ABSTRACT We analyze a procedure common in empirical accounting and finance research where researchers use ordinary least squares to decompose a dependent variable into its predicted and residual components and use the residuals as the dependent variable in a second regression. This two‐step procedure is used to examine determinants of constructs such as discretionary accruals, real activities management, discretionary book‐tax differences, and abnormal investment. We show that the typical implementation of this procedure generates biased coefficients and standard errors that can lead to incorrect inferences, with both Type I and Type II errors. We further show that the magnitude of the bias in coefficients and standard errors is a function of the correlations between model regressors. We illustrate the potential magnitude of the bias in accounting research in four commonly used settings. Our results indicate significant bias in many of these settings. We offer three solutions to avoid the bias.

We examine whether managerial sentiment is associated with errors in accrual estimates. Using public banks we find (1) managerial sentiment is negatively associated with loan loss provision estimates, (2) future charge-offs per dollar of provision are positively associated with sentiment when the provision is estimated, and (3) the effects of sentiment are greater for firms with more uncertain charge-offs. Results are similar for private banks, suggesting accrual manipulation related to capital market incentives is unlikely to explain the results. Although economic fundamentals explain most of the variation in the provision, we find sentiment has an incremental and economically meaningful effect.

ABSTRACT Cash flows play an important role in the accounting and finance literature. They are a major component of earnings, serve as a benchmark for earnings and accruals, and are an input to firm valuation. Although accrual measurement has been extensively studied, research on cash flow measurement remains limited. We review cash flow proxies researchers use in prior literature and highlight the importance of (1) identifying and defining the cash flow construct of interest based on the underlying research question and other constructs of interest and (2) selecting a cash flow proxy that minimizes measurement error and maps well into its construct. We also examine the challenges researchers face in measuring cash flows and demonstrate how measurement choices can affect the magnitude of coefficient estimates and statistical inferences. Finally, our study provides a framework researchers can follow to strengthen the construct validity of cash flow proxies. JEL Classifications: M41; G30; C01.