Now there is scientific evidence to back up the concept that stock market predictors for the future year are virtually worthless. The person who makes them is wasting his breath and, worse, the person who believes them is naÃ¯ve.
A potential new client wants to meet. Among the things he wants to discuss is my prediction for the stock market in 2013. That is, of course, a big order. Can I (or anyone else) make an accurate prediction?
Unfortunately, for those who believe in hope over reality, the answer is “No.”
Not only is there no crystal ball, there is scientific evidence to back up this concept: stock market predictors for the future year are virtually worthless. The person who makes them is wasting his breath and, worse, the person who believes them is naïve or stupid.
This is why: Joseph Davis, PhD, Roger Aliaga-Díaz, PhD, and Charles J. Thomas, CFA, from Vanguard analyzed the predictive value of market metrics for U.S. stock returns since 1926. Their paper — “Forecasting Stock Values: What signals matter and what do they say now?” —suggests that the usual criteria that managers and investors use when they make stock adjustments are of low or no predictive importance.
As an example, all of the one-year metrics had a predictive value less than 20% and of the 12, most were zero. These included price earnings ratio (P/E), government debt/gross domestic product, dividend yield, corporate profit margins, trailing one-year returns, and other less familiar terms. Rain-fall was even thrown in to show there was no correlation with it.
Ten-year predictors did better. P/E10, which uses trailing 10-year earnings, had the best score of 0.43. It is Yale University professor Robert J. Shiller’s cyclically adjusted P/E or “CAPE.” This means that 43% of stocks’ performance over the next decade can be explained by it. Of course, the majority, or 57%, cannot. Other indicators did even worse over a decade.
All of this leaves investors (and managers, too) with a clear message: Don’t believe in one-year predictions and be skeptical of ten-year ones.
Davis, one of the authors, says in Vanguard's long-range forecast for U.S. stocks, "…the fact that even P/E’s — the strongest of the indicators we examined — leave a large portion of returns unexplained underscores our belief that expected stock returns are best stated in a probabilistic framework, not as a 'point forecast,' and should not be forecast over short horizons."