From: Eric Klieber (message is being sent by David Rivera due to technical problems) <eklieber@buckconsultants.com>
Subject: Dramatic Reform Unnecessary
Conventional wisdom says that Social Security is approaching an inevitable crisis, and that the only way to avoid this crisis is to undertake major reform of the system now. Unfortunately, conventional wisdom, as is often the case, is wrong: the predicted financial crisis is unlikely to occur, and if it does, the resources to handle the crisis will be available without any need for reform at this time. The most likely result of reform at this time is needless tampering with one of the most successful social programs ever implemented in the United States.
Actuarial Study No.91 (Long-Range Estimates of the Financial Status of the Old-Age, Survivors, and Disability Insurance Program, 1983), published in April1984 by the Office of the Actuary at the Social Security Administration, presents the results of the OASDI actuarial valuation for 1983, taking into account the recently passed legislative changes. The study shows the system to be in actuarial balance through 2057. Further, the study shows the projected assets of the OASI trust fund as a percentage of expenditures as 227percent for the year 2000 (the report shows only every fifth year). The 1999 Trustees Report shows the projected assets of the OASI trust fund as a percentage of expenditures as 224percent for the year 2000. I congratulate the office of the actuary on the pinpoint accuracy of its 1983 study. (The results for the disability trust fund are less favorable, but this is due primarily to higher disability rates relative to the 1983 assumption.)
The above tells us two things. First, since the demographic trends generally cited as the cause of Social Security's financial problems were already well-known in 1983, the financial difficulties projected in the 1999 Trustees Report, at least through 2057, cannot be ascribed to demographics. Second, those financial difficulties are also not the result of poor experience since 1983. In fact, they result mostly from changes in economic assumptions, particularly the annual rate of increase in real wages, which declined from 1.5percent in 1983 to 0.9percent in the 1999 report. The increase in real wages, that is, the projected rate of increase in wages less the projected rate of increase in the cost of living, is not the same as, but is closely related to, the increase in the productivity of labor. The essential reason for the perceived financial problems of Social Security is the assumption used by the Social Security actuaries in the Trustees Report that the current high rate of productivity increase is a short-term anomaly and that the low rate of productivity increase in the 1980s is more indicative of the future. Both analytical and historical considerations point to this assumption's being incorrect.
As the nation's population increases, companies must produce more goods and services to maintain or increase the standard of living. There are two ways for increasing production: hiring more workers to increase the productivity of existing capital; or making capital investments to increase the productivity of labor. Rational managers choose whichever option is cheaper. When labor is plentiful and, hence, relatively cheap, managers hire more workers, but their productivity lags. This happened during the 1970s and 1980s. (The productivity of capital, on the other hand, soared, resulting in the long bull market.) We are now in a transition to a demographic situation where labor is less plentiful, although by no means unprecedentedly so by historical standards. Therefore, it is natural that managers should increase capital investment to make better use of relatively more expensive labor. (As a result, the productivity of capital should increase more slowly, cooling off the current red-hot stock market.) Based on this analysis, the current strong growth in productivity should continue as long as the supply of labor remains low.
Historical data bear out this analysis. Although the proportion of the population working has fluctuated since 1930, the growth in GDP has roughly tracked the growth of the total population rather than the growth of the working population. To put it another way, the economy has responded dynamically to the challenges posed by demographic shifts. The economic assumptions used in the Trustees Report predict the economy will remain static in the face of dramatic demographic changes, something that has never happened in the history of the United States.
There are roughly 4,000 enrolled actuaries in the United States who are also making projections of future economic conditions. Under ERISA, these enrolled actuaries must use their "best estimates" of future financial experience in making their projections. Right now, they are mostly using investment return assumptions in the range of 8 to 9percent, and salary increase assumptions in the range of 4 to 5percent. Since the salary increase assumption generally includes a longevity component, the implied rate of increase in wages for the economy as a whole is lower. This implies a spread between investment return and wage increases of at least 4percent. I know of no reasonable economic scenario under which these assumptions are compatible over a time period measured in decades with the GDP growth rates implied by the assumptions used by the Social Security actuaries (although I'd love to hear from you if you have one.) On a practical level, if the assumptions used by the Social Security actuaries are borne out, the assumptions used for most qualified pension plans will turn out to have been too optimistic, with the result that large contribution increases will be required at time when many companies may be experiencing financial difficulties due to slow economic growth. The Social Security assumptions spell financial problems not just for Social Security, but for private pension plans as well.
Those who believe Social Security is headed for financial crisis often add that the time to act on this problem is now, since the longer the lead time, the less traumatic the solution. Suppose we know for a fact that the projections in the Trustees Report are accurate to the dollar. How do we know what solution is the right one? If GDP growth falls to a level near or even lower than population growth, as predicted, living standards among workers are likely to stagnate or decline. This would be particularly true if the portion of the GDP tied up with non-consumable services, such as defense and environmental remediation, increases. In this situation, it may be entirely appropriate to reduce Social Security benefits, since why should workers accept a tax increase to support the living standard of retirees when their own living standard is falling. (AARP would probably disagree with this point.) Increasing taxes today could turn out to be the wrong solution, even if the problem is real.
Suppose we know for a fact not only that the projections in the Trustees Report are accurate to the dollar, but also that GDP growth will be sufficient to at least maintain the living standard of workers and retirees. I don't see how this would be possible, but let's suppose it anyway. Then, by definition, the economy must be generating enough income for workers that the necessary portion could be transferred to retirees to maintain the living standard of retirees without reducing the living standard of workers. So even if Social Security does run out of money as predicted, the resources will be available to keep the system going without hurting the workers regardless of what we do today. No doubt the workers would not accept the necessary tax increase gladly, but keep in mind that the initial twopercent Social Security payroll tax was levied during the worst economic depression the United States has ever experienced. Whenever I talked with my mother about Social Security, she never failed to mention the hardship caused by the withholding of the one percent employee portion of the tax from her meager paycheck.
So I conclude that: (1) the pessimistic view of Social Security finances in the Trustees Report is incorrect, and a broad consensus of enrolled actuaries is in apparent agreement; (2)even if the analysis in the Trustees Report is correct, it is difficult to support raising Social Security taxes at present, since reducing benefits may be the more appropriate solution; and (3)even if reducing benefits is not the appropriate solution, the resources to maintain promised benefit levels will be available when the system runs out of cash without any current corrective action. I hope you all sleep better tonight.
Eric J. Klieber, FSA, MAAA, EA
Consulting Actuary
Buck Consultants
eklieber@buckconsultants.com
