Confidence and Investors' Reliance on Disciplined Trading Strategies

Authors


  • We appreciate helpful comments received from Ray Ball (the editor), Tom Dyckman, Jeffrey Hales, Bob Libby, Jeff Wilks, an anonymous reviewer, and workshop participants at Cornell University, Emory University, Harvard University, the University of North Carolina at Chapel Hill, and the 2001 AAA Doctoral Consortium. We are grateful for financial support provided by the Johnson Graduate School of Management at Cornell University, and for programming support provided by Don Honeycutt.

abstract

Researchers and practitioners in accounting and finance often investigate or advocate particular disciplined trading strategies, but little work investigates the determinants of individual investors' trading-strategy reliance. We report two experiments, which provide evidence that the dual-source model of overconfidence (Sniezek and Buckley [1991]) predicts the circumstances in which investors are more likely to rely on disciplined trading strategies. Our results indicate that reliance is more likely when investors trade portfolios of securities rather than trading on a case-by-case basis, particularly when investors have received feedback that their previous (unaided) trading decisions have been unprofitable. These results are driven by the number of shares that investors transact rather than by investors' directional agreement with the recommendations of the trading strategy, suggesting that the effects of a portfolio approach and trading experience occur by mitigating investors' overconfidence. The effects violate an aspect of economic rationality because our experiments ensure that investors in all conditions trade the same set of securities based on the same set of information.

Ancillary