Over the past 75 years, government-sponsored pension systems have changed dramatically. Whether today's programme is unsustainable depends on how one views assumptions about economic growth and demographics far into the future.
But when thinking about retirement, it is important to keep two things in mind. First, someone has to bear the risk of providing income to the elderly; and second, the way in which we do it affects the size of the economy and productivity.
Risk-bearing and insurance are central to any pension system. Insurance allows us to engage in productive activities we would otherwise avoid. But sometimes the government needs to get involved. Most people agree that we should all have a guaranteed level of income even if our jobs disappear. But private insurers will not sell unemployment insurance for two reasons: policyholders could misbehave at work, increasing the odds of collecting on the policy; and in the event of a depression, the private fund would not be able to make the necessary payments.
The possibility of living a long life creates a risk that one's savings will be exhausted before dying. Well-designed pension systems insure that risk. But the government is the only organization that can reasonably pool the risks of the entire population.
Moving to a system of individual pension accounts shifts this income risk from the population as a whole back to individuals, eliminating the insurance component of the system. This is the compelling economic argument for a government-sponsored system; one that insures against elderly poverty.
How should such a system be financed? Current government-sponsored pensions systems are "pay-as-you-go": taxes on working people finance payments to retirees. In the US, there is also a trust fund that is accumulating assets. But as these assets are US Treasury securities, they must be paid for by future tax revenues from the next generation of workers.
Social Security has the added drawback that it masks the true size of the Federal government's budget deficit by nearly $150bn per year. The amount is estimated to rise to nearly $250bn per year by the end of the decade.
The problem with a pay-as-you-go system is that individuals treat their tax payments as retirement savings. Politicians encourage Americans to think of Social Security as equivalent to a private pension system in which payments accumulate in an account that is converted into an annuity at retirement. As a result people do not save. This lower private saving in turn lowers investment and leaves the economy less productive.
A well-designed government pension system must insure the incomes of the elderly in a way that does not reduce economic growth. At first, it might seem that balancing both parts of this objective leads to at least partial privatization. But political realities suggest that such a system will break down. If the private accounts do poorly, so benefits fall short of expectations, then the government becomes the insurer of last resort. In other words, when the stock market produces high returns, account holders win, and when the market does poorly, taxpayers lose.
The solution is for the government-run system to accumulate a real trust fund, one invested in privately issued stocks and bonds. Think of it as a giant version of the California Public Employee Retirement System, that has more than $150bn in assets to finance the retirement of 1.4m people.
I am not recommending that Social Security starts selling its US Treasury bonds and uses the proceeds to purchase stocks. This would just turn the government into a leveraged buyer of equities without increasing saving. Instead, as a start, why not invest the current surplus in privately-issued stocks and bonds? To avoid conflicts of interest, the funds would be invested in a way that tracks the broadest indices.
Since the Bush administration is committed to a public discussion of Social Security, make it a productive one. Our goal must be to emerge with a programme that ensures the elderly do not go hungry while at the same time increasing the level of saving, thereby encouraging economic growth.
Stephen G. Cecchetti is Professor of International Economics and Finance, Brandeis University, and Research Associate, National Bureau of Economic Research. He is a regular contributor to the Financial Times.