The Trust Fund

This is a figure from the Social Security Trustees' Report released today showing projected Trust Fund balances under three sets of assumptions:

Figure II.D7.-Long-Range OASDI Trust Fund Ratios
Under Alternative Assumptions
[Assets as a percentage of annual cost]

Click to enlarge

I =  SSA Low cost estimate
II = SSA Intermediate cost estimate (the one in the news)
III = SSA High cost estimate

As you can see, the assumptions regarding productivity, immigration, and so on matter - under the low cost scenario the Social Security and DI Trust Funds remains solvent. From the report, a note about the uncertainty of the projections represented by low, intermediate, and high cost options:

Significant uncertainty surrounds the intermediate assumptions. The Trustees have traditionally used low cost (alternative I) and high cost (alternative III) assumptions to indicate this uncertainty. ... The low cost alternative is characterized by assumptions that improve the financial condition of the trust funds, including a higher fertility rate, slower improvement in mortality, a higher real-wage differential, and lower unemployment. The high cost alternative, in contrast, features a lower fertility rate, more rapid declines in mortality, a lower real-wage differential, and higher unemployment.

These three alternatives have traditionally been constructed to provide a reasonable range of possible future experience. However, these alternatives do not address the probability that actual experience will be within or outside the range. As an additional way of illustrating uncertainty, this report includes estimates from a model of the trust funds that provides a probability distribution of possible future outcomes (see appendix E). The results of this model suggest that outcomes better than the traditional low cost alternative and outcomes worse than the high cost alternative have very low probabilities of occurring.

But the report says (pg. 161) not to take the probabilities in Appendix E too literally, the actual uncertainty (i.e. confidence bounds around the future projections) could be much larger than estimated, so that the main message is that there is a high degree of sensitivity to assumptions as exhibited in the figure:

The results from this model should be interpreted with caution and with a full understanding of the inherent limitations. Results are very sensitive to equation specifications, degrees of interdependence among variables, and the historical periods used for the estimates. For some variables, using the variations exhibited in a relatively recent historical period may not provide a realistic representation of the potential variation for the future. In addition, results would differ if random variations had been applied to additional variables other than those mentioned above (such as labor force participation rates, retirement rates, marriage rates, and divorce rates). Furthermore, additional variability could result from incorporating statistical approaches that would more fully model change in the long-range central tendencies of the variables. The historical period available for most variables is relatively homogeneous and does not reflect many substantial shifts. The time-series modeling reflects what occurred in the historical period. As a result, the variation indicated in this appendix should be viewed as the minimum plausible variation for the future. Substantial shifts, as predicted by many experts and as seen in prior centuries, are not fully reflected in the current model.