42
* Lancaster University Management School, Lancaster, LA1 4YX; Tel ++44 (0) 1524 594242; Email [email protected]. This paper has been prepared for the ICAEW Information for Better Markets conference held on December 16-17, 2013. I am grateful for comments and suggestions from conference organisers. The Drivers, Consequences and Policy Implications of Non-GAAP Earnings Reporting Steven Young This version: November 2013

Steven Young Icaew Ifbm2013hgjk Young Pro Forma v1

  • Upload
    jmams

  • View
    15

  • Download
    0

Embed Size (px)

DESCRIPTION

giufiuv

Citation preview

* Lancaster University Management School, Lancaster, LA1 4YX; Tel ++44 (0) 1524 594242; Email

[email protected]. This paper has been prepared for the ICAEW Information for Better Markets conference

held on December 16-17, 2013. I am grateful for comments and suggestions from conference organisers.

The Drivers, Consequences and Policy Implications of Non-GAAP Earnings

Reporting

Steven Young

This version: November 2013

The Drivers, Consequences and Policy Implications of Non-GAAP Earnings

Reporting

Abstract

Non-GAAP (pro forma) earnings form an increasingly important part of firms’ performance

reporting narrative. This paper reviews the academic and professional debate surrounding non-

GAAP earnings reporting by management. I argue that the demand for customized performance

reporting is a natural response to constraints imposed by a one-size-fits-all reporting system and

that the non-GAAP phenomenon forms part of a long-standing debate over the definition and

presentation of periodic performance. A review of extant research suggests non-GAAP

disclosures are driven by informative reporting and opportunistic motives. Opaque presentation

of non-GAAP earnings is associated with earnings mispricing, particularly among

unsophisticated investors groups. Regulations and governance systems designed to ensure

transparency are associated with higher quality disclosures and less mispricing. While

customized reporting behaviour is evident in many settings, I argue that such disclosures create

particular risks in a financial reporting context because they threaten the integrity of the

underlying reporting system. Prevailing regulatory approaches are reviewed and factors limiting

disclosure transparency are highlighted. The paper concludes with suggestions for further

research.

Keywords: pro forma earnings; street earnings; adjusted EPS; transitory items; Regulation G

1

The Drivers, Consequences and Policy Implications of Non-GAAP Earnings

Reporting

1. Introduction

Generally Accepted Accounting Principles (GAAP) govern how accounting constructs

such as periodic earnings are defined. GAAP ensure a level of reliability and consistency across

firms and time with respect to financial reporting outputs. However, by limiting reporting

practices, GAAP impose a degree of uniformity on entities whose operations and policies are

characterized by substantial heterogeneity. Where earnings fail to capture important firm-specific

aspects of performance, demand arises for customized measures of periodic performance. Non-

GAAP earnings (also known variously as pro forma earnings, street earnings and adjusted

earnings) that exclude non-cash and transitory items from the GAAP earnings number represent

a market response to this demand.

While the majority of extant research on non-GAAP reporting focuses on the practices of

US firms from the late 1990’s onwards, the phenomenon is neither a recent nor confined to the

US. The practice of firms disclosing adjusted GAAP numbers in their earnings announcements is

long established. The Securities and Exchange Commission (SEC) issued Accounting Series

Release No. 142 in 1973 warning of possible investor confusion from the use of financial

measures outside of GAAP:

[T]he unilateral development and presentation on an unaudited basis of various measures

of performance by different companies which constitute departures from the generally

understood accounting model has led to conflicting results and confusion for investors.

Additionally, it is not clear that simple omission of depreciation and other non-cash

2

charges deducted in the computation of net income provides an appropriate alternative

measure of performance for any industry either in theory or in practice...

In addition to the proliferation of firm-specific adjusted metrics, sector-specific non-GAAP

earnings numbers have also evolved as part of firms’ normal reporting practices (Gore and Scott

1998, Vincent et al. 1999, Serafeim 2011). Outside the US, non-GAAP earnings have been

studied in Australia (Cameron et al. 2012), New Zealand (Rainsbury et al. 2012), France (Aubert

2009), Germany (Hitz 2010) and the UK (Walker and Louvari 2003, Choi et al. 2007, Choi and

Young 2013). Isidro and Marques (2013) demonstrate a majority of firms in all major European

countries report at least one non-GAAP metric in their earnings releases.

Neither is the practice of excluding items from GAAP earnings an exclusively supply-

driven phenomenon. Widespread external demand for adjusted GAAP numbers also exists.

Equity analysts rely on street earnings metrics that adjust GAAP earnings for transitory items

(Bradshaw and Sloan 2002). Regulators frequently use GAAP-adjusted numbers to monitor

firms in the financial and utilities sectors. Value-based management consultants promote

customized earnings metrics (e.g., Boston Consulting Group’s Cash Flow Return on Investment

and Stern Stewart’s Economic Value Added©

), as do lenders when specifying contractual terms

in debt covenants and credit rating agencies when assessing solvency and liquidity. Commercial

databases including Compustat and Thomson Datastream also provide adjusted earnings metrics

(Brown and Sivakumar 2003, Choi et al. 2007), while investment associations such as the

Institute of Investment Management and Research have proposed their own measures of

permanent earnings (Lin and Walker 2000).

This paper explores the causes, consequences and policy implications of non-GAAP

earnings disclosures reported by management. Hereinafter I use the term non-GAAP earnings to

3

refer exclusively to management-reported adjusted earnings; I restrict use of the term street

earnings to analyst-reported earnings and I do not use the term pro forma earnings to avoid

confusion with pro forma statements required by securities regulators in the case of mergers, and

in recognition that outside the US the term pro forma earnings tends to be less widely employed,

particularly by regulators such as the International Accounting Standards Board (IASB).1

Depending on the prevailing regulatory environment, management may report non-

GAAP earnings information through a variety of channels including earnings press releases,

conference calls, the management discussion and analysis section of 10-K and 10-Q filings in the

US, and as part of published financial statements for firms reporting under International

Financial Reporting Standards (IFRS). Insofar as the properties, drivers and consequences of

such disclosures vary across disclosure platforms, care is required when comparing results from

different studies. I use the label non-GAAP earnings as an umbrella term for any form of GAAP-

adjusted earnings number reported by management.

The remainder of this paper comprises five sections. In the next section I provide an

overview of the non-GAAP reporting phenomenon including the arguments supporting

customized earnings reporting and the moral hazard problems of allowing management freedom

to determine exclusions, the broader performance reporting context in which the non-GAAP

earnings debate is located, and the regulatory landscape governing reporting practices. Section 3

1 Many studies use the terms pro forma earnings, non-GAAP earnings, and street earnings interchangeably when

referring to adjusted earnings metrics produced by management or reported by commercial analyst tracking services

such as IBES, First Call and Zacks. Research demonstrates a high degree of consistency between management-

produced non-GAAP earnings and analyst-produced street earnings metrics (Bhattacharya et al. 2003, Choi et al.

2007) and consequently studies have used street earnings to proxy for management-reported non-GAAP earnings

(e.g., Doyle et al. 2003, Frankel et al. 2010). Several factors make street earnings a noisy measure of management-

disclosed non-GAAP earnings. First, street earnings availability is independent of managements’ decision to

disclose non-GAAP earnings. Second, conditional on management reporting non-GAAP earnings, street earnings

adjustments differ from managements’ exclusions in approximately 35 percent of cases (Bhattacharya et al. 2003,

Choi et al. 2007). I focus specifically on understanding non-GAAP earnings disclosed by management, although

papers employing street earnings constructs are discussed where appropriate.

4

summarizes the properties of non-GAAP earnings and reviews evidence on management’s

reporting motives. Extant work presents substantial evidence supporting both opportunism and

informative reporting. Although helpful, results are limited insofar as they fail to provide

unambiguous insights regarding underlying disclosure motives. Section 4 examines the impact of

non-GAAP earnings disclosures on market pricing and investor behaviour, and reviews the

evidence on regulatory developments aimed at minimizing the risk of market mispricing.

Findings demonstrate that non-GAAP disclosures can lead to mispricing and that such effects are

driven largely by unsophisticated investors. Regulations aimed at improving disclosure

transparency reduce mispricing.

Section 5 considers the degree to which adjusted performance reporting is a problem

unique to financial reporting. The tendency to report customized metrics is common practice in

many non-accounting settings. A distinctive feature of non-GAAP earnings, however, is that the

adjustment process threatens the credibility and integrity of the underlying reporting system.

Minimizing this risk involves ensuring non-GAAP earnings are presented transparently and in a

way that encourages management to communicate their private information on key income

streams while alerting users to potential abuse. A series of factors that limit transparency under

current reporting regulations are discussed. Section 6 presents a non-exhaustive list of topics for

further research and section 7 concludes.

2. Background and context

2.1 Overview and definitions

Non-GAAP earnings create a headache for financial statement users and regulators

because management adjustments may reflect different underlying motives that are hard for

external parties to disentangle. On the one hand, a range of arguments highlight the potential

5

benefits of non-GAAP earnings reporting by management. First, separating out different income

streams can enhance reporting transparency (Bhattacharya et al. 2004). Former IASB Chairman

Sir David Tweedie has highlighted both the limitations associated with defining a single, bottom-

line measure of net income and the benefits of allowing management freedom to report

customized measures of periodic performance (Tweedie 1993); former SEC Chairman Harvey

Pitt suggested non-GAAP earnings represent a response to the growing complexity that makes it

more difficult for users to understand GAAP-based financial information (Pitt 2001); even

Warren Buffet has emphasised how non-GAAP earnings can improve transparency (Larcker and

Tayan 2010). Second, by excluding unusual events and arbitrary accounting adjustments from

GAAP earnings, non-GAAP disclosures enable cleaner cross-sectional and time series

comparisons of the core, controllable elements of periodic performance (Bray 2001; Halsey and

Soybel 2002).2 Third, accounting is inherently conservative and while timely recognition of

losses relative to gains is valuable in certain settings such as contracting, the bias and volatility

caused by accelerated loss recognition may render GAAP earnings less useful for valuation and

prediction (Black 1993). Non-GAAP reporting may therefore represent an ad hoc means of

reversing the asymmetry in earnings recognition and reducing earnings volatility. Fourth,

professional investors typically use GAAP-adjusted earnings without any reconciliation to

published earnings. Insofar as adjusted earnings form part of the investment vernacular, non-

GAAP disclosures may help reconcile alternative earnings constructs and enhance consistency in

the financial communication process.

On the other hand, allowing management scope to modify GAAP earnings introduces

moral hazard risks that range from mild impression management designed to portray periodic

2 Responding to criticism of non-GAAP earnings Groupon CFO Jason Child argued such disclosures supplement

GAAP metrics and help investors better evaluate company performance

(http://www3.cfo.com/article/2012/2/accounting-tax_groupon-cfo-jason-child-defends-non-gaap-metrics).

6

performance in a favourable light to outright obfuscation aimed at misleading investors.3 At the

heart of this view is the assumption that management prefer higher earnings because (they

believe) investors extrapolate better current-period performance into the future, leading to higher

stock valuations and a more valuable currency for raising capital and funding acquisitions

(D’Avolio et al. 2002: 144). Research suggests management seek to present results in a

favourable light generally (Schrand and Walther 2000) and use non-GAAP earnings in ways

consistent with this end (Bowen et al., 2005). The dilemma for investors and regulators is how to

give management freedom to report non-GAAP earnings capable of communicating their private

information regarding key earnings components while simultaneously limiting management’s

ability to employ such disclosures opportunistically. As such, non-GAAP reporting embodies the

perennial trade-off between relevance and reliability.

2.2 Part of a broader debate: Defining periodic performance

Non-GAAP earnings is part of the broader debate about the definition and presentation of

periodic financial performance. The question of how to define earnings and in particular whether

or not reported performance should include the effects of all periodic transactions and value

changes has vexed standard setters for decades. Two polar views exist. Some groups favour an

all-inclusive approach whereby net income reflects the effect of all periodic changes in

shareholders’ equity (other than transactions with owners). Others favour the current operating

concept, arguing that earnings should not be distorted by abnormal, non-recurring events and

transactions that confound the ability of current earnings to predict sustainable performance.

Relative support for each concept has varied across time and reporting regimes. The Financial

3 The SEC had brought two enforcement actions relating to non-GAAP reporting. The first was against Trump

Hotels and Casino Resorts, Inc in 2002; the second was against SafeNet, Inc in 2009 under Regulation G.

7

Accounting Standards Board (FASB) in the US has long promoted an all-inclusive approach. In

contrast, many non-US jurisdictions traditionally emphasized the current operating performance

concept. This approach, for example, underpinned UK financial reporting until the introduction

of Financial Reporting Standard 3 (FRS 3) in June 1993. Prior to FRS 3, certain transactions

deemed extraordinary in nature were disclosed separately and excluded from bottom-line

earnings.4 The approach led to a proliferation of items classified as extraordinary, the majority of

which were periodic losses. Moreover, subjectivity inherent in classifying items as extraordinary

led to inconsistent reporting and opportunism (Smith 1992, Beattie et al. 1994). Similar

opportunistic classification behaviour has been documented in the US in relation to special items

(McVay 2006) and in the UK post-FRS 3 for exceptional items (Athanaskou et al. 2007).

Non-GAAP earnings can be viewed as an extreme manifestation of the current operating

concept of income reporting applied in a voluntary disclosure setting (extreme insofar as

management are free to exclude any GAAP earnings component whereas under a regulated

application of the current operating performance concept the set of excludable items is typically

more tightly defined). Accordingly, the same arguments regarding lack of consistency and

opportunistic classification of excluded items that critics level at the current operating earnings

concept accompany non-GAAP earnings disclosures. The reemergence of old debates in the

guise of non-GAAP earnings highlights the inherent limitations of a one-size-fits-all approach to

earnings reporting. Valuation theory further emphasizes the problem of determining a single,

best measure of periodic performance: given clean surplus accounting, all earnings definitions

are equally valid from a valuation standpoint (Peasnell 1982, Ohlson 1995). The implications for

performance reporting in general and non-GAAP earnings in particular are profound. First,

4 The notion of extraordinary items was introduced by Statement of Standard Accounting Practice 6 (SSAP 6)

introduced in 1974 and modified by SSAP 6 (Revised) in 1986. Before SSAP 6, non-recurring transactions were

frequently excluded from periodic income and accounted for as movements on reserves.

8

customized earnings reporting by management in one form or another seems inevitable

regardless of the particular earnings construct accounting standard setters choose to adopt.

Perhaps this explains why regulators have avoided calls to outlaw non-GAAP earnings reporting

(e.g., D’Avolio et al. 2002) despite genuine concern about the scope for opportunism and

obfuscation. Second, since non-GAAP earnings violates clean surplus, transparency lies at the

heart of the non-GAAP reporting debate (Cornell and Landsman 2003). Third, accounting

standard setters must decide whether to embrace non-GAAP earnings as part of the financial

statements or to ring-fence GAAP earnings and relegate customized reporting to other

communication channels (and hence other regulators). I return to these issues in section 5.

2.3 Regulatory landscape

Non-GAAP earnings disclosure is subject to regulation that varies by country and

reporting channel. US securities’ rules prevent registrants from presenting non-GAAP metrics as

part of their audited financial statements. Non-GAAP earnings reported outside the financial

statements are governed by elements of the Securities and Exchange Act (effective March 2003)

including Regulation G, item 10(e) of Regulation S-K of the, and item 12 in Form 8-K.

Regulation G (Reg G) covers all public disclosures of non-GAAP financial measures including

press releases, conference calls, investor presentations, and other media. The regulation requires

non-GAAP earnings to be accompanied by the most directly comparable financial measure

calculated and presented in accordance with GAAP, together with a reconciliation of the non-

GAAP metric to the corresponding GAAP measure. Item 10(e) covers all SEC filings including

10-Ks, 10-Qs, and proxy statements, and requires non-GAAP disclosers: present a comparable

GAAP number with equal prominence; reconcile differences between non-GAAP and the

comparable GAAP earnings numbers; report the reason(s) why non-GAAP earnings is believed

9

to provide useful information; and disclose the purposes (if any) for which management use non-

GAAP earnings. Item 12 requires firms file Form 8-K within five business days of any public

disclosure of annual or quarterly operating results, and where the public disclosure contains a

non-GAAP financial measure management must provide the same supplementary disclosures

required by Reg G.

International rules governing non-GAAP earnings offer greater reporting freedom in at

least two respects. First, IFRS institutionalize non-GAAP earnings reporting as part of the

audited financial statements. Specifically, International Accounting Standard 33 (IAS 33)

permits management to report non-GAAP earning per share (EPS) metrics on the face of the

income statement or in the accompanying notes, as long as basic and diluted amounts per share

relating to any such metric(s) are disclosed with equal prominence along with a reconciliation to

a corresponding line item reported in the income statement.5 Second, international securities

regulations do not typically place restrictions on non-GAAP disclosures presented in

communications with investors.6 Moreover, as the scope of IAS 33 does not extend beyond the

audited section of firms’ annual report and accounts, management are able to discuss non-GAAP

earnings in unaudited report narratives without the need for accompanying definitions,

reconciliations, or explanations.

Cross-country differences in non-GAAP earnings regulation raise important policy

questions including whether disclosing non-GAAP earnings in audited financial statements

affects reporting behaviour and investor responses, and what combination of rules governing

5 IAS 33 borrows heavily from Financial Reporting Standard 3 (FRS 3) in UK GAAP.

6 For example, the Committee of European Securities Regulators issued guidance in 2005 that non-GAAP financial

measures should be disclosed in a way that is appropriate and useful for investor decision making but has stopped

short of issuing Reg G-type rules. While the Autorite des Marches Financiers in France has issued guidelines

requesting a reconciliation between non-GAAP and GAAP earnings, in practice such reconciliations are rare

(Aubert, 2010). In Germany, the securities act prohibits disclosure of misleading information but does not

specifically refer to non-GAAP earnings (Hitz, 2010)

10

financial statement disclosures and other communications yields the most reliable and relevant

information. For example, incorporating non-GAAP earnings in the framework of audited

financial statements could improve reporting transparency. Alternatively, allowing management

to report non-GAAP metrics in the financial statements alongside GAAP earnings could increase

the prominence of non-GAAP earnings and legitimise attempts by management to influence

investors’ perceptions of periodic performance (Bowen et al., 2005).

3. Properties and reporting motives

3.1. Descriptive evidence

The economic significance of non-GAAP earnings reporting is determined in part by the

frequency, nature, and magnitude of the components excluded from the GAAP number.

Bhattacharya et al. (2003) and Bhattacharya et al. (2004) provide the first large sample evidence

on the incidence and characteristics of non-GAAP earnings reporting by US firms. Prior

empirical work focused on street earnings provided by analyst tracking services (Bradshaw and

Sloan 2002) and core operating earnings constructs proposed by professional investment bodies

(Lin and Walker 2000) or supplied by commercial databases (Brown and Sivakumar 2003).

Non-GAAP earnings reporting is commonplace. The number of US firms reporting non-

GAAP earnings increased sharply over the period 1998 through 2000, although firms employing

a consistent quarterly reporting policy are rare (Bhattacharya et al. 2004).7 The reporting

frequency declined following implementation of Reg G and item 10(e) (Heflin and Hsu 2008,

7 The sample used by Bhattacharya et al. (2003) and Bhattacharya et al. (2004) comprises 1,149 quarterly non-

GAAP earnings disclosures identified using keywords “pro forma”, “pro-forma”, and “proforma”. Further analysis

leads Bhattacharya et al. (2003: 297) to estimate that their search string captures approximately half of all possible

non-GAAP EPS figures reported during their sample period. Marques (2006) confirms this conjecture. Specifically,

of the 4,234 observations in her sample of S&P 500 firms from 2001 to 2003, 2,475 disclose some type of non-

GAAP financial measure (with the most common being net income, either in its per share or aggregated form).

11

Marques 2006) before rebounding (Doyle et al. 2011). In the UK, the fraction of large non-

financial firms disclosing non-GAAP earnings on the face of the income statement under FRS 3

rose from 40 percent in 1993 to 75 percent by 2001 (Choi et al. 2007). By 2003, Isidro and

Marques (2011) find that approximately 80 percent of large European firms were disclosing at

least one non-GAAP performance metric in their earnings press releases, and following

transition to IFRS in 2005, almost 90 percent of UK firms report a non-GAAP earnings metric in

their financial statements (Petaibanlue et al. 2013).

Bhattacharya et al. (2004) provide evidence on exclusions by US reporters. Depreciation

and amortization was the most frequent adjustment category during the period 1998-2000 (21

percent of all adjustments), followed by stock-based compensation, shares outstanding, merger

and acquisition-related costs, and research and development (R&D) costs and write-offs of

purchased in-process R&D. Other categories with less frequent adjustments include restructuring

charges, tax, interest, gains and losses on asset sales, and stock-related charges such as

preference stock conversion costs and IPO expenses. The largest adjustments by magnitude

(relative to revenue) were associated with stock-related charges such as preferred stock

conversion charges and IPO expenses.

The majority of items excluded by US reporters are expenses that decrease GAAP

earnings. Of the 12 categories examined by Bhattacharya et al. (2003), only gains and losses on

asset dispositions are GAAP income-increasing on average. While many exclusions are likely to

be nonrecurring in nature, other items such as depreciation and amortization, stock-based

compensation, and R&D costs are more persistent. Management typically justify such exclusions

on the grounds they are historic cost-based estimates that are not indicative of current and future

12

expenditures or performance.8 Only 10 percent of repeat reporters make consistent adjustments.

Inconsistent reporting patterns are noteworthy but hard to interpret. On the one hand,

nonrecurring earnings components are less likely to generate consistent reporting behaviour by

their very nature. On the other hand, a high degree of inconsistency is suggestive of opportunistic

classification choices. Descriptive evidence on non-GAAP exclusions does not therefore afford

definitive conclusions on whether non-GAAP earnings promote or reduce reporting

comparability.

While international non-GAAP earnings reporting shares many common features with

findings documented for US registrants, notable differences are apparent. For example, whereas

US non-GAAP disclosers cluster hi-tech industries (Lougee and Marquardt 2004, Bhattacharya

et al. 2004), international non-GAAP disclosure appears less sector-specific (Choi et al. 2007,

Isidro and Marques 2013). European firms also adopt more consistent non-GAAP disclosure

policies than their US counterparts (Isidro and Marques 2013), while UK reporters rarely exclude

recurring earnings components such as depreciation and amortization (Choi et al. 2007).

Evidence presented by Isidro and Marques (2013) suggests institutional and economic factors

account for some of these international differences.

3.2 Reporting motives

Absent clear theoretical or practical guidance on managers’ dominant reporting

motives, researchers have sought to test whether non-GAAP disclosures are driven by

opportunism or the desire to provide incremental information on permanent earnings.

8 Christensen et al. (2011: 505) cite Akamai, Inc. (Feb 4, 2009), where management argue that depreciation and

amortization are based on estimates of useful economic lives of tangible and intangible assets, and that these

estimates could vary from actual performance of the asset. Management also argue charges are based on the historic

cost incurred to build up the company’s deployed network, and may not be indicative of current or future capital

expenditures. Whether errors due to prevailing measurement rules exceed the error from assuming such items to be

zero is moot.

13

While robust evidence supports both viewpoints, definitive conclusions regarding the

dominant reporting motive have proved elusive.

Proponents of informative reporting point to a large body of evidence demonstrating non-

GAAP earnings are informative. First, the high degree of overlap between adjustments made by

management and those made by analysts suggests non-GAAP earnings represent a step toward

permanent earnings. Second, non-GAAP earnings are considered more value relevant by

investors than GAAP operating earnings and are better able to predict future performance.

Conclusions hold using both indirect measures based on street earnings (Bradshaw and Sloan

2002, Brown and Sivakumar 2003) and direct tests based on actual non-GAAP earnings

disclosures (Bhattacharya et al. 2003, Choi et al. 2007, Marques 2006).9 Third, management are

more likely to disclose (Lougee and Marquardt 2004) and emphasize (Bowen et al. 2005) non-

GAAP measures when GAAP earnings have low value relevance. Fourth, incremental

adjustments by management over those made by analysts are also value and forecasting relevant

in some jurisdictions (Choi et al. 2007), consistent with such adjustments reflecting

managements’ superior information about the persistence of earnings components. In contrast,

Marques (2006) finds that US investors do not view incremental adjustments beyond IBES as

providing useful information.

Evidence confirming the incremental informativeness of non-GAAP earnings is

consistent with findings for unregulated disclosures in other settings. For example, Vincent

(1999) reports that funds from operations disclosed voluntarily by real estate investment trusts

(REITs) contain incremental information beyond GAAP earnings, while Serafeim (2011) finds

that the information asymmetry component of the bid-ask spread is lower for embedded value

9 Abarbanell and Lehavy (2007) and Cohen et al. (2007) argue that tests using street earnings as a proxy for non-

GAAP disclosures are likely to be biased in favor of concluding higher value relevance for non-GAAP earnings over

its GAAP counterpart.

14

reporters in the life insurance industry.10

Meanwhile, Barton and Waymire (2004) focus on the

unregulated financial reporting environment in the U.S. prior to the 1929 stock market crash and

find that managers respond to investor demand for information with voluntary financial

disclosures that promote investor protection.

Conversely, a large body of evidence also suggests that management report non-GAAP

earnings opportunistically to present a more favourable view of performance. For example, non-

GAAP earnings are typically higher than the corresponding GAAP number (Bhattacharya et al.

2003, Marques 2006, Choi et al. 2007, Isidro and Marques 2013); proxies for management

exclusions predict future performance consistent with non-GAAP disclosures excluding

recurring earnings components (Doyle et al. 2003, Landsman et al. 2007); management are more

likely to report non-GAAP earnings to overturn a GAAP loss, to report positive earnings growth

when on a GAAP basis growth is negative, and to meet or beat the consensus earnings forecast

when the GAAP surprise is otherwise negative (Lougee and Marquardt 2004, Black and

Christensen 2009, Barth et al. 2012, Doyle et al. 2011, Isidro and Marques 2013, Walker and

Louvari 2003); and optimistic non-GAAP disclosures are associated with higher audit fees and

auditor resignations (Chen et al. 2012).11

Studies examining both the prominence (Bowen et al.

2005, Petaibanlue et al. 2013) and ambiguity (Entwistle et al. 2006a) of non-GAAP earnings

disclosures also provide evidence consistent with opportunism. Finally, implementation of Reg

G was associated with a reduction in the incidence of non-GAAP reporting (Entwistle et al.

2006b), a decline in the frequency and magnitude of exclusions (Marques 2006, Heflin and Hsu

10

Serafeim (2011) also finds that information asymmetry is decreasing in the quality and comparability of

embedded value methods, and that the economic effect of embedded value reporting is larger than that from either

IFRS or U.S. GAAP adoption. However, results hold only for firms that certify embedded value calculations by

hiring an outside auditing or consulting firm. 11

Christensen et al. (2011) link non-GAAP earnings with attempts to influence analysts’ earnings forecasts but

remain silent on whether such influence assists or biases analysts’ estimated and investors’ stock valuation.

acer
Highlight

15

2008), an increase in the quality of exclusions (Kolev et al. 2008), a reduction in the probability

that non-GAAP earnings meet or beat forecasts (Heflin and Hsu 2008), an increase in investor

perceptions of non-GAAP earnings credibility (Marques 2006, Black et al. 2012), and a

reduction in ambiguous and potentially misleading non-GAAP disclosures (Entwistle et al.

2006a, 2006b). Collectively, these findings support the joint hypothesis that some non-GAAP

disclosures were motivated by opportunism in the pre-Reg G regime and that the SEC was at

least partially successful in its objective of improving the quality of non-GAAP reporting.

Extant research therefore supports both the informative and strategic reporting

explanations for non-GAAP disclosure. The apparent schizophrenic nature of non-GAAP

earnings has two implications. First, attempts to generalize reporting behaviour or identify a

single dominant explanation for non-GAAP earnings are likely to flounder because informative

reporting and strategic disclosure do not represent mutually exclusive explanations. Instead, both

motives likely co-exist with the particular driver varying across firms and time conditional on

prevailing reporting incentives. This is consistent with Lougee and Marquardt’s (2004) evidence

that disclosure probability is increasing in both the uninformativeness of GAAP earnings and the

presence of a negative GAAP earnings surprise. Second, extant research designs are typically

incapable of discriminating unambiguously between competing reporting incentives. For

example, while a higher likelihood of reporting non-GAAP earnings when GAAP earnings miss

a key benchmark is consistent with opportunism, it does not permit rejection of the informative

reporting hypothesis because disclosures that yield a better signal of permanent earnings could

also lead to benchmark-beating outcomes as a by-product (Black 1993).

The challenge for researchers seeking to better understand the motives driving non-

GAAP disclosures is to design empirical tests capable of disentangling competing reporting

acer
Highlight
acer
Highlight
acer
Highlight
acer
Highlight

16

explanations. Examining the link between non-GAAP disclosure and the presence of transitory

items in GAAP earnings provides a possible way forward (Curtis et al. 2011, Choi and Young

2013). For example, Curtis et al. (2011) study non-GAAP reporting in the presence of transitory

gains, where exclusions reduce reported performance and hence are more likely to be driven by

informative reporting rather than opportunism. Consistent with opportunism driving a significant

fraction of non-GAAP disclosures, 42 percent of firms with transitory gains appear to

strategically omit non-GAAP earnings information in an attempt to report higher performance.

Meanwhile Choi and Young (2013) partition firms on the sign of the GAAP earnings surprise

and examine variation in the strength of the association between non-GAAP disclosure and

transitory items in GAAP earnings. They find the positive relation between disclosure and

transitory items is more pronounced for the positive GAAP surprise partition, which supports the

view that informative (strategic) motives are more pronounced when benchmark beating

incentives are weak (strong).

3.3 The moderating effect of internal governance arrangements

If managerial opportunism drives a non-trivial fraction of non-GAAP earnings

disclosures, the quality of internal governance arrangements designed to maintain financial

reporting quality is expected to impact non-GAAP reporting behaviour.12

Research documents a

positive link between financial reporting quality and governance quality (Bushman and Smith

2001). Building on this theme, a growing body of work conditions non-GAAP disclosure on

internal governance arrangements designed to curb managerial opportunism including

12

Isidro and Marques (2013) demonstrate that country-level institutional and economic factors also affect non-

GAAP reporting incentives. In particular, they document a stronger link between non-GAAP earnings to achieve

earnings benchmarks in environments characterized by greater pressure to achieve earnings benchmarks and less

opportunity to manipulate GAAP earnings.

17

independent boards (Frankel et al. 2010, Isidro and Marques, 2011, Jennings and Marques 2011,

Entwistle et al. 2012), long-term compensation arrangements (Black et al. 2011), and auditor

oversight (Entwistle et al. 2012, Black et al. 2011, Chen et al. 2012).

Frankel et al. (2010) predict and find that firms with less independent boards are more

likely to opportunistically exclude recurring items from non-GAAP earnings, while Black et al.

(2011) report that compensation contracts and auditor effort deter managers from using non-

GAAP disclosures aggressively. Investors also view better governed firms as providing more

credible non-GAAP exclusions (Entwistle et al. 2012). In contrast, Isidro and Marques (2011)

report mixed evidence on the constraining role of boards using an international sample: while

good boards reduce the probability of non-GAAP disclosure and limit impression management,

they do not moderate the positive association between compensation-related incentives and low

quality non-GAAP reporting. Meanwhile, Kyung et al. (2013) report how steps to improve

governance quality through voluntary adoption of executive compensation clawback provisions

reduce non-GAAP reporting quality as the costs of misstating GAAP earnings increase.

Finally, Frankel et al. (2010) and Jennings and Marques (2011) examine how firm-level

governance arrangements interact with regulatory oversight in the form of Reg G to determine

non-GAAP disclosure outcomes. Both studies report evidence consistent with a positive

(insignificant) association between governance quality and non-GAAP disclosure quality in the

pre- (post-) Reg G regime. Findings suggest substitution between internal and external

governance arrangements. Crucially, however, while the endogenous nature of internal

governance arrangements allows low quality non-GAAP reporting to persist in equilibrium,

regulatory intervention addresses reporting quality where improvements are most needed.

4. Investor consequences

acer
Highlight

18

Given the competing incentives for non-GAAP earnings reporting and the difficulty

researchers have encountered discriminating between managements’ reporting motives,

investors’ ability to understand fully the implications of GAAP exclusions for firm performance

and value is an open question.13

Further, given heterogeneity in market participants’ level of

financial reporting sophistication, it is possible that specific groups of investors respond

differently to non-GAAP earnings disclosures. Policymakers charged with protecting the

interests of ordinary investors are particularly concerned that non-GAAP earnings may confuse

and mislead this class of investor (Allee et al. 2007). This section examines whether investors

respond to non-GAAP earnings disclosures appropriately, whether pricing behaviour varies with

investor sophistication, and whether reporting constraints help protect investors interests’.

4.1 Are investors fooled by non-GAAP disclosures?

Evidence that managerial opportunism likely drives a fraction of non-GAAP disclosures

begs the question whether investors are able to identify and see through such behaviour. If

markets are efficient then prices reflect all available information and investors are not

systematically fooled by the form in which information is packaged and presented. To the extent

non-GAAP disclosures repackage publicly available information, the scope for mispricing may

be considered limited. Nevertheless, experimental evidence demonstrates that holding

information constant, the way periodic performance is presented can influence investors’

judgements (Hirst and Hopkins 1998, Maines and McDaniel 2000). Behaviourial psychology

theory also suggests cognitive biases may cause investors to be misled by non-GAAP

disclosures, particularly when there is a material impact on the direction of the performance

13

Researchers as a rule do not undertake methodical financial statement analysis aimed at determining whether a

particular non-GAAP exclusion or set of exclusions is appropriate. Instead, they rely on the law of large numbers

coupled with empirical designs that condition on reporting incentives to identify predominant reporting motives. It is

possible that investors operating at the firm level are better able to discriminate between reporting incentives.

19

signal such as when adjustments transform a GAAP loss into a non-GAAP profit. Further,

insofar as non-experts rely more heavily on heuristics, non-professional investors’ judgments

may be particularly susceptible to opportunistic non-GAAP earnings disclosures.

Research demonstrates that investors discount non-GAAP earnings surprises that

overturn a negative GAAP surprise (Bhattacharya et al. 2003), deviate from the street surprise

(Marques 2006), or coincide with higher levels of prior earnings management (Black et al.

2013), suggesting market participants understand managements’ reporting incentives and are

sceptical of non-GAAP disclosures when opportunism is suspected. Discounting by investors is

particularly pronounced when managers make aggressive earnings exclusions in the presence of

safeguards designed to limit opportunistic behaviour (Black et al. 2011).

Nevertheless, investors’ ability to see fully through opportunistic non-GAAP disclosures

may be compromised. Using street earnings to proxy for non-GAAP earnings, Doyle et al.

(2003) find exclusions from GAAP earnings predict future returns. Lougee and Marquardt

(2004) report similar results using a hand-collected sample of non-GAAP disclosures.

Meanwhile, Curtis et al. (2011) present evidence consistent with investors overvaluing GAAP

earnings when transitory gains are opaquely reported in GAAP earnings rather than being

transparently excluded via a non-GAAP earnings disclosure. Jennings and Marques (2011)

conclude that prior to Reg G investors were misled by non-GAAP earnings disclosures made by

firms with weaker corporate governance. Collectively, these findings suggest some investors

may be confused or misled by non-GAAP earnings.

If non-experts lack the necessary sophistication and experience to fully understand the

precision and reliability of their information set and are more reliant on heuristics, then non-

professional investors’ judgments may be particularly susceptible to mispricing. Research

20

supports this view. Experimental and archival studies reveal income-increasing non-GAAP

adjustments affect less-sophisticated investors’ judgments of earnings announcements

(Frederickson and Miller 2004, Allee et al. 2007, Black et al. 2013), and that this effect is

increasing in the prominence with which the non-GAAP number is presented relative to GAAP

earnings (Elliot 2006, Allee et al. 2007). Professional investors, on the other hand, appear less

susceptible to such effects. Frederickson and Miller (2004) report experimental evidence that

professional analysts do not fixate on favourable non-GAAP earnings. Similarly, Allee et al.

(2007) and Black et al. (2013) find that sophisticated investors’ trading responses to earnings

announcements are unrelated to the non-GAAP earnings forecast errors, while Christensen et al.

(2013) demonstrate that short-sellers actively exploit overvaluation caused by non-professional

investors’ failure to fully understand the implications of recurring exclusions for future

performance. In contrast, Andersson and Hellman (2007) study Swedish equity analysts’

investment decisions in an experimental setting and find evidence of fixation on non-GAAP

earnings when forecasting next period EPS.

4.2 Do reporting constraints protect investors’ interests?

Theory and evidence highlight disclosure transparency as an important factor

conditioning the degree of misunderstanding, with higher transparency moderating the risk of

investors ignoring or fixating on particular information signals. For instance, experimental

evidence reveals that less sophisticated investors’ tendency to fixate on prominently reported,

performance-enhancing non-GAAP earnings information is moderated by a quantitative

reconciliation to the corresponding GAAP number (Elliot 2006, Dilla et al. 2010). Interestingly,

Elliot (2006) finds that professional investors attribute higher valuations to non-GAAP earnings

21

accompanied by a reconciliation relative to either non-reconciled non-GAAP or GAAP-only

disclosures, possibly because they view greater transparency as a signal of higher reliability.

The expected benefits to transparency underpinned the introduction in Reg G, item 10(e)

and item 12 by the SEC for US registrants. Archival empirical evidence examining the impact of

the SEC’s intervention is broadly consistent with the benefits suggested by experimental studies.

In particular, Zhang and Zheng (2011) study mispricing in the pre- and post-Reg G regimes and

provide three pieces of evidence in support of the regulatory intervention. First, mispricing in the

pre-Reg G period was confined to firms with low reconciliation quality. Second, no evidence of

mispricing is apparent following implementation of Reg G. Third, mispricing declined for firms

where reconciliation quality improved in response to Reg G, whereas firms that consistently

provided transparent reconciliations were not associated with mispricing in either regime.

Jennings and Marques (2011) also conclude that Reg G eliminated mispricing associated with

non-GAAP earnings disclosures. More generally, research demonstrates that Reg G led to an

increase in non-GAAP earnings credibility (Marques 2006) and a reduction in the opportunistic

use of such disclosures aimed at presenting a favorable view of performance (Entwistle et al.

2006a and 2006b, Heflin and Hsu 2008, Kolev et al. 2008).14

Nevertheless, Heflin et al. (2008) and Kolev et al. (2008) also highlight evidence of

unintended regulatory consequences in the form of a reduction in informative non-GAAP

reporting and a decline in the quality of special items, respectively.15

In particular, Heflin and

Hsu (2008) report a decline in the exclusion of special items and a reduction in the association

14

A degree of caution is warranted when interpreting results in Heflin and Hsu (2008) given that they use IBES

actual EPS to proxy for non-GAAP disclosure, and in Kolev et al. (2008) because their evidence that firms which

ceased disclosing non-GAAP earnings following Reg G had lower quality exclusions in the pre-intervention period

is based on just 28 cases where management stopped disclosing. 15

In contrast, Fortin et al. (2009) find no evidence that Reg G deterred REIT firms from reporting non-GAAP

information, while the quality of such measures increased.

22

between returns and earnings forecast errors, suggesting Reg G reduced management’s

willingness to use non-GAAP earnings to convey information about permanent income. Kolev et

al. (2008) find that special items became less transitory following Reg G, consistent with

management shifting more recurring expenses (previously classified as other exclusions) into

special items. Paradoxically, therefore, Reg G may have resulted in less transparent reporting

(i.e., more camouflage for recurring expenses after the reconciliation) and potentially higher

street earnings if investors simply strip out special items. These findings are consistent with

Bennett Stewart’s view that Reg G reduced alternative earnings disclosures and hence lowered

transparency (Copeland et al. 2006: 79).

5. Regulatory considerations

5.1 A problem unique to accounting?

Notwithstanding the brief summary in section 1 acknowledging how the practice of

modifying GAAP earnings extends well beyond adjustments made by management for external

reporting purposes, the discussion so far has treated customized reporting as an issue specific to

accounting (albeit broadly defined). In reality, the convention of using adjusting metrics or

applying alternative definitions when presenting and analysing quantitative information extends

well beyond the domain of financial reporting as the following examples illustrate:

Commenting on UK economic performance, former Governor of the Bank of England Sir

Mervin King observed that the UK economy would have grown by 1.5 per cent in 2012 if it

had not been hit by a collapse in construction and a dramatic fall in North Sea oil

production (King 2013).

School league tables are common place in many countries and often focus on the

percentage of pupils receiving top grades. However, many stakeholders argue such

23

measures result in unfair comparisons because they say more about differences in intakes

than differences in teaching quality. In response, some schools independently report

“ability-adjusted” or “value-added” rankings that control for potential confounding effects

(and also typically result in a more favourable ranking for the reporter);

In December 2003, then Chancellor of the Exchequer in UK announced that henceforth UK

monetary policy would be based on the Consumer Price Index (CPI) rather than the Retail

Price Index (RPI). Crucially, CPI tends to be structurally lower than RPI (e.g., it excludes

mortgage interest payments). The motives for the switch were widely debated, particularly

because it severed political ends at the time and has continued to do so.16

Low cost airlines have routinely excluded charges such as baggage fees, taxes, credit card

fees, fuel surcharges, etc. from the first published fare passengers see in advertisements and

on their website. Often the full fare is only revealed late in the booking process. Airlines

argue that quoting prices pre-tax is standard practice elsewhere, while other costs are

excluded because they are either discretionary or uncontrollable. Critics, however, argue

that these policies represent blatant marketeering designed to fool customers by disguising

the full cost of the fare.17

The common practice of adjusting reported performance metrics and using alternative

methods of measuring the same underlying construct is in part a response by information

16

It has been argued that CPI is a more accurate measure: it uses a geometric mean whereas RPI uses an arithmetic

mean, with the former better reflecting changes in consumer spending patterns relative to changes in the price of

goods and services. CPI is also internationally more comparable because it employs methodologies and structures

that follow international legislation and guidelines. Sceptics, however, highlight how the structurally lower CPI

metric limits increases in index-linked government spending such as pensions and public sector wages. 17

In the UK, the Office of Fair Trading announced on August 14, 2009 that Jet2.com had agreed to ensure

customers are made aware of fixed, non-optional costs early in the booking process and to provide a link at the start

of booking process to a web page showing the prices of all optional charges passengers may incur on top of the

standard flight cost. Effective January 26, 2012 the US Department of Transportation decreed that all mandatory

taxes and fees assessed on a per-passenger basis must be included in the first published fare passengers see. Three

low cost operators subsequently filed lawsuits arguing that pre-tax prices are standard practice in other sectors and

that airlines should not be treated differently.

24

preparers and users to the application of standardized measurement rules that ignore

idiosyncrasies at the micro level. Viewed in this context, it comes as no surprise that corporate

management seek to loosen the shackles imposed by GAAP and present results in a more

customized form. Insofar as non-GAAP earnings are a manifestation of this general

phenomenon, an important question for accounting standard setters and security market

regulators is the extent to which such behaviour creates unique problems (and demands unique

solutions). Several factors suggest non-GAAP reporting may create particular problems in an

accounting context. First, the tension between obfuscation and informative reporting is

particularly pronounced for financial reporting. Relative to many other fields, the technical

nature of financial reporting increases the scope for confusion and hence the opportunity for

obfuscation. Disregarding such disclosures as uninformative is nevertheless risky for investors

given management’s information advantage and the potential for non-GAAP earnings to

communicate private information. The fact that researchers using sophisticated methods have

struggled to disentangle these effects highlights the problem facing investors and regulators.

Second, non-GAAP disclosures threaten the credibility and integrity of the reporting

system in a way that customization practices in other domains do not. Earnings as a construct

owes its pre-eminent position in financial analysis and contracting to rules and assurance systems

designed to uphold faithful representation, reliability, comparability, and timeliness. High

information costs render earnings a trusted and valued brand in the corporate information

environment. The proliferation of non-GAAP earnings metrics unconstrained by the normal

disciplinary forces of the accounting system risks contaminating the very brand accounting

regulators fight so hard to maintain.18

While evidence that investors attach relatively more

18

Consistent with this risk, the FASB has expressed concern that the proliferation of non-GAAP earnings

disclosures is undermining the quality of financial reporting (FASB 2002).

25

weight to non-GAAP earnings is less of a concern when management use these disclosures to

provide supplementary information on important earnings streams, it represents a major threat to

accounting when non-GAAP earnings are associated with low information quality and fraudulent

reporting. For example, the SEC Division of Enforcement has identified non-GAAP earnings as

an important fraud risk factor (Leone 2010). If stakeholder groups including politicians and the

media fail to discriminate between GAAP and non-GAAP earnings in the case of financial

scandals, or if by failing to adopt a proactive stance accounting standard setters are viewed as

implicitly condoning non-GAAP disclosures, then problems associated with non-GAAP earnings

could taint the reputation of both (GAAP) earnings and financial reporting more generally. If

non-GAAP disclosures lead to the earnings brand being hijacked by preparers (and to a lesser

extent users in the form of equity analysts) then the practice represents a material danger that the

accounting profession cannot afford to ignore. Crucially, part of this risk stems from earnings’

role in the general financial communication process, much of which lies outside the traditional

focus and remit of accounting standard setters. Non-GAAP earnings reporting therefore raises

questions regarding the boundaries of accounting regulation and the nature of interactions

between accounting standard setters and securities regulators.19

5.2. Are current regulations sufficient?

The tensions and reputational risks associated with non-GAAP reporting place regulators

in a difficult position. The impossibility of identifying a single, theoretically superior GAAP-

19

Non-GAAP earnings disclosures also raise questions about the boundaries of regulatory intervention more

generally. For example, section 404 in the Sarbanes-Oxley Act dealing with internal controls over financial

reporting is restricted to GAAP earnings; no requirement currently exists for management to comment on controls

over non-GAAP reporting (Bryan and Lilien 2005). Similarly, while Reg G requires reconciliation to GAAP

earnings, it is silent on the placement or format of graphical disclosures containing non-GAAP information. Dilla et

al. (2013) present experimental evidence that graphical non-GAAP disclosures influence both professional and non-

professional investors’ judgements.

26

based aggregate measure of periodic value-creation capable of satisfying all users’ needs

(Cornell and Landsman 2003), coupled with evidence supporting the incremental

informativeness of non-GAAP earnings, suggests that prohibiting such disclosures is neither

feasible nor desirable. The regulatory solution most likely involves ensuring non-GAAP metrics

are transparently reported so that users are not mislead and sceptical users do not overlook

information that might otherwise be helpful were it reported transparently. This view is

consistent with the philosophy underlying US securities market regulations and IAS 33.

Nevertheless, anecdotal evidence suggests further improvements in non-GAAP earnings

transparency are possible (PricewaterhouseCoopers 2012).

Non-GAAP disclosures may lack transparency under prevailing regulations for several

reasons. First, there is inconsistency in the way non-GAAP information is regulated across

reporting channels. IAS 33, for example, governs non-GAAP reporting in firms’ financial

statements but most non-US stock exchanges afford management freedom on how non-GAAP

information is presented in earnings announcements, profit warnings, trading statements, etc.

Transparency requires consistency across alternative reporting channels so that investors can

easily reconcile information from different sources. A similar problem exists in the US where

management are allowed to present non-GAAP information in stock exchange filings but are

prevented from disclosing similar metrics as part of their financial statements. This inconsistency

impairs transparency by limiting (unsophisticated) investors’ ability to easily reconcile financial

statement information with market disclosures. It also conveys a mixed message: non-GAAP

earnings are too unreliable to be included in firms’ audited financial statements yet they can form

the basis of key communications with market participants.

27

A second reason why current regulations fall short of full transparency is that

reconciliations focus on excluded items (typically expenses). While included transitory items

(typically gains) are equally important they are not covered by reconciliation requirements.20

Curtis et al. (2011) find that approximately 42 percent of US firms in their sample fail to report

transitory gains transparently at the earnings announcement, leading to mispricing that is only

resolved after firms publish their 10-Q/K filings that provide more structured information on

transitory items. Choi et al. (2007) also report evidence of UK firms strategically retaining

nonrecurring gains in non-GAAP earnings. Transparent disclosure of transitory items included in

non-GAAP earnings is arguably as important as information on non-GAAP exclusions.

Existing rules requiring reconciliations from non-GAAP earnings to the closest GAAP

equivalent also lack transparency because they place no requirement on management to justify

exclusions. Instead, investors are left to draw their own conclusions why a particular item has

been omitted from GAAP earnings. To the extant treatment of a particular earnings component

can vary across firms and time, a short explanation accompanying each excluded item would

help investors better assess the validity of the non-GAAP treatment. For example, exclusion

(inclusion) of losses (gains) on asset sales is questionable (reasonable) where the sale is unrelated

to normal operating activities and vice versa. Likewise, it would be useful if management stated

explicitly their reason for omitting recurring expenses such as depreciation and amortization.

Finally, transparency is compromised because non-GAAP earnings represent only part of

the earnings game played-out between management and analysts. To the extent non-GAAP

20

On January 19, 2002 the SEC issued a cease-and-desist order against Trump Hotels & Casino Reports, Inc. for

violation of the anti-fraud provisions of the Securities Exchange Act of 1934 following its October 25, 1999

quarterly earnings announcement in which Trump stated that earnings exceeded analysts’ expectations after

excluding a one-time charge of $81.4 million relating to closure of Trump World Fair. An associated press release

attributed the positive results to operational improvements. However, the earnings release failed to disclose that the

quarterly non-GAAP earnings number included a $17.5 million one-time gain resulting from termination of a lease

of a restaurant tenant at one of its casinos.

28

earnings speak directly to professional investors and the analyst community (Christensen et al.

2011), the corresponding street earnings forecast represents a key piece of contextual

information typically omitted from non-GAAP disclosures. Research identifies the consensus

earnings forecast as an important benchmark influencing non-GAAP earnings disclosures. Of

particular concern are those non-GAAP disclosures that overturn a negative GAAP earnings

surprise. While sophisticated investors likely understand the relation between non-GAAP

earnings and market expectations, the same is unlikely to hold for private, non-expert investors

because information about street earnings is harder to access. Requiring management to disclose

the most recent consensus street earnings forecast from one of the major analyst tracking services

alongside GAAP and non-GAAP earnings (together with details, where appropriate, of any

material difference between management’s non-GAAP methodology and the models used by

analysts) would provide investors with important contextual information, as well as acting as a

simple red flag with respect to management’s disclosure incentives.

It is unclear precisely how far regulators should go in requiring supplementary

disclosures designed to protect investors who fixate on non-GAAP earnings without taking the

trouble to distinguish fiction from fact. Greater transparency increases reporting costs and risks

information overload. Where additional disclosure is considered appropriate, it should not come

at the expense of increased complexity and redundant boilerplating. The most effective

innovations aimed at further enhancing transparency are likely to be those that result in simple,

objective, and unambiguous disclosures.

6. Further research

Much of the research to date on non-GAAP reporting has focused on understanding what

motivates management to adjust GAAP earnings and how investors interpret such disclosures.

29

While prior research provides a useful basis for predicting disclosure behaviour, a non-trivial

fraction of firms are misclassified based on existing determinants: some firms fail to report non-

GAAP earnings even when doing so would enable them to overturn weak or disappointing

GAAP earnings performance; others fail to disclose even in the presence of nonrecurring items.

Our understanding of the decision to report non-GAAP earnings is clearly imperfect. Further

work aimed at developing a more complete picture of reporting motives is required. Studying

off-diagonal cases that are misclassified using models based on known reporting determinants

could provide interesting insights. Reducing the set of firms on which to focus would enable

researchers collect more granular information capable of casting new light on reporting practices.

While understanding management’s reporting motives is crucial, other important questions

regarding non-GAAP disclosures remain answered. Considerable scope exists for interesting

contributions including:

Understanding the extent and impact of reporting consistency. IAS 33 gives managers the

option of reporting non-GAAP earnings as part of the audited financial statements in

addition to disclosing customized earnings metrics through unaudited channels such as

earnings press releases and conference calls. The impact of disclosure consistency across

alternative reporting channels and the extent to which financial statement disclosures play a

disciplinary role for other (unregulated) non-GAAP disclosures represents an important

issue for policymakers.21

Understanding the interactions with other financial reporting decisions. Although the

majority of extant research examines non-GAAP earnings in isolation from other financial

reporting decisions, these disclosures likely form part of a broader communication policy

21

In the US, auditors are potentially responsible for ensuring consistency of non-GAAP earnings in voluntary

disclosures such as press releases, with any non-GAAP numbers included in mandated disclosures such as the 10-

Q/K (Chen et al. 2012). Whether such consistency is upheld in practice is an open question.

30

with market participants. Accordingly, the extent of discretion exercised over GAAP

earnings could influence non-GAAP reporting behaviour. Black et al. (2013) and Isidro and

Marques (2013) present evidence of a substitute relation between opportunistic non-GAAP

reporting and other forms of earnings management. Similarly, Kyung et al. (2013)

document how the propensity to report non-GAAP earnings increases and the quality of

non-GAAP exclusions declines in response to voluntary adoption of compensation

clawback provisions that improve GAAP reporting quality by increasing the costs

associated with misstatement. Meanwhile, the SEC’s view that non-GAAP earnings

represent important fraud risk factor (Leone 2010) suggests a degree of complimentary

between accounting manipulation and aggressive non-GAAP reporting.22

Conversely, changes in non-GAAP reporting behaviour may have knock-on effects for

GAAP earnings quality (Ewert and Wagenhoff 2005). Consistent with this view, Kolev et

al. (2008) document how improvements in non-GAAP reporting quality following Reg G

were accompanied by a reduction in the quality of special items as managers adapted to the

new disclosure environment by shifting more recurring expenses into special items. A

better understanding of the interactions between components of the financial reporting

system is essential for predicting any unintended consequences of regulatory interventions

aimed at curbing opportunistic non-GAAP reporting behaviour.

Exploring narrative disclosures associated with non-GAAP reporting. Several aspects of

narrative reporting are relevant in the context of non-GAAP earnings. First, Reg G requires

management to disclose the reason(s) why non-GAAP earnings provide useful information

and the additional purposes (if any) for which management use such metrics. As far as I am

22

In related work, Petaibanlue et al. (2013) document how the value relevance of non-GAAP earnings disclosed by

UK firms on the face of the income statement is lower than non-GAAP earnings disclosed solely in the notes when

the disclosing firm is suspected of earnings management.

31

aware, these potentially rich disclosures have yet to be studied systematically. Analysis of

these narratives could help shed new light on the motives for non-GAAP reporting, as well

as the degree to which non-GAAP metrics are embedded in firms’ internal performance

management systems. Understanding whether investors find such disclosures useful is also

important in the broader context of evolving financial regulation, much of which seeks to

promote transparency through increased narrative reporting. The benefits of expanded

narrative reporting in such contexts remain unclear, however, particularly in view of

concerns over boilerplating.

Second, non-GAAP earnings form an increasingly central feature of firms’ performance

reporting narrative (Graham et al. 2005), either via earnings press releases, conference

calls, or financial statements published in accordance with IFRS. While prior research links

non-earnings with attempts to influence investor perceptions of performance (Bowen et al.,

2005), our understanding of how non-GAAP disclosures correlate with other forms of

impression management behaviour is limited. Assuming management use multiple

reporting levers to influence investor perceptions, studying non-GAAP earnings in the

broader context of narrative disclosures could shed further light on managements’ reporting

motives and provide potential red flags for investors. Evidence presented by Guillamon-

Saorin et al. (2012) of complimentarity between non-GAAP disclosures and five

impression management proxies for a sample of European firms’ earnings announcement

press releases represents an important first step in this direction.

7. Summary

The decision by management to supplement GAAP earnings with additional non-GAAP

disclosures that exclude certain GAAP earnings components is a global phenomenon, the

32

popularity of which appears to be increasing in many jurisdictions. Despite the large body of

research devoted to understanding non-GAAP reporting, unequivocal insights regarding

underlying reporting motives remain elusive. A significant fraction of non-GAAP disclosures

undoubtedly provide useful information about performance and value but discriminating between

informative reporting and opportunism represents a challenge for investors and researchers alike.

Investors place less weight on income-increasing non-GAAP earnings disclosures when the

incentive for managerial opportunism is particularly high (e.g., when GAAP earnings fall short

of a key earnings threshold), suggesting that market participants do not respond mechanically to

such disclosures. Nevertheless, some investors (mainly unsophisticated non-professionals) are

misled when non-GAAP earnings exclusions are presented opaquely.

The apparent schizophrenic nature of non-GAAP earnings creates the classic dilemma for

regulators with respect to relevance versus reliability. Preventing management from disclosing

non-GAAP earnings is neither feasible nor desirable. Instead, the solution appears to lie in

ensuring such information is presented clearly. Transparent reconciliations of non-GAAP

earnings to the corresponding GAAP earnings number can eliminate the scope for

misunderstanding and market mispricing, and are therefore particularly helpful to small,

unsophisticated investors. While prevailing accounting standards and securities market

regulations provide important steps in this regard, transparency remains compromised due to

scope for inconsistent disclosure across alternative reporting channels, insufficient clarity about

transitory gains included in non-GAAP earnings, no requirement for management to explain the

rationale underlying specific GAAP exclusions (inclusions), and continued opacity for non-

professional investors with respect to reported performance and market expectations.

33

While customized reporting and alternative measurement methods are commonplace in

many fields, non-GAAP earnings create particular challenges not least because opportunistic

disclosure behaviour threatens the credibility and integrity of the underlying reporting system.

Further efforts on the part of accounting standard setters and financial market regulators to

improve reporting transparency is paramount to avoid perceived problems with non-GAAP

earnings quality tarnishing the reputation of the GAAP earnings brand. However, since much of

the earnings communication process occurs beyond the boundaries of the financial statements

and annual report, this in turn raises fundamental questions about ultimate control of the

performance reporting process and the definition of earnings used by market participants.

Although it is unclear precisely how far regulators should go to ensure uniformed, naïve or

careless investors are not disadvantaged by non-GAAP earnings disclosures, the level of

reputational risk facing the financial reporting system suggests accounting standard setters

cannot afford to ignore the central role non-GAAP earnings play in the financial communication

process. Ring-fencing audited financial statements and viewing non-GAAP earnings as a

problem for other regulators to address is a dangerous path for accounting standard setters to

tread.

34

References

Abarbanell, J. S. and R. Lehavy 2007. Letting the “tail wag the dog”: The debate over GAAP

versus street earnings revisited. Contemporary Accounting Research 24(3): 675-723

Athanaskou, V., N. Strong and M. Walker, 2007. Classificatory income smoothing: The impact

of a change in regime of reporting financial performance. Journal of Accounting and Public

Policy 26(4): 387-435

Andersson, P., and Hellman, N., 2007. Does pro forma reporting bias analyst forecasts?.

European Accounting Review 16(2): 277-298.

Allee, K. D., N. Bhattacharya, E. L. Black and T. E Christensen. 2007. Pro forma disclosure and

investor sophistication: External validation of experimental evidence using archival data.

Accounting, Organizations and Society 32: 201-222

Aubert, F., 2009. The relative informativeness of GAAP and pro forma earnings announcements

in France. Unpublished working paper, available at:

http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1414423

M. E. Barth, I. D. Gow and D. Taylor, 2012. Why do pro forma and street earnings not reflect

changes in GAAP? Evidence from SFAS 123R. Review of Accounting Studies 17(3): 526–562

Barton, J. and G. Waymire, 2004. Investor protection under unregulated financial reporting.

Journal of Accounting and Economics 38: 65-116

Beattie, V., S. Brown, D. Ewers, B. John, S. Manson, D. Thomas, and M. Turner, 1994.

Extraordinary items and income smoothing: A positive accounting approach. Journal of

Business Finance and Accounting 21(6): 791-811

Bhattacharya, N., E. L. Black, T. E. Christensen and R. D. Mergenthaler, 2004. Empirical

evidence on recent trends in pro forma reporting. Accounting Horizons 18(1): 27-43

Bhattacharya, N., E. L. Black, T. E. Christensen and C. R. Larson, 2003. Assessing the relative

informativeness and permanence of pro forma earnings and GAAP operating earnings. Journal

of Accounting & Economics 36 (1-3): 285-319

Black, E. L., T. E. Christensen, T.T. Joo and R. Schmardebeck, 2013. The relation between

earnings management and pro forma reporting. Unpublished working paper, available at:

http://ssrn.com/abstract=2259379

Black, D. E. and T. E. Christensen, 2009. US Managers' use of ‘pro forma’ adjustments to meet

strategic earnings targets. Journal of Business Finance and Accounting, 36(3-4): 297-326

35

Black, D. E., E. L. Black, T. E. Christensen and J. F. Waegelein, 2011. The effects of executive

compensation contracts and auditor effort on pro forma reporting decisions. Unpublished

working paper, available at SSRN: http://ssrn.com/abstract=1312562

Black, F.. 1993. Choosing accounting rules. Accounting Horizons 7(4): 1-17

Bowen, R. M., A. K. Davis and D. Matsumoto, 2005. Emphasis on pro forma versus GAAP

earnings in quarterly press releases: Determinants, SEC intervention, and market reactions. The

Accounting Review 80(4): 1011-1038

Bradshaw, M. and R. Sloan, 2002. GAAP versus The Street: An empirical assessment of two

alternative definitions of earnings. Journal of Accounting Research 40(1), 41-66.

Brown, L. and K. Sivakumar, 2003. Comparing the value relevance of two operating income

measures. Review of Accounting Studies 8(4), 561-572

Bray, C. 2001. SEC looking at pro forma earnings with purpose in mind. Dow Jones Newswire

(November 8).

Bryan, Stephen H. and Lilien, Steven B., 2005. Characteristics of Firms with Material

Weaknesses in Internal Control: An Assessment of Section 404 of Sarbanes Oxley.

Unpublished working paper, available at SSRN: http://ssrn.com/abstract=682363

Bushman, R.M. and A.J. Smith, 2001. Financial Accounting Information and Corporate

Governance. Journal of Accounting and Economics 32: 237-333

Cameron, R., M. Percy and P. Stevenson-Clarke, 2012. Do large Australian companies

emphasise non-GAAP financial measures over statutory net profit (GAAP) in annual reports?

JASSA: The Finsia Journal of Applied Finance 2012(1): 19-25

Chen, L., G. V. Krishnan, and M. Pevzner, 2012. Pro forma disclosures, audit fees, and auditor

resignations. Journal of Accounting and Public Policy 31(3): 237-257

Choi, Y-S., S. Lin, M. Walker and S. Young, 2007. Disagreement over the persistence of

earnings components: evidence on the properties of management-specific adjustments to

GAAP earnings. Review of Accounting Studies 12: 595-262

Choi, Y-S. and S. Young, 2013. Transitory earnings components and the two faces of non-

GAAP earnings per share. Accounting and Finance forthcoming

Christensen, T. E., K. J. Merkley, J. W. Tucker, and S. Venkataraman, 2011. Do managers use

earnings guidance to influence street earnings exclusions? Review of Accounting Studies 16(3):

501-527

36

Christensen, T. E., M. S. Drake and J. R. Thornock, 2013. Optimistic reporting and pessimistic

investing: Do pro forma earnings disclosures attract short sellers? Contemporary Accounting

Research forthcoming, DOI: 10.1111/1911-3846.12009

Cohen, D. A., R. N. Hann and M. Ogneva, 2007. Another look at GAAP versus the Street: an

empirical assessment of measurement error bias. Review of Accounting Studies 12: 271-303

Copeland, T., B. Stewart, T. S. Harris, S. F. O'Byrne, J. Pettit, D. Wessels and D. Chew. 2006.

Financial Management Association Roundtable on Stock Market pricing and Value-based

Management. Journal of Applied Corporate Finance 18(2): 56-81

Cornell, B. and W. Landsman, 2003. Accounting and valuation: Is the quality of earnings an

issue? Financial Analysts Journal 59(6): 20–28

Cornell, B. and W. Landsman, 2006. Accounting and valuation: How helpful are recent

accounting rule changes. Journal of Applied Corporate Finance 18(4): 34-43

Curtis, A., S. McVay and B. Whipple, 2011. Non-GAAP earnings: Informative or opportunistic?

An analysis of transitory gains. Unpublished working paper, available at:

http://ssrn.com/abstract=1965046

D’Avolio, G., E. Gildor and A. Shleifer, 2002. Technology, information production, and market

efficiency, in Economic Policy For The Information Economy Symposium, sponsored by the

Federal Reserve Bank of Kansas City, August 2002. 125-160. Available at:

http://www.kc.frb.org/publicat/sympos/2001/papers/S02shle.pdf

Dilla, W.N., D. J. Janvrin and C. G. Jeffrey, 2010. Pro forma accounting disclosures: The effects

of reconciliations and financial reporting knowledge on nonprofessional investors’ judgments.

Unpublished working paper, available at:

http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1646618 (permission required before

citing)

Dilla, W. N., D. J. Janvrin and C. G. Jeffrey, 2013. The Impact of Graphical Displays of Pro

Forma Earnings Information on Professional and Nonprofessional Investors' Earnings

Judgments (July 21, 2010). Behavioral Research in Accounting 25(1): 37-60

Doyle, J., R. Lundholm, and M. Soliman, 2003. The predictive value of expenses excluded from

pro forma earnings. Review of Accounting Studies 8(2-3): 145-174.

37

Doyle, J. R., J. Jennings and M. T. Soliman, 2011. Do managers define non-GAAP earnings to

meet or beat analyst forecasts. Unpublished working paper, available at:

http://ssrn.com/abstract=1933882

Elliott, W. B. 2006. Are investors influenced by pro forma emphasis and reconciliations in

earnings announcements?. The Accounting Review 81(1): 113-133

Entwistle, G., G. Feltham, C. Mbagwu, 2006a. Misleading Disclosure of Pro Forma Earnings:

An Empirical Examination Journal of Business Ethics 69(4): 355-372

Entwistle, G., Feltham, G., Mbagwu, C., 2006b. Financial reporting regulation and the reporting

of pro forma earnings. Accounting Horizons 20: 39–55

Entwistle, G., Feltham, G., Mbagwu, C., 2012. Credibility Attributes and Investor Perceptions of

Non-GAAP Earnings Exclusions. Accounting Perspectives 11(4): 229–257

Ewert, R. and A. Wagenhofer, 2005. Economic effects of tightening accounting standards to

restrict earnings management. The Accounting Review 80(4): 1101-1124

FASB, 2002. Financial Performance Reporting by Business Enterprises. Available at:

http://www.fasb.org/board_handouts/07-10-02.pdf.)

Fortin, S., P. Liu and D. Tsang, 2009. SEC intervention and industry guidance: The effect on

non-GAAP financial disclosures. Unpublished working paper, available at:

http://ssrn.com/abstract=1268911

Frankel, R., S. McVay and M. Soliman, 2011. Non-GAAP Earnings and Board Independence.

Review of Accounting Studies 16(4): 719-744

Frederickson, J. R. and J. S. Miller, 2004. The effects of pro forma earnings disclosures on

analysts' and nonprofessional investors' equity valuation judgments. The Accounting Review

79(3): 667-686

Gore, R. and D. M. Scott, 1998. Toward a more informative measure of operating performance

in the REIT industry: net income vs. funds from operations. Accounting Horizons 12(4):323-

339

Graham, J. R., C. R. Harvey and S. Rajgopal, 2005. The economic implications of corporate

financial reporting, Journal of Accounting and Economics 40(1-3): 3-73

Guillamon-Saorin, E., H. Isidro and A. C. Marques 2012. Impression Management and Non-

GAAP Reporting in Earnings Announcements. Unpublished working paper, available at:

http://ssrn.com/abstract=2011889

38

Halsey, B. and G. Soybel. 2002. All about pro forma accounting. The CPA Journal 72: 13.

Hirst, D. E. and P. E. Hopkins, 1998. Comprehensive income reporting and analysts’ valuation

judgements. Journal of Accounting Research 36(supplement): 47-75

Hitz, J.-M., 2010. Information versus strategic reporting: Determinants of the disclosure of pro

forma earnings by large German corporations. Unpublished working paper, available at:

http://ssrn.com/abstract=1549675

Heflin F. and C. Hsu, 2008. The impact of the SEC’s regulation of non-GAAP disclosures.

Journal of Accounting and Economics 46(2-3): 349-365

Isidro, H. and A. C. Marques, 2011. The effects of managers’ compensation and board quality on

non-GAAP reporting decisions. Unpublished working paper, available at SSRN:

http://ssrn.com/abstract=1730163 or http://dx.doi.org/10.2139/ssrn.1730163

Isidro, H. and A. C. Marques, 2013. The Role of Institutional and Economic Factors in the

Strategic Use of Non-GAAP Disclosures to Beat Earnings Benchmarks. Unpublished working

paper, available at SSRN: http://ssrn.com/abstract=1940514 or

http://dx.doi.org/10.2139/ssrn.1940514

Jennings, R. and A. C. Marques, 2011. The joint effects of corporate governance and regulation

on the disclosure of manager-adjusted non-GAAP earnings in the US. Journal of Business

Finance and Accounting 38(3-4): 364-394

King, M. 2013. BoE Governor: Recovery in 'sight' for UK economy, 15 March 2013, available

at: http://www.itv.com/news/2013-03-15/boe-governor-recovery-in-sight-for-uk-economy/

Kolev, K. S., C. A, Marquardt and S. E. McVay, 2008. SEC scrutiny and the evolution of Non-

GAAP reporting. The Accounting Review 83(1): 157-184

Kyung, H., H. Lee and C. A Marquardt, 2013. The effect of voluntary clawback adoption on

non-GAAP reporting. Unpublished working paper, available at:

http://ssrn.com/abstract=2279528

Larcker, D. F. and B. Tayan, 2010. Pro forma earnings: What's wrong with GAAP? Rock Center

for Corporate Governance at Stanford University Closer Look Series: Topics, Issues and

Controversies in Corporate Governance No. CGRP-09. available at:

http://ssrn.com/abstract=1678066

39

Landsman, W. R., B. L. Miller, and S. Yeh, 2007. Implications of Components of Income

Excluded from Pro Forma Earnings for Future Profitability and Equity Valuation. Journal of

Business Finance and Accounting, 34(3-4): 650-675

Leone (2010) What’s on the SEC’s radar? CFO.com: September 29

Lin, S. and M. Walker, 2000. FRS3 earnings, headline earnings, and accounting-based valuation

models. Accounting and Business Research 30(4): 299-306.

Lougee, B. and C. Marquardt, 2004. Earnings informativeness and strategic disclosure: An

empirical examination of ‘pro forma’ earnings. The Accounting Review 79(3), 769-795.

Maines, L. A. and L. S. McDaniel, 2000. Effects of comprehensive-income characteristics on

non-professional investors’ judgements: The role of financial-statement presentation format.

The Accounting Review 75(2): 179-207

Marques, A., 2006. SEC interventions and the frequency and usefulness of non-GAAP financial

measures. Review of Accounting Studies 11(4): 549-574

McVay, S. E., 2006. Earnings management using classification shifting: An examination of core

earnings and special items. The Accounting Review 81(3): 501-531

Ohlson, J. A., 1995, Earnings, book value and dividends in equity valuation. Contemporary

Accounting Research 11(2): 661-687

Peasnell, K. V., 1982.. Some formal connections between economic values and yields and

accounting numbers. Journal of Business Finance and Accounting 9(3): 361-381

Petaibanlue, J., E. Lee and M. Walker, 2013. Non-GAAP earnings and impression management

in the financial statements of UK firms. Unpublished working paper, available at: xxx (ask

permission before citing)

Pitt, H., 2001. Remarks before the AICPA governing council, October 22, 2001. Available at:

http://www.sec.gov/news/speech/spch516.htm

PricwaterCoopers, 2012. Trust through transparency What does your reporting say about you?

PricewaterhouseCoopers LLP

Rainsbury, E., C. Hart and S. Malthus, 2012. Non-GAAP earnings disclosures by New Zealand

Companies. Unpublished working paper, available at:

http://www.massey.ac.nz/massey/fms/Colleges/College%20of%20Business/School%20of%20

Accountancy/ARA%202012/Rainsbury_Non-

GAAP%20Earnings%20Disclosures%20by%20NZ%20Cos.doc

40

Serafeim, G. 2011. Consequences and Institutional Determinants of Unregulated Corporate

Financial Statements: Evidence from Embedded Value Reporting. Journal of Accounting

Research 49(2): 529-571

Schrand, C., and B. Walther. 2000. Strategic benchmarks in earnings announcements: The

selective disclosure of prior-period earnings components. The Accounting Review 75: 151-177

Smith, T. 1992. Accounting for Growth. Century Business: London

Tweedie, D., 1993. Tweedie: the story so far. Accountancy 111(1197): 120-121

Vincent, L. 1999. The information content of funds from operations (FFO) for real estate

investment trusts (REITs). Journal of Accounting and Economics 26: 69-104

Walker, M. and E. Louvari, 2003. The determinants of voluntary disclosure of adjusted earnings

per share measures by UK quoted companies. Accounting and Business Research 33(4): 295-

309

Zhang, H. and L. Zheng, 2011. The valuation impact of reconciling pro forma earnings to GAAP

earnings. Journal of Accounting & Economics 51(1-2): 186-2