Dechow and Dichev (2002) model earnings quality as the magnitude of estimation errors in accruals, and provide empirical estimates of this construct based on the relation between accruals and cash flows. I characterize the innovation and limitations in this approach, and provide empirical evidence of measurement error in their empirical specification. I also adapt their model to assess the specification of the Jones' (1991) model and document that this model provides estimates of discretionary accruals that are significantly associated with cash flows, which are likely to be substantially nondiscretionary. I conclude with suggestions for future research on earnings quality and earnings management.

This content is only available via PDF.
You do not currently have access to this content.