Unhappy Returns

The Real Alternative to Social Security Is Genuine Privatization

Harry Dolan is a writer and editor in Bowling Green, Ohio.

On August 13, 1920, a confidence man named Charles Ponzi was arrested for running a pyramid scheme that had cheated investors out of millions of dollars. Ponzi had promised his investors a 50 percent return after 45 days, and he was able to deliver, at least in the beginning. As news of these fabulous returns spread, more and more investors were lured into the scheme. Of course, the money Ponzi collected was never invested in any wealth-creating enterprise: the “profits” of the earlier investors were paid with money collected from later investors. Ponzi kept his scheme going for less than a year. He was tried and convicted on federal and state charges and served a pair of lengthy prison terms. He was released in February 1934.

Later in 1934, by a kind of charming historical coincidence, President Franklin Delano Roosevelt appointed a Committee on Economic Security to study ways of dealing with financial insecurity, particularly among the unemployed and the elderly. The Committee’s recommendations were embodied in the Social Security Act, which Roosevelt signed into law in August 1935. Under the resulting Social Security system, the payroll taxes paid by workers and their employers were not invested in any wealth-creating enterprise; rather, the contributions were used to pay the benefits of retirees.

The members of the Committee were hardly con artists; they were sober and thoughtful people grappling with a serious problem. In the depths of the Great Depression, the plight of elderly Americans without reliable sources of income or family support was dire. But the remedy the Committee proposed was not some modest and temporary measure designed to relieve suffering: the system they helped put into place has become a (seemingly) permanent arrangement by which wealth is transferred from workers to retirees.

When Social Security was established, it may have seemed plausible that such a system could be made to work. In 1940, when the first monthly benefit payments were sent out, there were only 222,000 beneficiaries. The age at which one could retire and begin to collect benefits was 65, four years beyond the average life expectancy. The system was financed by a relatively small (2 percent) payroll tax, paid half by the employee and half by the employer, and the amount of income subject to the tax was capped at $3,000. Today there are over 43 million beneficiaries. The retirement age is set to rise to 67 in the coming decades—but even so, people are living far longer (around 76 years on average) and thus spending far more time in retirement. The payroll tax, after a series of increases over the years, rests at 12.4 percent, and the amount of income subject to the tax has climbed to more than $60,000—and still, no one expects the system to remain solvent long enough to shepherd today’s younger generation of workers through their retirement. The Social Security Administration’s own best estimates indicate that the system will fall far short of being able to pay its present level of retirement benefits beyond the year 2032.

The Culprit: Demographics

Unlike Charles Ponzi, the Social Security system never promised to make anyone rich—and it hasn’t. At present, the average Social Security benefit replaces roughly 43 percent of the beneficiary’s pre-retirement income (a replacement level of 60 to 85 percent is needed to maintain one’s pre-retirement standard of living). The key to understanding this humble benefit level—and the system’s long-term insolvency—lies in the demographic changes that have taken place in the United States since the system’s inception.

These changes are actually a broader phenomenon, common to nearly all developed countries, and thus the Social Security crisis in the United States is mirrored by similar troubles in the old-age social-insurance programs of western European nations, Japan, and numerous other countries. The fundamental “problem” is that people are living longer and, at the same time, deciding to have fewer children. This leads to a general aging of the population, as more and more people live to collect retirement benefits—and go on collecting them for longer than ever before—while at the same time, proportionally fewer young people are in the work force paying into the system. The result is a declining “support ratio”—the ratio of current workers to current beneficiaries. In the United States, this ratio dropped from a high of more than 40-to-1 in the late 1930s, to 16-to-1 in 1950, to 3.4-to-1 today; it is expected to fall to 2-to-1 by 2030.

The declining support ratio is problematic for government pension programs because, as noted, these are essentially systems for transferring wealth from workers to retirees. In the language of policy analysts, they are predominantly “pay as you go” plans. Over time, as there are fewer and fewer workers to support each retiree, something has to give: payroll taxes must be increased, benefits must be cut, or the age of retirement must be pushed back (which is really just another way of cutting benefits). In the United States, Social Security benefits were essentially set at a fixed level in 1972, with mandated cost-of-living adjustments (COLAs) put into place to ensure that benefits keep pace with inflation. The primary method of dealing with the declining support ratio has been payroll-tax increases, more than a sixfold rise since the beginning.

The burden imposed on workers by the current 12.4 percent payroll tax should not be underestimated: many workers pay more in payroll taxes (which are levied from the first dollar of wages, without exemptions) than they do in federal income taxes. And while payroll taxes are nominally paid half by the worker and half by the employer, in reality the worker bears the full impact of these taxes. The share paid by the employer is part of the worker’s overall compensation package, and it necessarily reduces the amount of money available for wages and other benefits.

The most recent payroll-tax increase was passed as part of the 1983 Social Security amendments, which were intended to address the looming problem of how to finance the retirement of the “baby boom” generation. At their present level, these taxes generate a yearly surplus, which is invested in U.S. Treasury bonds (held in the Social Security Trust Fund). Thus, the current system is not entirely pay-as-you-go, yet it is still predominantly so, since the lion’s share of payroll taxes (80 percent in 1998) is paid out directly to current retirees. Nevertheless, this “partial reserve” method of financing is expected to postpone the system’s demise. According to the Social Security Administration’s projections, yearly outlays (benefits) will begin to exceed income (payroll taxes) in 2013. At that time, in order to continue to meet the system’s obligations, it will be necessary to begin redeeming the treasury bonds in the trust fund. This process is expected to extend the system’s life until 2032, at which time the fund will be exhausted and income from payroll taxes will meet only 70 to 75 percent of obligations to retirees.

It is worth noting, however, that most critics of Social Security view the trust fund as a dubious accounting device. As they point out, when surplus funds from payroll taxes are invested in treasury bonds, the federal government uses the money to finance its day-to-day operations. When the time comes to redeem the bonds, the government must come up with the money either through other borrowing, tax increases, or spending cuts. Depending on how palatable these various options are, the need to start cashing in the treasury bonds in the trust fund may precipitate a crisis long before 2032.

Poor Investment

The payments that individual workers and their employers make into the Social Security system are not, in a strict sense, investments. Unlike funds invested in equities, corporate bonds, or even savings accounts, payroll taxes are not put to productive use and do not earn a market return for the people who pay them. Nevertheless, those who retire after paying into the Social Security system do receive benefits, and these can be thought of as the “return” on their payroll-tax “investments.” If we think of Social Security this way, we see that it has, over the years, become an increasingly bad investment—especially if we consider the rates of return that would have been available through other forms of investment.

During the early phase of any old-age social-insurance system, the typically high support ratio (the high number of workers per retiree) allows an excellent rate of return for retirees—as might be expected, since some people are able to collect full benefits from the system after contributing to it for only a few years.1 As it turns out, however, the very best rates of return go to those who are between 30 and 50 years old when the program begins. This is because social-insurance programs tend to begin modestly (since a radical new program would be politically unpalatable), and benefits are gradually increased over time. Eventually, as the support ratio worsens, payroll taxes must be increased if benefits are to be maintained at the same level. Workers who come into the system late, after payroll taxes have been increased, will pay much more over the course of their working lives. Their benefits—even if adjusted each year to keep pace with inflation—will represent a terrible return.

Social Security in the United States has followed this pattern. For example, workers who retired in 1975 received, in just two years, benefits equal to their entire lifetime contributions. In contrast, a worker who retires in 2035 will take anywhere from seven to 17 years (depending on his income level) to get back the money he put into the system. That’s assuming he lives long enough—and that the system is still around in 2035.

The gravity of the situation for young workers becomes clear when we look at the annual real (inflation-adjusted) rate of return on their contributions to Social Security. This rate is estimated to be 1 to 2 percent for most workers (although it may be as much as 3 percent for the very lowest earners).2 And the situation may even be worse: economist Martin Feldstein estimates that those who are presently in their 40s or younger will earn a negative return: that is, they will actually receive less in benefits than they and their employers have paid into the system.3 Compare this with the real rate of return on stocks, which has averaged 7 percent from 1926 to the present, or on a mixed portfolio of stocks and bonds, which comes in at around 5 or 6 percent.4 The magnitude of the difference means that if workers were free to invest their money, rather than being compelled to contribute to a system that merely transfers their wealth to others, they could earn returns many times larger than those promised by Social Security.

In fact, William Shipman estimates that if workers born in 1970 were able to invest an amount equal to their Social Security contributions in stocks, they would (assuming they earned historically average returns) receive almost six times the benefits they would get under Social Security. Low-income workers would do less well, but would still receive almost three times what Social Security offers.5

Harnessing the Market

It is the prospect of these kinds of returns that has led some critics of Social Security to call for privatization: they argue that it should be replaced by a system of individualized private retirement accounts, financed (as the current system is) through compulsory contributions. The difference, of course, would be that contributions would actually be invested (in some combination of stocks, bonds, and mutual funds), rather than being paid out as benefits to current retirees. The amount contributed, plus accumulated earnings, would be used to fund the individual’s retirement. The model for this type of privatization is the system instituted in Chile in 1981.

Critics of this type of proposal point out that investing in stocks can involve substantial risks—to see this, we need look no farther than the 19 percent drop in the market between July 17 and August 31 of last year. That kind of volatility, critics say, is hardly a recipe for retirement security. The trustees of the Social Security Trust Fund appeal to this sort of reasoning in their 1998 annual report: “Social Security,” they write, “is a compromise that assures all workers a modest retirement base on which they can, if possible, add a private pension and personal savings. The tradeoff for this safety is a lower retirement benefit than at least some knowledgeable individual investors might build over their lifetimes.”

This is at least half true. The “retirement base” provided by Social Security is indeed very “modest,” as we’ve seen—and is sure to get even modester after 2032 if nothing is done. But would only “knowledgeable investors” reap better returns by investing in private markets than they would through Social Security? In assessing the risks of investing in stocks, we need to look at the long term—and in the long term, things look good even for novice investors, assuming they recognize the limits of their knowledge and invest in broadly diversified mutual funds or stock index funds that track the performance of the market as a whole. Even during the worst 20-year period of stock market returns, which ran from 1929 to 1948, the market as a whole yielded a real annual rate of return of more than 3 percent—which suggests that even under the gloomiest of conditions, workers wouldn’t do worse on their own than they would under Social Security. Indeed, if Feldstein is right and workers in their 40s and younger can expect a negative return from the current system, then any investment that merely keeps pace with inflation would offer a better deal than Social Security. If the dollar could be relied on to hold its value, you could beat Social Security’s return by stuffing your payroll-tax money into mason jars and burying them in the back yard.

Reform Proposals on the Table

What can be done to address the serious problems facing the Social Security system? Congress is expected to take some action on the matter this year. The two main alternatives that seem to be emerging both involve using the potential gains from investing in stocks to shore up the system without radically altering its basic structure. The first alternative would simply invest some part of the present annual Social Security surplus in private financial markets, in the hope that higher returns would help to cover the shortfall that the system is expected to experience as the baby boomers retire. The best that can be hoped for from this option is to preserve the current system as it stands—to perpetuate the current meager level of benefits, which could not otherwise be maintained beyond 2032.

The second, more ambitious alternative is backed by Senators Daniel Patrick Moynihan and Bob Kerrey. It would attempt to salvage the current system by supplementing it with private investment accounts—into which an individual could divert 2 percentage points of the 12.4 percent payroll tax. Individual workers could choose how to invest the money in their accounts—at least within government-approved limits. The returns from these accounts would be added to the individual’s standard Social Security benefits. Since diverting 2 percentage points of the payroll tax would actually worsen Social Security’s long-term financial imbalance, the difference would have to be made up somewhere. Accordingly, the age of retirement would gradually be raised to 70 between now and 2029, and any further shortfall would be paid for out of general tax revenues (the plan assumes that the overall federal budget will be running at a surplus).

A Better Alternative

Neither of these proposals challenges the fundamental assumption underlying Social Security (and old-age social insurance in general): the idea that society is responsible for ensuring some level of retirement security for all its members. Even a radical Chilean-style privatization plan (one that diverted all of an individual’s contributions into a private retirement account) would not challenge this assumption, so long as the contributions remained compulsory.6

But the path to security does not lie in surrendering responsibility to society. Indeed, to live a human life is to embrace the responsibility of living long range: it is to realize that there may come a time when one will not want—or simply will not be able—to continue working. Planning for such a contingency is a profoundly moral issue. It is a weighty undertaking; there are any number of things that might go wrong. The assumption behind Social Security is that it is too heavy a burden for any one person to bear, so the collective will relieve the individual of the need to think and plan for the long run.

The irony is that the creators of Social Security failed to think and plan long range. The system they designed has—in less than the span of a human lifetime—reached a crisis. The burden of its payroll tax robs many people, particularly the poor, of the opportunity to make investments that might actually provide them with a comfortable retirement. It represents a deal that no one would accept willingly and that no one should be forced to accept.

The inevitable response to the suggestion that Social Security should be abandoned is that its elimination would betray current and future retirees who have assumed they would receive its benefits. Indeed, elimination of Social Security would require a costly transition. This fact is an indictment of the system itself, and can hardly be used as an objection against the system’s critics.

What should ultimately replace Social Security? The Chilean-style privatization alternative mentioned above—essentially a compulsory savings plan—would undoubtedly provide a superior return for workers if it were properly implemented. Yet it fails to challenge the idea of collective responsibility that lies at the heart of Social Security, thus opening the door to large-scale government interference in private financial markets. The temptation to regulate investment for “the good of society” when the future of the nation’s retirees is at stake would be strong. What’s more, a compulsory private system would share Social Security’s assumption that individuals are not to be trusted to plan their own lives. The regimented existence that any compulsory system must impose is, at root, incompatible with human freedom and human responsibility.

The real alternative to Social Security is privatization in a literal sense—the realization that planning for retirement is a private matter, that it should be left to individuals and their families, who would be free to seek the soundest advice they could find and make the best arrangements that were within their means. Such a system—which is really not a “system” at all—would be far more secure than Social Security, since it would not leave people at the mercy of demographics and support ratios—or of politicians.


  1. For an excellent discussion of old-age insurance, from which the account in this paragraph is drawn, see Daniel Shapiro, “Can Old-Age Social Insurance Be Justified?” in The Welfare State, ed. Ellen Frankel Paul, Fred D. Miller, Jr., and Jeffrey Paul (New York: Cambridge University Press, 1997).
  2. Peter J. Ferrara, Social Security Rates of Return for Today’s Young Workers (Washington, D.C.: National Chamber Foundation, 1986); Michael W. Lynch, “Retirement Plans,” Reason, August 1998, p. 56.
  3. Martin Feldstein, “How to Save Social Security,” New York Times, July 27, 1998, p. A17.
  4. Testimony by Carolyn L. Weaver before the Senate Committee on Finance, September 9, 1998.
  5. William Shipman, Retiring with Dignity: Social Security versus Private Markets (Washington, D.C.: Cato Institute, 1995).
  6. For details on how a compulsory private system might work, see Peter J. Ferrara, “Privatization of Social Security: The Transition Issue,” in The Welfare State.

Further Reading