Number 15 in Series, "The Retirement Project"
The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.
Numerous collisions, advisory groups, and individual policy analysts have concluded that Social Security reform is essential, but many politicians do not want it to happen on their watch. Meanwhile, the public remains confused. Many young people believe that the system will have disintegrated by the time they retire, while many older people fear that reform will force them to eat dog food.
Policy analysts know that these fears are ungrounded, but their debates tend to be cantankerous and ideological. Conceptually, the differences between the left and right seem bridgeable, but no bridges are in sight. Demagoguery during the 2002 election campaign has further poisoned the debate.
The Problem
Agreement about what is wrong is widespread. Social Security is basically a pay-as-you-go program. About 80 percent of its tax revenues are being directly paid to today's beneficiaries.1
The potential generosity of a pay-as-you-go system depends on how many elderly there are and how much tax revenue comes in, which in turn depends on the number of workers, their wages, and their tax rate. Throughout almost all of Social Security's history, the elderly have gotten a good deal. The number of workers supporting them has grown rapidly, especially while the baby boomers were entering the labor force; wages have risen at a healthy rate; and the tax burden on wages has increased substantially throughout most of the system's history.2 Consequently, today's retirees enjoy benefits financed by much higher taxes than they themselves paid over their lifetime.
Unfortunately, Social Security's golden age is about to end. The number of beneficiaries will
surge after 2010, as baby boomers begin retiring and life spans continue to lengthen. The number
of workers supporting them will grow slowly because baby boomers had fewer children than their parents. Wages should continue to grow, but raising tax burdens as rapidly as they rose during
the program's first 50 years is not thinkable politically.
Between 2010 and 2030, the number of Old-Age, Survivors, and Disability Insurance (OASDI)
beneficiaries will rise by 65 percent while the working, taxpaying population will rise less than
8 percent. The number of taxpayers per beneficiary will fall from 3.4 in 2001 to 3.1 in 2010 and to 2.1 in 2030. Meanwhile, Social Security costs are rising in tandem with the much heavier costs of Medicare and Medicaid; without reform, the three programs will absorb an estimated 6 percent
more of the gross domestic product (GDP) by 2030 than they did in 2000 (Congressional Budget Office 2002).
The Solution
Analysts also widely agree on how to approach the problem. Whether leaning left or right, most
say we should rely less on the pay-as-you-go approach and try to fund a portion of benefits—that is, invest part of individual contributions in real assets so that lifetime benefits depend more on the rate of return to real capital and less on such demographic factors as birth rates and expected life spans.
Yet all honest analysts must admit that an increase in funding requires some sacrifice. Increased investment in real assets must be financed by increased saving, and few Americans warm to the idea of cutting consumption. Nor would it be fair or politically possible to reduce
traditional benefits immediately. Thus, workers will have to provide considerable support to the
already retired while also investing for their own retirement.
Social Security's current troubles compound the problem. Today's dedicated revenues will not
nearly cover the benefits promised by current law. To the extent that we continue to rely on the traditional system—and most politically plausible approaches do so to a considerable degree—some promised benefits must be cut or tax burdens increased.
Those few politicians courageous enough to advocate reform, including President George W. Bush, have made clear that people in or nearing retirement should not lose any currently promised benefits. That implies that workers and their families will have to shoulder more of the burden of reform.
But how much sacrifice does reform require? And compared with what? The current system automatically provides ever greater average real benefits to beneficiaries. Thus, someone who retired in 2000 gets a higher real benefit than someone who retired with exactly the same lifetime earnings profile in 1990. This promise of ever increasing average real benefits is what makes the system so expensive. Real benefits increase primarily for two reasons: (1) benefit levels grow with nominal wage levels—that is, generally faster than inflation; and (2) as life expectancy increases, people draw benefits for more years.
Yet, though a slowly growing labor force will have to support a rapidly growing retired population who are living longer, maintaining future retirees' living standards at today's absolute level will not be hard. Today's projections of dedicated tax revenues suggest that they are sufficient to finance about a 9.5 percent real increase in average annual benefits by 2030.3 The young need not fear that the well will have run dry by the time
they retire, and tomorrow's elderly can expect to live a better life than today's if the economy continues to grow.
But the benefits affordable with today's tax rates are not projected to grow as fast as wages
earned just before retirement. The whole purpose of funding part of the system is to reduce the
drop in income relative to wages when someone retires. That is to say, the goal is to do even better than the 9.5 percent increase in real benefits affordable with current tax rates.
Different Approaches
Although the right and the left generally agree that a portion of the system should be funded,
they disagree bitterly about how this should be accomplished. The left would have government
try to increase national saving, while the right would rely on individual accounts.
President Bill Clinton's proposals might be used to typify a leftish approach to the problem,
although he was vague on the details—especially on any that might cause pain in the long run. He advocated transferring revenues from the rest of the budget into the Social Security trust fund. Without saying how to do it, Clinton advocated balancing the budget outside the trust fund. National saving and future production would then be higher than they would be if the overall budget were balanced, making higher traditional Social Security benefits easier to afford.
The key question raised by a Clinton-type proposal is whether future congresses and administrations would be disciplined enough to balance the non-Social Security budget, especially when it was being burdened by transfers made to the trust fund. History is not reassuring in this
regard. In the 1960s, surpluses in government trust funds became significant and politicians
started using them to finance deficits elsewhere in government. The movement got support from the
1967 Commission on Budget Concepts, which argued that the budget used to plan macroeconomic
policy should include trust fund activity. The overall, or unified, budget balance was more explicitly made a target in the Budget and Impoundment Control Act of 1974 and the target
was given teeth in the Gramm-Rudman-Hollings Act of 1985.
Learning from Experience
The mixed record of state public pension fund management affords some clues about how judiciously government can be expected to invest (Romano 1993). Twenty-five state pension funds subsidize businesses to relocate to their areas. Some also invest in state agencies. Numerous funds restrict investments in particular countries or certain types of businesses (e.g., tobacco companies). Given the obvious risks created when funds start making politically motivated investment decisions, advocates of individual accounts consider it far safer to leave equity investing up to individuals, as most European and Latin American countries with reformed pension systems do. Canada and Japan have, however, created independent bodies to manage equity investments for their trust funds.
Of the many European reforms, Sweden's is the most ingenious and has been emulated in former communist countries such as Poland. It combines individual account contributions equal to 2.5 percent of wages with a pay-as-you-go defined contribution system. Workers are given credit for their tax payments and are paid an interest rate on these notional accounts equal to the rate of growth of wages. If the trust fund falls below a certain level, the interest rate is lowered. The cumulated balance is used to buy an annuity whose generosity is determined by actuarial assumptions appropriate to each age cohort. Thus, the system automatically adjusts to changes in life expectancy, birth rates, and economic growth. |
|
When the unified budget balance is chosen as the policy target, it is difficult to argue that trust fund surpluses add to national savings, since other changes in tax law or spending policies will offset random variations in trust fund balances. In the late 1980s, Social Security was taken "off budget" to shift the focus to the non-Social Security budget balance. Nevertheless, budget agreements in 1990, 1993, and 1997 still focused on trying to balance the unified budget. Not until a unified surplus emerged unexpectedly in 1998 did talk of balancing the non-Social Security budget get serious. Balance was achieved in 1999 and 2000, but budget discipline evaporated quickly as taxes were cut, spending soared, and (far less significantly) the costs of responding to the attacks of September 11, 2001, mounted. What type of fiscal rule will emerge from the current chaos remains anybody's guess. But rebalancing the non-Social Security budget now seems implausible unless current budget projections get far rosier very quickly.
If government did manage to increase national saving and investment, the additional capital income created would be divided among interest, dividends, and capital gains. The increased
income would also generate extra tax revenue. Moreover, every new dollar of capital income resulting from added domestic investment would be associated with roughly two dollars of additional before-tax wage income because of the effect of new investment on worker productivity.4
There is no guarantee that the additional income generated by funding would be used to enhance future Social Security benefits. It could be used for defense, the poor, tax cuts, or any other purpose. Initially, Clinton hoped to capture a higher portion of the additional income for Social Security by allowing its trust fund to invest in equities, which historically have paid a higher rate of return than the government bonds that are now held. But this proposal created much controversy. Advocates argued that an independent board could insulate the trust fund's investment policy from political pressures. Others, most influentially Federal Reserve Chairman Alan Greenspan, argued that politics would inevitably affect which equities were purchased. Clinton dropped the proposal in later versions of his plan (see box on how states have managed their public pension funds).
What if government fails to increase national saving and production but still enhances the trust fund with general revenue payments and returns from equities? Future beneficiaries would then get a larger claim on national production, but national production would not have increased.
Therefore, the bulk of the benefit increases would have to come out of the hides of future workers and their families.
Relative Risks
If it is doubtful that government fiscal policy can fund Social Security adequately, what are the odds that individual accounts can? In a typical "carve-out" approach, as advocated by President Bush, payroll taxes would be reduced by 2 percentage points if an individual puts the money into an individual retirement account. The government deficit would increase immediately by the amount of the forgone payroll tax revenues, and national saving would remain constant if the whole contribution to the individual account represented an increase in private saving. In other words, lower public saving would be offset by higher private saving. But if the new individual account substituted for other forms of retirement saving, the increased deficit would not be entirely offset by higher private saving.
That is the fear, but it is probably misplaced. For one thing, reforms of this type begin to
reduce traditional benefits as individual accounts accumulate and begin throwing off significant
retirement income. Government starts saving considerable sums in the long run, and the deterioration of the budget balance reverses itself. If Congress tries to balance the unified budget, the initial loss of revenues may also be offset by reducing tax cuts and spending increases elsewhere in the budget. Finally, most rational people can be expected to use most of their new individual accounts to increase net saving simply because the cuts in traditional benefits imply that they will lose retirement income if they substitute too much of the new individual account for other retirement saving.
How Much of General Revenues?
Lost in the debate over government funding versus individual accounts is a cardinal point: There is no iron link between the approach chosen and the size of future benefits. Both sides typically rely on general revenues to help fund their proposals. The Clinton approach does so explicitly. Most individual account proposals use general revenues until the income from individual accounts builds enough to make cuts in traditional benefits acceptable.
For example, Martin Feldstein of Harvard University proposes using individual accounts to guarantee a level of retirement income from traditional benefits and individual accounts higher
than the benefits promised today, partly financed by commandeering some corporate tax revenues.
The president's recent Commission to Strengthen Social Security (2001) presented three individual
account options of varying generosity that use general revenues as individual accounts build. All
are expected to deliver considerably higher benefits than today's tax levels can finance. Legislative proposals by Representatives Jim Kolbe (R-AZ) and Charles Stenholm (D-TX) and by Senators John Breaux (D-LA) and Judd Gregg (R-NH) are less generous and so require less general revenues. Under the typical individual account proposal, the general revenues used during the transition are eventually repaid, though payback for the most generous proposals can take decades. (Clinton's proposal did not specify what happens to benefits or funding over the long run.)
Of course, the basic problem with using general revenues is that someone has to finance them.
Future generations may get stuck with the bills if the initial funds are borrowed. Given that burden, how generous should we be to the elderly? Already, much of the nation's tax capacity supports an elderly population that is growing more affluent,5 is living longer, and is healthier and able to work longer in an economy where most jobs are not physically demanding (Steuerle, Spiro, and Johnson 1999). Yet this population is retiring
earlier than was the practice before the 1980s. Do we really wish to spend our national treasure
supporting ever longer retirement as life expectancy increases? The desirability of encouraging
later retirement will grow when we begin losing the skills and experience of retiring baby boomers.
Other Risks and Uncertainties
Whether particular reforms increase national saving will be difficult to determine because nobody will know how high it would have been without reform. (Economists still argue about whether IRAs and 401(k) plans increase national saving [Poterba, Venti, and Wise 1996]!)
This is another reason to be cautious about promising overly generous benefits to future
retirees. Despite our most energetic efforts to fund the system, we might not succeed. Then the
resources necessary to back up our promises would not exist.
The growth of traditional benefits can be slowed in many ways. Indexing methods can be altered, the age of entitlement to full benefits can be increased, lifetime income can be averaged over longer periods, and so on. Different approaches have different effects on rich and
poor, male and female, married and single, and other groups. Most individual account proposals
reduce benefits most for those with high lifetime earnings. Several proposals increase the minimum benefit for those with low lifetime earnings.6 Others also reduce perceived inequities between married couples and singles, improve benefits for poor widows, or reduce inequities among different classes of divorcees (Favreault, Sammartino, and Steuerle 2002). Indeed, decisionmaking is complicated precisely because there are so many options.
The stock market is also uncertain. Its downward slide in 2001 and 2002 reminds us that
stocks are risky investments, whether purchased by government or individuals. Advocates of government purchases argue that government can better handle risk by slowly spreading the pain of
shocks across generations through changes in benefits or tax burdens. In practice, however,
politicians rarely brave reform until a crisis hits. As a result, benefit and tax changes can be
painfully abrupt. Put differently, if government were good at spreading risk, Social Security would have been reformed long ago.
Despite government's deficiencies in handling risk, most proposed reforms involving individual
accounts and those adopted in other countries transfer considerable amounts of investment
risk from individuals to government. The array of approaches includes creating a generous safety
net for the needy, guaranteeing a minimum return, subsidizing below-average rates of return
on funds toward an average, or pegging reductions in traditional benefits to each person's return from the individual accounts.7 In some approaches, the uncertainty imposed on government budgets is substantial. (The recent fall in stock prices makes it very difficult to sell the notion of making equity investment in either government or individual accounts, but lower stock prices actually enhance the prospects for a better return on investments over a working life, so the worst time to invest in equities would have been before the recent bubble burst, not now.8)
Returns from the traditional Social Security system are also uncertain. Eventual benefits
depend on lifetime earnings, the growth in the individual's earnings relative to average earnings, marital status, spouse income, and other factors. More important, no one knows how and when necessary tax and benefit reforms will be implemented.
Administrative Costs
A governmental approach to reform has one advantage over individual accounts: It costs less
to administer. Nevertheless, the cost of administering individual accounts can be held down if
government runs the information system, limits investment choices and trading frequency, and
requires less frequent reporting. Such a barebones system is likely to annually cost less than
0.3 percent of the value of individual investments. Of course, voters may well want more services
and be willing to pay for them.
Consequences of Inaction
The Congressional Budget Office projections imply that the combined cost of Social Security,
Medicare, and Medicaid will rise by 6 percent of GDP between 2001 and 2030 without reform.
Maintaining the same relationship between revenues and total spending would require increasing
the federal tax burden by an implausible 30 percent. But if spending is not curbed and taxes
are not increased, national debt will explode, making hyperinflation the only way to reduce the
debt burden—unthinkable now, perhaps, but less so with every year that passes without reform.
Possible Compromise
Any hope for compromise requires nailing down what the right and left find hard to stomach
about each other's proposals. The right believes that government lacks the discipline to increase
saving and feels nervous about government owning substantial equity investments. What the right
likes about individual accounts is that they spread capital ownership throughout the population
and can be bequeathed if a person dies before retiring.
The left has not shown the same hostility toward individual accounts as the right has
toward government intervention. Indeed, President Clinton proposed an individual account
system to supplement Social Security. It generously subsidized low-income workers' contributions
to the accounts. The left has often been at the forefront of Social Security reforms around the
world that have established individual accounts.
The left's main worry is that the leading U.S. proposals for individual accounts divert payroll tax revenues from the traditional Social Security system, which this side wants to preserve as much as possible. Even though individual account proposals can provide generous total
retirement benefits to the extent that general revenues replace payroll taxes, the left fears that eventually more benefits would come from individual accounts and less from the traditional system, especially since the accounts are sure to prove popular, particularly with the more affluent. The left worries that the traditional system and the protections it offers those who earn less would eventually wither away.
Could the left be induced to accept a compromise that combined President Bush's carve-out
approach and President Clinton's add-on approach? Current protections for low-income
groups could be bolstered by enhancing Supplemental Security Income, establishing a
new minimum benefit under Social Security, or subsidizing poorer workers' contributions to individual accounts. The president's recent Commission to Strengthen Social Security (2001)
proposed an option that combined an add-on approach with a carve-out approach and attempted
to make the whole system more progressive. The option included a reduction in traditional
benefits that could be easily converted into a proposal to index the retirement age to life
expectancy—often a feature of more liberal plans (Aaron and Reischauer 2001).
Some variant of this compromise might work politically if both sides were ready to search for
common ground. But today's emotional divides are great.9 This polarization is tragic, because every year of waiting makes reform harder and more painful.
Endnotes
This brief is adapted from remarks at the Graying of America Seminar, Knight Center for
Specialized Journalism, University of Maryland, September 2002.
1. The payout ratio has been even higher throughout most of the system's history.
2. In 1950, only 59 percent of wages were covered by the Old Age, Survivors, and Disability Insurance (OASDI) payroll tax, and the total employee-employer tax rate was 2 percent. In 1986, the portion of wages taxed peaked at 87.3 percent. It drifted down to 84.8 percent in 2000. The OASDI combined tax rate is now 12.4 percent.
3. The projections used here are consistent with the intermediate projections of the 2002 U.S. Board of Trustees of the Federal Old-Age, Survivors, and Disability Insurance Trust Funds. See U.S. Board of Trustees of the Federal Old-Age, Survivors, and Disability Insurance Trust Funds (2002).
4. As much as half of the added saving could reduce capital inflows from abroad (Helliwell 1991). This portion would not affect domestic wages, but it would add to U.S. wealth by reducing liabilities to foreigners.
5. Work done at the Urban Institute for the Social Security Administration suggests that elderly income will become less equally distributed as average incomes rise in the future. This will occur because of the increase in the number of never married, divorced, and single mothers among the retired (Toder et al. 1999).
6. This approach will convey some benefits to women from affluent households who worked outside the home during only part of their careers or earned much less than their husbands while working.
7. With this type of guarantee, government must protect itself by imposing restrictions on an individual's choice of investment.
8. The stock market's fall further reduced the political appeal of equity investments by disrupting the retirement plans of older workers who were heavily invested in equities. However, the average 401(k) investor wisely takes less risk as retirement approaches. Forty-two percent of those in their sixties hold no equity funds and reduce investments in company stocks.
9. The left vigorously attacked the President's Commission. See, for example, Diamond and Orzag (2002).
References
Aaron, Henry J., and Robert D. Reischauer. 2001. Countdown to
Reform: The Great Social Security Debate. New York: The Century Foundation Press.
Congressional Budget Office. 2002. The Budget and Economic Outlook: Fiscal Years 2003-2012. Washington, D.C.: U.S. Government Printing Office.
Diamond, Peter, and Peter Orzag. 2002. "Reducing Benefits and Subsidizing Individual Accounts: An Analysis of the Plans Proposed by the President's Commission to Strengthen Social Security." Washington, D.C.: Center on Budget and Policy Priorities and the Century Foundation.
Favreault, Melissa M., Frank J. Sammartino, and C. Eugene Steuerle. 2002. "Social Security Benefits for Spouses and Survivors: Options for Change." In Social Security and the Family, edited by Melissa M. Favreault, Frank J. Sammartino, and C. Eugene Steuerle (177-228). Washington, D.C.: Urban Institute Press.
Helliwell, John. 1991. "The Fiscal Deficit and the External Deficit: Siblings but not Twins," published in Penner, Rudolph, ed. 1991. The Great Fiscal Experiment. Washington, D.C.: Urban Institute Press.
Poterba, James M., Steven F. Venti, and David A. Wise. 1996. "How Retirement Savings Programs Increase Saving." Journal of Economic Perspectives 10(4): 91-112.
President's Commission to Strengthen Social Security. 2001. "Strengthening Social Security and Creating Personal Wealth for All Americans, Final Report." Washington, D.C.: President's Commission to Strengthen Social Security.
Romano, Roberta. 1993. "Public Pension Fund Activism in Corporate Governance Reconsidered." Columbia Law Review 93(4): 795-853.
Steuerle, C. Eugene, Christopher Spiro, and Richard W. Johnson. 1999. "Can Americans Work Longer?" Straight Talk on Social Security and Retirement Policy No. 5. Washington, D.C.: The Urban Institute.
Toder, Eric, Cori E. Uccello, John O'Hare, Melissa Favreault, Caroline Ratcliffe, Karen E. Smith, Gary Burtless, and Barry Bosworth. 1999. Modeling Income in the Near Term: Projections of Retirement Income through 2020 for the 1931-60 Birth Cohorts, Final Report. Washington, D.C.: Urban Institute Press.
U.S. Board of Trustees of the Federal Old-Age, Survivors Insurance, and Disability Insurance Trust Funds. 2002. 2002 Annual Report. Washington, D.C.: U.S. Government Printing Office.
About the Author
Rudolph G. Penner is a senior fellow at the Urban Institute, where he holds the Arjay and Frances Miller chair in public policy.
The Retirement Project
The Retirement Project is a research effort that addresses how current and proposed retirement
policies, demographic trends, and private-sector practices affect the well-being of older individuals, the economy, and government budgets.
Latest Research
Browse by Topic:
PDF
Printer-friendly summary
