SYSTEMS of financial support for old people are in trouble worldwide. To ensure that these systems continue to protect the old and promote economic growth, many countries need to consider comprehensive pension reforms.
Over the next 35 years, the proportion of the world’s population over 60 will nearly double, from 9 percent to 16 percent. Populations are aging much faster in developing countries than they did in industrial countries. As today’s young workers near retirement—around the year 2030—80 percent of the world’s old people will live in what today are developing countries. More than half will live in Asia and more than a quarter in China alone. (See Charts 1 and 2.) These countries need to develop their old-age systems quickly and make them sufficiently resilient to withstand rapid demographic change.
Chart 1Rapid aging of population foreseen
Source Adapted from the World Bank population database
1 Japan is included with member countries of the Organization for Economic Cooperation and Development (OECD) not with Asia.
Chart 2Graying of population seen accelerating
Source: World Bank, Averting the Old Age Crisis: Policies to Protect the Old and Promote Growth, Oxford University Press, New York, 1994.
Note: Year in which percentage of population over 60 was or will be doubled is shown in parentheses.
Yet, even as the number of old people has been growing, systems of financial support for the old are in trouble worldwide. Both extended families and village support networks, on which two thirds of the world’s old people depend exclusively, tend to break down under pressures of urbanization, industrialization, and increased mobility. When that has happened in the past, governments have stepped in. Public spending on pensions (and, to a lesser extent, on health) has increased rapidly as incomes have grown and populations have aged, and now exceeds 15 percent of GDP in some industrial and transitional economies. But the public systems, too, have been breaking down.
Old-age-security systems affect the welfare of the old and they also affect the entire economy by influencing productivity and the size of the GNP pie. Therefore, countries should use two overarching criteria for old-age programs: they should protect the old and they should promote (or, at least, not hinder) economic growth. Existing programs in many countries fail both tests, yet countries that are starting new systems are on the verge of making the same mistakes.
A recent World Bank study concludes that both criteria would be better satisfied if governments used multiple instruments, or “pillars,” for providing the three main functions of old-age-security programs—saving, redistribution, and insurance. A mandatory, but privately managed and fully funded, pillar would have the primary responsibility for handling saving; a publicly managed, tax-financed pillar would have the primary goal of reducing poverty among the old; and voluntary saving-annuity plans would exist for people who want more protection.
Most formal systems of old-age security are publicly managed, pay pensions that are positively related to the worker’s earnings, and are financed by payroll taxes on a pay-as-you-go basis—meaning that today’s workers are taxed to pay the pensions of those who have already retired. But these systems are plagued by problems today, and many of the same problems are found in industrial and developing countries alike. These problems are not accidental; they are inherent in the political economy of the systems, reinforced by the short time horizons of politicians and their willingness to promise present benefits even if this means incurring large future costs.
High payroll taxes. Under pay-as-you-go systems, when populations are young, small contributions from the large number of workers allow for payment of generous benefits to a few pensioners. As populations age and systems mature, however, these systems must charge high taxes to pay the same benefits to the growing number of retirees. Payroll taxes for pensions are already over 25 percent of gross wages in Brazil, Egypt, Hungary, Italy, the Kyrgyz Republic, and Russia. As populations age over the next 30 years, contribution rates in almost all regions will have to rise dramatically if pay-as-you-go systems and current benefits are retained. High payroll taxes mean less take-home pay or more unemployment.
Evasion. High payroll tax rates that are not linked to benefits also lead to evasion. In many Latin American countries, over 40 percent of the labor force works in the informal sector, partly to avoid payroll taxes, and the informal sector is growing rapidly in Eastern Europe. In Argentina, prior to recent reforms, more than 50 percent of workers evaded their contributions to the pension system. Evasion undermines the system’s ability to pay pensions, makes it necessary to raise payroll taxes still further, and hurts the economy, since people who work in the informal sector are often less productive.
a US citizen, is Lead Economist in the Poverty and Human Resources Division of the World Bank’s Policy Research Department.
Early retirement. In Hungary, more than a quarter of the population are pensioners; the average retirement age is 54; and the payroll tax rate is 31 percent. In Turkey, many people below the age of 50, or even 40, retire with generous pensions. Public sector employees in many countries can retire at 55, or earlier. Early retirees stop making contributions and begin drawing benefits, thus doubly hurting the scheme financially while also depriving the economy of their experience.
Misallocation of public resources. In 1990, Austria, Italy, and Uruguay spent more than one third of their public budgets on pensions. The burden spilled over to their general treasuries as taxes earmarked for old-age systems failed to cover the systems’ growing expenditures. High public pension spending can hurt the economy if it squeezes out government spending on growth-promoting public goods such as infrastructure, education, and health services.
Lost opportunities. Many countries believe that their inadequate national saving rates hamper growth, but they have not used their old-age systems to induce people to save more, and some economists believe existing systems have actually induced people to save less.
Redistribution. Publicly managed systems financed by tax revenues are sometimes justified on the ground that they help the poor. But, despite seemingly progressive benefit formulas, studies of public pension plans in the Netherlands, Sweden, the United Kingdom, and the United States have found little, if any, redistribution from the lifetime rich to the lifetime poor. In fact, studies of such plans in the United States, Colombia, and other countries have found that, in the early years of these plans, high-income groups have benefited the most.
Generally, covered workers who retire in the first 20–30 years of a scheme get back much more than they have contributed, but their children and grandchildren get back less than they paid in and earn lower rates of return than they could have obtained elsewhere. In addition, because of the labor market distortions, the misallocation of public resources, and the failure to increase savings, the future GNP pie is likely to be smaller than it would have been otherwise.
As a result of all these forces, old-age systems are in serious financial trouble. The situation has been most acute in Latin America and Eastern Europe, where systems have been close to bankruptcy. But problems also loom for the member countries of the Organization for Economic Cooperation and Development (OECD), whose implicit public pension debt (the present value of amounts promised to current retirees and workers) far exceeds the explicit conventional debt and sometimes even exceeds 200 percent of GNP. Current payroll tax rates are much lower than those needed to pay off these debts, and most countries will have a hard time raising them enough to close the gaps. Pension benefits or other public goods will therefore have to be reduced in the future. Reforms are necessary to avoid further growth of the debt, but the already large implicit pension debts in many countries make reform difficult.
In fact, many countries have already reneged on their promises by allowing inflation to take place without indexing benefits. In Venezuela, for example, the real value of public pensions fell 60 percent during the 1980s because of inflation. This is devastating to all the old people who depended on these pensions.
Private pension plans have not been a panacea either. In most countries, privately managed plans are voluntary; they cover a minority of workers, mainly from high-income groups; they are only partially funded; their benefits are not always transferable when an employee moves from one job to another; and the tax expenditures involved are large.
Framework for reform
Reform efforts should focus first on the three functions that old-age security systems should perform—saving, redistribution for alleviating long-term poverty, and insurance. Existing public, pay-as-you-go systems have tried to perform all three functions within a single program and, as a result, have not succeeded in doing any of them well. The best way to ensure that the three functions are carried out is through a multipillar system having separate administrative and financing mechanisms, or pillars, for redistribution and saving (Chart 3):
Chart 3Three pillars of old-age-income security
Source: World Bank, Averting the Old Age Crisis: Policies to Protect the Old and Promote Growth, Oxford University Press, New York, 1994.
• a mandatory privately managed, fully funded pillar that requires people to save and manages their savings;
• a publicly managed, tax-financed public pillar for redistribution, to keep old people out of poverty; and
• a voluntary pillar for people who want more savings and larger pensions.
Pillar for mandatory saving. First, there would be a pillar for saving, which would differ dramatically from most existing systems. It would be fully funded, would link benefits actuarially to costs, and would be privately and competitively managed through personal saving plans or occupational pension plans.
• Why mandatory? Because it would require people to save for old age, which everyone should do but some are too shortsighted to do. This would greatly reduce the burden on the tax-financed public pillar.
• Why link benefits to contributions? To discourage evasion and labor market distortions.
• Why fully funded? To make costs clear, so that countries will not make promises that they will be unable to keep later; to prevent large intergenerational transfers, which is especially important in countries with rapidly aging populations; to help build national savings; and to enable higher pensions to be paid by financing them partially through investment returns.
• Why privately managed? To produce the best allocation of capital and the best return on savings. The World Bank study’s data show that most publicly managed pension reserves earned less than privately managed reserves and, in many cases, lost money throughout the 1980s because they were required to be invested in government securities or loans to failing state enterprises at low nominal interest rates that became negative during inflationary periods (Chart 4). Usually they are not permitted to invest in equities, real estate, or foreign assets, thus making it difficult for them to earn high yields and protect themselves against inflation. In addition, publicly managed funds run the risk of encouraging deficit finance and wasteful government spending, because they constitute a hidden and exclusive source of funds whose use is dictated by political, rather than economic, objectives. Competitively managed funded pension plans, in contrast, should spur financial market and private sector development and benefit from international diversification of investments, thereby reducing country-specific risk and enhancing economic growth.
Chart 4Returns on public and private pension plans compared
Sources: World Bank, Averting the Old Age Crisis: Policies to Protect the Old and Promote Growth, Oxford University Press, New York, 1994; and “Technical Annex to Averting the Old Age Crisis” forthcoming discussion paper, 1995.
Note: Simple annual averages are computed for countries with at least five years of data. India, Kenya, Malaysia, Singapore, and Zambia are provident funds. Rates reported are returns credited to worker accounts. Ecuador, Egypt, Peru, Turkey, the United States, and Venezuela are publicly managed reserves of partially funded pension plans. Rates reported are not adjusted for administrative costs. For privately managed occupational plans in the Netherlands, the United Kingdom, and the United States, estimated average net returns have been reported by subtracting one percentage point from simulated average gross returns. Actual average net returns, after expenses, are reported for the Chilean AFPs: average gross returns were 12.3 percent. For the occupational plans and AFPs, actual returns and expenses vary by fund.
1 Old-Age and Survivor’s Insurance.
2 Administradoras de Fondos de Pensiones (Pension Fund Administrators).
Three caveats are essential here: Countries must have at least rudimentary capital markets; considerable government regulation is essential to avoid investments that are overly risky or managers who are fraudulent; and a public pillar is needed to provide a social safety net in case investments fail.
Pillar for redistribution. Second, there would be a public pillar that resembled existing public pension plans, in that it would be publicly managed and tax-financed. Unlike most current systems, however, the reformed public pillar would focus on redistribution, thereby providing a social safety net for the old, particularly the old whose lifetime incomes were low or whose investments in the saving pillar had failed. To accomplish this, the benefit formula could be flat or means-tested, or could provide a minimum pension guarantee. But the benefit formula should not be positively related to earnings, as most public pensions are today.
Why is this important? Because a formula that is positively related to earnings inevitably redistributes to high-wage earners, especially in the first 20–30 years of a pension program. High-income people enter the labor force later and so contribute for fewer years; they live longer and so receive benefits for more years; and they have steep age-earnings profiles; consequently, they end up with high lifetime pension incomes relative to their life-time contributions. And, giving high benefits to high-wage earners while also keeping low- wage earners out of poverty requires a high payroll tax rate, which introduces all the efficiency problems discussed previously. Taxpayers will get more for their money by having an unambiguous and limited objective for the public pillar, while higher benefits are provided for high-wage earners through savings accumulated in the first mandatory pillar.
Pillar for voluntary saving. Finally, there would be a third pillar that would offer additional protection, through voluntary occupational pension or personal saving plans (possibly tax-advantaged), for people who wanted more income in their old age.
The insurance function would be provided jointly by all three pillars. They would all provide annuities to protect the old against costs incurred owing to increased longevity, as well as disability and survivors’ insurance. Broad diversification remains the best way to insure in a very uncertain world. Countries need to remember the old adage that modern finance has rediscovered: don’t put all your eggs in one basket.
This is not an ivory-tower proposal. Real- world examples of multipillar systems include Argentina and Chile, which use mandatory personal saving plans in their funded pillars; Australia and Switzerland, which use mandatory occupational plans; and Denmark and the Netherlands, in which collective bargaining has made occupational plans quasi-mandatory.
How to get there
While they all share the same ultimate goal, regions vary widely in the speed at which their populations are aging, in the problems faced by existing systems, and in the capacity of their governments and private financial sectors to undertake reform.
Problems are most pressing in the southern part of Latin America and the former centrally planned economies, where contribution rates, deficits, and evasion are high and growing. Argentina, Bolivia, Colombia, and Peru, whose old systems have broken down, are now in the process of implementing sweeping changes similar to those already made in Chile, which restructured its system 13 years ago. And Eastern Europe is weighing the radical change against more moderate reforms—such as raising the retirement age, reducing benefits, and improving record-keeping methods—but has not yet decided which way to go. The challenge for these countries is to find a politically viable transition path to a more economically efficient system. However, they are finding it difficult to move away from their pay-as-you-go systems, because of their large public pension debts: old people are unwilling to give up the pension promises that were made to them years ago, while young people are unwilling to pay the taxes needed to sustain or reform the systems. In some countries, the result has been political paralysis and economic deterioration.
The OECD countries will face similar problems as their populations age. High tax rates and low returns to the contributions of later generations will pose economic and political problems. If these countries start now, they can accomplish the transition more gradually, using productivity growth to reform the system—raising the retirement age, flattening out the benefit structure, and holding average benefits constant as wages rise, while raising contributions slightly and allocating a growing part to the mandatory saving pillar. The challenge for these countries is to start building this pillar soon, so the transition can be gradual and a crisis can be averted.
East Asia is the most rapidly aging and growing region. It will have to act quickly to meet the double challenge of aging populations and erosion of the extended family. Some of the East Asian countries are presently evaluating alternative directions for their new and expanding systems. For them, a system that has a large funded component will be better protected against rapid demographic change than a pay-as-you-go system. These countries already have, or could quickly develop, the financial and regulatory capacity to institute a competitively managed mandatory saving scheme. The challenge they face is to avoid the problems that confront Latin America and Europe, recognizing that it would be better to start down a different path because, once it is adopted, a pay-as-you-go system is difficult to reverse.
In sub-Saharan Africa and South Asia, informal systems are still strong and formal systems are limited in scope. The big danger is that these countries will expand their formal systems in ways that are not desirable or sustainable. The challenge they face is to avoid accelerating the decline of informal systems, rushing ahead with complex formal systems that they lack the capacity to administer, and making overly generous pension promises that they will be unable to keep. They should keep their public plans modest and simple while developing the human capital, infrastructure, financial markets, and regulatory capacities that will allow them to establish a private pillar later on.
Although the kind of reform that is needed will vary across countries, all countries should have a vision of where they are going in the long run. This vision should include separate mechanisms for redistribution and saving, and it should share responsibility between the public and private sectors.
The right share of responsibility for each pillar will not be the same for all times and places. It depends on a country’s objectives, history, and current circumstances, particularly on the relative strength of its redistributive and saving goals, its degree of financial market development, and its taxing and regulatory capabilities. The pace at which a mandatory multipillar system should be introduced will also vary—from quick in middle- income countries whose systems are in serious trouble, to gradual in high-income countries where the problems are less pressing, to very slow in low-income countries that lack the necessary financial and regulatory capacities.
Tackling pension reform will not be easy, and talk of reform elicits strong emotions, especially from people whose expectations about receiving generous pensions have been built up through the years. Nonetheless, change in the current systems is inevitable, and the longer reform is delayed, the more urgent and difficult it becomes. Countries can avert the old-age crisis. To do so, they should begin planning and educating the public now.
This article is based on a World Bank study. Averting the Old Age Crisis: Policies to Protect the Old and Promote Growth, Oxford University Press, New York, 1994.