The Commission to Strengthen Social Security focused on the latter approach, describing three plans to redirect Social Security contributions into individually owned, privately invested retirement accounts.
Regardless of one's views about the wisdom of private investment accounts, it is important to recognize that nothing prevents people from saving more than their Social Security contributions now.
In fact, existing tax-deferred plans -- individual retirement accounts, Keoghs and the like -- offer attractive subsidies for private saving. Despite these significant incentives, a remarkably large component of the population saves almost nothing for retirement, making them highly dependent on Social Security benefits.
Why are personal savings so low?
Stephen F. Venti and David A. Wise, economists at Dartmouth and Harvard, respectively, recently released a fascinating glimpse of savings behavior based on data from the Health and Retirement Survey, a continuing study of approximately 7,700 households nearing retirement age. (''Choice, Chance and Wealth Dispersion at Retirement,'' National Bureau of Economic Research Working Paper 7521.)
Mr. Venti and Mr. Wise started their analysis by estimating the lifetime income of each household, then sorted the households into 10 equal-sized groups based on their estimate. Their most striking observation was the extreme variation in total asset accumulation within each income group. For example, the wealth held by the top 10 percent of households in the group just below the median was 35 times the wealth held by the bottom 10 percent of that same income group. The dispersion of wealth for higher-income groups was slightly smaller, but still high.
(The authors consider various definitions of wealth, some limited to financial assets, some including housing, some including the actuarial value of Social Security benefits. I use the most inclusive definition, but the other definitions tell the same story.)
It has long been known that most Americans do not save much. What is more remarkable is that even a substantial fraction of relatively wealthy people do not save much. The bottom 20 percent of every income group has zero or negative wealth, with the only exception being the households with the highest lifetime income.
Just as some high-income households had minimal accumulated wealth, some low-income households had substantial wealth. The top 10 percent of the lowest income group had accumulated more than $150,000.
Even though nearly every group in the sample had the opportunity to contribute to an I.R.A. or a 401(k) plan, only about half the households did so.
Mr. Venti and Mr. Wise consider three explanations for the large variations in wealth accumulated for retirement: household wealth was significantly affected by chance events like illness or inheritances; household investments performed far differently; and some people saved substantially more than others.
They cannot offer a definitive answer, because their data was only a snapshot at a single point in time, but they offer suggestive evidence that the first two effects were relatively small. Even after one accounts for chance events and investment performance, the dispersion in wealth remains large. They concluded that most of the observed variation in retirement wealth is primarily a result of different propensities to save.
Reluctance to save is costly. In the data Mr. Venti and Mr. Wise examined, the middle-income households had, on average, financial assets totaling about $45,000. If these households had saved 10 percent of their income and invested it in a stock market index fund, they would have had more than $250,000 of wealth by retirement age. Even if they had invested in long-term corporate bonds, they would have accumulated over $100,000.
These households were generally aware that their savings were inadequate: over three-fourths of the respondents said they had saved too little, and virtually none said they had saved too much. Most people were worried by their minimal savings: only a quarter of them thought that ''Social Security or employer pensions would take care of my retirement income.''
It seems that people just have very different attitudes toward saving: some people do it, and some do not. I have my own theory about why there is such a low personal saving rate: I blame it on advertising. Businesses spend billions of dollars encouraging people to spend, and many people just cannot resist the temptation.
By contrast, only a pittance is devoted to advertising that encourages people to save. A typical ad for an I.R.A. shows a gray-haired middle-aged man playing golf. A more realistic ad might depict a bag lady sitting next to a grocery cart. O.K., that is a bit extreme -- but that sort of ad would send the right message: look what happens to you if you do not save.
Other analyses of consumer behavior show that I.R.A.'s and 401(k) plans have increased aggregate savings. But less than half the population takes advantage of such plans. There should be more plans aimed at the lower end of the income distribution.
It is not impossible to save $5 or $10 a day -- many people spend that much on lottery tickets and cigarettes. But $10 a day is $3,650 a year, and that amount can accumulate significantly over a lifetime.
As the Venti-Wise study shows, many low-income households save this much now. And many high-income households, to whom $10 a day is surely insignificant, do not bother to save even that tiny amount.
President Bush's Social Security commission was not asked to consider plans to encourage more voluntary saving. Instead, it focused on how mandated savings -- Social Security -- will be structured. One would hope that some part of whatever plan emerges from Congress's deliberations will offer better ways to encourage Americans to save more in addition to their Social Security contributions.