Trading Complex Assets∗

We perform an experimental study of complexity to assess its effect on trading behavior, price volatility, liquidity, and trade efficiency. Subjects were asked to deduce the value of a particular asset from information they were given about the composition and price of several portfolios. Following that, subjects traded with each other anonymously in a well-defined, simple bargaining process. Portfolio composition ranged from requiring simple analysis to more complicated computation in order to deduce the value of the asset. Complexity altered subjects’ bidding strategies, decreased liquidity, increased price volatility, and decreased trade efficiency. However, in follow-up experiments, we show that uncertainty over private values does not lead to the same changes in trading behavior. Therefore, while complexity induces estimation errors and higher uncertainty, this is not what drives our results. ∗We would like to thank Kim Donghyun for excellent research assistance and finance seminar participants at the University of Texas at Austin, Michigan State University, and 2010 Miami Behavioral Finance Conference for their comments and suggestions. †Anderson School of Management, University of California, Los Angeles, 110 Westwood Plaza Suite C-413, Los Angeles, CA 90095, [email protected], (310) 825-7246. ‡McCombs School of Business, University of Texas, Finance Department, B6600, 1 University Station Austin, TX 78712, [email protected], (512) 232-6839. §McCombs School of Business, University of Texas, Finance Department, B6600, 1 University Station Austin, TX 78712, [email protected], (512) 232-6823.