But it is hard to talk up dud firms when markets are falling, and anyway, there is little business to be won at such times. So it might have been reasonable to assume that analysts' recommendations are better in bearish markets than bullish ones. New research, alas, suggests this is not so: the advice analysts give in bad times seems to be even worse than the boosterism they peddle in good.*
Roger Loh of Singapore Management University and René Stulz of Ohio State University looked at analysts' forecasts of profits and the buy or sell recommendations they issued for the period 1983-2011. Their predictions, it turned out, were less reliable in falling markets than in rising ones, even after making allowances for increased volatility in such times. Analysts' forecasts of profits for the next quarter were out by 46% more during periods of financial crisis than at other times, for instance.
The drop in accuracy may be linked to cuts in research budgets. During downturns banks spend less on research. For instance, in the most recent crisis budgets were cut by around 40%, according to Neil Scarth at Frost Consulting, largely by replacing more experienced (and more expensive) analysts with younger, greener ones. The fear of being fired may also befuddle rather than focus minds.
Ironically enough, Messrs Loh and Stulz also found that investors pay more attention to analysts' opinions when times are tough. Normally only one change in ten in analysts' stock recommendations moves the price of the share in question. But the proportion increases to one in seven in falling markets, even though there are more changes during market routs. Just as drivers value maps more when it is foggy, investors pay more heed to research during periods of increased uncertainty, reckons Mr Stulz. Unfortunately for them, that is also when their maps are most likely to be wrong.
Click here to subscribe to The Economist
