The Influence of Overconfidence on Herding Behavior in Financial Decision-Making

Abstract

In financial decision-making, herding behavior is commonly observed where investors mimic each other's actions due to perceived superior information held by others [1]. Traditional approaches in financial decision-making have often utilized statistical analyses to understand such behavior [2]. However, these methods may overlook the nuanced impacts of cognitive biases on herding. One of the most robust findings in psychology is that people are overconfident [2]; this is why we are focusing on how overconfidence affects herding behavior. Overconfidence manifests in various forms: overestimation, overplacement, and overprecision—the latter describing an excessive certainty about one's knowledge accuracy [3]. Unlike overestimation and overplacement, overprecision directly impacts decision-making by influencing the tendency to overestimate the precision of one’s signal [3].

This study proposes a behavioral model that incorporates overprecision to simulate the financial decision-making process under controlled conditions. We will base our experiment on the Monte Carlo simulation, wherein we will exogenously manipulate overprecision. We will explore how overprecision specifically affects herding behavior in financial contexts, thereby addressing a gap in current research that often conflates all three types of overconfidence. This approach allows for the examination of herding behavior as a dynamic response to changes in perceived information precision.

In the experimental procedure, agents will receive a private signal indicating whether to invest in a risky or risk-free asset. They will observe the actions of their predecessors and, based on their private signal and the actions of others, will have to make a decision. The majority of the agents will be characterized as overprecise and quasi-rational Bayesian maximizers. The primary hypothesis posits that increased overprecision will lead to a noticeable decline in herding, as participants will be more inclined to rely on their private information rather than on the collective actions of others.

The significance of this study lies in its experimental methodology, which diverges from the traditional statistical analysis of herding in financial decision-making [2] by providing a controlled environment to directly observe the variables. This approach will enhance the internal validity of the findings, offering a detailed understanding of how specific types of overconfidence influence financial decisions. Additionally, by differentiating overprecision from other types of overconfidence, this study contributes to a more granular understanding of cognitive biases in financial decision-making. It is expected that these insights will not only extend academic discourse but also inform practical strategies in financial markets, where recognizing and mitigating overprecision can lead to more rational investment behaviors.

References

[1] A. Devenow and I. Welch, “Rational herding in Financial Economics,” European Economic Review, vol. 40, no. 3–5, pp. 603–615, Apr. 1996. doi:10.1016/0014-2921(95)00073-9

[2] P. T. Komalasari, M. Asri, B. M. Purwanto, and B. Setiyono, “Herding behaviour in the capital market: What do we know and what is next?,” Management Review Quarterly, vol. 72, no. 3, pp. 745–787, Mar. 2021. doi:10.1007/s11301-021-00212-1

[3] D. A. Moore and D. Schatz, “The three faces of overconfidence,” Social and Personality Psychology Compass, vol. 11, no. 8, Aug. 2017. doi:10.1111/spc3.12331