A Review of Overconfidence in Behavioral Finance

Behavioral finance can be dichotomized into limits to arbitrage and cognitive psychology. While limits to arbitrage describes when markets are inefficient, cognitive psychology refers to how people think. In this review we focus on one well-established bias of cognitive psychology, namely overconfidence. Throughout the literature there are three distinct ways in which overconfidence has been defined: (1) overestimation, (2) overplacement and (3) overprecision. We shortly summarize the ideas of conventional theory about overconfidence, and present the most significant findings of the relevant literature. Hereby we differentiate between studies that investigate overconfidence on the micro level (i.e. individual-level data) and the macro level (i.e. aggregated-level or market-level data). We will also discuss inconsistency across different measures of overconfidence and the reverse phenomena of overconfidence, i.e. underconfidence. To stay within the scope of this work we place our main focus on experimental studies.