What we don’t know

Do you know how long you will live? Do you know how long the average American will live 50 years from now? Do you know what birthrates will be for the next 50 years? Do you know the rate of immigration for the next 50 years? Do you know the rate of economic growth for the next 50 years?

Of course, neither you nor anyone else the answer to any of the above questions. However, those who tell you we do not need to change Social Security or need only make minor adjustments to the existing system can honestly do so if they know the answers to the above questions — which they do not.

Let’s start with what we do know. The present Social Security system is a “pay-as-you-go” system, in which the taxes paid by workers and their employers are used to fund the monthly benefit checks for the existing retirees. We know that in 1950 there were 16.5 workers paying Social Security taxes for every retired person receiving benefits. We know there are now only about 3.3 workers for each person receiving benefits and there probably will only be 2.2 workers for each benefit recipient 25 years from now in 2030, if we continue with the existing system. We also know the so-called Social Security Trust Fund actually contains no money, because Congress has spent all the money (the surplus from the Social Security tax over actual outlays) on other things since the program’s 1937 inception.

To fully understand the problem, let us go back to what we do not know.

One of the biggest unknowns is future life expectancy. The Social Security actuaries estimate only a two-year increase in life expectancy between 2000 and 2050. However, for the last 125 years, life expectancy has been increasing by about three months per year. If this trend continues, which many highly respected biochemists believe it will, average life expectancy would increase by about 12 years between now and 2050, not two years. Up until the 1940s, life expectancy for all ages increased because of improved water, sewage, food handling and vaccinations. By the 1970s, U.S. infant and child mortality was so low the increased life expectancy was nearly all at the upper age groups as heart disease and cancer prevention programs became widespread. Now, great advances are being made against the aging process, and there is no reason to believe these efforts will not continue increasing life expectancy at the present rate and perhaps even accelerate it.

Another unknown is future birthrates, which have fallen sharply in all developed countries in the last half-century. At present, the average American woman has slightly less than two children, a little bit lower than replacement (2.1 children per woman). But what happens if birthrates fall as low as they have in Italy (1.2 children per woman)? Where will the future workers come from to support all of the retirees?

This leads to the question of immigration. Because of high legal and illegal immigration into the U.S., the working population continues growing at a good pace, despite low birthrates among native-born American women. There is growing political pressure to reduce immigration, and if this succeeds it will only increase pressures on the existing Social Security program.

The future rate of economic growth is also one of the great unknowns. If we could sustain a 4 percent growth or more, the problem of a growing number of retirees will be manageable. But if our average economic growth falls to 2 percent or less, as it has in “old Europe” or Japan, we will be in very deep trouble. Those who advocate higher taxes to “bail out” the existing Social Security system often fail to recognize the higher taxes will slow economic growth, which may make things worse rather than better.

Given the future unknowns and the lessons from our previous experience, it is clear if we had to start over again with a retirement system, it would have the following characteristics: (1) Individual citizens would have their own accounts to which they would contribute over their working lives (there probably would be both mandatory and voluntary components of the contribution). Permissible investments would be some combination of government and private bonds and other securities.

(2) Individuals could select their own retirement age, depending on the return of their retirement investment portfolio, their health and their choice. The longer you worked and the more you contributed, the more retirement income you would have.

Some people would choose to retire at 50 and others at 90. The strictly government program would be limited to those whose lifetime earnings were too low to provide a minimally adequate retirement, and those through premature ill health or other misfortune were not able to provide for themselves.

As we travel toward Social Security, the first step should be to recognize explicitly what we do not and cannot know in advance, and then design a system to deal with that reality.

Then, we should go back and work out the least painful and economically destructive transition system.

Richard W. Rahn is a senior fellow of the Discovery Institute and an adjunct scholar of the Cato Institute.