Does the Pricing of Financial Reporting Quality Change Around Dividend Changes?


  • We thank Jennifer Francis, Ryan LaFond, Per Olsson, and Katherine Schipper for sharing the AQ factor returns data and I/B/E/S for providing us with the analyst data used in this project. We appreciate the comments received from the Editor, Ray Ball, and an anonymous referee. We are also grateful for comments received from Doron Nissim; K. R. Subramanyam; Michael Kirchenheiter; and workshop participants at the University of California, Berkeley, the Columbia 2004 Burton workshop, and the 2005 UBCOW conference.


We examine whether accrual earnings quality is a priced information risk factor in a dividend change setting. We define information risk as the probability that firm-specific financial statement information pertinent to investor pricing decisions is of low precision, and use the factor-mimicking portfolio returns formed on the Dechow-Dichev [2002] accrual quality (AQ) metric to proxy for the information risk (IR) factor returns. We augment the Fama-French three-factor model with this IR factor, and find that dividend initiation and increase firms exhibit a decrease in the factor loadings on the IR factor while dividend decrease firms exhibit an increase in the corresponding factor loadings, but such changes in the factor loadings occur months prior to the dividend change announcements. The results are robust to further controls for operating risk and using an alternative measure of information risk. Further analysis on changes in information characteristics such as AQ, the probability of informed trading score (PIN), forecast dispersion, and return volatility surrounding dividend change events are consistent with the asset pricing results. Overall, we interpret our results as being consistent with investors treating the information risk associated with the precision of financial statement information as a priced risk factor, with both the precision and pricing changing in predictable directions around dividend changes. However, while we attempt to control for operating risk changes in additional tests, we cannot completely rule out changes in operating risk as a competing alternative explanation for our observed results.