THE BUSINESS CYCLE, FORECAST HORIZONS AND STOCK RETURN PREDICTABILITY
MetadataShow full item record
This paper investigates the impact of the business cycle when forecasting equity returns over different forecast horizons. Weigand and Irons  show that relative valuation matters when forecasting long-term returns to the market, a conclusion that is verified in this study with reference to the period 1934-1949. This study reveals that the business cycle matters when forecasting short-term (one year or less) market returns, over the entire period of study (1934-1999). An econometric issue first noted in Weigand and Irons  is verified and pinpointed more precisely in time, to the year 1950. The market earnings yield (the inverse of the market P/E ratio), one of the foremost predictors of future market returns, is shown to be ineffective in forecasting returns over any forecast horizon in the second half of the 20th century. This result is further evidence of the impact of investors' belief in the Fed Model, which equates the yields on bonds to the returns on stocks. The term spread of interest rates, one of the business cycle proxies used in this paper, is shown to have a significant impact when forecasting market returns over any horizon during the period 1950-1999, which is consistent with the Fed Model.