Abstract
This Selected Issues paper on the United States analyzes the measures of potential output, natural rate of unemployment, and capacity utilization. Traditionally, measures of resource utilization have been used as indicators for the potential build-up of inflation pressures, and hence as guides for the formulation of macroeconomic policy. The paper highlights that the most commonly used indicators of resource utilization in the United States are the output gap, the employment gap, and capacity utilization in industry. The paper also analyzes the wage and price determination and productivity trends in the United States.
VII. “Fixing” Social Security1
1. Without changes in the current structure of benefits and revenues, the Social Security system is expected to begin running large deficits in the next 15 years that will grow rapidly over the longer term.2 This long-term financial imbalance reflects the significant demographic shift as the baby-boom generation begins to retire in increasing numbers around 2010. Although the U.S. Social Security system faces a significant longer-term financial imbalance, a number of other industrial countries, particularly Germany, Italy, France, and Japan, are expected to face even larger imbalances as their dependency ratios are projected to grow more rapidly than that for the United States (Figure 1). The longer action is delayed in the United States, the greater the tax increase facing future generations, or the deeper real benefits cuts would have to be in order to restore the financial viability of the system. Recognizing the importance of prompt action, in its FY 2000 budget presented in February 1999 the Administration proposed a plan to significantly improve the long-term financial outlook of Social Security that retains the basic structure of the program, but introduces two significant changes in the way the system is financed. On June 28, 1999, with the release of the Mid-Session Review of the budget, the Administration modified the details of its proposal, but retained its core elements.
Selected Industrial Countries: Classic Dependency Ratio, 1950–2050
Citation: IMF Staff Country Reports 1999, 101; 10.5089/9781451839579.002.A007
Source: United Nations (1998).2. The Administration’s plan would transfer general revenues to the Social Security Trust Fund over the next 15 years and allow a certain portion of the Trust Fund’s assets to be invested in equities. These proposed measures raise some concerns. Opening the doors to general-revenue financing could compromise the effectiveness of a budget constraint that probably has helped to restrain increases in Social Security benefits over the years. There is also a possibility that reliance on general revenues might further loosen the perceived link between benefits received and payroll taxes paid, possibly exacerbating the tax distortion associated with Social Security funding. Moreover, the proposal to invest a small share of Trust Fund assets in equities raises the question whether an effective “firewall” can be built to insulate such investments from political influence.
A. The Administration’s Approach
3. Currently, Social Security benefits are financed principally through payroll taxes, but interest income on Social Security Trust Fund assets and income taxes on Social Security benefits also help fund the system. The Social Security payroll tax is paid by both the employer and employee (each paying 6.2 percent) on gross yearly wages up to a ceiling of $72,600 in 1999. The wage income ceiling is adjusted automatically each year based on the increase in the average wage for all workers. Although the current system does establish a direct linkage between the amount paid in over a worker’s lifetime and the benefits received, there is a significantly more favorable pay-back to low-wage workers, especially low-wage workers with families (i.e., progressivity in benefits).3 Because of the wage ceiling, however, the payroll tax itself is regressive.
4. The approach to restoring the long-term financial viability of the Social Security system outlined in the Administration’s FY 2000 Budget departs from past fixes in two substantive ways. First, general revenues would be used to bolster the funding of Social Security for the first time since the inception of the program. This is intended by the Administration to be a response to a one-time demographic episode; namely, the retirement of the baby-boom generation. Second, a small portion of Social Security Trust Fund assets would be invested in private equities. Trust Fund assets are currently restricted by law to be invested in so-called “special issues;” U.S. Treasury securities specifically designated for the Social Security Trust Fund.
5. The plan outlined in the FY 2000 Budget calls for transferring 62 percent of projected federal budget surpluses over the next 15 years ($2.8 trillion) to the Social Security Trust Fund. Of these amounts, one-fifth would be invested in private equities so as to improve the expected risk-adjusted return on the Trust Fund’s assets. The intention is not to allocate 62 percent of the yearly budget surpluses ex post as they accrue, but to legislate ex ante dollar amounts to be transferred to the Social Security Trust Fund on a yearly basis. According to Administration estimates, the annual transfers to the Trust Fund and the higher expected returns associated with the investment of Trust Fund assets in private securities would bring the Social Security system into actuarial balance over a 5 5-year horizon. This is an improvement over the current outlook, but falls short of the 75-year actuarial balance that is the norm for assessing the long-term financial viability of the system. The Administration estimates that the remaining funding gap under its proposal would be equivalent to a ¾ percentage point increase in the contribution rate.
6. The Administration modified the details of its proposal in the Mid-Session Review of the FY 2000 Budget, but retained the core elements (namely, equity investments and transfers from general revenues). The new proposal calls for preserving all of the projected Social Security surpluses, with each dollar of that surplus used to reduce federal government debt held by the public (the so-called “lockbox”). Moreover, the Administration’s new proposal also calls for transferring the full amount of the interest saved from this debt reduction to the Social Security Trust Fund beginning in 2011. Between 2011 and 2014, a transfer of $543 billion from general revenues to Social Security would occur. Thereafter, transfers would be $189 billion annually. To further improve the financial outlook for Social Security, the Administration would invest these transfers in equities until these equity investments reached a maximum of 15 percent of the total Trust Fund. According to the Administration’s estimates, this approach would achieve actuarial balance over a 53-year horizon.
7. Both plans would establish a framework for retiring a significant share of federal government debt held by the public. This strengthens the capacity of the federal government to meet fixture Social Security obligations by easing future debt-service obligations, thereby avoiding economically burdensome levels of future taxation and/or public debt. Nevertheless, the effectiveness of the plans will remain subject to the resolve of future Congresses to sustain unified federal government budget surpluses.4 In the absence of full Ricardian equivalence, the Administration’s proposals will help to increase national saving, as would any plan that targets sustained budget surpluses, and would thereby stimulate investment and growth.5
8. The fiscal effectiveness of these plans to a large extent hinges on their capacity to achieve public debt reduction so as to strengthen the government’s capacity to meet its future Social Security obligations. Since Social Security is entirely pay-as-you-go on a consolidated federal government basis (i.e., the federal government including all government agencies), the unified federal government has no net assets to help defray future Social Security obligations, and thus must achieve a low level of public indebtedness in order to position itself to absorb the rising tide of Social Security outlays as they occur. On the other hand, if private securities are also purchased by the Trust Fund, debt held by the public would be reduced more slowly than otherwise, but this would be offset by the accumulation of net assets to help defray future Social Security obligations.
B. An Economic Assessment
9. The Administration’s approach would preserve current taxation and benefit levels, and leave the fundamental structure of the system essentially intact.6 An alternative approach would be to combine a Social Security payroll tax increase, or an increase in the payroll tax ceiling, with benefit cuts. In view of the large fiscal surpluses under current policies, these measures would create the fiscal capacity to cut income taxes and/or increase government spending on non-Social Security items while improving the outlook for debt reduction.7 Given these alternatives, an economic assessment of the Administration’s approach hinges on the equity (horizontal and vertical) and efficiency implications of: (i) higher payroll taxes versus higher income taxes; (ii) maintaining Social Security benefits at current levels versus cutting benefits; and (iii) investing Trust Fund assets in equities versus maintaining current restrictions.
10. The Administration’s approach has implications for the distribution of income. By relying on general revenues, the Administration is in effect selecting progressive income taxes to (partially) close the financing gap. Summers (1999) has indicated that the Administration prefers not to raise the payroll tax for equity reasons; namely, because it would hit low- and middle-income workers proportionately harder than upper-income households. Compared with the alternatives of higher payroll taxes and/or reduced benefits, the Administration’s approach seeks to preserve the existing redistributional structure of the system. By relying on income taxes instead of the payroll tax, the plan may also achieve a degree of “tax smoothing,” since reliance on the payroll tax with its smaller tax base would require a greater adjustment in the tax rate to achieve comparable revenues. Tax smoothing and maintaining a higher degree of progressivity are desirable on efficiency and equity grounds, respectively. However, payroll taxes with a perceived fee-for-service link could be less distortionary than equivalent income taxes, which carry no such link,8 Moreover, since taxes tend to impose a deadweight loss on the economy, cutting benefits to improve the finances of the system instead of raising taxes would improve efficiency.
11. An alternative approach that might satisfy the Administration’s equity objectives without resorting to general revenues would be to remove the ceiling on the Social Security payroll tax. At the same time, income taxes would be cut to offset the increase in receipts from payroll taxes. Eliminating the ceiling on the payroll tax base clearly would directly affect only higher-wage households.9 All workers with gross yearly wages of at least $72,600 in 1999 would face a direct increase of 6.2 percentage points in the marginal tax rate on labor income. The labor market distortion associated with the existing payroll tax ceiling, which makes it proportionately more costly to hire lower-wage workers, would be removed. On balance, removing the ceiling would put downward pressure on market wages at the upper end of the wage scale that in the long run would approach 6.2 percent of the amount in excess of the ceiling.10 However, because the payroll tax excludes nonlabor income such as dividends, short-term capital gains, interest, or property income, lifting the payroll-tax ceiling would discriminate against higher-income households whose income is derived largely from labor, and would create an additional incentive to seek nonwage compensation. It would treat households with high levels of nonlabor income relatively more favorably than would the Administration’s reliance on the income tax. Moreover, since higher income households tend to have a higher share of nonlabor income than lower-income households, the payroll tax with the ceiling lifted would remain regressive relative to an income tax base. However, to the extent that existing income tax loopholes already enable higher-income households to avoid income taxes, these distributional implications would be mitigated.
12. Some have suggested that the Administration’s approach could “undermine fiscal discipline” since it would eliminate the direct link between the assessed viability of the system and projected payroll tax revenues.11 Providing access to general revenues could tend to loosen a long-term budget constraint that has helped to restrict the growth in Social Security benefits over the years.12 The Administration has taken the view that the proposed transfers from general revenues are designed to address a one-time demographic episode associated with the aging of the “baby-boom” generation. Although the original proposal limited transfers to a 15- year period, the revised approach calls for transfers over an indefinite period. In either case, once the precedent of transferring general revenues to Social Security has been set, a practice that would benefit an influential interest group that would be growing in size relative to the rest of the population (retirees), it may well prove difficult to restrict these transfers (whether in size or duration) as originally intended,
13. Investing a share of Trust Fund assets in equities has been criticized on the grounds that it would likely have little, or no, effect on national saving and would merely lead to a shift in the asset composition of public and private portfolios. Equity accumulation by the Social Security Trust Fund would result in the substitution of government bonds for equities in private portfolios as a whole. Under the current defined-benefit structure of Social Security, equity investments would help to relieve future taxpayers of the tax burden otherwise required to close the system’s financial shortfall. At the same time, future Social Security recipients (current taxpayers/savers) would retire with unchanged Social Security benefits and a lower expected stockpile of financial wealth.13
14. Political economy considerations are also raised by the proposal to invest Trust Fund assets in equities. Some have argued, including Greenspan (1999), that it may be difficult to insulate investment decisions from political considerations. To address this concern, the Administration has proposed an institutional framework similar to the existing Federal Retirement Thrift Investment Board, whereby investment decisions would be made by an apolitical, independent investment board comprised of private-sector investment managers selected through a competitive bidding process. Furthermore, the Administration has proposed that investments would be limited to broad-based, widely used index funds, eliminating the possibility of individual stock picking. Moreover, the share of Trust Fund assets that would be invested in equities would be relatively small.
15. A further political-influence consideration may arise once the Social Security system has accumulated private equities and must eventually decide on how to draw Trust Fund resources to finance expenditures. When the system reaches a stage of negative cash flow, a choice will have to be made on the balance between net selling of the Trust Fund’s special issues and net equity sales. Since this decision could have direct implications for stock and bond markets and would also affect changes in federal government debt held by the public—a transparent and politically sensitive variable—political considerations again may come into play. To avoid this situation, Congress could legislate a rule specifying how much of each type of asset would be sold, well in advance of the date when the Trust Fund’s assets would need to be liquidated.
List of References
Aaron, Henry, 1998, “Social Security: Tune It Up, Don’t Trade It Li,” Brookings Institution (May).
Auerbach, Alan, and Kotlikoff, Laurence 1985, “The Efficiency Gains From Social Security Benefits” National Bureau of Economic Research, Working Paper No. 1645.
Ball, Robert, 1999, “The Future of Social Security for This Generation and the Next: Examining Proposals Regarding Personal Accounts,” Testimony before the Subcommittee on Social Security, Committee on Ways and Means.
Congressional Budget Office, 1999, An Analysis of the President’s Budgetary Proposals for Fiscal Year 2000, (April).
Gramlich, Edward M, 1999, “The Budget Surplus and Social Security,” Testimony of Governor Edward M. Gramlich before the Committee on Finance, U.S. Senate (February 9).
Greenspan, Alan, 1999, “On Investing the Social Security Trust Fund in Equities,” Testimony of Chairman Alan Greenspan before the Subcommittee on Finance and Hazardous Materials, Committee on Commerce, U.S. House of Representatives (March 3).
Hogan, Vincent, and Tokarick, Stephen 1998, “Alternative Approaches to Social Security Reform,” International Monetary Fund, SM/98/188.
Leidy, Michael P., 1997, “Investing U.S. Social Security Trust Fund Assets in Private Securities,” IMF Working Paper (WP/97/112) (September).
Marchildon, Lori, Sargent, Timothy and Ruggeri, Joe 1996, “An Economic Analysis of Payroll Taxes,” Department of Finance Canada, Economic Studies and Policy Analysis Division.
Nichols, Orlo R., 1994, “Do People get Their Money’s Worth From Social Security?”, OASIS Magazine Ask the Actuary, Office of the Chief Actuary, U.S. Social Security Administration (December).
Summers, Lawrence K, 1999. Testimony of the Deputy Treasury Secretary Lawrence H. Summers, Senate Special Aging Committee, (March 1).
United Nations, 1998, “World Population Prospects: 1998 Revision,” New York: Population Division of the Department of Economic and Social Affairs of the United Nations Secretariat.
Prepared by Michael Leidy and Stephen Tokarick.
The 1999 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds estimates that without changes in benefits or revenues, the cash flow of the system will shift to a deficit in 2014 and the assets of the system will be exhausted in 2034. The “actuarial balance” of the system compares the expected present value of all revenues to the expected present value of outlays over a given time horizon. This calculation also requires that at the end of the period (e.g., 75 years) the Trust Fund balance equals one year of projected outlays. The 1999 Actuarial Report indicates that an increase of 2.07 percentage points in the payroll tax would be needed now to eliminate the 75-year actuarial deficit of the Social Security program.
Nichols (1994) provides summary statistics on the progressivity of expected Social Security benefits across income groups. For example, a low-income married male with a family could expect, on average, to pay about $l91/2 thousand in Social Security payroll taxes over a 40-year work life, and would receive about twice that amount (all dollar amounts are 1994 present values) in retirement benefits. A comparable upper-income male would pay about $104 thousand in payroll taxes and receive about the same amount in retirement benefits.
Under budget accounting rules that have been in place since the 1960s, only those transfers used to purchase equities would result in lowering the reported unified budget deficit. The reported on- and off-budget surpluses would, however, be affected.
It is noteworthy, however, that the federal government’s budget outlook under current policies (projected unified budget surpluses) is better than under the Administration’s budget proposals with Social Security measures. The CBO (1999) estimates that debt held by the public under the Administration’s budget with Social Security proposals (correcting for Social Security assets held in private equities) would exceed that under current policies by more than $700 billion in 2009.
Plans that include more radical overhauls of the existing system are not considered in this paper. Hogan and Tokarick (1998) discuss three broad approaches to reforming the Social Security System, including altering the parameters of the current system, moving to fuller funding, and privatization.
A number of alternative reform plans include some combination of payroll-tax increases or benefit reductions, including the plan put forth by Robert Ball (1999). Changes to payroll taxes and benefits also constitutes one of the three approaches to reform contained in the 1996 Advisory Council Report on Social Security.
Assuming that maximum benefit levels were left unchanged, the Office of the Chief Actuary of the Social Security Administration has estimated that removing the ceiling on gross wages subject to the payroll tax in 2000 would be equivalent to an increase in the tax rate of 2.02 percentage points, essentially eliminating the current 75-year financing gap. If benefits were adjusted in line with current policies, lifting the ceiling on taxable payroll would be equivalent to an increase in the tax rate of L53 percentage points.
This assumes that the long-run elasticity of supply of higher-income workers is nearly zero and that labor markets are competitive. If, instead, the long-run aggregate labor-supply curve for upper-wage labor were somewhat elastic, lifting the ceiling would put less downward pressure on wages and would tend to reduce the equilibrium level of employment.
Gramlich (1999) and CBO (1999).
Perhaps to help minimize this possible problem, the Administration’s proposal also calls for changing budget accounting rules so that transfers to the Social Security Trust Fund would reduce the reported budget surplus.
Leidy (1997) examines the macroeconomic and intergenerational effects of investing Social Security Trust Fund assets in equities.
