Blog - Investing Notes

October 12, 2004 – Fed Model: Predictive or Not?

Many investors monitor the Fed Model, based on the relationship between the earnings yield of stocks and the bond yield, for long-term stock market timing signals. Does this model really work? Notable contrary arguments are found in the December 2002 paper entitled "Fight the Fed Model: The Relationship Between Stock Market Yields, Bond Market Yields, and Future Returns" by Clifford S. Asness and the 2004 paper entitled "A Tactical Implication of Predictability: Fighting the Fed Model" by Roelof Salomons. These two papers present similar analyses and conclusions, as follows:

• Investors should use fundamental measures of valuation, such as the price/earnings (P/E) ratio, for long-term forecasting of returns from stocks.
• The Fed Model helps explain why P/E varies over time, especially when adjusted for stock and bond relative volatility (risk), but it is not a useful predictor of long-term future returns from stocks.
• A risk-adjusted (in other words, complicated) application of the Fed Model can identify short-term opportunities for returns from stocks based on fluctuations between stock and bond yields.

In summary, the Fed model is inferior to fundamental valuation in predicting long-term stock returns, but it may have some tactical value.

Salomons extends the efforts of Asness, but Asness is clearer in presentation than Salomons, perhaps due to translation of the latter.

For a collection of recent research related to the Fed Model, see Blog Synthesis: Gunning for the Fed Model?.