Washington loves 10-year projections like the CBO’s, which said the health-care bill will actually save $120 billion or so in the next decade.
Unfortunately, 10-year economic predictions are as worthless as 10-year meteorological ones, and for the same reason: too many variables interacting in too many unpredictable ways. The CBO might as well sacrifice a chicken and read its entrails to determine what things will look like 10 years out.
A good example of this is Social Security. Since the mid-1980s, Social Security has been running huge surpluses to build up a reserve to care for the baby boomers as they start to retire. The first baby boomers were born in 1946 and are already increasingly retiring. The surplus has been invested in special, non-marketable federal bonds.
In 2008 — only two years ago – the CBO predicted that the Social Security surplus would disappear in 2019, and the Treasury would have to start redeeming those bonds then. But the surplus is disappearing this year, not nine years from now, thanks to the recession, which has caused FICA revenues to fall as unemployment rose, and the number of people applying for Social Security rose, as well.
The CBO predicts that for fiscal 2010, Social Security will run a surplus of $92 billion. But that is an accounting illusion because the government will be paying it interest of $120 billion on the bonds held in the trust fund (not in cash, of course, but in still more federal IOUs). That means that the treasury will have to come up with real money — $28 billion — to make up the difference. It will do this by, of course, selling more bonds to the public.
If the economy comes roaring back, Social Security will go back into surplus for a few years. But as more and more baby boomers retire, it will soon go into permanent deficit (unless it is reformed), and the surplus will be gone by — according to the latest predictions — 2037. That, of course, is a 28-year prediction. It would require reading the entrails of two chickens to come up with a prediction as reliable as that.
