MR. REISCHAUER: This will be a relatively informal discussion, and Howard will start by asking some questions to people around the table.
Howard.
MR. GLECKMAN: Thank you, Bob. And welcome everybody.
This really is a remarkable group. I think everyone here knows more about Social Security than I do. So, I'll just be the interlocutor and leave it at that.
Just a couple of ground rules. As Bob suggested, this is not going to be the typical panel discussion where you all sit and listen for 45 minutes while people do opening statements. What we're going to do is throw the table open, make it a real roundtable, and I encourage everybody to participate. One bit of housekeeping, I'm told that these microphones are rather directional, so keep them close to you when you talk. And please, because it's being taped, please introduce yourself. And speaking of introductions, I think it might be a nice idea if we just went around the table and had everybody introduce themselves. I'll start and we'll head over towards Bob and go around.
I'm Howard Gleckman.
MR. REISCHAUER: I'm Bob Reischauer, and let me say that Steve Goss and Bill Niskanen are somewhere caught in traffic. So I will introduce them.
MS. MOON: I'm Marilyn Moon from the Urban Institute.
MS. MARCUSS: Rosemary Marcuss from the Bureau of Economic Analysis.
MR. TODER: Eric Toder from the Urban Institute.
MR. THOMPSON: Larry Thompson from the Urban Institute.
MR. PRIMUS: Wendell Primus, Center on Budget and Policy Priorities.
MR. BRIEN: Mike Brien from the Council of Economic Advisors.
MR. IDASZAK: Jerry Idaszak, Kiplinger Letter.
MR. RICH: Spencer Rich, National Journal News Service.
MR. MINARIK: Joe Minarik, OMB.
MS. GORDON: Nancy Gordon, Census Bureau.
MR. CULLINAN: Paul Cullinan, Congressional Budget Office.
MR. WILLIAMS: Bob Williams, Congressional Budget Office.
MR. MATZZIE: Tom Matzzie, Campaign for America's Future.
MS. FITZPATRICK: Christina FitzPatrick from the National Women's Law Center.
MS. KUSKO: Andrea Kusko from the Federal Reserve Board.
MS. RENO: Virginia Reno from the National Academy of Social Insurance.
MS. BAKER: Linda Baker, General Accounting Office.
MR. BAKER: Dean Baker, Center for Economic and Policy Research.
MR. SAMMARTINO: Frank Sammartino, Urban Institute.
MR. HELMS: Bob Helms, AEI.
MS. TANAKA: Susan Tanaka, CBO.
MR. CRIPPEN: Dan Crippen, CBO.
MS. KOSTERLITZ: Julie Kosterlitz, National Journal.
MR. SEIDMAN: Burt Seidman, National Council of Senior Citizens.
MR. ROSS: Stan Ross, Social Security Advisory Board.
MR. BOSWORTH: Barry Bosworth, Brookings.
MR. HOLZER: Harry Holzer, Urban Institute.
MR. LEMIEUX: Jeff Lemieux, Progressive Policy Institute.
MR. GOSS: Steve Goss, Social Security Administration.
MR. PENNER: Rudy Penner, Urban Institute.
MR. STEUERLE: Gene Steuerle, Urban Institute.
(Several people introduced themselves off mike.)
MS. FAVREAULT: Melissa Favreault, Urban Institute.
MR. GLECKMAN: Okay. I think we've got everybody.
I brought along four pieces of paper that are merged into the response to the trustees' report. And, strangely enough, Dan Crippen also brought the same pieces of paper to show me this morning. One of them is a Wall Street Journal article entitled, "The Future Looks Rosier for Social Security and Medicare." The next one is a Washington Post article entitled "Rosier Future for Medicare Is Forecast." If you think you see a trend, I also have. "New Social Security Trustee Report Shows Long-Term Problems Remain." That's from the Concord Coalition. And "Social Security Reveals Growing Long-term Deficits." That's from Congressman Jim Kolbe.
So, now Steve Goss has come in, just in the nick of time, and with that level of consensus that I've just described in mind, I would like to get started by asking, Steve, if you could run down what the trustees actually did have to say on March 30th.
MR. GOSS: Thank you very much, Howard.
I make any remarks to you humbly, given Marilyn Moon, I assume, is here who was until very, very recently one of our trustees, and Gene Steuerle, of course, is here too, who is the chair of the technical panel that brought up the latest batch of advice and recommendations to the trustees.
This year's trustees report, all the reports that Howard mentioned are exactly right. Things are a bit better, but the significant problems for the more distant future are still very much out there. We did, for this year's trustees report, have some improvement, first of all, in the date at which the cost of the program will first begin to exceed the tax revenues. In the '99 report, that was 2014; that's one year further out this year to 2015. And the year in which the trust funds are projected in the intermediate assumptions to be exhausted and no longer be available to augment those tax revenues to pay benefits will be the year 2037, three years further out in the future than the 2034 that was projected in last year's trustees report.
The actuarial balance that's referred to oftentimes as sort of a measure of the whole 75-year status of the program improved fairly significantly from a 2.07 long-range deficit as a percentage of payroll down to a 1.89 percentage payroll deficit. This is something like the third or fourth year in a row in which there have been some sort of small improvements in the long-range prospectus of the program over the 75-year period, but roughly 2 percent of payroll is still a very, very large number, and one very significant additional measure is the extent to which the cost of the program exceeds the revenue that is scheduled to be coming in as of the end of the 75-year period, which decreased this year, fortunately, from about 6.4 percent of taxable payroll down to about 6.1 percent of taxable payroll. So there was sort of some good news all the way around.
The changes that were made in the trustees report, there were a variety of them, and I'm sure to one extent or another we'll be talking about a few of those today. A number of them, at least in part in response to recommendations made by the technical panel late last year, there were some changes in the mortality assumption resulting in about a one-third increase in the rate of decline in death rates that are projected under the intermediate assumptions from what was used in last year's trustees report that, in fact, worsened the long-range cost prospect of the program. However, there was, in addition on the demographics side, a small increase in the expected total fertility rate, or birth rates that are expected for the future, from 1.9 to 1.95 children per woman in the long run.
On the economic side, there were some fairly significant changes made. The productivity assumption for the trustees was increased by 2/10ths of 1 percentage point, which, in and of itself, would have resulted in an improvement in the long-range economic forecast equivalent to what had been recommended by the technical panel. However, the trustees, in looking very carefully at the assumptions, decided that there was one additional assumption, and offset half of that gain, which was to change the relationship between the CPI and the GDP price index, which the trustees had been projecting a .1 percentage point faster growth rate in the CPI than in the GDP price index in the '99 report, and for the 2000 report they increased that to 2/10ths.
In addition, the last thing I'd want to mention is, there were a number of—each individually fairly small—changes in methodologies, and as an actuary at the Social Security Administration, I have to sort of raise my hand on this one because the trustees make the decisions, the ultimate decisions, on all of the assumptions that go into the trustees report. However, the methodologies that were used to translate these assumptions into the projections are things that are largely, or really exclusively, under the control of the staff at Social Security, of the actuaries. And each year we look very carefully at all the methods we use. Any problems that we find we move immediately to fix. And this year we found five problems; unfortunately, four of them all went in the same direction. It's nice if they sort of split 50/50. And the sum total of those moved towards actually improving the long-range financial outlook as it is estimated for the trustees report.
But I would hasten to mention that if you look collectively, since 1983, the totality of all changes that have been made since then actually go the other way to a fairly significant extent, having worsened the long-range outlook of the Social Security trust funds—and this percent of payroll concept by a little bit over a half of a percent of payroll since 1983. Our hope is that these changes in methodologies over long periods of time tend to sort of wash out. We'll find as many that improve the system as hurt the system. This year it went a little bit in the direction of improving the system.
MR. GLECKMAN: Thank you, Steve.
I would like to ask Marilyn Moon, who was a public trustee, I guess just retired public trustee, a little bit about the decisionmaking process that happens there. The technical people present you with this information, what do you do with it?
MS. MOON: Well, we struggle with a great deal. We start in the fall looking at the long-range assumptions and talking about what the long-range assumptions ought to be. We spend quite a bit of time looking at past trends, and the question is, do you look over 75 years into the past, do you look 20 years into the past, do you look 10 years into the past, and you always get different answers, depending upon which period you choose. And, in fact, some of the variables that we look at in terms of these assumptions go up and down in fairly unpredictable but dramatic ways over a long period of time, in 10-year, or 12-year, or 15-year increments.
So, it is always a puzzle to try to figure out what happens. We also, I think, have taken the tact in general to think that the changes, unless there's something that really comes out like a technical change in the way something is measured, we try to be very careful about making changes from one year to the next in dramatic ways because it's very hard to argue that from one year to the next you suddenly have a dramatic difference in terms of what our 75-year outlooks. Even very good economic times can't necessarily translate into a 75-year time frame.
I was very impressed in the process of all five years that I worked on the trustees reports that contrary to what I've sometimes read in the press, there was really not a lot of political discussion at all. It was really wrangling over these numbers and looking at them, usually each assumption separately first, and then thinking about the total package. But with a lot of emphasis on just the particular numbers and, in fact, sometimes being surprised at how they fit together at the end.
The other thing I would say is that this is a process in which we could probably take some satisfaction that we've done somewhat the right thing because we've been criticized on both sides of being either too optimistic or too pessimistic. For example, the mortality assumptions, I'm sure there are people that would like, as in the technical report, for us to have gone further on, and on the productivity side, there are some concerns about whether we are too pessimistic. The other thing that I found that was really interesting in discussing this over time is, when you bring in outside experts, which we occasionally did, both inside and outside the government, to give us some advice, the first thing people would say is, I feel very confident that over the next four quarters, dash, dash, dash. And then you'd say, well, how about five years? And you could probably get them to tease out something over five years. And then, inevitably, when you'd say 75 years, they would laugh. So there really aren't very many people around who do these kinds of really long-range looks, and it's very hard to get people to focus on that.
So, I think that I would say that the actuaries do a very good job of providing us voluminous amounts of information, and the discussions are quite serious and not very political in nature, and then you get what you see here today.
MR. GLECKMAN: Thanks, Marilyn.
One group that was courageous enough is something with a really wonderful title of the Technical Panel on Assumptions and Methods Report to the Social Security Advisory Board. Gene Steuerle, Barry Bosworth were both on that panel. And I wonder if Gene could talk a little bit about the courageous assumptions that you made. I get the sense that you felt a little freer to get out on a limb on some of these issues, but talk a little bit about what you suggested.
MR. STEUERLE: Well, actually, Marilyn had it exactly right in terms of the 75-year numbers; that's probably the last thing the panel wanted to do. In fact, one of the difficulties is, we made dozens of recommendations, and most of them were with respect to modeling and recommendations for improving the ways that Social Security might be able to do things, for instance, do more microsimulation and analysis, things like this. That often hits the press—when you say "microsimulation analysis," they're really quite excited about that.
And then we had a lot of recommendations about presentation of the report. We did feel like some data was misinterpreted by people and might be presented in a better way. And those recommendations came fairly late in the trustees process, and how the trustees adopted it and will accept them over the years will only be known over time.
Finally, we made a few recommendations with respect to these sort of intermediate assumptions. That's really not where we spent most of our time. That's what got most of the attention from our report, mainly because one recommendation we made with respect to mortality ended up affecting the long-range cost estimates and increasing costs somewhat. But, in point of fact, if you dodge that issue, we didn't really make a recommendation of any bottom line. That's just sort of the way our numbers came out. If you actually look at what we recommended, and what the trustees did, they actually moved a significant degree in directions we had suggested. They did move in the direction of improving mortality. They did move in the direction of saying that productivity was increasing. We also said that you should take it slow and adjust over time, which they did.
A couple of areas we examined I think perhaps led to decisions made by the trustees, even though we didn't make the step. For instance, we were almost ready to recommend an increase in fertility rates. At the last minute, we had some additional data that sort of had us back off, and the trustees sort of went halfway between where we thought we were going to go at the last minute, and where we didn't go. So I think our discussion may have influenced that process. And then we recommended to the Social Security actuaries that they examine a couple of areas that looked funny to us. For instance, the labor force participation of women in older ages. The lack of later data on marriage and divorce. We didn't have recommendations there. We just thought that some improvements could be made, and those recommendations could have influenced the actual technical adjustments, which are actually what drove the trustees report actually into saying there was an improvement.
If you actually take out the technical adjustments, so-called "technical adjustments," they made, they basically had about the same long-term actuarial imbalance they had before.
So I think the two processes worked as they should have. We were a group of advisors. We made a set of recommendations, and the trustees accepted some of them. Some of them they thought, well, we'll watch over time and see whether we agree with them. Some of them they probably rejected. A lot of them they just didn't have time to get around to even examining, particularly with respect to the presentation issues.
Maybe Stan might want to add; Stan was chair of the advisory board that set up the technical panel. I don't know whether he wants to add to that or not.
MR. ROSS: I think what you're saying is a process that works pretty well and even-handedly. I was Marilyn's predecessor as a public trustee and did it for five years, like she has, and I would have given the same answer. It's largely nonpolitical. In fact, it's entirely nonpolitical. I was appointed by President Bush and did a transition to President Clinton, where David Walker went from being with the majority to being the sole Republican. I went from being the sole Democrat to being with a lot of other Democrats. But it didn't matter. In this process it's pretty nonpartisan for the most part.
And this technical panel, I think, in some ways, reflected the change in the 1994 legislation. They traditionally were appointed by advisory councils in a sort of, I don't know, an irregular fashion in terms of timing, but they seemed to happen enough. We now, as a board, have an institutional responsibility to do this periodically, and maybe to do some follow-up. And I'm kind of curious about that. I would put one problem on the table, which I think flows from what you've heard so far. Which is, I think at the level of the people around the table here, the process works very well—not a lot of big changes from year to year. People understand it. It kind of grinds itself through over the course of the year. But there's this tremendous gap between these books and the public, and the public perception.
Those four article headlines you read, one could argue, are all correct. It depends. You know, it's in the eye of the beholder, and frankly, in terms of that, the last innovation was 10 years ago when Walker and I thought of this short summary, and the first year did it ourselves because the institutional network process didn't think you could possibly summarize this and do it justice. And this has sort of become useful to a lot of people because it's short.
But as I sit here and listen to all this, it seems to me that the place where attention ought to be put is how to get from here to something that the public can really understand whether there has been much difference, or does it make any difference, or what the hell is going on with all this arcanery. And then, I think, it's been a subject that's received too little attention, although Gene's technical advisory panel did have some thoughts in that direction, which have largely been unnoticed, that I would like to see noticed.
I'll stop.
MR. GLECKMAN: Thank you, Stan.
I think now that we've got a pretty good sense of the process, let's see if we can drill down a little bit and talk about some of the assumptions. And I wonder if maybe Barry Bosworth could talk to us about one of them, which is the new economy. My magazine, among others, is, of course, madly in love with the new economy. And one of the great criticisms one always hears about not just the Social Security report but the CBO forecasts and the OMB forecast is, you know, what's the matter with these people, don't they understand that productivity has reached a new plateau? And, you know, don't they get it?
Barry, I wonder if you could talk a little bit about the assumptions that are in here about the new economy, and where you see them?
MR. BOSWORTH: Well, I think it was like Marilyn said, the point that she made that you are torn in these projections between short-run and long-run. I personally think, going through this process, just after a while you're arguing over the tenths of a percentage point and growth rates. I think it's sort of futile going out 75 years to do that. One of the things that we had suggested was just adopting a rule. So, quit pretending that we know how fast to put in something like the new economy, adopt a rule of the last 50 years, or you could use geometric weighting if you wanted to give more attention to the short run.
But maybe all this discussion of a tenth here, a tenth there is hardly worth it in some respects. What we found on the new economy, sort of arguments that people make, for example, if you look at the last five years, it's unambiguous, I think, that there's been a dramatic improvement in rate of growth of productivity in the U.S. economy. The five years prior to that, though, is pretty damned bad, it was somewhat below the historical average. So, do you want to take a five-year average? Do you want to take a 10-year average? If you take a 30-year average, it looks awful, because then you're back to the 1973. If you want to do 50, it starts to look a little bit better again. And so it's very difficult to know with these short-run trends how much to factor them in.
My only, I guess, anyway, sort of criticism maybe of these, I think that the revisions were very conservative on the economic side. And the reason is that the technical panel was actually meeting before the national income account revisions were done. And there we thought there was a case that there has been a pick-up in the rate of growth of productivity, and there are some other changes that would argue for a faster rate of growth in real wages by about two-tenths. But then along comes the national income account revisions, and they raised the rate of growth of productivity in the U.S. economy by about two-tenths. Now, that doesn't do anything to the real wage projections. That just changes the productivity number, because nobody's real wage is going to grow faster in the future just because the Bureau of Economic Analysis says GMP is growing faster.
But, still in these components, there was two-tenths in productivity. So I look at it as the technical panel just put in the revisions as far as the productivity growth, but the real wage assumption is only up a tenth. That's a pretty small change. I mean, I think you could have made a case that the new economy is doing a bit more than that, but I think the basic one is how are you going to fight to the death over a tenth of a percentage point here or there. It's very hard.
So I would argue that Marilyn's final point was the right one, that you want to have revisions from year to year, but you want to make them sort of continuous and smooth, no big changes unless some really new piece of information comes along. I think the direction of the changes in the future is obvious. I'll bet you that the economic growth projections steadily go up, and next year will probably be a little bit higher than this year, reflecting the fact that one year after another you get better news. But we didn't have any big differences, I don't think, with the existing assumptions.
MR. GLECKMAN: Let me ask Rosemary Marcuss for a little different perspective, from the BEA's perspective, about what you are seeing, in terms of the new economy, and whether a tenth here or a tenth there really does make a difference, and how far you can see?
MS. MARCUSS: Thank you, Howard.
I have to be very careful here, because I'm the only person in the room who has actually been called "Rosie Scenario." All the wise things said heretofore have really made my case a lot easier to make. Now, if voting for a new economy is tantamount to buying the NASDAQ, my job just got a little easier. One thing I want to mention is the word "new": that can mislead. I mean, we all know that economies morph all the time. Everything changes, things are new. But here we don't mean just new. When people talk about the new economy they're talking about something that is excellent beyond all generally accepted assumptions up to a few years ago. That's a higher-risk bet.
Now, as to what BEA is seeing, I'm daunted by looking around this room. It is chock full of very astute observers of the current economy. So I will just say that BEA is seeing just about what you are seeing. But, the story here is really about people who are involved in the very serious job of long-term economic projections. And as Barry and Marilyn, Jane, really everyone has emphasized, that's a world of compounding everything. It's the world where the jumping-off point is all-important for that 75-year story. It's a world (in fact, Marilyn said this and I just so identified with it) of constantly having to explain how just one year made a difference in your forecast of some year 40 years out. That's painful. I've done it many, many times.
So back to the new economy. It is still a hypothesis. Notwithstanding the fact there's a lot of evidence, as Barry points out, in recent years, of very fine machinery moving better than it used to, that's true. But as the Social Security Commission's technical panel selects your portfolio of assumptions, I think you want to diversify those assumptions to keep the variants of your projections over time smaller rather than larger. I think this is kind of what Marilyn and Barry are emphasizing. So by taking the last 5 years, where we've clearly got things booming along, and projecting that for 75, you're really not minimizing the variants of your assumptions. You're making your life a little bit harder.
Now, another daunting aspect of what you're doing is you're not just forecasting productivity, you're forecasting death rates, you're forecasting the relationship of the CPI to the GDP deflator. It makes productivity look like an easy assumption. So I'll just summarize by saying, yes, we definitely do see a lot of new things happening. But in this context, to project the good news of the last several years out for 75 years is clearly something that no one who is doing it thinks is a good idea.
MR. GLECKMAN: Let me ask Dean Baker, who I think may have a little different perspective on this, to talk a little bit about whether Rosemary is right. If everything we have heard so far is correct, should we be so cautious, should we be so careful about forecasting out like this?
MR. BAKER: A couple of points. First I should say, I'm going to have to leave in about 10 minutes, so people shouldn't take unnecessary offense at that. But there's a few things. First off, I'm not a big new-economy fan myself. But, I think it's important to realize, we do have very, very pessimistic assumptions in light of the historic experience. And one of the points I've made is that if you look back to '96, we had a 1 percent real wage growth; I'm skipping over the productivity and just focusing on the real wage growth, because obviously that's what matters here. We had a 1 percent real wage growth projection, that's what we have again, but we've changed the CPI by at least five-tenths of a percentage point by just about anyone's reckoning. So we've made things actually much, much worse over the last five years, or four years, rather than better.
But, the deeper question, we're looking at 75-year projections; all of us agree we don't know anything about the year 2040, 2050, 2060. And the question is whether this is a really good way to focus policy. I'm sort of struck. I clipped an article out of the Post, "The Bidding War," which criticized the vice president's proposal to extend benefits to older women. And it was striking because their position was they agree this is a good idea, but they say first we have to get our financial books in order. Now, keep in mind, we're all looking out 30, 35 years; the books are in order for that far—50 years, 60 years, maybe not.
Let's say we were going back to 1965 and debating Medicare. And we applied the same standards of fiscal prudence, and at that time we were much more optimistic, but let's imagine we actually saw the future, so it's 1965 and we realize that the baby boom is going to come to an end, we realize that the golden age of productivity growth is going to come to an end. Well, if we applied the Post standard, there's no way on earth we would have adopted Medicare. And the question I'd pose is, would that have been the responsible thing to do?
In my mind I'd have to answer no. But unless you think that would have been the responsible thing to do, to drop Medicare because it was fiscally irresponsible to adopt in 1995, then you would have to reject the Post logic in saying that because we don't have this balance for 75 years we can't go ahead and maybe do something that let's say at least many of us think would be a good idea.
MR. GLECKMAN: Does anybody else want to respond to that; does anybody want to jump in?
MR. STEUERLE: One thing the technical panel tried to get more focus on, it perhaps didn't succeed, because it wasn't the first to do this, is to try to figure out some way to demonstrate more in our language to focus on the variants rather than the mean. That is, the uncertainty of policy itself is something that we hope—the uncertainty of outcomes is something that we hope we'd be better able to demonstrate. For one reason, because policy is something that can actually amend it. Policymakers tend to think they need to get this intermediate projection down to a zero long-term deficit, when in fact you could get to a zero, 75-year actuarial balance and still have a 50 percent probability that you would be out of balance the next year, and perhaps a 30 percent probability that you'd be out of balance by 2 percentage points of payroll, or whatever measure you had. And those measures actually are—that is, that variants are something that's amenable to policy.
An example being, the easiest one perhaps,especially in the demographics, if you adjust as life spans go up, if you index for mortality, for instance, you reduce substantially that uncertainty of future outcomes. So the uncertainty of future outcomes is not just an issue of whether we can predict the future, it's also an issue that's amenable to policy. And all we tried to do as a technical panel is to try to figure out ways that we could demonstrate that better to policymakers. So they thought of that as not just something they couldn't deal with, because nobody knew what the right answer was, because Barry doesn't know what productivity is, or Rosemary doesn't know what productivity is, but that you could actually think about designing a system that doesn't have that degree of uncertainty in terms of long-term deficits, if you want that to be a target of policy.
MR. GLECKMAN: Dan Crippen, if you want to jump in here, you wrestle with these issues in your own context at CBO, what do you see there, what is your ability to forecast out? I know you talk about how hard it is to forecast out five years.
MR. CRIPPEN: We like to let everybody else do it. We have, as you know, Howard, and many of you do around the table as well, under way an effort that others are as well, at the Urban Institute, Frank and Rudy, trying to enhance our capabilities in the longer-term projection of both Social Security, which is the easier case, of course, and then Medicare. With the help of Steve and his colleagues, we hope to be able to at least mimic the actuary's results as a way to anchor or test or put parameters on our other work. We hope to be able by the end of the year to have a microsimulation capability, not unlike what Urban is working on, and others, and Steve and Barry are working together on. And then ultimately to be able to, as I refer to it, go above the actuarial model and work on what the macroeconomic effects of particular policy changes, or outlook changes even for the trust funds, might be.
So we're all, I think, getting a little more sophisticated in our ability to ask, perhaps, better questions about policy—what is the import of this change, does it necessarily adjust the impact it has on trust fund balances, on the economy, on distribution, between cohorts, among people—a whole range of other questions which we heretofore, while we may well have had the questions in mind, we certainly didn't have the capability of answering very well. We're hoping now to be able to refine those questions and, in turn, answer some of them through better modeling techniques, through enhanced computer power, frankly, and so we may be here a year from now saying exactly the same thing.
Certainly, these models are not going to take the variants out of forecasts for 75 years from now. But they may help enlighten the debate. In 1982 and '3, when many of us, or at least some of us, were involved in the Greenspan Commission effort, we didn't ask very many questions other than what effect does this have on the trust fund balance. That is, any policy proposal generally was aimed at restoring some sense of solvency, because we were looking at impending insolvency. It turns out that's one question, but only one of many, and we didn't ask many of the others then. Hopefully, as we improve our capabilities, we will be able to address some of the more potentially important economic questions.
MR. BOSWORTH: I was just going to mention, it used to be that most of the focus in the technical panel was the same. It was always on the very long run, because if it was an anywhere pay-as-you-go system, where there's no reserve, the long-run assumptions tend to dominate. But nowadays you have a very heavily funded system, and that is not, I think, as you look at the report, you'll find there's probably not enough attention to what's going to happen over the next 10 years. Very quickly it drops to the long-term assumptions. Whereas, if the economy continues to grow rapidly for just the next 10 years, you'll accumulate a lot more reserves than you had previously anticipated and that compounds, too.
Interest rates, rate of a term, now make a much bigger difference in the projections than they used to. Ten years ago nobody paid any attention to what the interest rate assumption was, it didn't matter, [in] a pay-as-you-go system. But it's in-between now. And so I think one is, maybe we focus too much on the long run and how terribly uncertain it is, and adopt some sort of rules. But the shorter transitional period also has to get more emphasis in the modern projections.
MR. GLECKMAN: Bob?
MR. REISCHAUER: Just a comment on what Gene said, and to a lesser extent Dan. Both of them were talking about the desirability of maybe designing systems which reduced the variants of the estimate from the government standpoint. And all of us policy wonks can see that as a desirable objective, but let's keep in mind that you don't want to do that by increasing the variants of replacement rates for future beneficiaries. You know, you can reduce—make the variant zero from a government standpoint—and have a really crappy program for the basic purpose of Social Security, which was to provide people with some benefit that they could count on, replacing X percent of their wages.
MR. GLECKMAN: You know, we've been talking about the uncertainty, and I think one of the interesting issues here is of course the policymakers, the politicians do not treat this as if there's any uncertainty; they treat this as if this is carved in stone from year to year. And they make their decisions, or I guess in the case of Social Security don't make their decisions, based on what they see in this report. And I wonder in some sense if the report isn't, I don't want to say misleading, because it's obviously for what it is not misleading, but I wonder if it lends itself to a danger and I wonder if maybe someone here could talk about the risks.
Yes, Bert?
MR. SEIDMAN: I'm Bert Seidman, with the National Council of Senior Citizens. I agree very much with Barry Bosworth that we should be focusing on a much more understandable period than the 75-year period. I spent my life in the labor movement. And when we were in educational sessions we just ridiculed the 75-year idea. And the easiest way to do it is to subtract 75 years instead of ahead 75 years. And the people we talked to simply felt that it was absolutely ridiculous to be doing this. On the other hand, I don't think we should ignore the periods after the short period. But I don't know that we should attach figures to them. In other words, we could say, well, after the 10 years, with much less certainty, we think this, and this, and this is likely to happen, which would affect the Social Security trust fund in a manner that could be this or could be that. It seems to me you'd get a much more understandable report if you had that.
The other thing that I think is unfortunate, and the technical committee brought this out a number of times, it seemed to me, is the assumption that policy doesn't change. Whereas, we know that over and over again the policy of Social Security has changed when we saw problems with the status of the trust fund. And to have figures come out every year that's on the assumption that the policy doesn't change when we know that the policy will change, it seems to me, also detracts from the credibility of the report.
MR. PENNER: I think Bert's last point is very important. But I think there are some other problems as well. The report draws attention to when the trust fund goes broke, what the actuarial deficit is, and so forth. And these things are really not very good indicators of what the system costs society and the economy. Now, it's really hard to measure the true economic costs. A naive way of looking at it is to look at benefits relative to GDP. And what that shows you is that the problems start long before the trust fund goes broke. I think there's a tendency to believe there is no problem until the trust fund is empty. But if you look at when the benefits really start to rise, that happens as early as 2010. Now, there has been some improvement even in that measure in this new report. Indeed, quite a bit of improvement, the ratio of benefits to GDP go down by about 5 percent roughly, both in 2010, and 2030. Interestingly enough, they go down in the short run, indicating that the problem, while at a lower level, gets worse a lot faster. In fact, that may draw our attention to it.
But even benefits relative to GDP are not a very good measure. In the ideal we'd like to know, what does this system do to savings, what does it do to work effort, these are the real measures of the economic costs of the system. But there isn't much academic consensus on these matters, unfortunately. So we, I think, the report tends to focus on relatively unimportant things, when we should be striving to know a lot more about the really important economic phenomenon connected to the system.
MR. MATZZIE: I just wanted to respond to this political—this point about—I'm sorry, my name is Tom Matzzie, Campaign for America's Future—interacts with the public's perceptions of the Social Security system. The public does not think about what the system costs society and the economy. They think about the Social Security system in terms of their retirement, their parents' retirement, their family's retirement, survivorship, disability. And I think that they are willing to tolerate some economic inefficiencies, so to speak, if you think it's discouraging work or if it's a drain on the budget. They're willing to let Social Security budgets soar, even if the federal budget is not balanced. And those perceptions, I think, are important to understand for policymakers, and also for those of us around the table who influence the perceptions of the system. It's not something to forget.
There are two observations we've had in the last couple of years. One, longer projections create a downward pressure on benefits. The longer you project out, the uncertainty of outcomes increases. And so from our perspective, Heidi Hartmann is sitting over here, we want to increase benefits. We agree with the vice president that benefits ought to be improved. So we would like to see some discussion of why 75 years. Why do we look out 75 years at all? It seems entirely arbitrary to me. In fact, I've studied the history, and it still seems to me—I mean, I've talked to Bob Meyers about it, and it still seems arbitrary to me. Why not 30 years? Government bonds, we only insure them for 30 years. Is there any other liability that goes past 30 years?
MS. HARTMANN: I was going to make the same point, that it doesn't seem to me that we need a 75-year projection, and maybe we should consider modifying the law or whatever it is that makes us have a 75-year projection. And when Bert Seidman said, we've made changes in policy when we've taken a look at the solvency issue in the trust fund, we've also made enormous changes to increase benefits, or as the case may be decrease them, although I think historically increases in benefits have been far more common than decreases. Of course, recently we've had the increase in retirement age, which is a decrease in benefits; we took away survivor benefits from children in college, which is kind of nonsensical when you look at how important a college education is, and how important survivor benefits are, particularly to minority children, how important it is for minority children to go to college. So we have had some unfortunate benefit cuts. But I think that the general trend has been toward benefit increases, and we would certainly argue that benefit increases should be on the table.
I mean, much as with Medicare, I think offering a prescription drug benefit puts a lot of political muscle and political motivation behind solving the Medicare solvency crisis. I think the same thing would happen with Social Security if people saw some much-needed benefit improvements; they would start thinking about, okay, what do we need to do to pay for those benefit improvements? So I think it's kind of a mistake not to have benefit improvements on the table, and part of it is an adjustment to changes in family structure, which I know the Urban Institute is also working on. We're going to see not less in the future but more never-married women, more divorced women. Right now most women's benefits are based on their husbands' earning records; that's not going to work for women in the future. So there are a lot of reasons to be talking about enhancing benefits, particularly for single women; those are the group that have the highest poverty rate right now.
MR. THOMPSON: I've done a lot of work in other countries in the last four or five years. And I've been impressed by thinking back about the people who set up our Social Security system, and the number of wise things they did that you wish other countries had done at the same time, or even more recently. I won't go into all the details. But one of them is this tradition of these public actuarial projections off into the future. And I'd like to say, I think they've served a very valuable purpose, and we should not forget that, so that we shouldn't make modifications which might undercut that purpose. They have made the Congress behave really responsibly. And quite in contrast to other topics, at other parliaments, with respect to their Social Security programs they've always been told, when they increased benefits they always had room or they financed it somehow. I watched a congressman, a senator withdraw his amendment to make some minor change in widows' benefits because the actuaries told him it would cost .007 percent of payroll. And he said, if it was going to be that expensive he didn't want it.
So it serves a very valuable purpose, this tradition of issuing these annual reports. Unfortunately, the corollary to that is they're really misinterpreted and misused. I mean, we have a report here which says we're going to project two independent trends. They're inconsistent trends. As soon as you get to 2037 it's inconsistent. We're now making projections of things that can't possibly happen, because we're projecting benefits that can't actually be paid. But it's useful for you to know what happens to these two trends off into the future. And the policies will change, they have to change, the law requires them to change.
The trustees' report is projecting a world that cannot happen, as soon as the trust fund becomes exhausted. And as we begin to lay more and more criticisms on this, we run the danger of losing sight of the original purpose, which was to just project these trends, and make the Congress and the people focus on where the current policies would take us, and whether they're consistent or inconsistent, not how the inconsistency gets resolved.
As far as how long we project into the future, 25 years ago I was a staunch opponent of proposals to shorten the period, because that would have shortened it to the point where we would have truncated the retirement of the baby boom generation. Now, it seems to me you can have an intelligent conversation—I don't think 10 years. I think 10 years is too short. But 50 years is probably ridiculously long, as it is, and will capture most of what you have to capture in terms of pretty predictable demographic changes. And in this business, the farther you go the wider the variance gets, because the more one-tenth of a percent in a growth rate causes things to just balloon. Either they shrink to nothing or they just balloon out. So I would propose that some shortening of the period is something to be considered now.
And I think, along with what Gene and Bob said, that the public policy debate under-appreciates the importance of the uncertainty in these projections. We deal in a world in which we may have a big problem, we may have almost no problem at all, and although one can have academic discussions about how to refine the range, the point of the matter is it's still quite large. And public policy likes to deal with point estimates, and is there a way that we can at least as analysts be aware of the uncertainty and try to deal with it. Now, I'm not sure indexing the system is—I will say, in support of what Bob said, recognizing that a defined benefit system allows you to take the uncertainty and spread the misses across generations, you don't have to concentrate them all on a particular cohort of retirees.
Beyond that, I think it's asymmetric; you're better off having people oversave than undersave, even though there's no reason in economics why it should be that way. I just feel it probably is, which would mean that you might be better off overfinancing the system than underfinancing it, so that adjustmentsare sort of—or adequate advance warning. I think it's an interesting idea, but it's a bad idea, because it will conflict with the first principle of the forecast, which is to keep the Congress behaving responsibly, and I'm not sure they'd behave responsibly toward a system that was intentionally overfinanced. So I'm stuck with, because people misinterpret these projections, and rather than just saying, here's what happens when two trends are projected independently, they reach the conclusion that there won't be any benefits there for me, which is not true.
So maybe what we ought to do is we ought to be putting in balancing proposals. And I suppose we'd have to agree on some combination of an out-year tax increase and an out-year benefit cut. I'd hope the benefit cut was a really obvious one, because the one that's going into effect right now is kind of a sneaky one, the retirement age increase. You have to have an obvious one, as obvious as the tax increase, so that as you get closer to the effective date of these things you have another debate about whether that's the right way to go, but at least you take some of the pressure off of the benefits won't be there for me when I retire.
MR. GLECKMAN: Wendell Primus?
MR. PRIMUS: I'd like to make just a couple of comments and that is just about what this means in terms of the politicians doing something to close the deficit. I think it's going to be extremely difficult. It's hard enough to get politicians to do painful decisions, in terms of raising the retirement age, or reducing benefits, or reducing taxes; now we have three or four, I forget, Steve, reports in a row where the actuarial deficit has gone down, if you plotted that little time trend of three or four points. The next one that comes out, if I was a politician, is going to be 1.6. I mean, we're on the right way. And Barry tells me the economic projections are probably a little too pessimistic even in this report. So if I'm a politician, why would I want to do anything painful?
And the second reason, I think, is the policy divide; it used to be the difference between Bob Meyers and Bob Ball. And in retrospect, that wasn't a very big difference. Now, it's a difference between the privatizers and the nonprivatizers. I mean, the consensus about what to do next on Social Security is just extremely wide. And therefore, I mean, for both of those reasons I don't see any reason, or any event, that will get Capitol Hill and the executive branch to agree on a set of proposals.
And I guess the last thing I'd say is, I hope that—and I'd like to improve benefits for some elderly widows, reduce poverty, the prescription drug benefit—but I hope that when we do those, we do them on at least a pay-as-you-go basis, or maybe at the same time improve the solvency of Medicare or Social Security as we do those political sweeteners. But I'm not so sure that will even happen. I like to make sure there's kind of a pay-as-you-go rule, if you will, within the system, in both systems, so that when we do those legislative changes that are obviously going to cost money, we do nothing to hurt the solvency.
MR. GLECKMAN: I'd like to ask Bill Niskanen to pursue the same issue, about how these reports are changing the political dynamic here.
MR. NISKANEN: First let me say that this report, as is the case in the past, is a conscientious, professional attempt to come up with an actuarial estimate. I find it wholly unhelpful, however, in guiding anybody on policy issues. In the immediate case, it has the effect of deferring attention to Social Security, because it looks like we've got another year of grace between now and the time that we have a negative cash flow. This report, however, over the years has contributed to the illusion that we have assets in the Social Security trust fund. It is still bound in the trust fund rhetoric, and arithmetic, which is quite irrelevant to anything in the world, other than the perceptions of some people. It does not add technical information that the group is capable of doing about what would happen if you make particular changes. There's no information in here that provides sort of partial estimates of what happens if longevity is longer, or real wage increases are longer. The optimistic pieces are put together in low cost projections, but those optimistic pieces include optimism about real wage growth and pessimism about how long you'll live. And so these are not pulled apart.
I think in general, with respect to our budget making in general, and Social Security in general, we should make our decision processes very much less forecast-dependent. To acknowledge that we cannot forecast with that much accuracy, however conscientious and professional the group is, and that we should look at the critical features of whatever program or policy we are addressing and say, what are the appropriate changes that ought to be made? Now, nothing in this report, for example, conveys the impression that this Social Security system is a very bad deal for most people under the age of 40. It has extremely low rates of return for most people under the age of 40. It has very low rates of return for two-worker families. It has low rates of return for people who have low expected life spans. Nothing in the report contributes any information to the critical variables that will affect policy choice on the structure of this program, and its benefit structure, or its tax structure, or the alternatives to the program.
Given the way that Social Security has been treated in the past, as I say, this is a conscientious, professional technical report that is difficult within that framework to see how to improve. But I think that that does not help the average congressman or senator or executive official to address whether Social Security should be opened up for examination, or in what way it ought to be changed. I heard a story yesterday about the challenge of translating some theological writings from German and Flemish into English. And the standard was apparently not very tight; he said that this material ought to be understandable by bishops. I think that the standard for what ought to be understandable about Social Security ought to be the average congressman, which may not be quite as demanding.
In any case, I think that the report doesn't help that person very much. It reads like it is written by a pension lawyer. The words are in English, and the sentences sometimes are coherent, but it doesn't help address, should I be worried about this system, what should be done about this system, and where do we go from here?
MR. GLECKMAN: We've got a lot of people who want to respond. Let's let Steve go first.
MR. GOSS: Just a couple of quick items. I'm not sure sort of where to start. First of all, Bill, we do have in this report, and for a lot of reports now, a section called the sensitivity section, which does individual assumption by individual assumption indicate the extent to which the long-range cost and the medium-term cost and the 25-year cost would be modified by changing one assumption at a time.
Just sort of another, almost kind of like an aside, the nature of the report really, I think, when we step back and look at the report we have to ask, what is this report really out there to do, what is desired to have happen from this report? The law actually requires the trustees to report annually to the Congress on the financial status of the program. And it requires that it provide a five-year projection of what the costs are. And what it requires beyond that is to indicate what the actuarial status of the program is. Now, when it says actuarial status it doesn't say should this be a 10-year, or a 1,000-year, or a 75-year analysis. It just says report on what the actuarial status is. In fact, for a number of years, I think it was prior to 1972, the trustees report actually provided estimates that went on into perpetuity; they looked forever in terms of what the cost of the program is. And that's been reeled back to 75 years. There are still, by the way—there are proponents, there might be some here today, who say we really should be looking to perpetuity. So this is another case where we're real happy when some people say you should go longer, some people say you should go shorter.
A couple of the rationales for the 75-year period that have been fairly convincing to the trustees, or at least some of the trustees—Marilyn and Stan might want to address this—is that 75-year period is something like the remaining expected lifetime in general of people entering the workforce, who are becoming new participants in the program, therefore giving some indication of what will happen during their lifetime perhaps makes some sense, well, lifetime to which they might expire, probably 75 to 80 percent of them.
MR. ROSS: Well, the life expectancy at age 22, the average age of death, is higher than at birth.
The other aspect of the 75-year forecast that I think is significant is that an awful lot of policy changes, or potential changes in assumptions that are considered by politicians or by the trustees, take a long time to really mature and have their full effect. And therefore having a longer-term forecast is really an important thing to have.
As for having, I think Bert mentioned the idea of the trustees report should be anticipating policy changes, and of course really, the trustees report has always been based on the notion of looking at what happens under current law. The very idea of it is to have it not anticipate what policy changes are, so that we can see what happens if we leave things as they are, so that we will know, therefore, when we do have to make changes in the future. This is really just an indication of what will happen if we don't make changes so we'll know when we must make changes.
And one other thing, too, about sort of the variability of the costs, I think Wendell had mentioned that the costs had been coming down for I think the last three years, since 1997. Since 1994 this actuarial balance has been a 2 percent deficit, give or take .2 or something like that. So really, the level of this deficit, if you step back from it, the .1s and the .2s, has been really fairly stable for almost a decade. And in terms of the uncertainty for the future, the nature of the program right now does have some of the things that Gene has mentioned about reacting to changes that might occur in certain economic variables in the future; it is wage indexed, so it does respond to some things. In the demographics the program is not designed now to respond to changes in the demographics, and that's why really the demographics are the most important variable for the future. But, on the other hand, the baby boom did happen, birth rates are down to a relatively low level now, and therefore much of what we need to know for the next 50 and even 75 years for the demographics is pretty much laid out there.
So there is clearly uncertainty in a lot of it for the future, but to sort of complete the loop and go back to the sensitivity section, one thing to keep in mind is the extent of the sensitivity, this actuarial deficit that we talked about, about 2 percent of payroll. If we had, say, about one-tenth, or even two-tenths, increase in the real wage growth assumption, each one-tenth increase in the real wage assumption reduces that 2.0 percent deficit by .1, by only 5 percent of the size of that deficit. So the sensitivity section of the financing department is not really all that enormous.
MR. GLECKMAN: I think Jim Klumpner in the back wanted to jump in first.
MR. KLUMPNER: Yes, this topic may have been overtaken a bit by others. But, one of the problems in the political arena—I work for the House Budget Committee—is that having a very long-term forecast like this, where the out years are subject to tremendous uncertainty, as everybody in this room well understands, one of the problems with this in the political arena is that it creates the premise that we should do all-or-nothing kinds of solutions. That it completely pushes incremental approaches to the problems off the table, that there's a presumption amongst political actors, at least in the debate and what they say, that this generation has a responsibility to solve the problem for all future generations.
And that, I think, really does have a pernicious effect, because it pushes these incremental approaches off the table.
MR. GLECKMAN: CBO wants to be heard from. Susan Tanaka first, then Dan Crippen, and we hope they agree with one another.
MS. TANAKA: I want to echo what Rudy and Bill have said and go back to Howard's question about what happens with this report. I think if the federal government only offered Social Security, this might be a fine document. But the problem is I think there's too much weight placed upon this document, because it tells us lots of interesting and useful information, but it doesn't tell us what kinds of budgetary and political pressures are going to be on the population and our policymakers in 10, 20, 30 years, because it acts as if the government only does Social Security. It seems to me that it's kind of like using a thermometer to tell what time it is. It's measuring something, but it's not terribly relevant when you try to think about the overall budgetary pressures and the programmatic needs that the federal government is going to face in the future.
So somehow with all this collective brainpower, and all this understanding of the uses, and the limitations, and the abuses of this information, it seems to me we ought to be able to come up with a better measurement to help the politicians so they can come up with a better goal besides save the Social Security surplus, or extend the solvency of the trust fund. We need to find something that guides them and provides them with a structure to make more informed policy decisions.
MR. CRIPPEN: Just a few seconds, I did want to return, as Susan just did, to the proverbial elephant that Rudy introducd to the room, which is that the trust fund status is not particularly relevant for economic, maybe even public finance, purposes. We can look at 30 years instead of 75, and those 30 years are enough to inform you that you're probably going to have to change the nature of either the programs or government as we know it. Programs for the elderly will grow from something like 7 percent of GDP today to something like 14 percent of GDP by 2030. And that's 14; it could be 13; it could be 15. It's probably more like 18, but the point is we're going to have to do something different than we're doing today in terms of government. We're either going to have to borrow more, raise taxes, or cut other spending, and those activities by the government are going to be the same whether or not there's a balance in the trust fund.
So that from those points of view, while there is clearly political relevance, maybe moral relevance, maybe lots of other relevance, but from the point of view of the economy and maybe even government finance, the trust fund balance and status is of less import, as Rudy said, than perhaps the naive measure of being the percent of GDP dedicated to these programs.
MR. GLECKMAN: I'd like to pursue that. I think it's a very interesting question. Does this report answer the wrong question? Does it give a good answer to the wrong question? And maybe Bob Bixby wants to take that on, as well.
MR. BIXBY: I think the trustees report accurately measures the trust fund and the trust fund balance. But for policymakers you really need to look beyond that, and look at the impact of the program on the budget and the economy. The real question is not so much whether there's spending authority in the trust fund for 30 or 40 years, but what is the implication of the cost burden on the economy 30 or 40 years down the road when those benefits become due, and what else will policymakers have to come up with in the federal budget to pay for it at that time. Social Security, of course, is one piece of the pie. Medicare is another big piece of the pie, which is going to add more to it. Medicaid? What are we going to do about long-term care for all of us boomers, and even for the World War II generation that's living longer and longer? As the president has said many times, those are good problems to have, but they are problems. There's nothing wrong with an older population, but it does have fiscal consequences that we can be dealing with now.
So I think the trustees report in the traditional sense that it's been looked at is becoming less and less relevant. And I think one of the key—and that's not to say that the trustees report is wrong, or bad, or inaccurate, it's a question of how do policymakers take the information that's in the trustees report and turn it into things that are more relevant to the types of reforms that, as Wendell said, there's a whole lot of range of reforms being proposed right now that are on the table, so the idea should be to take the trustees report and make it more relevant for those things, looking at the cost burden on the economy, the percentage of taxable payroll, you know, what are the real world consequences, aside from just what happens to this internal government bookkeeping device called the trust fund, that's only one factor.
How we do that, I'm not really sure. I think the technical panel made some extremely useful suggestions in that regard. The trustees, I don't know that the trustees would want to make those additional assumptions about impact on savings, impact on the budget. Maybe the trustees need to be supplemented with—what would you make about a trustees report with political assumptions. I mean, I know you really can't do that, but maybe there's got to be some sort of an addendum to the trustees report that says, okay, here is what we have said on these technical matters, but also consider these other issues, so you get a broader picture of what's going on here.
MR. GLECKMAN: Seventy-five-year political assumptions, that could be interesting.
MR. BIXBY: Well, why not?
MR. GLECKMAN: Let's ask a couple of trustees. Marilyn?
MS. MOON: I think that the value of the trustees report is that it doesn't satisfy all the political desires of folks, because I think you want to try to keep that out of it. This is the basis on which all these kinds of arguments go on, and all of us would be out of work if this was the be-all and end-all, and there was nothing else to be said after this came out. I think that there are some things in here that should be emphasized more or less over time. I think that a greater focus, for example, on the 25-year and 50-year numbers would be useful. A greater focus on looking at some of the specific measures that are in here would be useful. As a researcher, what I've always found important about this is that this is the basis upon which a lot of people have some agreement, and use as a starting point from which to do further analysis. And these are numbers that most of us cannot parse out on our own. And so they are extremely useful.
I find the trust fund report—for example, both Medicare and Social Security—undesirable because it doesn't focus on not so much political or the economy, but what happens to the beneficiaries in terms of these things over time. Is it adequate over time, is it not adequate over time, but I expect that that's my role as a researcher then to take that and move that forward. So I think that I would make a pitch that when this is looked at to make it more user friendly, and I agree that it should be, and I think there's an interest in doing that by the trustees and the future trustees, I hope that the emphasis will be on providing information that is useful to people, and for which there is some consensus, and for which it depends upon these tough modeling exercises that most people don't want to undertake. When you start getting into it yourself you really don't want to do it in many cases. And leave then and have the expectation that this is interpreted after it's released.
MR. GLECKMAN: Stan?
MR. ROSS: I come out very squarely for the fact that we've developed a useful tool. It can be improved surely, but people have to make their policy arguments in addition to it or off of the data contained there. The one time in recent history when we've sort of seen policy start to get interjected into what was otherwise a pretty green-eyeshade exercise was the early '80s with the supply-siders in Treasury. We wound up with two best estimates for a while. And when I became public trustee in 1990, the first thing we did is cooperate with the Treasury to get back to one best estimate, feeling two best estimates added to public confusion. So I think it's good for what it is, and we ought to just take it from there.
I did ask the technical panel to do something which I am disappointed didn't get noticed, which is to contribute to the policy debate by providing some yardsticks for measuring privatization plans for Social Security on a level playing field. The actuaries increasingly have been called upon with a lot of these privatization proposals, particularly starting with Simpson-Kerry, to provide estimates not just along the lines of the traditional trust fund reports, but of the effects of the private plans. And knowing that they've been pushed into this brave new world, where they don't like to make policy assumptions, I thought a group of experts ought to break some new ground, and try to provide similar conventions for the privatization side of proposals, as the traditional plan.
I must say, with Gene's help, we tried twice to get the press interested in this part of the technical panel's work—once at a press conference, when it was released, and once at a forum on Capitol Hill—and got precious little attention. And I would sort of almost ask the press people here, this is the area that really is relevant to a lot of the policy discussions, why haven't you sort of noticed that this group made a really first, strong effort? We included financial economists for the first time, people who study markets, on this panel to sort of try to bring more strength to this area of their report. And the press, I thought, ought to pick up on the fact that there are no conventions for measuring the claims that privatization proposals have, and why not sort of have a discussion and see if we can get better benchmarks for those proposals. And nobody seems to want to care about that.
MR. GLECKMAN: It's boring.
MR. CRIPPEN: But it's important. And I agree, that's where the debate is.
MR. GLECKMAN: Let me quickly ask Harry Holzer, who also was a trustee representing the Secretary of Labor, your sense of this? Is it satisfactory, are there things this report should be doing that it's not?
MR. HOLZER: I basically agree that it's a useful exercise designed to answer one very specific question that, I think, does merit answering, and beyond that there's a wide range of issues that we shouldn't expect to be addressed from this. And if our expectations are appropriate to the job, no need to keep debating that point. I think, having participated in this process one time, I was surprised—and again, this just repeats a lot of the comments made about the 75-year projections and how meaningless they become, but these one-tenth differences.
I was very active in an argument about the hours-worked assumption, which went on over several meetings, and the argument was whether the assumption should be 0.0 or minus 0.1, because over a 10-year period that goes out to create a 10 percent difference in hours worked. And the reason we kept arguing about that was that I think there was at least one agency in the process that was very frustrated that we weren't making much more optimistic assumptions for some of the reasons you said, and maybe for some political reasons, too, although I don't want to speculate too much on that. And they were looking hard for a place to—the idea was, we know what the outcome is, we want to be able to make this statement, and we're looking for a place where we can make it. And many of the rest of us were trying to fend off using this variable or that variable to make it. But it does become this juggling act of not just uncertainty over one variable, but when you have a dozen variables all of which have a lot of uncertainty, and you aggregate across those variables.
Some of us would say, well, we really don't know, if we're a little more optimistic on this one, we should be a little less optimistic on the other, and maybe they'll net out. Whereas, those agencies are players that do come in with agendas [and] have more room to push them. And in that process I was struck by how much room there was to push agendas. But I think the process worked and they were successfully fought off. But, again, on larger issues the report answers the one question, and we shouldn't look for more in that report.
MR. GLECKMAN: Gene, you want to jump in?
MR. STEUERLE: What the press people are going to respond to, Stan, why don't you do that?
MR. GLECKMAN: Did somebody want to respond to this?
MR. RICH: What I was going to say, the reason that you didn't get more coverage on the technical panel is, I think, that the press is now going for generally more simple explanations of things. This is really technical. And the editors say it's too complicated; nobody is going to know what the hell you're talking about.
The other part of it is that, as far as the trustees report, I think most reporters who cover this take that as the gold standard. It's very obvious that some of the organizations that are interested in, one way or the other, that are interested in what you can do with privatization, are making up the numbers, and nobody really trusts those. Who writes about this regularly, I think if you look in the Times, you will see that, and in other publications, like my colleague here, Julie, you will not find them just off the wall taking some projection made by some junior member of Congress who has studied it for about three weeks and decided he wanted to be an anti-privatizer or a pro-privatizer. I would much more trust the Congressional Budget Office to do this kind of stuff, or an advisory panel, where you could assemble all the information in the ways that people have been talking about here to go much further in what you should do than simply looking at the trustees report, but that's a terrific base.
MR. GLECKMAN: Julie Kosterlitz.
MS. KOSTERLITZ: I would just like to add onto that, that at National Journal, where our motto is Dare to Be Dull, we did, in fact, mention the implications of if the trustees were to include some kind of benchmark for measuring these various proposals. And, in fact, that would be very useful to journalists, I think, even if they didn't write about the proposal. If you went ahead and did it, they would be more than happy to have it, because it would clear a lot of things up.
I just wanted to make one other comment about other yardsticks that the technical panel suggested be included, and that other people mentioned, such as share of GDP that goes to benefits, or impact on the budget. And in that regard, I think that Marilyn's point that if you don't also include what the implications are for beneficiaries, you find yourself back in 1992 or '3 talking about healthcare reform, and having this mondo gap between the public and the policymakers on what reform means. And you have the policymakers and the public agreeing that healthcare costs too much, and of course the government means that you've got to cut benefits. You've got to find some way to constrain what you spend on people. And the public thinks, oh, great, they're going to reduce my doctor bills.
So I'm not sure if having share of GDP makes any of the political decisions any easier because it's still fundamentally a values choice about what share of your national wealth you want to spend on a population that's aging so much. And those things in the abstract are going to be I don't know how much more viable from looking at a GDP measure than from looking at the trust fund measures.
MR. GLECKMAN: I was just thinking how nice it was that it was about time to wrap up and a journalist was going to get the last word.
But Bob, I know, wants to jump in. It's his hall, so let's—
MR. REISCHAUER: This is just a comment on the criticism of CBO, which struck me as a little strange, that the trustees report doesn't go far enough because it didn't put it in the larger context of the budget.
Let's keep in mind that CBO puts out a long-range budget forecast book that takes the whole budget, Social Security, and everything else, and looks forward under a couple of different assumptions, as does OMB in the budget appendix. And nobody pays any attention to either of these publications, none of these things. Maybe if that's what we want, it really shouldn't be loaded on Steve and the trustees. There should be, every five years, a commission or something that would do this with some high-level visibility and say, putting together the trustees reports, and the other information we have, this is what our future looks like.
MR. GLECKMAN: The Reischauer Commission. Well, it's 10:30 and we need to wrap up. I would like to thank everybody for coming. I thought it was really interesting.
Latest Research
Browse by Topic:
Printer-friendly summary
