that you gentlemen have been looking at in conciliation? I'm looking at the Brookings' simulation. I mean we're running budget deficits of probably three percent of GNP now, whether it (inaudible) simulations, they look at a budget deficit of 1-1/2 percent, which is essentially the offset case, the validation case is 1-1/2 percent of positive surplus, assuming that roughly three percent surplus in the trust fund as a percent of GNP.

If we just went to a balanced budget, that would be a three percent improvement in the deficit. That would give you more savings. It would still leave half of the Social Security surplus for Bob's investment in human capital, infrastructure and all that. Why do we have to go all the way to the surpluses? Why aren't we content to just push us close to zero as we can get?

But it seems to me there's enough room to both get the more investment that you want in plant and equipment, and also some room for the infrastructure and human capital structure.

PROFESSOR SUMMERS: Once again, I do not wish to be portrayed as believing that plant and equipment investment is more important than education, infrastructure and other things. I think there are a host of ways of promoting those forms of investment. I think the lessons in a large number of states suggest that many of those ways do not involve the Government paying the whole thing, but involve various ways in which the Government can lever its funds and catalyze private efforts. So it is not my view that plant and equipment investment is the only important thing.

I guess the question I would look at, and I'm not looking at it with the horizon of next year or the year after, would be suppose that -- I didn't quite understand your figures -- suppose we came to have a 1-1/2 percent budget deficit of GNP, and that was relative to current 3.4 percent. We'd then have, if other things stay the same, a national saving rate of about four percent. When you consider that that's about 40 percent, somewhere between a third and 40 percent of the national saving rate of the average nation that we compete with, and that it's less than a quarter of the Japanese national saving rate, that does not seem to me like it's quite enough.

MR. DAHLMAN: One more question.

MR. KORMENDI: Yes. (inaudible) of the University of Michigan. There's two issues. The second part of Bob Eisner's question I'd like to hear the answer to, if in fact you're not going to raise taxes, how does one maintain that full amount of investment.

The other thing is that if in fact consumption is private sector behavior, an issue of private sector behavior, there has to be either some sort of incentives, or something that changes the consumption behavior as opposed to the investment part of the equation, and how would you --

MR. BOSWORTH: It's called taxes.

MR. KORMENDI: That's the issue.


PROFESSOR SUMMERS: The question is what is the structure of the taxes that should be.

MR. KORMENDI: Oh sure.

MR. BOSWORTH: Don't answer.


PROFESSOR SUMMERS: Let me just say one thing. If you look at what's happened to the budget over the last eight years, the budget -- the increasing budget deficits is just one sign of the extent to which the "me" generation is creating the "now" society. If you look at those components of the Government budget that are forward-looking -- manpower training, Federal support for education, support for civilian research and development, infrastructure -- if you look at any of that, you find that it also has declined sharply over the last eight years.

While there is, I believe, a need to make tough choices on spending in cutting, in reducing the budget deficit, I think it is critical that we avoid making those choices in a way that would cut back further the investment component to the budget. That component is, however, a relatively small part of the budget, now less than ten percent of the budget. So I don't think that that would be the dominant area on which the budget reductions would fall.


I had a question for the two presenters this morning, just a comment, and that is these forecasts have huge standard errors. The degree of uncertainty in these simulations is enormous. I think it does not do us any service to present one number without giving some indication of the degree of uncertainty of those numbers. Those forecasts in 2040 I wouldn't put much money on myself. They might be right, but they are probably wrong.

Anyway, the two papers before us in this session are kind of fun to read and I enjoyed them, but I don't necessarily agree with them. The first one is by Eisner, who I think is a magnificent national income accountant. The first part of his paper is really very important, I think, in raising various issues of income and wealth, connections with what is measured and how these things relate to international saving comparison. I'm much less enamored; in both these papers there is a lot of rhetoric. I guess the whole topic lends itself to a certain amount of rhetoric and I will contribute my share.


But I'm much less enamored of Eisner's Keynesian macroeconomic proselytizing about the virtues of deficit financing. The issue before us concerns long-run solvency of the Social Security system, not job creation, forced savings, or any other aspects of short-run macrostabilization policy.

Payroll taxes are not exactly commendable instruments for fiscal policy. Many of the concepts he espouses there, it seems to me, have very dubious theoretical or empirical virtues, even for his purposes. For Longer, I'd would say look what happened to the Polish economy and the secular stagnation, I think, is really not an issue that is much related to what we are dealing with here in fact. To me, the issue of secular stagnation is much more related to the size of governments, and the surpluses being generated and forecasted under this scheme do not give me much pleasure.

For Herman Leonard's paper -- by the way, I'm a person who has been accused of having two last names -- and Herman Leonard has two last names and two first names.


Leonard's paper has the exuberance of a person who has recently learned about how our Social Security system actually works and who is out to spread the news. I think this is great and I really got a kick out of it, having felt the same exuberance myself when I first learned how the system worked. I thought the rhetoric involved in the paper was a little excessive about the connections between Social Security, baby booms and Newtonian physics. In my circles Ponzi isn't mentioned in the same breath with Newton and Einstein.


So on the other hand, if you strip away the rhetoric in that paper, I think Leonard raises some very important points about the politics of reserve funding of Social Security and the difficulty of assessing any fiscal validation. True, the Social Security Act suggested partial reserve funding, if I am not mistaken, and I think -- but that fell by the wayside as any surpluses were spent really early.

I think that same force is one of the reasons why the scale of Social Security has grown over the years. I think you can make a case for the big benefit increases in the 1960s and 1970s along those lines. Now whether widespread understanding of the demographics in the impending crisis in the system will be sufficient to overcome these myopic tendencies remains to be seen. But I think there are some big pressures working in that direction.

Now even such a consummate politician as Ronald Reagan has refused to touch any Social Security issue with a ten-foot pole, though some of his confidantes inform me that it is not his favorite social program. The gray lobby is powerful because it's so sharply focused on only two issues: Social Security and Medicare. Its constituents have plenty of time to devote to these issues and they vote. By the way, they're after me. I get an invitation to join every week, I think. Maybe I will get half-fare on street car rides. Were it only that American Association of Retired Persons (AARP) and related organizations were Barro-Ricardo equivocators, knowing full well that they were taking it from themselves.

These political pressures will indeed make it difficult for Congress to validate the policy. Now Leonard refers to Social Security as an entitlement program and it is, I guess. But I think it clarifies one's thinking to think of it as a purely debt-financed program. That's really the source of its popularity. Of course, behind the scenes lies the intergenerational transfer that we all are familiar with.

But this pure debt-finance would be perfectly clear if, following a suggestion that James Buchanan made many years ago, the Social Security system actually issued paper certificates. Every time you paid your taxes, you would get a piece of paper debt, just like a bond, that pays off when you are old. Well, we know that the return on those bonds have fallen in every successive generation, in that the crisis now is related to the fact that that return's going to be very low for these baby boom people.

Now what can be done about that? Well, basically nothing. Those people are screwed. There are too many of them, and there's too many of them and the only way that we could solve this problem somehow is for other generations to transfer resources backward to them.

Now none of my colleagues are here, so I can feel a little safe to say that I understand the virtues of the Barro hypothesis and I think there is something to it. But I have never been a person who thought that Social Security actually did increase private saving. It seems to me, at best, it would have a neutral effect and probably reduces it.

So what this program of running a big fund is doing, it seems to me, this generation knows that it is in trouble. It knows that it has to accumulate private retirement accounts that are going to have a direct effect on private capital. What the fund build-up and this change in the law has done to

-- increase taxes, is for the Social Security system to go into competition with Teachers' Insurance Annuity Association (TIAA) and those kind of private accumulation funds. I do not think that is a good idea.

To me this was a great opportunity for the Social Security system to be wound down to a level that was more commensurate with its primary goals of providing a basic floor of retirement support, and not the primary source. Thank you.


In the musical "Once Upon A Mattress," King Sextimus was cursed by a witch and was condemned to silence for seven years, and when the spell was finally broken, his first words were "I've got a lot to say." But I've had seven years of discretion beat into me too, so you're only going to get a few of them. While my remarks have been in preparation for seven years, they're not yet back from the xerox room. So I do hope that by lunch time or thereabouts you do get the paper and can look at a couple of the graphs that I won't wave in the air, but some of them are fairly interesting.

I prepared the paper on the offchance that I'd be agreeing with the papers and the presenters, which in fact I did largely. So I want to say a few brief words about those papers, and then say a little bit about the trust funds, not where we'd like them to go but what they are. I am sorry that Larry Summers left. I did agree with one thing he said. I agree that Massachusetts has not had a better fiscal policy than the rest of the country. One other point. I think R&D spending has about doubled in real terms in the last eight years.

The Leonard paper has made a few very good points that I think bear repeating. He describes a round-robin financing, where he says that if Treasury didn't borrow from Social Security and instead Social Security lent to the private sector, the private sector would have more money to lend to Treasury. You've got a triangle, A, B and C and instead of A borrowing from C, C lends to B and B lends to A. So I think that's worth mentioning.

He says the trust funds may or may not be real saving, but they can surely become real spending. I think that is perhaps the danger of a sudden realization by the Congress that you have these large numbers floating around. I confess to be partly responsible for that, having helped prepare the format or suggesting the format of the trustees' reports for a number of years. The extras, of course, do all the real work.

But for a number of years they've published an actuarial note at our request, giving us the information in dollars as to where the system was headed. This was separate from the report. This year, with computer upgrades and a little nudge, perhaps, from Dorcas Hardy and the rest of us, they've accelerated their work and they have published it as an appendix to the trustees' report. There was a little bit of betting going on, on side among the staff people, as to whether more people would notice it this year than in the past. Apparently so, because there have been a flood of articles in the press about where the $12 trillion come from. It's been reported for five years. Where did the $12 trillion come from and what do we do with it? I think that the idea that this sudden burst of realization might indeed become real spending is a real danger, and I want to commend the paper for making that point very sharply.

The Eisner paper, beginning with the discussion of the concept of a real deficit and the real budget picture, is an important reminder that we are suffering from many an illusion all to often. At Treasury there was some work done. There was a paper by the Assistant Secretary of Deficits and Saving Behavior, which tries to lay this out and put the deficit into a real framework. I think the more often we can hear that, the better.

Dr. Eisner makes another important point. He says that he is uncomfortable with the notion that Government should play big brother and have Government telling people how much to save. I think that could be made stronger, and I'll refer again to Bob Barro, who is the ghost here at the proceedings. In addition to the Government perhaps should not tell people how much to save. Barro would say they can't tell people how much to save because the public reacts to fiscal policy changes, and a tax increase, contrary to Mr. Summers' attitude, particularly if it's structured improperly, could definitely come primarily out of saving and much less out of consumption. The idea of the appropriate structure of tax or spending changes, which is just never addressed in these rather loose national income shifting exercises, is critical as to what the impact of the tax or spending decision might be.

Indeed, I think there is an answer as to whether or not Social Security affects the savings rate. It is a tax; it is funded, partly by a tax on labor. That results in less employment which lowers the rate of return on the complimentary factor, capital, which results in less capital stock, which results in less saving. The flow is from the capital investment to the saving, not the saving to the capital investment.

Dr. Eisner has explained very clearly that Social Security's financing or funding or resource transfer occurs in the year in which the benefits are paid and that has to be financed in the year in which the benefits are paid. There has been some discussion about the financing of the system, and the funding of the system and the pre-funding of the retirement benefits of the baby boom.

The system is not funded ahead. No financing is done by Treasury in advance of the benefit check, and the trust funds do not occupy that slot in the real world. So let me turn to the trust fund very briefly and try and figure out what are these things. As mentioned, words don't always mean what the reality is.

One of the graphs in the paper, when you see it, is a graph of two mountains. One is Mount Everest, which you climb and climb and climb, and then roll down on the other side a great distance. The other is a little molehill underneath which you climb and then slip down the other side a heck of a lot faster.

The trust fund build-up that is reported in the reports in Table G-1 in the Appendix, is of course from the perspective of the system that is doing the reporting. It shows the entire trust fund build-up, the spending authority available to the Social Security Administration for the payment of benefits authorized by the Congress. That includes both the tax income of the system and the interest income of the system.

The system then reports both of those incomes and the total income and the outlays and the trust fund build-up.

Well, that's fine and that is the spending authority of Social Security. But from a total budget perspective, only the tax income minus the outlays represent the Federal budget's net income from the system that year. The interest transfer from the Treasury to the Social Security Administration is an intra-governmental bookkeeping shift. It does not represent a net budget surplus. It is not part of the impact of the system on the credit markets, and it is not what Wall Street should have looked at.

This comes, then, to the reason we had the pool. We were wondering what the credit market people would think if they really got to focus on the trust fund build-up, and their response was predictable. Two years ago, the Treasury would hold the refunding dinners for the discussion of the quarterly financing and the people from New York would come down and they'd tell us what we ought to sell them.

They were always after us. These terrific deficits; they're structural. They're all going to infinity. We won't be able to absorb your debt. Of course, most of the deficit was structural. It was not structural, it was cyclical. It was on the way out with full employment returning and they needn't have worried.

But they were hysterical. They weren't sure that they could handle all the inventory we were throwing at them. This spring we got a very different picture. We were getting phone calls from overseas and from all over the country and the quarterly refunding dinner was not the pleasure it might otherwise have been, because the public securities dealers came down and said $12 trillion dollar surplus. You're going to buy back the whole national debt. What will we sell? We won't have any inventory. We'll be out of business.

We told them that unfortunately that wasn't true or that we didn't mean it quite the way it sounded. But there was something else going on here. OASDI is projected to run annual cash surpluses of income over outlays for the next 30 years, through 2018. The nominal annual surpluses will peak at just over $160 billion in 2009. Correct it for inflation. That's about $70 billion in 1988 dollars. That's the peak. The average surplus will be a modest 8/10ths of one percent of GNP. It will total just over $3 trillion cumulated over the whole period. It's the annual surpluses and deficits that are OASDI's total budget impact.

I've been in Washington too long and I tripped up. I used unified budget the old way to mean the total budget. The new unified is the total budget including the off budget but not the off-off budget, which is where Social Security now is. If you look at the paper, when I say unified I mean the whole thing.

Now you get about $3 trillion in real surpluses. Now the trust fund goes to $12 trillion. In fact it keeps building past 2018 until 2031. That figure is not only not cash for spending; it's not really there. By 2031 the whole trust fund will consist of interest paid by the general funds to the trust funds. All the prior cash surpluses will have been followed by ongoing cash deficits. It's just that in that later 15 years or so the interest transfer will be larger than the cash deficit, so the trust fund will build up even though the Government is still borrowing money, because the system's not running a surplus.

The annual cash surpluses or net income, they reduce the borrowing requirement of the Government. The inter-governmental interest transfers are not net income; they are an accounting artifice. To see the artifice a little more clearly, the law says we have to pay a certain interest rate to the trust funds. The law could have said we had to pay twice that much or none or any number you care to pick, the trust funds would balloon or shrink or do anything you would like them to do. But it isn't actual cash flow.

Now once you adjust the trust funds for reality, I think, you get a slightly lower number. As I say, the cumulative is about the $3 trillion, not the $12 trillion. Another thing you might look at is, since again AARP and others are focused on all of the elderly's programs, the Medicare deficits are going on at a nice clip, at about the same time. If you do add in the Medicare, the numbers look even smaller. The combined cumulation on the trust fund in real dollars would be about $1.9 trillion, not the twelve nominal trillion that you look at when you only get the OASDI itself.

Now even the $12 trillion or the $3 trillion real surplus, it is bigger than the current national debt. But the national debt isn't standing still. Even if you assume Gramm-Rudman takes hold, you still get a build-up in the existing extant debt held by the public to about $3 trillion from the current roughly $1.8 trillion. The total debt is 2.5, but 500 is held by the trust funds and $200 billion is held by the Federal Reserve and the public's holding 1.8.

If you get up to $3 trillion and then you hit Gramm-Rudman, and you are in total budget balance thereafter, the $3 trillion sits there getting rolled over. The trust funds have to be compared, perhaps, to that. In Dr. Eisner's paper, he refers to the Gramm-Rudman affecting the unified budget excluding Social Security, but in fact Gramm-Rudman applies to the total budget, and therefore we get a slightly different picture.

The trust funds, as they build up to $12 trillion, would indeed be able to have $12 trillion in Treasury securities to hold. The interest we pay the trust funds works like this. Interest is due on the debt. We do not go out and borrow money and give it to Social Security and borrow it back. There's no credit market intact at all. We simply said "oh, we owe you some interest. Would you take a bond instead?"

We give them a new bond, which is an internal creation, and hand it to them. That part of the debt is not affected by Gramm-Rudman. It has nothing to do with the $3 trillion debt to which the public debt would build by 1993 under Gramm-Rudman total budget, unified in that sense, balance. So you could put the trust funds into these special Treasury bonds if you felt like it.

Indeed, unless the unified budget itself in the total sense went into surplus, you would not be buying back national debt at all. So there is nothing to invest in the private sector under those circumstances, except in the sense that if we put the trust fund money into the private sector, we'd have to borrow an equal amount back out to fund the actual spending, unified total budget of the Government. So in that sense it is moving the markers around on the game board.

The question then arises, and I think this is where the scenarios in the two papers presented this morning are kind of critical. The ones that are attracting the most attention are should we validate the surplus and be running huge total Government budget surpluses to actually try to force an increase in national saving.

To very quickly say Larry, I disagree with that concept. Dr. Eisner's pointed out that we should be investing the trust funds where they do the most good, the most return on them. I think that's true and also in the real sense he talked about, where he was talking about infrastructure, education and so forth.

The same point can be made with Government spending, and saving in general in the economy. If plant and equipment spending is yielding the highest return, or Federal aid to education or state aid to education or private sector education or infrastructure, the whole spectrum, the saving should go toward where it has the highest rate of return.

In attempting to force higher saving by the Government, you run into the Barro problem. I don't quite agree with Barro's analysis of how you get there, but it is more than likely tax increases will come out of saving. If you have an increase in the corporate tax rate or a weakening of the depreciation allowances, you come dollar for dollar out of corporate saving. Personal tax increases of various types are likely to come overwhelmingly out of personal saving.

In other words, the public won't let you push them around that way. So it's not likely that you could in fact push up the national saving with a tax increase if you allow the public any freedom of action. If you could, should you? There the answer is probably not. If certain things need to be done through Government because of externalities, by all means invest through the Government.

But generally speaking, you would not want huge numbers of trillions of dollars to be steered by the Government into particular types of plant and equipment investment in the private sector. If you are going to have those trillions kicking around, please lower the tax rate on capital, not necessarily in the negative territory. There are still positive taxes on capital which could come off and induce the private sector to spend the money in the most efficient way possible.

I think that would be the preferable way to grow and prepare for the baby boom. Thank you.


I don't have much, I think, to add to this. I would like to return a little bit to the original papers that we were asked to discuss.

One of the questions that has come up here in terms of a Barro hypothesis effect, I think is an important one to consider, the possibilities that surpluses would just lead to offsetting private reductions. I would like to put forth, though, an alternative that seems to me an equally reasonable theory that comes out of the data of the 1980s. Government budget deficits lead to declines in private saving because the reasoning is if the Government can do it, why can't I do it. Now that may sound like a foolish theory but there is a lot more evidence in favor of it in the 1980s than there is in the Barro hypothesis.

Second, I think that these forecasts are very uncertain, and that is a very good thing to keep in mind. But that is a reason to think of the issue not in terms of the forecast, but in terms of the deviation from it. What difference does it make? That is a lot less uncertain.

Also, the other one is if the world is as optimistic as the II-B projections have turned out, I think it is sort of right to say we really would not care, because our standards of living would be rising so rapidly because of technological change, that the implication of this issue would be small. On the other hand, if economic growth continues at the same rate that it has since 1973, this is going to be a very big issue, because if on top of that, future generations have to pay for the baby boom's retirement at standards of living no higher than today's. It's going to be a big issue. So actually it cuts to me, it seems, a little bit the other way.

I look at this really as sort of we have three options that we could consider at the present time. One of them is that we can continue to run a surplus in the Social Security system and we could try to save it. The second one is that we can run a surplus in the Social Security system and consume it. The third one is we can abandon the whole thing and go back to pay-as-you-go financing.

I believe the first option does offer, for the reasons that Larry Summers gave, great opportunities to increase future incomes, and that there is a very high rate of return for doing that. It may be that politically we cannot do it, and we will end up with option two. We'll just use the surplus and finance to general fund.

If that was going to happen, my personal preference would be to go to option three. I do not like the idea of using a regressive employment tax to finance general funds revenues. I do not like the type of inter-generational debate that will occur 30 or 40 years from now when my generation, with great righteousness, says "that money's mine. Give it back to me." So I think the way to think of this is in terms of three options and the relative merits of attaching them.

Bob Eisner's paper, what I take to be the critical issue, is whether or not we can utilize the surplus effectively. Bob takes a short-term perspective. I would prefer a somewhat longer-term perspective that a lot of the short-run fiscal problems we have had are sort of Keynesian economics. We are talking here about long-run targets. In any given year you may have to back off from the goal, because you have an unemployment problem or some other difficulty to deal with.

Again here, I think of it as in the past, we talked about a goal of balancing the budget. We never did it. We missed a little bit and our average deficit has been about 1.5 percent. What we are talking about here is what should be the goal. We won't do it. The question is should be aim to balance the total budget deficit, a la Gramm-Rudman, or should we aim to balance the operating budget deficit. I think we should aim to try to balance the operating budget. We won't succeed, and we will probably run a deficit in it of about 1.5 percent like we have in the past.

I think the biggest thing that would go for is simply to reduce foreign borrowing. At the present time, I do not see why this issue comes up, "can you increase private investment that much?" What you can certainly do is quit borrowing so much overseas. That is a very easy adjustment to make.

The second one is I think Bob, though, is quite right to stress that human capital and research and development are equally good uses of this surplus. They, too, are investment. What we worry about here, though, is that there would be a tendency in the Federal budget to say "that's true. Okay, I am going to use the Social Security surplus to finance all the education and R&D expenditures that I used to do out of the operating budget."

You just shift funds. You claim Social Security is being used to finance education, when in fact you've just shifted some accounts around and end up using the money for consumption. If we could be sure that we were going to use the surplus for added additional expenditures on education, over and above what we would otherwise do, and added additional R&D expenditures over what we would otherwise do, that would have the same economic effect as spending it on physical capital. I do not see that it makes a lot of difference and I agree with Bob that that may be a way in the future, that these funds get used.


My comments are brief also, and they are primarily directed to Bob Eisner's paper. We, for our part in our study, would certainly endorse the idea of using a much broader definition of saving and capital accumulation. In our study, validation broadly refers to translating the trust fund surpluses into saving and capital accumulation, and that could include Government saving.

In our full report, we in fact refer to investments in human capital and infrastructure as potential uses for the capital that is being accumulated. With regard to the simulation model, we are constrained to use data as reported, imperfect as they are. So we have to use the savings and investment concepts as published by the national income accounts.

In estimates of effects on output and substitutability among capital and labor, we're of course constrained to using the existing data and existing measures of business investment and production. Government capital accumulation, then, could fit into the conceptual framework, but the problem is that it is very difficult to measure and it is very difficult to measure the outputs of additional Government investment.

This is more than simply a technical modeling point. I think all of us agree there is evidence that we could usefully reallocate our national capital in the direction of more investment in human capital and perhaps in environmental capital, in R&D and local infrastructure.

But there are two major problems. The first question is how much. The absence of a market test for public investment makes the ultimate decision a political one, and makes it somewhat arbitrary. Even though private investment may be pushed to the point where the marginal return is zero or less due to inefficient Government investment incentives, there is a lot of evidence that investment in public enterprises, particularly from countries where public enterprises play a much bigger role -- such as the less-developed countries -- that public investment can easily be pushed to the point where marginal return is less than zero.

The second problem we would have, while endorsing the notion of using a broad definition of investment, is that public investments are likely to involve a much greater political element and be much more subject to political input, either with the outcome of supporting pet projects of important and powerful constituencies, or powerful political leaders.

So when we fall back upon it, as long as we -- if there is a general consensus that private investment in plant and equipment is inadequate, and that foreign borrowing is excessive, we can begin with a great deal of confidence looking in that direction.


MR. MYERS: My name is Bob Myers. I come here somewhat like the little boy in the fable of Emperor's Magic Clothes. I assert that all the analysis and discussions are founded on a false premise. This false premise is that in 1983, the financing provisions were developed to build up a mammoth fund to take care of the baby boomers. This is not so at all.

Rather, the major effort in 1983 was to solve the short-run problem by using pessimistic assumptions for the financing provisions. Then to solve the long-range problem, on the average -- and I emphasize on the average -- you might ask why didn't Congress and the National Commission do a more thorough job in 1983? Well, the situation was that the ship was about to hit the iceberg. At that time, you worried about dodging the iceberg not how to redecorate the dining salon, namely, long-range funding procedure.

I would challenge anybody find anything in the Report of the National Commission on Social Security Reform that said the intention of developing the financing of the program was to build up a mammoth fund to take care of the baby boomers. Nor will you find any of this in any of the Congressional discussions, the debates, the Committee reports. All the fabric has been made up subsequently.

The real answer to it all is what Barry Bosworth said, the third one. Go back to current-cost financing. In fact, the Board of Trustees has said this for years. They said immediately after the 1983 amendments for a while that that's what we were having. We should have current-cost financing with a small contingency reserve. If we need to take care of the saving situation in the country and I bow to all the economists here, let's do it forthrightly, not by regressive payroll taxes. Get out there through other payroll taxes that will reduce the debt in the public sector held privately, which the essence of what this would do if Social Security builds up a big fund.

Now some people might say "well, this is what was done in 1977," and that's not the case either. Because in 1977, the financing long-range problem was not attacked. Instead, a fund would have been built up that would amount to about two or three years' outgo, and then would have been drawn down. The fund balance would have been exhausted in the year 2028, just when the baby boomers, whom we are worrying about, would be at the peak of their outgo.

One final important fact that I think that people should keep in mind is that the cost rates of the Social Security system do not peak with the baby boomers. Rather, they plateau at a very high level. So to build up a big fund and then use it up just does not make sense. If you go back to what Barry Bosworth said or what some of the Brookings people were saying, build up a big fund and maintain it, which I think has great political dangers, as Steve Entin pointed out.

But if you do that, we are just reinventing the wheel of what was done in 1935, because that was the exact financing basis then, and it was dropped over the years. Thank you.

MR. DAHLMAN: Sounds to me you would like to rescind the scheduled tax increase of Social Security.

MR. MYERS: Exactly. That's what I've been saying for the last couple of years.


MR. PECHMAN: I want to just ask Sherwin a question and then I want to make an entirely different comment. Sherwin said that baby boomers are being screwed. Sherwin, look at Figure 3 (of the Brookings handout). The baby boomers, according to my authority who is not quite among the baby boom -- Alicia Munnell was born between 1946 and 1962. They will retire between the years 2013 and 2029, giving them the life expectancy of 20 years, that's 2033 to 2049. Now you look at this chart. Why are they being screwed? That is precisely the time --

PROFESSOR ROSEN: They are paying into the system and they are getting it back. Who would wants --

MR. PECHMAN: But is that being screwed?

PROFESSOR ROSEN: --Social Security when you could invest in a private retirement plan.

MR. PECHMAN: Okay. I just wanted to clear up the point. I just wanted to clear this point that they are getting it back and you implied that they were not.

PROFESSOR ROSEN: No, I didn't mean to say that.

MR. PECHMAN: Well, the point I wanted to make --

PROFESSOR ROSEN: They didn't get as good a deal as you and I are going to get out of the system.

MR. PECHMAN: Oh, I'm sure they're not going to get as good a deal as you and I. That's too bad.


But the point I wanted to make is I wanted to call attention to Figure 5 and Gary Burtless' presentation and relate it to the term "crisis" that occasionally appears, and this is partly for the benefit of public trustees.

If you look at Figure 5, even if you take the needs for -- if you go Bob Myers' way and the Bosworth way, pay-as-you-go financing, the tax rate never exceeds, according to this diagram, something like 15 percent. That's total. Now that tax rate is lower, the tax rate that we would have to have 30 or 40 years from now is lower than it is now in most of the countries in Europe.

According to this chart, there is no crisis. All you need to do is to have modest tax rate increases. You can actuarially balance the system for the next 100 years. I just would like to know what people are talking about when they talk about crisis.

MR. HARRIS: I was born in 1956; that was the peak. So I can speak on behalf of the baby boom generation, since that was the peak year. My name is Ethan Harris. I'm also the only guy from Wall Street here, so I will speak on behalf of Wall Street.


Larry Summers earlier talked about the "me" generation, but paradoxically it is the elderly who are saying "me" the loudest now, not the baby boomers, because it's the elderly who have seen the biggest increase in their incomes. What I like to ask people on the panel is the big no-no question, which is how many of you would advocate cutting Social Security and Medicare?

MR. DAHLMAN: Let's take that one right away. Anybody want to? Stand up and be counted here.

PROFESSOR ROSEN: I certainly would not, and I think that is outrageous, and I don't know why more people don't say it. If there were a budget crisis, which I don't believe, there is no reason why you should try to solve it at the expense of a particular group. If you believe --

MR. HARRIS: Well, Social Security's the only thing that's been sacred in the budget debate.


MR. HARRIS: I'm just talking about making Social Security part of the general budget cuts.

PROFESSOR ROSEN: Making Social Security part of general budget cuts means making the cuts in a particular group of the population. The general budget cut, the idea of a general budget cut is that you're presumably cutting everything equally. If it excludes something that strikes at a particular group, you're not doing that.

PROFESSOR LEONARD: I have a slightly different reaction to the question. My own reaction is that I try to understand what is going on in the economy and to respond to questions that seem to be relevant politically. I look at the politics of the system and I just don't see any prospect for balancing whatever shortfalls you see in the out-years, by drastically reducing the benefits.

There are a hundred million taxpayers out there taking the pledge every day. They are not going to vote against themselves when the time comes to collect their benefits. Those are not transfers to them. They regard them as righteously earned entitlements. They are as result of contributions. The notion that you could do it seems to me on its face wrong, people's preferences about it notwithstanding.

MR. ENTIN: What time frame are you talking about? Because in 1981, we tried to take a bite out of Social Security for other than Social Security reasons by having an ultra- pessimistic forecast, what Bob Myers mentioned a bit earlier. They did not look at the long-run part of the system, looked only at the short-run part and Congress said 96 to 0 "no way."

In 1983, the Commission looked at the short end again and did not look at the long end. Now we hear constant talk about nickel-and-diming the COLAs. People already retired, some of them 80 or 85 years old, seeing an erosion of their benefits which may be the primary source of their income, and no way to jump back into the labor market. It is not necessary for the budget, it is not necessary for the Social Security system.

The important question is, and getting back to the other Professor's point, a two percent tax increase here on payroll? Great. A couple of million jobs lost. How about another 6 percent tax on payroll to balance the Medicare deficit? You've got two systems --

MR. PECHMAN: I'd like to say that you're dead wrong. A two percent increase to payroll tax will not reduce employment --

MR. ENTIN: Remember, you've got two systems here competing for the same tax base, the payroll tax. When you look at the two deficits, they're very large. I think what you have to do is to look at the benefit formula, what the future retiree, who's the current taxpayer. What does he want? Does he want higher taxes and higher out-year benefits, does he want the higher out-year benefits in retirement, or does he want them in health insurance? Or does he want to keep the taxes low and trim one or the other?

The point is that the politicians today are thinking about the current retiree, and no one is asking the current worker or the current younger generation what do they want for themselves later. I wish someone would ask them. I don't know what the answer is going to be, but nobody's is asking.

PROFESSOR LEONARD: But what they say now is not going to be what they demand then. They are going to be there in the hundred of millions. That is the nature of this problem, okay?

MR. AARON: We will be there. We will be there. Henry Aaron. I was just going to say, listening to the comments about how payroll taxes are going to reduce the number of jobs reminds me of various statements that we have shipped ten, fifteen or twenty million jobs overseas, which if we only had back would give us a negative unemployment rate. There are other processes at work, and that is from price theory.

The issue that I wanted to speak to was the question of whether the baby boomers are going to be screwed, and in what sense, and what rates of return. Somebody I heard say that they could expect to receive negative rates of return from the Social Security system. That issue has been examined. There are real numbers. Bob Myers is very familiar with them and maybe the best thing I could do is defer to him, but he'll correct me, I'm sure, if misstate the facts.

The analyses suggest that cohort by cohort, the internal rate of return that successive cohorts of retirees can expect to receive is going to trend down, that under plausible assumptions, under the II-B assumptions, the rate of return remains positive, cohort by cohort. It does not turn negative. So it is not the case that it is going to be a negative rate of return. Given the assumptions that are used in the projections, for any cohort of retirees looking ahead, to say that they are going to receive a negative rate of return is incorrect.

It is the case -- well, speaking of opportunities, where does the -- what is the riskless rate of return which is commonly used in economics? Somewhere in the range of two to three percent. That happens to be the rate of return toward which the internal rate trends for successive cohorts of retirees. Now it is the case, of course, that within cohorts

shake your head no; ask Bob Myers, who is I think a fairly reliable witness on this.

Within cohorts, it is certainly the case that some groups receive higher rates of return then do others, and it is even possible that some members of some cohorts could receive back in benefits negative rates of return. It is the case that Social Security has since the beginning, and remains today, a mechanism for providing higher benefits for some groups of retirees than for others as a matter of social policy. Since, as we know from Chicago and elsewhere, lunch is not free, [tape ends]

MR. ENTIN (continued): who managed to retire without having any beneficiaries other than themselves at all throughout their lifetime. It is that social policy does redistribute benefits from them to large families and to low income recipients.

But the important thing to keep in mind is that the statement that there are going to be negative rates of return is a misstatement if one is speaking of the rate of return that the baby boom generation is going to receive. One should not make that error.

MR. SALISBURY: Under the II-B assumptions, is what you're saying?

MR. ENTIN: That is correct, under the II-B assumptions. One could make assumptions that the rate of growth was very low or negative, in which case interest rates are going to be extremely low all over the economy, not just in Social Security. I think it is important to understand that this tide raises a whole bunch of boats besides Social Security.

MR. DAHLMAN: Well, I used the word negative. I stand corrected. Well, as Sherwin Rosen answered, (inaudible) the cost issue.

PROFESSOR ROSEN: It's just a question of what the riskless rate is and whether you think Social Security is riskless. I don't know. Perhaps it is not.

MR. BURTLESS: I hate for Brookings to monopolize the floor, but since it's given to me. There's an assumption that was implicit in which Dutch Leonard said and explicit in what Sherwin said, that somehow the political economy of the Social Security system is dominated by some kind of fraudulent Ponzi scheme, which by and large the taxpayers don't understand. That is just empirically untrue, surprising as it may seem anywhere around Cook County.

Surveys show that by an 8 to 1 margin, adult Americans are aware that their tax payments go to finance the current benefits received by current beneficiaries. There is not a sense -- there's no deception there. The same surveys show that the political support of Social Security is quite strong among young Americans, even though young workers have a very low assessment of the probability that they will receive benefits.

You might ask "why are they so self-sacrificing, that they are unwilling to accept cuts in Medicare and Social Security benefits?" Those are the two programs that are consistently found to have the highest ratings among taxpayers young and old alike. Why is that? I think that they also think that there is a self-interest in preserving the well-being of elderly Americans.

It's not righteous entitlement that Dutch Leonard pointed to. There are a lot of countries on this planet that have social insurance systems that provide for the elderly, that are equally popular even though there is no direct pay-in and pay-out, there is no insurance scheme that we have in the United States. In England and in Sweden, for example, the basic pension is payable to people when they reach a certain age. It has nothing to do with their contributions.

So I just do not think that the mechanism that is implied here based upon the deception of taxpayers is actually very important in causing the large support for the system.

PROFESSOR ROSEN: I did not mean to imply there is deception. I think the question that Dutch and I second is the control of political pressures. So it's an agency problem; that's what we're really talking about. How does the public transfer its votes to politicians?

MR. BURTLESS: Well, you're the one who mentioned the word Ponzi. There's some sense in which one generation or one group is left holding the bag and that is not --

PROFESSOR ROSEN: It's this generation; it's this young generation.

PROFESSOR LEONARD: No, but Jerry I mean, I agree with everything you said what people understand about this system. They understand that the money goes in now and that it isn't waiting there for them. That doesn't change their perception that there is a set of benefits that also ought to be waiting for them, as a result of the fact that they've paid into that system. I am not talking about a feeling of righteous entitlement at the age of 28. I'm talking about the feeling of righteous entitlement at the age of 65.

They're right. Okay, they did pay into the system over a long period of time. They did support it and it is -- they are deserving of having some of it -- something come back to them. I know it's a problem of politics about a fraud. I am looking at the politics of that and saying "boy that is a locomotive that I wouldn't want to try to stand in front of!"

MR. BURTLESS: But the political support among 28 year-olds for the system cannot arise on the fact that they believe that they're going to get benefits, because if you ask them, they don't think they will.

PROFESSOR LEONARD: No, I agree with that. It's a lot more complicated. Their support at various levels are more complicated than that, and a lot of it has to do with the fact that they want to see their parents taken care of, so they don't have to pay for them.

MR. DAHLMAN: You've been waiting a long time and I want to get this gentleman a chance.

MR. DAVIS: Yes. I'm Al Davis and I'm with the House Budget Committee, which I will deny if anyone quotes me. I'm struck by how the real economic problem of paying for future Social Security and retirement benefits has given rise to an extraordinary number of schemes that are basically accounting gimmicks, that are supposed to somehow influence the Government to change its behavior.

Having sort of been in the trenches for a while, I would like to suggest that when you play accounting games,

other folks can do that too with the opposite objectives. Secondly, it's very hard to predict how Congress will really respond. If you put Social Security off-budget and you go around telling Congress the true budget doesn't include Social Security, how do you know they're not going to raise benefits, for example, or reduce Social Security taxes? I mean that's a risk that's involved with these kind of games.

The second thing I would like to point out, in particular about one of the Leonard proposals which is very ironic and fun. He's suggesting that the Social Security system lend money to the private sector. Well, where would Social Security get the cash balances? Well, they'd have to get them from the Treasury.

So what you have is a case where Social security would lend and the Treasury would borrow. Most of us economists don't think this really would have a great effect on the credit markets or national saving or anything else. It is true that under the current scorekeeping, that would make the deficit go up.

Now that is an interesting game, and might affect perceptions. But there are plenty other people out there that realize that if you do things like sell financial assets and manipulate on and off-budget agencies, you can reduce the budget deficit basically by going through accounting manipulations. All you have to do is think about what has been done with FSLIC and what was done in some of the deficit reduction plans to realize that playing accounting games is a two-way street.

PROFESSOR LEONARD: I take the point that accounting gimmicks are gimmicks. There is a process called the problem of developing the face of the issue. The face of the issue, as people understand it to be is, I think, one of the only important determinants of what we actually do. The whole game is to shift perception in the direction that you think is closer toward the reality that you see.

That's what leadership is all about. It's about understanding what the problem is, understanding what people currently believe and figure out how to move their perceptions in that direction. My perception is that the non-OASDI budget deficit is much bigger and more damaging than most people think because it is currently conjoined with Social Security surplus. I think we would be better off looking at the numbers head-on, rather than separately.

In spite of the fact that I take quite seriously a number of Bob Eisner's observations, about the fact that the non-OASDI budget deficit itself is not a particularly well formed number. It is, however, a number that has some influence and it seems to me it's worth looking at, how to get this program and that one jointly and separately understood by the people who have to make those decisions.

PROFESSOR EISNER: I have a few remarks I'll try to put together very quickly. Larry, on distribution as between cohorts in the population. First, I'd just like to raise one thing, something that was almost never mentioned and I wonder why we don't worry about it. That is I would suggest that blacks get screwed in this system, because they do not live as long and any other people who don't live as long don't get very well treated.

VOICE: They get disability, though, at a higher rate.

PROFESSOR EISNER: Now wait a minute. I'd like to see some calculations in terms of longevity, in terms of when people start working and contribute to those systems. On the actual distributions between cohorts, I welcome the last remarks about accounting gimmicks. I might also say that should be extended to whatever pledges we have, whatever things you have in the scenarios.

You get down to the year 2020 or 2030, the amount that the elderly can get, a hundred million lobbyists or not, is going to be very heavily influenced by the economic situation, by what's available, by what that production is. If we do not provide the capital, or we don't import the Mexican workers or somebody else to produce them, there won't be that much around and the political realities of what the pie is and who the population is, is going to determine that distribution rather than anything that we have put down here.

I do have to come back to these budget deficits. I hate to sound like a broken record. I think a lot of you people may be familiar with my views on this point and with the fact we can somehow ignore them. The way we are measuring the budget deficit is absurd. No private business measures it that way, state and local governments do not, and yet we are all caught up in it. We talk about balancing a budget which will not be balanced with any other measure.

We ignore the state and local government budgets. Let me give you a few quick numbers. The total national income account deficit last year was a little over $100 billion, including the state and local. $100 billion, you begin to compare that on an international sphere and we're not getting that much higher, if higher at all, than other countries.

Secondly, you make an inflation adjustment, which any of you who are economists know you have to make. That is the real value of the existing Government debt, even with a modest four percent inflation rate, four percent of 1,800 billion outstanding, 2,000 billion outstanding. You're talking about an $80 billion inflation tax right there. That virtually wipes out the deficit.

You add to that the fact that we don't account for capital expenditures as any private business does, you substitute the current depreciation for capital spending on the investment type expenditures indicated by the Office of Management and Budget. You have another $70 billion difference.

You can quickly find that the real deficit, correctly calculated, is not a deficit at all. It's about a $50 billion surplus. Now some of you might want to say that we should be having that kind of surplus. Okay. But know what you are talking about and don't paint these pictures of this disastrous budget deficit. Now God knows you should now that much as I disagree with Larry, I'm not saying that as any partisan of the Reagan administration.

But I think the failure to face the facts, to keep repeating conventional wisdom, repeating numbers that have no meaning is appalling in terms of economic analysis. It miseducates the American public to the extent that they pay any attention to economic affairs, and it really gives us no guidance as to how to proceed.