Adding Up the Fed Model's Flawed Math

Analysts who track the so-called Fed model, a valuation tool comparing corporate earnings with interest rates, say that now is a good time buy stocks. Should you believe them?

While it's certainly possible that stocks could rise over the next few weeks or months, relying on the Fed model for clues about the market's direction is probably unwise. After all, the model says nothing about absolute expected stock returns, and it has been a poor predictor of the stock market over long periods of time.

"There is no logical basis for the model, nor was there ever a logical basis," said Larry Swedroe, director of research at Buckingham Asset Management in St. Louis and author of What Wall Street Doesn't Want You to Know.

The Fed model compares the yield on the 10-year Treasury note with the earnings yield of the S&P 500. When the earnings yield, defined as the S&P's expected earnings divided by the price, is higher than the Treasury yield, stocks are considered undervalued relative to bonds. Conversely, when the earnings yield is lower, stocks are considered relatively overvalued. While the Federal Reserve has never formally admitted to using the tool, a report to Congress in 1997 suggested that it was something the bank looked at.

Right now, the model is suggesting that stocks are cheap in comparison with Treasuries. The earnings yield on the S&P 500 is sitting at 5.6% compared with a yield of 4.6% for the 10-year note.

Even if the model is right and stocks are cheap relative to bonds, that doesn't mean stocks are cheap on an absolute basis. The model could simply be saying that stocks aren't as wildly overvalued as bonds. This point is often lost on many analysts, who use the Fed model to gauge absolute stock returns.

Of course, there were times when the model seemed to predict turns in the market. In early 2003, for example, it was showing that stocks were more than 40% undervalued relative to bonds. By the end of the year, the S&P 500 had climbed 26%. The Fed model also seemed to foreshadow big losses in August 1997 and in March 2000.

< Previous

1 2

Next >