Forecasting the Aggregate Stock Market
The early 2002 forecast for 2002 earnings per share for the S&P 500 portfolio was about $53.16 The 10 year Treasury bond yield was about 5.2%. Because the earnings yield on the S&P 500 has been about 1 percentage point below the 10 year Treasury yield, a first guess for the earnings yield on the S&P 500 might be 4.2%. This would imply a P/E ratio of 1/.042 23.8. Our forecast for the S&P 500 index would then be 23.8 + 53 1,261.
Of course, there is uncertainty regarding all three inputs into this analysis: the actual earnings on the S&P 500 stocks, the level of Treasury yields at year end, and the spread between the Treasury yield and the earnings yield. One would wish to perform sensitivity or scenario analysis to examine the impact of changes in all of these variables. To illustrate, consider Table 18.4, which shows a simple scenario analysis treating possible effects of variation in the Treasury bond yield. The scenario analysis shows that forecasted level of the stock market varies inversely and with dramatic sensitivity to interest rate changes.
Some analysts use an aggregate version of the dividend discount model rather than an earnings multiplier approach. All of these models, however, rely heavily on forecasts of such macroeconomic variables as GDP, interest rates, and the rate of inflation, which are difficult to predict accurately.
Because stock prices reflect expectations of future dividends, which are tied to the economic fortunes of firms, it is not surprising that the performance of a broad based stock index like the S&P 500 is taken as a leading economic indicator, that is, a predictor of the performance of the aggregate economy. Stock prices are viewed as embodying consensus forecasts of economic activity and are assumed to move up or down in anticipation of movements in the economy. The government’s index of leading economic indicators, which is taken to predict the progress of the business cycle, is made up in part of recent stock market performance. However, the predictive value of the market is far from perfect.
A well known joke, often attributed to Paul Samuelson, is that the market has forecast eight of the last five recessions.