Rx for Social Security: The First Steps

Report Social Security

Rx for Social Security: The First Steps

December 14, 1981 36 min read Download Report
Peter G.
Senior Fellow and Director of Government Finance Programs

(Archived document, may contain errors)

1 159 December 14, 1981 Rx FOR SOCIAL' SECURITK I I THE FIRST STEPS INTRODUCTION The Social Security system has enjoyed overwhelming public support since its inception and is frequently called the single mos t successful income security program in American history.

However, as the system's deficiencies become more apparent public confidence in it is being rapidly eroded. wide poll conducted by Sindlinger and Company, Inc for The Heritage Foundation revealed t hat 84 percent of those surveyed expressed little or no confidence in the financial soundness of Social Security. Several members of the Administration have testified that unless immediate action is taken the most devastat ing bankruptcy in history could occur on or about November 3 19

82. This could be a disaster for the many people dependent on current or expected future benefits covers 90 percent of the labor force and provides income support to 36 million elderly or disabled Americans and their dependents.

Although Congress has repeatedly insisted that it would never allow the Social Security system to fail, it is becoming increas ingly clear that.unless the practice of providing excessively generous benefits is reversed, the resulting financial burden may lead to the system's collapse flicting objectives to be both an insurance program and a welfare program. Even though constant constituent pressure to raise benefits, whether earned or unearned, has effectively destroyed the link between contributions and benefits, there are still many who complain that their benefits are inadequate. The fact is that most current retirees are earning an extremely high return on their "investment," but economic.realities and demogra- phic shifts will make this all but i m possible in years to come. Stanford University economist Michael Boskin predicts that "if we wait until the baby-boom generation.retires before we begin to A recent nation The proqram currently Social Security's underlying problems result from its con 2 d e al with the tremendous long-term deficit in Social Security, we will see the greatest tax revolt and age warfare in the history of the United States.lI1 In short, the government has promised benefits that will be difficult to provide under projected condi - tions. The only way to avoid the cataclysm described by Boskin is to examine the system thoroughly and then to reform it.

President Reagan's appeal for immediate action in dealing with the problems facing the economy is equally applicable to Social Secur ity Can anyone here say that if we can't do it, someone down the road can do it? And if no one does it, what happens to the country would face an eventual collapse a challenge, but ask yourselves: If not us, who? If not now, when All of us here know the e c onomy I know it's a hell of BACKGROUND The Social Security Act was signed into law on August 14, 1935, and established a federal Old Age Insurance (OAI) system designed to provide workers with monthly benefits upon retirement at age 65 program to replace a portion of workers' earnings lost as a result of retirement. Participants were urged not to rely solely on Social Security for old age support, but to supplement their retirement incomes from other sources. President Roosevelt alluded to this point in hi s message to Congress on June 8, 1934 Social Security originally was intended as an insurance Ample scope is left for the exercise of private initia tive. In fact, in the process of recovery, 1 am greatly hoping that repeated promises of private investment and private initiative to relieve the government in the immediate future of much of the burden it has assumed will be f~lfilled Similar sentiments were expressed by members of Congress report of the House Ways and Means Committee on April 5, 1935 describe d the purpose of Social Security The While humanely providing for those in distress Social Security] does not proceed upon the destructive theory that the citizens should look to the government for everything. On the contrary, it seeks to reduce dependency and to encourage thrift and self-support.4 1 "The Crisis in Social Security," Newsweek, June 1, 1981, p 25. Charles Alexander Making it Work," Time, September 21, 1981, p. 38.

Cited in Peter J. Ferrara, Social Security: The Inherent Contradiction (San Fra ncisco, California: CAT0 Institute, 1980 p 18. Cited Ibid p. 19. 3 In sum, the basic goal of Social Security was to provide safeguards against the worst misfortunes and vicissitudes of life, but not to provide the sole means of support for the elderly.

Se curity payroll tax, which is assessed against a portion of earned income. This tax was to be shared equally by employee and employer alike, although economists generally agree that the employee bears the full burden of the tax.5 pay a flat tax rate that h a s ranged from two-thirds to three- fourths of the combined employer-employee tax. In 1937, the combined tax rate for employers and employees was 2 percent levied against the first $3,000 of earned income such benefits because it captures the guid pro quo n ature of an insurance program. As a result, people have come to view their contributions to Social Security as premiums. on insurance, giving them title to annuities from the government in their old age. Over the years, the program has extended protection to cover its participants and their dependents from costs associated with disability, hospitalization, and death.6 At the same time, Social Security taxes have soared. (See Table 1.) The inexorable growth of Social Security taxes has imposed an increasing l y severe burden on workers. In the early years, these taxes grew relatively slowly. In 1937, the maximum tax payable was only $60, which remained constant until 1950 when it w.as raised to 90. Since then, the maximum has grown at a steady and ever more ra pid rate, increasing an astounding 900 percent between 1965 and 19

81. Inclusion of hospital insurance raises the total to $3,950.10 today. This extraordinary increase can be attributed primarily to the program's departure from strict insurance principles and the inclusion of a sizable unearned component in the payment o f benefits.

The future costs of financing Social Security appear even more disturbing. Table 2 projects tax rates of the Social Security system if benefit payments and financing methods are not altered.

By the year 2030, Social Security payroll taxes appr oach one quarter of taxable payroll. This number could surpass 40 percent if more pessimistic but actuarially possible assumptions are accurate The program has been financed through a special Social The self-employed A payroll tax is the appropriate mecha nism for financing This point will be discussed in more detail later.

Brittain, "The Incidence of Social Security Payroll Taxes," American Economic Review LXI (March 1971), pp. 110-125 and Ferrara, op. cit pp. 405-412.

Benefits were added for survivors of deceased workers in 1939, at which time the program became known as the Old-Age Survivors Insurance (OASI program. Disability insurance (DI) was added in 1957; hospital insurance (HI) in 19

66. Each has its own trust fund and is financed by its own tax, although all three are assessed together as if one tax tax Also, see John A the OASDHI 4 Table 1 Tax Rates percent of taxable payroll) Maximum Tax Maximum Employee and for employees and Years Taxable Income Employer Combined Self-employed employers combin e d 1937-49 1950 1951-53 1954 1955-56 1957-58 1959 1960-61 1962 1963-65 1966 1967 1968 1969-70 1971 1972 1973 1974 1975 1976 1977 19 78 1979 1980 1981 3,000 3,000 3,600 4,200 4,200 4,200 4,800 4,800 4,800 4,800 6,600 6,600 7,800 7,800 7,800 9,000 10,800 13, 2 00 14,100 15,300 16,500 17,700 22,900 25,900 29,700 2.00 3.00 3.00 4.00 4.00 4.50 5.00 6.00 6.25 7.25 8.40 8.80 8.80 9.60 10.40 10.40 11.70 11.70 11.70 11.70 11.70 12.10 12.26 12.26 13.30 2.25 3.00 3.00 3.375 3.75 4.50 4.70 5.40 6.15 6.40 6.40 6.90 7.50 7 . 50 8.00 7.90 7.90 7.90 7.90 8.10 8.10 8.10 9.30 60.00 90.00 108.00 144.00 168.00 189.00 240.00 288.00 300.00 348.00 554.40 580.80 686.40 748.80 811.20 936.00 1,263.60 1,544.40 1,649.70 1,790.10 1,930.50 2,141.70 2,807.54 3,175.34 3,950.10 Since its incept i on, Social Security has also fulfilled a social adequacy function, paying some individuals benefits solely on the basis of need, whether or not they have paid for these benefits through their taxes. In fact, many of the benefits provided by Social Securit y are completely unrelated to a worker's contributions and are largely responsible for the emasculated condition of the Social Security trust funds. The Social Security system is an inappropriate vehicle for achieving these putative welfare objectives beca u se it is financed by a regressive payroll tax.7 While such a tax is suitable for the insurance goal of Social Security, there can be no reasonable justification for providing welfare benefits by a tax that places its .heaviest burden on the very group it i s designed to help. increasing instability of the Social Security system is linked to In short, the This tax is considered regressive because it is levied proportionately up to the ceiling, at which point the tax rate becomes zero and thus takes a greater proportion of income at the lower end increased when unearned income is included as part of total income.

The regressivity is 5 Table 2 Projected Tax Rates Necessary to Finance Present Social Security Program Tax Rates percent of taxable payroll Calendar Employee and Year Employer Combined Self-employed 1982-84 1985 19 86- 89 1990 2000 2010 2020 2030 2040 2050 13.40 14.10 14.30 15.30 15.60 17.20 21.10 24.60 24.70 24.40 9.35 9.90 10.00 10.75 10.75 11.50 14.10 16.40 16.40 16.20 NOTE: Figures from 1980 throu g h 1990 are the tax rates scheduled in present law. The figures for the year 2000 and later represent the tax rates necessary, based on the intermediate assumptions in the the 1979 Trustees Reports, to finance benefits and administrative expenses assuming n o change is made in present law. This does not include the taxes necessary to support the Supplementary Medical Insurance program, which is not I financed by payroll taxes. I SOURCE: Adapted from A. Haeworth Robertson, The Coming Revolution in Social I Se curity (McLean, Virginia: Security Press, Inc., 1981 p. 63 its two conflicting goals: individual equity and social adequacy.

The key to reforming Social Security thus is to separate the transfer and annuity functions and finance them through general revenu es and payroll taxes respectively. The transfer component then could be completely needs-oriented this Rubicon will Social Security resemble the retirement program it was intended to be Only after crossing THE PROBLEM OF INEQUITY Social Security suffers f rom serious inequities, which occur between people within the same generation as well as between those of different generations.

Social Security's quixotic social adequacy function, which often pays benefits regardless of whether or not they are earned. Th ese transfers are advocated in the belief that recipients need financial assistance. However, no proof of need is ever required because it is determined by surrogate measures supposedly reflect- ing need, such as age, family size, and earnings in employme n t covered by Social Security. Distributing benefits in such a These inequities result from 6 manner is very inefficient and often leads to payment of consider able unearned benefits to people who would not qualify as needy under most definitions make it b o th a poor tool of income redistribution and an unsound insurance program Social Security's contradictory goals Benefit Formula A worker's benefit is determined on the basis of his earnings record in covered employment. Once a worker's average indexed mont h ly earnings (AIME) are computed, a primary insurance amount PIA which is the worker's benefit, is found by applying the AIME to a special formula In 1981, this formula was 90 percent of the first $211 of AIME, plus 32 percent of the next $1,063 of AIME, p lus 15 percent of all AIME over $1,274.8 This formula is adjusted annually by the rate of increase in wages each month benefits are received before the age of 65 and is raised by one-fourth of 1 percent for each month receipt is delayed.

The benefit schedule is clearly weighted to favor lower income classes. This bias is evident in in Table 3, which compares benefits of two workers with differing earnings, one with an AIME of $1,100 and the other with an AIME of $2

20. Although the former worker paid about five times as much in taxes, his benefit is only about two-and-a-half times as large. The benefit-to-taxes ratio falls still further as a recipient's AIME crosses the 1,274 threshold and is pushed from the 32 percent benefit bracket to the 15 percent bra cket.

Other factors, however, tend to work against the poor. They are more likely to pay payroll taxes over a longer period than the rich, who can delay their entry into the labor force by extending their schooling. The typically shorter life span of lower -income groups, moreover, means that they do not collect benefits for as long as the rich do. Finally, the tax-exempt status of Social Security benefits is a more valuable feature to wealthy beneficiaries in higher marginal tax brackets A person's PIA is r educed by five-ninths of 1 percent for To ensure individual equity, benefit payments should be closely related to past contributions proportional benefit structure would be a more equitable approach to disbursing benefits Adopting a strictly This formula i s used for those reaching the age of 62 after 1978 those attaining age 62 prior to 1979, a more munificent benefit structure is used to determine benefits time from 1979 through 1983 have a choice of either formula for calculat ing retirement benefits For In addition, those becoming 62 at any 7 Table 3 Comparison of the Primary Insurance Amount of Workers With Different AIMEsg AIME of $220 AIME of $1,100 90 percent of $211 $189.90 $189.90 32 percent of 9 2.88 32 percent of $889 284.48 Primary Insurance Amo u nt $192.78 $474.38 Retirement Test There is much controversy regarding Social Security's so called earnings, or retirement, test, which limits benefit payments to otherwise eligible Social Security recipients earning more than a specified amount. In 1981, those aged 65 to 71 will lose 1 for every 2 earned in excess of $5,500 The amount exempt for beneficiaries under 65 is $4,080 perceived as unfair by more than 65 percent of those surveyed in the Heritage Foundation poll. However, the penalty is justified b y proponents of the test for several reasons. First, because Social Security was originally intended to partially replace earnings lost as a result of retirement, individuals continuing to work have earnings that remove the need for Social Security benefi t s. Second, by inducing the elderly to retire, problems with unemployment are relieved somewhat by making more jobs available for younger workers. Finally, members of the National Commission on Social Security argue in a recent report that repealing the ea r nings test would cause intergenerational inequities by allowing higher-earning workers to remain employed while receiving full benefits and compelling their younger lower-paid co-workers to finance these benefits through their contributions to Social Secu r ity. The real motivation for hanging onto the retirement test, however, is the cost of paying increased benefits. These may be difficult to finance in light of the impending crisis in Social Security.ll This limitation was This is an updated version of an example in J. W. Van Gorkom, Social Secur ity Revisited (Washington, D.C American Enterprise Institute, 1979 p. 14.

National Commission on Social Security, Social Security in America's Future Washington, D.C March 1981, p. 150.

The first-year cost would be about 6 to 7 billion dollars and more in later years. For those aged 65 and older, however, the cost in the first year is only about 2 billion dollars lo l1a Though there are reasons for the earnings test, it is one of the mor e inefficient and inequitable provisions of Social Security. Supporters of the earnings test contend that by continuing to work beyond the age of 65, the elderly restrict the job market for younger workers. This view is an example of the lllump of labor fa l lacy,Il which falsely holds that there are only a given number of jobs available in the economy and that one person's gain is another person's loss set limit effectively raises marginal tax rates for older workers. When this reduction is combined with the income and Social Securi ty payroll taxes, which they also pay, these workers become one of the nation's most heavily taxed groups. As a result, many of the elderly who otherwise would have continued working are involun tarily forced into retirement. peop l e to provide more for themselves, the earnings test makes it difficult, if not impossible, for them to supplement their income in many cases, it may actually lower their standard of living. Alicia MUMell, vice president and economist of the Federal Reserv e Bank of Boston, has accurately described some of the pernicious effects of the retirement test The 50 percent benefit reduction rate on earnings above a Rather than encouraging older In sum, there is good reason to be concerned about the provision of Soc i al Security law that discourages labor force participation of people over sixty-two who prefer to continue working. sources, such a deterrent reduces the welfare of the elderly low-income people, who seldom have other sources of retirement income such as p rivate insurance, pension benefits, or savings. In addition, any provision that encourages a smaller labor force in 'future years will force a significantly higher tax.rate in the long run.12 By limiting available income The burden falls particularly heav ily on Moreover, this constraint on earnings misallocates resources and lowers potential output by distorting the labor-leisure decision of older people. This loss is particularly egregious given the experience and knowledge the elderly.have to offer.

The greatest inequity of the earnings test is that it denies the elderly benefits they have paid for during their working years. This problem again arises from a confusion of insurance and welfare objectives. Paul H. Douglas, an economist who helped draft the amendments to the Social Security Act and who later became a U.S. senator from Illinois, elaborated on this problem l2 Alicia Munnell, The Future of Social Security (Washington, D.C Brookings Institution, 1977 p. 82.

The 9 This provision, however, is in p art a confusion of the idea of relief with that of insurance. The workers will have made direct contributions for half of their annuities and indirectly will have paid for most of the employers' contributions as well. When the system is thoroughly establi s hed, they will therefore have earned their annuities employment is, in reality, attaching a condition upon insurance which they have themselves bought.13 To require them to give up gainful Peter Ferrara compares this to a bank withholding cash from a cust o mer because it felt that the customer did not need the money.14 Is it fair for the government to coerce people into Social Security and then deny them benefits if they choose to work beyond a certain age? after age 72 regardless of whether the recipient h as substantial earnings or not. This inconsistency is a direct antithesis to the rationale advanced by advocates of the retirement test.

The earnings test limits benefits on the basis of earned income, yet allows full benefits even if there is enormous inc ome from other sources such as dividends, interest, capital gains, or rents. This also discriminates against low-income workers since they rely pr.imarily on wage income, while the wealthy are able to supplement their retirement incomes through various in vestments.

What justification is there, moreover, for a distinction between earned and unearned income? Both can be viewed as a return on an investment. Marshall Colberg, professor of economics at Florida State University, writes Investment in the individu al is now seen to be similar in many ways to investment in material resources.

Formal education, vocational training, on-the-job training, and even expenditures on health, on migration and on searching for information about prices and incomes have been vi ewed as investments in human capi- tal Interest earnings on material and human capital are consequently not inherently unlike In addition, full benefits are received For many persons and under many conditions during their lifetime, material resources and h uman resources are practical alternatives for investment. Yet interest from following the former course is considered by law to be "unearned incomeif while equivalent interest derived from the latter course is called "earned income' I Interest on this for m of capital encounters the problem of the Social Security earnings test while interest on material capital escapes the test.15 l3 l4 l5 Marshall. Colberg, The Social Security Retirement Test Washington, D.C American Enterprise Institute, 1978 pp. 2-3.

Ferrara, op. cit., p. 244.

Colberg, op. cit., pp. 23-25 Right or Wrong 10 A more productive economy could be achieved by eliminating this restraint on the elderly.

Two further inequities associated with the retirement test should be mentioned. First, distor tions arise from differences in the cost of living across the nation is in current, rather than real dollars, beneficiaries in some areas are unfairly more restricted than in others. Second, Social Security's treatment of earned income differs from that o f many other retirement plans, which allow workers to take other employment without sacrificing their annuities the earnings test to income and Social Security payroll tax collections; costs to the Social Security Administration would decline by reducing t h e administrative burden of enforcing the test; and perhaps most compelling of all, an unfair and costly onus would be lifted from the aged Because the limitation Finally, there are several immediate benefits from expunging increased labor force participat i on would add Spouse's Benefit16 The spousels benefit, added to Social Security in 1939 consists of two parts: a retirement benefit, which awards the wife of a retired worker 50 percent of her husband's benefit; and a survivorls benefit, which grants a wid o w 100 percent of her husband's benefit. These annuities become available in full after participants reach age 65, although reduced benefits are provided under each part for women who are 62 and 60 years of age respectively. In essence, this measure is an u nearned benefit awarded on the premise that a worker with a wife is more needy than a single worker. This provision distorts the nexus between benefits and contributions while adding considerably to the system's cost ties. First, married workers receive g r eater protection than single workers under Social Security A married worker is entitled to receive 50 percent more in benefits than a single worker contributing an equal amount in taxes. subsidize their married counterparts is clearly a violation of deter m ining payments based on insurance principles As a result, the spousels benefit creates a number of inequi Forcing single workers to Second, there are inequities associated with the unequal treatment of working and non-working wives find that the protectio n they receive based on their earnings record adds little or nothing to the coverage they already have Married women may l6 Similar benefits are available for children, grandchildren, parents and divorced wives For simplic.ity, the discussion here is restr icted only to the spouse's benefit The size of the benefit depends on the worker's PIA. 11 from the spouse's benefit her to a benefit that is equal to or less than one-half of her husband's, she will get no return on her contributions at all.

If she qualif ies for more than the amount available from the spouse's benefit, her net gain is only the difference between her benefit and the benefit she could have received based on her husband's earnings record It is unfair, moreover, to have working wives, who ten d to come from lower-income households, subsidize benefit payments for non-working wives, who are more likely to come from higher-income households If a wife's earnings record entitles Finally, a family with two earners may receive less in benefits than a one-earner family with the same total family earnings. For example: an equal share of the family income, while in Family B the husband is the sole supporter. equivalent yearly earnings and that all four individuals reach age 65 at the beginning of 19

81. W ith AIMEs of $400, both husband and wife in Family A receive an annual benefit of $250.30 for a combined payment of $500.76. has worked, both husband and wife receive a greater benefit his AIME entitles them to $378.38 per month plus 50 percent, or $189.1 9, for his spouse, for a total monthly retirement benefit of $567.

57. Family B will receive an annual benefit $801.72 greater than Family A even though both families have presumably paid an equivalent amount in taxes anomalies in the current benefit struc ture. This benefit was created on the basis of the traditional family model of the 1930s when it appeared that most women would marry and that few would then participate actively in the labor force. This no longer holds today. Family structures and women' s work patterns have changed tremendously since 1939, dramatically reducing women's dependence on their husbands for instance, has nearly tripled since 1940, yet the spouse's benefit remains and continues to discriminate against them in Family A both husba n d and wife earn Assume that the two families have had In family B, where only the husband Eliminating the spousels benefit would reduce some of the The number of married women working One possible remedy is to treat each household as a single economic uni t. For a married couple, the benefit earned by either would be divided equally among both large welfare element from Social Security.

This would remove a Interqenerational Transfers An inequity often overlooked in Social Security is the inherent subsidization of an older generation by a younger one.

This has been defended on the grounds that elderly recipients are generally in need. Indeed, a common justification for Social Security is that it transfers money from a younger, wealthier generation of workers to an older and poorer generation of retiree s . As a consequence, young, unskilled workers, struggling to make ends meet may be compelled to subsidize retired doctors, lawyers, and others who may be far better off than the workers themselves. 12 Just how this transfer works is clear from the case of a n individual with average earnings 62 years old .in 1937, who retired at age 65 in 1940.17 contributions at the .interest rates then prevailing, he would have accumu1ated.a retirement fund of 68.36 by 1940, yielding him $6.59 per year. From the standpoint of social adequacy, however, such benefits were considered too meager. The average annual Social Security benefit actually awarded in 1940 to a 65-year-old male was 270.60 yielding the beneficiary a $264.01 windfall. But since these benefits were subject t o periodic changes, it is more useful to compare capitalized savings and benefits. The present value of lifetime Social Security benefits for this individual would have been $2,962.09, resulting in a pure transfer of $2,893.73 or 97.7 percent of benefits received.

However, the relative size of this unearned component has been falling over time as the system has been maturing.

A factor contributing to a significant reduction in the welfare component is the longer period over which more.recent retirees have paid taxes. A retiree in 1960, for instance, could only have contributed to Social Security for a maximum of twenty- three years, whereas a retiree in 1981 may have participated for as many as forty-four years mature, these windfalls from short earnings h istories in covered employment will be smaller. Moreover, the percentage of payroll taxes has steadily grown, lowering the return on Social Security still further If he invested his Social Security Now that the system is nearly These transfers were initia l ly accepted because a large pool of workers supported a relatively small recipient group. In 1940, for example, the ratio of covered workers to retirees was 300:l. This ratio radically declined in subsequent years: 16:l in 1950; 4:l in 1965; and 3.2:l tod a y. By the year 2030, this ratio may fall to 2:1, or still lower if the projected demographic trends prove to be too optimistic ratio continues to fall, the burden on future generations will become increasingly greater As the worker to retiree Michael Bosk i n calculated the return on Social Security contributions that workers of'various age groups in 1977 can expect to receive under current law, including tax increases already legislated but not yet implemented (see Table 4). He computed the amount each age c ohort has paid (or will pay) into the OASI system, and then compounded each year's contributions by an interest rate will collect over a lifetime. The return on contributions for most workers appears to be quite good. The average current He then estimated the benefits each age group l7 This example is taken from D. Parsons and D. Munro, "Intergenerational Transfers in Social Security," in Michael Boskin, ed., The Crisis in Social Security: Problems and Prospects, (San Francisco: Institute for Contem porary Studies, 1977). 13 Table 4 Old Age and Survivors Insurance: Relationship of Taxes Paid to Benefits Received (1977 Dollars 65+ 64-55 54-45 44-35 34-25 Under 25 Average Tax Family 7,058 18,345 33,883 53,326 73,843 Average Benefit/ Family 49,400 47,639 55,60 0 66,321 73,577 Avg. Net Benefit Family 42,343 29,294 22,718 12,994 -267 large, negative Avg. Net Benefit as Tax/Family 600.0 160.0 67.0 24.4 36 Total Taxes pd by Cohort (billions) 172 235 349 389 540 552 Total Benefits pd to Cohort (billions) 1,282 629 57 0 473 503 Transfers as of Total Benefdts 86.2 62.7 38.8 19.4 -7.39 large, negative SOURCE: Michael J. Boskin, John B. Shoven, Marcy Avrin, and Kenneth Cone Separating the Transfer and Annuity Functions of Social Security,"

Department of Economics, Stanford University and National Bureau of Economic Research, p. 28 retiree receives a net transfer of about $42,000 per family, an amount six times larger than what he paid in plus interest. The size of this return, however, will steadily decline for those retir i ng in the future. An average family of the 55-64 age cohort, for example will receive a transfer amounting to only 1.6 times what it paid in plus interest. Workers in the 25-34 age cohort will be the first group as a whole to get a negative return on its S ocial Security contributions, albeit a relatively small loss times as high as that paid by those currently retired, yet their benefits are only one-and-a-half times as great. For those under the age of 25, the loss will be so great that Boskin simply list s it as lllargell and I1negative Indeed the average tax paid by this group is ten The threat of the Social Security system going bankrupt because of changing economic conditions and demographic shifts could have been avoided by a fully-funded system. Provi d ing overly generous benefits during the start-up phase of Social 14 Security, however, resulted in a pay-as-you-go system. Contribu tions to the program are not saved and invested, but used to pay benefits to today's recipients. The consequences .of this p olicy are now becoming apparent as the tax burden worsens and many more people begin getting lower returns on Social Security contributions than they otherwise would have been able to obtain in the market It would have been far more efficient and equitabl e if the first generation of retirees had been taxed more than 2 percent of their earnings during their working years, and their benefit payments reduced to more accurately reflect their past contribu tions in the program's early years it would have equali z ed the return across generations. Those deemed to be in need could have been more appropriately cared for through means-tested welfare programs Nevertheless, the first cohorts of retirees were allowed to collect benefits that were far in excess of anythin g "justified by their brief taxpaying experience. Ill Though this may have resulted in very low benefit payments EFFICIENCY CONSIDERATIONS Savings and Capital Formation One harmful side effect of Social Security is the reduction in savings that results fro m the pay-as-you-go nature of its financing. As a result, funds available for capital formation are reduced, thereby depressing economic growth and national output.

Social Security payroll taxes lower disposable income, leaving individuals with less money to allocate for o'ther purposes Social Security reduces private savings in two ways First l8 l9 John A. Brittain, The Payroll Tax for Social Security (Washington, D.C The Brookings Institution, 1972 p 9 There is still considerable debate on the nature and extent of the effect of Social Security on private saving and its corresponding impact on capital formation.

The Impact of Social Security on Private Saving Series, with a reply by Martin Feldstein (Washington, D.C Enterprise Institute, 1978 Michael Darby , The Effects of Social Security on Income and Capital Stock (Washington, D.C Institute 1979 Martin Feldstein Social Security, Induced Retirement and Aggregate Capital Accumulation Journal of Political Economy, Vol 82 (Sept.-Oct. 1974 pp. 905-926; Martin F eldstein Social Security Induced Retirement, and Capital Accumulation National Bureau of Economic Research, Working Paper 579, November 1980 and Alicia H. Munnell, The Effect of Social Securi'ty on Personal Saving Cambridge, Massachusetts: Ballinger, 1975 ) . The evidence that Social Security does not substantially retard saving remains weak excellent critique of these alternative theories see Ferrara, op. cit pp 76-104 For a more complete review see Robert J. Barro American Evidence from U.S. Time American Enterprise A Correction and Update,"

For an 15 including saving. The magnitude of this reduction depends on the marginal propensity to save.

More important, many individuals view their contributions to Social Security as a form of forced savings for retir ement and are therefore likely to reduce the savings they otherwise would have accumulated for retirement. Savings may be depressed by the full amount of Social Security taxes. For example, consider an individual with an income of $10,000 a year who, with o ut Social Security, might save 10 percent of his earnings for retirement.20 With Social Security, he is forced to save more than the 1,000 he originally intended to save because he is required to pay a greater amount in Social Security payroll taxes In th is case, the net impact is likely to be a reduction on savings equal to the full $1,0

00. Because the government uses the contributions to pay current beneficiaries, there is no corresponding increase in savings to balance the resulting decline in savings. On the aggregate level, pay-as-you-go financing may lower savings on a dollar-for-dollar basis with Social Security contributions In an empirical study, Martin Feldstein estimated that Social Security had reduced personal savings by 55 billion in 1976.21 this loss amounts to 58 percent of total actual private saving and a 37 percent reduction of the potential total private saving of $150 billion. That same year, employee and employer contribu tions to Social Security OASI) amounted to $63 billion. Thus, t he estimated 55 billion decline corresponds to 87 percent of private saving and supports the hypothesis that Social Security lowers private savings almost by the full amount of the tax.

Covered workers may think of their contributions to Social Security as part of their overall personal wealth. Itsocial Security wealth1' can be defined as the present value of the annuity stream that an individual expects to receive in the future It is fungible with ordinary wealth and allows one to reduce his own accumulat ion or personal wealth by an equal amount.

Unlike ordinary wealth, however, Social Security wealth is not a tangible form of wealth. Rather, it is simply an implicit promise that the next generation will tax itself to pay the benefits that were promised by an earlier generation. Because Social Security wealth is not real, the amount of Social Security wealth represents the stock of personal wealth lost because of the program. Feld stein estimates the value of Social Security wealth in 1976 to have been 3,2 3 8 billion (in 1976 dollars). By inducing people to substitute Social Security wealth for real wealth, the Social With $95 billion in total private savings that year 2o 21 This example is taken from Martin Feldstein Social Security," in Boskin op. cit p. 2 2 Martin Feldstein Social Security, Induced Retirement, and Capital Accumulation: Research, Working Paper 579, November 1980 A Correction and Update," National Bureau of Economic 16 Security system has had an enormous adverse impact on the size of the nati on's capital stock.

By decreasing saving, Social Security lowers the amount of money available for capital formation. In essence, this redistri butes income from labor to owners of capital, as the relative scarcity of capital drives up its price, and the s maller capital stock leads to a decline in worker productivity which reduces per capita output and the real wage rate. This redistribution is particularly harmful to low-income workers, who rely heavily on wage income for support, whereas the wealthy typi cally receive a much larger share of their income through capital investment.

Lower wages are also likely to contract the supply of labor by lowering the price of leisure relative to labor. Taken together lower productivity and less employment lead to slow er economic growth and a lower gross national product I Labor Supply The Social Security payroll tax also has's pronounced negative impact on labor supply because it drives a wedge between what an employer pays and what an employee receives as compensatio n.

This wedge includes both the employee's and the employer's share of the tax since the burden of the tax is shifted completely to la bor. According to a basic law of economics, employment is inversely related to the real wage rate, which is equal to the marginal productivity of labor wage rises, another must fall if existing levels of employment are to be preserved. In other words, an employer is able to pass the burden of the payroll tax onto an employee by lowering his observed wage, thereby maintaining a constant real wage.

Even if employers cannot prevent the real wage from rising somewhat, labor still cannot evade the consequences of the tax A rising real wage increases labor costs, thereby lowering employ ment opportunities. This simply shifts the burden to those either losing their jobs or those unable to find work by reducing a worker's disposable income, this wedge discourages e mployment. Consider an employee whose productive value enables him to command $10 an hour in the marketplace. Today, with a combined OASDHI tax rate of 13.3 percent, this individual will receive only $8.67 after deducting both the employer and employee po r tions of'the tax.22 By lowering compensation, the Social Security payroll tax precludes workers from receiving their full worth and induces some of them to drop out of the labor force as the price of leisure falls. New York attorney Peter Ferrara accurate l y characterized the consequences of the tax payroll tax is essentially a tax on employment and as always the result of taxing something is that there is less of it.1r23 This If one component of the real Moreover The 22 The size of this wedge may be somewh at smaller after taking into account the tax deductibility of the employer's share. This mitigates, but does not eliminate, the adverse effect on labor supply.

Ferrara, op. cit., p. 105. 23 17 employment effect would largely be negated if individuals could receive actuarially sound returns on Social Security, but the return on Social Security is uncertain and only tenuously linked to past contributions. Moreover, as the system fully matures the rate of return for an increasing number of people will fall be low the market rate of return, which will further exacerbate the distortion of the wedge effect on labor supply.

When considered along with the labor supply effects of the earnings test, which reduces employment among those over 65, it is apparent that Soc ial Security depresses employment, creates economic inefficiency by misallocating resources, and reduces total output. These problems of Social Security arise once again from Social Security's conflicting objectives of trying to be both an insurance and a welfare program AVENUES FOR REFORM Indexing One commonly discussed proposal to improve efficiency and equity within the Social Security system is to modify benefit indexation. Benefits are currently adjusted for inflation by indexing them to the Consumer P rice Index (CPI). Use of the CPI however, may improperly lead to excessive Social Security 'benefits if, as many economists assert, it overstates the true rate of inflation. One of the major flaws in the CPI is its treatment of homeownership. The CPI over s tates housing costs by ignoring the investment value of the home. Other criticisms of the CPI include outdated buying patterns (determined in 1972-73 failure to account for consumer substitution when faced with higher prices and limited applicability to c e rtain subgroups, such as the elderly. In this connection, it should be noted that only a very small proportion of the elderly are in the housing market, a category heavily weighted in calculating the CPI ing the inflation rate, the CPI may lead to the sub stantial overcompensation of Social Security recipients by the Bureau of Labor Statistics, endeavors to circumvent some of the more serious problems connected with the housing component.

This approach attempts to separate the consumption and investment mot ives in purchasing a home by using market rents as a proxy for the shelter services of a similar owner-occupied home By exaggerat A "rental equivalencet1 (CPI X-1) index, now being developed Other economists have suggested using the Personal Consump tion E xpenditure (PCE) chain index on the National Income and Product Accounts. This index would be preferable to the CPI for several reasons: 1) its coverage is somewhat broader as it includes all goods and services currently produced for consumption 2) it emp l oys the rental equivalency approach used in computing the CPI X-1; and 3) it uses current consumption patterns rather than those determined in the 1972-73 survey period. 18 In lieu of changing the CPI or replacing it with a new index, President Reagan and other policymakers are considering a three-month deferral of cost-of-living increases for Social Security beneficiaries. savings, it ignores the basic flaws of the CPI as an escalator.24 Although this would result in large Raising the Retirement Aqe Incre a sing longevity among the elderly and the trend toward early retirement have contributed greatly to the actuarial imba lance of Social Security by lengthening the period over which .benefits are received and reducing the number of years during which taxes a re paid into the system originally conceived in 1935, the average retiree spent 12.8 years in retirement; now.it is 16 years.25 Moreover, this expec ted increase in longevity will extend the retirement period still further, thereby increasing the threat o f insolvency. As a result, one of the most frequently discussed proposals under consideration is delaying retirement by raising the age at which full Social Security benefits are paid. This would help offset the projected reduction in the ratio of workers to beneficiaries.

To allow those nearing retirement age sufficient time to adjust this reform could'be phased in gradually.26 When Social Security was Raising the retirement age for Social Security recipients can be justified for several reasons. First, me dical advances not only have increased life expectancy, but also have enabled people to work longer and manufacturing to trade and service jobs has reduced the proportion of the labor force in physically demanding and dangerous employment. Finally, the de m and for higher education has effec tively shortened the working years for many people by delaying their entrance into the labor force Second, a shift in employment from mining The primary disadvantage of raising the retirement age however, is that it does not consider the special needs of the elderly who may not be able to postpone retirement for health reasons or other circumstances by allowing workers to retire at any age they choose and to receive actuarially fair benefits. -Unfortunately, this may not be possible unless the transfer and annuity portions of Social This disadvantage can be avoided 24 For a detailed discussion of the CPI and indexing, see Adjusting the Consumer Price Index ,I' Heritage Foundation Backgrounder No. 152, October 15, 1981.

President's Commission on Pension Policy, Coming of Age Retirement Income Policy, Washington, D.C., February 26, 1981, p. 23.

Representative J. J. Pickle (D-Tex Chairman of the Ways and Means Subcommittee on Social Security, and Senator Lawton Chiles (D-Fla.) have both introduced bills raising the retirement age gradually from 65 to 68.

Rep. Pickle would implement this change from 1990 to 1999, while Sen.

Chiles recommends that this be done from 2000 to 2012 Peter Germanis 25 26 Toward a National 19 Securit y are separated because the flexibility would be more than the cost of maintaining this system could support. Reducing the number-of years individuals can collect benefits may be a more politically feasible alternative than cutting benefits by eliminating the current welfare component.

Taxation of Benefits Employees now pay income taxes on the part of their earnings also subject to the Social Security payroll tax; employer's contributions, however, are considered business expenses and thus escape taxation. Including half of all Social Security benefits in taxable income would approximate the current tax treatment of private pensions and benefits from other government programs.

This measure would have few, if any, ramifications on the low income group since the tax code already has several provisions that take into account an individual's ability to pay, e.g progressive tax rates, a zero-bracket amount, and a personal exemption that is doubled for those over 65. Excluding Social Security benefits from taxat i on simply helps those recipients with relatively high incomes. Security benefits in taxable income would reduce the after-tax benefits for some recipients, it also could be viewed as a tax cut because it would reduce the need to raise payroll taxes in the future Although including half of Social Universal Coverage Universal coverage requires the extension of Social Security protection to all workers. At present some seven million workers are not covered primarily those permanent civilian employees of the f ederal government, employees of a number of state and local governments, and non-profit organizations choosing not to participate.

Making coverage compulsory for all workers would eliminate gaps in protection experienced by workers moving between covered a nd uncovered employment ty and most pension plans require a minimal period of employment before eligibility is established that accrue to individuals with short work periods in covered employment or those with low earnings histories arising from time spen t in non-covered employment. qualified for benefits under other retirement plans and, despite being relatively well-off, receive unearned benefits designed to provide a subsistence income for workers with low lifetime wages. The following examples illustra te the present inadequacy.

Mr. Jones worked his entire life as an unskilled laborer and at age 65 retired with an AIME of $211, which entitled him to a monthly benefit of $189.

90. Mr. Jones is the type of worker the weighted benefit formula was designed to assist These gaps arise because Social Securi Universal coverage also would largely eliminate windfalls Many of these people have Mr. Smith, on the other hand, spent ali his l ife as an employee of the federal government and earned a civil service pension of $1,500 per month. In addition Mr. Smith qualified for Social Security benefits by working in covered employment in his spare time. With an AIME of $211, Mr. Smith also rece i ved a monthly annuity of $189.90 for Social Security. The progressive benefit structure was not intended to yield him such a generous award, yet he was allowed to take advantage of it This type of abuse of Social Security is not uncommon approximately 45 p ercent of those receiving a civil service pension also get benefits under Social Security.27 Universal coverage would greatly increase contributions by expanding the tax base while only gradually increasing disbursements. If mandatory coverage were to bec o me effective January 1, 1982, for example, incremental revenues of over $100 billion could be realized by 1987, thereby providing immediate relief for the short-run financing problems.28 run because it would be many years before a sizable number of newly- c overed workers would reach retirement. long-term effects are not as significant the Social Security Administration estimates that universal coverage could lead to a reduction in payroll taxes of 0.5 percentage points during the next seventy-five years.29 i mprove the actuarial balance of Social Security, however, is by giving the newly-covered workers a "bad deal and it is therefore sure to be resisted This is possible in the short Although the The only way this proposal could Spreading the burden to uncove r ed workers thus is not a solution to Social Security's problems. Its main advantage is that it points out the great need to separate the welfare compo nent from the program so that no group is unfairly penalized by participating in the program coverage a moot question.

This would then make universal Accounting Chanqes Accounting adjustments may provide relief in the short run without altering either benefit payments or already scheduled tax rates.

Interfund borrowing would allow the three trust funds to This borrow from one another whenever any runs short of funds 27 28 Colberg, op. cit., p. 16.

Lowell Jones and Michael Romig, "Social Security Financing and Options for the Future," Statement of the Chamber of Commerce of the United States of America, July 10, 1981, p. 9.

U.S. Department of Health and Human Services, The Desirability and Feasi bility of Social Security Coverage for State and Local Government and Private Non-profit Organizations, Report of the Universal Social Security Coverage Study Group (Washington, D.C., 1980 p. 47 29 21 change is aimed at augmenting the projected shortfall in the OASI fund, the largest of the three trusts, and would enable it to receive temporary transfers from the more solvent disability insurance DI) and hospital ins u rance (HI) funds. Another alter native is to realign the proportion of the payroll tax going to each of the three trust funds. These reallocations may resolve the short-run financing problems, but fail to address the more serious and imminent underlying p roblems facing Social Security.

General Revenue Financinq One of the many nostrums advocated for curing the Social Security system of its financial ills is financing part or all of the program through general revenues. Through general revenues goes the arg ument, part (or all) of the burden of the payroll tax would be shifted from the low wage worker, who can least afford a reduction in disposable income, to those in society better able to absorb the loss. This proposal, however, would further weaken the al r eady tenuous connection between payments made into the system and benefits paid out, while substantially advancing its conversion to a pure welfare program. If the objective of this reform is to aid the poor, moreover, then this can be more effi ciently a c hieved through other needs-related programs financed by general revenues such as the Supplemental Security Income (SSI) program, food stamps, and the Aid to Families with Dependent Children (AFDC) program. In particular, SSI provides a guaranteed level of income for the aged, blind, and disabled because it recognizes that Social Security alone may not be sufficient.

CONCLUSION The key to establishing a fair and efficient retirement program is to eliminate the transfer function of Social Security altogether and pay benefits that are directly related to an individual's total contributions plus interest. This would require dropping the redistributive benefit structure, the spouse's benefit, the earnings test, and other features based on welfare principles. be n efits to those generally not considered needy transfers without regard to need raises serious questions of equity, while further financially emasculating the Social Security system would more appropriately be helped through needs-oriented programs, such a s SSI and food stamps, which are financed from general revenues. of course, but there could be enormous savings running probably into the tens of billions of dollars by precluding retirees who are not poor from receiving undeserved welfare benefits. A majo r objection to this reform, however, is that many of the elderly poor would be subject to Ildegradingll needs-related tests. Even though some of these people may feel embarrassed to accept finan These often have led to paying substantial unearned Continuin g The destitute elderly would not be ignored. Instead, they Funding for these programs would have to be expanded, 22 cia1 assistance from welfare programs, they are, nevertheless now receiving an implicit welfare subsidy through Social Security.

There need be no stigma attached to receiving welfare. In any event, if 'the rest of the nation's needy citizens receiving assistance are obliged to recognize their plight, why should the elderly be exempted?

The final step in bringing about a comprehensive reform of Social Security is to make it voluntary and partially, if not completely, privatized. Some 60 percent of those surveyed in the Heritage poll feel that Social Security should be made voluntary; an even greater number believe that private pension alterna t ives could provide for retirement more efficiently. Though extensive analysis of such reform is beyond the scope of this paper, there are a few advantages to this option that are readily apparent.36 There is no reason that government-approved private insu r ance alternatives cannot adequately provide for retirement or any of the other contingencies covered by Social Security. People could be given a choice of either continuing their participation in Social Security (without the welfare subsidy) or investing a portion of their income in private plans. This would greatly expand individual liberty by allowing people to choose from a variety of plans and to purchase the one best suited to their needs and desires In addition, fully funded private plans would stimu late economic growth by reducing the adver.se incentives for capital formation of the pay-as-you-go nature of the current system.

Although some of the reforms outlined in the preceding section of this paper might strengthen the financial soundness of Socia l Security, they do not address the main flaw of the program its schizoid attempt to pursue both insurance and welfare objec tives. Yet many ignore this flaw and oppose genuine reform. The National Commission on Social Security, for example, contends that Social Security is "sound in principle" and that any substan- tial changes would violate the implicit compact made between generations S.ocia1 Security is based upon a social compact between generations. benefits on those who are working, just as their ta x es paid the benefits to those who came before them. the younger generation, the deduction from their earn ings for Social Security is justified by the understand ing that the system will support them when they retire.31 This so-called compact, however, wa s made by a generation Those who are retired depend for their For that is now reaping a tremendous windfall at the expense of the 30 31 For a compendium of proposals to achieve these goals, see Ferrara op. cit pp. 311-397 National Commission on Social Secu r ity, op. cit p. 131. 23 up-and-coming generations. The obvious question is: Why should a younger generation be bound by a compact made without their consent and one that will yield them an unjustifiably low return in their retirement? Basing benefit payme nts on past contribu tions plus interest is the key to achieving a fair and efficient retirement program.

Peter G. Germanis Policy Analyst


Peter G.

Senior Fellow and Director of Government Finance Programs