Price momentum as documented by Jegadeesh and Titman (1993) and earnings momentum as originally documented by Ball and Brown (1968) and later confirmed by Bernard and Thomas (1989) and others are the two prominent anomalies that cannot be explained by the Fama and French (1993) three-factor models. A number of explanations have been proposed to explain these two anomalies, including rational models and behavioral models. Behavioral explanations try to reconcile the anomalies with investor cognitive biases such as investors’ overconfidence and self attribution bias as suggested by Daniel, Hirshleifer and Subrahmanyam (DHS, 1998) and investors’ initial underreaction to new information as suggested by Barberis, Shleifer, and Vishny (BSV, 1998) and Hong and Stein (1999). Rational explanations, however, seek solutions using either the serial correlations in time-varying expected returns or the cross sectional differences in unconditional expected returns. To rationalize the positive serial correlation, Berk, Green, and Naik (1999) develop a model based on optimal asset portfolio decisions of the firm, while Johnson (2002) introduces the growth rate shock that is episodically persistent. From a different perspective, Conrad and Kaul (1998) find a nontrivial role of cross sectional dispersion in mean returns and use this finding to justify that momentum profits are attributable to risk.
Given the fact that realized returns can be decomposed by an expected return component and a return innovation component (i.e., an unexpected return component), return decomposition may provide a framework in which different explanations can be potentially tested. If the expected return component plays a dominant role in explaining momentum profits, momentum profits are more consistent with the risk-based explanation. In contrast, if the return innovation component plays a dominant role in explaining momentum profits, then momentum profits might be more consistent with the behavioral justification.
The traditional approach of return decomposition is to decompose stock returns using a vector autoregression (VAR) method to model the discount rate news component as suggested by Campbell and Shiller (1988) and Campbell (1991) and to use the residual from the return regression as a proxy for the cashflow news component. To circumvent the model misspecification or low predictive power problems, Chen and Zhao (2009) propose a return decomposition method that is based on accounting valuation models and analysts’ earnings forecasts. Chen and Zhao (2009) find empirical evidence that supports the new intuition on the time-series behavior of stock price movements. In this paper, we extend the Chen and Zhao's (2009) method to investigate momentum anomalies in the cross section. Specifically, we separate fundamental drivers for momentum anomalies through decomposing realized stock returns into the expected return component, the cash flow return component and the discount rate return component. The three return components capture stock price movements that are associated with the compensation for risk, the change in expected earnings prospects, and the change in risk, respectively.
The expected return component, a measure related to the implied cost of equity, is a proxy for ex ante expected compensation for risk. The discount rate return component reflects the risk-adjustment effect that may incorporate investors’ time-varying risk aversion or sentiment. The underlying dynamic is that a positive discount rate return will decrease the future expected return. In other words, a decrease in risk causes the current stock price to go up and the expected future return to go down.
Using this return decomposition method, we expand the dimension where differences in price momentum and earnings momentum can be inferred. Chan, Jegadeesh and Lakonishok (1996) document that past returns and past earnings surprises contain orthogonal information that predicts drifts in future returns. Chordia and Shivakumar (2006) focus on the systematic components in price momentum and earnings momentum. Their results from factor regressions find that when the earnings momentum factor is included as one of the factors in the regressions, the price momentum no longer exists. In contrast, when the price momentum factor is included as one of the factors in the regressions, the earnings momentum still exists. The results lead them to claim that earnings momentum subsumes price momentum. In this paper, we attempt to examine the patterns of different return components in distinguishing these two related momentum anomalies. We aim to explore whether the two anomalies differ in terms of cash flow news or discount rate news or both. Since cash flow news dominates in explaining stock price movements in long horizons (Chen and Zhao (2009)), the relative contributions of the cash flow return component and the discount rate return component might provide useful implications for the long-run performance of these two momentum strategies.
Download
PDF Ebook Price and Earnings Momentum: An Explanation Using Return Decomposition
