Determining the effective return that an average "investor" can expect to earn from their "investment" in Social Security turned out to be the most difficult part of developing the Social Security vs. Private Retirement Accounts comparison calculator.
For starters, the whole Old Age and Survivor Insurance (OASI) Trust Fund within Social Security (the part of the program that pays retirement benefits) is set up to pay the program's current recipients as soon as the money collected through your dedicated payroll taxes hits the books. The taxes collected in excess of those required to pay off the current recipients of Social Security are used to either pay the program's administrative costs or are otherwise siphoned off into other federal spending programs. The "siphoned" funds are actually borrowed by the U.S. government, which replaces the funds in the OASI Trust Fund account with "special-issue" interest-bearing Treasury bonds.
This all sounds well and good, but the portion of your payroll taxes that actually gets "invested" this way is just a small fraction of what you (and your employer on your behalf) paid in, and is getting lower and lower as the number of people receiving benefits increases. When you also consider that the actual returns for each recipient are tied to a series of factors that can vary wildly (age when benefits are first received, life expectancy, marital status, etc.), it becomes extraordinarily difficult to work out what the actual return on your individual investment might be. That the administrators of Social Security are reluctant to calculate an average individual rate of return because of these issues only further complicates the matter.
But why not?! Fortunately for me, others have looked at the same issue, and have come up with what the average Social Security investor's effective rate of return on contributions to the OASI Trust Fund account on their behalf might look like. The data in the following table was taken from a MSN Moneycentral article by Tom Woodruff, and illustrates the average rate of return for "investors" who retired in the indicated year:
| Social Security's Effective Annualized Rate of Return | |
|---|---|
| Year of Retirement | Annualized Rate of Return |
| 1940 | 135% |
| 1950 | 24% |
| 1960 | 15% |
| 1970 | 10% |
| 1980 | 8% |
| 1990 | 6% |
| 2000 | 4% |
I took this data, and used several different regression techniques to work out a mathematical formula using Microsoft Excel that can be used to estimate the effective rate of return from Social Security into the future. Of course, the formula below is only as good as the limited number of data points used to determine it, and the best "fit" between formula and data came when I excluded the 1940 data point (Editor's Note: That doesn't mean the 1940 isn't there at all, it has just been slightly shifted from where it was. Here's the proof: enter 1940.0001, or some other similarly small offset from 1940, for the YEAR in your calculator when you enter the formula. Keep trying new "YEARS" until you reach a rate of return of 135%. It turns out to be a pretty good trade-off between Microsoft Excel's curve-fitting and the projected boundary conditions!):
Rate of Return = -11.009 * LN(YEAR-1940) + 48.589
In the formula above, "YEAR" is the expected year of retirement and "LN(YEAR-1940)" represents the natural logarithm of the number of years between the expected year of retirement and 1940. I selected this particular equation as being representative of the general trend for two reasons:
- Since Social Security is primarily a "pay-as-you-go" program, I believe that any estimation of the average rates of return in the program should reflect when the program's cash flow turns negative.
- The formula conservatively reflects the expected trend in the Social Security program by turning negative in 2023, just five years after current projections indicate that program outlays will exceed revenues. It also projects that the rates of return will grow worse slowly over time, which I believe to be the best case scenario for "investors" in the program.
You can compare the data table above with results from the formula, or project your effective return in the future with this simple calculator:
Note: The negative cash-flow situation described above doesn't mean that the benefits paid out will be cut beginning in 2018. What will happen is that the OASI Trust Fund administrators will begin spending down the fund's principal, cashing in the special-issue treasury securities to meet its obligations, which will also reduce the effective rate of return of the "investment" as this real source of investment revenue is consumed. This practice will continue until the trust fund is depleted, which is currently expected in 2042. At that point, the payments out will be forced to equal the taxes in, and benefits will need to be cut or the difference made up from other sources, such as general tax revenue or additional government borrowing, assuming no changes are made in the funding of the program.
Update (February 25, 2005): A special hat tip goes out to pat m., who, via the comments at Patrick Ruffini's blog pointed to a useful table from the Social Security Administration that provides the year-by-year future projection of the value of the special-issue treasuries maintained in the OASI Trust Fund under the agency's intermediate cost assumptions. The Trust Fund, which in 2004 contained some $1,684 billion dollars, will continue growing at a rapid clip up until 2018, and then will continue growing at slower and slower rates until it reaches a peak value of slightly over $6,599 billion dollars in 2028. Amazingly, the table indicates that the Trust Fund will be fully depleted by 2042, just 14 years later, in order to continue sustaining the promised retirement benefits of the baby boom generation. This cash burn rate is phenomenal when you consider that it will have taken the Trust Fund over 44 years (from 1984 to 2028) to reach its peak!
Note: The footnotes to the table indicate that under the agency's high cost assumptions, the Trust Fund will be depleted as early as 2031.
Labels: tool
Welcome to the blogosphere's toolchest! Here, unlike other blogs dedicated to analyzing current events, we create easy-to-use, simple tools to do the math related to them so you can get in on the action too! If you would like to learn more about these tools, or if you would like to contribute ideas to develop for this blog, please e-mail us at:
ironman at politicalcalculations
Thanks in advance!
Closing values for previous trading day.
This site is primarily powered by:
CSS Validation
RSS Site Feed
JavaScript
The tools on this site are built using JavaScript. If you would like to learn more, one of the best free resources on the web is available at W3Schools.com.
Other Cool Resources
Blog Roll
