Anchoring bias happens when people overemphasize a starting point when they make predictions or decisions. In the case of analysts, Ling Cen, Gilles Hilary and K.C. John Wei show that their forecast earnings per share (FEPS) are influenced by the industry median, which in turn leads them to underestimate future profitability for firms with abnormally high FEPS compared to the industry norm, and overestimate those with lower FEPS.

“Estimating the future profitability of a firm is a complex task and involves a high degree of uncertainty, suggesting market participants may anchor on salient but irrelevant information,” they said.

“Our empirical results consistently support the idea that anchoring, a significant cognitive bias in psychology literature, affects the decision-making of individuals in this important economic setting.”

They developed three tests of anchoring and applied them to a large U.S. data sample from 1983-2005. The results showed anchoring was indeed taking place and managers were reacting to it.

First, analysts’ forecasts of EPS were found to be closer to the industry norm than they should be, leading them to underestimate future earnings growth of firms with a high FEPS.

Second, expectations of future earnings were biased and, once true earnings were revealed, stocks with a high FEPS tended to outperform similar stocks in the same industry that had low FEPS.

And third, managers appeared to be aware of these biases and to act on them by splitting shares, so that their FEPS was lower relative to that of other firms.

“We find when a company’s current FEPS is already much higher than those of its industry peers, analysts appear to make insufficient upward adjustments even if such adjustments are well supported by fundamental information. They also make insufficient downward adjustments for firms with lower FEPS,” the authors said.

They also found an anchoring bias in investors. In general, the anchoring bias was stronger when the industry norm was more stable or when market participants were less sophisticated.

They authors not only provide new insights into analyst and investor behavior, but also suggest profitable strategies that could be followed based on an understanding of the anchoring bias in forecasts:

“Our results suggest a hedge portfolio that goes long on firms with a high cross-sectional anchoring bias and short on forms with a low bias could, over the period studied, have generated a risk-adjusted return of 0.71 per cent per month, or 8.52 per cent per year. The profitability of such a trading strategy remains significant for investment horizons spanning at least 12 months.

“Our results also suggest a corporate strategy based on stock splits for managers of firms with a high level of FEPS. Such a strategy can mitigate under-valuation and sometimes over-valuation by influencing analysts’ earnings forecast or revisions.”