Don’t Confuse Brains with a Bull Market: Attribution Bias, Market Condition, and Trading Behavior of Individual Investors

Abstract:     
Theory predicts that attribution bias — inflated confidence in one’s own skill — creates overconfident traders. This paper tests this prediction by comparing the trading behavior of individual investors in different market conditions. In a bull market, investors incorrectly attribute trading successes (luck) to their own abilities and therefore should be more overconfident than they are in a bear market. Using the trading records of Chinese individual investors from January 2005 to November 2008, we find that individual investors exhibit more overconfidence in a bull market than in a bear market, where overconfidence is measured by excessive trading. Specifically, we find that in the bull market the stocks bought by individual investors significantly under-perform those sold in the subsequent one month and three months. In the bear market, however, individual investors do not make similarly suboptimal trading decisions. In addition, the poor trading decisions made in the bull market are due to poor security selection, not to poor market timing. Finally, we demonstrate that individual investors turn over their portfolios more frequently and that their risk-adjusted performance net of transaction costs is significantly worse in the bull market than in the bear market. Overall, these results support the argument that attribution bias creates overconfident traders.

Number of Pages in PDF File: 37

Keywords: Attribution bias, individual trading behavior, bull market, bear market

JEL Classification: G02, G11, G12

working papers series