For the middle of the Great Depression, it was a masterstroke. The savings of millions of people had been wiped out. Few people owned their own home. The post war system of corporate pensions did not exist.

Worse, some of those people were getting old. Many were too old for the tough demands of manufacturing jobs--- and manufacturing jobs dominated the economy.

The solution was Social Security.

Put a small tax on all workers. Use it to provide a basic income for all retirees. So from 1937 through 1949 the combined employer/employee tax of 2 percent of the first $3,000 of wages cost each worker $30. It also cost his employer $30.

On the other side of the tax, the benefits were amazing. A tiny 'investment' in Social Security employment taxes produced an enormous virtual return in tax-free lifetime income for retirees. For workers born in 1880 and retiring in 1945 the 'return' was about 32 percent.   And that was after adjustment for inflation.

In effect, the employment tax created a 'virtual portfolio' for workers --- a guaranteed lifetime income that was like having a large portfolio of stocks, bonds, and cash without the bother of saving, managing, and taking risk.

Retirees who otherwise would have been penniless had money for food, clothing, and shelter. Retiree benefits created demand that boosted employment and made the economy stronger.

Unfortunately, it didn't stay that way.

The combined rate for the employment tax went to 3 percent in 1950. It hit 6 percent in 1960. It doubled again to 12.1 percent in 1978. It reached its present level, 15.3 percent, in 1990. Today, the employment tax (which includes Medicare as well as retirement benefits) takes 15.3 percent of the first $84,900 in wages. The tax that topped out at $60 in 1940 now tops out at $12,990.   Today, most workers pay more in employment taxes than they pay in federal income taxes. And employment taxes now loom large in federal revenue collections.

As the tax rate has risen, the 'return' from Social Security has declined. For the retirees of 1955, the "return" was just over 15 percent. By 1965 it was under 10 percent. By 1995 it was under 3 percent. And by 2005 it is expected to be less than 2 percent. To put that in perspective, the real, inflation-adjusted, return on bonds has averaged 2 percent   over the last 75 years.

We are now entering a period in which actual returns in the private economy are better than "virtual" returns of Social Security.

As virtual returns have declined, so have net lifetime Social Security benefits. By one set of calculations, the net value of all payments peaked in 1981.   It declined sharply for retirees of 1985, and continued to decline. Those who retire in 2005 will be the first Social Security retirees who will have paid in slightly more than they are likely to receive in benefits. If you were born after 1960 the prognosis is worse: you will lose an amount similar to the amounts gained by the first 50 years of Social Security retirees.

In effect, the Social Security program that boosted retiree income for sixty years in the last century is now beginning to suppress retiree income in this century. Invested for real (instead of virtual) returns, our employment tax dollars might earn enough to keep up with our increasing life expectancy. You can understand this by taking a close look at the table below.

If the virtual return on Social Security is zero but the real return on actual savings is 5 percent, private investment will support more consumption with less investment. While it would take 31 percent of income to provide 17.6 years of consumption at a 0 percent virtual return on Social Security, it would only take 12 percent of income to do that same with private savings.

The Magic Bullet of Investment Return: Social Security vs. Private Investment
This chart shows the percentage of income that must be saved to provide a sum equal to the number of years of life expected at 65. In each case the amount of income to be replaced has been reduced by the amount of income saved. As a consequence, the amount that can be consumed rises rapidly with real return. The model used did not consider rises in living standard due to productivity or rising costs due to inflation. It also did not consider return on investment after age 65. Allowing for return after age 65 would reduce the required savings rate further.
   Life Expectancy at Age 65
Real Return 12.6 years (1935) 17.6 years (2000) 20.6 years (2040)
Social Security, 0% 24% 31% 34%
Pension Portfolio, 5%    9% 12% 15%
Lost Consumption (15%) (19%) (19%)
Source: author life savings model

The difference is 19 percent of lifetime consumption. However you jigger with the numbers, this is a major problem.