We provide new evidence about how analysts incorporate and improve on management ETR forecasts. Quarterly ETR reporting under the integral method provides mandatory point-estimate forecasts by management, but firms must record certain “discrete” tax items fully in the quarter in which they occur, polluting these forecasts. We investigate management ETR accuracy, analysts' decisions to mimic management's estimate, analysts' accuracy relative to each other or to management, and dispersion. Our comprehensive analysis reveals that analysts deviate from management more and are more accurate relative to management as complexity increases, with real effects on EPS accuracy and dispersion. In contrast to prior research that analysts ignore or are confused by taxes, we provide evidence that analysts pay attention to taxes and improve on management estimates. Based on our evidence that management's quarterly ETRs have less predictive value in the presence of discrete items, we suggest standard-setters reexamine the discrete item exception to require more disclosure.
In this study, we examine whether banks’ use of the loan loss provision (LLP) to manage earnings is associated with (a) the extent to which banks hold assets subject to fair value reporting and (b) the use of an industry specialist auditor. We find that banks with a greater proportion of assets subject to fair value reporting (i.e., higher fair value exposure) use less LLP-based earnings management but more transaction-based earnings management (i.e., earnings management achieved by timing the realization of gains/losses). We also find that banks engaging industry specialist auditors use less LLP-based earnings management. Our findings suggest that banks’ use of the LLP to manage earnings is more limited when they have access to alternative earnings management tools and when they engage an auditor with more industry knowledge. Our results should be informative to regulators, members of the banking industry, and academics interested in the earnings management behavior of banks.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.