ABSTRACT: I examine whether firms alter their behavior in response to changes in accounting standards that mandate new financial statement disclosures. While prior research suggests that new recognition rules lead to changes in firm behavior, there is limited evidence that disclosure rules can impact firm behavior. This study helps to fill this void in the literature by examining the economic consequences of the mandated disclosures of pension asset composition required under SFAS 132R. Under pension accounting rules, the composition of pension assets is a key determinant of the assumed expected rate of return (ERR) on pension assets. I find that when firms disclose asset composition for the first time under SFAS 132R, firms that were previously using upward-biased ERRs respond by increasing asset allocation to high-risk securities and/or reducing the ERR assumption. While disclosure requirements arguably create less powerful incentives to alter firm decisions than recognition requirements, these findings offer evidence that firms alter behavior in response to disclosure standards. Data Availability: The data used in this study are publicly available from the sources indicated in the text.

Earnings distributions commonly exhibit statistically significant discontinuities at zero earnings, which are widely interpreted as evidence of earnings management to avoid a loss. In contrast, Durtschi and Easton, 2005, Durtschi and Easton, 2009 assert that discontinuities are instead explained by some combination of prior researchers׳ choice(s) of scaling and sample selection as well as a scaling-related effect due to a systematic relation between the sign of earnings and market prices. Resolution of the conflicting interpretations of discontinuities is important because (1) it affects how investors, regulators, and scholars view earnings management and (2) it demonstrates the importance of a close linkage between theory and research design choices. We point out that DE provide no evidence that scaling or selection create discontinuities, but only evidence showing that changes in scaling or selection eliminate discontinuities. We demonstrate why the research designs used by DE eliminate discontinuities and why alternative designs using the same data yield statistically significant discontinuities that cannot be attributed to either scaling or selection.

Abstract The accounting literature includes numerous examples of discontinuities at prominent benchmarks that are widely interpreted as evidence that earnings are managed to meet those benchmarks. However, there are a few examples where discontinuities do not exist, which are sometimes interpreted as inconsistent with earnings management. Alternative explanations for discontinuities have also been suggested. This paper reviews, evaluates, and integrates the available evidence and concludes that the theory that earnings are managed to meet benchmarks provides the simplest and most complete explanation for the body of discontinuity‐related evidence.

ABSTRACT Experts have long conjectured that pension accounting rules, by which pension expense depends on a managerial estimate that is directly tied to the riskiness of plan assets (i.e., the expected rate of return, or ERR, on plan assets), encourage risk-taking with pension investments. The recent passage of IAS 19, Employee Benefits (Revised) (hereafter, IAS 19R) eliminates the ERR and replaces it with a managerial estimate unrelated to plan asset riskiness (the discount rate). We demonstrate that a sample of Canadian firms affected by IAS 19R reduces risk-taking in pension investments post-IAS 19R, compared to a control sample of U.S. firms unaffected by IAS 19R. Therefore, removing firms' ability to recognize immediately in net income the expected higher returns from risk-taking (via a higher ERR) reduces their propensity for that risk-taking—providing some of the first empirical evidence on the economic consequences of eliminating the ERR-based pension accounting model.

This paper examines the characteristics of management forecasts available on Thomson First Call’s Company Issued Guidance (CIG) database relative to a sample of forecasts hand-collected through a search of company press releases. Due to the significantly lower cost of using CIG (relative to hand-collecting data), academics have increasingly relied on this database as a source of management forecasts. However, it is important for researchers to consider the properties of this database (such as coverage, accuracy, and breadth) when evaluating whether it is an appropriate data source for their study. Overall, our results suggest systematic differences between forecasts reported on CIG and forecasts gathered from company press releases. We suggest several sample criteria that will remove or mitigate these biases.