Social Security Financing Reform: Lessons from the 1983 Amendments

97-741 EPW CRS Report for Congress Received through the CRS Web Social Security Financing Reform: Lessons From the 1983 Amendments July 24, 1997 David Stuart Koitz Specialist in Social Legislation Education and Public Welfare Division Congressional Research Service ˜ The Library of Congress Social Security Financing Reform: Lessons From the 1983 Amendments Summary In April 1997, the Social Security board of trustees released its latest appraisal of Social Security’s financial condition. As in earlier reports, the board projected that the system will face long-range problems under its “best guess” or middle-of-the-road forecast. When changes were last made to Social Security in 1983, the prognosis was that it would be solvent for 75 years. Since that time, actuarial deficits have reemerged and grown worse. The current report projects insolvency in 2029. As the prospects grow for another round of reform, a number of observations can be made about the 1983 amendments that may be helpful in considering future changes. Various misperceptions of their intent have developed over the years, among them being that Congress wanted to create surpluses to “advance fund” the benefits of post World War II baby boomers. This view has affected congressional debates ranging from cutting Social Security taxes to adopting a constitutional amendment to balance the budget. Some would argue that the budget deficits the government has run since 1983 have subverted the amendments. There is, however, little evidence to support the view that the surpluses were intended to pay for the baby boomers’ retirement. The record suggests that the goal was to assure that the system would not be threatened by insolvency again, not to advance fund future benefits. Other little-understood aspects of the amendments include how Social Security’s financing problem was measured. While using pessimistic assumptions was considered prudent in dealing with the near term, using them to assess the long run would have been seen as exaggerating the problem. However, using middle-of-theroad assumptions left no room for a later worsening of assumptions, which in fact occurred. Moreover, discussions in the key congressional committees revolved around the average 75-year deficit and how much the various options would affect it. There was little understanding that a period of surpluses would be followed by a period of deficits — or that solvency was not achieved on a pay-as-you-go basis. The role of a select panel formed by President Reagan and congressional leaders in an attempt to reach a bipartisan solution also is not well understood. The 1982/1983 National Commission on Social Security Reform is often cited as a model for resolving otherwise intractable political problems. Although the Commission brought various factions together and served as a framework for later action, its plan may have been the product of only a few of its 15 members working with officials of the Reagan Administration. Moreover, the threat of insolvency — projected then to occur in less than 6 months — may have been the real catalyst for action. Finally, while today there is widespread recognition that looming demographic shifts may significantly raise federal entitlement spending early in the next century, this was much less of a concern in 1983. This is not to suggest that the demographic bulge of retiring baby boomers was not observed. It was, but Social Security’s financing problems were viewed and tackled in isolation. Today, they are seen much more as a segment of the strain that total federal retirement and health care spending may create. Hence, to some extent the context for reform may have shifted. Contents The Emergence of Social Security Surpluses — by Accident or Design? . . . . . . 2 The Amendments Provided Little Cushion for Misestimates in the Long Run . . . 6 The Use of “Average” 75-Year Estimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 The Role of the National Commission on Social Security Reform . . . . . . . . . . 12 Social Security Within the Larger Context of Federal Entitlements . . . . . . . . . . 13 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 List of Tables Table 1. Projected Reserve Cushion of the Social Security Trust Funds Made Upon Consideration of the 1983 Amendments . . . . . . . . . . . . . . . . . . . . . . 4 Table 2. Comparison of Actual Social Security Trust Fund Balances to Those Projected After Enactment of the Social Security Amendments of 1983 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Table 3. Long-Range Impact of 1983 Amendments Projected at Various Stages in Their Legislative Development . . . . . . . . . . . . . . . . . . . . 7 Table 4. Comparison of Major Assumptions Underlying Long-Range Social Security Projections in 1983 and 1997 Trustees’ Reports . . . . . . . . . 8 Table 5. Long-Range Actuarial Status of the OASDI Trust Funds as Shown Under the Intermediate Assumptions in the Trustees’ Reports Issued Over the Period From 1983 to 1997 . . . . . . . . . . . . . . . . . . . . . . . . 9 Table 6. Social Security Projections Made After Enactment of the Social Security Amendments of 1983 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Social Security Financing Reform: Lessons From the 1983 Amendments In April 1997, the Social Security board of trustees released its latest annual appraisal of the Social Security system’s financial condition. Like their previous 7 reports, the 1997 report projected that the system would face significant long-range financing problems. Under the central (or best guess) forecast in the report, the system would have an average actuarial deficit of 2.23 percentage points of taxable payroll.1 In layman terms, it would need, on average, 17% more income than projected to cover its costs over the next 75 years. The trustees estimated that the Social Security trust fund balances would peak in 2019 and decline thereafter until they were exhausted in 2029. At that point the system’s ongoing income would cover only 75% of its projected outgo. When major changes were last made to Social Security in 1983, the prognosis was that it would be solvent both in the near term and long run. The actuaries of the Social Security Administration (SSA) estimated that on average the system’s income and outgo would be closely matched over their 75-year valuation period. Since that time, actuarial imbalances have re-emerged and have progressively worsened. Although income is projected to exceed outgo for the next 2 decades, the costs of supporting an increasingly elderly population is projected to lead to the program’s insolvency in 2029. As the prospects grow for another round of reform, there are a number of observations that can be made about the 1983 amendments that may be helpful in the consideration of future changes. Over the years a number of misperceptions of the goals of those amendments have developed, among them being their purported intent to create surpluses to “advance fund” the benefits of post World War II baby boomers and later retirees. This particular misperception has affected congressional debates on a wide variety of issues — ranging from reductions in Social Security taxes, investment of the trust funds, raising the federal debt ceiling, adoption of a balanced budget amendment to the Constitution, to passage of annual budget resolutions and reconciliation measures. Other little-understood aspects of their development include (1) how Social Security’s problem was perceived and defined in 1983; (2) what measurements of the problem were considered critical in reaching a consensus; (3) 1 Its average costs would equal 15.60 percentage points of taxable payroll; its average income would be 13.37 percentage points; leaving an average deficit of 2.23 percentage points. Taxable payroll represents the total amount of wages, salaries, and self employment income in the economy subject to Social Security taxation. It is estimated to be $3.2 trillion in 1997. Thus, having an average deficit of 2.23% of taxable payroll would be the equivalent of having a $71 billion shortfall in 1997 and correspondingly larger amounts in each of the next 74 years. CRS-2 what the role was of a blue-ribbon panel established by President Reagan and congressional leaders in an attempt to reach a bipartisan solution (the National Commission on Social Security Reform)2; and (4) how the future financing needs of Social Security were perceived within the context of the potential strain of future entitlement spending generally. This report discusses these aspects of the amendments. The Emergence of Social Security Surpluses — by Accident or Design? The last major reform of the Social Security system occurred in 1983 with enactment of the Social Security Amendments of 1983 (P.L. 98-21). In response to the system’s imminent insolvency and a long-range actuarial imbalance (as reported by the system’s trustees), Congress made major changes to the system’s taxing and benefit provisions. In fact, it has become conventional wisdom that Congress deliberately The goal of the 1983 Social Security intended to built up large amendments was not to create balances in the trust funds, not surpluses, but to assure that the just for the near term, but to help finance the benefits of the system would not be threatened by post World War II baby boomers insolvency again in the event and later retirees. Moreover, adverse conditions arose. some would contend that by running deficits in the rest of the government over the past 14 years, this goal of the 1983 amendments has been (and is being) subverted. There is, however, very little evidence to support the view that the surpluses arising after passage of the amendments were intended, and that they were designed to be on top of a balanced budget in the rest of the government. To the contrary, a review of the record of congressional proceedings would suggest that the goal was not to create surpluses, but to assure that the system would not be threatened by insolvency again in the event adverse conditions arose. In the 2-year period leading up to the amendments, the system’s immediate financial outlook had deteriorated rapidly. This occurred in spite of major remedial 2 President Reagan proposed the creation of the Commission in September 1981 after a summer of heated partisan debate over the necessity of changing the system. It had been precipitated by a controversial set of changes, mostly benefit reductions, proposed by the Administration four months earlier to address the system’s financing problems. After bipartisan consultations with congressional leaders, the Commission was established by executive order on December 16, 1981. It was comprised of 15 members, seven of whom were then sitting members of the House or Senate. Alan Greenspan, currently chairman of the Board of Governors of the Federal Reserve System, served as its chairman. CRS-3 changes enacted in 1977 and other changes enacted in 1980 and 1981.3 As the system’s outlook worsened in 1981 and 1982, a great deal of skepticism developed about the effectiveness of the previous steps that had been taken, particularly the 1977 changes. The common perception was that they had failed, at least initially, since the expectation upon enactment was that they would assure solvency until early in the next century. Their failure was particularly troublesome to many Members of Congress who had voted for tax increases that had already taken place by 1982.4 Moreover, there was still considerable uncertainty about economic conditions as the nation was just emerging from back-to-back recessions in 1980 and 1982. As a result, there was a general feeling that a cushion of sorts needed to be built into the next set of Social Security changes. Congress largely accomplished this in 1983 by assessing the potential impact of the changes it was considering using fairly pessimistic assumptions for the near future. The House Ways and Means and Senate Finance Committees had asked SSA’s actuaries for projections of the impact of their proposals under both the Social Security trustees’ traditional middle-of-the-road assumptions — the so-called “intermediate II-B” assumptions — and their pessimistic ones (and neither expressed interest in projections under optimistic conditions). Their primary concern was whether solvency would be maintained in the 1980s under pessimistic conditions. Surpluses of income over outgo were projected using the intermediate II-B assumptions but not using the pessimistic ones.5 In his memorandum transmitting estimates to the House Ways and Means Committee (of the impact of the committee’s bill), SSA’s deputy chief actuary stated “The projections... indicate that the OASDI program would be in a very borderline situation under H.R. 1900, if future economic conditions were to follow the [pessimistic] assumptions. It is not possible to state with certainty whether benefits could or could not be paid on time under these assumptions... it is very possible that the OASDI trust funds would be unable to pay benefits on time in late 1987 and early 1988...”6 3 The Social Security Amendments of 1977, P.L. 95-216; the Social Security Disability Amendments of 1980, P.L. 96-265; the Reallocation of Social Security Tax Receipts Act, P.L. 96-403 (providing for a 2-year tax reallocation from DI to OASI); the Omnibus Budget Reconciliation Act of 1981, P.L. 97-35; and P.L. 97-123 (providing for interfund borrowing among the trust funds, including the Medicare Hospital Insurance (HI) trust fund). 4 When the 1977 amendments are viewed as a whole, the tax increases were a relatively small part. Most of the major provisions involved changes in Social Security’s benefit computation rules designed to remove a flaw that threatened to greatly inflate the system’s future costs. The tax increases, however, went into effect immediately. They were mostly short-run in nature, whereas the benefit changes were mostly long-range in nature. After enactment, there were tax rate increases in 1978, 1979, 1981 and 1982, and by 1982 the maximum level of earnings subject to the Social Security tax had been raised by $8,100 more than it would have under the old law ($32,400 versus $24,300). 5 These were preliminary 1983 trustees’ report estimates given to the House Ways and Means and Senate Finance Committees (identical letters dated February 18 and 22, 1983 from Social Security Commissioner John Svahn to Representative Daniel Rostenkowski and Senator Robert Dole). 6 Memoranda from Richard Foster, Deputy Chief Actuary, SSA, “Estimated Short-Range (continued...) CRS-4 Apprehension about the estimates was so strong, particularly in the Senate Finance Committee, that Congress ultimately adopted various “fail-safe” measures. Although modified in the final bill, one such measure reported from the Finance Committee was estimated under pessimistic assumptions to result in $100 billion in benefit reductions in the 1985-1989 period alone.7 Table 1. Projected Reserve Cushion of the Social Security Trust Funds Made Upon Consideration of the 1983 Amendments House bill (H.R. 1900) Calendar year Intermediate II-B projections Pessimistic projections Senate bill (S. 1) Intermediate II-B projections Pessimistic projections* (Balance of funds at beginning of year as % of outgo during the year) 1983 15% 15% 15% 15% 1984 22% 21% 22% 21% 1985 21% 18% 22% 20% 1986 22% 16% 23% 21% 1987 22% 12% 25% 22% 1988 22% 9% 23% 19% 1989 29% 12% 33% 32% 1990 38% 15% 43% 46% 1991 50% 20% 59% 61% 1992 64% 26% 76% 76% Source: Series of Memoranda from Richard Foster, Deputy Chief Actuary, SSA, Estimated ShortRange Financial Effects of H.R. 1900 and S. 1, March 4, 7 and 15, 1983. (Reflect bills as reported from the House Ways and Means and Senate Finance Committees). * The reason these figures appear as large as those under the Intermediate II-B projections is that they assume large benefit constraints would be triggered by the various “fail safe” measures proposed by the committee. See footnote 7 below. 6 (...continued) Financial Effects of H.R. 1900 as Reported by the Committee on Ways and Means on March 4, Based on the 1983 Alternative II Assumptions,” March 7, 1983. OASDI is the acronym for Old-Age, Survivors, and Disability Insurance, the title in the law for Social Security cash benefits. 7 The reductions would have gone on indefinitely. As reported out of Committee, the provision required cutting back benefit increases to the extent necessary to keep the balances of the trust funds from falling below 20% of annual expenditures. In the final bill, a so-called “stabilizer” provision was enacted requiring that benefit increases be based on the lower of wage or price growth if the trust funds’ balances fell below 15% of annual expenditures before 1989 and 20% for 1989 and thereafter. The measure has never been triggered. CRS-5 As it happened, the surpluses of income over outgo and the corresponding increases in trust fund balances that arose during the first 10 years after enactment greatly exceeded even those projected under the intermediate II-B projections (as shown in Table 2 below). The economy went though a sustained growth period after the 1982 recession — Gross Domestic Product (GDP) grew at rates ranging from 3% to 6.8% a year from 1984 through 1989 (rates of 3% to 3.9% had been projected) and the inflationary spike of the late 1970s and early 1980s subsided to a much greater degree than projected (the Consumer Price Index rose at rates ranging from 1.6% to 4.8%; rates of 5% to 6.2% had been projected).8 As a result, surplus Social Security income exceeded the projections made in 1983 and the balances of the trust funds grew to $331 billion by the end of 1992, $86 billion more than projected.9 Table 2. Comparison of Actual Social Security Trust Fund Balances to Those Projected After Enactment of the Social Security Amendments of 1983 1983 Intermediate II-B Projections Actual Calendar year Trust fund balances, end of year ($s in billions) Balance at beginning of year as % of annual outgo Trust fund balances, end of year ($s in billions) Balance at beginning of year as % of annual outgo 1985 $34 22% $42 24% 1990 $134 39% $225 75% 1992 $245 64% $331 96% Source: 1997 OASDI trustees’ report, and Memorandum from Richard Foster, Deputy Chief Actuary, SSA, “Short-Range Financial Status of the Social Security Program Under the Social Security Amendments of 1983,” April 6, 1983. Hence, while trust fund balances of some magnitude were shown to be possible under one set of projections provided to Congress in 1983, these were not the projections around which the 1983 amendments were designed. To suggest that these balances were intended to finance or “advance fund” the benefits of the baby boomers and subsequent retirees presumes that the authorizing committees (and the Congress generally) designed the measures deliberately to create significant excess income and 8 Assumption used in 1983 derived from memorandum from Harry C. Ballantyne, Chief Actuary, SSA, “Short-range OASDI Estimates Based on 1983 Assumptions,” February 14, 1983. 9 It is worth noting that the trust funds’ balance is now fairly close to that projected under the intermediate II-B assumptions in the 1983 trustees’ report. Although not published in that report, projections of long-range, dollar denominated balances were included in an actuarial note published by SSA’s chief actuary in October 1983. They showed estimated trust fund balances reaching $589 billion at the end of 1996. The actual level was $567 billion. (See Actuarial Note #117, “Long-Range Projections of Social Security Trust Fund Operations in Dollars,” by Harry C. Ballantyne, Chief Actuary, Office of the Actuary, SSA). CRS-6 believed that this excess income would be isolated from the financial operations of the rest of the government such that it would have accumulated as a “nest egg.” Neither of the reports from the House Ways and Means and Senate Finance Committees made any reference to such “advance funding.” A lengthy discussion of the potential for substantial trust fund accumulations under pre-1983 law did take place among members of the National Commission on Social Security Reform on September 20, 1982, but considerable skepticism was expressed about the projections because of their “vulnerability to economic fluctuations,” i.e., that they might be too optimistic (from unpublished minutes of the proceedings of the Commission). No record exists showing that Congress seriously considered “advance funding” the system in 1983 or how such action would actually save resources. The Amendments Provided Little Cushion for Misestimates in the Long Run Estimates made in 1983 of the long-range condition of the system, as well as estimates of the impact of various measures to improve it, were expressed as 75-year “averages” and as percentages of taxable payroll. At the time that the amendments were considered, the system’s average 75-year cost was estimated to be 14.38% of taxable payroll; its average income was 12.29%; and its average deficit, 2.09%. The general approach taken by the Little cushion for long-range Ways and Means and Finance Committees was to enact enough changes to the misestimates was built into system so that (1) the possibility of nearthe 1983 amendments. term insolvency (i.e., the trust funds running down to a zero balance) would be remote and (2) the long-range average actuarial imbalance of 2.09% of taxable payroll would be eliminated in subsequent trustees’ reports. While both the intermediate II-B and pessimistic assumptions were used to measure near-term effects, only the more optimistic (intermediate II-B) assumptions were used to measure long-term effects. The precedent for the two different approaches had largely been set by the National Commission. There had been a lengthy and contentious debate during the preceding 18 months both in the Commission and in Congress over the extent of the long-term problem and the underlying assumptions. Reaching a consensus on how to view and measure it had been considered a major accomplishment of the Commission. By using only the middle-of-the-road, intermediate assumptions, a framework had been established for working toward a long-range solution. The earlier remedial changes enacted in 1977 had not eliminated the long-range problem. A projected shortfall of 1.45% of payroll had been left unresolved at the time of enactment, and in the intervening years between 1977 and 1983, that gap had not increased significantly (i.e., growing to 2.09% of payroll by early 1983). It was the short-range situation that had markedly deteriorated. In essence, the failure of the 1977 changes was viewed more as a short-run phenomenon than a long-range one. Whereas the use of pessimistic assumptions in 1983 was generally seen as an act of prudence in dealing with the near term problem, planning for the long run on a pessimistic basis would have been seen as “exaggerating” the problem and as an “assault” against the system. CRS-7 There was so much skepticism during that period that some of the system’s supporters suggested shortening the valuation period, for instance, from 75 to 50 years (or to an even shorter period). This would have eliminated or greatly reduced the problem (the presumption being that the estimates would inevitably be wrong anyway). Thus, the use of these middle-of-the-road projections for measuring longrange changes was both a political and substantive compromise.10 Table 3. Long-Range Impact of 1983 Amendments Projected at Various Stages in Their Legislative Development House Action on H.R. 1900 Stage of process Amount of improvement created by bill Actuarial balance (+) or imbalance (-) Senate action on S. 1 Amount of improvement created by bill Actuarial balance (+) or imbalance (-) (% of taxable payroll) Long range imbalance before amendments -2.09 -2.09 With changes passed by Committees +2.12 +0.03 +2.17 +0.08 With changes passed by House or Senate +2.08 -0.01 +2.17 +0.08 With passage of conference agreement +2.07 -0.02 +2.07 -0.02 Source: House Ways and Means and Senate Finance Committee reports to their respective chambers on the Social Security Act Amendments of 1983 (House report no. 98-25, March 4, 1983 and Senate report no. 98-23, March 11, 1983); Comparison of Provisions of H.R. 1900, the Social Security Act Amendments of 1983, March 23, 1983; memoranda from Francisco R. Bayo, Deputy Chief Actuary, SSA, “Long-Range OASDI estimates for H.R. 1900 as Passed by the Congress,” March 25, 1983, and “Final Estimated Long-Range OASDI Cost Effect of Social Security Act Amendments of 1983 (P.L. 98-21),” April 26, 1983. The estimates in the table were based on preliminary intermediate II-B assumptions of the 1983 trustees’ report. The actual 1983 trustees’ report, issued on June 24, 1983, showed the system as having an actuarial balance of +0.02% of taxable payroll under its intermediate II-B projections. 10 In 1981, the Reagan Administration had introduced “worst case” projections. These were 6-year projections based on more adverse economic assumptions than contained in the trustees’ pessimistic scenario. While they may have been designed from the perspective of a prudent manager, many viewed them as an attempt to make Social Security’s financial condition look worse than it was to justify making deep cuts in benefits. An atmosphere of distrust emerged and persisted throughout the period leading up to the 1983 legislation. CRS-8 However, the downside of using them was that they left no room for misestimates. The projections furnished to the committees at each step in the process repeatedly showed that the bills would achieve actuarial balance, but the margin was narrow. Simply stated, there was no room for a later worsening of assumptions. When the effect of the amendments was shown under pessimistic assumptions three months after enactment (with the release of the 1983 trustees’ report),11 it reflected an average long-range deficit of 3.51% of taxable payroll and projected that the trust funds would become insolvent in 2027. As a matter of interest only, this insolvency point is remarkably close to that shown in the latest trustees’ report, i.e., 2029. As the following table shows, however, this is by accident only, since the assumptions underlying the two sets of projections are very different. Table 4. Comparison of Major Assumptions Underlying Long-Range Social Security Projections in 1983 and 1997 Trustees’ Reports Long-range assumptions 1983 Intermediate II-B projections 1983 Pessimistic projections 1997 Intermediate projections Annual increase in: —wages in covered employment 5.5% 6.0% 4.4% —Consumer Price Index 4.0% 5.0% 3.5% Unemployment rate 5.5% 6.5% 6% Annual interest rate 6.1% 6.6% 6.2% Fertility rate (births per woman) 2.0 1.6 1.9 —women 84.4 89.7 83.5 —men 76.3 81.3 78.1 400,000 350,000 900,000 Life expectancy in 2060 (at birth): Annual net immigration Source: 1983 and 1997 OASDI Trustees’ Reports. In the years following the 1983 legislation, the long-range projections have progressively worsened. Starting in 1984, annual trustees’ reports again showed long-range actuarial imbalances. Although small at first, for the past 4 years they have 11 The conference committee on the amendments met and reached agreement on March 24, 1983. The 1983 trustees’ report was issued on June 24, 1983. CRS-9 reached a level comparable to the imbalance that Congress addressed in 1983 (see Table 5 below and chart of page 10). Table 5. Long-Range Actuarial Status of the OASDI Trust Funds as Shown Under the Intermediate Assumptions in the Trustees’ Reports Issued Over the Period From 1983 to 1997 Average Actuarial income rate Average balance or (income cost imbalance Year of report from taxes) rate (-) (% of taxable payroll) Imbalance as % of average tax income Year that trust funds were projected to become insolvent OASI DI Combined 1983 preliminary* 1983 final* 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 12.29 12.87 12.90 12.94 12.96 12.89 12.94 13.02 13.04 13.11 13.16 13.21 13.24 13.27 13.33 13.37 14.38 12.84 12.95 13.35 13.40 13.51 13.52 13.72 13.95 14.19 14.63 14.67 15.37 15.44 15.52 15.60 -2.09 0.02 -0.05 -0.41 -0.44 -0.62 -0.58 -0.70 -0.91 -1.08 -1.47 -1.46 -2.13 -2.17 -2.19 -2.23 17.01% *** 0.39% 3.17% 3.40% 4.81% 4.48% 5.38% 6.98% 8.24% 11.17% 11.05% 16.09% 16.35% 16.43% 16.68% 1983 ** ** 2050 2054 2055 2050 2049 2046 2045 2042 2044 2036 2031 2031 2031 ** ** 2050 2034 2026 2023 2027 2025 2020 2015 1997 1995 1995 2016 2015 2015 1983 ** ** 2049 2051 2051 2048 2046 2043 2041 2036 2036 2029 2030 2029 2029 Source: 1983-1997 OASDI Trustees’ Reports, intermediate or intermediate II-B projections. Figures for 1983-1990 represents the intermediate II-B forecast. There were two intermediate forecasts made annually from 1981-1990 — II-A and II-B. The II-B forecast was considered the one most closely aligned with traditional “intermediate” forecasting. * 1983 “preliminary” are estimates made prior to enactment of 1983 amendments; 1983 “final” are estimates appearing in the 1983 trustees’ report published in June 1983 (after enactment). ** The fund was not projected to be exhausted within the 75-year projection period, i.e., from 19832058. *** There was a positive balance, representing 0.16% of average tax income. CRS-10 Projected 75-Year Average OASDI Deficit as % of System's Total Tax Income 18% 16% 14% 12% 10% 8% 6% 4% 2% 1996 1994 1992 1990 1988 1986 1984 1983 after 1983 before 0% Year of trustees' report Source: OASDI trustees' report, 1983-1997, intermediate projections (1983 "before" represents the forecast made prior to enactment of the 1983 amendments) The Use of “Average” 75-Year Estimates The reports on the 1983 amendments from the House Ways The use of “average” 75-year and Means and Senate Finance estimates obscured the deficits Committees contained year-to-year estimates of the impact of their projected to arise in the latter respective bills only through 1992 as half of the projection period. did the materials submitted to the Committees by SSA’s actuaries prior to their deliberations.12 There was only minimal recognition that there would be trust fund surpluses after 1992 under projections of “current law” and no projections or discussion of the year-toyear funds flow after 1992 resulting from any of the measures being considered.13 Those estimates first appeared in the 1983 trustees’ report (which came out 3 months 12 The eventual report from the conferees for the two bodies contained no estimates of the impact of the bill, and the “markup” document used by the conferees contained year-to-year estimates only through 1989. Average 75-year impacts, provision by provision, expressed as a percent of taxable payroll, were used to show long-range effects. 13 One table among a number furnished to Representative Rostenkowski and Senator Dole in the February 18, 1983 letters from Commissioner Svahn (see footnote 5) showed long-range projections in 5-year increments out to 2060. These preliminary 1983 trustees’ report estimates showed some surpluses would arise under pre-1983 law during the 1990 to 2010 period. The letters also provided projections of the impact of the National Commissions’ recommendations, but they only showed the average 75-year impacts. CRS-11 after the bill was enacted). The discussion in Committee markups revolved around the average 75-year deficit and how much the various options would affect that figure. Hence, there was very little understanding that a period of surpluses would be followed by a period of deficits—or that “actuarial balance” was not achieved on a pay-as-you-go basis. If year-to-year or 5-year interval projections had been available, it might have been observed that surpluses of income over outgo would arise for about 35 years, followed by an indefinite period of deficits beginning in or around 2020 to 2025 (see “excess income/ or deficit” column in Table 6 below). It also would have been apparent that future projections of actuarial balance (i.e., the 75-year average) would deteriorate in future trustees’ reports if for no other reason than the addition of a deficit year at the back end of the projection period and the dropping of a surplus year at the front end. Table 6. Social Security Projections Made After Enactment of the Social Security Amendments of 1983 Calendar year Income Outgo Excess income/ or deficit (—-) Trust fund balance as % of annual expenditures (% of taxable payroll) 1983 11.24 11.49 (0.24) 15% 1985 11.58 11.33 0.25 21% 1990 12.71 11.27 1.44 38% 1997 12.79 10.36 2.42 117% 2000 12.78 10.08 2.71 234% 2010 12.82 10.31 2.51 491% 2020 12.95 12.76 0.19 538% 2030 13.08 14.73 (1.65) 437% 2040 13.14 15.17 (2.03) 314% 2050 13.16 15.27 (2.11) 192% 2060 13.17 15.44 (2.27) 54% Source: 1983 OASDI trustees’ report, intermediate II-B projections. In nominal dollars, the trust fund balance was shown to grow steadily to a peak of $20.7 trillion in 2045 and then decline to $6.8 trillion in 2060 (from Actuarial Note #117, loc. cit). Former SSA chief actuary, Robert J. Myers, makes the point that if the projections had been extended out further, they would have shown that the trust funds would have been exhausted just a few years beyond 2060). CRS-12 The Role of the National Commission on Social Security Reform The 1982/83 National While the 1982/83 National Commission on Social Security Reform is often cited as a model Commission on Social Security for lawmakers in achieving Reform played an instrumental role consensus about controversial in bringing various factions legislative matters. Although it together, and in the end, provided a played an instrumental role in bringing various factions together framework for congressional in 1982 and, in the end, provided action, it did not reach consensus a framework for congressional on a total solvency package. action in 1983, the Commission did not reach consensus on a total solvency package. For much of the period over which it met and deliberated (February 1982 to January 1983), it acted more as a fact-finding and “educational” forum in which the system’s financing problems and related issues were identified and debated. It was only in the final weeks, and by some accounts final days, of its existence that its members made significant progress in putting together a package of changes that a majority could agree on. Moreover, it is not clear that the Commission itself was the forum in which the solution evolved. Even as late as a week before the Commission was due to expire, there were press accounts of stalemate and great likelihood that the Commission was not going to be able to reach a consensus.14 By executive order the Commission’s report was due January 15, 1983.15 It may have been the actions of a few of the Commission’s members working with the White House that permitted the basic elements of a package to emerge. It was reported that members of the Reagan Administration played an instrumental role in those final days in working out an agreement with a small group of the Commission’s members.16 14 See, for example, Washington Post, “Social Security Panel Fails to Compromise,” January 9, 1983. 15 In the hope that a consensus plan could be reached with a little more time, on December 23, 1982 President Reagan extended the Commission’s reporting date by 15 days (making it January 15, 1983 rather than December 31, 1982 as required by his original executive order establishing the Commission). A consensus package was agreed to on January 15, 1983 and the President extended the Commission for another 5 days to provide time to put the report together. The report was transmitted to the President and congressional leaders on January 20, 1983. 16 By one account, Senator Robert Dole, Chairman of the Senate Finance Committee and a member of the Commission, stated that the Commission had come very close to failing to come up with a compromise. He stated that the Commission would not have arrived at a package without the active participation of President Reagan, through his intermediaries, and Speaker of the House, Tip O’Neill (from a transcript of The MacNeil-Lehrer Report, January (continued...) CRS-13 It also is not commonly understood that in its final report, the Commission was quite fragmented. While 12 of its 15 members supported a so-called consensus package, that package left one-third of the long-range deficit unresolved. That portion was left to the respective authorizing committees of Congress and ultimately was resolved only after a contentious floor debate. The largest faction proposing a complete package consisted of only eight members. Finally, in contrast to current circumstances, the Commission’s attention was drawn more to the system’s short-range condition than to its long-range problems, since the trust funds were expected to be depleted very quickly. It was only through enactment of a tax reallocation in 1980 and interfund borrowing from the HI portion of Medicare in late 1982 that the Social Security trust funds retained a balance throughout the period in which the Commission and Congress deliberated on the problem. Even with these measures, the trust funds were projected to be depleted by the following July (1983).17 In essence, the Commission and later the Congress were acting to forestall a crisis — imminent insolvency. This is in stark contrast to current circumstances, in which the Social Security trust funds are projected, under the intermediate estimates, to grow steadily until reaching a peak in 2018 or 2019 and would not become insolvent until 2029. Social Security Within the Larger Context of Federal Entitlements While today there is widespread recognition that looming demographic shifts early in the next century may significantly raise federal entitlement spending, this was much less of a concern in the development of the 1983 Social Security amendments. This is not to suggest that the demographics of a retiring baby boom generation was not understood. It was; there was considerable discussion of the future decline in the ratio of workers to Social Security recipients. However, the potential impact of the baby boomers on entitlement spending generally was not well understood and had little effect on the development of the amendments. As viewed in 1983, the projected strain on the Social Security system resulting from retirement of the baby boomers and a progressively aging society was certainly symbolic of the financial strain that retirement and other federal entitlement programs would undergo early in the next century.18 However, it was newly emerging as a concern in congressional deliberations. Long-range estimates of budget expenditures were for the most part non-existent. Even long-range Social Security estimates were viewed with substantial skepticism, given their duration. 16 (...continued) 17, 1983). Also see New York Times, “Gain is seen on 2-party accord on Social Security,” January 10, 1983. 17 The interfund borrowing provision was deliberately designed to assure adequate financing only through June 1983. 18 For example, see two articles by Peter G. Peterson appearing in the New York Review of Books: “Social Security: The Coming Crash” (Dec. 2, 1982) and “The Salvation of Social Security” (Dec. 16, 1982). CRS-14 The preparation of longrange Social Security estimates Where Social Security’s problems dates back to the beginning of the were viewed independently in program; 75-year projections 1983, today they are seen much date back to 1965.19 However, budget estimates by the more as a segment of the future Administration and CBO rarely strain entitlement spending transcended more than 5 years generally may create. Hence, to into the future. Medicare some extent the context for reform trustees’ estimates were routinely made for only 25-year periods, has shifted. and it was only about the time of the 1983 amendments that requests were made to take them out longer (the first trustees’ report to include 75year HI projections was the 1983 report, which came out after enactment of the amendments).20 The SSA and Health Care Financing Administration (HCFA) actuaries typically resisted making longer-range Medicare projections because of the greater uncertainty they saw with projections of health expenditures. But the failure to make more than 25-year projections obscured the potential demographic shift looming for the 2010-2030 period. Twenty-five year forecasts made in the period 1980 to 1983 ended in 2005 to 2008. The first cohort of baby boomers would reach age 65 in 2011. Long-range projections for Medicaid, as well as national health expenditures generally, also were non-existent. Some members of the National Commission voiced concern that by “ignoring the cost of the HI program, the potential tax burden of the entire Social Security program might not be properly assessed when making reforms...”21 Two members made the point that if the Medicare problem were taken into account, “the reserve needs of the system would be considerably larger” than what the Commission was assessing.22 19 According to Robert J. Myers (who assisted in the development of the original program estimates and was SSA’s chief actuary from 1947 to 1970), initially year-by-year figures were developed for a 44-year period (from 1937 through 1980), and it was assumed for valuation purposes that annual tax income and outgo remained constant thereafter. He states that the valuation period was indefinite in duration (i.e., it went into “perpetuity”). This procedure was used until 1965, at which point the valuation period was set at 75 years (from a conversation with Mr. Myers, July 22, 1997). 20 Estimates prepared in November 1982 by the staff of the National Commission on Social Security Reform (with assistance from the actuaries of the Health Care Financing Administration) were among the first to portray 75-year HI forecasts. These estimates were later arrayed in a Senate Finance Committee staff document to make the point that the problem was much larger than reflected by the discussion of Social Security. 21 22 Report of the National Commission on Social Security Reform, pps. 3-2 and 3-3. National Commission report, supplementary statement #5 by Senator Robert J. Dole and Representative Barber B. Conable, Jr. Also see supplementary statement #11 by former Representative Joe D. Waggonner, Jr. He wrote that “Demographics is the long-term problem... The baby-boom represents a tidal wave of future beneficiaries... While this Commission has not addressed the financing problems facing Medicare, I recommend that the (continued...) CRS-15 However, five members of the Commission objected to the preparation of 75-year HI estimates, stating that “the trustees consider that the degree of uncertainty concerning future hospital costs, relative to the remainder of the economy, is so great as to make projections beyond 25 years thoroughly misleading.”23 In its final report the Commission did not address HI’s problems, and specifically deferred attention to them to an upcoming Quadrennial Advisory Council on Social Security. Moreover, it had long been the practice of Congress to assess the financial condition and needs of Social Security in isolation. Separate trust funds for Social Security and Medicare had been in existence from the start of each program, and it was not until 1969 that Social Security and other federal trust funds were aggregated with other government program in a “unified” federal budget. In 1983, as was typical in the past, the focus was on resolving Social Security’s troubles. It was the possibility that its trust funds would be depleted in the short run and that the trustees’ were repeatedly projecting long-range actuarial imbalances that was most bothersome. The trustees’ annual warnings were affecting public opinion. The fact that long-range costs would progressively rise as the demographics changed was not really of primary concern. The pattern of income and outgo was certainly important in the short run, but the fact that long-run costs would rise even though the system was brought into “average” balance was not even observed (since there were no estimates of such until after the amendments were enacted). Today there is much more appreciation and concern about the impact that the demographic shifts looming early in the next century may have on entitlement programs as a whole. Medicare estimates are now routinely prepared for 75-year periods, and in recent years numerous estimates of the long-range cost of entitlements and federal expenditures overall have emerged. Congressional Budget Office projections made in 1996 suggest that the costs of Social Security, Medicare, and Medicaid alone could rise from 8% of GDP today to the 15%-16% range in 2025. Similar estimates are reflected in the 1997 Economic Report of the President.24 A 1994 commission appointed by President Clinton — the Bipartisan Commission on Entitlement and Tax Reform — spent nearly a year examining and debating the longterm costs of federal entitlement programs and ways to curb them, including limiting Social Security’s growth. And there are few congressional fiscal policy debates — 22 (...continued) policy implications of Medicare be reflected in OASDI legislation. The long-term deficit for the Hospital Insurance portion of Medicare is almost three times as large as the OASDI deficit... That deficit occurs despite massive cost shifts and despite assumptions that predict that health care costs will ultimately be controlled.” 23 National Commission report, supplementary statement #4 by Robert M. Ball, Martha Keys, Lane Kirkland, Senator Daniel Patrick Moynihan, and Representative Claude Pepper. 24 See 1997 Economic Report of the President, Chapter 3: Economic Challenges of an Aging Population, GPO, February, 1997 (See also the 1996 edition—Chapter 3: Making Fiscal Policy Choices Within and Across Generations); Long-term Budgetary Pressures and Policy Options, CBO, March 1997; The Economic and Budget Outlook: Fiscal Year 1997-2006, CBO, May 1996; Bipartisan Commission on Entitlement and Tax Reform, Interim and Final Reports, August 1994 and January 1995; and Eugene Steuerle and Jon M. Bakija, Retooling Social Security for the 21st Century, The Urban Institute Press, Wash., D.C., 1994 (p. 59). CRS-16 be they on annual budget resolutions or reconciliation bills, measures to raise the Treasury’s borrowing authority, or constitutional amendments to require a balanced federal budget — that fail to raise concerns about the present level of entitlement commitments made to the baby-boom generation and subsequent retirees. Where Social Security’s financing problems were viewed independently in 1983, today they are seen much more as a portion of the overall strain that entitlement spending may create. Hence, to some extent, the context for reform may have shifted. Conclusion With the completion of the work of the 1994-1996 Advisory Council on Social Security following on the heels of the 1994 Bipartisan Commission on Entitlement and Tax Reform, the prospects for reform of Social Security loom again. How significant those prospects are is uncertain. Few who have followed the stream of recent adverse trustees’ reports or polls showing public sentiment about the survival of the system have suggested that the status quo can be maintained. Although three different factions evolved within the Advisory Council, all proposed major changes to address Social Security’s long-range financial problems. None said “do nothing.” And while the Bipartisan Entitlement Commission was unable to reach a consensus on whether and how to constrain future entitlements, 24 of its 32 members concluded that “A bipartisan coalition of Congress, led by the President, must resolve the longterm imbalance between the government’s entitlement promises and the funds it will have available to pay for them.” They further concluded that balance needs to be restored to the Social Security trust funds and confidence in the system strengthened.25 Also apparent from the deliberations and reports of the advisory council and entitlement commission is that there are a multitude of options for altering Social Security that represent a wide range of ideological choices. This too was the situation in 1983. However, unlike the situation in 1983, there is no urgency today that might work to diminish ideological differences. Simply put, there is no immediate crisis that might help policymakers to narrow the list of options around which a consensus can be formed. As important as the range of options might be, how the problem and options are aired, debated, measured, and arrayed for congressional consideration are vital to achieving a viable solution. It may seem obvious now, but the year-to-year pattern of income and outgo that the options are projected to produce is more important in achieving a lasting restoration of the system than measurements of its “average” condition over 75 years. Moreover, the necessity of building a “cushion” against the possibility that today’s projections may prove overly optimistic has been demonstrated by repeated failures of past efforts to provide lasting remedies. Similarly obvious is that a consensus for reform is likely only if emanating from a bipartisan forum. Recognizing that political “windows of opportunity” may have been lacking, the recent advisory council and the entitlement commission may have 25 Bipartisan Commission, final report, pps. 1-3. CRS-17 been too large and unwieldy to achieve a consensus on reform. The 1983 commission, although reaching more of a consensus than these recent panels, also may have suffered by its size. Whether Congress should set up yet another independent panel, or simply work through the traditional committee process, is also a relevant question. There are numerous examples of major bipartisan efforts that have emanated directly from the legislative process. Whatever the forum, the principal question is whether, in the absence of a crisis, there is a will to consider reform. Two subtle forces are now stimulating the call for change: (1) persistent skepticism about the survival of the current system among baby boomers and younger segments of the population (driven in part by a rising battery of criticism that the system may not be a good deal for them), and (2) growing recognition that if the system’s long-range deficit is to be remedied, corrective action can be implemented in small marginal — and even deferred — steps if begun today. Whether these subtle forces are significant enough to trigger reform any time soon is a matter of conjecture