Some historical perspectives on the Bond-Stock Earnings Yield Model for crash prediction around the world

Abstract

We provide a historical background on Ziemba’s experiences and research on the bond-stock earnings yield differential model (BSEYD), from the time he first used it in Japan in 1988 through to the present in 2014. Over this period, the model has called many crashes, but not all. Those called have had high interest rates in long term bonds relative to the trailing earnings to price ratio. In general, there will almost always be a crash if the model is in the danger zone. The model predicted the crashes in China, Iceland and the US in the 2006–09 period. Iceland had a drop of fully 95%. For the US, the call was on June 14, 2007, and the stock market fell 56.8%. A longer-term study for the US, Canada, Japan, Germany, and the UK shows that, over long periods, being in the stock market when the bond-stock signal is not in the danger zone, and in cash when it is in the danger zone, provides a final wealth which is about double that of a buy and hold strategy for each of these five countries. The best use of the model is for predicting crashes. Finally, we compare Shiller’s high PE ratio crash model to the BSEYD model for the US market from 1962–2012. While both models add value, the BSEYD model predicts crashes better.