Elsevier

Advances in Accounting

Volume 46, September 2019, 100432
Advances in Accounting

Non-GAAP earnings disclosures around regulation G – The case of “implicit non-GAAP reporting”

https://doi.org/10.1016/j.adiac.2019.100432Get rights and content

ABSTRACT

We identify a phenomenon related to non-GAAP earnings disclosure and examine its prevalence around Regulation G (RegG). Specifically, we analyze to what extent firms only disclose adjustments to GAAP earnings instead of entire adjusted earnings figures thereby not providing the ideal non-GAAP to GAAP reconciliation promoted by RegG. We refer to this reporting behavior as “implicit non-GAAP reporting” and ask three different questions: How is “implicit non-GAAP reporting” related with the adoption of RegG? What type of firm reports implicit non-GAAP measures? What are the motives for “implicit non-GAAP reporting” post-RegG? Our analyses yield three key findings. First, the frequency of “implicit non-GAAP reporting” spikes after the regulatory intervention but to a lesser degree also existed before. Second, during the post-RegG time period, the prevalence of “implicit non-GAAP reporting” is much higher among firms who only started to report non-GAAP earnings after RegG was enacted (starters) than among those, which continued to disclose non-GAAP earnings across the regulatory intervention (continuers). Third, we show that only for starters, “implicit non-GAAP reporting” is associated with motives of beating analyst earnings forecasts as well as experiencing GAAP losses. Our study provides important insights for regulators, firms and academics into "implicit non-GAAP reporting" by examining properties and determinants of implicit vs. explicit non-GAAP earnings for different types of firms around RegG.

Introduction

The Securities and Exchange Comission (SEC) has specifically regulated the voluntary disclosure of so-called non-GAAP financial measures1 through the enactment of Regulation G (thereafter RegG) in 2003. While RegG still allows management to report non-GAAP financial measures,2 it strives to promote transparency. Specifically, it does so by mandating a detailed, quantitative and tabular non-GAAP to GAAP reconciliation whenever firms report non-GAAP earnings. Experimental research (Elliott, 2006) as well as archival research (Aubert & Grudnitski, 2014; Malone, Tarca, & Wee, 2016 and Zhang & Zheng, 2011) has shown that bespoke reconciliations are useful to investors and analysts alike.3 In particular, they may serve to identify managers that deliberately tailor their non-GAAP earnings to beat GAAP results (e.g. Doyle, Jennings, & Soliman, 2013), for instance with the ultimate goal of boosting share price. In this paper, we identify a phenomenon that enables firms to report non-GAAP earnings without providing an explicit non-GAAP EPS figure and by that also no reconciliation from non-GAAP to GAAP information. We refer to this way of disclosing non-GAAP information as “implicit non-GAAP reporting”.

Specifically, we infer “implicit non-GAAP reporting” if firms only disclose adjustments to GAAP earnings, so called non-GAAP adjustments, instead of entire adjusted earnings figures and by that do not apply the non-GAAP to GAAP reconciliation required by RegG. For instance, in its fourth quarter 2004 Kraft Foods Inc. reports implicit non-GAAP EPS of $0.49 (Fig. 1).

In fact, the firm clearly suggests that investors view the charge of $0.12 separately from its result of $0.37 and effectively excludes it to arrive at bespoke implicit non-GAAP EPS of $0.49. In addition, Kraft Foods does not provide any further reconciliation or disaggregation of the adjustments. Hence, investors are e.g., unclear about the type and quantity of impairment as opposed to restructuring. Further, this reporting behavior is in contrast to the firm's previous assurance that it would stop presenting non-GAAP financial results (Fig. 2):

Therefore, it seems that some firms use “implicit non-GAAP reporting” to disclose non-GAAP earnings as opposed to explicitly stating the latter including the reconciliation required by RegG. Consistent with this idea, we explore three different research questions regarding “implicit non-GAAP disclosure”: How is “implicit non-GAAP reporting” related with the adoption of RegG? What type of firm reports implicit non-GAAP measures? What are the motives for “implicit non-GAAP reporting” post-RegG? We structure our analyses around these questions. However, given that there is no priors on “implicit non-GAAP reporting”, we abstain from formulating specific ex ante expectations i.e., the paper is explorative in nature.4

We are specifically interested in the role of “implicit non-GAAP reporting” in view of a regulatory change and therefore set our sample period around the enactment date of RegG in March 2003. Specifically, we hand-collect non-GAAP data including information on “implicit non-GAAP reporting” from earnings press releases using a sub-sample of 229 S&P 500 firms for the 1999 to 2005 time-period. With respect to our first question, we observe the evolution of “implicit non-GAAP reporting” before and after the adoption of RegG. We find that “implicit non-GAAP reporting” spikes after the adoption of RegG. Further, compared to the pre-RegG time-period, the share of implicitly reported non-GAAP earnings remains at significantly higher levels after the regulatory intervention.

Because non-GAAP reporting remains a form of voluntary disclosure, firms' decision to self-select into the group of non-GAAP reporters may simultaneously drive their decision to report non-GAAP earnings implicitly. Therefore, we apply a methodology similar to Black, Black, Christensen, and Heninger (2012) and Black, Christensen, Kiosse, and Steffen (2017) to assign firms into those that only reported non-GAAP earnings before the regulation was enacted (stoppers), those that started to report only after the regulatory intervention (starters), those that continued throughout both time-periods (continuers) and those that never reported (non-reporters).5 Ultimately, we focus on 637 firm-quarters that pertain to starters and continuers after the regulatory intervention when “implicit non-GAAP reporting” in view of the reconciliation requirement by RegG is particularly interesting. Consistent with prior literature arguing that RegG might have acted as an unwanted safe harbour i.e., as a catalyst rather than a deterrent to non-GAAP disclosures (e.g. Black et al., 2017; Cazier, Christensen, Merkley, & Treu, 2017), we reason that starters are less inclined than continuers to report non-GAAP earnings implicitly.6 However, interestingly and contrary to our expectation, we find that the share of “implicit non-GAAP reporting” among starters is much higher than among continuers.

Next, we proceed to disentangle the motives that underlie starters' and continuers' differential propensity to disclose non-GAAP earnings implicitly. Specifically, we investigate to what extent starters' and continuers' choice to report non-GAAP earnings implicitly is associated with a) beating analyst earnings forecasts on a non-GAAP basis and b) the experience of concurrent GAAP losses. We find that particularly for starters there is a significant and positive association between the decision to engage in “implicit non-GAAP reporting” and the beating of analyst earnings forecasts. The results also hold in the presence of concurrent GAAP losses but are not as strong. This suggests that for the group of starters "implicit non-GAAP reporting" may be differently motivated compared to continuers.

Our results contribute to the emerging literature on unintended consequences of RegG (Baumker, Biggs, McVay, & Pierce, 2013; Black et al., 2017; Heflin & Hsu, 2008; Kolev, Marquardt, & McVay, 2008). However, unlike prior literature, our novel and self-devised concept of “implicit non-GAAP reporting” is related to an explicitly regulated aspect of RegG i.e., its reconciliation requirement. In contrast, prior literature examines non-GAAP reporting practices that at least at the time of the respective observation period, were not explicitly regulated. For instance, this applies to the idea that excluding recurring items constitutes opportunistic behavior which Black et al. (2017) strongly draw on in their study covering the period 2002–2013. However, the SEC confirmed this view only in 2010 when it published Compliance & Disclosure Interpretations (C&DIs). Hence, in absence of any regulatory guidance, managers would have had to judge by themselves. Undeniably, this is a difficult process, which could certainly have yielded different results across firms. In fact, the recent call by Black and Christensen (2018) for the SEC to change its stance on the exclusion of recurring expenses illustrates the difficulty and judgement involved in this issue.

In practical terms our results may encourage regulators in jurisdictions, which only recently adopted comparable non-GAAP regulation, to stay vigilant and potentially tighten enforcement. In particular, this applies to the European Securities and Markets Authority (ESMA) with its recently adopted “Guidelines on Alternative Performance Measures” effective as of July 2016. Contentwise and in particular with respect to the reconciliation requirement, ESMA guidelines resemble the U.S. rules. Yet, while the U.S. regulation is legally binding, ESMA guidelines are not binding per se but de facto binding as national financial market authorities are committed to incorporate them into their supervisory practice (European Parliament, 2010; ESMA, 2015a; ESMA, 2015b).

This paper proceeds as follows: Section 2 provides background on the institutional setting as well as details on “implicit non-GAAP reporting”. Section 3 contains related literature while Section 4 introduces our data. Section 5 presents our analyses and Section 6 concludes.

Section snippets

Regulation of non-GAAP measures

Non-GAAP reporting is understood as the reporting of numerical measures of a registrant's future or historical performance, financial position or cash flows that either excludes or includes amounts that are included or excluded in the most directly comparable GAAP measure (SEC, 2002a). Thus, non-GAAP measures are adjusted for certain gains and losses at management's discretion and therefore do not conform to financial accounting standards. In particular, this definition subsumes adjusted net

Related literature and research questions

Since non-GAAP earnings are about the only form of specifically regulated voluntary disclosure, the adoption of RegG has provided and continues to provide a fruitful ground for research. However, often studies either document consequences consistent with the intentions of the regulatory intervention (e.g. Entwistle et al., 2006; Marques, 2006) or focus on opportunistic non-GAAP disclosures only during the post-RegG time-period (e.g. Guest, Kothari, & Pozen, 2018; Shiah-Hou & Teng, 2016). In

Data

Our observation period spans around the enactment date of RegG in March 2003 ranging from 1999 to 2005. We begin our sample selection procedure by identifying all S&P 500 firms as of December 1998 via Compustat's “Index Constituents” function. Next, we remove 165 firms (4620 firm-quarters), which did not form part of the S&P 500 during the whole sample period to arrive at a reduced sample of 337 firms.15

Analysis

Our analyses on “implicit non-GAAP reporting” are sub-divided into three different sections. First, we provide descriptive evidence on the evolution of “implicit non-GAAP reporting” across the regulatory intervention in 2003 (Section 5.1). Then, we proceed to focus on the post-RegG period to analyze whether “implicit non-GAAP reporting” changed after the regulatory intervention requiring a reconciliation from non-GAAP to GAAP. Specifically, we analyze whether starters and continuers exhibit

Conclusion

In this paper, we explore the use and motives of a non-GAAP phenomenon, which we refer to as “implicit non-GAAP reporting”. Specifically, “implicit non-GAAP reporting” enables firms to report non-GAAP earnings without reporting the reconciliation from non-GAAP to GAAP in the form required by RegG. We find that a) many but not all managers seem to apply “implicit non-GAAP reporting”, b) it is more prevalent among starter than continuer firms and c) only for starters it is associated with

References (40)

  • D.E. Black et al.

    US managers' use of ‘pro forma'adjustments to meet strategic earnings targets

    Journal of Business Finance & Accounting

    (2009)
  • D.E. Black et al.

    Non-GAAP reporting: Evidence from academia and current practice

    Journal of Business Finance & Accounting

    (2018)
  • E.L. Black et al.

    Has the regulation of non-GAAP disclosures influenced managers' use of aggressive earnings exclusions?

    Journal of Accounting, Auditing & Finance

    (2017)
  • N.C. Brown et al.

    Investor sentiment and pro forma earnings disclosures

    Journal of Accounting Research

    (2012)
  • R.A. Cazier et al.

    Litigation risk and non-GAAP reporting

  • A.B. Curtis et al.

    The disclosure of non-GAAP earnings information in the presence of transitory gains

    The Accounting Review

    (2013)
  • W.B. Elliott

    Are investors influenced by pro forma emphasis and reconciliations in earnings announcements?

    The Accounting Review

    (2006)
  • G.M. Entwistle et al.

    Financial reporting regulation and the reporting of pro forma earnings

    Accounting Horizons

    (2006)
  • Ernst et al.

    SEC reporting update: 2017 trends in SEC comment letters

  • ESMA

    ESMA guidelines on alternative performance measures

    (2015)
  • Cited by (0)

    View full text