16. Appendix C briefly discusses the proposal to introduce individual accounts, which would make a fundamental change to the Social Security program, and other ideas have been proposed. For example, there could be more income taxation of the benefits received by people at higher income levels, or some or all of the proceeds from the federal estate tax could be devoted to Social Security. However, given the magnitude of the continuing funding gap, these two revenue sources could contribute only relatively small portions of the needed funds. From another perspective, several proposals have been made in recent years to enhance Social Security benefits for vulnerable populations, such as very old beneficiaries, workers with very low wage histories, and widows. It was beyond the scope of the committee’s charge to assess these or other enhanced benefit proposals. We note, however, that if benefits were enhanced for such vulnerable populations, corresponding reductions in benefits or tax increases would be needed to finance these changes.


17. However, for high earners, Option 1 would reduce the inflation-adjusted level (not just the growth) of currently scheduled benefits for future retirees.


18. Because a later retirement age provides more time for workers to become disabled and to apply for disability insurance and because it delays the age of “conversion” from disability to retirement benefits, raising the Social Security retirement age will increase spending in the disability insurance program. However, the only net effect of delayed conversions on the program’s financial position comes from differences in indexing between retirement and disability benefits for this option and two others (see Appendix C).


19. This “progressive indexing” option is not to be confused with a somewhat similar proposal called “progressive price-indexing” that has been proposed and analyzed elsewhere. For a critical analysis of that proposal, see Appendix C and Furman (2005a).


20. The chained Consumer Price Index would replace the older fixed-Weight Consumer Price Index. The newer index reduces the latter’s general overstatement of price inflation by roughly 0.3 percentage points per year; see a study by statisticians at the U.S. Bureau of Economic Analysis and Bureau of Labor Statistics (McCully et al., 2007:26-33) and see National Research Council (2002).


21. However, in comparison with currently scheduled benefits, this provision would reduce monthly benefits the most for long-lived beneficiaries, who are disproportionately widows. But note that low- and middle-level workers (and their spouses) generally fare best under progressive indexing—another benefit change under Option 1.


22. Lifetime—not just monthly—benefits are important, too, of course. For instance given the longer longevity projected for 2050, Option 1’s delayed retirement ages, relative to the unsustainable current law, produces a greater downward tendency for its lifetime than for its monthly benefits. This helps allow it to sustain program finances without a payroll-tax increase.


23. Throughout this chapter, the illustrations of benefits, earnings replacement rates, and payroll tax paid use the Social Security Administration’s definitions of representative workers at different earnings levels. The “low,” “medium,” “high,” and “very high” illustrations are of lifetime earnings levels, scaled to reflect changes in the overall wage distribution over time. “Maximum earners” are defined differently, however, because they reflect steady earnings at each year’s taxable maximum, as does “worker at taxable maximum,” when the formula for the taxable maximum changes under some of the study’s reform options. We note, however, that the Social Security “medium” earning level has been shown to be higher than the actual average level (Mitchell and Phillips, 2006, 2009). However, because our analysis uses the various Social Security definitions of earning levels only for comparison with each other, this finding does not affect the comparisons.


24. We present Social Security earnings replacement rates to gauge the degree to which beneficiaries—especially those without pensions or savings—can rely on Social Security benefits.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement