The Federal Reserve Board eagle logo links to home page

Remarks by Vice Chair Alice M. Rivlin
Social Security
Before the Business Roundtable, Social Security Symposium, Washington, D.C.
April 6, 1999

This conference focuses on the question: "What should we do about funding Social Security?" That sounds like a well-defined issue for a conference. There are serious Social Security reform proposals on the table and choices to be made. Experts can be brought in to analyze the options and evaluate the pros and cons. Business leaders can pick the best ideas and try to influence politicians to support them.

I believe, however, that focussing too narrowly on the Social Security funding question -- in isolation from the more fundamental economic challenge of an aging population -- risks muddling the problem and perhaps picking a wrong answer. The fundamental challenge is that in about a decade the ratio of older people to those of working age will begin to rise rapidly as the baby boom generation retires from the work force. Moreover, because people are living longer, that ratio is almost certain to go on rising indefinitely, albeit at a much slower rate after the whole baby boom generation has retired.

In any given future year, say 2050, a larger proportion of older people will be competing with the workforce and the rest of the population for shares of the GDP in that year. Whatever is produced in 2050 will have to suffice for all the claimants. Societies cannot consume more than they produce for long, nor can consumer goods feasibly be stockpiled.

This challenge raises two policy questions of immediate importance:

  • What can we do now to increase future GDP so that there are more goods and services to be distributed among the claimants in future years?
  • What, if anything, should be done now to lock in the claims of particular groups, such as retirees, on that future GDP?

The first question -- how to move to a higher economic growth path -- is obviously the most important, as well as the most urgent. If we can find ways to make the future workforce more productive, both they and future retirees will benefit. Moreover, since growth is cumulative and even small increases in growth rates have large effects over time, the sooner we get started, the greater the impact on any given future year.

What to do about funding Social Security is an aspect of the second question -- whether to try to lock in the claims of older people on future GDP. Its main urgency is that some solutions contribute to higher growth and some do not. It's important to choose a pro-growth solution and choose it soon. Locking in claims may also have some intrinsic merit, since it reduces uncertainty and future tension among claimants, but any deal struck now could come unstuck under future political pressures.

The Important Objective: Higher Sustained Economic Growth
It we want to have a bigger GDP to divide among the elderly and the non-elderly in the future, the obvious thing to do is to save more now and invest it wisely. Higher saving can come in the form of public sector surpluses used to reduce debt held by the public. Public debt reduction makes more funds available for investment and puts downward pressure on the cost of capital. Alternatively, higher saving can come from the private sector deciding to consume less and save more, which also channels more funds into investment and puts downward pressure on capital costs. From a macroeconomic point of view, it doesn't much matter whether the saving is public or private. The choice is pragmatic. We know how to increase public saving by running surpluses in the public budget; we are not at all sure how to raise private saving in the United States. We have tried a lot of different incentives in the last decade or two without any visible effect. Private saving remains low. Hence, paying down public debt has the obvious merit of being something we know how to do.

Increasing national saving isn't a guarantee of higher growth in the future, nor is it the only prerequisite for a more productive economy. The increased saving has to be used well. It should be combined with a steady stream of innovations ready to be turned into products and processes, successful investment in human capital that results in upgrading workforce skills, and a widespread managerial focus on continuous productivity improvement.

That all sounds like what is happening in the U.S. economy right now, so one way to contribute to a higher GDP over coming decades may be to figure out what we are all doing so right in the current U.S. economy and keep doing it.

Tight labor markets have long been seen as harbingers of inflation (as higher wages led to higher prices) and lower productivity (as less skilled and experienced workers were drawn into the labor market). But recent experience suggests that, at least in the context of intense global and domestic competitiveness, as well as a continuing revolution in computers and telecommunications, tight labor markets can provide incentives for managerial innovation, skills acquisition and higher productivity, thereby leading to higher growth with little inflation.

Locking in Claims on Future Output
The challenge of an aging society is sometimes equated with locking in claims on future output, especially those of Social Security beneficiaries. Current estimates are that the Social Security system will continue to take in more than it is paying out (including interest) until about 2022, thus building up balances in the fund. Then, as the baby boom generation swells the retirement rolls, it will begin paying out more than it is taking in, drawing down its resources (selling the Treasury securities in the fund). By current calculations, the fund will be used up in 2034; the Social Security system will be unable to pay all the claims out of current inflow. If nothing is done to fix the imbalance before then, either benefits would have to be cut drastically or payroll taxes increased precipitously (or some combination of the two) to keep the fund solvent and the checks flowing.

Proposals to "fully fund" Social Security are designed to avoid the pain of abrupt changes in either taxes or benefits by ensuring the fund is large enough to go on paying benefits for the foreseeable future (or at least for 75 years).

There would be advantages to knowing that the system was fully funded. The scare talk about "Social Security going bankrupt" would diminish. Younger people might become more confident that they would get their Social Security benefits when they retire (many, if not most, currently doubt that Social Security will still be there when they retire), and people of all ages would likely do more accurate planning for their retirement.

Full funding, however, does not necessarily contribute to making future GDP larger, although it could do so if it was done in a way that added to national saving. Full funding achieved by increasing tax inflows into the fund and/or reducing benefit outflows would build up larger balances in the fund and would add to national saving, provided these larger surpluses in the Social Security trust fund were not offset by larger deficits in the rest of the federal budget. Full funding achieved by transferring general revenues to the Social Security fund would add to national saving only if other expenditures were cut commensurately so that no offsetting deficit was created in the rest of the budget. Raising the rate of return on Social Security assets by investing in them in private sector securities would help achieve full funding with a smaller tax increase or benefit cut, but would do nothing by itself to increase national saving.

The merits of any particular full funding proposal, therefore, depend on:

  • how much it contributes, on balance, to national saving (the more it does, the more it helps to ease the burden of paying all future claimants, elderly and non-elderly alike);
  • how the evaluator weighs the future consumption of the elderly against that of other groups (the young, the middle aged, the sick, the poor) or against other activities (national defense, scientific research).

Of course, priorities chosen now won't necessarily be those of future populations and their elected representatives. Current law Social Security benefits, even if fully funded by enactment of future tax increases, might begin to seem too costly as the future difficulty of paying other claims became apparent. The opposite scenario, seems more likely, however. Near-term decisions to fully fund the system by reducing future benefits could easily come unstuck as the political clout of seniors inevitably increases with their rising proportion of the voting-age population.

A third criterion that appeals to me is preserving the structure of the Social Security system as much as possible. Social Security has served us well. It has been enormously successful in enhancing the well-being, independence and dignity of older people. It is extremely popular and well entrenched in our culture, with other retirement and insurance programs built around and on top of it. Its universality, its contributory nature and its redistributive effects seem to me huge pluses worth preserving. We should be reluctant to scrap or drastically alter a program that has worked so well for more than half a century, when options for preserving its merits, while correcting its defects, are available.

The Plans on the Table
The Administration's plan gets high marks on the most important criterion: adding to national saving. It does that by creating a politically attractive motive for not reducing currently projected surpluses; it preserves them by allocating them largely to saving Social Security and Medicare. The Administration would extend the life of the Social Security trust fund to 2049 by legislating large transfers of general revenues into the fund. It allocates nearly two-thirds of the projected surplus to augment the Social Security trust fund (and another 15 percent for Medicare).

Whether this commitment adds to national saving depends, of course, on what you think would happen to the projected surpluses in the absence of a powerful argument to "use" them to replenish the social insurance funds. If you think (and I find this scenario compelling) that the congress would likely use the surpluses to fund tax cuts and other federal spending, then it follows that the Administration's plan gives a huge boost to national saving. It would reduce the debt held by the public -- indeed current projections eliminate it entirely by 2018 -- thus making huge sums available for private (and state and local) investment and lowering the cost of capital.

The Administration's plan also preserves the existing Social Security system on the benefit side by locking in the claims of future beneficiaries to the benefits specified under current law or at least by making it much less urgent to reconsider those benefits in the name of fully funding the system. To be sure, the Administration's plan does not achieve full funding, even for 75 years, so benefit cuts or tax increases would still be necessary at some point, but it puts off the day of reckoning.

On the income side, the plan does violence to the traditional contributory nature of Social Security by using a huge injection of general revenues to preserve claims to future benefits. Social insurance purists will worry about this. I'm inclined to believe that it would be worth sacrificing a bit of purity to get the addition to national saving inherent in using such a politically attractive excuse for not turning the projected surpluses into tax cuts or current spending.

Indeed, whether or not its specific plan is adopted, I believe the Administration has achieved a significant breakthrough in increasing the national understanding of the importance of raising national saving to meet the challenge of an aging society. The mechanics of the Administration's plan are complex and hard to grasp, which made unfair charges of budget gimmickery hard for them to refute, but in the process of defending it, Administration officials have sold the plan's main feature well. Using surpluses to pay off debt held by the public adds to national saving and enhances future growth; the Administration's plan has that big strength.

In terms of sorting out the claims to GDP in, say, 2050, the Administration makes a clear choice: it favors future Social Security recipients over other claimants. If the economy continues to do well (enhanced by the additional saving generated by public debt reduction) this tilting of priorities toward the elderly may not be questioned, especially by an electorate increasingly heavily weighted with seniors. If the economic good news peters out, however, and the surpluses disappear, the priorities issue will become more stark. If the commitments to transfer general revenue resources to the Social Security fund have to be met by raising general taxes or drastically cutting spending for other purposes (including investment in infrastructure, worker skills and scientific research that arguably enhance future growth) then many might argue that the priorities were wrong and should be altered. Generations X and Y will likely be gray by then, but their younger counterparts may be equally vocal.

The Administration also proposes universal saving accounts, apparently designed to encourage private saving, but has yet to spell out details of the proposal. Others have offered plans that would substitute compulsory contributions to privately managed savings accounts for a significant portion of Social Security benefits, leaving Social Security as a back-up plan for low earners and unsuccessful investors.

Martin Feldstein's plan is ingenious and, like the Administration's proposal, adds to national saving by using the projected surpluses to finance retirement savings. In particular, his plan would use the surpluses to fund individual accounts. These accounts would receive government contributions equal to 2.3 percent of the previous year's wages, up to the Social Security maximum. At retirement, the proceeds of each account would be used to purchase an annuity; and the government would guarantee that the combination of traditional benefits and the annuity would be at least as large as the Social Security benefit promised under current law. The Feldstein plan would effectively use the proceeds from the individual accounts to make up the expected shortfall in future Social Security benefits. The individual accounts would presumably be invested, at least partially, in private securities, so like the Administration's proposal, this plan uses the higher rate of return on private securities to help fund the Social Security system.

Like the Administration's plan, the Feldstein plan adds to national saving only if one believes that without the proposed transfers to Social Security, the surpluses are likely to disappear into tax cuts and other federal spending. It differs from the Administration plan in that it substitutes privately invested individual accounts for much of the current Social Security structure. If successful, it would presumably be a transition to a retirement system based primarily on compulsory contributions to individuals accounts invested in private securities, with Social Security as we know it becoming a residual safety net.

Those of us who believe that the original concept of social insurance has served us well and should continue would be fearful that such a transition would lose the benefits of a universal, redistributive (but not means-tested) system.

Where do I come out? In the days before surpluses of such magnitude were anticipated in the Federal budget, solving the Social Security funding shortfall seemed to me a pretty simple problem -- at least conceptually, if not politically. The objective was to preserve the structure of the system, get it into at least 75 year actuarial balance, and add to national saving in the process. Since it seemed inadvisable to raise payroll taxes, I thought it would be necessary to trim future benefits (by raising the retirement age, making the calculation of benefits marginally less generous, making sure that benefits paid were not over-indexed for inflation). The benefit reduction could be minimized by making sure that coverage was extended to all possible recruits and by investing a portion of the proceeds in private markets to increase the rate of return.

Admittedly, the private investment aspect was a little suspect, because it did nothing to raise national saving, but it did make the narrow Social Security funding problem easier to solve. The argument that public funds invested in private securities could lead to undesirable attempts to exercise political influences over private firms seemed to me an argument to take seriously, but I regarded the problem as fixable with restrictions and safeguards, as does the Administration.

As for individual accounts, I thought the proposal of my colleague, Ned Gramlich, was a sound one, since it involved an add-on to current payroll tax contributions and therefore a net addition to saving. I thought the funds among which an individual would choose ought to be centrally managed to minimize the administrative cost; but I did hope that the experience of owning private securities might induce additional private saving over time. It also might be useful to consider providing a matching government contribution to low-income workers, both to enhance the progressivity of individual accounts and to reduce the disincentive to hire low skilled workers.

Now that we have the surpluses in view, however, I believe we have a unique opportunity to increase national saving by reducing the public debt. Combining this opportunity with a solution to the Social Security funding problem makes both political and economic sense. A bipartisan solution that combined the Administration's idea for locking in the surpluses with modest benefit cuts and some type of add-on individual accounts now strikes me as both desirable and fortuitous.

Return to topReturn to top

1999 Speeches