Budget Crunch: Population Aging in Rich Countries

Over the next several decades the populations of the major industrial countries will grow considerably grayer. By 2030 the ratio of people past age 64 to those ages 15-64 will be just over 30 percent in the United States, close to 40 percent in France and Great Britain, and nearly 50 percent in Germany and Japan. The increased cost of retirement benefits will put enormous pressure on public sector budgets at a time when the workforce is scarcely growing or even shrinking. But though all the big industrial countries share the prospect of an aging population, no two face exactly the same future. Variations in the size and timing of the demographic changes, as well as important differences in public programs for the elderly, mean that population aging has different implications in each country.


Over the next several decades, the aged dependency rate—the ratio of people over age 64 to those of working age—will rise most steeply in Germany and Japan. Not only will the aged population increase, but the working-age population will fall. Indeed, the economic issues of population aging will be compounded by the possibility of actual population decline. Fertility rates in both Germany and Japan are far below the “replacement rate” needed to maintain a constant population, now about 2.1 children per woman. Official projections assume the fertility rate will gradually return to the replacement rate in Japan, but remain close to the current level (1.4) in Germany. German forecasts also assume substantial (but declining) immigration, an annual net ow of about 2 immigrants per 1,000 residents, compared with 5.6 earlier this decade. Immigration is assumed to be negligible for Japan.

France and the United Kingdom face less dramatic population change. Although fertility rates in both have declined (to 1.8), over the next quarter century the total populations of France and Britain are expected to grow, while the working-age populations remain roughly unchanged. Dependency rates will rise because of the growing number of the elderly.

Though the aged population is projected to grow fastest in the United States, the U.S. aged dependency rate will grow the least. The U.S. fertility rate is now above 2.0, and immigration remains strong (4.4 per thousand residents), so the working-age population will continue to grow, although much more slowly than in the past.

Official forecasts in all ve countries suggest life expectancy improvements will slow. In Japan, for example, life expectancy is predicted to improve over the next 30 years at one-sixth the rate of the past 30. In the United States, it is predicted to rise at half the rate of the recent past. Germany projects no gains in life expectancy. Many demographers believe these projections understate likely improvements in longevity, implying an even greater rise in the aged dependency rate.

Pension Burdens

France, Germany, Japan, the United Kingdom, and the United States have all tried to improve the lot of their aged citizens over the past half century. By liberalizing public pensions and health insurance, all have sought to provide the elderly a living standard comparable to that enjoyed by the working-age population. By the mid-1980s, they had largely achieved their goal. Poverty had fallen sharply among the elderly, and living standards of typical aged and non-aged households had substantially converged.

The five countries differ widely in their spending on public pensions, however. Pension costs as a share of GDP are much higher in France and Germany than in Japan, Britain, or the United States (line 1, table 1). Not only do France and Germany provide more generous pension benefits (line 2), they also use public pensions to finance early retirement for the long-term unemployed, a practice that has depressed employment rates among people in their late 50s and early 60s.

Table 1. Public Spending on Pensions, 1995-2040
  France Germany Japan United
1. Public pension costs—1995
   (Percent of GDP)
10.6 11.1 8.6 6.5 5.1
2. Net replacmenta
78 63 55 50 50
3. Pension cost projections
   (Percent of GDP)
2000 12 11 11 6 5
2010 13 12 14 6 5
2020 15 13 15 6 6
2030 17 14 16 6 7
2040 20 14 n.a. 6 7
4. Present value of net pension liabilities
   (Percent of GDP)
115 110 105 5 25
a After-tax value of public pension as a percent of after-tax wage while at work for average-wage worker.
n.a. Not Available

Public pension costs are less burdensome in Japan, Britain, and the United States, in part because pensions are less generous, but also (in the case of Japan and the United States) because the populations are younger. The United States spends the smallest portion of GDP on public pensions. Its aged dependency ratio is the lowest, and its public pensions are, along with those in Britain, the least generous.

Each country’s public pension system is distinctive (see box at the bottom). In France, Germany, and Japan, population aging and the generosity of the pension formula mean that public pension costs must rise rapidly. Reforms in the U.S. system in 1977 and 1983 will hold down spending increases. The normal retirement age, for example, will rise gradually starting early in the next century, reaching 66 for people born in 1943 and 67 for people born in 1960. That reform essentially reduces future pensions 12-14 percent. On the other hand, steep increases in the cost of Medicare will send U.S. spending on the elderly soaring in spite of only moderate growth in public pension costs. Britain took major steps during the 1980s to scale back public pension spending. The Conservative government cut future public pension commitments and offered financial incentives for workers to opt out of part of the public system.

The projected costs of public pension programs vary widely (line 3). France, Germany, and Japan are expected to encounter severe financing problems within the next few decades. In all three countries, estimated unfunded liabilities of the public systems exceed current GDP (line 4). The unfunded liability of the U.S. system is one quarter of GDP. Britain’s net liability is near zero.


To restore long-term solvency to public pensions, policymakers confront a choice among four reform alternatives. Three—cutting benefits, increasing contribution rates, or raising the age of retirement—can be implemented within the present pay-as-you-go framework. The fourth moves away from pay-as-you-go toward advance funding of retirement obligations—either within the public system or in privately owned and managed pension funds.

Benefit cuts. Public systems are now the main source of income for most retirees, providing over 40 percent of total retirement income in the United States and up to 70 percent in Germany. Because many old people have fairly modest incomes (often just above the poverty line), most countries cannot reduce minimum pensions without increasing poverty. Many proposals for scaling back benefits therefore emphasize means-testing or modifying the inflation index.

Moving from an earnings-related to a at-rate benefit could reduce public pension costs, but it introduces important incentive problems. Severing the link between a worker’s earnings (and tax contributions) and his or her retirement pension discourages work and encourages tax evasion. It will be opposed by high-wage workers, who would pay a tax proportional to their earnings but receive only a minimum, at-rate pension. To keep the contribution rate down in a reformed system, these workers would almost certainly press their political representatives to keep the basic pension low, putting many old people at greater risk of becoming poor.

Means-testing public pensions on the basis of retirees’ current income can also lower costs. But it would discourage private pensions and saving, which might, perversely, make middle-income workers more reliant on public pensions.

Benefits can also be trimmed by reducing the inflation adjustment. To justify that step, some observers argue that nonmedical consumption needs decline with age. In the United States some also claim that the consumer price index overstates increases in the cost of living. But reducing the annual inflation adjustment would progressively reduce the real benefits of aging retirees as they grow older. Because most other retirement income is not indexed, this would exacerbate a pattern in which retirees’ real income declines with age—and poverty rates rise.

Higher contributions. Raising contribution rates is a second option. But tax rates are already so high in many countries that raising them further would be very unpopular and, possibly, counterproductive. Often, though, the high tax rates apply to a tax base that is far less than 100 percent of total labor compensation. Wages are taxed only up to a certain limit. Fringe benefits are usually not taxed. (In fact, the gradual shift from taxable wages to untaxed benefits accounts for more than a third of the long-term U.S. retirement fund deficit.) Broadening the tax base could obviously close some of that gap. But some of the extra revenue would be offset by higher pension payments to workers credited with higher average wages. And many fringe benefits, such as health insurance, are hard to value, making it hard to calculate a worker’s pension contributions.

Raising immigration or the birth rate could also add contributors and revenue to the public system. But immigration is unpopular in Europe, in part because of high joblessness, and it has become much less popular in the United States in the past decade. Large-scale immigration has never been permitted in Japan. Germany now offers major incentives for childbearing, but it is hard to see any effect on the birth rate. And childbearing incentives may encourage women to stop working, partially offsetting the benefits of a higher birth rate.

Delaying retirement. Increasing the retirement age is another way to reduce pension costs. Although expected longevity at age 60 has increased about one-fth since 1960, the earliest age for claiming pensions has been left unchanged or even reduced in the ve big industrialized countries. Germany, Japan, and the United States plan to raise the age of entitlement for a full pension, but much less than the longevity increase over the past few decades. Because workers in most countries prefer to retire before the “normal” retirement age, raising the age of entitlement for full benefits without raising the early retirement age amounts to reducing benefits.

Increasing the retirement age can certainly cut costs. The United States could erase its 75-year Social Security deficit by raising the retirement age to 67 today, increasing it gradually to 70 in 2030, and raising the early retirement age from 62 to 67. Delaying retirement is widely unpopular, however, especially among workers with physically demanding jobs. The evidence of the past few decades suggests, in fact, that most workers strongly prefer earlier retirement options.

Advance Funding

Proposals to address the financing problem through benefit cuts and tax increases are inherently divisive, because they force generations and income classes to vie over who will have to make the larger sacrifice.

It is possible to mitigate these divisions by increasing the national income that will finance the consumption needs of future workers and retirees alike. To achieve this, the current generation can increase its saving to finance more of its own retirement. Larger accumulations in retirement systems, whether public or private, over the coming decades would raise the nation’s capital stock and raise national output. In the next century, the nation would be spending more on pension programs, but paying for it out of a larger economic pie, leaving a bigger slice of the pie for future workers.

The current pay-as-you-go system of financing public pensions does not increase national saving. In all the national systems under discussion here, payroll taxes from today’s workers go almost entirely to pay for pensions for today’s retirees. During the 1950s and 1960s, pay-as-you-go looked like a good idea. The labor force was growing briskly, and real wages were climbing 2-5 percent a year. The return on contributions once the system was mature was expected to be 4-7 percent a year, far more than ordinary workers could earn on their own savings.

Declining labor force growth and the dramatic slowdown in labor productivity growth have eliminated those advantages of a pay-as-you-go system. The rate of return has fallen below 2 percent a year in most countries and may soon become negative. Private investment alternatives offer workers and pension fund managers real returns exceeding 3 percent a year. In view of the difference in expected rates of return, many of today’s workers and young voters would choose prefunded retirement accounts over pay-as-you-go.

Unfortunately, the pay-as-you-go system has inescapable consequences. Governments have accumulated huge pension liabilities to retirees and older workers. Democracies are unlikely to default on these obligations. Over the next several decades, current and future workers will pay for the promised pensions, regardless of whether governments adopt new advance-funded systems. The double burden of paying off those obligations and saving in advance for their own retirement makes it costly for younger workers to move cleanly from pay-as-you-go financing to advance funding.

Nonetheless, today’s workers could increase the portion of retirement income they expect to derive from capital income and reduce the portion coming from payroll contributions of future workers. Governments could move toward partial funding of future retirement obligations either by modifying current public systems or by converting them fully or partially to private systems. In either case the central question is whether the increment to funding would really add to national saving and capital formation and boost future national income or whether it would be offset by reduced public or private saving elsewhere.

Advanced funding is simplest to implement within existing public programs because it would leave accrued benefit claims intact. Increased contribution rates or reduced benefits (or both) would create a reserve, which should be strictly separated from other government accounts. The reserve would then be invested in either public or private securities. From the point of view of economywide gains, it matters little which. If the public pension fund purchased government debt, more private saving would go to finance private investment. If the public fund invested in private debt or equities, private savers would be forced to purchase more government debt.

Public management of a huge retirement fund, however, raises thorny political issues. Politics might skew investment decisions. Even worse, public officials might use reserve accumulations to offset deficits in other government accounts.

Private retirement accounts can reduce these political risks. In addition, they offer workers flexibility in managing their own retirement savings. Partial privatization, as in the two-tier system adopted by Chile in the early 1980s, is a possibility. A first-tier public program could provide a at benefit or one related to the number of years of participation; the second-tier program could support a private defined-contribution pension program, with individual accounts invested in a range of capital market assets by the individual contributors.

But privatization too carries risks. Explicitly separating out the redistributional component could create strong pressure to reduce or eliminate it. A two-tier partially privatized system may not provide adequate income security for retirees with low lifetime earnings. Workers may make bad investment decisions. Converting individual accounts into annuities when workers retire or become disabled presents a huge challenge. Solving this and other problems entails high administrative costs that may eat into the returns of small accounts.

Individual retirement accounts have obvious appeal to high-wage workers, especially those with investment expertise. Setting up millions of IRAs—effectively, defined-contribution pension plans—eliminates concerns about the public costs of retirement programs. Benefits are determined by contributions and market interest rates, not by government contributions. But if millions of workers suffer losses on their investments, democratically elected governments may face enormous pressure to compensate them.

The Key: Economic Growth

Public programs for the elderly already consume a large and growing portion of public expenditures. For France, Germany, Japan, and the United States, the percentage of public budgets targeted on the elderly will rise sharply in the coming decades unless current laws are changed.

Britain is the exception to this general pattern. Recent policies have curbed the future growth of public spending on first-tier public pensions and encouraged workers to opt out of the second-tier public fund. As a result, future U.K. retirement incomes will depend increasingly on returns earned in privately managed and invested pension funds.

If all goes as planned, Britain will accumulate substantial reserves in its (increasingly private) pension system. If the budget deficit is kept low, those growing accumulations can help boost national saving, which in turn can spur economic growth. But there is a risk. Workers who invest their retirement funds badly may have to retire on pensions substantially lower than their preretirement incomes. A lengthy period of low or negative private market returns may leave an entire cohort of workers facing the prospect of low retirement incomes. This shortfall may be more than a problem for the unfortunate workers. If voters demand good incomes for workers in retirement, it might also create huge problems for the public budget.

The United Kingdom, like other countries, must support its retired population out of the national income available to it. Whether retired workers get most of their income through public pensions, as in Germany, or private pensions, as in the U.K., their consumption will be derived from the output of future workers and the future capital stock. If future productivity grows rapidly, the elderly can be generously supported while workers enjoy steady increases in their after-tax incomes. If productivity grows slowly, future workers will have to accept lower after-tax incomes or retirees smaller pensions unless workers can be persuaded to delay their retirement. The implications of slow growth will be the same whether pension incomes come from public or private sources. One way to increase economic growth is to accumulate larger pension fund reserves to boost national saving. Whether the reserves are accumulated in a public or private fund, they can contribute to future national income only if they yield an increase in overall saving.