Herd behavior, where investors mimic the trades of others rather than relying on their own information or analysis, is a widespread phenomenon that can lead to suboptimal investment decisions and market inefficiencies. Hence, understanding how to reduce this behavior is crucial for improving individual investment outcomes and sustainable market performance. Though the literature suggests that self-reflection and financial literacy have potential to mitigate herd bias, it is unclear as to how these two perspectives interact to minimize this irrational behavior. By drawing on insights from the adaptive market hypothesis, transformative learning theory, concept of bounded rationality and dual process theory, this study attempts to enhance the understanding of this phenomenon by introducing a conditional mediation model that promotes self-directed learning. It predicts that investors can reduce their herd bias through self-reflection on their past stock trading experiences, which they can practice by engaging intuitive logical thinking and strengthen further through their financial literacy. Since herding is generally predicted to be more prevalent in frontier stock markets, the study was conducted in a frontier market - the Colombo Stock Exchange. The data was obtained through a self-administered questionnaire from 253 active individual investors, and analyzed by applying the PROCESS procedure. The findings reveal the significance of self-reflection as a mediating variable across its antecedents, which provides strong evidence for its central role in reducing herd bias. Notably, the findings indicate that self-reflection has a stronger effect on reducing herd bias of investors with lower level of financial literacy. It implies that investors with low financial literacy are more susceptible to herd bias, and self-reflection facilitates them to recognize and minimize their herd bias. Accordingly, the study concludes that promoting self-reflection can empower individual investors to become more financially literate and thereby mitigate their herd bias.
Read full abstract