Every society faces the difficult task of providing support for older people who are no longer working. In an earlier era, retirees lived with their adult children, providing childcare and helping around the house. But those days are largely gone. Retirees and their adult children alike prefer living independently.

In a rational economic world, individuals would save during their working years, accumulating enough to purchase an annuity that finances a comfortable standard of living when they retire. But that is not what most people do, either because of their shortsightedness or because of the incentives created by the government social security programs.

European governments since Otto von Bismarck and US governments since Franklin Roosevelt have therefore maintained pay-as-you-go (PAYG) retirement pension systems. More recently, Japan has adopted such a system.

But providing benefits to support a comfortable standard of living for retirees with just a modest rate of tax on the working population depends on there being a small number of pensioners relative to the number of taxpayers. That was true in these programs’ early years, but maintaining benefit levels became more difficult as more workers lived long enough to retire and longer after retirement, which increased the ratio of retirees to the taxpaying population.

Life expectancy in the United States, for example, has increased from 63 years in 1940, when the US Social Security program began, to 78 years in 2017. In 1960, there were five workers per retiree; today, there are only three. Looking ahead, the Social Security Administration’s actuaries forecast that the number of workers per retiree will decline to two by 2030. That implies that the tax rate needed to achieve the current benefit structure would have to rise from 12% today to 18% in 2030. Other major countries face a similar problem.

If it is not politically possible to raise the tax rate to support future retirees with the current structure of benefits, there are only two options to avoid a collapse of the entire system. One option is to slow the future growth of benefits so that they can be financed without a substantial tax increase. The other is to shift from a pure PAYG system to a mixed system that supplements fixed benefits with returns from financial investments.

A US example shows how slowing the growth of benefits might work in a politically acceptable way. In 1983, the age at which one became eligible to receive full Social Security benefits was raised from 65 to 67. This effective benefit reduction was politically possible because the change began only after a substantial delay and has since been phased in over several decades. Moreover, individuals are still eligible to receive benefits as early as age 62 with an actuarial adjustment.

Since that change was enacted, the life expectancy of someone in their mid-sixties has increased by about three years, continuing a pattern of one-year-per-decade increases in longevity for someone of that age. Some economists, including me, now advocate raising the age for full benefits by another three years, to 70, and then indexing the future age for full benefits to keep the life expectancy of beneficiaries unchanged.

Consider the second option: combining the PAYG system with financial investments. Pension systems operated by private companies achieve benefits at a lower cost by investing in portfolios of stocks and bonds. A typical US private pension has 60% of its assets in equities and the remaining 40% in high quality bonds, providing a real (inflation-adjusted) rate of return of about 5.5% over long periods of time. In contrast, taxes collected for a PAYG system produce a real rate of return of about 2% without investing in financial assets, because real wages and the number of taxpayers rise.

It would be possible to replace the existing PAYG systems gradually with a pure investment-based system that produces the same expected level of benefits with a much lower tax rate. Unfortunately, the benefits produced by that contribution rate would entail significant risk that the benefits would be substantially below the expected level.

First « 1 2 » Next