SMOKE AND MIRRORS….I was emailing with WM’s editor today about Social Security, and one of the things I mentioned is that I’m skeptical of “free lunch” proposals. A free lunch proposal is one that ? when carefully examined ? essentially proposes that we can fix Social Security without any tax increases or benefit cuts.

All of these proposals rely on at least one heroic assumption, and in the case of privatization the assumption is that the average return on private accounts will be about 7% per year. Is this reasonable? Over at MaxSpeak, Dean Baker is properly skeptical. The following prose is pretty impenetrable to financial non-gurus, but that’s the way it goes with these things, and you should probably treat the numbers in the following paragraphs the same way you treat Russian names in Tolstoy novels:

I have a test of my own that I have been trying to get economists to take (thus far unsuccessfully), in which I ask proponents of privatization to write down the set of dividend yields and capital gains that will give them the 6.5-7.0 percent real stock returns that they conventionally assume. Such returns were possible in the past because the price to earnings (PE) ratios have historically been much lower and profit growth was much faster.

The price to earnings ratio averaged about 14.5 to 1 over the last seventy years, compared to more than 20 to 1 today. This is important, because if 60 percent of profits are paid out as dividends (or used for share buybacks), this gets you a dividend yield of over 4.0 percent with a PE ratio of 14.5 to one. It gets you just 3.0 percent with a PE ratio of 20 to 1, and of course less when the PE ratio is higher.

He goes on to suggest that profit growth (and thus stock appreciation) is likely to be about 1.4% in the future, which gives you a total return of about 4.4%. In other words, the first question you should ask about any privatization scheme is: What if average returns over the next 40 years are only 4.4% ? or less?

It’s a good question, because you can “fix” Social Security pretty easily if you’re allowed to simply make any future growth assumptions you want. For example, if productivity growth and labor force growth are just a bit higher than the Social Security trustees currently assume, the system will remain solvent forever. Look ma, no crisis!

(On a related note, here’s a variation on Dean Baker’s question: please provide a projection of future economic growth rates that makes it reasonable to assume that (a) private accounts grow 7% a year but (b) Social Security as it’s currently funded eventually becomes insolvent. You can probably do this if you’re willing to fiddle with a spreadsheet long enough, but you’d have to twist your brain into a pretzel in the process.)

The fact is that there’s nothing necessarily wrong with private accounts being part of a Social Security package, but only if they’re based on reasonable assumptions about how much money they’ll raise. If they’re properly accounted for, tightly regulated, and honestly funded, they might be worth taking a chance on.

Needless to say, though, “properly accounted for, tightly regulated, and honestly funded” is not what we’ve come to expect from Bush administration economic policy. So unless you hear otherwise, you’d best keep your hands on your wallet when their actual proposal comes down the pike.