Behavioral bias-driven trading and return momentum

In this paper, we provide evidence that trading driven by investors' behavioral biases contributes to stock return momentum. In particular, we focus on two types of irrational trading, momentum trading and confidence-influenced trading, which could be driven by psychological biases introduced by several previous studies. We develop a model to study when there are more than one traders engaged in both types of trading, the aggregation and interaction of the trading, and test our model implications accordingly. The major implication of our model is that aggregated momentum trading would drive momentum in return. Aggregated momentum trading, however, is affected by two factors: the information processing cost and traders' aggregated confidence. Using R&D intensity and the Consumer's Confidence Index (CCI) as proxies for information processing cost and aggregated confidence respectively, our empirical study yields the following results that are consistent with our model: momentum profits are positively associated with firms' R&D intensity and/or the CCI. These results are robust to alternative proxies for information processing cost, the GDP adjusted CCI, the industry adjustment, and the macroeconomic variable adjusted returns. Our study also yields evidence on market mispricing due to over- or under-reaction. Collectively, our findings suggest that investors' behavioral limitations contribute to the return momentum.