Figure 3-18 also plots the most recent Census Bureau projection of the OADR as a purple line. It is just at or below the lower 25 percent bound of the committee forecast, perhaps because the life expectancy forecast in the Census Bureau projection is lower than in the committee’s. The Census projection does not come with a range. Figure 3-18 further shows the OADR as projected by the Social Security Trustees in its 2012 high, intermediate, and low-cost variants, all indicated by diamonds. The intermediate projection is very close to the committee’s median through 2030, and then transits to the Census Bureau projection at the lower 25 percent of the committee’s range. The Trustees’ low-cost scenario is slightly below the lower 2.5 percent bound for the committee’s forecast, while the high-cost scenario rises above the committee’s upper 2.5 bound before dipping below this bound around 2040. The Trustees’ projection is centered a bit lower than the committee’s owing to less projected gain in longevity (life expectancy of 82.2 years in 2050, versus the committee projection of 84.5 years, as discussed earlier). The Trustees do not assign a probability to the range for the OADR, but Figure 3-18 suggests that its probability coverage is about 95 percent. This is the probability that the OADR in any given year will fall between the high-cost and low-cost brackets. This does not mean, however, that there is a 5 percent chance that the OADR would generally lie outside this range in every year between 2010 and 2050. That probability would be far lower, because a typical trajectory of the OADR would wander around within that range, with offsetting upward and downward variations.

This last point can be seen clearly in Figure 3-19, which shows a probabilistic forecast of the weighted support ratio based on the age profiles for 2007 shown earlier in Figures 3-10 and 3-11. It uses the same 1,000 stochastic trajectories and plots 20 of them for illustrative purposes. The trajectories often can be seen to fluctuate rather than to be persistently high or low. As in Figure 3-18, the solid black line is the median projection. The dashed blue lines indicate quartiles (so there is a 50 percent chance that the future outcome will lie between them in a given year), and the dashed green lines represent the upper and lower 2.5 percent bounds and define the 95 percent probability interval. The value of the ratios has been adjusted so that the ratio in 2007 is 1.0. That is, the plotted values show the ratio relative to the ratio in 2007.9

The expected decline in the support ratio between 2010 and 2050 is 12 percent (the same as reported earlier in this chapter), and there is a two-


9Of course, the age profiles of labor income and consumption will change over the next four decades, and any attempt to project their future levels would involve substantial uncertainty. However, the support ratio for future years is calculated using the baseline (2007) age profiles so as to isolate the effects of demographic change. Therefore the uncertainty surrounding future values of the age profiles is irrelevant. For a discussion of the construction and use of support ratios in this context, see Cutler et al. (1990).

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