SSP No. 4 July 26, 1996

Social Security

Privatizing Social Security: A Big Boost for the Poor

by Michael Tanner

Michael Tanner is director of health and welfare studies at the Cato Institute.

Executive Summary

Critics of Social Security privatization often warn that such proposals hold serious dangers for the elderly poor. However, a closer examination of the evidence indicates that the poor would be among those who would gain most from the privatization of Social Security.

By providing a much higher rate of return, privatization would raise the incomes of those elderly retirees who are most in need. Although the current Social Security system is ostensibly designed to be progressive, transferring wealth to the elderly poor, the system actually contains many inequities that leave the poor at a disadvantage. For instance, the low-income elderly are much more likely than their wealthy counterparts to be dependent on Social Security benefits for most or all of their retirement income. But despite a progressive benefit structure, Social Security benefits are inadequate for the elderly poor's retirement needs.

In addition, the progressivity of Social Security is undermined by differences in life expectancy. Because the wealthy generally live longer than the poor, they receive more total Social Security payments over the course of their lifetimes. In a privatized system, an individual's benefits would not be dependent on life expectancy. Individuals would have a property right in their benefits. Any benefits remaining at their deaths would become part of their estates, inherited by their heirs.

Finally, Social Security drains capital from the poorest areas of the country, leaving less money available for new investment and job creation. Privatization would increase national savings and provide a new pool of capital for investment that would be particularly beneficial to the poor.

For those reasons, Social Security privatization should be viewed as a big boost to America's poor.


The benefits that privatization of Social Security would bring to most of society are obvious. First, privatization would bail out a system hurtling rapidly toward insolvency.1 Second, privatization would provide retirees with a much higher rate of return on their investments.2

Those arguments have been made frequently by advocates of privatization. However, there is another reason for privatizing Social Security: privatization would greatly benefit the poor; they could retire with dignity and leave estates to their heirs, thus enabling them to lift themselves out of poverty.

As envisioned by most supporters, a privatized Social Security system would essentially be a mandatory savings program. The 10.52 percent payroll tax that is the combined employer-employee contribution to the Old-Age and Survivors Insurance (OASI) portion of the Social Security program would be deposited in a personal security account (PSA) chosen by the individual employee. (The disability and Medicare programs are separate issues and should be dealt with separately.)

PRAs would operate much like current Individual Retirement Accounts (IRAs) or 401(k) accounts. Individuals could not withdraw funds from their PSAs before retirement, determined either by age or by PSA balance requirements. PSA funds would be the property of the individual, and upon death, any remaining funds would become part of the individual's estate.

PSAs would be managed by the private investment industry in the same way 401 (k) plans and IRAs are. Individuals would be free to choose the fund manager that best met their individual needs and could change managers whenever they wished. The government would establish regulations on portfolio risk to prevent excessive speculation and protect consumers. Reinsurance mechanisms would be required to guarantee fund solvency.

If necessary, the government could continue to provide a minimum safety net, through either a guaranteed minimum benefit or a floor benefit. Under a guaranteed minimum benefit, if upon retirement the balance in an individual's PRA were insufficient to provide an actuarially determined retirement annuity equal to a minimum benefit--for example, a benefit equal to the minimum wage--the government would provide a supplement sufficient to bring the individual's monthly income up to the level of the minimum wage. A floor benefit would provide a minimum standard flat benefit to all retired Americans regardless of income.

Social Security's Benefit Structure

Social Security is intended to have a progressive benefit structure. That is, lower income recipients are to receive a return on taxes paid into the system that is proportionately higher than the return received by higher income people. Thus, in addition to simply providing retirement benefits, Social Security attempts to redistribute income from wealthy retirees to poorer ones.

In determining an individual's Social Security benefits, the Social Security Administration first calculates that person's Average Indexed Monthly Earnings (AIME). All earnings on which the person has paid Social Security taxes between 1951 and the year in which the individual reaches age 60 are indexed to account for past inflation and real wage growth. The indexing formula is based on the ratio of the average national wage in the year the individual turns 60 to that in the year to be indexed. Wages earned after age 60 are not indexed but left in nominal dollars. The Social Security Administration then selects the 35 years in which wages were highest. The wages for those years are totaled and divided by 420 months. The result is the AIME.3

A progressive formula is applied to the AIME to determine the person's Primary Insurance Amount (PIA). For people turning 65 in 1995, that formula was 90 percent of the first $387 of the AIME, plus 32 percent of the next $1,946 of the AIME, plus 15 percent of the remaining AIME.4

In the case of early or late retirement, benefits are adjusted accordingly. For example, if a worker retires at age 62 (the earliest allowable age under Social Security), the worker's benefits will be permanently reduced to 80 percent of his or her PIA. If the worker delays retirement until age 67, his benefit will be increased to 109 percent of PIA.5

Spouses and survivors of workers are also eligible for benefits. A spouse who has not worked long enough to be eligible for her own benefits will, at age 65, receive benefits equal to 50 percent of the worker's PIA. If the worker dies, the spouse receives survivor's benefits equal to 100 percent of the worker's PIA. If the widowed spouse is under age 65, she is eligible for benefits equal to 75 percent of PIA if she is caring for children under age 16. Minor children of a deceased worker are also eligible for survivors' benefits equal to 75 percent of PIA. However, a family's total benefits are capped at between 150 percent and 187.5 percent of PIA, according to a mildly progressive formula.6

Depending on Social Security

Social Security currently accounts for approximately half of all retirement benefits. More than half of the elderly receive no private pension, and more than one-third have no income from assets.7 But that tells only part of the story. Low-income workers are far more likely to be dependent on Social Security when they retire than are their high-income counterparts. As Figure 1 shows, the poorest 20 percent of the elderly depend on Social Security for 81 percent of their income, while Social Security provides only 20 percent of income for the wealthiest 20 percent of retirees.8

Figure 1

Social Security as a Percentage of Aggregate Retirement Income
by Income Quintile


Source: Neil Gilbert and Neung-Hoo Park, "Privatization, Provision, and Targeting: Trends and Policy Implications for Social Security in the United States," International Social Security Review 49 (January 1996): 22.

That trend has been slowly but steadily growing worse, with the poorest elderly depending more and more on Social Security. Projections indicate that the disparity between poor and wealthy seniors will grow wider in the future, with low-income elderly people remaining extremely dependent on Social Security while those with higher incomes rely increasingly on private pensions and other sources of income.

The question then becomes whether Social Security benefits are sufficient to provide an adequate retirement income for those elderly poor who have no other income. Perhaps the easiest way to look at the question is to examine Social Security's "replacement value," that is, the proportion of a person's preretirement income that Social Security benefits equal.

Because of the progressive benefit formula described above, Social Security replaces a higher proportion of income for low-wage earners than for high-wage earners. The actual replacement rate fluctuates slightly from year to year because of a number of factors. For an average-wage earner who retires this year, Social Security can be expected to replace approximately 42.4 percent of preretirement income. A low-wage worker will receive benefits equal to 57.1 percent of preretirement income, while the replacement rate for a high-wage worker will be 25.6 percent.

For people who are 25 years old today and will retire in 2037, projected replacement rates are 55.7 percent for low-wage workers, 41.5 percent for average-wage earners, and 27.5 percent for high-wage earners.9 However, the relatively higher replacement rate for low-wage workers should be considered in light of those workers' greater dependence on Social Security. As shown above, low-wage workers depend on Social Security for 81 percent of their income. Thus, total annual retirement income for the 25-year-old low-wage worker (from Social Security and other sources) will equal only 69 percent of preretirement income. In comparison, Social Security accounts for only 20 percent of postretirement income for high-wage earners. Therefore, when a high-wage worker retires, his income will actually equal 138 percent of his income before retirement despite the lower replacement rate.

Most financial planners suggest that if one's preretirement standard of living is to be maintained, retirement benefits of between 60 and 85 percent of preretirement income are probably necessary.10 Clearly, then, Social Security fails to provide sufficient income to afford poor workers a dignified and secure retirement.

Perhaps that is one reason why poverty rates remain higher for people over the age of 65 than for people aged 18 to 65. Although we have made extraordinary strides in reducing poverty among the elderly, in 1992, 12.9 percent of seniors were poor, compared to 11.7 percent of 18 to 65 year olds.11

Consider the advantages that privatization of Social Security would bring to the elderly poor. In his 1995 study for the Cato Institute, financial analyst William Shipman of State Street Global Advisors compared the Social Security benefits that an individual is scheduled to receive under current law with the potential return that the individual would received if he or she had been allowed to invest an amount equivalent to the OASI portion of payroll taxes in either stocks or bonds.

Shipman found that the 25-year-old low-wage worker described above is scheduled to receive Social Security benefits of $769 (1995 dollars) per month upon retirement at age 67. However, if that young worker were allowed to invest his payroll taxes in bonds, he would be able to purchase an annuity upon retirement that would provide benefits of $1,085 per month, assuming that the average performance of the bond market over the next 40 years remains approximately the same as the average over the last 60 years. Again assuming historic rates of return, investing in stocks would provide that worker with benefits of $2,419 per month (Figure 2).12

Figure 2

Monthly Benefit Comparison of Returns from Social Security
and Capital Markets for a Low-Wage Worker


Source: William G. Shipman, "Retiring with Dignity: Social Security vs. Private Markets," Cato Institute Social Security Paper no. 2, August 14, 1995, p. 4.

With those higher returns, the worker's retirement income would equal 92 percent of his preretirement income if he had invested in bonds and 188 percent of preretirement income if he had invested in stocks. Clearly, the higher rate of return would benefit the elderly poor, providing them with a higher postretirement standard of living.13

Moreover, any discussion of scheduled benefit levels or projected replacement rates assumes that Social Security will be able to pay all the benefits promised. But only 70 percent of future benefits are fully funded.14 Thus, unless additional funding is found, benefits will have to be reduced and replacement rates will fall, perhaps by as much as 30 percent.15

Lifetime Return on Investment

Another way to look at Social Security benefits is to consider the total benefits that a retiree receives over his lifetime. Clearly, a person who dies the day after he turns age 65 receives less money from Social Security than does an individual who lives to be 100.

For a variety of reasons, there is a well-established relationship in this country between wealth and longevity (Figure 3).16 Put simply, wealthy people live longer than poor people.


Figure 3

Males Surviving of 100,000 Reaching Age 20

Source: Daniel Garrett, "The Effects of Differential Mortality Rates on the Progressivity of Social Security," Economic Inquiry 33 (July 1995): 458.

Therefore, wealthy retirees will probably receive Social Security for a longer period of time than will their poorer counterparts. As a result, the question becomes whether the longer period of returns offsets the system's progressive benefit structure. In short, do the wealthy get a better deal than the poor?

Analysis of that issue has provided conflicting answers. One of the first economists to examine the question was Henry Aaron of the Brookings Institution. In a 1979 study, Aaron concluded that the poor ultimately receive a lower return in retirement benefits on their past tax dollars than do those with higher incomes.17 However, Aaron now warns against overreliance on his study, saying that the calculations were incomplete and the issue needs to be studied further.18

The next major study of the issue was by Eugene Steuerle and Jon Bakija of the Urban Institute. In their 1994 book, Retooling Social Security for the 21st Century, they concluded that while differentials in mortality rates do weaken the progressivity of Social Security, the differentials are not sufficient to completely offset the progressive benefit structure. Therefore, the system still results in a net redistribution of wealth from rich to poor. The transfer is smallest for single men and largest for one-wage-earner couples.19

However, Daniel Garrett, writing last year in Economic Inquiry, provided a different perspective.20 Whereas Steuerle and Bakija based their conclusions on the overall transfers over an entire income quintile, Garrett looked at the percentages of winners and losers within each quintile. According to Garrett, when mortality differentials are considered, slightly more than half (53.7 percent) of poor single-earner couples receive a negative return from Social Security. Because single-earner couples receive the highest return on their Social Security payments, the impact on other types of households is likely to be even greater. Garrett concluded that "differences in mortality eliminated the progressive spread in returns across income categories."21

Both the study by Steuerle and Bakija and the one by Garrett agreed that the long-term progressivity of Social Security benefits is reduced by the tendency of low-income workers to enter the labor force at a young age.22 The poor tend to start work right after high school, if not before, while the middle class and the wealthy are more likely to delay full-time entry into the workforce until after they have completed college. Therefore, the poor will begin paying taxes several years earlier than will the wealthy, but paying taxes for those additional years does not earn any additional benefits.23

A February 1996 study by the RAND Corporation also concluded that there was a positive correlation between total benefits and length of life. As a result, the researchers found that, although Social Security benefits retain some degree of progressivity, if such factors as marital status and life expectancy are taken into account, "the OASI system is not as progressive" as generally believed.24 Indeed, the study found that for unmarried men, the rate of return for high-wage earners was actually higher than for low-wage earners, making Social Security regressive for that category of recipients.25

The RAND study also concluded that the current benefit structure disadvantages African-Americans, who have lower life expectancies and marriage rates. According to the study, whites consistently earn higher rates of return than do blacks.26 In fact, over a lifetime, the income transfer from blacks to whites is as much as $10,000.27 Moreover, although evidence on the current long-term progressivity of Social Security benefits may be ambivalent, there is no doubt that proposed reforms such as raising the retirement age would tilt the equation strongly toward higher income retirees.

Thus far, we have considered the return for individuals who live until retirement age. However, the poor also have much higher death rates before age 65 than do the nonpoor.28 Individuals who die before age 65 never receive any retirement benefits at all. Their survivors will, of course, receive survivors' benefits, which at least partially offset the loss of retirement benefits. However, for a tiny fraction of what a person paid in OASI taxes, he or she could have purchased a life insurance policy that would have paid at least as much as Social Security. Moreover, the major part of OASI taxes goes, not to pay for survivors' benefits, but for retirement benefits that the worker will never receive. Indeed, if a person is single or has no survivors, he or she receives nothing at all in exchange for years of payroll taxes. Survivors' benefits accounted for only 5.3 percent of all OASI payments in 1995. Therefore, a person who dies before age 65 receives the worst deal of all from Social Security, and that person is far more likely to be poor than rich.29

Privatization provides an answer to the problem of lifetime return in three ways. First, by transforming Social Security from a defined benefit to a defined contribution plan, privatization would disconnect total benefits from life expectancy. The benefits an individual would receive would depend on what was paid into the system plus the investment return on those payments, not on how many years the person received benefits.

Second, under a privatized system, individuals would have a property right to their benefits. If a person were to die with money still in his or her retirement account, that money would become part of an estate, to be inherited by that person's heirs.

Finally, individuals who begin work earlier, and therefore contribute for additional years, would earn additional benefits as a result of their contributions.

Capital, Investment, and Economic Opportunity

Social Security's pay-as-you-go financing mechanism reduces national savings, which leads to a decline in capital investment, national income, and economic growth. Among the areas that suffer most as a result are America's inner cities and other regions of high poverty and unemployment.

Virtually everyone agrees that countries that save and invest more grow faster and have more rapid improvements in their standards of living. Yet, as shown in Figure 4, the United States has the lowest national savings rate in the industrialized world.30

Figure 4

National Savings Rates


Source: Bipartisan Commission on Entitlement and Tax Reform, Interim Report to the President (Washington: Government Printing Office, August 1994), p. 8.

The principal reason for the low national savings rate is the large deficits run by the federal government. However, there has also been a significant decline in personal savings. That is particularly important because, with the government a negative saver, personal savings have become an increasingly important component of national savings. Indeed, because government savings are now negative, personal savings provided $1.24 of every net dollar of national savings in the 1980s ($1.24 private savings - $0.24 government dissavings = $1.00 net national savings), compared to 77 cents in the 1970s and 59 cents in the 1960s.31 However, as Figure 5 shows, personal savings have declined to barely more than 4 percent of GDP from a high of more than 8 percent during the 1960s.32

Figure 5

 Personal Savings as a Percentage of Gross Domestic Product

Source: Bipartisan Commission on Entitlement and Tax Reform, Interim Report to the President (Washington: Government Printing Office, August 1994), p. 9.

Clearly, the imposition of the Social Security tax reduces private savings. Workers are required by law to pay Social Security taxes.33 That precludes their putting those lost wages in private savings or investments.34

Some people might argue that would not matter, if private savings and investment were replaced by government savings and investment. However, even granting that rather dubious premise, approximately 86.5 percent of the money collected in Social Security taxes is not saved or invested in any sense of the word; it is simply paid out in benefits.35

Moreover, "investment" of the remaining 13.5 percent is more semantic than real.36 That money is used to purchase federal Treasury obligations that are credited to the Social Security Trust Fund; the government then uses the money it has borrowed from the trust fund to meet current operating expenses.37

In addition, the trust fund surpluses are a temporary phenomenon. Beginning as early as 2013, every penny collected in Social Security taxes will be used to pay benefits. Indeed, the payroll tax will not be sufficient to pay all the benefits that are promised; the federal government will be forced to turn in the bonds in the trust fund to obtain the cash needed to finance benefits.38

Therefore, the current Social Security system is helping to reduce private savings and limit the pool of capital available for new investment. The lack of capital is of particular concern in high-poverty areas.

As Jesse Jackson says, "Capitalism without capital is just an 'ism.'"39 Most economists recognize that when investment capital grows scarce, the areas that are the first to feel the lack are areas where investments are most risky--inner-city neighborhoods with high crime rates, a poorly educated workforce, and high business bankruptcy rates.40 As economist Jude Wanniski explains, "The people who lose the most . . . are the poorest, the youngest, those at the beginning of their careers, those who are furthest from the sources of capital."41

In addition, by reducing savings and capital accumulation, Social Security reduces the ratio of capital to workers, which leads to a reduction in productivity. As a result, wages are lower than they would otherwise be.42 Interestingly, however, the reduction in capital may actually drive up interest rates and, therefore, the return on capital. The poor depend primarily on wages, while the wealthy receive a far higher proportion of their income from capital. Thus, Social Security may perversely redistribute wealth from the poor to the rich.43

Of course, devotees of Keynesean economics might argue that the loss of investment is "offset" by increased consumption brought about by the payout of benefits. That is, beneficiaries spend the money they receive from Social Security and thereby benefit the economy. However, even under that scenario, poor communities are left behind.

The geographical distribution of Social Security benefits is also far from even. For example, a 1995 study by the youth advocacy group Third Millennium found that the 10 wealthiest congressional districts received $451 million in Social Security retirement benefits in December 1993, compared with $314 million for the 10 poorest congressional districts. Even if you adjust for the varying number of retirees in each district, the disparity remains. Social Security pays out an average of $769.75 per retiree in the 10 wealthiest districts but only $602.51 per retiree in the poorest districts.44

Indeed, in all congressional districts with average incomes above $50,000, the average monthly Social Security retirement payment was $770.42, while the average payment in districts with incomes below $20,000 was only $603.83.45 Of course, that distribution does not necessarily mean that Social Security is regressive. However, it does mean that Social Security payments are not going to poor districts as much as they are to wealthy ones.

Last, it is important to recognize that one key to improving the condition of the poor in America is economic growth. The United States is currently undergoing one of the slowest economic growth periods since the Great Depression. As Figure 6 shows, the slowdown in economic growth has had its strongest impact on the poor. During the first half of the 1980s, a period of robust growth, the real family incomes of all economic groups rose substantially. However, between 1989 and 1994 as economic growth slowed, only the wealthiest Americans saw their real family incomes increase. Americans with incomes in the lowest 20 percent experienced a decline of nearly 6 percent.46

Figure 6

Change in Aggregate Real Income

Source: Derived from Bureau of the Census, Current Population Reports, Series P-60, March 1996, Table F-1.

In contrast to the current system, a privatized Social Security system would significantly increase savings, investment, and economic growth. The movement of so much capital into private markets would have a significant impact on economic growth. Martin Feldstein, for example, estimates that "the combination of the improved labor market incentives and the higher real return on savings has a net present-value gain of more than $15 trillion, an amount equivalent to 3 percent of each future year's GDP forever."47

Look, for example, at what happened in Chile after that country successfully privatized its pension system in 1981. As Robert Genetski, former chief economist at Chicago's Harris Bank, points out, "The buildup of funds in the workers' retirement accounts has produced a 29% savings rate. . . . Instead of resenting the rich, Chile's workers themselves are becoming rich."48 The cumulative assets handled by Chile's fund managers exceed $23 billion, or roughly 41 percent of GDP.49 During the past decade Chile's real GDP growth has averaged over 7 percent, more than double that of the United States.50

Privatizing the Social Security system can have the same beneficial economic impact here.51 Although all Americans will benefit from increased savings, investment, and economic growth, the poor will be among those receiving the biggest boost.

Giving the Poor Ownership in America

An important side benefit of Social Security privatization is that it would give every American--including poor Americans--an opportunity to participate in the economy by owning a part of it. In effect, a privatized pension system would act as a nationwide employee stock option plan, which would allow even the poorest workers to become capitalists. Through Social Security privatization, workers would become stockholders. The division between labor and capital would be broken down.

As minister of labor, José Piñera was the architect of Chile's successful privatization of its pension system. He explains the change that privatization brought to Chilean workers:

The new pension system gives Chileans a personal stake in the economy. A typical Chilean worker is not indifferent to the stock market or interest rates. When workers feel that they own a part of the country, not through party bosses or a Politburo, they are much more attached to the free market and a free society.52

The same empowerment of workers is possible through privatization of Social Security in this country.

The Alternatives to Privatization: A Bad Deal for the Poor

Finally, Social Security privatization must be looked at in light of the alternatives. Because of Social Security's looming insolvency, failure to privatize the system will almost certainly result in a reduction in benefits, an increase in payroll taxes, or--more likely--both. Those alternatives would be particularly devastating for the poor.

Most poor people do not pay much, if any, in income taxes. They do, however, pay payroll taxes. In fact, 76 percent of Americans pay more in payroll taxes than they do in federal income taxes.53 The payroll tax paid directly by individuals is matched by payments from employers. That adds significantly to the cost of employment and leads to a loss of jobs. According to the Congressional Budget Office, payroll tax increases between 1979 and 1982, for example, resulted in the permanent loss of 500,000 jobs.54 A subsequent study of the 1988 and 1990 payroll tax hikes, by economists Gary and Aldona Robbins, estimated job losses at 500,000 per year and the reduction the U.S. GNP of $30 billion per year.55

Nearly all economists agree that individuals who are hired pay the cost of the so-called employer's share of the payroll tax through reduced wages. Therefore, an increase in the payroll tax would result in less money in workers' pockets.

It is also important to remember that the payroll tax is an extremely regressive tax. First, it is a tax only on wages, leaving other income sources, such as capital gains, interest, and other profits on investment, untaxed. Since wages represent a high proportion of the poor's income, a payroll tax will take a higher percentage of total income of the poor than of the wealthy. That effect is compounded because the amount of income subject to the payroll tax is capped.

When Social Security began, the payroll tax was quite small, a maximum combined employer-employee payment of $60. However, since then the payroll tax has been increased 13 times. Even adjusted for inflation, the maximum tax has increased nearly 900 percent. The amount of earnings subject to the tax has also been increased 26 times, rising from $3,000 in 1937 to $61,200 today.56

According to intermediate projections, to fund all promised benefits, the combined employer-employee OASDI portion of the payroll tax would have to be increased from the current 12.4 percent to nearly 18 percent.57 Adding the amount required to provide promised Medicare benefits would yield a total payroll tax of nearly 28 percent. Moreover, the so-called pessimistic assumptions, which assume slightly higher unemployment and inflation rates and an increase in life expectancy at retirement much closer to actual experience in recent decades, indicate that payroll taxes could reach 40 percent.58 The burden of a payroll tax increase would fall disproportionately on the poor, who would be the most likely to lose their jobs and who can least afford reductions in pay.

Likewise, as we have seen, a reduction in benefits would disproportionately hurt the poor since they are more likely than the wealthy to be dependent on Social Security benefits. Those elderly poor who receive most or all of their retirement income from Social Security can ill afford any reduction in benefits. Yet benefit cuts of as much as one-third may be required to keep the system solvent. A one-third reduction in Social Security benefits would leave the elderly poor with a postretirement income equal to only 50 percent of their preretirement wages. Such a benefit cut would likely plunge millions of elderly Americans into poverty.

Reducing benefits indirectly, for example, by raising the retirement age, would also disproportionately affect the poor, who are more likely to hold physically demanding jobs. Raising the retirement age would also work to the severe disadvantage of African-American men. As Figure 7 shows, the life expectancy of a black man today is just 65.8 years.59 Therefore, raising the retirement age to 70 would mean that many black men would pay Social Security taxes their entire lives but never collect benefits. (The impact is less significant for black women.)

Figure 7

 Life Expectancy by Race and Sex

Source: Based on data from the U.S. National Center for Health Statistics cited in Bureau of the Census, Statistical Abstract of the United States, 1995 (Government Printing Office, 1995), Table 114, p. 86.


Social Security privatization offers substantial benefits to the poor. Privatization would benefit them directly by providing them with higher benefits and removing the link between benefits and longevity. As a result, the poor elderly would have a more secure and more comfortable retirement.

Privatization would also benefit the poor indirectly by increasing both overall economic growth and the pool of investment capital available in poor neighborhoods. That would result in more jobs and higher wages for the working poor, while helping to revitalize our inner cities. At the same time, Social Security privatization would provide poor workers with greater participation in the American economic system by making them stockholders. Finally, privatizing Social Security would avoid other potential Social Security reforms, such as raising taxes or cutting benefits, that would hurt the poor.


1. As has been frequently pointed out, as early as 2013 the system will begin to pay out more in benefits than it takes in revenue. At that point the so-called Social Security Trust Fund must start turning in bonds to the federal government to obtain the cash needed to finance benefits. But the federal government has no cash or other assets with which to redeem those bonds. It can obtain the needed cash only by borrowing and running an even bigger deficit, increasing taxes, or cutting other government spending. See Mark Weinberger, "Social Security: Facing the Facts," Cato Institute Social Security Paper no. 3, April 10, 1996.

2. See William G. Shipman, "Retiring with Dignity: Social Security vs. Private Markets," Cato Institute Social Security Paper no. 2, August 14, 1995.

3. U.S. House of Representatives, Committee on Ways and Means, 1994 Green Book: Overview of Entitlement Programs (Washington: Government Printing Office, 1994), pp. 10-12.

4. Ibid., pp. 12-13.

5. Ibid., p. 13.

6. The monthly benefit cap formula is 150 percent of the first $544 of PIA, plus 272 percent of PIA between $545 and $785 per month, 134 percent of PIA between $786 and $1,024, and 175 percent of PIA above $1,024. 1995 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Trust Funds (Washington: Government Printing Office, 1995), p. 68.

7. Josh Weston et al., "Who Will Pay for Your Retirement?" Committee for Economic Development, Washington, 1995, p. 32.

8. Neil Gilbert and Neung-Hoo Park, "Privatization, Provision, and Targeting: Trends and Policy Implications for Social Security in the United States," International Social Security Review 49 (January 1996): 22.

9. U.S. House of Representatives, Committee on Ways and Means, 1994 Green Book, Table 1.7, p. 16.

010. A. Haeworth Robertson, Social Security: What Every Taxpayer Should Know (Washington: Retirement Policy Institute, 1992) p. 218.

11. U.S. House of Representatives, Committee on Ways and Means, 1994 Green Book, Chart H-1, p. 1159.

12. Shipman, Figure 3, p. 4.

13. The figures given are for full privatization. There are currently before Congress several bills that would allow private investment of a portion of Social Security taxes. Those bills represent an important step forward. However, they would provide a smaller boost to the poor than would full privatization, since the returns on investment would be far smaller than those cited here.

14. Stephen Entin, "Social Security: Problems and Opportunity," Institute for Research on Economics of Taxation, Washington, June 19, 1995, p. 3.

15. Ibid.

16. See, for example, Harriet Duleep, "Measuring the Effect of Income on Adult Mortality Using Longitudinal Administrative Record Data," Journal of Human Resources 21 (Spring 1986): 238-51; Evelyn Kitagawa and Philip Hauser, Differences in Mortality in the United States: A Study in Socioeconomic Epidemiology (Cambridge, Mass.: Harvard University Press, 1973); Eugene Rogot, Paul Sorlie, and Norman Johnson, "Life Expectancy by Employment Status, Income, and Education in the National Longitudinal Mortality Study," Public Health Reports 107 (July-August 1992): 457-61; and Howard Iams and John McCoy, "Predictors of Mortality among Newly Retired Workers," Social Security Bulletin 54 (March 1991): 2-10.

17. Henry Aaron, Demographic Effects on the Equity of Social Security Benefits (Washington: Brookings Institution, 1979).

18. Personal note from Henry Aaron, April 10, 1996.

19. C. Eugene Steuerle and Jon Bakija, Retooling Social Security for the 21st Century: Right and Wrong Approaches to Reform (Washington: Urban Institute, 1994), pp. 115-19.

20. Daniel Garrett, "The Effects of Differential Mortality Rates on the Progressivity of Social Security," Economic Inquiry 33 (July 1995): 457-75.

21. Ibid., p. 457.

22. Steuerle and Bakija, p. 119; Garrett, p. 458.

23. One of the earliest to write about that problem was Milton Friedman. See Milton Friedman and Wilbur Cohen, Social Security: Universal or Selective? (Washington: American Enterprise Institute, 1972), pp. 26-27.

24. Constantijn W. A. Panis and Lee Lillard, "Socioeconomic Differentials in the Return to Social Security," RAND Corporation Working Paper Series no. 96-05, February 1996, p. 20.

25. Ibid.

26. Ibid., p. 14.

27. Ibid., p. 20. The study also found a lifetime transfer of $50,000 from men to women.

28. See, for example, Iams and McCoy, p. 8.

29. Peter Ferrara, Social Security: The Inherent Contradiction (Washington: Cato Institute, 1980), pp. 229-30. That person is also likely to be an African-American. Indeed, the average life expectancy for an African-American man today is just 65.8 years.

30. Organization for Economic Cooperation and Development, National Accounts, Main Aggregates 1960-1992 (Paris: OECD, 1993), vol. 1, pp. 8-9.

31. Congressional Budget Office, "Assessing the Decline in the National Savings Rate," April 1993, p. 2.

32. Ibid., p. 3.

33. Internal Revenue Code, Secs. 3101 and 3102.

34. Many economists also argue that Social Security benefits reduce savings. Since Social Security provides beneficiaries with retirement income, people feel less need to save for retirement. Martin Feldstein, "Social Security, Induced Retirement, and Aggregate Capital Accumulation," Journal of Political Economy 82, no. 5 (September-October 1974): 905-26; Martin Feldstein, "Inflation, Tax Rules and Investment: Some Econometric Evidence," Econometrica 50, no. 4 (July 1982): 825-62; and Peter Diamond and J. A. Hausman, "Individual Retirement and Savings Behavior," Journal of Public Economics 23, nos. 1-2 (February-March 1984): 81-114. The reduced-savings effect would apply equally to the privatized system described above.

35. 1995 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Trust Funds, Table II.F1, p. 77.

36. That amount was $60.7 billion in 1994. Ibid., p. 40.

37. For a detailed explanation, see Weinberger, pp. 4-5.

38. Of course, the federal government has no cash or other assets with which to pay off those bonds. It can only obtain the cash by borrowing and running a bigger deficit, increasing taxes, or cutting other government spending.

39. Cited in Dick Armey, "The GOP's Fiscal Challenge," Washington Post, October 25, 1995.

40. Stephen Moore and John Silvia, "The ABCs of the Capital Gains Tax," Cato Institute Policy Analysis no. 242, October 4, 1995, p. 34.

41. Jude Wanniski, Testimony before the Senate Finance Committee, February 15, 1995.

42. Martin Feldstein, "Social Insurance," Harvard Institute of Economic Research Discussion Paper no. 477, May 1976, pp. 30-35.

43. Martin Feldstein, "Social Security and the Distribution of Wealth," Journal of the American Statistical Association, December 1976, p. 65.

44. "From Safety Net to Golden Hammock: Social Security Benefits Distributed across Congressional Districts," Third Millennium, Washington, August 3, 1995.

45. Ibid.

46. Bureau of the Census, Current Population Reports, Series P-60, March 1996, Table F-1.

47. Martin Feldstein, "The Missing Piece in Policy Analysis: Social Security Reform," American Economic Review 86 (May 1996): 12.

48. Robert Genetski, "Privatize Social Security," Wall Street Journal, May 21, 1993, p. A10.

49. Luis Larrain, "Social Security Reform," in The Chilean Experience: Private Solutions to Public Problems, ed. Cristian Larroulet (Santiago: Center for International Private Enterprise, 1991), p. 45.

50. José Piñera and Mark Klugmann, "The Chilean Private Pension System," International Center for Pension Reform, Santiago, 1995.

51. There is one important caveat, however, in considering how privatization of Social Security would increase the U.S. savings rate. Any additional government debt used to finance the transition to a private system will offset a portion of the increased private savings.

52. José Piñera, "Empowering Workers: The Privatization of Social Security in Chile," Cato's Letter no. 10, April 1996, p. 17.

53. U.S. House of Representatives, Committee on Ways and Means, "Background Materials on the Federal Budget and Tax Policy for the Fiscal Year 1991 and Beyond," February 6, 1990, p. 24.

54. Congressional Budget Office, "Aggregate Economic Effects of Changes in Social Security Taxes," August 1982, p. 30.

55. Aldona Robbins and Gary Robbins, "Effect of the 1988 and 1990 Social Security Tax Increases," Institute for Research on the Economics of Taxation, Washington, 1991, pp. 14-15.

56. 1995 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds, Table II.b.1, pp. 35-36.

57. Ibid. p. 19. Note that the entire FICA payroll tax would increase from 15.3 percent to nearly 20 percent. The FICA tax and the various trust funds it supports are often confusing. The tax is currently 15.3 percent divided equally between employer and employee. While generally considered a single tax, the payroll tax actually funds several different trust funds: (1) The Old-Age and Survivors Insurance (OASI) Trust Fund is used to pay monthly benefits to retired workers and their spouses and minor children and to survivors of deceased workers. Those are the benefits most commonly thought of as Social Security. That portion of the tax currently amounts to 10.52 percent. (2) The Disability Insurance (DI) Trust Fund is used to pay disabled workers and their spouses and minor children and to provide rehabilitation services to the disabled. The DI Trust Fund is often linked to the OASI Trust Fund as OASDI to distinguish it from the Hospital Insurance Trust Fund. The payroll tax for support of OASDI is 12.4 percent. (3) The Hospital Insurance (HI) Trust Fund is used to pay for hospital costs under Medicare Part A. That trust fund was established in 1965 and is supported by 2.9 percent of the payroll tax.

58. Ibid.

59. Bureau of the Census, The American Almanac: Statistical Abstract of the United States, 1995 (Austin: Reference Press, 1995), Table 114, p. 86.