It is almost an optical illusion: looming on Japan's horizon, and on Europe's and on America's, is a pensions crisis. The problem is real, though exaggerated. The illusion is in some of the plans being devised to deal with it.
The main question is whether privatising pension systems, as George Bush has proposed for social security in the United States, would solve the problem or merely make matters worse. With many countries pondering whether to adopt variants of the Bush plan, the question requires careful examination.
By itself, privatisation is clearly not the solution. America's troubled private pension system - now several hundred billion dollars in debt - already appears to be heading for a government bail-out. There was a time when privatisation - allowing individuals to set up individual savings accounts - seemed better than social security, which invests in lower-yielding Treasury bills (government bonds). Advocates of privatisation argued that funds would do much better if invested in stocks, predicting a return of 9%.
But the stock market does not guarantee returns; it does not even guarantee that the stock values will keep up with inflation - and there have been periods in which they have not. America's social security system insulates individuals against the vagaries of the market and inflation, providing a form of insurance that the private market does not offer.
It does so with remarkable efficiency. The costs of managing the social security system are far smaller than those likely to be associated with privatised accounts. This is understandable: private investment firms spend an enormous amount on marketing and salaries.
It is possible that to reduce these transaction costs, Bush will propose restricting choice, which was the main argument for privatisation in the first place. But these limited kinds of choices - for example, a T-bill fund with 90% in T-bills and 10% in an indexed stock fund - could easily be introduced into the public social security system.
Bush says that reform is urgently needed, because the system will be insolvent in about a quarter of a century. But the problem depends on America's growth rate: if the growth rates of the late 1990s return, there is no problem. Even if there is a problem, it can easily be fixed: spending a fraction of the money that went into Bush's two tax cuts would have fixed social security for 75 years; slight benefit cuts, adjusting the age of retirement, or minor adjustments in the level of contributions could fix the system permanently.
Moreover, Bush's proposals won't fix social security - unless they are accompanied by drastic benefit cuts. For how could they? He proposes diverting almost a third of the social security tax to private accounts. That means less money coming in. If benefits are not reduced, the gap between receipts and expenditures will increase. One doesn't need a Nobel prize to figure that out.
So privatisation would not protect retirees against the social security system's insolvency; it would merely add enormously to today's fiscal deficit, because partial privatisation entails diverting money to private funds that would have been used to close the gap between government expenditures and revenue.
The anticipated increase in the fiscal deficit is striking. The central plan discussed by Bush's council of economic advisers would - according to the council's own estimates - increase America's fiscal deficit by $2 trillion over the next decade. Advocates of privatisation claim to believe in markets, but they are proposing budget gimmickry that would move those losses off the books, as if markets could be easily fooled.
America and the world should remember: Argentina's privatisation of its pension system was at the centre of its recent fiscal woes. Had Argentina not privatised, its budget would have been roughly in balance. The US is starting on its privatisation venture with a fiscal deficit of 4% of GDP.
Privatisation advocates insist, however, that investments in stocks would yield sufficiently higher returns to give individuals the same retirement income as before, with the surplus used to fill the gap. But if markets are working well, then returns will be higher only because risk is higher. There is still no free lunch in economics.
With higher risk, there is a chance that, 40 years from now, many individuals will find themselves with less than they need to retire. But if one really thinks that free lunches exist, there is still no reason to privatise: the government could get the additional returns by investing in the stock market itself. Indeed, President Clinton proposed doing just that.
With increased transaction costs, worsening solvency for the system, increased budget deficits and decreasing benefits and security for retirees, why the drive for privatisation? One reason is the interest financial markets have in grabbing a piece of all those transaction costs. A second is the Bush administration's ideological hostility to the modest amount of wealth redistribution implied by the public system. America's social security programme has been so successful in reducing poverty because the poor get back a little more than they contribute, and the rich get back a little less.
Even with social security's mildly redistributive effect, poverty and inequality in America are increasing. Privatisation will only make matters worse.
Bush has tried to scare America about the magnitude of the problem, and he has tried to fool America about how privatisation would solve it. The social security deficit pales by comparison with the deficits created by Bush's huge tax cuts for upper-income Americans or in comparison with the deficit in Medicare, which provides healthcare for the aged. Why has he ignored these problems? Is there another agenda?
Joseph Stiglitz is professor of economics at Columbia University and a Nobel prize winner. Project Syndicate www.project-syndicate.org
© 2005 Guardian Newspapers, Ltd.