Share This
September 2001 Cover Art

Search Finance & Development

Advanced Search

About F&D


Back Issues

Write Us

Copyright Information

Free Email Notification

Receive emails when we post new items of interest to you.

Subscribe or Modify your profile

Finance & Development
A quarterly magazine of the IMF
September 2001, Volume 38, Number 3

The Truth About Pension Reform
Nicholas Barr

Aging populations in industrial and transition countries have provoked heated debate about pension reform—in particular, about the desirability of abandoning pay-as-you-go schemes in favor of private, funded pensions. What kind of pension plan would best meet the needs of future retirees?

There are two ways we can provide for a secure old age. We can save part of our wages each week and draw on the accumulated funds after we retire to buy goods produced by younger people. This is the principle underlying funded pension plans. Or we can obtain a promise—from our children or our government—that, after we retire, we will be given goods produced by others. This is broadly the way pay-as-you-go (PAYG) systems, with pensions paid out of tax revenues, are organized. Both types of pension plan thus exchange current production for claims on future production, but there is considerable controversy about which is the better choice. This article investigates arguments suggesting that funded plans are superior. Although it finds that those arguments can be overstated, it does not mean to discredit funded plans themselves, merely some of the claims that are made about them.

Funding versus PAYG

The working-age population in most industrial countries will soon have to support increasing numbers of pensioners, arousing fears that PAYG systems will become unsustainable. Many—including the new U.S. administration—are arguing that future retirees would be better off investing at least part of their savings in private, funded pension plans. But funded plans face many of the same problems as PAYG systems; some of their claimed advantages are more myth than fact. And PAYG state social insurance systems are more flexible than most people realize; many of the problems they now face can be solved, as Sweden demonstrates (see box).

How does Sweden's PAYG system work?

Sweden introduced a "notional defined-contribution" scheme in 1998. The state pension is financed by a social insurance contribution of 18.5 percent of earnings, of which 16 percent goes into the public scheme. Although this year's contributions pay this year's pensions, the social insurance authorities open a notional account that keeps track of each person's contributions. The account attracts a notional interest rate reflecting average income growth. A person's pension is based on his or her notional lump sum at retirement (in other words, social insurance mimics annuities) and on projections about life expectancy and future output growth. There is a safety-net pension for people with low lifetime earnings, and periods spent caring for children carry pension rights. The remaining 2.5 percent of a person's contribution goes into a funded scheme—either a private account or a government-managed savings fund. The individual can choose to retire earlier or later, with the pension being actuarially adjusted accordingly.

Thus, Sweden has a defined-contribution scheme with a safety net. It is a publicly organized, PAYG analogue of Chile's privately managed, competitive, individually funded arrangements. Either method gives people choices, for example about their preferred trade-offs between the duration of retirement and living standards in retirement, while presenting them with the actuarial costs of those choices.

Myth 1: Funding resolves adverse demographics. Population aging reduces the workforce and, as a result, a country's output. The effect on PAYG schemes is that the contribution base shrinks. The effect on funded schemes is more subtle but equally inescapable. When a large generation of workers retires, it liquidates its financial assets to pay for its pensions. If those assets are equities, sales of financial assets by the large pensioner generation will exceed purchases of assets by the smaller younger generation, leading to falling equity prices and, hence, to lower pensions. Alternatively, if those assets are bank accounts, high spending by the large pensioner generation will generate inflationary pressures and—again—reduce the value of pensions.

Myth 2: The only way to prefund is through pension accumulations. Future retirees can protect themselves against shifting demographics in ways that do not involve pension funds. First, they can invest in countries with younger populations. Second, governments can cut future public spending to offset expected increases in PAYG pension spending—for example, by paying off some public debt now to reduce future debt interest payments. Third, government can set aside resources to meet increased future demands. Norway, for example, puts part of its oil revenues into a fund to smooth taxes in the face of demographic change.

Myth 3: There is a direct link between funding and growth. It is often regarded as self-evident that saving—and, hence, economic growth—will be higher with funded schemes. In a famous 1973 paper, economist Martin Feldstein claimed that the United States' PAYG social security system reduced personal saving by about 50 percent, and the country's capital stock by 38 percent. The connection between funding and growth is complex, however. First, savings rise only during the buildup of funded schemes; once schemes are mature, saving by workers is matched by payments to pensioners. Second, even during the buildup, increases in mandatory saving may be at least partly offset by reductions in voluntary saving. Third, saving does not necessarily lead to new investment (a British trade union once famously invested part of its pension fund in old masters). Finally, investment does not lead automatically to growth: during the last days of communism, investment rates in the centrally planned economies were high but growth was stagnant or negative. Even in well-run economies, it cannot be assumed that pension fund managers make more efficient choices than other agents in channeling resources into their most productive uses. Separately, funded schemes could assist growth by contributing to the development of capital markets—but only if other conditions are right. Thus, although the argument that funded pensions boost growth might have some validity, it should not be seen as automatically or always true.

Myth 4: Funding reduces public spending on pensions. It is true that private pensions reduce public pension spending in the longer term, once private schemes are mature. In the shorter term, however, introducing private pensions is likely to increase budgetary pressures: if workers' contributions go into their individual pension accounts, they cannot be used to pay for the pensions of the older generation; thus, governments have to finance pensions for the transition generation through taxation or borrowing.

Myth 5: Funded schemes offer better labor market incentives. Labor market distortions are minimized when contributions bear a clear actuarial relationship to benefits. Private pensions may have these characteristics, but so do some state schemes that pay benefits strictly proportional to contributions. In contrast, badly designed schemes cause labor market distortions. Some pension plans—public and private alike—encourage early retirement by increasing pensions for work beyond the normal age of retirement by less than the actuarial amount. Many employer plans encourage labor immobility (public schemes, being universal, do not have this problem). Thus, labor supply depends more on pension design than on whether a pension is private or public.

Myth 6: Funded pensions diversify risk. The argument of risk diversification should not be overstated. First, PAYG systems and funded schemes are both vulnerable to macroeconomic, demographic, and political shocks. Second, private pensions face additional risks. Fund management may be fraudulent, or it may be honest but incompetent. Thus, substantial regulation is required to protect consumers. Even if a fund is managed competently, there is a risk that its investment performance will not meet expectations or that stock market fluctuations will cause benefits to vary widely. Two individuals with identical lifetime contribution profiles could end up with very different pensions. Finally, the risk-diversification argument applies to state pensions as much as to private pensions and is thus logically incompatible with the view that PAYG pensions should be minimized.

Myth 7: Increased choice is welfare-improving. An increase in the number of pension plans may be desirable if consumers know enough to choose well. However, pensions are complex even for financially sophisticated consumers. The Chairman of the U.S. Securities and Exchange Commission has said that more than 50 percent of Americans do not know the difference between a bond and an equity. A second issue involves the cost of such choices. Pension plans in Chile and the United Kingdom, both of which rely to a significant extent on individual funded accounts, have high administrative costs; and because the costs of maintaining a pension account are largely fixed, they bear most heavily on small pension accounts—those of low earners.

Myth 8: Funded schemes do better than PAYG systems if real returns exceed real wage growth. It is often argued that funded schemes provide larger pensions than PAYG systems because stock market returns are higher than the returns offered by state schemes. Though often true in a brand-new world, this argument is not necessarily true in a country that already has a PAYG scheme and is moving to a funded scheme.
Contributions and benefits under a simplified pay-as-you-go system

Source: Orszag (1999).
The table illustrates the argument at its simplest. In period 1, the $1 pension of older generation A is paid by the $1 contribution of younger generation B. In period 2, when generation B has retired, its pension is paid by the contributions of generation C. Suppose that the real rate of return on assets is 10 percent and that we are generation C. As members of the PAYG system, we pay contributions of $1 in period 2 and receive a pension of $1 in period 3; the real rate of return is zero. If, in contrast, we join a funded scheme, we would save $1 in period 2 and get back $1.10 in period 3; the real rate of return, it appears, is 10 percent.

But the real rate of return is not 10 percent. If generation C contributes to funded pensions, generation B's pension must be paid from some other source. If that source is government borrowing, and if the interest on that borrowing is paid by the older generation, generation C receives a pension of $1.10 but has to pay interest of 10 cents on the debt that financed generation B's pension. The real return—as under the PAYG scheme—is zero. The lower return on the PAYG system is the cost of the initial "gift" to generation A.

The argument can be extended to cases where the move to funding is paid out of taxes. The real return may also be less than it appears once account is taken of the differences in risk and administrative costs between the two systems. The benefits to future generations of a move from PAYG to funding, again, are neither automatic nor inevitable.

Myth 9: Private pensions get government out of the pensions business. It is well known that public schemes are vulnerable to government failure—but so are private pensions. Fiscal imprudence can lead to inflation, eroding the stability of private funds. In addition, if government regulation is ineffective, financial markets will fail to channel savings into efficient and productive investment, thus squandering the gains private pensions were intended to engender. Effective government is essential for any type of pension scheme.

It is sometimes argued that funded schemes are safer from government depredations than PAYG schemes. This is not necessarily true. Governments can, indeed, renege on their PAYG promises, but they can also reduce the real return to pension funds—for example, by requiring fund managers to hold government financial assets with lower yields than they could earn on the stock market or by reducing the fund's tax privileges.

Pension design: essentials

Policymakers have many choices in designing pensions—but only if key prerequisites are met. In the public sector, first, the state pension plan must be fiscally sustainable. This does not mean that public pension spending must be minimized (as opposed to optimized), but that it must be compatible with continued economic growth. Second, pension systems must be politically sustainable. Reform of pensions—whether PAYG or funded—is not an event but a process, requiring support from all levels of government. Both fiscal and political sustainability depend on the government's ability to collect contributions and maintain macroeconomic stability and, for private pensions, to regulate financial markets effectively.

Private pensions have additional private sector prerequisites. There need, first, to be adequate public understanding of, and trust in, private financial instruments. Second, there have to be financial assets and financial markets. (Some poor countries try to get around their lack of domestic assets and financial markets by using pension savings to buy financial assets in more advanced countries; however, this approach does not stimulate domestic investment and employment.) Third, private sector administrative capacity is essential, given the considerable administrative requirements of funded pensions.

Pension design: policy choices

The core objectives of old-age security are poverty relief, consumption smoothing over an individual's lifetime, and insurance. Since the World Bank's highly publicized study of pension reform in 1994, many observers have advocated a multipillar pension system, "optimally consisting of a mandatory, publicly managed, unfunded pillar, and a mandatory but privately managed funded pillar, as well as supplemental, voluntary, private funded schemes. . . ." Substituting the word tier for pillar (unlike pillars, tiers can be additive and set up in whatever constellation one wishes), I characterize the first-tier pension as intended primarily to provide poverty relief; though normally set up as a public PAYG system, it can take other forms, including finance through general taxation. The second tier provides consumption smoothing; it can be publicly or privately managed, funded or PAYG, and integrated into or separate from the first tier. The third tier is private, funded, and voluntary.

Although the prerequisites discussed earlier are inescapable, policymakers have a wide range of choice, as the following sample of questions illustrate.

First, should the first tier be a guarantee, available only (or mainly) to those who need it, or a base on which to build other pension income? In Chile, only the poorest receive state pensions; the first tier is designed to be a state guarantee of income for individuals in private schemes. Australia awards state pensions on the basis of an affluence test (the best off are not eligible). In other countries, the first-tier pension is paid at a flat rate to all pensioners: the flat rate falls below the poverty line in many poorer countries, is broadly equal to the poverty line (the United Kingdom), or is above the poverty line (New Zealand).

How redistributive should the first tier be? The smaller the pension and the more closely benefits are linked to contributions, the less redistribution of income there is between rich and poor or between different generations. A flat-rate pension financed by a proportional contribution is more redistributive and a flat-rate pension financed by progressive general taxes more redistributive still.

Should a second-tier pension be mandatory? Most commentators say yes, because imperfectly informed young people will make suboptimal choices about saving, or because a mandatory pension provides insurance against unforeseeable events, or for other reasons.

How should the second tier be designed? In defined-contribution schemes, pensions are based on the size of pension accumulations; in defined-benefit schemes, pensions are related to wages. Most of the risk associated with defined-contribution funds is borne by the individual, while most occupational schemes are defined-benefit plans that share risks broadly among industry workers, company shareholders, and taxpayers.

Should a second-tier pension plan be PAYG or funded? In the United States, the first- and second-tier pensions are rolled into one, both mainly PAYG. In Canada, a first-tier state pension provides poverty relief, and a mandatory, publicly organized PAYG second-tier pension provides consumption smoothing. Australia and several Latin American countries have privately managed, funded, mandatory second-tier pensions. The United Kingdom has a mixed system: the basic flat-rate state pension is mandatory; the second tier is also mandatory, but consumers can choose among the state earnings-related pension scheme (which is PAYG), approved occupational schemes (private, funded, frequently defined benefit), or individual funded accounts.

Another key question is whether, to what extent, and how governments should protect pensioners against inflation. When governments provide indexation or other assistance, funded schemes acquire an unfunded element.

Although countries have a wide range of choices, they cannot pick and mix without regard for their particular circumstances. Countries with mature PAYG systems and aging populations need to adopt policies that increase output and reduce the generosity of the pension system (for example, by raising the age of retirement); prefunding future pension needs could be one element in the policy mix. Countries with large, unsustainable PAYG systems have little choice: they must reduce benefits, increase contributions, or both. Countries with limited institutional capacity also have little choice. In the poorest, administratively weakest countries, the issue is how to organize poverty relief; as their ability to collect taxes increases, they might move to a tax-financed (probably flat-rate) pension; as public administrative capacity grows, they may adopt a contributory system; and as incomes rise and private administrative capacity grows, private pensions become an option.

This article draws on Nicholas Barr, 2000, "Reforming Pensions: Myths, Truths, and Policy Choices," IMF Working Paper No. 00/139 (Washington: International Monetary Fund). For a fuller discussion, see Nicholas Barr, 2001, The Welfare State as Piggy Bank: Information, Risk, Uncertainty, and the Role of the State (New York: Oxford University Press) (

Martin S. Feldstein, 1973, "Social Security, Induced Retirement, and Aggregate Capital Accumulation," Harvard Institute of Economic Research Discussion Paper No. 312 (Cambridge, Massachusetts: Harvard University).

Peter Orszag, 1999, Individual Accounts and Social Security: Does Social Security Really Provide a Lower Rate of Return? (Washington: Center on Budget and Policy Priorities);

World Bank, 1994, Averting the Old Age Crisis, World Bank Policy Research Report (New York: Oxford University Press for the World Bank).

Nicholas Barr is a member of the Economics Department at the London School of Economics and the author of numerous books and articles on the economics of the welfare state ( He was a Visiting Scholar in the IMF's Fiscal Affairs Department in the spring of 2000 and a principal author of the World Bank's World Development Report 1996: From Plan to Market.