When I posted on Social Security as a Ponzi scheme on March 11, I didn’t expect the degree of interest I got. It also led to a discussion of what to do now that we’re in a mess.
So I’ve decided to post the rest of my chapter of The Joy of Freedom: An Economist’s Odyssey. I’ll do it in installments. The last installment discusses what to do about it.
Here’s the next installment.
Flawed from the Start
How did we get into this mess? It started in 1935, when President Franklin D. Roosevelt, together with Congress, explicitly designed Social Security as an intergenerational “chain letter.” That, more than any other single feature, virtually guaranteed a big mess for future generations. Interestingly, when the proposal was debated, its chain-letter aspect was little discussed. Politicians in neither the Democratic nor the Republican party seemed upset about that crucial aspect of the plan. At the time, some of its proponents thought of the Social Security tax as a way of extending the income tax to lower-earning people. W. R. Williamson, an actuarial consultant to the first Social Security Board, stated that Social Security extends Federal income taxes “in a democratic fashion” to the lower-income brackets.[1]
Roosevelt and Congress also rejected the Clark amendment, named after Missouri Senator Bennett Champ Clark, which would have exempted employers and employees who had government-approved pension plans. Although the Senate backed this amendment by a vote of 51 to 35, it was later removed. Had that exemption been in the law, many fewer people would have been in the Social Security program and, in fact, with the growth of private pensions, the fraction of the workforce in Social Security would probably have shrunk over the years.
Roosevelt strongly believed in a payroll tax as the way to finance the program. Calling the taxes “contributions,” which the federal government did from the start, would make people think of Social Security as an annuity that they had paid for and that they therefore had a right to. That’s also why Roosevelt wanted to use a special payroll tax rather than general revenues. If people paid a payroll tax earmarked for Social Security, reasoned FDR, they would think themselves entitled to benefits from the program. FDR stated,
[T]hose taxes were never a problem of economics. They are politics all the way through. We put those payroll contributions there so as to give the contributors a legal, moral, and political right to collect their pensions….With those taxes in there, no damn politician can ever scrap my Social Security program.[2]
Roosevelt was saying, in effect, that once the entitlement mentality had taken hold, it would be very difficult ever to cut or eliminate Social Security. He was right. What he didn’t say—but what the chain-letter financing implied—was that the other reason Social Security would be entrenched was that older people would press politicians for continued benefits, which would necessitate continued taxes on working people, who, when they retired, would push for further taxes on the next generation, and on and on forever. In short, FDR implemented a system of passed-on intergenerational abuse that is still with us today.
Presidents Johnson and Nixon made the problem worse. Between 1967 and 1972, Congress and the President raised Social Security benefits by 72 percent (37 percent after adjusting for inflation). When Wilbur Cohen, Johnson’s Secretary of Health, Education, and Welfare, proposed a 10 percent hike in Social Security benefits, Johnson replied, “Come on, Wilbur, you can do better than that!”[3] President Nixon added to the problem by getting into a bidding war with Wilbur Mills, a powerful congressman who was jockeying for the 1972 Democratic presidential nomination. The net result was a 20 percent increase in benefits.
MIT economist Paul Samuelson added some of the intellectual backing for these policies. “The beauty about social insurance is that it is actuarially [italics Samuelson’s] unsound.” Samuelson’s point was that if real incomes were growing quickly, each generation could get more out of Social Security than it paid in. While its critics attacked Social Security as a Ponzi scheme, Samuelson beat them to the punch in 1967 by blessing it as one. “A growing nation,” wrote Samuelson, “is the greatest Ponzi game ever contrived.”[4]
We are now paying through the nose for that “beautiful” Ponzi game. If we include the portion paid by the employer, over 62 percent of families now pay more in payroll taxes (most of which is for Social Security) than they pay in federal income taxes.[5]
The initial payroll tax rate when the program first began was 2 percent on the first $3,000 of income, split equally between employer and employee. In the year 2001, the tax rate for Social Security was 10.6 percent on income up to $80,400 and zero after. This increase in income taxes is not simply an adjustment for inflation. Three thousand dollars in 1938, adjusted for inflation, is less than $38,000 today, or only about half the base income that is taxed today. The maximum tax, employer and employee combined, is $8,077 today versus $60 when the program first started. Had the tax been increased just for inflation, but no more, it would be only about $750 today. See Table 14.1.
Table 14.1 Maximum Tax for Social Security (excluding Disability Insurance)
| Calendar Year | Maximum Tax | Maximum Tax in 2000$ |
| 1939 | $60 | $735 |
| 1950 | $90 | $636 |
| 1955 | $168 | $1,066 |
| 1960 | $264 | $1,518 |
| 1965 | $324 | $1,750 |
| 1970 | $569 | $2,496 |
| 1975 | $1,234 | $3,902 |
| 1980 | $2,341 | $4,835 |
| 1985 | $4,118 | $6,512 |
| 1990 | $5,746 | $7,483 |
| 1995 | $6,438 | $7,187 |
| 1997 | $6,932 | $7,348 |
| 2001 | $8522 | $8,274 (estimated) |
Source: Tax rates and tax base from Social Security Board of Trustees Report, various issues; inflation adjustment from Economic Report of the President, various issues.
Ponzi versus Stocks
Many critics of Social Security have claimed that the current elderly are getting a windfall from the system, but that the younger you are, the worse a deal you will get. They’re half right. The younger you are, the worse your deal. But many of the current elderly are also hurt. The reason is that the return from Social Security compares very unfavorably to the returns available in the stock market.
In a 1987 article in the National Tax Journal, Stanford economists Michael Boskin (later to be chairman of the first President Bush’s Council of Economic Advisers), Douglas Puffert, and John Shoven, and Boston University economist Laurence Kotlikoff presented data on the rate of return earned from Social Security taxes[6]. The real rates of return varied from minus 0.79 percent to 6.34 percent and depended crucially on the person’s age (older is better), income level (low income is better than high income), and marital status (being married with one spouse not working is better than either being single or being married with both spouses working). Interestingly, even the person who did the best—someone born in 1915, the sole wage earner for a married couple, earning only $10,000 a year in 1985 dollars—received a return of 6.34 percent. Every other category of income earner they considered, including those slightly younger or with a slightly higher income, earned a lower return from Social Security taxes.
In a more recent study,[7] Harvard economist Martin Feldstein and Dartmouth economist Andrew Samwick found that the average rate of return on taxes paid will be as shown in Table 14.2.
TABLE: Average Real Rate of Return on Social Security Taxes Paid
| Year of Birth | Pre-1915 | 1915 | 1930 | 1945 | 1960 | 1975 | 1990 |
| Real Rate of Return | 7.0% | 4.21% | 2.52% | 1.67% | 1.39% | 1.39% | 1.43 |
Source: Feldstein and Samwick, “The Transition Path in Privatizing Social Security,” National Bureau of Economic Research, Working Paper # 5761, September 1996.
Compare these rates of return with what you could have earned with an indexed portfolio of stocks. According to Ibbotson Associates, a Chicago-based firm that computes stock market returns, the average rate of return on stocks between 1926 (before the 1929 crash) and 1997 was 11.0 percent, or 7.7 percent when adjusted for inflation.
For shorter periods, of course, the rate of return has been higher and lower than this, but for no 30-year period has the real rate of return ever been below 4 percent. So a rate-of-return comparison shows private investment in stocks to be superior to the government system for people who invest for 30 years or more. Of course, you can find 5-year periods and even 10-year periods during which you would have done considerably worse. According to Ibbotson Associates, the worst 10-year period was October 1, 1964 to September 30, 1974, when the annual inflation-adjusted rate of return in stocks was -4.3 percent.[8] The moral of the story is that you shouldn’t put all your savings in stocks if you plan to draw on the funds in 10 years or so.
Another equally valid way to compare makes the contrast starker: look at the effect that Social Security taxes and benefits have on your wealth. Economists do the computation in three steps. First, they compute the present value of Social Security taxes paid by you and your employer—the value at retirement age of all the previous taxes paid, assuming that they earn compound interest. Second, they estimate the present value of Social Security benefits—the value at retirement age of a stream of future income—using the same rate of return they use for the taxes. Finally, they subtract the present value of taxes from the present value of benefits.
The crucial variable for such a calculation is the interest rate. A pessimistic real rate to use is 4 percent. Why? Because, as noted above, you could have earned over 4 percent with a portfolio of stocks for the worst 30-year period for stocks. Shawn Duffy, a student at the Naval Postgraduate School, using an inflation-adjusted rate of return of 4 percent, found that someone born in 1929 who paid the maximum Social Security tax his or her whole working life and who retired in 1994, would have been $120,000 better off with a private savings plan instead of Social Security. Someone who worked at the average wage his or her whole life would have been $54,000 better off without Social Security. And even a 1994 retiree who earned the minimum wage for the whole of his or her working life, supposedly the quintessential social-security-windfall king, would have been about $9,000 better off with a private savings plan.[9] With a more realistic 6 percent real rate of return, the Social Security caused the maximum-earning 1994 retiree to lose $262,000 in wealth, caused the average earner to lose $160,000, and caused the minimum-wage earner to lose $66,000.
It’s true that the earliest recipients of Social Security did very well. That’s because they had paid into the system for only a few years, but received substantial benefits for many years. Miss Ida Mae Fuller, for example, the first recipient of Social Security, received, by the time of her death at age 100, $20,000 in benefits in return for $22 in taxes paid. But now that all future and most current beneficiaries have paid taxes over a working lifetime (when this happens, economists who study Social Security call the system “mature”), there is no windfall for current and future retirees.
[1] “26,000 in Brooklyn Defy Security Law,” New York Times, November 29, 1936, p. 37.
[2] From Arthur M. Schlesinger, Jr., The Age of Roosevelt, vol. 2, The Coming of the New Deal (Houghton Mifflin, 1959), pp. 309–310, referenced in Martha Derthick, Policymaking for Social Security, Washington, D.C.: Brookings Institution, 1979, p. 230.
[3] This story is told in Peter G. Peterson, Will America Grow Up Before It Grows Old?, New York: Random House, 1996, pp. 93–99.
[4] Samuelson quotes are from Newsweek, February 13, 1967, and are quoted in Derthick, p. 254.
[5] Andrew Mitrusi and James Poterba, “The Distribution of Payroll and Income Tax Burdens, 1979-1999, National Bureau of Economic Research, Working Paper No. 7707, May 2000, p. 24.
[6] Boskin, Michael, Laurence Kotlikoff, Douglas Puffert, and John Shoven, “Social Security: A Financial Appraisal Across and Within Generations,” National Tax Journal 40, 1987, pp. 19–34.
[7] Martin Feldstein and Andrew Samwick, “The Transition Path in Privatizing Social Security,” National Bureau of Economic Research, Working Paper # 5761, September 1996, p. 20.
[8] I thank Heather Fabian, public affairs manager at Ibbotson Associates, for providing the computations.
[9] Shawn P. Duffy, Social Security: A Present Value Analysis of Old Age Survivors Insurance (OASI) Taxes and Benefits, Naval Postgraduate School, Masters Thesis, December 1995.