School of Accounting Seminar Series. IFRS and non GAAP earnings disclosures: Determinants and consequences. Ann Tarca

Transcription

1 Australian School of Business School of Accounting School of Accounting Seminar Series Semester 2, 2013 IFRS and non GAAP earnings disclosures: Determinants and consequences Ann Tarca The University of Western Australia Date: Friday 8 th November 2013 Time: 3.00pm 4.30pm Venue: ASB 216

2 IFRS and Non-GAAP Earnings Disclosures: Determinants and Consequences Lance Malone, a Ann Tarca b and Marvin Wee b* 30 October 2013 a Employee, Commonwealth Bank of Australia. The opinions expressed in this article are those of the authors only. They do not reflect the opinions of the Commonwealth Bank of Australia. b Business School, University of Western Australia, Stirling Highway, Crawley, Western Australia * Corresponding author: Marvin Wee, UWA Business School, M250, 35 Stirling Highway, Crawley, Western Australia Marvin.Wee@uwa.edu.au Tel: We acknowledge the financial support of the BT Financial Group Victor Raeburn Honours Scholarship, the Hackett Foundation Alumni Honours Scholarship and the UWA Business School. We thank seminar participants at the AFAANZ 2012, Justus-Liebig-Universität, Giessen, Monash University, the University of Queensland, the University of Western Australia and the Australian National University 2011 College of Business and Economics Honours Colloquium and David Emmanuel, Martin Glaum, Majella Percy and Anne Wyatt and for their helpful comments. We appreciate the valuable research assistance of Gabriella D Orsogna and Jeremy Tan. 1

3 IFRS and Non-GAAP Earnings Disclosures: Determinants and Consequences Abstract We investigate the relationship of companies and analysts measures of non-gaap earnings and the incidence and amount of IFRS remeasurements (related to financial instruments, impairment and revaluation of investment property and agricultural, pension and insurance assets) and non-recurring items. Our sample includes 613 Australian company-years from 2008 to We find non-gaap earnings disclosure is more likely for firms with a higher incidence (but not magnitude) of remeasured and non-recurring items. Companies non-gaap disclosures appear to be useful in highlighting remeasured and non-recurring items to analysts and are associated with lower analyst forecast error in the subsequent period. We find that while the difference between GAAP and non-gaap earnings is value relevant for non-financial firms, remeasured items adjusted by companies and analysts are generally not associated with share price. Our findings suggest release of non-gaap disclosures is part of efficient information exchange rather than motivated by opportunism. Key words: IFRS, Non-GAAP, Pro forma earnings, voluntary disclosure, analyst forecast accuracy and dispersion, value relevance. JEL Codes: M40, M41 2

4 IFRS and Non-GAAP Earnings Disclosures: Determinants and Consequences 1. Introduction This study investigates the relationship of IFRS 1 financial reporting and the disclosure of non- GAAP earnings. Non-GAAP earnings are figures reported by management that exclude items required to be recognised under IFRS. For example, non-gaap earnings may be GAAP earnings before deducting particular items such as interest and tax (EBIT) or GAAP earnings before non-recurring items, sometimes called core, underlying or future maintainable earnings. In the US, non-gaap earnings are commonly investigated as operating, pro forma or street earnings, the latter being analyst-adjusted earnings. The reporting of non-gaap earnings has increased following adoption of IFRS, leading to questions by regulators and extensive discussion within the financial reporting community about the purpose and usefulness of non- GAAP reporting (CESR 2005; EFRAG 2009; PWC 2007; Fitch Ratings 2012; ASIC 2005, 2011; FINSIA 2008, 2009; KPMG 2010). Prior studies investigate many aspects of non-gaap reporting, particularly with regards to whether non-gaap earnings provide useful information for evaluating companies performance and predicting earnings (Bradshaw and Sloan 2002; Bhattacharya et al. 2003; Landsman et al. 2007). Other studies have focused on the opportunistic use of non-gaap reporting to influence analysts forecasts and investors decision making (Bhattacharya et al. 2007; Black et al. 2010; Doyle et al. 2013; Guillamon-Saorin et al. 2013). While many studies explore aspects of non- GAAP reporting, ours is the first to explore the link between non-gaap reporting and the 1 International Financial Reporting Standards, issued by the International Accounting Standards Board (IASB). 3

5 specific measurement requirements of IFRS. We consider the actual adjustments made by listed companies and by security market analysts in determining non-gaap earnings, not just categories of items (e.g. total exclusions, special items and other exclusions from Compustat). We examine the items occurrence, timing and magnitude and consider their economic importance by investigating the relationship of firms and analysts adjustments for remeasured and non-recurring items with share prices and properties of analysts forecasts. Our study is based on ASX 200 companies in the period (613 firm-years) for which we could obtain data about companies and analysts non-gaap earnings adjustments. We focus on several groups of items commonly adjusted under IFRS because they reflect fair value remeasurements and preparers judgements and estimates (gains and losses on financial instruments; amortisation; impairment; and revaluation of tangible assets, investment property and agricultural assets) and are sometimes argued to not reflect firm performance or future earnings. We also consider non-recurring items (relating to mergers and restructuring and a catch-all category of the remainder of non-recurring items) that are argued to be irrelevant for predicting earnings. We collect data about the incidence and magnitude of the items according to (i) firms (as disclosed in their statutory accounts, earnings announcements and investor presentations) and (ii) analysts (adjustments recorded in the Morningstar-Aspect Huntley FinAnalysis database). We investigate the extent to which these specific groups of non-gaap items are relevant for information users, based on tests of their association with properties of analysts forecasts and share prices. Prior studies have investigated non-gaap reporting in terms of the association 4

6 between market returns and GAAP or non-gaap earnings (Bradshaw and Sloan 2002; Arbarbanell and Lehavy 2002). Other studies examine predictive ability of non-gaap earnings for future profitability and returns (Burgstahler et al 2002; Brown and Sivakumar 2003; Lougee and Marquardt 2004). Landsman et al. (2007) explore the relevance of non-gaap items for forecasting and their relationship with share price. Our main interest is decision usefulness of the items that are adjusted to arrive at non-gaap earnings so we attend to the second of the matters considered by Landsman et al. (2007) (i.e., value relevance of non-gaap items) and conduct tests of the adjusting items relationship with properties of analyst forecasts and with share price. In terms of considering both company and analyst non-gaap earnings and focusing on remeasurements, the prior study most similar to ours is Barth et al. (2012) who explore the relationship of pro forma earnings, street earnings and expensing of stock options under SFAS 123R. 2 Australia provides an excellent setting for our investigation, beyond the fact of the availability of the preparers and analysts data about the non-gaap adjustments. The Australian market, although small by world standards, has a key role in providing finance to Australian listed companies, all of which are legally required to use IFRS-equivalent standards. During our study period, the disclosure of non-gaap earnings was widely practiced and discussed but not regulated, in contrast to other jurisdictions such as the US where non-gaap reporting follows directions of the Securities and Exchange Commission (SEC) (Bowen et al. 2004; Barth et al. 2 Like our study, Zhang and Zheng (2013) consider companies individual reconciliation items (between GAAP and non-gaap earnings). However, they study US GAAP and use only dummy variables for the items and do not investigate the items magnitude. 5

7 2012). 3 In addition, our study period includes the 2008 financial crises period, when market and asset price volatility could lead to more sensitivity about IFRS remeasurements. Thus our setting allows for the investigation of incentives for, and consequences of, voluntary non-gaap disclosures, in relation to the adoption of a new set of accounting standards that required greater use of fair value measurement and introduced more judgement and estimates into financial reporting numbers. We expect and find firms with higher incidence of the selected profit or loss items (remeasurement and non-recurring items) in their financial statements are more likely to provide non-gaap disclosures. The total dollar amount of the item-groups is not a significant explanatory factor for non-gaap disclosure. In relation to the usefulness of non-gaap earnings, we find lower analyst forecast error in the following year for firms disclosing non- GAAP earnings. The amount of the non-gaap/gaap difference is positively associated with share price in a pooled sample and in each period in the study for non-financial firms. Firm adjustments for IFRS remeasurement items (relating to financial instruments, impairment and revaluation) are not value relevant, providing support for their exclusion by preparers. Similarly analysts adjustments for remeasurements for financial instruments and asset impairment are not associated with share price. Our evidence suggests non-gaap earnings disclosures relating to remeasurement are more likely to reflect information efficiency than opportunism. Our study adds to studies of non-gaap disclosure because we consider the relevance of individual items commonly adjusted in non-gaap disclosures. We extend earlier studies using 3 The SEC s Regulation G requires firms disclosing non-gaap earnings to label the disclosure as non-gaap and to reconcile it to GAAP earnings (Barth et al. 2012). 6

8 US data that have focused mainly on aggregate measures of non-gaap earnings. Our study extends Barth et al. (2012) who consider a specific adjusting item (share-based payment expense) and both companies and analysts non-gaap earnings. We show that non-gaap disclosures have a role in communicating information that is used by analysts, thus providing empirical evidence in support of practitioner claims about the need for non-gaap disclosures. Our results are more in line with studies that support the usefulness of non-gaap disclosures than the counter arguments of managerial opportunism and misleading disclosure. As such, our results do not suggest regulatory intervention to limit non-gaap disclosure is necessary, at least in Australia. The remainder of the paper is organised as follows. Section 2 presents background and research predictions, section 3 outlines data sources and statistical models, section 4 presents results and section 5 concludes. 2. Background and Research Predictions 2.1 Non-GAAP disclosures A number of US studies indicate that investors find non-gaap earnings more strongly associated with returns, share price and future earnings than GAAP earnings. Bradshaw and Sloan (2002) conclude that street earnings are more strongly associated with returns than US GAAP earnings. Similarly, Brown and Sivakumar (2003) find operating earnings are more strongly associated with share price than GAAP net income. They suggest GAAP net income contains many non-operating items that reduce its value relevance compared to operating 7

9 earnings. Lougee and Marquardt (2004) suggest the higher value relevance of non-gaap earnings derives from their incremental information content relative to GAAP, thus they are useful only for firms with low-gaap informativeness. Thus, it appears that non-gaap earnings are potentially an important source of information about firm performance. However, studies also suggest non-gaap disclosures are used opportunistically by management to shape investors perceptions, particularly in relation to less sophisticated retail investors (Bhattacharya et al. 2007). While the exclusion of non-recurring losses may be argued to be justifiable, some firms also exclude standard recurring items, such as depreciation and common operating expenses (Black et al. 2010a). Doyle et al. (2003) find that excluded items are predictive of future cash flows and abnormal returns, raising questions about their classification as non-recurring. These studies suggest non-gaap earnings may be noisy information and difficult for market participants to interpret. For example, there appears to be some confusion among market participants about the relevance of the components of non-gaap earnings. Studying US listed firms, Burgstahler et al. (2002) conclude prices do not fully reflect the implications of excluded items (Compustat s special items) for future earnings. Doyle et al. (2003) also consider the relevance of excluded items for forecasting and conclude investors underreact to the excluded components, indicating market mispricing. Landsman et al. (2007) consider both forecasting and value relevance implications of excluded items (Compustat s total items, special items and other exclusions). They find the items are relevant for forecasting but significant coefficients not with the predicted sign for the excluded items lead the authors to conclude the items are mispriced. 8

10 Zhang and Zheng (2013) extend this line of research and show that mispricing is less for firms with higher quality reconciliation statements. The opportunistic (or aggressive) use of non-gaap earnings has been measured mainly though the exclusion of recurring items and the use of non-gaap earnings to meet benchmarks (Black and Christensen 2009; Brown et al. 2010; Doyle et al. 2013). Firms with stronger corporate governance have been found to have less aggressive exclusions (Frankel et al. 2010; Jennings and Marques 2010). Black et al. (2010b) show that mangers are less likely to disclose aggressive non-gaap earnings after the enactment of the Sarbanes-Oxley Act (post-sox) and the implementation of Regulation G in the US. However, Black et al. (2010b) still find use of aggressive non-gaap earnings disclosures post-regulation. Interestingly, Black et al. (2010b), replicating Bhattacharya et al. (2003) in a post-sox environment (i.e. one subject to greater regulatory constraints and oversight), find that although investors still pay more attention to non- GAAP than to GAAP operating earnings, they appear to discount non-gaap earnings when they perceive that earnings exclusions are overly aggressive. Barth et al. (2012) add to the evidence on opportunistic non-gaap reporting by demonstrating that firms are more likely to exclude share-based payment expense from their non-gaap earnings to manage investor perceptions while analysts are more likely to exclude the expense when the exclusion results in a measure of earnings that has greater predictive ability for firms future performance. 2.2 Adoption of IFRS These questions have generally not been investigated in an IFRS setting. Despite many similarities between US GAAP and IFRS, it not clear that the above findings will hold in IFRS 9

11 reporting countries because of differences in the institutional settings in which financial reporting takes place 4 and differences between countries in the impact of application of IFRS (e.g. considering a country s prior national GAAP). If firms experience a large change in accounting policies and financial reporting numbers (including earnings) when they apply IFRS, it could be argued that they have more to explain and communicate to investors, possibly leading to more non-gaap reporting. Studies indicate incidence of non-gaap reporting is higher under IFRS compared to prior national GAAP (KPMG 2010 for Australia; Hitz 2010 for Germany). Theories underpinning disclosure research are relevant to predicting how IFRS reporting may incorporate non-gaap earnings disclosures. The voluntary disclosure of private information by firms is a common equilibrium strategy in many theoretical models of disclosure (e.g. Milgrom 1981; Milgrom and Roberts 1986; Jovanovic 1982). Models incorporating cost of disclosure can generally be interpreted as showing a trade-off between the benefits obtained and the associated costs (Dye 1985, 1986; Verrecchia 1983, 1990) and that the equilibrium of the trade-off is achieved when the marginal benefits of additional disclosure are equal to the marginal costs (Diamond and Verrecchia 1991). Considering non-gaap disclosure, a partial equilibrium would be expected to occur when the benefits of ostensibly reflecting the real underlying earnings exceed the costs of preparation and dissemination of this information. Moreover, the benefits of releasing non-gaap earnings would likely be greatest when they are higher than statutory earnings, which is generally the case, because expenses and losses are added back to GAAP earnings (KPMG 2010). In this 4 Recall that enforcement of IFRS reporting requirements and national security market regulation is primarily the responsibility of national institutions. 10

12 sense, non-gaap earnings, like all discretionary disclosures, could represent an attempt to manage perceptions about earnings (Doyle et al. 2013). Guillamon-Saorin et al. (2013) show that managers of European firms who exclude recurring components from non-gaap earnings make use of more impression management in earnings press releases. The authors find investors and analysts identify management s intentions and react negatively to non-gaap disclosures. Reimsbach (2013) concludes non-gaap disclosures improve German investors evaluation of company performance but investors may consider these disclosures strategically motivated and thus weight non-gaap information less heavily. Isidro and Marques (2013) find that non-gaap disclosure is more likely when board compensation is linked to firms market performance and that an efficient governance structure reduces the propensity to disclose non-gaap earnings. Some analysts and firms maintain that the adjustments to GAAP earnings are necessary to modify the effects of accounting entries (required by accounting standards) that do not relate to business operations or accurately reflect the underlying business reality, and are therefore less relevant to investors (FINSIA 2009; Hitz 2010). Isidro and Marques (2013), considering non- GAAP reporting in press releases for 318 European firms , conclude that these disclosures are potentially opportunistic in 27 per cent and potentially informative for 48 per cent of cases. In IFRS reporting, the contentious items relate to remeasurements (gains/losses on financial instruments through profit or loss, impairment expense and remeasurements of assets such as investment properties and agricultural assets) and non-recurring items (gains/losses on mergers, integrations and divestment of business operations; and redundancies and restructuring 11

13 expense). To the extent that these transactions in IFRS are uninformative or misleading for the purposes of predicting future returns (see Barth et al. 2012, for US GAAP reporters exclusions for share-based payments), we posit that their exclusion can generate benefits for users of financial information (particularly analysts), and hence a firm s exposure to these items provides an incentive for the disclosure of non-gaap earnings. Thus we expect firms with more (more items or a larger amount) of (a) IFRS remeasurements or (b) non-recurring items are more likely to release non-gaap earnings. 5 Our first hypotheses can be stated formally as: H1: Non-GAAP disclosing firms are more likely to have more IFRS remeasurements considering (a) incidence and (b) amount. Some commentators contend that IFRS are flawed, particularly in relation to fair value remeasurement requirements (KPMG 2010). If these views are valid and remeasurements and non-recurring transactions distort reporting of firms underlying economic reality, additional disclosures that reveal the impact of these items may be an important supplementary source of information for forecasting. Andersson and Hellman (2007) report that, based on an experiment with Swedish analysts, those receiving both GAAP and non-gaap earnings measures provided a higher earnings forecast, implying additional information from non-gaap disclosures. Similarly, Reimsbach (2013) finds forecast earnings higher when non-gaap earnings are disclosed in an experiment with non-professional German investors. He reports that investors judgements are most favourable when both GAAP and non-gaap disclosure are provided. These studies suggest that managers can influence analyst exclusions through non-gaap earnings disclosure. Therefore we expect that, after controlling for other factors known to affect 5 We do not distinguish between gains or losses on these items as both should be informative. 12

14 properties of analyst forecasts, forecasts will be more accurate and have less disagreement for firms releasing non-gaap earnings. Taking the perspective that the information contained in non-gaap earnings is useful for market participants (Bradshaw and Sloan 2002), we expect the difference between GAAP and non-gaap earnings to be value relevant. However, consistent with the argument that the items are removed from earnings because they are non-recurring, not part of core operations and interfere with predictive ability, we do not expect the individual components of non-gaap earnings to be associated with share price. Our hypotheses can be stated formally as: H2: Non-GAAP disclosing firms are more likely to have lower analyst (a) forecast error and (b) dispersion; H3: The difference between GAAP and non-gaap earnings is positively associated with share price; and H4: The items adjusted as part of non-gaap earnings are not associated with share price. 3. Data and Method 3.1. Sample Selection and Data Collection We study large Australian firms (from the ASX 200, based on the largest 200 Australian firms by market capitalisation) because they are the most economically important and are more likely to have diverse shareholders, substantial financing needs and to be followed by security analysts. 6 We hand-collected disclosure of non-gaap earnings from firms annual reports, earnings announcements (i.e., preliminary financial statements and investor presentations) and 6 The ASX 200 represents approximately 80% of the market capitalisation of the Australian Securities Exchange. 13

15 investor presentations lodged with the ASX (accessed through the Securities Industry Research Centre of Asia-Pacific (SIRCA) Australian Company Announcement database and company websites). We searched for non-gaap earnings using Adobe Acrobat Pro text search software and terms such as underlying earnings and normalised profit. 7 The total sample comprises 613 firm-years from 2008 to 2010, of which 388 firm-years (63.3%) released non-gaap earnings information (Table 1). We identified the items expected to be associated with the release of non-gaap earnings based on practitioner studies (Ernst & Young 2007; KPMG 2010) and observed that they are related to both remeasured and non-recurring items. We classified the relevant financial statement items in firms statutory accounts (using profit or loss data extracted from the Aspect Huntley database) and all of the items recorded by Aspect Huntley analysts (in their database) as analysts adjustments to earnings into two categories (a) IFRS remeasurement items and (b) non-recurring items. Thus our first source of data is extracted from the Aspect FinAnalysis database of Profit and Loss items. For each firm, we record the amounts for remeasured and non-recurring items shown in profit and loss in the annual firm accounts. The second source is the non-gaap/gaap reconciliation (when provided by the company) in the annual report, earnings announcement or investor presentation, obtained from the firm s website. Our third source of data is the list of 7 We determined the search terms from a review of the literature about non-gaap earnings and a pilot study of the largest 20 firms, which confirmed the most commonly used terms. KPMG (2010) gave a list of key words commonly used in reporting non-gaap information, specifically: underlying earnings, normalised profit and cash earnings. Using these words, we examined the Annual Report, Preliminary Final Report and Investor Relations slide show presentation of the sample firms. The pilot study confirmed the use of these words and revealed use of the terms: before significant items and core earnings which were then added to our word search list. 14

16 analysts adjustments for each firm-year compiled by the Aspect Huntley analysts following the firm. The analyst adjustments are defined as items which are part of the organisation s operations but are considered abnormal (Aspect Huntley 2011) and reflect analysts views of the items not being a part of maintainable or underlying earnings to be used to predict future earnings. In this study, we refer to these items as analyst adjusted amounts. Our coding identifies six groups of items from within (1) the firms statutory profit and loss accounts and non-gaap/gaap reconciliation statements and (2) the analyst adjustments file. There are four groups of IFRS remeasurement items namely (i) gains or losses on the remeasurement of financial instruments to fair value through profit and loss under (IAS 39) AASB 139 Financial Instruments, (ii) impairment expenses under (IAS 36) AASB 136 Impairment of Assets, (iii) amortisation expense under (IAS 38) AASB 138 Intangible assets, 8 and (iv) revaluation of investment property under (IAS 40) AASB 140 Investment Property, agricultural assets under (IAS 41) AASB 141 Agriculture, pension assets under AASB 119 Employee Benefits and insurance assets under AASB 4 Insurance Contracts. There are two groups of non-recurring items. The first includes gains or losses associated with mergers, integrations, divestment of business operations, redundancies and restructuring costs. The second is a catch-all category for any remaining items, called other. This group contains the items that have not been classified as belonging to any of the previous five categories, such as tax effects, litigation expense, donations and losses from natural disasters. 8 Amortisation is not a remeasurement consistent with AASB 139, 140 or 141. However, we include amortisation expense because it is an item that has been commonly adjusted in the past (e.g. analysts add back amortisation of goodwill, see Cotter et al. 2012). 15

17 To determine the association of potential non-gaap information (in firm accounts) and analyst adjustments with properties of analyst forecasts, we included firms with two or more analysts. Data to calculate analyst forecast error and dispersion at three, six and nine months prior to firm financial year end are obtained from the I/B/E/S database. Share prices are obtained from the SPPR database provided by SIRCA and other financial data are obtained from the Aspect Huntly database Data Analysis Prior to running the models to test our hypothesis, we run binary logistic regression models to explore whether release of non-gaap earnings is associated with exposure to IFRS remeasurements and non-recurring items. The models are as follows: NONGAAP = α + α FININST + α IMPAIR + α AMORT + α REVAL + α MERGER it, 0 1 it, 2 it, 3 it, 4 it, 5 it, + α OTHER + α FIN + α MINING + α LOSS + α ACHEARN + α SIZE 6 it, 7 it, 8 it, 9 it, 10 it, 11 it, (1) + α NUMEST + α VARCFO + α PRE _ CRISIS + α POST _ CRISIS 12 it, 13 it, 14 it, it, it, it, β0 + β1 it, + + εit, 15 it, + α PREVAFE + α PREVFD + ε NONGAAP = COUNT controls (2) NONGAAP = γ + γ MAGNIT + controls + ε (3) it, 0 1 it, it, Where: NONGAAP!,! FININST!,! Dummy variable equal to 1 for firms that have released non-gaap earnings, zero otherwise. Current year s net gain or loss taken to profit and loss for the fair value remeasurement of financial instruments for firm i, scaled by the firm s absolute value of cash flow from operations for the year. 16

18 IMPAIR!,! AMORT!,! REVAL!,! MERGER!,! OTHER!,! COUNT!,! MAGNIT!,! Current year s impairment expense or reversal for firm i, scaled by the firm s absolute value of cash flow from operations for the year. Current year s amortisation expense or reversal for firm i, scaled by the firm s absolute value of cash flow from operations for the year. Current year s net gain or loss on revaluation of investment property, agricultural, pension and insurance assets for firm i, scaled by the firm s absolute value of cash flow from operations for the year. Current year s gains or losses associated with mergers, integrations, divestments of business operations, redundancies and restructuring for firm i, scaled by the firm s absolute value of cash flow from operations for the year. Current year s items in the Aspect Huntley database or firm reconciliation statement that are not included in any of the five variables above for firm i, scaled by the firm s absolute value of cash flow from operations for the year. Count of the current year s total number of non-zero items for non- GAAP variables (FININST, IMPAIR, AMORT, REVAL, MERGER and OTHER) from 0 to 6 for firm i. Net sum of the current year s total value (in dollars) of non-zero values for non-gaap variables (FININST, IMPAIR, AMORT, REVAL, MERGER and OTHER) for firm i, scaled by the firm s absolute value of cash flow from operations for the year Control variables: FIN!,! MINING!,! LOSS!,! ACHEARN!,! Dummy variable equal to 1 for firms in the GICS Financials Industry Group, zero otherwise. Dummy variable equal to 1 for firms in the GICS Metals and Mining industry, zero otherwise. Dummy variable equal to 1 if the current year s earnings per share is negative, zero otherwise. The absolute value of the difference between the current year s actual 17

19 earnings per share and last year s actual earnings per share, deflated by the share price at the end of the current year. SIZE!,! The natural log of the firm s market capitalisation at the beginning of the year. NUMEST!,! The number of analyst earnings forecasts included in the consensus forecast. VARCFO!,! Standard deviation of cash flows from operations over the previous 10 years at financial year end. PRE_CRISIS!,! Dummy variable equal to one for observations with financial year end prior to 1 July POST_CRISIS!,! Dummy variable equal to one for observations with financial year end post 30 June PREVAFE!,! Absolute forecast error (AFE) for firm i from the previous corresponding financial year. PREVFD!,! Forecast dispersion (FD) for firm i from the previous corresponding financial year. In Model 1 we include COUNT, calculated as 0 (none of the six items) to 6 (the firm has a nonzero value for all six items). In Model 2 we replace COUNT with MAGNIT, which is the sum of the dollar value of all six items, to test the impact of the net value of the items. In Model 3 we include the six items (FININST, IMPAIR, AMORT, REVAL, MERGER and OTHER) as individual amounts. Control variables include industry dummy variables (FIN and MINING) as we expect firms in the financial sector and mining sector to be more likely to have remeasured amounts included in the six items. We include SIZE and NUMEST to control for differences between firms in analyst following, which may relate to the non-gaap disclosure. Non-GAAP disclosure may be more likely when firms experience losses or have variability in earnings (Loguee and Marquardt 2004) so we include variables for incurring a loss (LOSS), change in 18

20 earnings from last year (ACHEARN) and variability of earnings over time (VARCFO). 9 The incentives to provide non-gaap disclosures (particularly remeasurements, which are sensitive to economic conditions) may change over time (Bradshaw and Sloan 2002) so we include dummy variables (PRE-CRISIS and POST-CRISIS) to capture the effects of the uncertainty associated with the global financial crisis period (1 July June 2009). We use Ordinary Least Squares (OLS) regression models with robust standard errors to explore the relationship of analyst forecast error (AFE) and forecast dispersion (FD) and release of non- GAAP earnings disclosures: AFE = δ + δ NONGAAP + δ LOSS + + δ VARCFO + + δ ACHEARN + it, it, 2 i, t 1 3 i, t 1 4 i, t 1 + δ PREVAFE + δ NUMEST + δ SIZE + δ ADR + δ PRE _ CRISIS (4) + δ POST _ CRISIS + ε 5 it, it, it, it, it, 10 it, it, FD = ϕ + ϕ NONGAAP + ϕ LOSS + ϕvarcfo + ϕ ACHEARN it, it, 2 it, it, it, ϕ PREVAFE + ϕ NUMEST + ϕ SIZE + ϕ ADR + ϕ PRE _ CRISIS (5) + ϕ POST _ CRISIS + e 5 it, it, it, it, it, 10 it, it, Where 10 : AFE!,! FD i,t Absolute forecast error measured as (A i,t F i,t-j ) / P i,t-j where A i,t is firm i s actual EPS for the financial year ended t; F i,t-j is firm i s median consensus forecast for EPS for the financial year ended t, measured j months prior to time t, where j is 3, 6 and 9 months; and P i,t-j is firm i s price per share j months prior to time t. Forecast dispersion measured at j months prior to the end of the 9 We use variability of cash flows rather than net income as the control because net income includes the remeasurements which are our experimental variables. 10 Other variables as defined in Equation 1. 19

21 ADR!,! financial year t, where j is 3, 6 and 9 months; captured by the standard deviation of firm i s EPS forecast, scaled by P i,t-j. Dummy variable equal to 1 for observations for firms cross-listed in the U.S. as American Depository Receipts, zero otherwise. Forecast error (AFE) and dispersion (FD) are measured consistently with prior studies (Hope 2003; Lang and Lundholm 1996). We measure both nine months before the financial year end (with forecasts at six months and three months considered in robustness tests). Forecasts occur a minimum of nine months prior to the next year end as ASX listing rule requires a disclosing entity s documents to be lodged with ASIC in the three months after the end of the accounting period (ASX, Chapter 4). We use nine month forecasts in our primary tests because we want to focus on the period when the non-gaap information is most likely to be relevant. As the financial year unfolds, other information will become available that may reduce the value of the pro forma information. We expect that non-gaap information is likely to be more valuable closer to the end of financial year prior to the date of the forecast. In Equations 4 and 5 (for AFE and FD, respectively) we measure non-gaap earnings disclosure using a dummy variable equal to one if the firm provides non-gaap earnings for the previous financial year end. We expect that firms providing non-gaap earnings disclosures at year t are more likely to have lower error and less dispersion in forecasts at year t+1. Additional variables are included in Equations 1-5 to control for factors that may be associated with non-gaap disclosure and properties of analyst forecasts. We include size (SIZE) as larger firms provide more disclosure and are followed by more analysts (FOLLOW) (Lang and Lundholm 1996; Hope 2003). Changes in expected earnings, volatility in earnings and incurring 20

22 losses require explanation to market participants and thus may encourage release of non-gaap earnings. We include ACHEARN (change in EPS in current year compared to prior year), VARCFO (standard deviation in cash flows over the previous ten years) 11 and, LOSS (dummy variable = 1 if firm incurs a loss in the current year) to control for these factors. We include industry dummy variables (FIN and MINING) to control for difficulty of forecasting in these sectors (Chalmers et al. 2010; Barth et al. 2011). The economic turmoil of the financial crisis may also impact on analysts forecasts so we include time period dummy variables to distinguish the years before (PRE-CRISIS) and after (POST-CRISIS) the financial crisis period. In Equations 4 and 5 (AFE and FD) we add variables to capture the level of error and dispersion in the previous year (PREVAFE and PREVFD) and the number of forecasts (NUMEST) as studies suggest these are explanatory factors for current year error and dispersion (Brown et al. 1999). The variable ADR is included to control for any effects of cross listing in the US since the US environment may encourage provision of more information for analysts or, alternatively, discourage provision of non-gaap earnings disclosures (Barth et al. 2013). In addition, we use OLS regression models to explore the value relevance of non-gaap earnings, financial statement amounts for items commonly included in non-gaap earnings and analyst-adjusted non-gaap amounts. We use the price level model adopted in prior studies (Barth and Clinch 1998; Goodwin et al. 2008; Chalmers et al. 2010), which is derived from Ohlson (1995), adding a variable capturing the difference between non-gaap and GAAP earnings (NG_DIFF$) in Equation 6 as follows: 11 We use cash flows rather than NPAT to proxy for variability of earnings because NPAT is affected by non-cash remeasurements while cash flow from operations is not. 21

23 Price = α + α BVE + α NI + α PF _ DIFF$ + ε (6) it, 0 1 it, 2 it, 3 it, it, where Price = a firm i s share price three months after end of year t; BVE = book value of equity per share, at year end t; NI = earnings per share, for year t; NG_DIFF$ = non-gaap earnings less GAAP earnings per share, for year t. Price = α + α BVE + α NI + α FININST + α IMPAIR REC REC it, 0 1 it, 2 it, 3 it, 4 it, + α AMORT + α REVAL + α MERGER + α OTHER + ε REC REC REC REC 5 it, 6 it, 7 it, 8 it, it, (7) Price = α + α BVE + α NI + α FININST + α IMPAIR AA AA it, 0 1 it, 2 it, 3 it, 4 it, + α AMORT + α REVAL + α MERGER + α OTHER + ε AA AA AA AA 5 it, 6 it, 7 it, 8 it, it, (8) All other values are as previously defined. Based on prior research, we predict book value of equity and earnings to be relevant to share price, and expect positive coefficients on BVE and NI. Consistent with the approach of Goodwin et al. (2008) and Barth et al. (2011), where additional items are included in the value relevance model, we add NG_DIFF$ in Equation 6. If non-gaap earnings are informative, we expect NG_DIFF$ to be associated with price. In Equation 7 we add the amounts for financial statement items commonly included in non-gaap earnings (FININST REC, IMPAIR REC, AMORT REC, REVAL REC, MERGE REC and OTHER REC ). 12 We expect that the six items will be related to price as they are considered relevant for predicting future earnings. Finally, in a variation of Equation 7, we include the analyst-adjusted amounts for the six items (FININST AA, IMPAIR AA, AMORT AA, REVAL AA, MERGER AA and OTHER AA ). We expect these adjustments to be associated with share 12 The IFRS remeasurement and the non-recurring items, i.e. FININST, IMPAIR, AMORT, REVAL, MERGER and OTHER are on a per share basis. 22

24 price, as they are the adjustments made by analysts to arrive at predictions of future earnings. Because Goodwin et al. (2008) finds financial services firms are affected by fewer, more complex standards than most other industries, we expect the value relevance to be more evident in non-financial firms. As such, we run our models over the pooled sample and a sample without the financial firms. 4. Results 4.1. Descriptive Statistics A total of 386 out of 613 firm-years (63%) released non-gaap earnings (Table 1, Panel A). For these firms, the mean (median) statutory Net Profit after Tax (NPAT) was $ ($76.5) million, compared with firms mean (median) non-gaap earnings (NonGAAP_FS) of $ ($146.95) million and analysts mean (median) non-gaap earnings (NonGAAP_AA) of $ ($126.20) million (Table 1, Panel B). On average, firms and analysts non-gaap earnings are higher than NPAT. This result is not surprising, as the most frequent adjustments are to add back expenses and losses thus increasing profit. On average, companies record a higher maximum non-gaap earnings and lower minimum non-gaap earnings than analysts, that is, companies are more optimistic in their estimates of non-gaap earnings than are analysts. <Insert Table 1 about here> On average, NPAT, companies non-gaap earnings (NonGAAP_FS) and analysts non-gaap earnings (NonGAAP_AA) are highest in the later period (i.e., POST_CRISIS) at $ million, $ million and $ million, respectively. The difference between companies non- 23

25 GAAP and GAAP earnings (NG_DIFF) is largest during the CRISIS year ($ million). Median NPAT varies significantly over the period (Z = 7.01, p < 0.01) while companies and analysts non-gaap earnings (NonGAAP_FS, NonGAAP_AA) do not, indicating that non- GAAP earnings provide a more stable, and thus potentially more useful, measure for analysts. <Insert Table 2 about here> Table 2 presents descriptive statistics for the six groups of items included in non-gaap earnings. Panel A shows the amounts for the six groups in profit and loss in the statutory financial statements for the full sample (n=613). Panel B provides descriptive statistics for the 338 firms providing non-gaap reconciliations. It shows the amount for the six groups of items in the statutory profit and loss; in the reconciliation statement in the firm s statutory financial statements or investor presentation; and adjustments as measured by analysts. For the full sample, the profit and loss items (Panel A) show firms have an average incidence of groups of items of 2.68 (with a median of 3 and a maximum of 6). The mean (median) of the total of the six groups of items (MAGNIT) is -$228 million (-$25 million). For the sample of firms with non-gaap earnings (Panel B), the profit and loss item groups in the statutory accounts show firms have an average of 3.05 item-groups and the mean of the total of the six groups of items (MAGNIT) is -$261.8 million. For reconciliation items, firms have an average of 2.5 item-groups and the mean of the total of the six groups of items (MAGNIT) is -$ million. For analyst adjustments, the average number of adjusting item-groups is 1.94 and the mean of the total of the six groups of items (MAGNIT) is -$ million. Comparing 24

26 the firms reconciliation amounts with analyst adjustments, we see that firms make more adjustments than analysts and the adjustments are of greater magnitude. Both firms and analysts make substantial adjustments to profit and loss remeasurement items (relating to financial, revalued and impaired assets), that is, on average a larger proportion of the revenue/expense shown in the profit and loss statement is adjusted. On average, analysts adjust only some of the remeasured and non-recurring items included in firms GAAP earnings. However, some firms and analysts make no adjustments for some items, as shown in the median adjustment for many items being zero. Interestingly, many firms include amortisation as a non-gaap item, while analysts are less likely to do so. 13 Table 2 (Panel A) shows the largest average value of the financial statement item-groups is IMPAIR (mean -$95 million, median -$0.18 million) followed by OTHER (mean -$43 million, median $0) and AMORT (mean -$32 million, median -$4.75 million). The largest average itemgroup in firms reconciliation statements is IMPAIR (mean -$59.15 million, median $0) followed by OTHER (mean -$29.89 million, median $0) and REVAL (mean -$28.28 million, median $0). The largest average item-group in analyst adjustments is IMPAIR (mean -$80.31 million, median $0) followed by REVAL (mean -$26.86 million, median $0) and FININST (mean -$10.32 million, median $0). 13 Panel B1 financial statement items show the descriptive statistics for data based on the average value for the item for each firm. That is, in FININST one value is included for each firm and it is a net value (gains and losses are offset). For B2 reconciliation items and B3 analyst adjustments, a firm or analyst may choose to adjust one or more of the items included the net amount in the financial statements. All reconciliation items and analysts adjustments are items in the financial statements, but they may have different minimum, maximum, mean and median value because they are net items in the B1 financial statement category and may be individual items in the B2 reconciliation items and B3 analyst adjustments. 25

27 On average, firms record net losses (or net expenses) on all six item-groups (FININST, IMPAIR, AMORT, REVAL, MERGER and OTHER) in the financial statements and in the reconciliation statements. Similarly, analyst adjustments are negative for all item-groups except OTHER (other non-recurring items). A negative item (i.e., an expense or loss) increases profit when it is added back by firms and analysts while a positive item (i.e., a revenue or gain) decreases profit when it is removed. Thus, on average, the firms and analysts adjustments for remeasurement items and the non-recurring item MERGER increase profit. Stated another way, firms and analysts considered the downward remeasurement items in the accounts were too large and some portion of them was added back, increasing profit. <Insert Table 3 about here> Table 3 shows the mean value for the six item-groups for non-gaap earnings firms and other firms (i.e., firms that do not release non-gaap earnings) based on financial statement items (Panel A) and analyst adjustments (Panel B). Comparison of means generally are not significant except for MERGER in Panel A and FININST and AMORT in Panel B. The data show a positive relationship between non-gaap earnings disclosure and larger amounts of expenses in the financial statements. Comparison of the number of remeasurement and non-recurring items recorded by non-gaap reporters and other firms shows that the two groups are significantly different for all six itemgroups in relation to the number of firms with the items (Panel A) and the number of firms for which the analysts adjust the item (Panel B). Table 3 also shows that when a firm provides non- 26

28 GAAP earnings, analysts are more likely to record adjustments for the remeasured and nonrecurring items. If a firm has more items, then analyst adjustments are more likely for that firm. Data in Table 3 could be interpreted to suggest that non-gaap earnings disclosures bring analysts to focus on the remeasured amounts. <Insert Table 4 about here > Table 4 reports descriptive statistics for the other explanatory and control variables used in the regression models. There are significant differences between the non-gaap and other firms in the control variables (untabulated). Mean and median values show that non-gaap earnings disclosure firms are larger (SIZE), have higher analyst following (FOLLOW) and greater variability in cash flows from operations (VARCFO). They are more profitable (NI) on average, but include a higher proportion of LOSS firms. The non-gaap earnings disclosure group includes a higher proportion of financial industry firms and a lower portion of mining sector firms. Somewhat surprisingly, firms releasing non-gaap earnings have a lower mean change in earnings (ACHEARN). <Insert Table 5 about here > Table 5 reports descriptive statistics for AFE and FD in the year following the release of the non- GAAP earnings (year t). As expected, mean (median) AFE of (0.011) is higher at nine months before year end t+1, compared to three months and the amounts are significantly different between the periods (F = 3.34, p < 0.05; Z = 24.97, p < 0.01) for the pooled sample. 27

29 Mean FD of 0.01 at nine months is lower than at three month and again amounts are significantly different between the periods (F = 4.38, p < 0.05). At nine months, mean AFE (FD) is (0.006) in the PRE-CRISIS period, (0.014) in the CRISIS year and (0.011) in the POST-CRISIS period suggesting it was more difficult to forecast earnings in the year following the October 2008 financial crisis Factors associated with non-gaap earning disclosure Table 6 presents results for Equation 1, which tests the relationship of non-gaap earnings disclosure and profit or loss items relating to remeasurement and non-recurring items. Recall that there are four groups of items classified as remeasurement (FININST, IMPAIR, AMORT and REVAL) and two as non-recurring (MERGER and OTHER). We predicted greater incidence (H1a) and larger amounts (H1b) of these six items would be associated with the release of non- GAAP earnings. <Insert Table 6> Panel A presents results for incidence and amount of the items in firms financial statements. Model 1 shows firms with a great incidence of the items (COUNT) are more likely to release non-gaap earnings, providing support for H1a. Model 3 shows that dummy variables for incidence of IMPAIR, REVAL and MERGER items are positively associated with non-gaap earnings disclosure. That is, firms with these items are more likely to provide non-gaap earnings. Hypothesis 1b is not supported: total amount of the six items (MAGNIT) is not associated with non-gaap earnings disclosure (H1b) (Model 2). We interpret the significance of 28

30 COUNT but not MAGNITUDE as consistent with information communication motivations and not supportive of an opportunism explanation. If opportunism was a major incentive, we would expect the magnitude of the adjustments to be an explanatory factor for non-gaap earnings release because the magnitude of the adjustments affects the amount of the non-gaap earnings presented. Additional models (untabulated) investigating the importance of the magnitude of individual items show that the amount of the remeasurement items (FININST, IMPAIR and REVAL) are not associated with non-gaap release. Panel B provides similar results for analyst adjustments. Firms with a greater incidence of the items (COUNT) are more likely to release non-gaap earnings (Model 4) while total magnitude of analyst adjustments is not associated with non-gaap earnings release. Firms with analyst adjustments for REVAL, MERGER and OTHER (based on dummy variables recording incidence) are more likely to provide non-gaap earnings disclosure. The results are observed after controlling for factors that could also explain the release of non- GAAP earnings. The significance of these control variables is similar between the six models models and for the financial statement items analysis and the analyst adjustments analysis (Table 6, Panel A and B respectively). 14 Industry and time period do not appear to affect the dependent variables as might be expected. Financial sector or mining sector firms are not more likely to release non-gaap earnings nor is their release associated with the time period (pre- or postcrisis). Release of non-gaap earnings is not associated with firm size or having a larger change in earnings. Firms with more variable earnings are more likely to make a release, but loss firms 14 We explore the relationship of financial statement items and analyst adjustments and release of non-gaap earnings. We do not include reconciliation items in this table, because they are by definition associated with the disclosure of non-gaap earnings. 29