Covariance Risk, Mispricing, and the Cross Section of Security Returns

Working Paper: NBER ID: w7615

Authors: Kent D. Daniel; David Hirshleifer; Avanidhar Subrahmanyam

Abstract: This paper offers a multisecurity model in which prices reflect both covariance risk and misperceptions of firms' prospects, and in which arbitrageurs trade to profit from mispricing. We derive a pricing relationship in which expected returns are linearly related to both risk and mispricing variables. The model thereby implies a multivariate relation between expected return, beta, and variables that proxy for mispricing of idiosyncratic components of value tends to be arbitraged away but systematic mispricing is not. The theory is consistent with several empirical findings regarding the cross-section of equity returns, including: the observed ability of fundamental/price ratios to forecast aggregate and cross-sectional returns, and of market value but not non-market size measures to forecast returns cross-sectionally; and the ability in some studies of fundamental/price ratios and market value to dominate traditional measures of security risk. The model also offers several untested empirical implications for the cross-section of expected returns and for the relation of volume to subsequent volatility.

Keywords: covariance risk; mispricing; security returns; cross-section

JEL Codes: G1


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
covariance risk (C10)expected returns (G17)
mispricing variables (D46)expected returns (G17)
systematic mispricing (G19)expected returns (G17)
fundamental-price ratios (F31)expected returns (G17)
market value (D46)expected returns (G17)
non-market measures (P42)expected returns (G17)

Back to index