FRBSF Economic Letter
1997-14 | May 9, 1997
Proposals for Reforming Social Security
In early January, the Advisory Council on Social Security published a review of the system and reported that it is in jeopardy. Although the system is currently accumulating a surplus, actuaries at the Social Security Administration forecast that the present value of future obligations far exceeds the present value of future revenues. The magnitude of the projected shortfall is enormous, amounting to roughly half of the federal government’s current outstanding debt. These projections suggest that the current system is unsustainable, and the Advisory Council proposed a number of ways to reform it. This Economic Letter describes the problems underlying the current system, reviews the reforms proposed by the Advisory Council, and explains how those reforms might affect the U.S. economy.
As it is now constituted, Social Security is primarily a pay-as-you-go system, which means that current retirement benefits are paid mainly from current tax receipts on younger workers. In contrast, a fully funded program would tax workers when they are young, invest the proceeds on their behalf, and eventually pay their retirement benefits out of the accumulated interest and principal. Although the current system does a little of this, it is mainly a transfer program rather than an investment fund.
A key variable in determining the solvency of a pay-as-you-go system is the ratio of taxpayers to recipients. Thirty years ago, there were 5.5 workers for each recipient, and Social Security would be in much better shape if this ratio had remained at that level. However, this ratio has been falling for some time and will continue to fall. For example, there are now 4.5 workers per recipient, and demographers project that this will fall to 2.5 over the next 30 years. One source of this decline is that a relatively large cohort, the Baby Boom generation, was followed by a relatively small cohort, the Baby Bust. Another is that while the retirement age has remained constant, people are living longer and collecting retirement benefits over a longer span of time. With decreasing fertility and mortality, the population is becoming older, and the number of Social Security recipients is rising relative to the number of younger workers who support them. This demographic fact poses a dilemma for the system. In order to maintain benefits at current levels, younger workers would have to contribute a higher fraction of their income to the system. For example, we could close the gap by raising the payroll tax now by 2.2 percentage points, from 12.4% to 14.6% (combining the contributions of employers and employees), and leaving it there permanently. Postponing the tax increase would only increase its eventual magnitude.
Alternatively, if taxes were maintained at current rates, retirement benefits would eventually have to be cut. The system is currently running a surplus and investing the proceeds in a trust fund, and it will continue to do so until 2011. Between 2012 and 2018, the current system will provide enough revenue from tax collections and interest on the trust fund to sustain the current benefit schedule. After 2019, benefits could be sustained by withdrawing principal from the trust fund, but this would exhaust the principal by 2029. Thereafter, if the current tax rate were held constant, the system would have to reduce benefits. Without a tax increase, the system would have to cut benefits by around 25% by 2040 and by 30% by 2070. Thus, without reform, the grandchildren of the Baby Boom would face substantial benefit cuts when they retire.
The Advisory Council proposed three reform packages: the Maintenance of Benefits (MB) plan, the Individual Accounts (IA) plan, and the Personal Security Account (PSA) plan. The packages all incorporate tax increases, benefit cuts, and changes in investment strategy, but they differ in their objectives and in how they mix these elements.
The MB plan is the most conservative, in the sense that its primary objective is to eliminate projected deficits without altering the essence of the program. According to this plan, tax increases would close roughly one-third of the gap. Taxes would be increased in three ways: the payroll tax rate would be increased by 1.6 percentage points in the out years of the plan, the tax base would be broadened by bringing certain state and local government employees into the system, and benefits in excess of contributions would become taxable, in accordance with current laws governing other defined-benefit plans. Benefit cuts would close approximately 40% of the gap. These would include alterations in the basic benefit formula resulting in an average cut of 3%, reductions in cost-of-living adjustments, which are now believed to be overstated due to biases in the consumer price index, and reallocations of revenues to the retirement fund from other programs, which would presumably involve cuts in those programs.
The remainder of the gap would be closed by altering the trust fund’s investment strategy. Currently, the trust fund invests only in Treasury securities, which are safe but earn a lower average return than stocks, and the MB plan would allocate up to 40% of the trust fund to equities. This element is perhaps the most controversial. While equities offer higher average returns, they are also riskier. The government would not be in a position to guarantee the average real return on equities, and recipients would have to bear the risk. Households already have the option of holding equities in private portfolios and yet many choose not to. Therefore, if the trust fund continued to be centrally managed, as under the MB plan, and everyone were forced to hold identical portfolio shares, many households would be forced to bear more risk than they would like.
The IA plan has two objectives. In addition to balancing the existing system, this package also would introduce mandatory, supplementary savings accounts for all participants. To balance the existing system, the IA plan relies on many of the same elements as the MB plan. For example, it also alters the basic benefit formula, reduces cost-of-living adjustments, makes excess benefits taxable, and broadens the tax base. These common features close roughly three-quarters of the projected shortfall. In addition, the IA plan calls for greater benefit cuts, chiefly by increasing the retirement age and reducing payments to middle- and upper-income recipients by 20%. These provisions close roughly 30% of the gap and, when combined with the prior elements, would result in a modest surplus.
In addition, the IA plan would increase the payroll tax by 1.6 percentage points and allow households to invest the proceeds in a variety of government managed index funds. By allowing a limited degree of portfolio choice, this plan would not force households to hold riskier portfolios than they would like. Balanced against this advantage is the fact that these supplementary accounts would be more costly to manage, perhaps ten times as costly as the centrally managed MB plan. However, the estimated costs compare favorably with private sector mutual funds.
The PSA plan incorporates many of the features of the other plans, but it goes much further in changing the nature of Social Security. Essentially, this plan proposes a two-tier system for funding retirement. It would allocate 7.4 percentage points of the current 12.4% payroll tax to a flat-rate benefit that, by itself, would provide recipients with an income roughly 30% below the poverty line. The remaining 5 percentage points of the payroll tax would be allocated to personal retirement accounts which would be managed by individuals and placed with private companies, and income from these accounts would supplement the basic benefit. The diversion of 5 percentage points of current taxes into personal accounts would leave insufficient funds to cover current benefits, and the PSA plan calls for a $2 trillion dollar loan from the Treasury to smooth the transition. It pays for this loan by levying an additional 1.5 percentage point tax on payrolls. Thus, the PSA package closes two-thirds of the projected gap by raising taxes, makes very deep cuts in basic benefits, and replaces much of that with income derived from personally managed portfolios. This represents a step away from the traditional pay-as-you-go system, toward a funded system, and from a defined-benefit plan toward a defined-contribution plan.
Critics of the PSA plan object that it transfers too much investment risk from the government to households, but proponents see this as a virtue, insofar as it allows households to trade off higher degrees of risk for higher expected returns. Critics also point out that the cost of managing personal accounts is roughly ten times that of managing the mutual funds proposed by the IA plan, and they suggest that a less extensive menu of investment options would be sufficient to preserve freedom of choice while economizing on management fees.
The macroeconomic consequences of reform depend chiefly on how it will affect national savings. Government programs contribute to national savings when they run a surplus and subtract from it when they are in deficit. Without reform, Social Security is expected to run large deficits in the next century and become a drain on national savings. All the reform packages reduce those deficits and hence increase public saving.
The extent to which the reforms increase public savings depends on whether the system will continue on a pay-as-you-go basis or evolve into a funded system. Under a pay-as-you-go system, tax receipts are not invested but are simply transferred to current recipients, and there is no public saving (if the transfer budget is in balance). In a funded system, tax revenues are invested on a worker’s behalf, and that investment represents public saving. Therefore, to the extent that reforms shrink the pay-as-you-go portion of Social Security and replace it with a funded system, they are likely to increase public savings further.
The reforms are also likely to affect private savings. Social Security reduces private savings because taxes reduce the income of young workers, and the promised retirement benefit lessens the need to accumulate private funds for one’s own retirement. Holding taxes constant, a reduction in retirement benefits is likely to increase private savings, because young workers will want to replace some of the lost retirement income out of their own funds. Holding benefits constant, a tax increase is likely to reduce private savings, because it reduces the income of young workers. Hence the net effects of tax and benefit reforms on private savings are ambiguous.
Other elements of reform are likely to reduce private savings. For example, while mandatory supplementary accounts will increase public saving, households who now invest in private mutual funds can avoid the supplementary tax by reducing private savings by an equal amount. The supplementary account is very much like a private IRA, and no doubt some households will merely switch their retirement funds from privately managed mutual funds to publicly managed ones.
On balance, the reforms are likely to raise national savings, although the magnitude of the increase is uncertain. An increase in national savings would benefit the economy in a number of ways. The U.S. economy is now saving less than it invests and borrowing from abroad to make up the difference. An increase in domestic saving would lessen our reliance on foreign capital. Similarly, an economy that saves less than it invests also spends more than it produces and runs a trade deficit vis a vis the rest of the world. An increase in national saving would close the gap between what we spend and what we produce and would reduce the trade deficit. Finally, an economy that saves and invests more accumulates more capital. This increases productivity and raises the standard of living in the long run.
Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System. This publication is edited by Sam Zuckerman and Anita Todd. Permission to reprint must be obtained in writing.
Please send editorial comments and requests for reprint permission to
Attn: Research publications, MS 1140
Federal Reserve Bank of San Francisco
P.O. Box 7702
San Francisco, CA 94120