Return of the Fortune Tellers

The most interesting information in the new Congressional Budget Office report on the future of Social Security is on its cover. There the reader will find a picture that shows Social Security revenues and outlays projected ahead to the year 2100. In addition to the "best guess" projections, the reader finds a band, which shows where the CBO thinks revenues and outlays will land with 80 percent probability. That band gets wider and wider the farther we look ahead, until by the year 2100, it shows that outlays can be expected to lie somewhere between 5.5 and 9.5 percent of GDP-a huge difference. And the Social Security Trust Funds may be exhausted in as little as 30 years or they may still be intact at the end of the century.

The CBO has provided us with projections that look ahead 100 years. These are more optimistic than the 75 year projections coming out of the Social Security Administration's actuaries' reports: the CBO's best guess is that the Trust Funds will be exhausted in 2052 (versus 2042 for the SS actuaries) and that thereafter, payroll taxes will cover only 80 percent of promised benefits (vs. 70% for the actuaries).

These differences are chicken feed compared to what happens if projections look really far ahead. It has become fashionable recently to make projections to infinity - you did not misread, infinity, forever - and this method has produced some dazzling numbers. Economists Lawrence Kotlikoff and Kent Smetters, in particular, have produced estimates of the long-run deficits of Social Security and Medicare/Medicaid that, in today's dollars, are more than four times as large as today's GDP, 44 trillion dollars.

This sort of stuff is nonsense. Checking economic flows 75 or 100 years into the future is like eyeballing traffic 75 or 100 miles away from America's tallest building. And forecasting outlays and revenues to infinity is no more possible than standing on the Sears Tower and looking at the dark side of the moon. We have little idea of what will happen in 10 years and we have no idea of what will happen in 100. 

For example, as Brad Delong pointed out on June 15, both the SS actuaries and the CBO assume that there will be 900,000 immigrants per year from now on. "But net immigration today is roughly 2.1 million per year (we think)." Delong also criticizes the assumptions of the actuaries and of the CBO regarding wage growth, which in both cases fall far short of historical growth rates.

If the assumptions mattered little to the conclusions of these long run forecasts, we could look ahead with some confidence. In fact, as the cover of the CBO report illustrates, projections even a few decades from now are all over the map.

Both the SS actuaries and the CBO come to a similar conclusion about the future of Social Security. It would not take much to fix it so that we can meet its promises. As Robert Greenstein, Peter Orszag and Richard Kogan have pointed out, the new CBO numbers imply that the 75 year deficit of the SS system could be eliminated simply by letting the tax cuts expire for households with incomes above a million dollars a year. 

The real story from the CBO about the long run outlook for Social Security is not that Social Security will survive 50 instead of 40 years. We honestly don't know whether the future holds a moderate problem (unlikely), a small problem (likely), or no problem at all (unlikely). But if we want to keep Social Security working for the average American, we can do it easily. The challenge is not to our pocketbook-we can handle that-but to our will.