Bidding and Overconfidence in Experimental Financial Markets

Abstract

Overconfidence is a well-documented phenomenon in psychology. Psychologists define an overconfident individual as one who believes he has more accurate information than he actually does. Recently, behavioral economists have become interested in the implications of trader overconfidence for financial decision-making and the functioning of financial markets. To date, most financial market studies have been analytical in nature. These studies assume that traders are overconfident and model decision-making behavior accordingly. Rather than assuming the presence of overconfidence, we use experimental bidding data to determine the extent to which trader overconfidence exists, and what variables suggested by previous finance and psychology research relate to it. We find approximately 40% of subjects exhibited overconfidence. Variables that distinguish overconfident bidding from risk-averse and risk-neutral bidding include the traditional financial variables that explain bidding (expected value and standard deviation), non-traditional financial variables, and variables relating to the self-attribution bias and feedback. Contrary to what some analysts have suggested, experience did not reduce overconfidence.