Study of consensus predictions hints at how ordinary investors can be more spot-on than analysts
To guide their investment and portfolio-management decisions, many investors rely on the yearly financial-market outlooks offered by strategists and economists. But just because investors rely on those outlooks doesn’t mean that those predictions are reliable.
In a blog post published by the CFA Institute, Joachim Klement, CFA, noted that expert forecasts on the direction of a market in any given year tend to diverge, with many mainstream predictions foretelling that all will be just fine, and a few extreme ones swinging toward either doom or boom.
“[I’ve examined] the S&P 500 return forecasts of investment professionals over the last 20 years,” Klement said, citing an analysis from a forthcoming book. “In 13 of those years, beginning-of-the-year predictions were off by more than 10 percentage points. Often, they didn’t even correctly call the stock market’s direction.”
Klement took the analysis further, breaking down the data based on how often the annual analyst consensus predicted a rise or a fall, and how often predictions were right. With that breakdown, he found that consensus predictions on the stock-market direction were correct in just nine of the past 20 years.
“New research … shows that the consensus forecast of individual economists beats even that of the most skilled single economist,” he said. “So investors should rely on the wisdom of the crowd and follow the expert consensus forecast rather than any individual prediction.”
Given how often consensus calls go wrong, that may not be the most encouraging advice. But another piece of research, which Klement said never got wide coverage because it was published in German, suggests a simple method to outperform even experts’ consolidated forecast.
“Oliver Hein and Markus Spiwoks analyzed more than 150,000 stock market, interest rate, and exchange-rate forecasts compiled by the German ZEW Institute between 1995 and 2004,” he said. “These forecasts sought to predict six international stock markets and the interest rates in these markets, as well as the major exchange rates, for the next three and 12 months.”
The researchers found that the consensus forecast, which typically beat almost all individual predictions, were more correlated with where stock markets and interest rates stood at the time of the prediction than the time the prediction aimed to forecast.
The takeaway, according to Klement, is that “investors are better off assuming that nothing will change at all.” In other words, standing pat by predicting that stock markets or interest rates will not change between today and a year from now tends to be not just more accurate than the most skilled individual forecast, but also more accurate than the consensus.
“So when it comes to end-of-year forecasts, economists and analysts should avoid making them and investors should avoid listening to them,” he said.