This is the first of a series of reviews I've written to look into how well the various other calculators that claim to allow an individual to compare the benefits they would receive in retirement between the traditional Social Security (SS) program and the newly proposed Personal Retirement Account (PRA) option (and that are not written by me) perform. Since this is the first review of the series, I have provided more information regarding the reasons why I selected the particular test data used throughout the series to produce sample results for each comparison calculator, and I have also extended my comments beyond what may appear in the later reviews, as the context will be established.

Each review will evaluate each calculator's user interface, provide selected sample results, analyze the selected results, look at each calculator's underlying assumptions, discuss other potential issues that may affect the produced results, and will provide a conclusion regarding the calculator's overall usefulness. The first calculator considered is the Reid/Schumer Social Security Benefit Calculator, which has been advanced by Democratic U.S. Senators Harry Reid and Charles Schumer.

The User Interface

The Reid/Schumer Social Security calculator has a very simple interface. Users of the calculator only need to enter two pieces of information: their birth year and their expected lifetime average annual earnings. This latter value may be particularly difficult to estimate, particularly if you're young with most of your working life ahead of you, and I suspect that most people who use the calculator simply enter their current annual earnings. If you're young, you will most likely undershoot your lifetime average, if you're middle-aged, you'll probably be pretty close and if you're near retirement age, you'll be on the high end, assuming you don't actually go to the trouble of calculating your average annual earnings directly.

Although the simple interface makes it easy to come up with quick numbers, it also locks you into the calculator's key assumptions without any means of being able to alter them to better relate to your personal situation or to reflect what proposals may actually be adopted in the lawmaking process. These inherent restrictions limit the ability of the calculator to provide useful information for individual decision making, which is what these calculators are theoretically created to do in the first place.

Sample Results

In evaluating the calculator, I set $40,000 as the individual's average lifetime annual earnings and selected birth years that would correspond to the years where major reductions in benefit payments from Social Security have been projected. For example, 1975 corresponds to a scheduled retirement year of 2042, when the SSA projects retirement benefit payments from Social Security will be reduced by 27%. Likewise, 1985 corresponds to a scheduled retirement year of 2052, when the CBO forecasts that the OASDI Trust Fund will be depleted, causing benefit payments to be reduced by 22%. (I selected the other two birth years to see what results the calculator would determine for someone born 10 years ago and for someone born in this year.) That explained, I produced the following results:

Analyzing the Results

If you look at the table of data above, you'll notice that for each year listed, the combined annual amount of benefits you would receive from traditional Social Security plan and PRAs under the "Bush" plan for the salary listed is nearly identical from year to year (approximately $15,988). This combined number represents the core assumption of the Reid/Schumer calculator of what your future total Social Security retirement benefit would be if the benefits were indexed according to core rate of inflation rather than the rate of increase in average wages. This assumption appears to have been cherry-picked from several different options (available as a 12.5MB PDF document) put together by a presidential commission to suggest various ways of "fixing" the traditional Social Security retirement benefit program.

The last I checked, none of the various proposals have been adopted by the White House (which is why "Bush" keeps appearing in quotes above!) It could be that the Congress will decide to index benefit increases to the core rate of inflation. Then again, they could raise the scheduled age of retirement with full benefits even more (as is already being done); they could increase or remove the cap on income (currently at $90,000) that may have Social Security taxes applied; they could means-test recipients to restrict the amount that wealthy individuals could receive; they could hike Social Security tax rates; and so on. The bottom line: no one knows what's going to happen yet, and you can't call it the "Bush" plan until the White House gets behind one or more of the options available.

Looking at the combined "Bush" value again, you'll also notice that the PRA option directly offsets the amount received from traditional Social Security, as if your annual benefits from the "Bush" program will be capped this way. There is no indication as yet to suggest that this type of annual capping of total Social Security benefits may occur, which would be yet another method of actuarially "fixing" the program.

From what we do know about the President's proposal, a maximum amount equal to 4.0% of your annual income may be placed in a PRA when the introductory cap on PRA contributions is fully phased out, while a minimum amount equal to 6.6% of your annual pay would go toward the traditional program within Social Security that provides retirement and survivor's insurance benefits. This fact alone would seem to indicate that the portion of benefits that you would receive from the traditional Social Security program, while reduced from what you would receive if you opted to remain fully within the traditional program, would almost certainly be higher than the amount calculated by the Reid/Schumer comparison calculator, especially for younger generations.

Underlying Assumptions

In originally looking at the results produced by the Reid/Schumer calculator, I had wondered why the numbers for someone born in 1975 and 2005 would be different if they had the same lifetime average annual income. As it turns out, the Reid/Schumer calculator assumes that Social Security's promised retirement benefits will increase at an average annual inflation-adjusted rate of 1.23% indefinately. This assumption accounts for what I had earlier observed, as it factors in the average real rate of wage growth that forms the current basis for determining annual increases in Social Security retirement benefits.

There are several problems with this assumption. First, the 1.23% average annual rate of growth in Social Security benefits is a historical figure that matches up to long-term economic growth in the U.S., as that factor affects the increase in wages over time, which in turn affects the rate of growth in Social Security's retirement benefits. A 1.23% average annual increase in benefits matches up with the annual rate of economic growth in the U.S., which typically ranges between 2.0 and 5.0%. Since the supporting material (in the form of a 1-page PDF document) for the Reid/Schumer calculator indicates that they rely on the CBO's future economic growth forecast of 1.9% for its future projections, I can't help but observe that this figure falls below the low end of the U.S.' historical rate of economic growth. This fact would suggest that the calculator's assumption that Social Security's promised benefits will continue increasing at a real rate of 1.23% is highly optimistic.

The second problem with this assumption arises from the source of the calculator's assumed economic projections, the CBO itself. In accepting the CBO's economic growth forecasts, the creators of the Reid/Schumer calculator are obligated to also accept the CBO's projections for the long-term sustainability of Social Security's promised benefits. The CBO has forecast that as early as 2052, Social Security retirement benefits will need to be reduced by at least 22%. Unless, of course, there are other unstated means of supporting benefits at the "promised" level. This 1999 513KB PDF document, (GAO/HEHS-99-110), from the Congress' General Accounting Office (GAO), now known as the General Accountability Office, suggests that this level of promised retirement benefits may be sustained through a series of unspecified future increases in the Social Security tax rate. Here's the money quote from Page 23, which appears in the note appearing underneath Figure 2-1 (which depicts the implicit rate of return individuals born in the indicated years would receive if their benefits are maintained at the "promised" level):

Note: Inflation-adjusted rates, average for all workers in each birth year. These estimates do not include Social Security disability contributions and benefits. They do reflect tax rates that would keep the system in actuarial balance on a pay-as-you-go basis.

That increases in the Social Security tax rate will be required to sustain increasing retirement benefits is confirmed on page 25 of this report:

The estimates in figure 2.1 take these projected demographic changes into account and also reflect tax rates that would keep the system in actuarial balance. Under this scenario, tax rates would increase but so would lifetime benefits as people live longer.

Other Issues

The calculator's interface indicates that it will produce results that fit a predetermined outcome (this is evident from the text and graphics that accompany the results portion of the interface.) The calculator also assumes that you will retire at age 65. There is no good explanation for why this assumption was made, other than it made the calculations "easier." Having looked at the calculations, I'm not convinced that's the case. To the programmer's credit however, the level of "promised" benefits do appear to have been properly adjusted to account for this "early" retirement for those eligible to participate in the proposed Personal Retirement Accounts.

As we know it today, the President's proposal only permits individuals born in or after 1950 to participate in the PRA option. Currently, SS promised benefits are already in the process of being cut for this group as the age at which an individual may retire with full promised benefits from Social Security is being gradually increased from 65 to 67.

This assumption produces two negative effects. First, the benefits one would receive from Social Security are understated, since early retirement at age 65 for those affected by the proposed reform would reduce their annual benefits to the calculated level compared to what they would receive if they retired at their "scheduled" age. Second, the amount the affected individuals' PRAs would grow in the actual remaining time before scheduled retirement is excluded from that comparison. To see what kind of impact an additional one or two years might have on the value of a 30 to 50 year-long investment, feel free to play with the Investing: Future Value calculator with numbers from your own investment or savings accounts.

Conclusions

The Reid/Schumer Social Security Calculator is a highly flawed product. Rather than providing the user with real decision-making information that matches up with what we know about the President's plans so far, or allowing the user to alter any of the calculator's key assumptions, the results provided by the calculator are set to a predetermined fate, which is only minimally affected by the user's information. It is however, the only comparison calculator that currently considers one or more actuarial fixes that may be under consideration by members of Congress (tax rate increases, indexing future benefits to inflation vs. wage growth, absolute caps on benefits) which does provide some limited value. Since there is no guarantee that the actuarial fixes suggested by the calculator will be the ones ultimately chosen, I would advise anyone using it to not rely upon it as an authoritative source until we see what proposals will actually be adopted.

Labels: review

- posted by Ironman at 1:33 PM | Permalink | Home |