Report | Budget, Taxes, and Public Investment

Social Security and the Federal Deficit: Not cause and effect

Briefing Paper #273

Download PDF

Press release

(Updated August 9, 2010, 3:00pm)

The conventional wisdom in Washington is that an aging population will cause entitlement spending to balloon, driving our nation deeply into debt (see, for example, Brookings-Heritage Fiscal Seminar 2008, Peterson-Pew Commission 2009). While it is true that Medicare spending will soar if health care cost growth is not brought under control, the same is not true of Social Security spending, which is projected to level off as a share of GDP after the Baby Boomer retirement.

Social Security is running a surplus of $77 billion this year1 and amassing a trust fund large enough to last through the peak retirement years of the Baby Boomers. Social Security can only spend what it receives in tax revenues and has accumulated in its trust fund from past surpluses and interest earnings. It cannot add to the deficit if the trust fund is exhausted because the law prohibits it from borrowing (if current revenues and savings in the trust fund are not sufficient to pay promised benefits, these have to be cut). Though modest changes will be needed to put Social Security in balance over the 75-year planning period, the projected shortfall is less than 1% of gross domestic product (GDP).

The Social Security trust fund is projected to grow to a peak of about $4.2 trillion by 2024. At that point, Social Security will begin tapping its trust fund to help pay promised benefits. The trust fund itself is projected to run out around 2037. If Congress does not act to shore up Social Security’s finances before the trust fund runs out, then benefits would have to be cut by an estimated 22% because payroll taxes would be lower than benefit outlays, and Social Security would not be allowed to make up the shortfall by borrowing. As a result, Social Security cannot and would not add to the federal deficit when its trust fund is exhausted.

Even if Congress changed the law and turned to general revenues to pay promised benefits, this would increase the budget deficit by only about 4-7%. While this is more than a drop in the bucket, it would not be a major driver of the federal deficit, unlike rising health care costs or making the Bush tax cuts permanent.

We the authors—Chief Actuary of Social Security, President of the Economic Policy Institute, and an EPI economist, respectively­—have lived through previous attempts to gut the program in the name of saving it. Just as those earlier attempts failed when the facts became widely known, we believe that educating policy makers, opinion leaders, and the public about Social Security’s long-term viability is the best way to ensure that the program will be strengthened—not weakened—over its next 75 years.

In particular, we aim to dispel the myth that Social Security faces a demographic iceberg that will sink the program and our country along with it. In fact, Social Security’s actuaries fully anticipated both the Baby Boomer retirement and rising life expectancy. Though the actuaries now expect the trust fund to be exhausted sooner than they had earlier anticipated, this has nothing to do with the Baby Boomers or life expectancy, but rather a range of economic factors including growing wage inequality and rising health care costs.

Though life expectancy has risen, the labor force and productivity have grown fast enough to accommodate this trend and will continue to do so in the future. Social Security is on sound financial footing for years to come despite the Baby Boomer retirement and the worst economic downturn since the Great Depression. We should applaud this achievement while we prudently plan for the future—not weaken the only part of our retirement system that has withstood the test of time.

A pay-go system

Social Security is financed through dedicated payroll and benefit taxes2 and interest earned on previous surpluses. The biggest source of funds is a 12.4% payroll tax on earnings up to a cap (currently $106,800), split equally between employers and workers.

Social Security is primarily a pay-as-you-go (pay-go)3 system, meaning that benefits for current retirees are paid for mostly through taxes on current workers. A big advantage of a pay-go system is that it is less affected by ups and downs in financial markets than an advance-funded system. Thus, while employer-based plans and IRAs were rocked by the recent stock market downturn (see Figure A), this had no direct impact on Social Security.

The Social Security Trust Fund4

Though pay-go systems are largely immune to financial shocks, they can run into problems when demographic fluctuations raise the ratio of beneficiaries to covered workers, as with the retirement of the large Baby Boom generation born after World War II.5 However, Congress approved forward-looking reforms in 1983 that raised revenues and cut benefits in order to build up a large trust fund to help pay for the Baby Boomer retirement. Thus, Social Security is no longer a pure pay-go system, though it still receives most of its financing from current tax receipts and will continue to do so in the future.

The trust fund holds Treasury securities similar to those held by private citizens, businesses, and governments around the world.6 It holds roughly one-fifth of total Treasury debt outstanding.7 These Treasury securities are the closest thing to a risk-free asset, immune to the kind of volatility that that has wreaked havoc on 401(k) accounts and pension funds.8

Social Security and the federal budget

Social Security operates independently from the rest of the federal government, with dedicated funding sources and long-term commitments that are not subject to the annual budget process. This makes benefits more secure, since the amount spent each year on Social Security benefits is not limited to what Congress appropriates but is instead based on a benefit formula.9 Though Congress can rewrite the law to change this formula—as it did in 1983 in the form of a gradual increase in the normal retirement age10—it has been reluctant to cut benefits for older workers and retirees who would have the greatest difficulty adjusting to cuts.

The fact that Social Security benefits are an entitlement that workers earn through a lifetime of contributions does not make them an open-ended liability for taxpayers. To the contrary, the Social Security Administration is prohibited from borrowing, so if it does not have enough money to pay promised benefits in full, it will be forced to cut benefits. This could happen in 2037 when the trust fund is projected to run out, though it is much more likely that Congress will act to shore up the system’s finances before then.

It is important to understand that, because Social Security is prohibited from borrowing, it must balance its long-term budget. As Social Security Chief Actuary Stephen Goss succinctly put it: “Trust Funds enforce long-term budget neutrality. Total spending to date cannot exceed income to date” (Goss 2010). Social Security can run a short-run deficit only if it has previously run surpluses. Thus, when it is drawing down trust fund assets to pay for the Baby Boomer retirement, it will be contributing to the unified budget deficit, a measure that includes Social Security. But over time, Social Security cannot add to the federal deficit.

Short-term outlook: the recession

Compared to private-sector retirement plans, Social Security has weathered the recession well. Still, the extraordinary job losses of the recession have driven payroll tax receipts lower than had been projected. As a result, the trust fund balance is $300 billion below 2007 projections (2007 and 2010 Trustees Reports) and Social Security outlays will exceed tax revenues for the next two years, though Social Security will still run a surplus thanks to the interest it earns on its trust fund investments. But even with a sluggish recovery, a few years of lower-than-anticipated surpluses will not make an enormous difference in long-term Social Security finances, and the trust fund is projected to continue growing for the next 15 years, when it will reach $4.2 trillion (see Figure B).

Medium-term outlook: The Baby Boomer retirement

The retirement of the large Baby Boomer generation will cause Social Security spending to increase from 4.8% of GDP in 2010 to 6.1% of GDP in 2035. The Baby Boomer retirement was fully anticipated by Social Security’s actuaries and the members of the National Commission on Social Security Reform (“the Greenspan Commission”) appointed by President Reagan. As a result of reforms enacted by Congress in 1983 following the commission’s report, Social Security is in the process of building up a trust fund that will be large enough to cover benefits through the peak Baby Boomer retirement years.

Though the oldest Baby Boomers became eligible for retirement benefits in 2008, most Baby Boomers are still in the workforce. Around 2025, when the younger Baby Boomers reach retirement age, Social Security will begin drawing down the trust fund. At that point, Social Security’s outlays will start to exceed its tax revenues and interest from the trust fund, though the balance of the trust fund will ensure that full benefits can be paid through 2036 or so.

The Social Security trust fund will run out of assets around 2037. If Congress does not act before then to shore up the program’s finances, Social Security benefits would have to be cut by an estimated 22% to allow revenues to fully cover benefits. Though such an abrupt cut in benefits should certainly be avoided, the inflation-adjusted value of these benefits would still be larger than current benefits due to economic growth, though they would replace a smaller share of pre-retirement earnings (CBO 2009).

Revisions in actuarial projections for Social Security have been widely covered in the press. The trust fund is now projected to be exhausted a quarter century sooner than was anticipated in 1983, when the system was last in actuarial balance.11 This, however, has nothing to do with increases in life expectancy or the Baby Boomer retirement, which were anticipated by the Social Security Actuaries. In fact, the 1983 intermediate (“II-B”) projection has proved remarkably accurate, projecting a ratio of 33 beneficiaries per 100 covered workers for 2010, compared to 34 beneficiaries in the 2009 report.12

In an appendix to the report of the 1994-96 Advisory Council on Social Security, the Social Security actuaries pointed out that an unanticipated increase in the beneficiary-to-worker ratio—“the usual popular explanation”—was not the cause of the emerging shortfall. Instead, it was mostly the result of higher disability take-up, slower wage growth, a growing share of earnings above the taxable earnings cap, and a growing share of compensation going toward health insurance and other untaxed benefits (Advisory Council on Social Security 1997). This remains as true today as it was then.

Long-term outlook: Addressing shortfalls will require modest changes

Social Security spending as a share of the economy is projected to decline after the Baby Boomer retirement, leveling off at around 6% of GDP (see Figure C); this is a little more than 1 percentage point above current revenues as a share of GDP.

The Social Security actuaries have projected that an increase in revenues equal to just 0.6% of GDP will be sufficient to cover promised benefits over the 75-year planning period because of the savings built up in the trust fund. The Social Security shortfall averages 4% of the unified budget deficit each year, or amounts to roughly 7% of the long-term fiscal gap (the latter method gives more weight to later deficits when the economy will be larger).13

It may seem surprising that outlays are projected to level off despite rising life expectancy. The explanation is that increases in life expectancy are often offset by increases in the covered labor force and other factors. Since the mid-1980s, when Social Security was in balance, cohort life expectancy at 65 has risen steadily, but an increase in women’s labor force participation, immigration, and other factors helped stabilize the beneficiary-to-worker ratio. Social Security outlays as a share of GDP actually declined slightly over this period, from 4.5% of GDP in 1985 to 4.3% in 2007, before the recession hit.

A similar pattern is projected after the Baby Boomer retirement. Life expectancy at 65 is projected to keep rising steadily, but the ratio of beneficiaries to covered workers is expected to level off after the Baby Boomer retirement (a slight increase after 2063 is based in part on a conservative assumption that immigration will decline) (see Figure D).14 In short, the immediate demographic challenge is the retirement of the Baby Boomers, and we have met that challenge. Going forward, however, we will need to increase revenues to offset lower birth rates (Goss 2010b).

It is true that we will need to shore up the system’s long-term finances. A modest shortfall of less than 1% of GDP can be addressed by raising employer and employee payroll tax rates, by raising the cap on taxable earnings, or by some combination of the two. Raising the cap is an appealing option because in recent decades the lion’s share of increases in both earnings and life expectancy has gone to those at the top of the income distribution. For example, 10% of earnings were above the taxable earnings cap in 1983, whereas 16% were above the cap in 2008 (2009 Annual Statistical Supplement) (see Figure E). Meanwhile, the Congressional Budget Office (CBO) has warned that the growing disparities in life expectancy could affect the system’s long-term finances because those who are living longer are those with higher benefits (CBO 2008). Policies that would expand the covered workforce, reduce health care cost inflation, and promote broadly shared economic growth would also improve the system’s finances.

Americans across the political spectrum say they would prefer to address the shortfall by raising revenues rather than raising the retirement age or otherwise cutting benefits (Reno and Lavery 2009; Wright and Davies 2007). This is not surprising when you consider that the average Social Security retirement benefit is $14,000—less than the income of a full-time minimum-wage earner–yet it is the main source of income for two-thirds of seniors.

Life expectancy: A closer look shows why raising the retirement age is not an acceptable policy option

A number of people, including members of the National Commission on Fiscal Responsibility and Reform appointed by President Obama and Congress (“the Fiscal Commission”), have suggested that Social Security’s normal retirement age should increase because of rising life expectancy (Berry 2010; Sopelsa 2010). As just shown, longevity improvements are often offset by growth in the covered workforce and other factors. But even without offsetting factors, tying the retirement age to life expectancy would be overkill since it would freeze the years of covered retirement while the number of working years would continue to rise indefinitely. Meanwhile, it would not do much to improve the system’s shorter-term finances.

Moreover, increasing the retirement age would affect all workers, including those who have not seen improvements in life expectancy. A number of studies have documented that in recent decades these improvements have been concentrated among those with higher incomes and more education, especially men (CBO 2008; Cristia 2009; Cutler et al. 2010; Singh and Siahpush 2006; Waldron 2007).

Over the past quarter century, life expectancy at age 65 has increased by one year for lower-income men, compared to five years for upper-income men (see Figure F). Men in the lower half of the earnings distribution have not even caught up to where upper-income men were in 1982. In the case of women, life expectancy has grown slowly overall, with lower-income women actually seeing declines and upper-income women seeing only modest improvements. The general pattern appears to hold with older women as well, as Cristia (2009) found a substantial and statistically significant increase in mortality differentials by lifetime earnings for women aged 50-64, and a smaller and not significant increase among women aged 65-75.


Social Security cannot add to long-term deficits because it is prohibited from borrowing. Even if we assume Congress could turn to general revenues to pay promised benefits when the trust fund runs out, this would only increase the 75-year budget deficit by 4-7% per year on average.

Republican billionaire Peter G. Peterson, who has been calling for cuts to Social Security benefits for over a decade (Peterson 1999), sometimes acknowledges that Social Security is a small part of the fiscal problem (2010 Fiscal Summit). However, he and other deficit hawks usually lump Social Security in with Medicare and Medicaid in an effort to show unsustainable growth across entitlement programs (Figure G).

In reality, health care cost inflation and insufficient tax revenues are by far our biggest long-term budget challenges. Nevertheless, as Senator Max Baucus predicted, the president’s Fiscal Commission has painted a target on Social Security (Hulse 2009). The commission has set up a special task force on entitlement programs, and co-chair Erskine Bowles has indicated that Social Security must be part of a grand bargain on taxes and spending (Bowles 2010).

Why single out Social Security when the challenges it faces are modest? This reflects the success of the conservative campaign against this popular government program, aiming to undermine public confidence in its future (Butler and Germanis 1983). As Nobel–prize-winning economist Paul Krugman put it, “Doomsaying about Social Security—declaring that the program as we know it can’t survive the onslaught of retiring baby boomers—is regarded as a sort of badge of seriousness” (Krugman 2007).

Since it is impossible to argue that Social Security is a key driver of the long-term deficit, Fiscal Commission member Alice Rivlin and others claim Social Security cuts may be needed to “send a message” to bond markets (Sopelsa 2010; Blinder 2010). It is, however, unclear why bond markets should be swayed by an attempt to avoid the real problems, particularly health care cost inflation. It is also unclear why the bond market would prefer cutting benefits to raising revenues.

Meanwhile, those who would profit from a shift to privately managed retirement savings resort to mystification and alarmism in an effort to convince the public that the Social Security trust fund is suspect. One of the most startling moments in this debate came when Former Treasury Secretary Robert Rubin, who has spent much of his career on Wall Street, suggested at an event hosted by Peter G. Peterson that the U.S. Treasury could default on its obligations to the trust fund (Rubin 2010). The Treasury bonds in which the trust fund invests are backed by the full faith and credit of the United States government, which has never defaulted on any debt in its entire history.

The fact that programs like Social Security and Medicare have long-term, dedicated funding sources means that they are more susceptible to criticism for not being in balance, even when the projected shortfall is modest and distant, as in the case of Social Security. Perversely, policies like the Bush tax cuts that were never paid for with offsetting tax increases or spending cuts receive less scrutiny because no federal agency is required by law to estimate their impact on the federal budget 75 years into the future.

Addressing Social Security’s projected future shortfalls is not a case of Americans wanting to eat their cake and have it too. Polls have consistently shown that voters across the political spectrum oppose benefit cuts but would be willing to pay more to preserve and strengthen the program (Reno and Lavery 2009; Wright and Davies 2007). Nor is it is a problem of lack of foresight, even though Fiscal Commission co-chair Alan Simpson absurdly claims “they never knew there was a Baby Boom in ’83” (Berry 2010). The fact is that Social Security’s actuaries did foresee the coming Baby Boomer retirement wave, and Congress acted to build up the trust fund explicitly to help finance the Boomers’ retirement.

Meanwhile, ordinary workers, especially younger workers who will bear the brunt of any benefit cuts, cannot afford further reductions in benefits. Younger workers already face a higher normal retirement age of 67, and therefore a lower income-replacement rate than that of previous generations. The higher retirement age, coupled with a shift from traditional pensions to 401(k) plans, means that younger workers are much more likely to face a sharp drop in living standards at retirement (Figure H) than are the generations preceding them.

As Fiscal Commission co-chair Alan Simpson has said, we should worry about our grandchildren. This is why we should ignore his advice to raise the retirement age and instead strengthen Social Security for the long term.


  1. Unless otherwise noted, all statistics cited are from the 2010 Social Security Trustees Report, including the statistical supplement and single-year tables available on the Social Security Web site.
  2. A portion of the Social Security benefits of higher-income retirees is subject to income tax and the proceeds are dedicated to Social Security.
  3. In the federal budget process, pay-go can also refer to a requirement that direct spending or tax changes not add to the federal deficit. In this paper, pay-go will refer to pay-as-you-go pension financing.
  4. Technically, Social Security has two trust funds—one for the Old Age and Survivors Insurance (OASI) and one for Disability Insurance (DI). However, it is common to refer to the combined (OASDI) trust funds as a single fund.
  5. The U.S. Census Bureau defines a Baby Boomer as someone born between 1946 and 1964.
  6. Though these “special issue” securities are not publicly traded, the interest on them is based on the market yield on Treasury securities held by the public.
  7. Of the $11.9 trillion in federal debt, $2.6 trillion is held by Social Security, $1.8 trillion is held by other government entities, and $7.5 trillion is held by the public (OMB 2010; Social Security Trustees 2009).
  8. Bonds are subject to interest rate risk if not held to maturity. However, this is not an issue for the bonds in the Social Security trust fund, which are redeemable at face value at any time.
  9. To further secure Social Security benefits, the program was explicitly exempted by the “Byrd Rule” from the purview of the budget reconciliation process, which is used by Congress to cut direct spending programs.
  10. The retirement age is gradually increasing from 65 to 67 over a 22-year period beginning in 2000. This is equivalent to a 13% benefit cut for a worker retiring at age 65. Workers who were 45 or younger when the policy was implemented in 1983 were affected, with the full impact felt by workers then in their early 20s.
  11. The more conservative of the two intermediate projections (Alternative II-B) in the 1983 Trustees Report projected that the trust fund would be dwindling as the 75-year projection period came to a close.
  12. The 1983 report somewhat underestimated the growth in male life expectancy and disability take-up, but this was offset by slower-than-anticipated growth in female life expectancy and faster-than-expected labor force growth.
  13. The lower measure is equal to the average annual Social Security surplus (deficit) projected by CBO as a share of the unified budget deficit, based on the alternative fiscal scenario. The higher measure is the Trustees’ summary present-value estimate of the unfunded obligation (0.6% of GDP) divided by CBO’s summary fiscal gap measure (8.7% of GDP). CBO’s alternative fiscal scenario incorporates “changes to current law that are widely expected to occur or that would modify some provisions of law that might be difficult to sustain for a long period.” This includes indexing Alternative Minimum Tax and extending most of the Bush tax cuts. It also assumes that “several policies enacted in the recent health care legislation that would restrain growth in health care spending would not continue in effect after 2020” (CBO 2010). Though this is a controversial assumption, this is the projection used by the president’s Fiscal Commission.
  14. The Trustees’ intermediate projection assumes that net immigration will gradually decline from approximately 1.3 million a year before the recession to 1.0 million a year in 2085. Their low-cost projection assumes, among other things, that immigration will level off at around 1.3 million during the second half of the planning period, and that the beneficiary-to-worker ratio will also stabilize, albeit at a somewhat higher level than today. While immigration in recent years has been unusually high, it is not unrealistic to assume that these levels could persist or even increase given global population growth.


Advisory Council on Social Security. 1997. Report of the 1994-96 Advisory Council on Social Security, Volume I: Findings and Recommendations, Appendix 1, Developments Since 1983 (Washington, D.C., January 1997), available at

Bell, Felicitie, K. Mark Bye, and Al Winters. 2008. “Unisex life expectancies at birth and age 65,” Actuarial Note, Social Security Administration Office of the Chief Actuary, June.

Berry, John M. 2010. Simpson’s comments undermine commission efforts. The Fiscal Times, June 20.

Blinder, Alan S. 2010. Return of the bond market vigilantes. The Wall Street Journal, May 20.

Bowles, Erskine. 2010. Remarks made at the “Fiscal Summit: America’s Challenge and a Way Forward,” Sponsored by the Peter G. Peterson Foundation, Ronald Reagan Building, Washington, D.C., April 28.

Brookings-Heritage Fiscal Seminar. 2008. “Taking Back Our Fiscal Future.” April.

Butler, Stuart and Peter Germanis. 1983. Achieving a ‘Leninist’ strategy. Cato Journal. Fall.

Congressional Budget Office. 2008. “Growing Disparities in Life Expectancy.” Washington, D.C.: CBO. April 17.

Congressional Budget Office. 2009. CBO’s Long-Term Projections for Social Security: 2009 Update, Additional Information. Washington, D.C.: CBO. August.

Congressional Budget Office. 2010. The Long-Term Budget Outlook. Washington, D.C.: CBO. June.

Cristia, Julian. 2009. Rising Mortality and Life Expectancy Differentials by Lifetime Earnings in the United States. IDB Working Paper. Washington, D.C.: IDB. January (subsequently published in Journal of Health Economics).

Cutler, David M., Fabian Lange, Ellen Meara, Seth Richards and Christopher J. Ruhm. 2010. “Explaining the rise in educational gradients in mortality,” NBER Working Paper 15678, Cambridge, Mass.: NBER. January.

Federal Reserve, 2010. Flow of Funds, Tables L.118b and L.118c. Washington, D.C.: Federal Reserve. June.

Goss, Stephen C. 2010. “Social Security: Part I. The System, Solvency, and Sustainability,” Presentation to the National Commission on Fiscal Responsibility and Reform, May 12.

Goss, Stephen C. 2010b. “The Future Financial Status of the Social Security Program,” Social Security Bulletin, Vol. 70, No. 3.

Hulse, Carl. 2009. Baucus: Don’t outsource my committee. New York Times online, December 10.

Krugman, Paul. 2007. Played for a sucker. New York Times, November 17.

Office of Management and Budget. 2010. Analytical Perspectives. Budget of the U.S. Government, Fiscal Year 2011. Washington, D.C.: OMB.

Peterson, Peter G. 1999. Grey Dawn: How the Coming Age Wave will Transform America, and the World. New York: Crown Publishing Group.

Peterson-Pew Commission on Budget Reform. 2009. “Red Ink Rising: A Call to Action to Stem the Mounting Federal Debt.” April. Washington, D.C.: Pew Charitable Trusts.

Reno, Virginia P., and Joni Lavery. 2009. “Economic Crisis Fuels Support for Social Security: Americans’ Views on Social Security.” Washington, D.C.: National Academy of Social Insurance. August.

Rubin, Robert. 2010. Remarks made at the “Fiscal Summit: America’s Challenge and a Way Forward,” Sponsored by the Peter G. Peterson Foundation, Ronald Reagan Building, Washington, D.C., April 28.

Senate Aging Committee. 2010. “Social Security modernization: Options to address solvency and benefit adequacy.” May 13.

Singh, Gopal K., and Mohammad Siahpush. 2006. Widening socioeconomic inequalities in U.S. life expectancy, 1980-2000. International Journal of Epidemiology, May 9.

Social Security Administration. 2010. “Annual Statistical Supplement, 2009,” February.

Social Security Trustees. 2009. The 2009 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds, May 12.

Social Security Trustees. 2010. The 2010 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds. August 5 (including Single-Year Tables Consistent with the Report).

Sopelsa, Brooke. 2010. “Raise Social Security Retirement Age: Budget Expert.”, June 10.

Waldron, Hilary. 2007. Trends in Mortality Differentials and Life Expectancy for Male Social Security-Covered Workers, by Socioeconomic Status, Vol. 67, No. 3.

Wright, William, and Curt Davies. 2007. AARP. Retirement Security Survey Report, February 19. Washington, D.C.; AARP.