Privatizing Social Security: The Troubling Trade-Offs

Policymakers and the public show growing interest in the idea of replacing Social Security with a private system of individual retirement accounts. It is too soon to tell whether curiosity will lead to popular acceptance, but it is already plain that many Americans are thinking about private alternatives to Social Security for the first time.

Proponents of privatization see three main arguments, in addition to ideological advantages, for moving toward a private retirement system:

  • it can lift the rate of return workers obtain on their retirement contributions;
  • it can boost national saving and future economic growth;
  • it has practical political advantages in comparison with a Social Security rescue plan based on higher payroll taxes and a bigger accumulation of Social Security reserves.

Any transition to a private system must overcome a major financial hurdle, however. Social Security has accumulated trillions of dollars in liabilities to workers who are already retired or who will retire soon. To make room for a new private system, policymakers must find funds to pay for these liabilities while still leaving young workers enough money to deposit in new private accounts. This requires scaling back past liabilities – by cutting benefits – or increasing contributions from current workers. Most large-scale privatization plans also involve major new federal borrowing. Consequently, if a balanced budget amendment becomes part of the constitution, it would torpedo any attempt to replace most of Social Security with a private retirement system.

Privatizing Social Security can boost workers’ rate of return by allowing retirement contributions to be invested in private assets, such as stocks, which yield a better return than the present pay-as-you-go retirement system. Returns can be boosted still further if the government borrows on a massive scale to pay for past Social Security liabilities, allowing workers to invest a larger percentage of their pay in high-yielding assets. Exactly the same rate of return can be obtained, however, if the current public system is changed to allow Social Security reserves to be invested in private assets.

By shifting the retirement system away from pay-as-you-go financing and toward advance funding, privatization can boost national saving. Such a move will require a consumption sacrifice, through either a cut in benefits or a hike in combined contributions to Social Security and the new retirement plan. Privatization plans that do not impose a consumption sacrifice will not achieve a higher saving rate. Higher national saving can also be achieved by reforming the present public retirement system. The crucial change in policy is the move to advance funding, not the move to a private system.

The claimed economic advantages of privatization can be obtained in either a public or a private retirement system. In either case, a short-term consumption sacrifice is needed. The best political argument for privatization is that the consumption sacrifice will be more palatable if workers are given ownership rights over the extra contributions they will be forced to make.


How a Private Sytem Should Work

Privatization plans are based on a simple idea. Instead of contributing to a collective, pay-as-you-go retirement program, workers would be required to build up retirement savings in individually owned and directed private accounts. Workers could withdraw their funds from the accounts when they became disabled or reached the retirement age, and their heirs could inherit any funds accumulated in the account if the worker died before becoming disabled or reaching the retirement age. At the time a worker chose to start receiving a pension, some or all of the funds in the worker’s account would be converted into an annuity that would last until the worker died. If the worker is married, both worker and spouse might be required to accept a joint survivors’ annuity, that is, an annuity that would last until both the worker and spouse have died. Under some proposals, workers could choose to withdraw some of the funds as a lump-sum distribution when they become disabled or retire.

Workers would be free to decide how their contributions were invested, at least within broad limits. In some privatization plans, contributions would be collected by a single public or semi-public agency and then invested in one or more of a limited number of investment funds. A worker might be given the option of investing in, say, five different funds – a money market fund, a stock market index fund, a real estate investment trust, a corporate bond fund, and a U.S. Treasury bond fund. By pooling the investments of all covered workers in a small number of funds and centralizing the collection of contributions and funds management, this approach would minimize administrative costs, but it would limit workers’ investment choices. Another strategy is to allow mutual fund companies, private banks, insurance companies, and other investment companies to compete with one another to attract workers’ contributions in hundreds or even thousands of qualified investment funds. This strategy would permit workers unparalleled freedom to invest as they chose, but administrative costs might be high.

Privatization has attracted growing interest because many public retirement systems around the world have encountered serious financial difficulties. In Chile, a costly and failing public system was replaced by a less costly private system that has so far been quite successful. The United Kingdom has also shifted toward much greater reliance on private pensions. Advocates of privatization in the United States have proposed several plans for moving to a partially or fully private retirement system. Two plans were outlined in the just-published report of the 1994-1996 Advisory Council on Social Security. A majority of Advisory Council members voted to scale back defined-benefit Social Security pensions and require workers to contribute a percentage of their pay to private defined-contribution retirement accounts. Under one plan, workers would be required to contribute 1.6 percent of their covered wages to publicly managed but individually selected retirement accounts. The 1.6 percent contribution would be on top of the 12.4 percent combined tax that workers and employers already pay into Social Security. Under the more ambitious privatization plan, the overall contribution would rise by 1.5 percent of covered wages, but 5 percentage points of the higher tax would be redirected into individually selected and privately managed personal security accounts.

Privatization plans differ from Social Security in two important ways. First, the worker’s ultimate retirement benefit would depend solely on the size of the worker’s contributions and the success of the worker’s investment plan. Workers who made larger contributions would receive bigger pensions, other things equal. Workers whose investments earned high returns would enjoy more comfortable retirements than workers who invest poorly. Second, in a private system workers’ pensions are paid out of accumulations of their own previous savings. In contrast, Social Security pensions are financed mainly by the payroll taxes of active workers (see box 1). This difference between the two kinds of system implies that the savings accumulation in a private plan would be many times larger than the reserves needed in pay-as-you-go Social Security.

Because the connection between individual contributions, investment returns, and pension benefits is very straightforward in a defined-contribution pension program, a private retirement system offers less scope for redistribution in favor of low-wage workers. Redistribution in favor of low-wage or other kinds of workers must take place outside these accounts. The Social Security pension formula explicitly favors low-wage workers and one-earner married couples in order to minimize poverty among elderly people who have worked for a full career. To duplicate Social Security’s success in keeping down poverty among the elderly, a private system must supplement the pensions from individual retirement accounts with a minimum, tax-financed pension or with public assistance payments.

box 1: The Social Security System

Social Security, established in 1935, is the most costly item in the federal budget. After the income tax, it also provides the largest source of tax revenues. The program provides old-age, survivors, and disability insurance protection to most people who work in wage and salary jobs and to the self-employed. Workers and their employers each pay taxes equal to 6.2 percent of covered wages. The self-employed pay 12.4 percent of taxable self-employment earnings. Taxes are collected by the Internal Revenue Service and deposited in government-administered accounts called the Old-Age and Survivors and Disability Insurance (OASDI) Trust Funds.

Payroll tax revenues are used to pay benefits to people who are currently collecting Social Security pensions. (This method of financing is referred to as pay-as-you-go funding.) Any excess of taxes over benefit payments is invested in special U.S. Treasury bonds, which earn the average rate of return on publicly traded government debt. In the year ending September 1996 tax revenues totaled $380 billion, while administration and benefit costs were $350 billion, leaving a difference of $30 billion (or 1 percent of taxable wages) to be invested in Treasury bonds. In addition, interest payments totaled $36½ billion, providing the Social Security system with an overall annual surplus of about $66 billion.

Workers who accumulate enough earnings credits become eligible to receive a Social Security pension when they attain the early retirement age (62) or become too disabled to continue working, no matter what their age. A worker’s dependent spouse and non-adult children may draw monthly survivors’ pensions when the worker dies. Social Security is a defined-benefit pension program. That is, each worker’s pension depends on his or her average career earnings and on the age when the worker or worker’s dependents first obtain a pension. The exact amount of the pension is determined by a formula that is enshrined in law and is updated every year to reflect changes in economy-wide wages and consumer prices.

The Social Security benefit formula provides substantial redistribution in favor of workers with low career earnings, workers who face an abnormal risk of becoming totally disabled, and married couples with only a single earner. The generous treatment of these groups is affordable because high-wage workers, unmarried and childless workers, and dual-income married couples receive less favorable treatment under the system.

Transition to a Private System

The United States cannot immediately scrap its public retirement system and replace it with a private system. At the end of 1996, more than 43½ million Americans were collecting benefits under Social Security. About 1.6 million workers began to collect new retirement benefits during the year, and another 600,000 were awarded new disability pensions. Even if the country adopted a private system for young workers, people who are already retired or planning to retire within the next few years would continue to receive Social Security checks for several decades. Public funds must be appropriated to pay for these pensions, regardless of the system established for workers who will retire in the distant future.

The need to pay for the pensions of people who are already retired or near retirement age poses a challenge to all plans for privatizing Social Security. Money must be found for existing pension liabilities at the same time workers will be asked to contribute to a new type of private pension account. Because active workers will be required to finance pensions for retired workers and old workers nearing retirement, they may resent the obligation to pay for their own retirement pensions through contributions to new private accounts.

Some privatization plans would fund new retirement accounts by diverting a small part of the present payroll tax into private retirement accounts. In 1996, OASDI benefit payments exceeded Social Security tax revenues by $30 billion, or 1 percent of taxable earnings (see box 1). Thus, 1 percent of the 12.4 percent payroll tax could be invested in individual retirement accounts while still leaving enough taxes to pay for current pension payments. This source of financing for private accounts will not last forever, however. Even if workers under age 40 are completely excluded from collecting Social Security pensions, benefit payments will exceed Social Security taxes by around 2015 (see figure 1). Thus the strategy of diverting a small part of Social Security taxes can work only if current benefits are scaled back. In addition, workers must contribute much more than 1 percent of their wages if they hope to accumulate enough private savings to enjoy a comfortable retirement.

More ambitious privatization plans would divert half or more of the present Social Security payroll tax into private retirement accounts and slash Social Security payments available to young workers (for example, those under age forty-five). These plans would require borrowing or new federal taxes to pay for existing Social Security liabilities. The diversion of payroll taxes would starve the Social Security system of revenue, forcing the program to run huge deficits. To cover these deficits Congress would be forced to raise taxes or borrow funds. The need for extra taxes or borrowing would eventually shrink as pensioners collecting Social Security were replaced by pensioners who received benefits from the new private accounts, but this process would not be complete for several decades. In the interim, the federal government would need to impose extra taxes (temporarily replacing most of the lost Social Security taxes) or sell a large amount of additional public debt.

Arguments For Privatization

Any discussion of reform should begin by recognizing that the current retirement system is already a mixture of public and private plans. The public plan is universal but skewed toward protecting low-wage workers. Private or employer-sponsored plans cover about half the full-time work force, but they tend to leave part-time and lower-wage workers uncovered.

Advocates of privatization see a number of advantages in increasing the size of the private system and shrinking the size of the public one. For some proponents of privatization, ideological concerns are paramount. They are fundamentally opposed to public provision of retirement benefits. More common are people who see important economic advantages in privatizing Social Security. They believe workers will receive larger pensions and the economy will grow faster under a private rather than a public retirement system. Finally, some advocates of privatization believe the United States is more likely to take needed steps to prepare for a rising aged population if the retirement system is reformed to include a bigger private role.


A few critics of Social Security, who are particularly distrustful of public intervention, believe it is an unwarranted intrusion on personal freedom to require workers to contribute a fixed percentage of their pay to a retirement plan. They believe individual workers can make better judgments than public officials about the proper division of earnings between consumption during a worker’s career and savings for the worker’s retirement. Libertarians who hold this view oppose all mandatory saving schemes for retirement, whether or not the retirement funds are placed in private accounts.

Most advocates of privatization acknowledge that it makes sense to compel workers to save for old age, disability, and early death. In the absence of mandatory saving, many workers would save too little and could become destitute and be forced to rely on public aid when they stop working. Even if government authority is used to force workers to save for old age, however, it does not follow that government officials should be entrusted with the job of managing workers’ retirement funds. Most privatization advocates believe that decisions about the investment of retirement saving are best left up to workers and their employers.

Higher returns

Few voters share the ideological concerns of Social Security’s harshest critics. If they did, it would be hard to explain the program’s enormous popularity over the past few decades. Most voters’ opinions about the program are driven by their concern for people who collect Social Security and their practical interest in the affordability of contributions, the adequacy of benefits, and the sustainability of the present system. Nearly all advocates of privatization try to appeal to these interests. They argue that pension contributions would be more affordable or benefits more generous if the nation moved toward a private retirement system. Stated crassly, most workers could expect a better deal under a private system than they can obtain under Social Security.

This argument is based on a straightforward calculation. If workers invested 12.4 percent of their earnings in a private retirement account yielding a moderate rate of return (say, 3 percent a year after adjusting for inflation), most would collect larger pensions than they can expect under the present Social Security system.

Figure 2 shows the Social Security Actuary’s estimates of the average rate of return under Social Security for several kinds of workers. Estimates are displayed for workers born in a variety of years who earn steady wages at three different earnings levels. The low-wage worker is assumed to earn roughly the minimum wage; the average-wage worker earns the average covered earnings under Social Security; and the high-wage worker earns about two-thirds of the maximum taxable wage. (In 1995 these annual earnings amounts were equivalent to about $11,000, $25,000, and $40,000, respectively.) In performing the calculations, the Actuary converted each worker’s tax and benefit payments to constant, or inflation-adjusted, dollars using the actual or predicted Consumer Price Index. The Actuary then computed the interest rate that would be required so that the discounted value of real tax payments would be exactly equal to the discounted value of real benefit payments. This interest rate is the internal real rate of return on the worker’s tax contributions.

Two facts about Social Security stand out in figure 2. Low-wage workers get a better deal than high-wage workers, and workers born before 1930 get a much better deal than workers born later. The comparatively high rate of return enjoyed by low-wage workers is the result of Social Security’s redistributional benefit formula, which favors people who earn low lifetime wages.

Even more striking than the disparity between low-wage and high-wage workers is the difference in returns enjoyed by people born before and after 1930. The high return enjoyed by workers born before 1930 helps explain the current popularity of Social Security, especially among older voters. In particular, workers born around 1920 enjoyed two pieces of good fortune. When they entered the work force in the early 1940s, the combined employer-employee tax rate was just 2 percent. The tax rate was low throughout much of this generation’s early career because little revenue was needed to pay for Social Security benefits, since few retired workers yet qualified for a full pension. More recently, the tax rate has been raised to cover much more generous pensions to a far larger number of retirees. Workers born in 1955, for example, faced a combined contribution rate of 11.7 percent when they entered the work force.

People born in 1920 were also fortunate in enjoying rapid wage growth throughout most of their careers. Real annual earnings climbed 2 percent a year between 1940 and 1980, rising about 130 percent over four decades. Under the Social Security formula, each generation’s average pension is determined by the economy-wide average wage when it nears the retirement age. Rapid growth in real wages produces a good rate of return in a pay-as-you-go pension system.

Unfortunately, real wage growth slowed dramatically in the mid-1970s. The economy-wide average wage did not grow at all in the twenty years from 1973 to 1993. Slow wage growth will be reflected in slow growth in pensions and a lower rate of return on Social Security contributions for future generations of retirees. In fact, because taxes will have to be hiked or benefits cut to keep Social Security solvent, workers born after 1980 will probably receive lower returns than those shown in figure 2.

In order to bolster political support for the nation’s retirement system, advocates of privatization believe it is necessary to improve the returns young workers will earn on their contributions. A rapid rise in the number of workers collecting pensions after 2010 combined with slow labor force growth and anemic wage improvements mean that a pay-as-you-go retirement system must offer low or negative returns to many of today’s young workers. Privatization can raise expected returns by placing some or all of workers’ pension contributions in stocks, bonds, real estate, or other private assets that offer attractive rates of return.

Increased saving

Many advocates of privatization believe that full or partial privatization will boost U.S. saving rates. The current saving rate is low by historical standards, leaving only a small percentage of the nation’s income for investment in domestic and overseas capital. The low rate of capital accumulation contributes to the slow growth of national income and wages. If saving could be increased, income growth would accelerate, making it easier for the nation to afford the extra burden of supporting a large retired population in the next century.

Unlike the present Social Security system, which is largely financed on a pay-as-you-go basis, a private retirement system would involve huge accumulations of assets in individual retirement accounts. Because workers would be setting aside a percentage of their pay in private accounts for their own retirement instead of sending in contributions that are immediately spent on pension payments, the introduction of a privatized system could lead to a jump in saving.

Political feasibility

In theory, national saving can be raised within the existing Social Security system, even if there is no move toward privatization. This could occur if Congress raised the present contribution rate or reduced benefits, increasing the annual surplus of the program. The Social Security trust funds would accumulate larger reserves than are anticipated under current law. Instead of accumulating assets in tens of millions of individual retirement accounts, as in a private system, the saving would take place in a single public fund.

Advocates of privatization doubt, however, that the funds accumulated within a public fund would actually be saved. They fear Congress would use the funds to finance growing deficits in other government accounts, such as Medicare. In the absence of larger Social Security surpluses, the Congress would be forced to deal with the deficit in other programs, either by curtailing spending or by increasing taxes. A larger surplus in Social Security makes it easier for Congress to avoid this unpleasant choice. Privatization advocates therefore think it is safer for the accumulation to take place in tens of millions of privately owned accounts, outside the reach of a revenue-hungry Congress.

Privatization also offers a politically acceptable method of managing the accumulation of huge reserves and corporate stocks. In a system where the accumulation takes place in a single public system, legislators and public officials would be responsible for allocating the funds among investment alternatives and across individual companies. Their investment decisions might be guided by political rather than economic considerations, reducing the yield of the investments, diverting investments into unproductive uses, or intruding on the business decisions of company managers. In a private system of individual accounts, decisionmaking authority over the accumulation would rest on the shoulders of millions of workers. Through their choices among investment alternatives and specific investment funds, workers and private fund managers rather than public officials would exercise ultimate authority over the allocation of investments.

A private retirement system, with its broad dispersion of asset ownership, also has an advantage over a public retirement fund when it comes to accumulating corporate stocks. The U.S. government has traditionally achieved public goals by regulating businesses rather than by owning a percentage of their shares and exercising ownership rights in company elections. If retirement asset accumulation took place within a single public fund and if the public fund owned shares in thousands of companies, Congress or public trustees would have to decide how these shares should be voted. Voting decisions might be determined by political rather than economic criteria, possibly reducing the efficiency and profitability of American business.

Some advocates of privatization also maintain that voters would be more willing to accept an increase in their combined contribution to the retirement system if 100 percent (or more than 100 percent) of their extra contribution took the form of deposits into individually owned and managed investment accounts. While voters would reject a hike in the payroll tax, they will tolerate – and may actually welcome – compulsory saving in individually owned accounts. This argument for privatization is essentially pragmatic. Because the work force is growing older, it is important to raise national saving. Voters and Congress are more likely to take the steps needed to increase saving if workers have direct ownership of their extra contributions to the retirement system.

Weighing the Advantages of Privatization

Most public uneasiness about the future of Social Security results from workers’ fear that the present system is not sustainable. This fear is exaggerated but not completely unfounded. In order to pay promised Social Security benefits, the future contribution rate must be increased. Future voters might balk at paying higher taxes, and benefits would then have to be cut. The expected revenues of Social Security will fall short of expected benefit payouts by 14 percent over the next seventy-five years, a shortfall that is equivalent to 2.2 percent of taxable wages over the entire period. By 2030 benefit payments would need to be cut nearly one-quarter to keep the program solvent under the present payroll tax. This does not mean Social Security pensions must eventually be eliminated, as some young workers fear, but it does mean their taxes must be increased or their benefits cut if the system is to be preserved.

The two main economic arguments in favor of privatization are that it would increase real returns on pension contributions and boost national saving. Both arguments are valid, for some forms of privatization, assuming the public system to be dismantled is financed on a pay-as-you-go basis. There is no reason, however, that public retirement benefits must be supported with pay-as-you-go financing. The important distinction is between advance funding and pay-as-you-go financing, not between public and private management of investment funds. Millions of employees of state and local governments have advance-funded pensions, and their pension funds are publicly managed. Advance funding, combined with a more aggressive investment strategy, can offer a higher return to current and future contributors. The larger accumulation in pension funds, whether they are publicly or privately managed, can boost national saving.

Rate of return

In theory, the Social Security Trust Fund can obtain the same average rate of return that would be earned in individual workers’ private retirement accounts. If the reserve were invested in the same mix of assets that would be selected by workers, it would earn an identical rate of return. The net return would actually be somewhat higher, because the expense of maintaining a single public fund is considerably smaller than the cost of administering tens of millions of private accounts, many of which would be extremely small.

For Social Security to accumulate the same kinds of assets that workers would place in private retirement accounts, a change in Social Security investment strategy is needed. By law, Trust Fund reserves are invested in U.S. Treasury debt where they earn the rate of return on publicly held U.S. bonds. Workers seeking a high return on their retirement savings invest in other types of assets in addition to government securities. Based on the experiences of workers who invest in 401(k) pension plans, the Advisory Council estimates that 55 percent of the retirement savings of workers under age forty would be invested in equities. One of the three plans outlined by the Advisory Council proposed investing up to 40 percent of Trust Fund reserves in corporate stock, increasing the expected rate of return on reserves by about 1.9 percentage points to 4.2 percent.

The Social Security Actuary has calculated the rate of return that workers can anticipate under the current system and under alternative systems proposed by the Advisory Council. These calculations are helpful in understanding the potential gains from privatization and how they are achieved. Figure 3 shows the expected rate of return of an average-wage worker under two alternatives. One alternative assumes workers will continue to receive Social Security benefits under the present benefit formula but that taxes will eventually be raised (starting in 2025) to ensure that the OASDI Trust Funds are never depleted. This strategy keeps Social Security solvent, but it reduces the rate of return received by younger workers, because they must make larger contributions to obtain the same amount of benefits. Figure 3 shows that the rate of return under this policy will decline continuously for workers born in successive generations. Average-wage workers born in 2004 will typically receive a return of just 1.7 percent.

The second alternative assumes that 5 percent of the present payroll tax is diverted into private retirement accounts; an extra 1.5 percent payroll tax is imposed to help pay off Social Security’s past liabilities; and the remaining Social Security program is converted into a modest flat-rate pension plan that can be financed with a tax rate of just 7.4 percent. In 1996, this pension would have been $410 a month (or 65 percent of the poverty line) for a full-career worker, compared with an average retirement pension of $724 under the present system and a pension of $546 for a newly retired low-wage worker. The Actuary assumes that almost half of the funds in private retirement accounts will be invested in stocks and that stocks will yield an annual real return of 9.3 percent. The rate of return under these assumptions is predicted to fall for workers born in the 1940s and 1950s but then to rise above 3½ percent for workers born after the mid-1970s. For workers born in 2004, the rate of return under the partially privatized system is predicted to exceed the return under a solvent OASDI system by 2 percentage points.

Although an average worker’s return is higher under partial privatization than it is under a solvent Social Security system, it is far less than 9.3 percent, which is the assumed return on stocks. There are four main reasons for the difference. First, a private system will absorb some worker contributions for administration of the individual accounts. In addition, workers will not invest all their contributions in the stock market, preferring instead to invest some funds in less risky assets, like government bonds, where yields are lower. Risk-averse workers may invest much less than half their funds in stocks, especially as they near retirement. For middle-aged workers, returns will also be low because much of their retirement income will come from scaled-back Social Security benefits. As these benefits are reduced (for example, by raising the age of entitlement for full pensions), workers will be forced to accept a lower rate of return on their past Social Security contributions. Finally, workers will be required to contribute 1.5 percent of their pay to cover the past liabilities of Social Security. They receive no return on these contributions.

For average-wage workers born in the 1940s and 1950s, the transition to the private system actually requires workers to accept a reduction in the rate of return they can anticipate under the present system. Workers born in 1949, for example, would earn 0.2 percent less on their contributions than they would receive under the present system. The financial advantages of privatization only become sizable for workers born in the 1970s or later, who earn good returns on the privately invested part of their contributions throughout most of their careers. The higher returns of future workers are achieved at the expense of middle-aged and older workers, who would be asked to accept smaller Social Security pensions than promised by current law. By accepting smaller pensions, middle-aged and older workers make it possible to pay for the transition to a private system with a supplemental payroll tax that is just 1.5 percent. If they received full Social Security pensions, the supplemental payroll tax would be higher and the returns enjoyed by younger workers would be smaller than shown in figure 3.

The privatization plan described above requires substantial federal borrowing over a transition period that lasts about three decades. The Actuary estimates this would require the Treasury to issue about $2 trillion in extra debt, an amount equal to slightly more than 20 percent of national income at the point of peak borrowing. (The current publicly held Treasury debt is about 50 percent of GDP.) The Treasury is assumed to pay a real interest rate of 2.3 percent to borrow these funds. The assumed interest rate is important for two reasons. A higher interest rate increases the supplemental payroll tax needed to pay for the past liabilities of Social Security, which makes the private plan appear less attractive. At the same time, it increases the interest earnings of the OASDI Trust Funds under the current system and postpones the year that payroll taxes must be raised, which improves the rate of return enjoyed by workers under the present system. The more costly it is for the federal government to borrow from the public, the less attractive moving toward the private retirement system will appear.

Most privatization plans, like the one just described, involve four basic elements: a promise to retirees and older workers to pay all or most of the Social Security benefits they have earned; a cut in benefits to younger workers; a diversion of Social Security payroll taxes for younger workers into private investment accounts; and increased federal borrowing to offset the diversion of taxes into private accounts. Because younger workers are assumed to place their contributions in high-yield investments, they earn a better return than they could obtain in a mature pay-as-you-go retirement system. Their returns are higher because the federal government can borrow funds at a low interest rate while workers can invest the funds at a high rate. In essence, workers’ well-being has improved because they can borrow (through the U.S. government) at a 2.3 percent interest rate and earn returns that exceed 5 percent a year. If this magic works when 5 percentage points of the payroll tax is diverted into individual accounts, it is natural to ask why the diversion should be limited to 5 percent. Why not divert the entire 12.4 percent tax into individual accounts? Under the assumptions used by the Actuary, the returns earned by young workers would be even better than those shown in figure 3.

Many proponents of privatization hesitate to recommend diversion of the entire payroll tax into private accounts for two reasons. They wish to retain at least a small public system to redistribute some extra benefits to low-wage workers and protect them against abject poverty. Any part of the payroll tax that is retained for this purpose will reduce the returns enjoyed by average – and above-average – wage workers. Many who favor privatization also recognize that voters might reject any plan that adds massively to public borrowing. In fact, a balanced budget amendment to the constitution would probably make it unconstitutional to implement ambitious privatization plans (see box 2). In 1996 the OASDI payroll tax raised $360 billion. If the entire payroll tax had been diverted into private accounts, the federal deficit would have soared to $468 billion. In later years additions to the deficit would be even greater, because higher interest payments on a larger debt would increase federal spending requirements still further.

Voters’ reluctance to pursue this borrowing strategy makes good economic sense. We should be skeptical of suggestions that the government could issue trillions of dollars of new debt and allow the proceeds to be invested in equities without raising the costs of borrowing or reducing the return on equities. If the national debt were expected to climb enormously over the next few decades, investors who purchased government debt would probably demand a higher rate of return than 2.3 percent a year. As we have seen, a higher federal borrowing rate makes most privatization plans less attractive.

The predicted high returns of this financing proposal have little to do with privatization. If the public system borrows the same amount of money at the U.S. Treasury interest rate and follows the same investment strategy that individual workers would have pursued with the same funds, overall returns under the public system could be the same as or slightly higher than those in the private system. The more money that is borrowed at a low predicted interest rate and invested at a high assumed rate of return, the better the predicted returns under the system, whether the investment funds are managed publicly or privately.


Most versions of the balanced budget amendment require 60 percent majorities in both Houses of Congress if a budget is to be passed in which total federal outlays exceed total receipts. If this provision is interpreted literally, it would prohibit privatization of Social Security unless privatization would be accomplished without large-scale government borrowing. Alternatively, the House and the Senate would need to assemble large majorities for many successive years to support the annual borrowing that would allow privatization to take place. As a practical matter, these requirements make it unlikely that large-scale privatization could occur. Of course, a future Congress might redefine “receipts” and “outlays” to allow borrowing for privatization without 60 percent majorities. If it did so, however, Congress would be implicitly acknowledging that a balanced budget amendment has no practical effect on fiscal policy.

The balanced budget amendment would also make it more difficult, though not impossible, to invest part of the Trust Fund in corporate equities. Such investments are treated as a government expenditure under current budget rules. Dividends and income produced by sales of corporate stock are treated as government receipts. Buying corporate stock requires that the Trust Funds purchase less government debt, so the Treasury would be forced to sell more debt to the public. This could occur under the balanced budget amendment only if there were a surplus in the rest of the federal budget or if 60 percent majorities in both Houses approved.

Some Members of Congress propose to protect the Social Security system by adding a provision to the balanced budget amendment that specifically excludes OASDI receipts and outlays from calculations of total federal receipts and outlays. This provision would have the ironic effect of making Social Security privatization much easier to accomplish. Even without 60 percent legislative majorities, the OASDI Trust Funds would be permitted to borrow the funds needed to finance the transition to a private system.

Effects on saving

Well-informed proponents of privatization recognize that the transition to a private retirement system may not produce higher saving. For example, if workers are given a rebate of their Social Security taxes in order to fund their new individual retirement accounts, the Social Security system will be deprived of revenues that are needed to pay current pension obligations. Instead of enjoying an operating surplus, Social Security will then require a substantial infusion of funds from the Treasury, forcing the federal government to raise taxes, to reduce other spending, or to borrow extra money.

If the government borrows all the extra money, as some advocates of privatization urge, the policy could easily reduce national saving. To see why, it is important to understand the components of saving. The flow of national saving in a given year is the sum of saving that takes place in the private sector plus saving in the government. Federal government saving is the sum of saving in the Social Security system plus the surplus or deficit in non-Social Security operations. In 1996, for example, Social Security had a surplus of $66 billion, while the remainder of the federal budget had a deficit of $174 billion. The overall federal deficit was $108 billion.

If all the Social Security surplus were diverted into individual retirement accounts, the federal budget deficit would jump $66 billion, reducing saving in the government sector and forcing the government to sell more bonds. At the same time, the flow of funds into workers’ new retirement accounts would increase saving in the private sector. If private saving rose by the full $66 billion, national saving (which is the sum of government and private saving) would be unchanged. But it is unlikely that private saving would grow by $66 billion. Many workers already have compulsory or voluntary retirement plans connected to their jobs. Some of these plans, such as IRA, 401(k), or Keogh plans, are almost indistinguishable from the new compulsory retirement accounts that would be established in privatization. At least a few workers would reduce their contributions to existing IRA, 401(k), or Keogh plans if they were forced to save in new government-mandated accounts. Any reduction in the flow of saving into old retirement accounts would offset part of the effect of the flow of saving into the new retirement accounts. Private saving would then climb by less than $66 billion.

In order to boost national saving, a privatization plan must reduce someone’s consumption. The plan could reduce the pensions – and thus the consumption – of the people who are already retired or who will soon retire. Alternatively, it could increase the combined contributions that workers make to Social Security and private retirement accounts and thereby reduce their consumption. For example, if benefit payments in 1996 had been trimmed $10 billion (or about 3 percent), the Social Security surplus would have been $10 billion larger and the federal deficit $10 billion smaller. National saving would have been higher. If workers were required to contribute an additional $10 billion of their pay to Social Security and new retirement accounts, the revenue flowing into Social Security and the new retirement accounts would have been $10 billion larger than last year’s payroll tax revenue. This policy would raise national saving, too.

Some privatization plans have a good chance of boosting saving. The Advisory Council plan for reducing Social Security benefits and forcing workers to invest 1.6 percent of their pay in new retirement accounts would reduce the federal deficit and increase workers’ private retirement saving. It would almost certainly raise national saving. The alternative Advisory Council plan for diverting 5 percent of payroll taxes into private accounts would raise national saving in two ways. It would cut Social Security benefits promised to workers who will retire starting early in the next century. This will be accomplished by hiking the normal and early retirement ages for collecting Old-Age benefits and by reducing Disability Insurance benefits. These steps increase public saving by cutting future public expenditure. The plan also introduces a new payroll tax of 1.5 percent that is scheduled to expire in seventy-two years. The tax boosts federal revenues, another contribution to higher public saving.

Lower Social Security benefits and the new payroll tax are not enough to offset the 5 percent diversion of payroll taxes, however, so the federal government will be forced to borrow funds from private savers over the next several decades. The plan is nonetheless likely to raise total saving, because the government will borrow less money from savers than workers will be forced to deposit in their private retirement accounts. If workers do not reduce their saving in other retirement accounts, private saving will increase by 5 percent of taxable wages, while public saving will decline by less than 5 percent of taxable wages. Total saving should increase.

The same increase in national saving could also be achieved, of course, if the payroll tax were raised 1.5 percent and Social Security benefits temporarily reduced in the existing public system. Privatization is not needed to achieve a higher national saving rate. What is needed is a consumption sacrifice on the part of current retirees, workers who will soon retire, or workers who will retire in the distant future. Current retirees can be forced to make a consumption sacrifice if their Social Security benefits are cut, for example, by cutting the annual cost-of-living adjustment. Workers who will retire in the future can be forced to make a consumption sacrifice if payroll taxes are increased or if future benefits are trimmed, for example, by raising the retirement age. Either option will increase saving within the Social Security system and hence potentially boost national saving. The contribution of a retirement plan to national saving does not depend on whether the plan is public or private. It depends on the pattern of consumption sacrifice the plan requires over the next few decades.

Political realities

Increasing national saving is highly desirable. It is a practical way to increase the total amount of future income out of which the consumption needs of both future workers and future retirees will be financed. A straightforward policy to achieve this goal is to increase the saving of current workers so that they fund in advance a bigger percentage of their own pensions. Advocates of privatization sometimes argue that active workers are more likely to agree to such a plan and the consumption sacrifice it entails if they have ownership rights over the extra retirement contributions they are forced to make. If this reasoning is correct, it is a powerful argument in favor of privatization.

Some privatization plans will have little effect on national saving, however, and others would actually reduce saving. Under these circumstances there are many reasons to prefer the existing retirement system to an alternative system that involves a larger private role. The current system has been hugely successful in reducing poverty among the elderly. A scaled-back public system would have less scope for providing generous pensions to low-wage workers, unless the benefit formula is tilted more heavily in favor of low-income recipients. A stronger redistributional tilt in the benefit formula would make it even less favorable to average-wage and high-wage workers, which could undermine political support for the remaining public component of the retirement system. There is a danger that the explicit separation of the redistributional component of the retirement system into a small but highly visible public program could result in strong pressure to reduce or eliminate it.

Private accounts also raise the problem of how to manage the conversion of the accounts into annuities when workers retire or become disabled. If individuals are given the option of accepting lump-sum cash-outs, a private system for converting assets into annuities will encounter the problem of adverse selection. The price of annuities would have to rise, meaning that larger accumulations would be needed to pay for the same monthly benefits. This is not a problem faced by a public system of defined-benefit pensions, in which there is universal and mandatory conversion to annuities. Privatization also involves higher administrative costs. No private insurance or investment management company comes close to matching Social Security’s low operational costs.

Of course, many of the advantages of the current public system would be retained under a reform that moves only a small distance toward full privatization. As the size of the public system shrinks, however, it will become more difficult to achieve some of the important redistributional goals served by the present system.


The debate over reforming or replacing Social Security should not begin with exaggerated claims about the imminent “bankruptcy” of the program, but with a rational assessment of the goals of a sensible retirement system and the success of Social Security in achieving them. The program has been notably effective in improving living standards and reducing poverty among disabled and elderly workers and their families. It offered exceptional rates of return to early contributors, in part by imposing ever higher tax rates on workers who entered the system later. Exceptional returns are no longer possible, because wage growth has slowed and the number of pensioners relative to contributors has soared. Rates of return for workers just entering the system are expected to be low.

It would be desirable to improve the rate of return young workers will enjoy on their retirement contributions. It would also be desirable to increase national saving to help speed economic growth. Both goals can be accomplished by raising net saving within the retirement system. They can be achieved within the current public system by raising contributions and investing in assets that earn better returns. They can also be achieved by introducing a new private retirement system with individual investment accounts. From an economic perspective, the critical choice is between advance funding and pay-as-you-go financing of future pension obligations, not the choice between public versus private plans.

The choice between the public and private alternatives depends largely on political rather than economic considerations. Advocates of privatization are skeptical that elected officials can be trusted to manage the accumulation of a big retirement fund. They fear that larger Social Security surpluses will be spent on other government consumption (and hence not saved) or that fund accumulation will be invested unwisely. Opponents of privatization believe that explicit separation of the redistributional component of Social Security into a smaller public program will cause the public component to be stigmatized as welfare. This could undermine the popularity and perhaps even the sustainability of large-scale redistribution for the low-income elderly. A public plan offers stronger assurances to low-wage workers, but a private plan is more appealing to average-wage and high-wage workers who want a better return on their contributions.

Both public and private reform plans can raise saving and long-run returns, but only to the extent that they induce some sacrifice of consumption in the near term. The strongest argument for introducing a compulsory and private retirement system is that it may be a more popular and, hence, politically realistic method of boosting saving than a Social Security tax increase. The least credible argument for privatization is that higher saving and improved returns can be achieved without any short-term sacrifice.