Macro Conditions May Enhance Short-term Predictability Of The Shiller P/E -

Macro Conditions May Enhance Short-term Predictability Of The Shiller P/E

Aug 14, 2018Jeremy ParkinsonFinance, No picture0

King of the Mountain: The Shiller P/E and Macroeconomic Conditions

Robert D. Arnott, Denis B. Chaves, and Tzee-man Chow

Journal of Portfolio Management

A version of this paper can be found here

What are the research questions?

Is there a relationship between real yields and short-term market valuation?

Is there a relationship between inflation rates and short-term market valuation?

Does the predictive power of the Shiller P/Eimprove by using yields and inflation?

What are the Academic Insights?

YES. The authors describe a “mountain-shaped” relationship between stock market valuations and inflation rates and real yields. When levels of inflation and real yields are moderate, then the Market P/E rises above 16.7 (the unconditional historical average). When inflation or the real yield reside at extreme levels, the P/E is noticeably lower. For the U.S., the Market P/E is generally at it’s highest when real yields are approximately 3% to 4%. When real yields are below 1% or above 6%, the P/E drops from 19.6 to 10.7. Outside this narrow interval, median P/Es fall rather quickly from a peak of 19.6 to 10.7 when real yields are below -1% and to 10.5 when real yields are above 6%.

YES. A similar pattern exists for inflation as does for real yields, relative to market valuation. The median market P/E of 20.3 occurs when inflation is between 2% and 3%. However, as inflation increases the P/E declines faster than that observed for real yields. On the other side, the authors do not observe a steep decline in the P/E when the inflation rate also declines. Perhaps an equity investment is not as effective as a hedge against inflation as conventional wisdom suggests.

YES. Using a regression model with a defined Gaussian function conditional on current levels of inflation and real yields provides superior forecasts for forecast horizons of 3 years or less. At longer horizons, the conditional model produces inferior results for periods of 3 to 5 years or longer. At that span, the basic Shiller P/E model maintains its impressive well-documented performance.