Portfolio Selection with Narrow Framing: Probability Weighting Matters

This paper extends the model with narrow framing suggested by Barberis and Huang (2009) to also account for probability weighting and a convex-concave value function in the specification of cumulative prospect theory preferences on narrowly framed assets. We show that probability weighting is needed in order that investors reduce their holding of narrowly framed risky assets in the presence of negative skewness and high Sharpe ratios, which are typical characteristics of stock index returns. The model with framing and probability weighting can thus explain the stock participation puzzle under realistic assumptions on stock market returns. We also show that a convex-concave value function generates wealth effects that are consistent with empirical observations on stock market participation. Finally, we address the asset pricing implications of probability weighting in the model with narrow framing and show that in the case of negative skewness the equity premium of narrowly framed assets is much higher than when probability weighting is not taken into account.