Another look at anchoring and stock return predictability

Abstract

The superior performance of a momentum strategy long in stocks trading near their 52-week high prices and short in stocks trading far from their 52-week high prices is well-documented. In contrast, recent research finds that a similar strategy based on historical high prices exhibits subsequent reversals instead. This paper shows that after excluding low-priced stocks and/or January returns from the sample, the stocks trading near their historical high prices, in fact, exhibit significant outperformance. In particular, in a sample without low-priced stocks, a strategy long in 10% of the stocks with prices nearest to their historical high prices and short in 10% of the stocks with prices furthest from their historical high prices earns an average monthly return of 0.93% in non-January months. The performance of 52-week high momentum strategy also improves significantly upon exclusion of low-priced stocks and/or January returns. These findings have important implications for the anchoring-based behavioral explanations of these return patterns.