Statement by Stephen C. Goss,
Deputy Chief Actuary
Social Security Administration
before the Senate Special Committee on Aging
August 26, 1997
Mr. Chairman and Members of the Committee:
I appreciate your invitation to discuss the status of Social Security's Old Age and Survivors Insurance (OASI) and Disability Insurance (DI) Trust Funds. I plan to provide some background on how the trust funds operate and to discuss the 1997 Trustees Report
Background on Trust Funds
Social Security cash benefits are paid under two programs enacted as title II of the Social Security Act. Th programs are Old-Age and Survivors Insurance (OASI) ehich pays monthly benefits to retired workers and their dependents and to survivors of deceased workers, and DIsability Insurance (DI), which pays benefits to totally disabled workers and their families. OASI and DI are financed form Social Security payroll taxes (now 6.2 percent each on employers and employees) and self-employment taxes (12.4 percent), from interest on investments, and from the proceeds of income taxation of Social Security benefits.
Although most of OASI's and DI's current revenues are used to pay for immediate expenses, excess receipts are credited to the trust fund for each program. The excess receipts of the OASI and DI Trust Funds are invested in special issue United States Government bonds. The yield on these special issue bonds equals the average return on all outstanding long-term debt of the United States. The recent yield in real terms has been between 3 and 4 percent, and is assumed to average 2.7 percent in real terms over the long run.
The restrictions in the law limiting investments of excess Social Security receipts to interest-bearing obligations of the United States or obligations whose principal and interest are guaranteed by the United States has traditionally been justified as ensuring the most safety of principal and stability of interest. This policy has also been viewed as a way of avoiding intrusion into private markets of the affairs of State or local governments.
From its inception in 1935, the Social Security program has been financed on close to a pay-as-you-go basis, with current costs met from current revenues and relatively small excess amounts hled in the trust funds. Rather than providing a mechanism for advance funding future benefit obligations of the system, the trust funds have served laregly as a contigency reserve, an accounting of excess obligational authority on the books of the Treasury.
Most experts have agreed over the years that this contingency reserve shoudl equal about 100 to 150 percent of the following year's annual expenditures in order to assure continuation of benefits in the event of short-term adverse circumstances, of either an economic or non-economic nature, that night cause unanticipated revenue shortfalls or benefit expenditures.
The Social Security Amendments of 1977 and 1983 have created a temporary period wher ethe Trust Funds will build up reserves beyond the level needed for a contigency reserve. Currently, the combined OASI and DI Trust Funds are projected to peak at 265 percent of annual program cost in the year 2011.
The 1997 Trustees Report
The 1997 Trustees Report tells u sthat the assets of the combined funds increased by $70.9 billion, from $496.1 billion at the end of December 1995 to $567.0 billion at the end of December 1996. In 1996, the Social Security trust funds took in $424.5 billion and paid out $353.6 billion. In additiona, the combined assets of the OASI and DI Tryst Funds are expected to increase from the current level of about $600 billion to over $2 trillion in 2011, or from about 160 percent of annual expenditures to about 265 percent of annual expenditures. thus, the combined OASI and DI Trust Funds, as well as each fund seperately, are adequately financed and meet the short-range test for financial adequacy for the next ten years.
The Trustees also assess the actuarial status of the trust funds over the next 75 years, the period which is considered long range for program evaluation purposes. The OASI and DI programs are out of actuarial balance for the period ending in 2071. Actuarial balance is essentially the difference between annual income and costs summarized over a given period. If the balance is negative, as it is now, the fund has an actuarial deficit.
The deficit is generally expressed in terms of a percentage of taxable payroll rather than dollars because the value of a dollar changes over time. The deficit in this year's Report for the OASDI program changed very little from last year's Report, rising from 2.19 percent of taxable payroll to 2.23 percent.
In addition, the combined OASDI and seperate OASI and DI programs do not satisfy the long-range test of close actuarial balance. This test essentially requires that the expected income for the program be at least 95 percent of the expected cost of the program over the 75-year period as a whole. In the 1997 Trustees report, the income of the OASDI program is projected to be about 86 percent of the cost of the program over the long-range period.
As has been true throughout the program's histpry, administrative expenses for the OASDI program were small in relation to benefits. Administrative expenses actually decreased by $0.2 billion compared to last year and amounted to $3.0 billion in 1996, of about 0.9 percent of benefits paid during the year.
Interest earnings on the invested assets of the combined OASI and DI trust Funds were $38.7 billion in 1996. This represented an effective annual interest rate of 7.6 percent earned by the combined assets during the year. During the same period, the average interest rate on new securities purchased b the trust funds was 6.6 percent.
The Trustees develop three alternative sets of estimates based on varying economic and demographics assumptions to show a range of possibilities regarding the financial condition of the trust funds. These estimates range from low cost (alternative I) to high cost (alternative III). Alternative II, the intermediate set of assumptions, reflects the Trustees' best estimate of what future experience will be. The projections take into account fertility rates, mortality rates, net immigration rates, productivity increases, unemployment rates, cost of living increases, and other factors, all of which are difficult to predict with very much certainty.
Under the 1997 Trustees Report's intermediate assumptions, the annual combined tax income of the OASDI program will continue to exceed annual expenditures form the funds by more than $30 billion each year through 2006. This excess will then decline, until annual expenditures frm the OASI and DI Trust Funds are projected to become larger that the funds' combined annual tax income for 2012. However, total income, including interest, would continue to exceed expenditures until 2019. The funds would begin to decline in 2019 and would be exhausted in 2029. Each of these three dates is the same as was estimated in last year's Report.
Thus, tax income alone is more than enough to pay benefits in full through the year 2011, and, with the help of the accumulated trust funds, benefits are projected to be payable in full through 2028. Even if no action were taken between now and the year 2029, and the combined trust funds were allowed to become exhausted, continuing payroll taxes and income from taxes on benefits would be expected to generate more than $600 billion in revenues (in constant 1997 dollars) for the Trust Funds in 2029. This is enough income to pay approximately 75 percent of benefit obligations. Even with projected further increases in the cost of the OASDI program throughout the long-range period, the tax income under present law would still be enough to pay over two thirds of benefit obligations in the year 2071.
The Trustees concluded, ... there is ample time to discuss and evaluate alternative solutions with deliberation and care. The size of the long-range deficit is such that long-range balance could be restored within the framework of the present program, or through plans that involve some structural change. Nonetheless, the impact of any required changes will be less disruptive the sooner they are enacted. Addressing the long-range financial imbalance of the OASDI program soon will allow changes to be phased in more gradually and with more advance notice for those affected, allowing more time to make any desired changes in planning for retirement.
Changes from the 1996 Trustees Report
As mentioned earlier, the estimated long-range actuarial deficit of the OASDI program changed from 2.19 percent of taxable payroll for the 1996 Report to 2.23 percent of payroll for the 1997 Report, based on intermediate assumptions. A number of changes contributed to this small change. the mere passage of time, changing the ending year of the long-range period from 2070 to 2071, increased the long-range actuarial deficit by 0.08 percent of taxable payroll, because of the inclusion of the large annual deficit projected for the year 2071.
The net effect of all other changes resulted in partially offsetting improvements to the estimated long-range actuarial balance for the 1997 Report. These changes include (1) modifications based on the latest demographic and economic data, (2) changed CPI- and wage-growth assumptions related to changes made by the Bureau of Labor Satistics in 1995 and 1996 in the measurement of prices for the CPI, (3) new regulations that distribute the date for payment of benefits throughout the month, and (4) an increase in the assumed ultimate real interest rate on long-term Unted States government securities.
Changes from the 1983 Trustees Report
The last comprehensive amendments to the Social Security program, enacted in 1983, were projected at the time to put the OASDI program in actuarial balance for the long-range period ending 2057. The projections in the 1983 Trustees Report indicated that the trust funds would be solvent throughout the period, but would be declining rapidly toward the end of the period and would become exhausted almost immediately after 2057.
Since 1983, a number of changes have been made that have, on balance, worsened the long-range financial projections for the program. Contributing to the currently estimated actuarial deficit of 2.23 percent of payroll in about equal amounts are (1) the gradual change in the long-range valuation period from the period ending with 2057 to the current period ending with the year 2071, (2) changes in ultimate economic assumptions, principally a reduction in the assumed real growth rate in the average wage from 1.5 percent in the 1983 Report to the current assumption of 0.9 percent, (3) increases in the cost of the Disability Insurance program, largely due to higher rates of disability among women and for persons with mental impairments, and (4) changes in the methods used for making long-range projections, primarily improvements in the method for projecting future benefit levels. Modifications in demographic assumptions since 1983, principally higher than expected birth rates between 1984 and 1996 and the asumption that immigration will add more to the United States population, have resulted in partially offsetting improvements in the long-range OASDI actuarial balance.
Clearly, 75-year projections of the cost of the OASDI program should not be considered to be precise predictions or forecasts. Over a period this long, many substantial changes will occur that cannot be predicted. These projections serve rather as a basis for considering the implications of the provisions in current law if economic and demographic trends follow patterns that are deemed to be most likely by the Trustees. Due to the great uncertainty for the future, a range of assumptions is developed to indicate what might occur if trends move in directions that are favorable or unfavorable to the financing of the program.
The favorable, or low-cost, assumptions result in projections where the OASDI program would be well financed into the indefinite future. No change in the program would be needed. The unfavorable, or high-cost, assumptions would result in exhaustion of the combined OASI and DI Trust Funds in 2018, 11 years earlier than under the intermediate assumptions.
Under the intermediate projections, the SOcial Security program will continue to receive more in taxes than it pays out through 2011. The accumulation of this excess budget authority will allow the program to continue paying benefits in full through 2028. Before 2029, changes will be necessary. Early enactment of changes will allow the changes to be phased in more gradually, and will allow for more advance notice to the affected workers and beneficiaries, so that they can effectively adjust their plans for retirement.