Essays on the Usefulness of Non-GAAP Earnings
Author | : Felix Thielemann |
Publisher | : |
Total Pages | : 0 |
Release | : 2019 |
Genre | : |
ISBN | : |
Download Essays on the Usefulness of Non-GAAP Earnings Book in PDF, ePub and Kindle
Voluntary disclosure of adjusted earnings metrics; i.e., so-called non-GAAP earnings, is subject to ongoing controversy. In fact, critics allege that management uses these earnings metrics to portray an overly optimistic view of company performance whereas proponents argue that, relative to GAAP earnings, they are more indicative of recurring and/or operating performance. Hence, the usefulness of these earnings measures is ultimately an empirical question. Against this background, the three essays of this dissertation project explore the usefulness of a) management-provided non-GAAP earnings disclosure (Essays I & II) and b) Standard & Poor's (S & P) so-called Core Earnings metric, as a similarly adjusted but more credible, yet also standardised non-GAAP earnings measure (Essay III). In particular, Essays I & II offer a new perspective on Regulation G (RegG), which the Securities and Exchange Commission (SEC) introduced in 2003 to protect investors from the potentially misleading character of non-GAAP disclosures. While Essay I provides evidence supportive of the regulation's benefit, Essay II documents that it also enabled new opportunistic behaviour as an unintended consequence. Specifically, Essay I extends prior non-GAAP literature's exclusive focus on the equity markets by showing that the regulation alleviated the credibility problem of non-GAAP earnings to the point that bond investors incorporate them into their credit risk assessment. In contrast, Essay II explores the proliferation and motives underlying a self-devised strategy of regulatory avoidance, thereby contributing to the nascent literature on post-regulation opportunism and unintended consequences. Finally, Essay III compares the ability of S & P's Core Earnings metric to predict future operating cash flow against that of GAAP earnings. An in this setting novel out-of-sample estimation approach is applied, which yields no significant difference in predictive abil.