People everywhere are living longer, filling leisure communities and nursing homes, and alarming public policy analysts. The widespread failure to die in a timely fashion has inspired countless predictions of political chaos and economic ruin in the industrialized countries. The large American generation born after World War II will have to cope, it seems, not just with the usual intimations of mortality, but also with the unhappy consequences of their longevity for their societies and indeed for their own children.

But a measure of caution is appropriate because, as Bismarck, often credited with inventing social insurance, noted, politics is not an exact science. Neither is predicting economic, demographic, and social trends. Fertility experts, for example, foresaw neither the beginning nor the end of the American baby boom. And 2030 -- the year of peak U.S. boomer retirement -- is as far away from us now as 1964. How many, 30 years ago, could have predicted stagnation of wages, the decline in birthrates, the rise of computers, the fall of the U.S.S.R.?

The shift in the balance between retirees and workers is real, and the sooner adjustments in retirement programs are made, the smaller they will need to be. But the hit-'em-over-the-head-to-get-their- attention approach is having perverse effects. In the running-scared politics of the 1990s, the insistence on revolutionary changes may frighten off officeholders who need the reassurance of knowing that incremental revisions can preserve the pension safety net. In fact, it would be foolish to react to longer life by throwing out the basic features of successful and popular pension programs.

Although the details of advanced nations' pension plans differ, all rely on pay-as-you-go public systems in which firms, current workers, and often general tax revenues provide support for retirees. Some of these programs, like Social Security today, accumulate partial reserves for the future. All require adjustment and fresh thinking. But, despite superficial appeal and powerful marketing programs mounted by special interests, the most radical suggestions -- especially privatization of basic public programs and full funding of all future pension liabilities -- are neither necessary nor, perhaps surprisingly, less expensive. Retaining pay-as-you-go systems and the existing intergenerational compact, with modifications, remains the best course for preserving domestic tranquillity. Although the scale varies by country, adjustments in pensions can generally be accomplished without cramping economic growth or adding to inequality and poverty. The question, therefore, is not, as alarmists would have it, one of economic feasibility, but whether the rich nations choose to insure growing numbers of elderly against want.


Aging populations in most countries are a consequence of two simultaneous demographic trends: declining or stable fertility rates and the steady extension of the life span. The combination of more old people and fewer children means that two sets of statistics have become important: the widely cited ratio of retirees to workers (the elderly dependency ratio, or the number of elderly per 100 working-age individuals 15-64) and the ratio of all dependents to workers (the total dependency ratio, or youths 0-14 years old and elderly per 100 working-age individuals). While it is well known that the former figure is rising, in the United States and Canada total dependency in 2030 will not be much different from 1960 levels. The dependency ratios shown in the table reflect these trends. Germany and Japan, among others, will experience a greater rise in the proportion of elderly dependents than will the United States. The Japanese increase will be particularly rapid, with elderly dependents more than doubling by 2020.

In assessing the projected "burden" of more elderly, an even more general measure may be crucial: the ratio of all potential workers to all those not working. Today, 46 percent of Americans are in the labor force; when the boomers are all retired in about 2030, that number will decline slightly to 44 percent. In 1964, when the baby boomer population peaked, however, only 37 percent of Americans were in the labor force -- a ratio considerably "worse" than can be expected in the 21st century.

Moreover, the real burden on workers is best measured by looking at the overall size of the pie and the resources available per person. In the 1960s baby boomers' parents had less income per capita ($12,195, in today's dollars) than Americans do now ($20,469) and, even under very modest growth assumptions, much less than the projected average in 2030 ($35,659).


The best way to think about the issues posed by aging is to start with the basics about seniors in the United States. They are now 12 percent of the total population and receive, from all sources, about 13 percent of income. Assuming Swiftian solutions are out, the boomers' share of income, while it might be cut a bit, will also be approximately equal to their share of the total population when they retire. Thus, in 2030 seniors will constitute about 20 percent of the population and will probably receive a proportional share of income. To put this change in perspective, between 1970 and 1995, the share of national income for the richest 20 percent of American families increased from 41 percent to 47 percent.

Most workers retire without sufficient personal resources or private pensions to sustain a middle-class living standard; government provides the balance. All Group of Seven nations employ some variant of a package that includes pensions, health care, and disability and survivors' insurance. The United States' minimum old-age benefit, however, is well below the G-7 average, providing only about a third of median income, versus an average of over 50 percent in other advanced nations. The deduction from U.S. workers' paychecks for Social Security is 6.2 percent of the first $65,400 of payroll and is matched by an equal employer contribution. In Germany, by contrast, the workers' deduction is about 9.3 percent of covered payroll, in Japan about 8.25 percent. The payroll maximums are similar.

The scale of public-sector pension reform required depends in part on the other sources of income available for seniors, including personal wealth, private pensions, and continued job participation. Savings, of course, are the ideal source of old age support. High savings can finesse political issues and contribute to economic growth. Household savings in the G-7 countries currently range from a low of 5 percent in the United States and Canada to over 11 percent elsewhere. American baby boomers are currently saving about a third of what they would need to maintain their living standards in retirement. In 1994, the share of U.S. elderly income from all assets had fallen from a peak of 28 percent in 1984 to 18 percent, the 1974 level. Income sources for American retirees other than personal assets are Social Security (42 percent), private pensions (19 percent), and earnings (18 percent).

Private pensions, not normally indexed for inflation, are an important source of senior support. There are persuasive calls for imposing a broad pension requirement on businesses; Australia, for example, has mandated company programs. Today less than 40 percent of American men are vested in a private pension fund, although participation is higher among older workers. The expansion of personal retirement plans, especially 401k tax-deferred payroll deductions, may be improving the picture and even increasing savings. Still, overall, the contribution of private pensions remains less than a quarter of total senior income.

The British experience with personal pensions demonstrates that private plans are not always a recipe for saving public money. In 1986, the government encouraged contracting out of the State Earnings Related Pension Scheme by offering individuals reductions in national insurance contributions as a reward for participation in new personal pension plans. However, the British National Audit Office calculates that, because lost taxes have exceeded the reduction in pension liabilities, the program has resulted in a net loss of $9.6 billion. Britain also found that private plan sales agents captured an outsized share of worker pension contributions during the 1980s. Most troubling is an estimate that in 2030 the basic British pension will be worth a significantly smaller share of average earnings.

Like other countries, the United Kingdom has had to bail out several private pension plans. The 1991 collapse of the Maxwell publishing empire, for example, involved a loss of $880 million. This year Britain is increasing funding standards and enhancing government oversight of private pension plans. In the United States, public guarantees of private pension plans have been the vehicle for bailouts like that of the steel company LTV, which began in 1986 and cost $1 billion. On the plus side, the assets of private pension funds in the United States and the United Kingdom are now equal to more than 50 percent of GDP.

Another possible area of relief for government programs is higher levels of senior participation in the work force. While those levels have stabilized recently, for most of the past three decades workers have been leaving the work force at earlier ages. In France, there is a growing movement to reduce the already low retirement age. In the United States, Britain, Germany, Italy, and Japan, however, the official retirement age has been slowly increasing. Left unanswered is the question of job prospects for older workers. Only the United States has a comprehensive law against discrimination based on age.

Another reform proposal, as recommended in the 1997 report of the U.S. Advisory Council on Social Security (ACSS), a 13-member panel of experts and business and labor leaders appointed by Secretary of Health and Human Services Donna E. Shalala in 1994, is that equities should be added to the investment portfolio of the Social Security trust fund. While diversification is a reasonable idea, it is being oversold. To simplify, the Social Security crisis is premised on U.S. economic growth declining to less than two percent annually, but this solution depends on a continuation of past high returns on equities. Some economists question whether stock values can grow indefinitely in a slow market economy at the same rate achieved in a high-growth environment.

Finally, higher economic growth, combined with other developments such as more rapid population increases, theoretically could ease the demands on retirement plans -- high growth, that is, compared with the annual wage rise of one percent in the mid-level Social Security projection. In fact, under the assumptions of the "optimistic" official projection, the trust fund would enjoy a large surplus throughout the next century. Regrettably, there is no consensus on how to raise either savings or growth, or even how much would solve the pension challenge. Given stable or declining labor forces in most countries of the Organization for Economic Cooperation and Development, most growth projections are disheartening. Therefore, since the public sector will provide the lion's share of retirement support, the debate centers on the relative virtues of privatization and full funding versus modification of existing public pay-as-you-go plans.


The implicit argument for a pension revolution is simple: we're too generous. Seniors will get too big a slice of the pie. But given the actual income and asset profile of most seniors, the real question is whether we would somehow be better off with a lot of poor old people. Only seven percent of retired Americans maintain incomes of more than $75,000 a year. As the Concord Coalition stresses in its 1996 report, "The New Debate over Social Security Reform," half of American households have less than $1,000 in net financial assets; half of households headed by someone in their fifties have less than $10,000 in savings. Overall, median assets for all Americans are slightly over $50,000, including housing, a reduction, in real terms, since 1989. Are these the potentially savvy investors who, under privatization, will buy a share or two of stock a month (presumably at bargain, odd-lot prices) and pay their own way?

In general, privatizers advocate switching from publicly managed and guaranteed pay-as-you-go plans to some form of individual investment accounts. Normally responsible people often talk real foolishness about privatization -- implying that all participants in the universal system of private accounts will consistently outperform the market. While any businessperson or money manager knows that he or she would go broke or to jail by promising a perpetually high, risk-free annual payout to all comers, many privatizers blithely put forward such expectations for all workers. Taken straight, of course, privatization means that how much one can save and how well one does in the market determines what one gets in retirement. Social insurance, such as that at the heart of Social Security, is either eliminated or funded from outside the system.

In the industrialized world, therefore, most serious proposals involve only partial privatization, retaining a minimal safety net (below the poverty level) for those who lose out as either workers or investors. The model often cited by advocates of privatization for the G-7 countries is a system adopted more than 15 years ago in Chile by the military dictatorship of Augusto Pinochet.

In 1981 Chile abandoned a theoretically fully funded plan that had been wiped out by runaway inflation and adopted mandatory individual employee retirement accounts. Workers have contributed 20 percent of earnings to the system -- 10 percent for investment purposes, 3 percent for administrative charges and disability insurance, and 7 percent for health care. There is no employer contribution. Investment accounts were managed by a handful of government- regulated private companies (Administradoras de Fondos de Pensi¢nes, or AFPS). Additional backup pensions are funded from general revenues. Active workers who switched to the new system were awarded service credits for past contributions. To ease the transition, Chileans also received an across-the-board 18 percent wage increase.

The Chilean experience has demonstrated both the benefits of pension reform in a rapidly developing economy and the high costs of transition to privatization. The Chilean government's financial liabilities include large future pension obligations for retirees who participated in the old system, generous guaranteed minimum pensions (85 to 90 percent of the minimum wage), and additional pensions, like those for the military, which have not been privatized. The government has financed some of these costs through a budget surplus but has also borrowed, selling many of the new bonds to the new pension fund managers.

Although the overhead in the Chilean system has come down in recent years, the cost of administrative fees remains 13 percent of mandatory contributions. With 3.5 million contributors, AFPS employ 11,500 salespeople. Perhaps more to the point, since private pension investments are closely regulated, there is little justification for such high marketing and administrative expenses.

Chilean reforms have been touted as a success because, from 1981 to 1995, accounts yielded an average annual real rate of return of 12.9 percent. Recent returns, however, have been lower, and were minus 2.5 percent in 1995. Until the 1990s, virtually all funds were invested in special high-yield government bonds and bank paper. Despite the excellent overall performance of the Chilean economy, AFPS still have small investments in stocks -- 10 percent in 1990, 28 percent in 1996. Thus, the high returns are essentially due to high interest rates. Moreover, Chilean politics are relevant: the new system was imposed by a military regime after national labor unions had been outlawed. Finally, contribution evasion remains a problem, with some estimates suggesting that a quarter to almost half the work force avoid payments.

On the plus side, while some economists question the relationship, it seems reasonable to credit the forced deductions from disposable income for some improvement in savings. The new system is also properly praised for encouraging the development of financial and capital markets. Perhaps most important there is little evidence so far of the corruption and waste so characteristic of Third World systems.

For quite different reasons, in the United States and elsewhere, the push to privatize has considerable political and financial support. According to the December 23, 1995, National Journal, the Cato Institute has raised $2 million to promote privatization. The ACSS report included minority recommendations for full funding and privatization. Implementing these recommendations would not be cheap: one version proposed increasing taxes by $6.5 trillion over the next 72 years. But these advocates of privatization deserve credit for making explicit the requirement for higher taxes. Others who talk of privatization emphasize tax cuts in order to increase savings. But does anyone really believe that a worker earning $20,000 a year who received a cut in payroll taxes would have a 100-percent-plus propensity to save, saving the full amount of the cut and more, as would be required to increase net savings? Most fanciful of all are the privatizers who imagine a world in which individual savers capture the total return of the most productive companies, presumably pay no fees, and consistently exceed the gains of the stock market. Few privatizers mention the additional overhead involved. According to the January 7, 1997, Washington Post, actuary David Langer estimates that under the ACSS proposal favored by Wall Street, a system of individual accounts with a range of investment choices, management and administrative fees could total $240 billion from 1998 to 2010.

Social Security currently provides a larger share of senior income than it did in 1979; for 63 percent of retirees, it is the source of 50 percent or more of total income. The go-it-alone character of privatized systems disregards the dramatic success of programs like Social Security in alleviating elderly poverty, which has fallen from almost 40 percent in the 1940s to less than 12 percent today.

Another practical concern is what happens if, as in the 1970s, the stock market drops sharply. Imagine the political crisis if millions of new retirees were told, "Forget your middle-class dreams. The market went against you." The likely result of a market crash -- as with busted private pension plans, savings and loans, and insurance disasters -- would be a government bailout.

Given these realities, assuming privatization will work seems perverse. Privatizers' predictions depend on compulsion (higher taxes), fear (lower assured government payments), and universal market success. More likely, the result would be increases in need-based pensions and insurance. Regrettably, all privatization proposals fail to come to grips with the most fundamental reasons for public programs: market risk, wealth inequality, sluggish wage growth, and uncertainty about personal longevity.


In the G-7 countries, although all pay-as-you-go systems require maintenance, repair is cheaper than moving to privatized or fully funded systems. According to the ACSS, an increase of 2.2 percent of the combined worker-employer payroll tax would fix Social Security for the foreseeable future. On the other hand, when a nation makes a transition from a pay-as-you-go system to a fully funded one, it pays benefits to retirees who participated in the unfunded system while no longer receiving contributions from current workers, since they are contributing to the new, funded system. To finance benefit payments the government can raise taxes, cut spending on other things, or increase its national debt. In one way or another, a generation of workers will be asked to pay twice for Social Security, once in the form of higher taxes (or lower government benefits) to pay for the transition, and once in the form of a reformed Social Security contribution.

Many comparisons between public systems (pay-as-you-go or funded) and private schemes miss a central point. Universal plans are not merely savings or investment vehicles designed to provide the best return. They pool the risk that all workers share -- income shortfalls, market downturns, disability, and other misfortunes -- and they provide basic support, usually in the form of a lifetime retirement annuity that offers a return equal to inflation plus the real increase in wages. They normally include an insurance component that provides income for surviving spouses and the disabled, the face value of which, in the United States, is about $12.1 trillion.

One oft-repeated argument against pay-as-you-go systems is that they discourage savings. But the evidence is mixed. Mandatory payroll deductions can reduce current consumption by workers, and a modified pay-as-you-go program running a surplus, like Social Security today, can contribute to net national savings. The compelling politics of senior support programs provide one of the few justifications for raising taxes rather than borrowing. In the United States during the strongly anti-tax era since 1970, for example, payroll taxes have increased on more than a dozen occasions. The much higher tax rates that would have to accompany a shift to a fully funded system, however, are simply politically impractical.

The other key arguments against public pay-as-you-go systems are that they no longer provide a fair return, that they will become too great a burden for those working, and that they will create destabilizing fiscal burdens for governments. With regard to the first point, a comprehensive review of recent studies leads the economists Yung-Ping Chen and Steven C. Goss to conclude that Social Security "provides exceptional money's worth. Program administrative costs absorb less than one percent of every dollar." Even excluding insurance value, the return, although less favorable for future workers, will still be "reasonable."

With regard to the burden on workers, remember that total per capita GDP continues to increase and that total dependency rates in the United States and Canada will only return to 1960s levels. Moreover, when considering total burdens, future workers -- heirs to the immense economies of the G-7 countries and to the capital accumulated by the largest generation in history -- may be in an enviable position. The Great Depression, World War II, and the Cold War were somehow manageable for boomers and their parents. A larger pay-as-you-go pension obligation sounds a trifle tame in that company of generational challenges and is a small price to pay for that legacy.

On the fiscal risk point, the International Monetary Fund, police officer of government profligacy, recently published a study analyzing the coming gap in pension financing. After focusing on the relative cost of adjusting current pay-as-you-go systems versus moving to fully funded systems, the study concludes that U.S. and U.K. plans could stay solvent with modest increases in contribution rates. In 1995, the IMF calculated that the United States would have adequate funding through 2050 if it raised its contribution rate by 0.8 percent of GDP. How revolutionary would such a shift be? Well, as children, the baby boomers sparked an increase in the share of GDP devoted to elementary and secondary education from 1.4 percent in 1946 to 4.1 percent in 1970. The United Kingdom, according to the same IMF study, is in the best shape of the industrialized nations, requiring an increase in contributions of only 0.1 percent of GDP. Canada, Germany, Japan, Italy, and France would need more substantial increases of approximately 2 to 3 percent of GDP to avoid a buildup of pension debt. Other measures, such as raising the retirement age, lowering the income replacement rate, and indexing benefits to prices rather than wages, would reduce the revenue shortfall. Importantly, the IMF assumes reduced growth rates in all the nations under study -- a range of 1 to 1.5 percent.

The IMF also found that, over the next 50 years, changing to fully funded pension systems would involve large transition costs. If the United States made a gradual transition to such a system, it would need to increase its cyclically adjusted deficit, excluding net interest, to 3.4 percent of GDP, compared with only 1.5 percent of GDP to modify the current system. Similarly, to switch without debt accumulation, Germany, the United Kingdom, and Italy would need to increase their cyclically adjusted deficit by double the amount necessary to mend their pay-as-you-go systems. For all seven major industrialized nations, moving to a fully funded system would be more costly than modifying their current systems.


In modern democracies, any large group that is not self-sufficient is a source of political controversy. Sometimes the scale of programs involved bears little relation to their political significance. The welfare benefits of 14 million Americans, 9 million of them children, for example, have been at center stage of national politics for more than a decade, despite the fact that the dollars involved are less than one percent of the federal budget.

In this sense, the current pension debate is really about the prospects for income and wealth distribution. If recent sluggish wage and income growth and worsening wealth inequality were rectified, the transition to an older society would be easier. Of course, some argue that slow growth is inevitable because boomers will draw down assets in retirement and overall savings will decline. In the United States, pension savings are 3.6 percent of total wages today; they may be zero in 2024 and minus 3.5 percent in 2040. But, while aging may reduce household savings, the decline could be offset completely by lower demand for capital goods. Indeed, if current population projections hold, it will take no net savings to maintain a constant amount of capital per worker. Moreover, living standards can always rise because of technological change. Slow growth is not an inevitable byproduct of an older society. As the economists James Poterba, Larry Summers, and their colleagues put it, "A more definitive finding is the absence of any empirical support for the pessimistic view that aging societies suffer reduced productivity growth."< In fact, increased productivity growth of 0.1 percent per worker in the United States would completely offset the additional burden of an increased elderly population.

It is surprising, therefore, that there is not more debate over whether the G-7 nations have an iron destiny to grow less than 2 percent a year. Could future economic performance be enhanced by targeting investment to the young, perhaps reducing investment in health care facilities for the old? Couldn't we focus policy and resources on more savings, investment, training, and education to increase growth, rather than funneling premiums to Wall Street? The answer seems to be probably not, if such improvements require higher taxes. Another course is to expand the labor force through more immigration, but here too politics is an overriding factor. There is also a good case, if not a good political strategy, for more equal incomes, higher minimum wages, and other measures to make savings a more practical alternative for average workers.

There are many proposals, not dependent on growth, designed to work in combination to fix systems like Social Security: smaller income replacement ratios for upper earning brackets, higher taxation of high pension benefits, eliminating ceilings on taxable payroll, switching from wage-based to price-based adjustments, increasing the retirement age, and, in the United States, bringing millions of government workers, not currently covered by Social Security, into the program. All of these proposals, however, while sensible, depend on political will.


With the demise of communism, there are no systems left standing that promise citizens equal economic results. Capitalism, now favored by most of the world for good reason, certainly does not make such a claim. Everywhere it holds sway, there are wide disparities in income and wealth. About 85 percent of all financial assets, for example, are owned by the richest ten percent of American families. There is a certain illogic, then, in prescribing, as privatizers do, a cure for elderly poverty of taking a little more of the market and calling them in the morning.

In the democracies, at least, there is a strong consensus that extremes of wealth and poverty do not make for a healthy society. Just plain compassion is still compelling for some, particularly toward the young and old. There is also a widespread belief that those who have worked should be assured of enough income in retirement to maintain a reasonable standard of living. It may be rational to frighten people about aging in order to scare them into saving more, working harder, getting more education, even supporting additional payroll taxes. But it is silly to claim that the private market can make everyone rich. Fortunately, the appeal of such notions can stand neither close analysis nor the test of politics.

Programs like Social Security remain enormously popular. According to the results of a Europe-wide poll published in the January 27, 1996, Economist, most voters think that their governments do not do enough for seniors. And, despite repetition of the canard that Americans have given up on Social Security, more than 80 percent of them believe that government must keep the program, and an equal number approve of their own continuing payroll contributions. Seventy-one percent would pay higher taxes to prevent cuts.> The unequivocal commitment to Social Security and Medicare by both parties in the 1996 campaign was not a result of ignoring the polls.

Given the alternatives, continuing support for current policies -- with modifications -- may not be a bad outcome. The reform of existing public pensions looks possible and prudent. The democratic process is not likely to lead, as some contend, to disaster, but rather to an acceptable result. The likelihood of periodic adjustments in the demographic and financial estimates is a reminder of the abiding virtues of muddling through. For those who would be distressed if privatization and all those new commissions and management fees fail to materialize, there is the consolation offered by John F. Kennedy: "If a free society cannot help the many who are poor, it cannot save the few who are rich."

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now