Friday, March 05, 2004
The Sky's Not Falling (We're Doomed)
A sunny forecast on Social Security from Paul Krugman:
The annual report of the Social Security system's trustees reveals a system in pretty good financial shape. In fact, it would take only modest injections of money to maintain that system's current benefit levels for at least the next 75 years. Other reports, however, appear to portray a system in deep financial trouble. For example, a 2002 Treasury study, described on Tuesday in The New York Times, claims that Social Security and Medicare are $44 trillion in the red. What's the truth?


First, two words ? "and Medicare" ? make a huge difference. According to the Treasury study, only 16 percent of that $44 trillion shortfall comes from Social Security. Second, the supposed shortfall in both programs comes mainly from projections about the distant future; 62 percent of the combined shortfall comes after 2077.
The outlook's always grim if you're pessimistic about the effort. Seems to me that this whole question is not really about the perceived flatlining of Social Security or about how the market could do it so much better—though those are the talking points. Underscoring the debate, the bedrock question is whether or not by political function the government can override a contract with itself that involves this sort of multigenerational commitment. My assessment is sort of subsumed under category #2 of Brad DeLong's diagnosis of the argument, though it's presented as a reason for privitization:
  1. The first argument is: "Yes, the Social Security Trustees have bought Treasury bonds instead of (say) bank stocks with the Social Security Trust Fund, but this is a good thing. If they had bought newly-issued bank stocks, then the banks would have turned around and bought Treasury bonds with the money from the sale of stock. In that case, the banks would be earning interest from the Treasury on their holdings of Treasury bonds, and would be paying dividends to the Social Security Trust Fund out of those earnings. The banks would take 1% for administrative costs and their own profits. And so the rate-of-return the Social Security Trust Fund earns on its investments would be (roughly) 3% this year. But by investing directly--by elimination the middlebanker--the SSTF saves this 1%, and so earns 4% on its investments this year."

  2. The second argument is: "If the banks owed the SSTF money, and if the Treasury owed the banks money, then both of these obligations would be legal, enforceable contracts. And there would be no chance that politics would someday make the SSTF disappear. But since the way things are today the government is both the owee and the ower, we have lost this insurance--politics may someday make the SSTF disappear, and no court or lawsuit will ever be able to stop it."

  3. The third argument is: "The fact that politicians focus on the government's *debt held by the public* rather than the *total debt* makes them much more willing to increase spending and cut taxes than they would be otherwise. If the SSTF was invested in private assets, there would be no distinction between total debt and debt held by the public, and this problem would not arise."

  4. The fourth argument is: "You want the SSTF invested in private bonds and stocks? Are you crazy!? Do you want some government bureaucrat voting stocks and so electing corporate managers? That's just insane!"
Privatizing the arrangement might guarantee that politics can't cheat the kitty, but isn't that sort of circular?

And it's just bunk that the market can offer more stability than US Treasury bonds. And Halliburton/DynCorps have more or less proven that privatizing government functions (in this case, defense) can lead to cost inflation, favoritism, maybe outright cronyism—there's a thousand reasons not even related to economics why this is a bad idea, but as I try not to froth before lunch, I'll stop there.

If I'm still around by 2077 I better be getting my damned check.