Faculty Profile

This paper explores how managers alter the nature of their voluntary management guidance as they become more reliance on debt versus equity financing.

Prior research documents a negative contemporaneous relation between earnings changes and stock returns at the aggregate level. We find that the aggregate earnings-return relation turns significantly positive in recent years (1992-2013). As explanations for that shift we test over-time shifts in the properties of firm-level earnings and the magnitude of inflation rates. To do so we use the Bureau of Economic Analysis’ (BEA) corporate profits series as an alternative measure of aggregate earnings because it is unaffected by changes in GAAP measurement rules and is adjusted for changes in price levels. While changes in both aggregate GAAP earnings and the BEA’s corporate profits series exhibit a significant negative (positive) relation with aggregate returns in the 1970-1991 (1992-2013) period, changes in aggregate GAAP earnings have incremental explanatory power for aggregate returns beyond the BEA’s corporate profit measure in the 1992-2013 period. Additional analysis reveals that such incremental explanatory power stems from GAAP earnings’ inclusion of non-operating items. We also find that a shift over time in inflation, as well as in the proportion of loss-making firms, contribute to inter-temporal variation in the aggregate earnings-returns relation. The effect of changes in the properties of GAAP earnings and inflation rates over time on the aggregate earnings-return relation is new to the literature.

This paper analyzes how earnings guidance effects bond prices.

This paper studies the actions taken by managers, analysts, and investors in response to the information void created when firms cease proving earnings guidance.

This paper analyses how debt and credit analysts use earnings forecast information.

This paper examines alternative approaches to handle influential observations in accounting research and how inferences are affected by that choice.

Much recent research focuses on earnings management via the manipulation of real activities (hereafter real earnings management or REM). While tests of REM hinge critically on the underlying model of abnormal real activities there is no systematic evidence on the properties of commonly used REM measures or of the specification of tests based on them. This study provides such evidence by documenting which REM measures lead to well-specified tests, and which do not, across a wide variety of settings encountered in accounting research. The results indicate that the traditional REM measures used (to-date) in the literature are severely mis-specified in that their Type I error rates differ significantly from nominal significance levels (e.g., 1% or 5%). To address such mis-specification we analyze alternative REM measures based on performance-matching. While performance-matched REM measures are not well-specified in each and every setting (no REM measure) the weight of the evidence suggests they will provide better-specified tests, compared to traditional REM measures, across a wide variety of settings. To further document the importance of our analysis we replicate recent research examining whether managers manipulate real activities to avoid reporting losses. This replication illustrates the importance and benefits of performance matching and shows that inferences change when performance-matched REM measures are used. Our evidence facilitates a keener interpretation of the inferences drawn in prior REM studies as well as informing future researchers on the trade-offs involved in measuring REM. Overall, the findings indicate that performance-matched REM measures provide a more reliable basis from which to draw inferences about REM than traditional measures. In some settings, however, performance-matched REM measures (as well as traditional REM measures) may provide conservative tests of REM-related hypotheses.

We investigate how use of Non-GAAP earnings measures in executive compensation plans effects managers’ income statement reporting/classification decisions. We predict and find that use of such measures (often called pro-forma earnings) leads managers to shift losses from core earnings to non-core earnings and that such firms are more likely to report special items. We also find that restatements of special items are more highly associated with use of non-GAAP executive compensation measures. The findings are robust to a variety of sensitivity/robustness tests including use of a propensity score-matched sample.

This study examines the economic factors that determine the mix of the different types of voluntary management guidance issued across firms and through time.

This papers develops and tests hypotheses about the use of non-GAAP measures in executive compensation plans including: 1) What are the economic determinants of firms' adoption of non-GAAP measures in comp plans? 2) Do non-GAAP measures or plans heavily tilted toward non-GAAP measures lead to higher levels of future performance and/or improved/more efficient investment decisions? 3) Is there better pay for performance sensitivity after adoption of non-GAAP measures?

We document how voluntary disclosure of management earnings forecasts (MFs) affects the asymmetric timeliness of earnings measured using the Basu (1997) model. The rationale is that MFs affect the extent to which prices lead earnings, which in turn affects the asymmetric timeliness coefficient of the Basu (1997) model. We document the magnitude of the effect by examining MF subsamples formed based on the basis of horizon and news content. We find that for firms issuing MFs of future year earnings (forward-looking MFs), measured asymmetric timeliness is insignificant. In contrast, firms issuing MFs for periods that overlap with the Basu model’s return estimation interval display no evidence of reduced asymmetric timeliness. Further analysis reveals that the weaker asymmetric timeliness of firms issuing forward-looking MFs is concentrated in those that convey more negative news. Asymmetric timeliness is also absent for book-based (i.e., financial statement) measures of asymmetric timeliness, which is consistent with measurement concerns related to the effect of forward-looking MFs on stock returns. Additional tests reveal that the effect of forward-looking MFs on asymmetric timeliness is correlated with the fundamental determinants of conservatism. Our results document how managers’ MF disclosure decisions affect measured asymmetric timeliness separately and distinctly from managerial actions that influence the recognition of news in earnings via the effects of the conservatism imbedded in accounting rules.

We examine voluntary disclosure by IPO firms in the one-year period prior to filing a prospectus with the Securities and Exchange Commission. Specifically, we consider the determinants and consequences of pre-prospectus press releases and attendance at broker-hosted conferences by the IPO firm. These voluntary disclosure activities are undertaken when the firm is private. In contrast, prior research focuses almost entirely on voluntary disclosure activities undertaken by public firms.