Structural Reform of the Social Security System: The Time Has Come
by Jagadeesh Gokhale
February 15, 1997

Introduction
Basics of the U.S. Social Security Program
Current Status
Long-Range Projections
Structural Deficiencies
Undoing the Structural Damage
Conclusion
Footnotes

After months of wrangling, Congress and the administration recently reached an agreement on how to balance the federal budget by the year 2002.1 Although the compromise bill contains a $115 billion reduction in projected Medicare spending, it includes no initiatives regarding Social Security.

Why the apparent oversight? Do lawmakers believe that the need for Social Security reform is much less urgent than for Medicare, or is it simply expedient to defer changing such a politically sensitive program until change becomes unavoidable?

This Economic Commentary argues that reforming Social Security is one of the most pressing tasks facing the United States. Although the Social Security trust fund’s accounting conventions suggest that the program will remain solvent until 2029, this projection is misleading at best. Changes in the nation’s tax policy will be required much earlier to maintain benefits at mandated levels. Furthermore, the current system contains structural features that result in economic inefficiencies, the costs of which will compound over time.2 Any successful reform effort will have to include a number of specific elements aimed at eliminating these structural problems.

Return to Index

Basics of the U.S. Social Security Program

Social Security began in 1935 as a small pension program, but has expanded over time along several dimensions, including 1) the share of the workforce that is covered, 2) the size of payroll tax rates, and 3) the generosity and variety of benefits provided. Today, recipients are insured against old age, dependency, disability, and death. This wide variety of benefits is one of the features that has introduced inequities into the system and caused an inefficient allocation of resources.

Another important feature of Social Security is its pay-as-you-go structure, whereby current workers’ contributions are immediately and directly handed over as retirement and other benefits. This feature alters the allocation of resources in a way that impedes future productivity.

Return to Index

Current Status

In fiscal year 1996, Social Security’s reported income was $416 billion and its outgo was $350 billion, of which $343 billion was devoted to benefit payments.3 This $66 billion surplus pushed the system’s year-end assets to $550 billion. Assets of the Social Security trust fund consist exclusively of federal IOUs, because the government borrows the surplus each year and spends it on current operations.

This implies that all current contributions are consumed—either by beneficiaries or by the government—rather than invested in productive capital assets. If and when the IOUs have to be drawn down to finance future benefit payments, Congress will be forced to increase non-payroll taxes in order to redeem them.

Return to Index

Long-Range Projections

The Demographic Side
Before examining Social Security’s long-range financial projections, it’s instructive to look at the demographic trends that are driving them. Figure 1 shows population–age distributions for 1995 and 2025. The hump in the middle of the 1995 curve represents the baby boom generation, whose members are in their prime working years and are contributing substantial amounts of money to the system. As the boomers age, the population–age distribution will change dramatically. Over the next three decades, the increase in the number of young persons will be far outstripped by the skyrocketing number of retirees.

FIGURE 1:     POPULATION-AGE DISTRIBUTIONS

SOURCE: Social Security Administration.


These projected trends imply that the ratio of young contributors to old beneficiaries will plummet. Today, there are 3.3 workers per beneficiary. By 2025, that figure will fall to only 2.2—a potentially devastating development for any intergenerational pay-as-you-go program.4 Preserving current benefit levels will require sharp increases in payroll tax rates or major cuts in future benefits.


FIGURE 2:     OFFICIAL LONG-RANGE PROJECTIONS

a. Includes interest on government bonds held in the trust fund.
SOURCE: Social Security Administration.


The Financial Side
Figure 2 shows that because of the growing number of elderly Americans, Social Security outlays are expected to rise rapidly during the first two decades of the next century. Reported income will fail to keep pace, however, and annual surpluses will turn into deficits starting in 2020. At this point, the trust fund will be drawn down to finance the shortfall. Basing their judgments on the year the trust fund’s paper assets will be exhausted, many analysts and policymakers place the program’s date of insolvency at 2029. However, the major policy change—increasing taxes or cutting benefits—will be forced on the government in 2019. This should be considered the true date of insolvency.

The above analysis is not complete, however. Just as redeeming the trust fund’s IOUs will require non-payroll taxation, so will paying interest on the accumulated trust fund. The latter will also transfer the cost of financing benefits to the non-Social Security side of the government’s ledger. Hence, rather than comparing outgo with total income when estimating the date of insolvency, one should compare it with income net of interest income—payroll contributions plus taxes on Social Security benefits. The latter comparison places the date of insolvency at 2012.

Even this may turn out to be too optimistic, however. The above projections are based on the Social Security Administration’s (SSA) intermediate assumptions about future productivity and demographic trends. However, the SSA makes two other projections of income and outlays based on optimistic and pessimistic assumptions about these trends.5

Historically, Social Security’s realized income and outgo have fallen between the intermediate and pessimistic projections. Based on the latter assumptions, insolvency could come as early as 2000 (see figure 3).6 A conservative estimate would be 2006—midway between the intermediate and pessimistic predictions. Given that the oldest baby boomers will begin retiring after 2006, one must conclude that their retirement security is in jeopardy.


FIGURE 3:     PESSIMISTIC LONG-RANGE PROJECTIONS

a. Includes interest on government bonds held in the trust fund.
SOURCE: Social Security Administration.


Return to Index

Structural Deficiencies

Because of the wide variety of benefits offered, the link between households’ Social Security contributions and the benefits they receive is weak. Under current rules, individuals in some types of households receive benefits whether or not they worked and contributed to the system in the past.7

Married single-earner households, for instance, receive more benefits than married dual-earner couples (for a given level of contributions), since the nonearning spouses in the former qualify for dependent and survivor benefits based on the breadwinner’s earnings record. Such a benefit structure involves a work disincentive for secondary earners in every household.

Another example is single heads of households, who receive relatively few benefits despite their substantial contributions. For them, Social Security payments seem more like a tax and add to their marginal income-tax rates—again resulting in disincentives to work. Such a redistribution of lifetime resources across households breaks the connection between contributions and benefits and, ultimately, between work and its rewards. The lower supply of labor services that this induces implies an inefficient use of the nation’s human resources.

Pay-as-you-go Social Security also alters the allocation of resources between consumption and saving. As mentioned earlier, all current contributions are consumed rather than invested in real productive capital. However, a growing pay-as-you-go system brings about an additional reallocation in favor of consumption because it redistributes resources from younger and future generations toward the elderly. That is, growth in the size of taxes and benefits implies that people who paid into the system when tax rates were low will receive windfalls after they retire. The cost of paying these individuals a disproportionate share of benefits, of course, falls on subsequent working generations.


FIGURE 4:     RELATIVE RESOURCES BY AGE

SOURCE: Author's calculations based on the Economic Report of the President, 1995.


FIGURE 5:     AVERAGE CONSUMPTION PROPENSITIES

SOURCE: Author's calculations.


Figure 4 shows that an increasingly generous pay-as-you-go Social Security system, coupled with expanding Medicare benefits, has resulted in a transfer of resources from younger and future generations toward older Americans.8 On average, the elderly consume a larger fraction of their resources than do younger individuals, and the gap has been widening (see figure 5).9 Thus, a transfer of resources from the young to the old implies greater aggregate consumption and smaller aggregate saving. As figure 6 indicates, national saving has declined considerably since the early 1970s, when entitlement programs such as Social Security and Medicare began to expand rapidly.


FIGURE 6:     U.S. NET NATIONAL SAVING RATE

SOURCE: Author's calculations.


Return to Index

Undoing the Structural Damage

The work disincentives caused by Social Security’s current benefit structure reduce output directly, while lower national saving constrains capital formation and thus reduces labor productivity. It is important, then, to consider the structural shortcomings of the Social Security program when weighing the merits of different reform plans.

From this perspective, both of the obvious options to restore trust fund solvency—increasing payroll taxes and/or decreasing future benefits—seem inadequate. First, both would reduce the already low returns on contributions that current and future working generations will receive.10 Second, neither of these measures would alter the structural deficiencies that produce the undesirable effects on labor supply and saving.

Future consumption—retirees’ or anyone else’s—must come out of future output. Therefore, a greater number of retirees relative to workers implies that more output must be available for sustaining older generations’ living standards. This can occur without hurting younger generations only if output expands adequately—an unlikely scenario if recent productivity trends continue over the next three decades. Hence, to secure the retirement of the baby boomers without placing an intolerable burden on future working generations, any effective reform package must 1) result in more real investment, 2) restore individuals’ incentives to work, and 3) reduce or eliminate the intergenerational redistribution that leads to low national saving. These critical elements cannot be ignored—or delayed.

Return to Index

Conclusion

Securing the retirement of the baby boomers and future generations of Americans cannot be accomplished through mere tinkering with Social Security taxes and benefits. The many types of benefits provided under our current system have resulted in a poor linkage between contributions and benefits and have undermined individuals’ work incentives. Moreover, the intergenerational redistribution that has accompanied the expansion of pay-as-you-go Social Security has contributed to low national saving rates. Any effective reform effort must address all of these structural issues, and must do so quickly.

Return to Index

Footnotes

1. This article went to press in mid-June 1997.
Back to commentary

2. This Economic Commentary does not discuss any specific reform proposals. For an example of a reform plan that concentrates on rectifying Social Security’s structural problems, see David Altig and Jagadeesh Gokhale, "Social Security Privatization: A Simple Proposal," The Cato Project on Social Security Privatization, 1997 (forthcoming).
Back to commentary

3. The numbers reported here are based on the 1996 Annual Report of the Board of Trustees, Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds, table II.F11, pp. 104–05.
Back to commentary

4. Ibid., table II.F19, p. 124.
Back to commentary

5. These are also known as high- and low-cost assumptions. Essentially, the pessimistic (high-cost) assumptions incorporate a lower rate of future productivity growth, pushing projected payroll tax revenues below the intermediate case. It also assumes low future birth and mortality rates, both of which reduce the ratio of future taxpayers to future retirees. The optimistic (low-cost) assumptions assume higher productivity, birth, and mortality rates.
Back to commentary

6. See 1996 Annual Report of the Board of Trustees (footnote 3), table III. B2, p. 180.
Back to commentary

7. For a description of Social Security’s eligibility and benefit rules, see Social Security Administration, The Handbook of Social Security, Washington, D.C.: Government Printing Office, 1996.
Back to commentary

8. Resources are the sum of current net worth and the present value of earnings and pensions, minus the present value of taxes net of transfers.
Back to commentary

9. Ibid.
Back to commentary

10. See Dean R. Leimer, "Cohort-Specific Measures of Lifetime Net Social Security Transfers," Social Security Administration, Office of Research and Statistics, Working Paper No. 59, February 1994.
Back to commentary

Return to Index


Jagadeesh Gokhale is an economic advisor at the Federal Reserve Bank of Cleveland.

  The views stated herein are those of the author and not necessarily those of the Federal Reserve Bank of Cleveland or of the Board of Governors of the Federal Reserve System.


[Abstracts] [FRB ClevelandHomepage] [Corporate Communications Homepage] [Research Homepage]

http://www.clev.frb.org/research/com97/