Non-GAAP measures in Australia
Name: Kevin Walsh Student number: 11419741 Thesis supervisor: Mr. Jullens Date: June 25, 2018
Word count: 11285, 0
MSc Accountancy & Control, specialization Accountancy Faculty of Economics and Business, University of Amsterdam
Statement of Originality
This document is written by student Kevin Walsh who takes full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
Non-GAAP measures are increasingly being used in companies all over the world. According to managers, these measures are being used, to provide extra information to investors. However some regulators have concerns that these measures are being used in order to mislead investors or to produce Favourable numbers. Regulators and the various Accounting Standards Boards are trying to evaluate current non-GAAP measures, with a view to protecting investors. In the United States the SEC (Securities & Exchange Commission) has already intervened and implemented Regulation G, to try to improve the quality of non-GAAP measures. In Australia the Australian Securities and Investments Commission (ASIC) introduced Regulatory guide-230. This was introduced to promote full and clear disclosure of non-GAAP measures submitted by companies in Australia. This study investigates the claim that regulatory guide-230 improved the quality of non-GAAP measures in Australia. First I test to see if there is a change in the percentage difference between GAAP and non-GAAP earnings before and after the regulation-230 using a sample collected from various companies listed on the ASX 100. Then I consider the persistence number of non-GAAP earnings before and after regulation. I then test to see if non-GAAP numbers have higher earnings persistence than GAAP numbers. Higher persistence number means that it is of higher quality and is better in predicting future cash flows. Lastly I examine the data to see if non-GAAP numbers are more value relevant than GAAP numbers. I do this with an incremental value relevance test. An accounting term that has a significant association with share prices indicates that the information is relevant and reliable enough to be of value to investors.
Contents 1 Introduction ...5 2 Literature review ...8 2.1 Financial report ...8 2.2 Disclosure ...8 2.3 Agency theory ... 10 2.4 Non-GAAP measures... 10
2.5 Advantages of non-GAAP measures ... 11
2.6 Disadvantages of non-GAAP measures ... 13
2.7 Different non-GAAP research settings ... 13
2.8 Non-GAAP regulations... 15
2.9 Regulation in Australia... 16
3 Hypothesis development ... 18
4 Research design ... 20
4.1 Research method for hypothesis 1 ... 20
4.2 Research method for hypothesis 2 ... 20
4.3 Research method for hypothesis 3 ... 21
5 Results... 22
5.1 Descriptive statistics... 22
5.2 Non-GAAP measures in changing regulations ... 23
5.3 Persistence in non-GAAP and GAAP measures... 24
5.4 Value relevance in non-GAAP measures ... 26
6 Conclusion ... 29
7 Bibliography ... 31
1 Introduction
In recent years more and more listed companies supplement their financial statements disclosures with non-GAAP earnings. These are alternative methods to measure the earnings of a company. Many companies disclose these numbers in addition to their earnings calculated through generally accepted accounting principles (GAAP).
According to Audit Analytics 449, S&P 500 companies (88%) presented at least one non-GAAP metric between July and September of 2015. New analysis shows that 96% of the S&P 500 presented non-GAAP data in their earnings releases during Q4 2016. The increase of these types of disclosures in the United States got the attention of the SEC (Securities and Exchange Commission). According to SEC Chair Mary Jo White, the increasing prevalence of non-GAAP reporting “deserves close attention” and “may be a source of confusion.” This increase in the submission of non-GAAP data was seen not only in the United States, but also in Europe and Australia. Over the last couple of years the use of non-GAAP measures increased within Australian companies. In 2000 around 22% of companies listed in the ASX 500 used non-GAAP data, in 2014 around 42% (Coulton, Ribeiro, Shan, & Taylor, 2016). The Australian regulators wanted to ensure that non-GAAP measures were not used to mislead investors. So in December 2011 the Australian Securities and Investments Commission (ASIC) issued Regulatory Guide-230 Disclosing non-IFRS financial information. The goal of this guide is:
The purpose of our guidance is to promote full and clear disclosure for investors and other users of financial information and minimise the possibility of those users being misled by such information.
This was a big step in regulating non-GAAP earnings. Australia is together with the USA one of the few countries that has some sort of regulations regarding non-GAAP measures. South-Africa alone has full regulations for non-GAAP measures. Because so many studies focus on the USA, Australia seems like a good alternative candidate for this research. Research is difficult in Australia because of the absence of machine-readable data. Most of the studies done in Australia are performed with hand collected data, as is this analysis.
Some managers believe that the non-GAAP earnings provide a more accurate measurement of the company’s financial performance than regular GAAP earnings. This statement is confirmed by various studies. There are also different studies that researched opportunistic reporting by managers who used non-GAAP reporting in such a way as to convey more
favourable numbers. Not everyone is positive about non-GAAP measures. Some say that these are used to boost stock prices in quarterly reports.
“We are concerned that corporations are selectively reporting one-time and recurring items as non-GAAP financial measures that may make ‘core’ operations look more favourable and not disclosing one-time adjustments that would make ‘core’ operations look worse,” explained Tony Sondhi, who serves as the co-lead of the PCAOB’s Investor Advisory Group (IAG) and is also a member of the FASB’s Emerging Issues Task Force (EITF) (Weng, 2017).
The use of non-GAAP earnings is a form of providing extra disclosure. Companies provide non-GAAP earnings next to their regular earnings. They report these non-GAAP earnings by removing large one-off costs such as write-downs. They do this because managers think these costs should not be considered normal operating costs and that they can distort the true financial performance of a company. Companies say that they provide more disclosure to reduce information asymmetry and in this way give more value to their numbers, thus being more informative. Non-GAAP measures can provide extra information which investors can use to predict future cash flows or the functioning of the company’s core business. Countries such as the United States wanted regulations for non-GAAP measures. The SEC introduced Regulation G in 2002. This regulation established rules for non-GAAP metrics, such as pro-forma earnings. These rules require, among other things, that non-GAAP inpro-formation be presented with the same prominence as GAAP information. The emphasis was placed on pro-forma inpro-formation by the SEC (Bowen, Davis, & Matsumoto, 2005).
Non-GAAP measures are not regulated in Australia. But however there is the new guideline introduced in 2011 mentioned above (Regulatory guide 230). This guideline was introduced by the ASIC. The guide was introduced to promote full and clear disclosure for investors and other users of financial information. This was to reduce the chance of investors being manipulated. The SEC introduced similar regulations in the United States and several studies have been done to investigate the impact of these regulations (Heflin & Hsu, 2008).
This study will investigate whether Regulatory guide-230 has impact on non-GAAP measures disclosed by companies in the ASX 100. The ASX 100 is a stock market index which represents Australian large and mid-cap securities. This study will investigate whether non-GAAP measures give more value-relevant information prior to and following the introduction of Regulatory guide-230
The research question will be: Did Regulatory guide-230 have an impact on non-GAAP measures disclosed by Australian listed companies?
This research question tests the signalling hypothesis that managers disclose non-GAAP earnings in order to convey extra information to users. Non-GAAP measures can be used to provide stakeholders extra information about the company. This can be helpful for investors but also for other stakeholders such as employees, banks or even Governments. Non-GAAP measures could therefore reduce information asymmetry and help with agency problems. The downside of non-GAAP is that managers can use them to provide misleading numbers to investors and stakeholders. Since non-GAAP measures are mostly not audited, managers can be creative with these numbers. So there are potential upsides and downsides to non-GAAP measures. The aim of the regulatory guide-230 is to improve the quality of non-GAAP measures and make it more understandable for investors. This research paper will test if the above is the case.
This research paper adds to the literature of non-GAAP measures. There has been a lot of research done on the signalling and the manipulation hypotheses of non-GAAP measures. Australia has had little research done into non-GAAP measures. There is also little to no research done, studying the impacts of new regulations conforming non-GAAP measures in Australia. The introduction of the regulatory guide gives a unique opportunity studying the effects of regulations on non-GAAP measures in Australia. The analysis presented in this report should help to illuminate the functioning of the 2011 regulation. Also some research has been done as to whether regulator intervention has helped non-GAAP quality, whether if they improve the predictability of future cash flows or if these numbers give more value relevant information. This kind of research has mostly been done in the United States, but not, to my knowledge, in Australia. The main question will answer if and how regulations help with non-GAAP measures.
The next section will be the literature review explaining different components of non-GAAP, usefulness, downsides and other regulatory backgrounds from different countries. Also it will explain the Australian context. After that there will be the hypothesis development and research design, explaining how my research will be conducted. The results will follow with a conclusion.
2 Literature review
This section will discuss the several aspects affecting this study. First I will discuss the parts of financial reporting that are affected by non-GAAP measures, the financial report and disclosures. Also I will briefly discuss the agency theory. After that I will discuss what non-GAAP measures are and the pros and cons of non-non-GAAP measures. I will end the literature review with examples from other studies that covered non-GAAP measures in general and how non-GAAP measures are regulated in Australia.
2.1 Financial report
The financial report is a way of a company to interact with the outside world. It can be used to attract equity and communicate to its stakeholders how the company is doing. The challenge in any economy is the optimal allocation of savings to investment opportunities (Healy & Palepu, 2001). Companies want to attract money and/or savings from households or other institutions to fund their business. Business and savers most likely want to do business with each other but matching savings to investments can be difficult. This is because of two reasons; the first problem is that managers and companies have better information than the side that wants to invest. Managers tend to overstate their businesses to attract more investments. This is called an information problem. The second problem is the Agency problem. The Agency problem is present because investors do not play a role in the management of their funds. The responsibilities of managing the investor’s money lie in the hands of the company management. Management can have different ideas on how to maximize profit that could hurt the funds of the investor. For example; management sees an opportunity into buying a stake in a company that could lead to huge profits, but there is also a huge risk involved losing the money. The investor could lose all his money.
2.2 Disclosure
Disclosure is an important aspect of a company. It is critical for an efficient capital market. Firms can provide disclosure through various channels, for instance footnotes in financial statements, management discussions and other regulatory findings. A Non-GAAP measure is also a way of disclosing extra information to shareholders. These are voluntary disclosures. Other examples of voluntary disclosures are analyst’s presentations, press releases, internet and other (corporate) reports. (Healy & Palepu, 2001) argues that demand for financial reporting and disclosure arises from information asymmetry and agency conflicts. They find
that disclosure credibility is enhanced by auditors, Accounting Standards Boardsand/or regulators. For instance America has the SEC with Regulation G. Australia has the new Regulatory Guideline 230 provided by the ASIC. Both regulators try to enhance credibility of the voluntary disclosures. According to (Healy & Palepu, 2001) there are six forces that affect the managers’ propensity to add disclosures to the financial reports. These forces are; capital market transactions, corporate control contests, stock compensation, litigation, management talent signalling and proprietary costs. The first force, capital market transactions is about how the manager has a lot of information and he wants to reduce information asymmetry to reduce the cost of capital. To do this he provides voluntary disclosure so that he can also reduce information risks. The second force, corporate control, is how managers disclose information to stakeholders when the firm is not functioning well. Managers are responsible for the company and its stock performance so they should be able to explain why the company is not doing well. Stock compensation is the third force. Managers can be rewarded for their work by receiving stock. Managers are willing to disclose information to increase the liquidity of the stock of the company. Managers also have the incentive to disclose this information to reduce the contracting costs associated with stock compensation. The fourth force is litigation costs. Manager’s disclosure decisions can be affected by legal actions against mangers for late disclosure or bad disclosure. This can lead to more voluntary disclosures provided by managers. The second is litigation risks that come with voluntary disclosure. Litigation risks can make managers less inclined to add voluntary disclosures, they also tend to give less forward looking information. The fifth force is management signalling. (Healy & Palepu, 2001) argue that talented managers have an incentive to make voluntary disclosures earnings to reveal their type. When investors think managers have a good read on the company’s environment, investors think the manager can anticipate future changes, thus increasing the firm’s market value. The last force of (Healy & Palepu, 2001) is proprietary costs. Firms can decide not to disclose information to investors that can damage the company’s competitive position. Another study carried out to investigate what manager’s different incentives were for disclosing voluntary disclosures was done by (Berger & Hann, 2007), They also found similar incentives to the research of (Healy & Palepu, 2001). The study of (Berger & Hann, 2007) found two incentives from not disclosing them. (Berger & Hann, 2007) studied that managers withheld voluntary disclosures because of the high agency and proprietary costs. They argued that managers would be less inclined to disclose items if the costs should exceed the benefits. Managers were also afraid of providing competitors with too much information.
2.3 Agency theory
Managers claim the non-GAAP measures are disclosed to provide more information to investors and stakeholders. The agency theory is a theory that explains the relationship between principals (shareholders, investors) and agents (managers). The agency theory is concerned with resolving problems between these two. According to (Eisenhardt, 1989), the key idea of the agency theory was; principal-agent relationship should reflect efficient organization of information and risk-bearing costs. According to Eisenhardt there are two problems the agency theory tries to resolve. First is the problem that occurs when the principals and agents have different goals and objectives. The agent and principals can have differing views on how the company should be run. The second problem that occurs is the problem of risk sharing. This can happen when the agent has personal incentives to gain as much profit as possible. This can make the agent more risk tolerant, possibly damaging the company with bad investments. The principal can limit divergence from his interests by establishing appropriate incentives for the agent and spend money on monitory costs to limit opportunistic behaviour (Hill & Jones, 1992). The agent can spend bonding-costs to guarantee that he will not take certain actions that would harm the principal. Despite these interventions there can still be divergence between the agent and principal. The welfare of the principal that is lost to this is called residual loss. Agency costs are the sum of the principal, the monitoring costs, the bonding costs and any residual losses.
2.4 Non-GAAP measures
Non-GAAP measures, as said before, are not new. They are increasingly common with companies. This research will look into non-GAAP earnings. Examples of these are; EBITDA, cash earnings, operating earnings and adjusted operating income. Managers can exclude certain expenses to create non-GAAP disclosures. The manager may think that this will give a more informative number than regular GAAP disclosures. For instance, in the press release of Netflix 2009 second quarter income statement, Netflix disclosed a GAAP net income of $28.7 million. They disclosed a non-GAAP net income of $34.4 million. Management of Netflix believed that non-GAAP net income is a useful measure of operating performance because it excludes the non-cash impact of stock option accounting. Netflix also discloses a free cash flow measure, which is also non-GAAP. Netflix believes that free cash flow is a useful measure of liquidity because it excludes the non-operational cash flows from purchases and sales of short-term investments, cash flows from investment in business and
cash flows from financing activities. Non-GAAP measures are not only recorded in press releases. It is also disclosed in Annual statements. In the annual statement of Rio Tinto from 2009/2010, they disclose underlying earnings of 14.2 billion and statutory earnings of 13.8 billion. The reason Rio Tinto discloses underlying earnings is as follows:
Underlying earnings is a measure that provides insight into the underlying business performance of the Group’s operations and is the key financial performance indicator used across the Group. This KPI provides insight to cost management, performance efficiency and production growth. It is therefore an indicator of financial and operational excellence and growth.
According to a study of (Whipple, 2015), the difference between non-GAAP earnings and GAAP disclosures are in the exclusion of “special’’ and “other’’ items. According to Whipple, special items are disclosed to make numbers more understandable. With other items it is not clear why they are excluded. The study of (Bradshaw & Sloan, 2002) documents an increase in non-GAAP measures. They also document a very strong bias toward the reporting of a non-GAAP number that exceeds the GAAP earnings number. They show that the market response to the non-GAAP earnings number has displaced GAAP earnings as a primary determinant of stock prices. Finally (Bradshaw & Sloan, 2002) show that management has taken a proactive role in defining non-GAAP numbers with analysts and investors.
2.5 Advantages of non-GAAP measures
Statements are made that non-GAAP measures provide more informative numbers. This was confirmed in several studies (Veenman & Leung, 2017; Black, Christensen, Ciesielski, & Whipple, 2017). The study of Veenman & Leung illustrated that in loss firms non-GAAP measures are highly informative about future performance. Also the study showed that loss firms reporting non-GAAP profits have better future performance than GAAP-only loss firms and that they are not overvalued. (Black et al, 2017) studied how the discretion afforded in non-GAAP reporting influences earnings consistency. They provided evidence that companies modified their non-GAAP calculations over time to provide more informative earnings metrics. This is in contrast to the concern that managers inconsistently calculate non-GAAP earnings to obscure financial performance. Also the study of (Bhattacharya N. , Black, Christensen, & Larson, 2003) showed that non-GAAP pro forma earnings have information relevant of core earnings for investors disclosed in financial statements. (Brown & Sivakumar, 2003) found that non-GAAP earnings provide more
relevant information than regular GAAP measures. In the research of (Brown & Sivakumar, 2003) they compared the value relevance of GAAP net income with operating earnings. A number is value relevant when it can predict future returns/earnings or the future value of a company. They found that operating earnings reported by managers and analysts are more value relevant than a measure of operating earnings derived from firm’s financial statements, as reported by Standard and Poor’s, because the reported earnings from managers and analysts had fewer transitory components than the reported earnings in the annual statements. Transitory components are mostly one-time gains, loss or expenses on the income statement that is nonrecurring. It is not a part of the company’s on-going business. When a company has less transitory components it is considered more value relevant (Brown & Sivakumar, 2003). The research of (Curtis, Mcvay, & Whipple, 2014) also came to the conclusion that non-GAAP measures are provided to inform investors in the presence of transitory gains. However, a significant proportion of firms appear opportunistic in that they only disclose non-GAAP earnings information when it increases investors’ perceptions of core operating earnings. Another research which studied the effects of non-GAAP earnings on investor perspectives is the study of (Frederickson & Miller, 2004). They did an experiment that examined the effect of pro forma earnings disclosures on the judgements of analysts, divided into sophisticated investors and nonprofessional. Sophisticated investors have more knowledge, more experience and more sophisticated tools available to them than nonprofessional investors. The results indicated that nonprofessional who received an earnings announcement that contained GAAP disclosures assessed a higher stock price than did nonprofessional who received an announcement containing only GAAP disclosures. Professional investors did not change their assessment. The pro forma disclosure did not cause nonprofessional investors to assess a higher earnings number for determining a stock price, but rather caused nonprofessional to perceive the earnings announcement as more favourable, which in turn caused them to convert earnings or some other performance metric into a higher stock price. This effect appears to be due to unintentional cognitive effects, rather than non-professionals relying on pro forma earnings information because they perceived it to be informative (Frederickson & Miller, 2004). (Schwartz & Johnson, 2005) also studied if investors look differently at non-GAAP measures disclosed in pro forma announcements. The results of their research cast doubt on the notion that investors are misled by pro forma earnings. This contradicts the views of regulators who keep warning investors that non-GAAP measures can be misleading.
2.6 Disadvantages of non-GAAP measures
Other reasons to use non-GAAP earnings are to manipulate investors. They report opportunistically to give a better picture of themselves to investors. (Bhattacharya N. , Black, Christensen, & Mergenthaler, 2004) did a study that shows that non-GAAP reporting seems to increase when firms have share price declines and earning declines. They also show that non-GAAP reporting increases when there are analysts’ predictions to meet. (Bowen, Davis, & Matsumoto, 2005) showed that when firms issued press releases, they put emphasis on the performance measure which portrays a more attractive performance. So non-GAAP earnings are disclosed at the beginning of press releases to emphasise a positive performance when GAAP earnings would fall short. In 2001 the SEC send out a warning where they underline the importance that investors should be cautious with non-GAAP/pro-forma earnings. They said that pro-forma financial information, by its nature, departs from traditional accounting conventions; its use can make it hard for investors to compare an issuer’s financial information with other reporting periods and with other companies (Katz, 2001). The paper of (Johnson & Schwartz, 2005) investigated the claim that investors are misled by pro-forma (non-GAAP) measures in the United States. They came to the conclusion that investors are not being misled by these measures. In the research of (Black E, Christensen, Joo, & schmardebeck, 2016), they studied how real earnings management and accruals management influences the probability that a company will disclose non-GAAP measures in its earnings press release. They found that when solid operating performance alone allows firms to meet expectations, managers meet expectations using real and accruals management, they are significantly less likely to report a non-GAAP measure. They also found that when companies were not performing as expected, companies are more likely to employ non-GAAP reporting. This suggests that the timing and costless nature of non-non-GAAP reporting allows managers to meet expectations on a non-GAAP basis when GAAP earnings do not meet the forecasts. Finally they found that companies report non-GAAP earnings when they are unable to use real or accruals earnings management, are constrained by prior-period accruals management, and the companies operating performance is poor.
2.7 Different non-GAAP research settings
The research is set in Australia; this is because there is little research done in this setting. This paragraph will show the different studies that are already done about non-GAAP measures and where they have been done. A lot of research has already been done in the
United States. (Heflin & Hsu, 2008) studied the impact of SEC regulations on non-GAAP measures. They came to several conclusions. Declines in the frequency of special and other item exclusions, decline in exclusion magnitude, decline in the probability that disclosed earnings meet or beat forecasts and a decline in the association between return errors and forecast errors. Their research suggests that regulations reduce the use of non-GAAP disclosures they also reduce the willingness of firms to use non-GAAP earnings to convey permanent earnings. There is also some research done in Europe. For instance the paper of (Isidro & Marques, 2012) looked into a sample of large European firms to investigate the influence of countries’ institutional and economic factors on manager’s GAAP disclosures. They found that managers are more likely to use non-GAAP measures to meet earnings benchmarks than GAAP earnings in certain countries. This would be the case in countries with efficient law and enforcement, strong investor protection, developed markets and good communication (Insidro & Marques, 2012). Another paper (Isidro & Marques, 2013) based on the European market found that board directors who are linked to a firm’s market performance are associated with a higher probability of disclosure of non-GAAP figures in the press releases of earnings announcements. This can be associated with opportunistic disclosure. In the Australian setting there are a couple studies such as the research of (Cameron & Percy, 2011) who studied if non-GAAP measures were used in Australia and how often. They did their analysis on a sample of the 50 biggest Australian mining companies. They showed that the most frequently used disclosed metric variants were EBIT, EBITDA or NPAT but with adjustments, calling them underlying, adjusted or normalised. The most common was underlying earnings. The paper showed that more emphasis was set on pro-forma earnings but it is not clear if the purpose was to be more informative or to be opppertunistic. (Johnson, Majella, Peta, & Robyn, 2014) carried out research to determine whether the disclosure of non-GAAP earnings affects information choices of users of financial statements with less knowledge than an analyst. (Johnson et al, 2014) suggested that when non-GAAP disclosure is provided, investors tend to select non-GAAP information rather than GAAP earnings to identify probability. (Rainsbury, Hart, & Malthus, 2013) carried out research in New-Zealand concerning non-GAAP measures. They studied the motivation for the increased reporting of non-GAAP earnings by New Zealand companies following the adoption of International Financial Reporting Standards (IFRS). (Rainsbury et al 2013) came to the conclusion that non-GAAP earnings are better predictors of future earnings and contribute value to analysts’ forecasts.
2.8 Non-GAAP regulations
Since 2011 there is a new set of regulations in Australia that companies have to follow with non-GAAP measures. This is called Regulatory Guide 230- Disclosing non-IFRS financial information. The major requirements of this regulation are the prominence given to non-GAAP earnings, the terminology used by firms the disclosure of a detailed reconciliation between GAAP and non-GAAP earnings and the consistency across time of the adjustments made by managers (Marques A. , 2017). Non-GAAP measures are voluntary and regulated in Australia and the United States. Listed firms in Europe have to consider only a set of recommendations. In Australia and the United States there is more scrutiny on non-GAAP measures. The United States has Regulation G, which was implemented in 2002. A research studied this new Regulation in the United States and at how new regulation could impact non-GAAP disclosures. Regulation G was introduced by the SEC and the research of (Marques A. C., 2005) studied the impact on investors. She found out that after the SEC’s intervention there was a decrease in the probability of disclosure of non-GAAP financial measures and this decline accelerated after more SEC intervention. She also found that investors do not value firms more or less because of the disclosure of non-GAAP financial measures. Another study done in the United States, studying the impacts of SEC interventions (Kolev, Marquardt, & Mcvay, 2008) examined the effects of intensified scrutiny over non-GAAP reporting on the quality of non-GAAP earnings exclusions. They found that, on average, exclusions were of higher quality following intervention by the SEC. (Kolev et al, 2008) also found that firms that stopped releasing non-GAAP earnings after the SEC intervention had lower quality exclusions in the pre-intervention period. The objective of the SEC, as well as ASIC, is to improve the quality of non-GAAP measures. The results of (Kolev et al, 2008) are consistent with the SEC’s objective. In South-Africa it is mandatory by law to supplement the financial statement with GAAP and non-GAAP disclosures. It is also subject to audit. South-Africa is the only country where this is regulated. The study of (Venter, Emanuel, & Cahan, 2014) looked into the South-African market and results showed that headline earnings and headline earnings exclusions have different incremental value relevance, which suggests that the mandatory disaggregation of headline earnings exclusions from GAAP earnings, and the separate disclosure, is useful to investors. In Europe the ESMA (European Securities and Markets Authority) introduced the ESMA Guidelines on Alternative Performance measures in 2015. The purpose of this guideline was to promote usefulness and transparency of alternative performance measures like non-GAAP. ESMA
said that it will also improve reliability, comparability and comprehensibility. This will help to protect investors from misleading numbers. According to ESMA this will benefit users and promote market confidence. Regulators are increasingly checking non-GAAP numbers, to protect investors. More regulations will probably come in the future. For instance, the IASB set up a study which began in December 2016: While the project is multi-faceted, we have tentatively decided to focus on improvements to the income and cash flow statements with much of our attention on the income statement (‘the P&L’). This is where we will start addressing non-GAAP earnings measures, often referred to as ‘alternative performance measures’ (APMs), such as income before non-recurring items (Kabureck, 2017). More regulations will probably come in the future or the South-African system may be used.
2.9 Regulation in Australia
In 2004, a supplementary amendment to AASB 101 Presentation of Financial Statements superseded AASB 1018 Statement of Financial Performance. It stated that ‘an entity shall not present any items of income and expense as extraordinary items, either on the face of the income statement or in the notes’ (AASB 101, paragraph. 78). As a result, after the adoption of A-IFRS, Australian reporting entities have been prohibited from separately disclosing either abnormal or extraordinary items (Coulton et al, 2016). On January first of 2005, Australia adopted IFRS. All publicly listed Australian companies are required to adopt this new standard. GAAP disclosures were heavily regulated, but non-GAAP disclosures were not. For instance under paragraph 73 of AASB 133 companies were allowed to disclose, in addition to normal Earnings per Share, amounts per share using a reported component of the statement of comprehensive income. After the adoption of IFRS, companies were increasingly using non-GAAP measures which caught the attention of regulators. In 2005 Consultation paper 69 (Disclosing pro forma financial information) was released by ASIC. This was done to provide guidelines on the production of non-GAAP information, its disclosure requirements and type of non-GAAP measures which could be published (Coulton et al, 2016). This was to prevent misleading non-GAAP measures. In March 2009 The Financial Services Institute of Australasia (FINSIA) and the Australian Institute of Company Directors (AICD) issued a paper to guide companies on how to use non-GAAP disclosures. This guide was non-mandatory and its aim was to make non-GAAP reporting more informative. They also wanted to make non-GAAP reporting more transparent and consistent and to make sure that non-GAAP reporting was not misleading to investors. The main aim of
this guide was to force companies to explain the use of Underlying Profit and to encourage companies to produce clear explanations of what the difference was between Underlying Profit and statutory profit. In 2011 Regulatory guide 230 was presented along with Paper 150 ‘Disclosing Financial information Other than in accordance with Accounting standards. The main aim of Regulatory guide 230 is to focus on the prominence given to non-GAAP earnings, the terminology used by companies like underlying profit, normalised profit and underlying result. There needs to be disclosure of reconciliation between GAAP and non-GAAP earnings where you can see the adjustments made in the non-non-GAAP earnings. There are some statutory obligations in Australia for the use of non-GAAP measures. There are requirements that the information should not be misleading. Mangers have to comply with the Corporations Act, where managers are obligated and liable to not give out information that is misleading. Non-GAAP measures are not required to be audited. However, under ASA/ISA 720, the auditors’ responsibility is to read and verify other information presented in documents containing an audited financial report. When there are material problems in the audited report under ‘other information’, auditors are required to make ‘another matter(s)’ disclosure (Coulton et al 2016).
3 Hypothesis development
The research question tests the signalling hypothesis that managers disclose non-GAAP earnings in order to convey all relevant financial information to users and to check if the new Regulatory Guide-230 has an impact on this. The aim of this new regulatory guide was to improve non-GAAP disclosures and to provide clear and full information to investors. Managers can have different incentives for disclosing non-GAAP measures. One of these is to be more informative to stakeholders which in turn will reduce information asymmetry. The other possibility is that managers wish to paint a more positive picture of their results and in this way, manipulate their stakeholders (Bhattacharya, et al, 2004). There are a couple of studies which studied the impact of regulations implemented for non-GAAP measures such as Regulation G in the United States (Kolev et al, 2008; Heflin & Hsu, 2008). The aim of Regulatory Guide-230 is to improve non-GAAP disclosure in order to promote full and clear disclosure, thus enhancing information quality and reducing information asymmetry. Therefore the first thing that will be tested is if Regulatory Guide 230 changes the way managers disclose non-GAAP measures. Secondly this research will test if non-GAAP measures provide more relevant information than statutory GAAP measures. Therefore the first hypothesis will be:
Hypothesis 1: Regulatory guide 230 changes the way managers’ report non-GAAP measures. The second hypothesis will look into the prediction value of non-GAAP earnings as seen in the (Veenman & Leung, 2017) study that Non-GAAP earnings were highly informative about future performance of loss making companies. This hypothesis will also take into account the new regulations of 2011 as well as the research of (Black et al 2017) which studies how the discretion afforded in non-GAAP reporting influences earnings consistency. The study provided evidence that companies modified their non-GAAP calculations over time to provide more informative earnings metrics. Managers and companies claim they disclose non-GAAP measures to be more informative to investors. Managers adjust certain GAAP numbers to non GAAP by deducting certain costs that they feel are not important. If this is the case, the information that comes from non-GAAP measures should be informative about future performance. According to (Dechow, Ge, & Schrand, 2010) one of the future metrics which can be measured is future cash flow with earnings persistence. A more persistent earnings number is considered to be of higher quality. The theory is that with persistent numbers it’s easier to predict future cash flows. With this in mind I will check if non-GAAP
earnings provide a more persistent number than GAAP numbers. Therefore the second hypothesis will be.
Hypothesis 2: Non-GAAP earnings provide a better prediction of future earnings than GAAP earnings after 2011.
Bhattacharya et al (2003) showed that non-GAAP earnings provide information relevant to investors, in financial statement disclosure. This was also in the study Brown et al (2003), where he found that non-GAAP earnings provide more relevant information than regular GAAP measures. When considering the various studies, non-GAAP earnings seem to provide more value relevant information than GAAP earnings. Since reported earnings provide information relevant for equity valuation, I am going to examine if non-GAAP earnings provide more relevant information than GAAP earnings. This can be helpful for investors as they want to know if the company will deliver value over time. Regulatory Guide-230 is to help investors with their financial decisions, so non-GAAP earnings should be more value relevant or at least be seen to be improving in the direction of value relevancy, following implenmentation. The third hypothesis will examine if this is the case, i.e. are non-GAAP earnings more value relevant than GAAP earnings? I am going to test this with the market capitalization at balance date metric. Therefore the third hypothesis will be:
Hypothesis 3: Non-GAAP earnings provide more value relevant information than GAAP earnings after 2011.
4 Research design
The research will be an archival study. This study will examine the annual reports of a number of Australian listed companies. Since there is no single database available for non-GAAP measures in Australia, the data has to be manually collected. The time period covered is from 2009-2012. This time period has been chosen because it precedes the new regulation by two years and also covers a period of two years after the regulation. The data will be sought in the annual statements of companies listed in the ASX 100. In every annual report a word search will be carried out on the following words: “Adjusted”, “core earnings”, “cash earnings”, “normalized”, “underlying”, “EBITDA”, “recurring”, “operating profit” and “non-IFRS/GAAP”. These are the most common financial terms used in non-GAAP reporting. In the following tests these non-GAAP earnings will be compared with the GAAP earnings each year. For instance, when there is an underlying EBIT, this will be matched against the statutory EBIT. This process will be carried out for each of the four years only. Using this approach 44 observations were identified. Each observation represents a non-GAAP measure occurring in the period two years before and two years after the issuance of the new regulatory guide. Manual collection of the data has resulted in a small dataset; therefore the results may be less reliable than studies with a more extensive dataset.
4.1 Research method for hypothesis 1
To answer H1, I will test to see if there is a significant difference between the period before and after the introduction of the new Regulatory guide. The procedure that I will follow is to calculate the percentage difference between the GAAP and non-GAAP measure for each year and observation. I will divide the observations into two groups, such that the first group will contain the differences before the new regulation and the second group the differences after. Thereafter I will carry out a paired t-test on the means of the two groups. The t-test will show if a significant difference takes place following the introduction of Regulatory guide-230.
µDiff_before = µDiff_after
4.2 Research method for hypothesis 2
A more persistent earnings number is considered to be of higher quality. The underlying theory is that it will be easier to predict future cash flows of a firm which has more persistent or sustainable earnings. This in turn will make it easier to forecast cash flows for equity
valuation and reduce valuation errors (Dechow, Ge and Shrand, 2010). Earnings persistence is measured by the slope coefficient from a regression of future earnings (year t+1) on current year earnings (year t). For current year earnings regressions are repeated using GAAP and non-GAAP earnings. The earnings will be taken from the annual statements.
GAAPt+1AUSGAAP = β0+β1GAAPAUSGAAP+εit
GAAPt+1AUSGAAP = β0+β1NGEt+εit
Where:
GAAPt+1/tAUSGAAP = earnings under AUS GAAP at time t+1.
NGEt+1/t = non- GAAP earnings at time t+1.
εit = error term
The coefficient β1 measures persistence with a higher coefficient considered a more persistent
earnings stream.
4.3 Research method for hypothesis 3
In this paper we use an earnings and book value relation with market value to test hypotheses 2. An incremental value relevance test (Hung and Subramanyam, 2007) is used to examine the additional information that non-GAAP earnings provide above and beyond that provided by income reported under regular GAAP. This test was chosen because the variables can be easily found in the annual statements and Compustat.
MVt=β0+ β1NPATtAUSGAAP+β2(NGEt-EarningstAUSGAAP)+β3BVt+εt
Where:
MVt = Market capitalization at balance date
EarningstAUSGAAP = earnings under AUS GAAP
NGEt-EarningstAUSGAAP = the difference between the non-GAAP earnings under AUSGAAP.
BVt = book value of equity under AUSGAAP.
εt = error term
GAAP, non-GAAP and book value of equity are found in the annual statements. Market capitalization will be collected from Compustat by dividing the amount of shares at balance date with the shares’ closing price at balance date.
5 Results
5.1 Descriptive statistics
This study used manually collected data from annual statements. Some data was collected from Compustat. The research is carried out in the Australian context; this is because of the limited number of studies performed on non-GAAP measures in this region. As a result of the dearth of studies carried out in Australia, there is no single database where non-GAAP measures used by companies can be accessed. A sample had to be chosen which could produce enough variables to test and where time would not be an issue. The ASX 100 was chosen. There are 100 companies in this index with a mid to large market capitalization. A period of four years was chosen to make possible the production of sufficient data. I started with searching commonly used non-GAAP measures in the annual statements of 2009/2010 and 2011/2012 of all the companies listed on the ASX 100. In the end I had 36 companies with one or more non-GAAP measures I could use. The search process in the annual statements involved the collection of the most commonly used non-GAAP metrics. The following were found and used in this research;
- (Underling/normalized) Earnings before interest (EBIT): before significant items, excluding significant items.
- (Underling/normalized) Earnings before interest and tax (EBITDA): before/excluding significant items
- Underlying profit: Underlying profit is the statutory profit adjusted to exclude significant items.
- Adjusted earnings: excluding exceptional items such as finance costs.
- Underlying/normalized net profit after tax: net profit before/excluding significant items. - Operating profit before special items: adjusted profit before special/significant items. - Operating profit: An adjusted version of regular profit.
Companies are not always clear with what they mean with significant/special items. Mostly they explain themselves with lines such as: Underlying EBIT and NPAT performance reflect statutory results adjusted for individually significant items and other items that are relevant to a proper understanding of business performance.
Other variables that were found in the annual statements were the equity book value and the GAAP measures that were related to the non-GAAP measures found. Adjusted
EBIT/EBITDA is related to statutory EBIT/EBITDA. Underlying profit is related to statutory profit. Compustat was used to find the share prices and amount of shares needed to calculate the market capitalization at balance date.
5.2 Non-GAAP measures in changing regulations
The goal of the new Regulatory Guide-230 issued by ASIC is to make non-GAAP measures more clear and informative and to make it more difficult for companies to mislead investors with these measures. (Marques A. C., 2005) studied the impact on investors. She discovered that following an SEC intervention there was a decrease in the probability of disclosure of non-GAAP financial measures and this decline accelerated in proportion to the number of interventions. This paragraph will examine if Regulatory guide-230 has an impact on the non-GAAP measures companies provide. One would expect companies to adjust their non-non-GAAP measures because of this new Regulatory guide-230. The first hypothesis that I test is;
Hypothesis 1: Regulatory guide 230 changes the way managers’ report non-GAAP measures.
To test this I conducted a paired t-test on the means of the percentage differences between GAAP and non-GAAP measures before and after the new introduced guideline. The results can be found below in table 1.
Table 1, paired t-test of hypothesis 1
Variable Obs Mean Std. Err. Std. Dev. [95% Conf. Interval]
Before 86 -.3323 .7715 7.1544 -1.866 1.202
After 86 -.2716 .1887 1.7497 -0.402 0.348
T value= -0.3874 P value=0.6994
In this table we can see that the percentage difference between GAAP and non-GAAP before the regulation has a mean of 33%. This means that the average non-GAAP number was 33% higher than its reported GAAP number. This is in line with multiple studies which concluded that non-GAAP earnings would be higher than GAAP-earnings (Bradshaw & Sloan, 2002; Brown & Sivakumar, 2003). Also in the research of (Chen & Pevzner, 2012) they showed that when non-GAAP earnings were higher than GAAP earnings it could be an indicator for opportunism. Therefore you would expect that pre regulation non-GAAP earnings are higher than GAAP earnings, which is the case here. When non-GAAP numbers are higher than GAAP numbers, it does not necisarley have to indicate that the non-GAAP numbers are
opportunistic. After the regulation non-GAAP earnings were on average 27% higher than their GAAP earnings. This shows a small decline of 6% which could indicate that managers are taking steps to make non-GAAP more in line with GAAP measure. However, Since P(0.6994)>0.05, it can be concluded that there is not a statistically significant difference between the two groups, so H1 is rejected. The t-test shows that there is not a difference in percentage differences before and after the new regulation. So there is not a significant chance that the difference between non-GAAP and GAAP measures is affected by the new Regulatory guide 230. Although after the regulation, non-GAAP measures are on average 27% higher than GAAP earnings; this does not have to mean that non-GAAP measures are now more opportunistic. The aim of the new regulatory guide 230 was to make the non-GAAP measures more informative. There could be a chance that non-non-GAAP measures provided by companies before the Regulatory guide were already highly informative and in accordance with the new guide, so companies did not have to adjust their non-GAAP much after the guide was introduced. Since this research has a relative small sample due to the data had to be collected by hand, this could also affect the outcome of the t-test.
5.3 Persistence in non-GAAP and GAAP measures
Companies can disclose non-GAAP measures to give more information for potential investors. They want to know how the company will function over a longer period. However companies can also manipulate non-GAAP earnings for their own advantage (Bhattacharya et al, 2004). One of the aims of Regulatory Guide 230 was to protect investors from manipulative non-GAAP measures. In the research of (Kolev et al, 2008), they found that the intervention of the SEC improved the quality of non-GAAP numbers. Like the SEC, the goal of the ASIC is to improve the quality of non-GAAP measures for investors. A high quality number can be a measure that predicts the future cash flows of a company. This is helpful for investors, because they want to know how the company will do in the future. A more persistent earnings number is considered to be of higher quality. The underlying theory is that it will be easier to predict the future cash flows of a firm that has more persistent or sustainable earnings which will make it easier for forecasting cash flows for equity valuatio n and reduce valuation errors (Dechow, Ge and Shrand, 2010). You can assume that companies who provide earnings with high persistence are less likely to be trying to mislead their investors because of the fact that high persistent earnings are more value relevant. Earnings persistence is measured by the slope coefficient of a regression analysis of future earnings.
To test this, a regression analysis was carried out. The results can be found below in table 2. The second hypothesis is;
Hypothesis 2: Non-GAAP earnings provide a better prediction of future earnings than GAAP earnings after 2011.
Table 2 Regression earnings persistence
Total sample Sample before Sample after
Dependent
variable GAAP GAAP GAAP GAAP GAAP GAAP
Explanatory variables βNonGAAP 0.904 (0.000)** 0.989 (0.000)** 0.853 (0.000)** βGAAP 0.846 (0.000)** 0.884 (0.000)** 0.828 (0.000)** Constant -1.49 (0.137) 1.07 (0.285) -2.66 (0.009) 2.62 (0.010) -1.15 (0.252) -0.53 (0.597) N 168 168 84 84 84 84 F-value 742.68 (0.000)** 416.89 (0.000)** 3711.09 (0.000)** 294.41 (0.000)** 218.15 (0.000)** 178.99 (0.000)** Adj R2 0.8162 0.7135 0.9781 0.7795 0.7235 0.6820
* and ** highlight significance at the 0.05 and 0.01 levels where the level of significance is two-tailed
The tests have been applied to 3 samples (table 2). The “total sample”, including all 168 observations from 2009 to 2012, followed by the “sample before Regulatory guide 230” (2009-2010) and “sample after Regulatory guide 230” (2011-2012). The table shows the earnings persistence of GAAP earnings and GAAP earnings. In every sample the non-GAAP have a higher persistence number from which you can conclude that non-non-GAAP earnings are of higher quality. The results also show that non-GAAP earnings are of higher quality than their GAAP counterparts. You can conclude that companies in this sample providing non-GAAP earnings are doing this to give more value relevant information and are not trying to manipulate investors. However when you compare the before and after sample in table 2, there is a decline in the non-GAAP earnings persistence Beta. The non-GAAP metric taken after the introduction of Regulatory Guide-230 has a lower persistence number
than before its introduction. Non-GAAP measures following introduction of Regulatory Guide-230 continue to have a higher persistence number than GAAP earnings, but the persistence has declined in comparison to before the guides introduction. Looking at the results the conclusion should be to reject H2, as the persistence number of non-GAAP measures after the introduction of the guide is lower than before it was introduced. We can conclude that non-GAAP measures have higher persistence than GAAP numbers, but after the regulation the persistence declined. This contradicts what was expected. Regulatory guide-230 was introduced to make non-GAAP measures more value relevant for investors. You would expect that this would be the case here. Again the small sample size of this study could affect the results. Therefore, for now we reject H2. Non-GAAP earnings do not provide a better prediction of future earnings than GAAP earnings following the implementation of Regulatory guide-230.
5.4 Value relevance in non-GAAP measures
One of the key reasons managers use non-GAAP measures is to provide more information to investors (Black et al, 2017; Bhattacharya et a, 2003, Brown & Sivakumar, 2003). This extra information can help investors to decide whether or not they want to invest in the company. It also helps reducing information assymetry between the company and investor. Moreover, the aim of Regulatory Guide-230 was to improve the information that is given to investors through non-GAAP measures to prevent misrepresentation of the numbers by opportunistic managers. The last hypothesis will look into the equity value and reported earnings. An accounting amount that has a significant association with share prices indicates that the information is relevant and reliable enough to be of value to investors. The incremental value relevance test is used to test whether non-GAAP earnings provide extra information than GAAP earnings. This is done by performing a regression analysis with as dependent variable the market capitalization at balance date with explanatory variables; non-GAAP-GAAP earnings, Book value of equity and GAAP earnings. The explanatory variable non-GAAP-GAAP tells if non-non-GAAP-GAAP discloses more information relevant for investors. The results can be found in table 3, below. The last hypothesis is:
Hypothesis 3: Non-GAAP earnings provide more value relevant information than GAAP earnings since 2011.
The test is again divided into 3 samples; “Total Sample”, “Sample before” and “Sample after”. The “Total Sample” has 168 observations with all years included (2009-2012).
“Sample before” (2009-2010) and “Sample after” (2011-2012) both have 84 observations. The explanatory variables are the “book value of equity at balance date”, “market capitalization at balance date”, “GAAP earnings” and “non-GAAP minus GAAP earnings”. The last variable, “non-GAAP minus GAAP earnings” tells us if non-GAAP earnings provide more value relevant information than GAAP numbers. The results show that in all the samples the results are not significant in the context that non-GAAP earnings provide more information relevant for investors. The aim of Regulatory guide-230 was to improve non-GAAP measures disclosed by companies. The information given by non-non-GAAP measures has to be clear and full. You would expect that non-GAAP measures provide extra value relevant information after this regulation, but there is no signifcant evidence that this is the case. This goes against other studies who looked at interventions of external bodies like the ASIC, such as the research of (Kolev et al, 2008), where they found that the SEC intervention improved the quality of non-GAAP measures. . (Rainsbury, Hart, & Malthus, 2013) also did a research studying if non-GAAP measures provide more value relevant information. They did the same value relevance test as this study, but their setting was set in New-Zealand. They also did not get signifciant results. Unfortunately they did not give an explanation why this could have happened. Their sample had almost the same amount of observations as the sample used in this study, the study of (Rainsbury et al, 2013) had a sample of 272 observations. This study did a scaled version of this test, scaling the variables with the amount of shares, because this should give better results. (Rainsbury et al, 2013) also scaled their test with the amount of shares. A non scaled version of this test can be found in the appendix. In that version there are significant results that non-GAAP measures provide more value relevant information. However, since scaling makes the test more reliable we take the results of the scaled version. The test in this research shows that the introduction of the guide did not improve non-GAAP measures after it was implemented. A reason why this could be is that regulatory guide-230 made non-GAAP measures less value relevant for future values and made non-GAAP measures more information relevant for the current year. Also according to (Marques A, 2017) the major requirements of this regulations are the prominence given to non-GAAP earnings, the terminology used by firms the disclosure of a detailed reconciliation between GAAP and non-GAAP earnings and the consistency across time of the adjustments made by managers. The emphasis seems to lie on how managers come to these non-GAAP measures and explain themselves. Maybe there is not enough emphasis in Regulatory guide-230 about the value relevance and extra information non-GAAP measures can give to investors. The ASIC made this guide to protect investors against opportunistic mangers, so they made them
more informative how companies came to these numbers. But it seems these numbers do not give a lot of information anymore about future earnings and company value. The small size of the sample can also intervene with given results. But for now we reject H3. Non-GAAP earnings do not provide more value relevant information than GAAP numbers after the introduction of Regulatory guide-230.
Table 3 incremental value relevance test, scaled by number of shares Total sample Sample before Sample after
Dependent variable MarketCap MarketCap MarketCap
Explanatory variables BV 8.32 (0.000)** 8.12 (0.000)** 5.06 (0.000)** GAAP -0.47 (0.642)** -3.84 (0.000)** -0.58 (0.565)** (NonGaap-GAAP) -1.15 (0.252)** 0.45 (0.657)** -0.76 (0.449)** Constant 7.22 (0.000)** 1.39 (0.167)** 5.37 (0.000)** N 168 84 84 F-value 123.9 (0.000)** 99.85 (0.000)** 52.57 (0.000)** Adj R2 0.6832 0.7772 0.6454
* and ** highlight significance at the 0.05 and 0.01 levels where the level of significance is two-tailed MarketCap=market capitalisation at balance date scaled by amount of shares
BV= Book value equity at balance date scaled by amount of shares GAAP= GAAP earnings scaled by number of shares
6 Conclusion
Non-GAAP measures are increasingly being used by companies all over the world. The outlook is for increasing use of this accounting system. Regulating authorities are trying to standardize these measures in order to protect investors and other stakeholders. It is important for regulators to stay alert as non-GAAP measures can be used to mislead investors and make managers more opportunistic. Alternatively non-GAAP numbers can also be more informative for investors. Companies distribute extra information to reduce information asymmetry. Investors can use this information to aid their investment choices. Extra information can also be used to determine the future value of the company. The SEC introduced Regulation G in 2002 to improve non-GAAP measures. It helped, according to several studies, in improving the quality of non-GAAP measures (Heflin & Hsu, 2008). In Australia, Regulatory Guide-230 was introduced by the ASIC in 2011. The purpose of this guide was to promote the production of full and clear non-GAAP measures. This was to protect investors from misleading data produced by company managers using non-GAAP measures. In this study I reviewed the theory that non-GAAP numbers have improved financial reporting since the introduction of the guide in 2011.
Using manually collected data I examined 100 companies listed in the ASX 100. I used the annual reports from the years 2009-2012, two years before and two years after the
introduction of Regulatory guide-230. I conducted a word search in these annual reports looking for non-GAAP numbers. I only used companies which used the same non-GAAP numbers over all of the four years. Thereafter I looked up the related GAAP numbers. 43 observations were found. One observation is a non-GAAP measure that is used over four years by a company with a related GAAP number.
First I checked if there was a significant change in the percentage difference between GAAP and non-GAAP numbers. (Bradshaw & Sloan, 2002) showed that non-GAAP numbers were higher than GAAP numbers. I also concluded that non-GAAP were on average higher than GAAP numbers. But there was no significant evidence that there was a significant change after the Regulatory guide was introduced. The average difference between GAAP and non-GAAP numbers did decline in my sample, but since the test is not significant we can not conclude that the new regulation had nothing to do with it.
Next I examined the quality of non-GAAP earnings versus GAAP earnings. The first quality measure is the prediction of future cash flows of a company. A more persistent
earnings number is considered to be of higher quality. Underlying theory informs us that it will be easier to predict the future cash flows of a firm when it has a more persistent earnings number (Dechow, Ge and Shrand, 2010). The total sample shows that non-GAAP earnings have a higher persistence number than GAAP earnings, making non-GAAP more persistent thus of higher quality. But comparing the persistence number between the “before” and “after” sample, reveals a decline in the number. In the “after” sample, non-GAAP earnings still have a higher persistence number than GAAP earnings. But comparing it to the ‘before’ sample, it declined. Because of this decline I conclude that Non-GAAP earnings do not provide a better future prediction of earnings than GAAP earnings since the implementation of Regulatory guide-230.
Finally I checked to see if non-GAAP earnings provided more value relevant information than GAAP earnings. I tested this with an incremental value relevance test (Hung & Subramanyam, 2007). An accounting amount that has a significant association with share prices indicates that the information is relevant and reliable enough to be of value to investors. Again I divided the test into 3 samples; “Total”, “Before regulation” and “After regulation”. None of the samples showed that non-GAAP earnings provide value relevant information. There is no evidence that “after” regulation, non-GAAP earnings provide more value relevant information than GAAP earnings. So my conclusion on the last hypothesis is; I have to reject H3, Non-GAAP earnings do not provide more value relevant information since the introduction of Regulatory guide-230.
The conclusions of this study are not what were expected. The expectation was that Regulatory Guide-230 would improve non-GAAP measures and that this would become clear when comparing the “before” and “after” data. In other studies where the impact of regulations, like regulation G, and other SEC interventions were taken into account, non-GAAP quality improved. However this is not the case here. This study concludes that the new Regulatory guide-230 did not improve non-GAAP measures after its introduction. One of the reasons why this may be, is that Regulatory guide-230 tries to improve disclosure of non-GAAP earnings. Managers need to explain how they derive certain numbers. This seems to be the major aim of this guide. It does not set guidelines on how to make non-GAAP earnings as informative as possible. However it is strange that the ‘before’ sample has better conclusions than the ‘after sample’. Secondly the sample used in this study is small. For future research it may be better to extend the sample and/or collect more variables.
