The use of non-GAAP measures by the technology industry

(1)

The use of non-GAAP measures

by the technology industry

Name: Shanita Somai Student number: 11411651

Thesis supervisor: prof. dr. D. Veenman Date: June 25, 2018

Word count: 18.165

MSc Accountancy & Control, specialization Control

(2)

Statement of Originality

This document is written by student Shanita Somai who declares to take 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

(3)

Abstract

Over the past few years, there has been an increase in companies that disclose non-GAAP information. Previous studies have found opposite findings for why companies disclose non-GAAP information. That is why there is still an ongoing debate on whether non-GAAP measures are really that informative for companies and its investors and on why companies disclose non-GAAP information. Through an archival study and a correlation- and regression analysis on 12.603 observations across 771 companies per fiscal quarter, I examine whether the factors R&D expenses, Intangible assets and ROA are related to the disclosure of non-GAAP information. I decided to focus on the technology industry in the US and collected data from WRDS and Bentley et al. (2017) from the time period 2006 to 2016. First, I tested the effect of the independent variables on the dependent variable separately. After this, I tested the combined effect of the independent variables on the dependent variable. Overall, I find a positive linear relationship between high investments in R&D expenses, a high percentage of intangible assets and a low ROA with non-GAAP disclosure. This implies that companies with high R&D expenses, high percentage of intangible assets and a low ROA actually disclose more non-GAAP information to be more informative towards investors. For the multiple regression analysis, with all variables included, I find that R&D Expenses, Intangible assets and ROA all have a positive linear relationship with non-GAAP disclosure. The ROA shows different results throughout the regression analysis, turning from a negative relationship to a positive relationship. I therefore cannot draw a definitive conclusion for the effect of ROA on non-GAAP disclosure. Furthermore, the results do not explain much of the non-GAAP disclosure variable. The regression model with all variables included explains only 13,15% of non-GAAP disclosure, and 86,85% is explained by other, unknown, factors. It shows that the probability of non-GAAP disclosure is most probably dependable on far more variables.

Keywords: non-GAAP measures; non-GAAP disclosures; R&D expenses; Intangible assets; Return on assets (ROA).

(4)

1 Introduction

The SEC often makes statements regarding companies that use non-standard accounting practices. Regulators are noticing that companies are making much more use of other accounting methods next to the internationally recognized Generally Accepted Accounting Principles (GAAP). These other accounting methods tend to include pro forma assumptions, exclusions or expectations rather than measurable data. These methods together are called non-GAAP measures (Isidro & Marques, 2008). Nowadays, these non-GAAP measures bring a number of challenges and risks to companies. So why do companies still use them and what is influencing the use of the non-GAAP measures by companies?

The purpose of this thesis is to examine why companies in the United States use non-GAAP measures in their reporting. There are many industries in the United States. One industry in particular is at the centre of the continuing wave of innovation, and that is the technology industry. More than anything, technology companies are associated with innovation and invention (Lantz & Sahut, 2005). Because of this, many say that technology companies need to invest heavily in research and development (R&D). This matches the statement of Chen et al. (2001) that technology industries tend to make large investments in R&D. One of the reasons why technology companies will continue to dominate other industries is because they continually reinvest much of their profit in research and development (Acker et al., 2017). The investments in R&D can lead to future earnings. However, these future earnings are uncertain (Chen et al., 2001). The research of Chen et al. (2001) finds that companies disclose non-GAAP information when future earnings are relatively more uncertain. I therefore assume that, because technology companies need to invest heavily in R&D to stay innovative and inventive, their future earnings will more likely be uncertain. According to Chen et al. (2001) this would suggest that these companies would disclose non-GAAP information. That is why I assume that high investments in R&D will lead to an increased use of non-GAAP measures.

Furthermore, technology companies often have a high percentage of intangible assets (Chen et al., 2001). This is because a company also requires organizational structure, employees, patents and other factors that are called “intangible assets” (Brynjolfsson et al., 2002). However, intangible assets have their known problems, like the

(8)

definition, recognition and accounting treatment. This has led to the fact that the current accounting models are less informative about the intangible assets (Francis & Schipper, 1999).

The research of Chen et al. (2001) also finds that companies disclose non-GAAP information when current earnings are less informative. Concluding, I assume that, because technology companies have a high percentage of intangible assets, they have current earnings that are less informative. According to Chen et al. (2001) this would suggest that a company would decide to disclose non-GAAP information. That is why I assume that having a higher percentage of intangible assets will lead to an increased use of non-GAAP measures.

Next to this, because of the continuing reinvestments and the fact that technology companies are growing fast (Lanz & Sahut, 2005), the performance of most of the technology companies should also be very high. This is consistent with the findings of Child (1974), which states that growth is associated with performance of a company. However, the research of Konstantindi & Pope (2016) finds that loss firms are associated with increased uncertainty about future earnings compared to profitable firms. According to Chen et al. (2001), companies disclose non-GAAP information when future earnings are more uncertain. This means that non-GAAP disclosures are more important for low performance (loss) firms than for high performance (profit) firms. This matches the findings of Leung and Veenman (2018), which state that non-GAAP disclosures are particularly more informative than the non-GAAP disclosures for loss firms. Moreover, the research of Heflin & Hsu (2008) states that the performance of a firm can be improved by excluding more expenses in the non-GAAP disclosure. This means that a company with low performance would be more likely disclose non-GAAP information, to improve their performance, than a company with high performance. In conclusion, I assume that, because of the uncertainty about future earnings, a company with a low performance would more likely decide to disclose non-GAAP information. That is why I assume that having a low performance will lead to an increased use of non-GAAP measures.

So, the factors R&D expenses and intangible assets could have a positive relationship with the use of non-GAAP measures, in contrast to the performance factor. A high performance would have a negative relationship with the use of non-GAAP measures.

(9)

That is why, more specifically, this thesis will investigate which factors could influence the use of non-GAAP measures by technology companies. This will be an attempt to answer the following research question:

Why do companies that belong to the technology industry in the United States use non-GAAP measures in their disclosures?

Providing an answer to this research question is important, because non-GAAP measures are recognized as very informative by previous studies. However, there is a lot of criticism towards non-GAAP disclosure, because recent studies suggest that some managers disclose non-GAAP earnings opportunistically (Frankel et al, 2011). Next to this, the financial press also questions the usefulness of non-GAAP disclosures (Leung & Veenman, 2018). Moreover, Marques (2017) found that in certain circumstances the publication of non-GAAP information can mislead investors. This is why it is interesting to find out why technology companies still use non-GAAP measures in their reporting and what is influencing them to do so.

For this research, I focus solely on the United States because many technology companies are located in the United States. The United States is also known for being one of the most innovative countries in the world. I chose to focus on the technology industry because it is the largest single segment of the market. In addition to that, technology companies continuously reinvest in R&D to stay innovative and tend to have a high percentage of intangible assets, which contributes to a company’s growth and performance.

I included three variables in my research namely, R&D expenses, intangible assets and performance. By combining the literature and my thoughts I developed corresponding hypotheses for these variables. This study is an archival research. To answer the research question, I gathered the data from two data sources. The first data source contains common financial data of technology companies (e.g. R&D expenses, intangible assets, net income, total assets) and the second data source contains data on the publication of non-GAAP information by technology companies. I inspected and filtered both datasets and merged them into one dataset.

I also added three control variables to the dataset: Firmsize, Profit/Loss and Trend. The main research is not on these control variables, but I assume that these control variables are related to the dependent variable. I included these variables to control

(10)

for their effects on the dependent from the model (Bentley et al., 2017). I chose to control for firm size because I expect that bigger companies have more capabilities to invest more in R&D (Chen et al., 2001; Acker et al., 2017). Next to this, I assume that non-GAAP disclosures are more often used by low performance (loss) firms than by high performance (profit) firms (Konstadini & Pope, 2016; Chen et al., 2001).

This is because Non-GAAP disclosures are particularly more informative than GAAP disclosures for firms that operate at a loss (Leung & Veenman, 2018), hence the inclusion of the Profit/Loss variable. Furthermore, non-GAAP measures have become more popular overtime (Bradshaw & Sloan, 2002; Zhang & Zheng, 2011). That is why I assume that companies in general started to disclose more non-GAAP information overtime. To control for this effect I included the Trend control variable. In order to examine whether there is a relationship between these variables, I conducted a correlation- and regression analysis.

The results of the correlation analysis shows that there is little relationship between R&D Expenses, Intangible Assets, ROA, Profit or Loss and the probability of Non-GAAP disclosure. There is a weak positive relationship between the Firm Size, Trend and probability of Non-GAAP disclosure.

The regression analysis between the R&D Expenses and Non-GAAP shows results as expected. R&D Expenses seems to have a strong positive linear relationship with the probability of Non-GAAP disclosure. In conclusion, R&D expenses are positively related to the use of non-GAAP disclosure by technology companies. I find similar results for Intangible Assets and Non-GAAP disclosure, which also has a positive linear relationship with Non-GAAP disclosure. This implies that a high percentage of intangible assets leads to a higher probability of non-GAAP disclosure. For the multiple regression analysis, with all variables included, I find that R&D Expenses, Intangible assets and ROA all have a positive linear relationship with non-GAAP disclosure. The ROA shows different results throughout the regression analysis, turning from a negative relationship to a positive relationship. I therefore cannot draw a definitive conclusion for the effect of ROA on non-GAAP disclosure.

The results of this thesis are relevant for stakeholders and companies themselves. Stakeholders (investors) might discover why technology companies provide them with pro forma information and what is influencing companies to use non-GAAP

(11)

measures. Is it because of their high percentage of investments and intangible assets? Or is it because they want to emphasize the positive points and hide the negative points of their performance? If this is the case, they are actually misleading the investors. Companies, on the other hand, might acquire a better insight or understanding of how the industry uses non-GAAP measures.

My research makes several contributions to prior literature. Firstly, it contributes to extend the literature on the subject of non-GAAP measures. Secondly, prior research is done in a different way than I did my research because I examined the relationships between certain factors and the use of non-GAAP. This research may develop new insights for companies that invest much in R&D and intangible assets.

The next section of my thesis describes the literature review. Here, I will explain what US-GAAP and what non-GAAP measures are. Furthermore, I will explain more about technology companies, R&D expenses, intangible assets and the Return on Assets. In the following section, I will describe how I developed my seven hypotheses. I will explain what my thoughts are on the subject and support them with theoretical arguments. This will eventually lead to my hypotheses. The fourth section will set out my research methodology. I will explain which data sources I used and how I inspected and prepared these data sources for the correlation and regression analysis. I will show the results of the correlation and regression analysis in section five. Lastly, in section six, I will draw my conclusions.

(12)

2 Literature review

2.1 U.S. Generally Accepted Accounting Principles (GAAP)

As mentioned earlier, I focus solely on technology companies in the United States. Therefore, in this section, I will describe the origin of the US GAAP. Hereafter, I will explain what financial reporting is and why this is important for the economy. Finally, I will describe other accounting standards, like IFRS, and compare it with the US GAAP.

2.1.1 The origin of U.S. GAAP

In 1929, the stock market crash occurred in the United States. During this period, many people lost their faith in the stock market and in the US economy. The government decided that they needed to rebuild that lost faith, and so, the Securities and Exchange Commission (SEC) was established. The purpose of the SEC was to regulate financial practices among publicly traded companies. Later on, the SEC started to ask for help from the American Institute of Accountants (AIA) in examining the formation of financial statements. The AIA created a subcommittee to specifically create the GAAP principles. This subcommittee was named Committee on Accounting Procedure (CAP) and comprised 18 accountants and three accounting professors. The CAP created the first set of GAAP standards. In 1973, the SEC decided to replace the CAP with the Financial Accounting Standards Board (FASB), which is still in place today (SEC, 2000). The FASB establishes financial accounting and reporting standards for public and private companies and not-for-profit organizations that follow the Generally Accepted Accounting Principles (GAAP) and aims to improve financial accounting standards in the United States for current and potential investors, financiers, donors, creditors and stakeholders of financial reporting (FASB, 2014).

These accounting principles have become the guidelines for companies in America, because all the companies in the United States are required to comply with Generally Accepted Accounting Principles (GAAP). It was developed from US fundamental laws and regulations with many specific rules. Thus, it can be concluded that U.S. GAAP are rules-based (Gaeremynck & Van der Meulen, 2003).

(13)

2.1.2 The objective of financial reporting

Healy and Palepu (2001) state that in an economy there should be an optimal allocation of savings to investment opportunities. The savings have to be matched to business investment opportunities. However, this is very complicated because of the information asymmetry problem and communication between both parties. Next to that, agency conflicts (conflicts of interests) are also a problem. A potential solution to these problems is regulations that require companies to fully disclose their private financial information. Disclosure of financial information between companies and investors play an important role in mitigating the above problems (Healy & Palepu, 2001). GAAP can solve this problem by allocating the capital by lowering transaction- and information costs between suppliers and users of capital, which is consistent with previous researches from the SEC, FASB and IASB (Kothari et al., 2010)

According to Bradshaw & Miller (2008), the U.S. GAAP makes it possible to compare financial reports worldwide despite the differences among the businesses in countries. However, the analysis of Kothari et al (2010) states that GAAP is no facilitation system for firms, but it serves as a “control” system. This means that the financial reports that are prepared according to the GAAP standards provide information that is useful for performance evaluation (efficient contracting). This result supports the conclusion of the research of Holthausen & Watts (2001).

2.1.3 International Financial Reporting Standards (IFRS)

After the Enron scandal in 2002, there was much criticism on the U.S. GAAP. More and more scandals kept occurring. As mentioned earlier, the U.S. GAAP is rules-based. Because of this, companies could stick to just the rules and secretly hide huge debts. In addition to that, companies also used earnings management, which means that managers could change the structure of transactions by making use of their own judgment (Platikanova & Nobes, 2006). In order to prevent further scandals (such as Enron) and criticism, the European Parliament mandated all European listed companies on 19 July 2002 to apply IFRS to their consolidated financial statements from 1 January 2005 on (Soderstrom & Sun, 2007). The purpose of IFRS was to prevent future scandals and the use of earnings management, to improve the quality of financial reports and to facilitate the comparison of financial reports (Picker et al, 2016).

(14)

2.1.4 Differences U.S. GAAP and IFRS

Contrary to the U.S. GAAP, the IFRS is not a national set of standards. The International Accounting Standards Board (IASB) is responsible for preparing the IFRS. There is no strict regulator that provides interpretations of standards in Europe. The distance to 'politics' is therefore greater in Europe. Yet the relationship between politics and financial reporting is not completely absent. After all, in Europe, the European Commission can block the certain unacceptable standards. In addition, strong criticism from the business world, from accountancy firms or from others can also influence the content of the standards.

However, the FASB and IASB have been attempting to minimize the differences between GAAP and IFRS for internationalization (FASB, 2013).

According to JeanJean & Stolowy (2008), the use of IFRS has little influence on the prevention of the use of earnings management. Besides that, Leuz (2010) concludes that it will always be hard to compare financial reports worldwide despite the same accounting standard. However, he also states that it is unfeasible to apply the same accounting standards worldwide, so he supports the differences between U.S. GAAP and IFRS.

2.1.5 Summary

Due to the stock-market crash in 1929, the US government wanted people to have more confidence in the financial market. Eventually, the FASB was established. The FASB created financial accounting and reporting standards for public and private companies and not-for-profit organizations that follow Generally Accepted Accounting Principles (GAAP) and financial accounting standards in the United States for current and potential investors, financiers, donors, creditors and stakeholders of financial reporting. GAAP became the guidelines for companies in the US and all the companies in the United States are required to comply with it. After criticism arose at GAAP, the IFRS was developed, but it is still criticized. Despite the criticism, both standards are still used, GAAP in the US and IFRS in Europe.

Unfortunately, financial information is becoming more complex and diverse. This is why stakeholders want additional financial information from the companies. This leads us to the next section, which will discuss non-GAAP information.

(15)

2.2 Non-Generally Accepted Accounting Principles (non-GAAP) financial measures

In this section, I will first explain what voluntary disclosure is and why companies are providing it. Next, I will describe what non-GAAP financial measures are. Furthermore, I will provide an example of a non-GAAP measure used by a real world company. In addition to that, I will describe why and how companies are reporting non-GAAP financial measures. Finally, I will name the risks of non-GAAP financial measures.

2.2.1 Voluntary disclosure

Users of annual reports need a lot of financial and non-financial information in order to assess the value of an investment in a company (Myburgh, 2001). The demand for information and transparency from listed corporations started to increase. In spite of an increased demand for mandatory disclosures from regulators, corporations chose to voluntarily disclose additional information in order to satisfy the demands from the capital market (Broberg et al, 2010). Furthermore, the additional information can be relevant for the decision-making of users of the company's annual reports. The research of Bhattacharya (2003) has shown that non-GAAP measures provide information that is relevant for investors next to the GAAP numbers. Therefore, I assume that non-GAAP financial measures are useful for investors and stakeholders to better capture a company’s value. But when and how were non-GAAP financial measures developed?

2.2.2 The origin of non-GAAP financial measures

In 2001, the SEC sent a warning to public companies that used other methods than the Generally Accepted Accounting Principles (GAAP). Companies used these principles to present earnings and results of operations. The warning was related to non-GAAP, which is also known as pro forma. There is no official law that states that non-GAAP measures are illegal. The SEC, on the other hand, does state that non-GAAP measures can be misleading for investors. The use of non-GAAP measures can make it difficult for investors to compare different periods or companies with each other. The SEC

(16)

also states that companies should be aware of their obligations to share correct information with investors.

In 2002 a warning statement regarding non-GAAP measures was released by the International Organisation of Securities Commissions (IOSCO Committee). The IOSCO noticed a development regarding the use of non-GAAP measures. More companies started using their own specific measurements for press releases and reports to shareholders. According to the IOSCO, this development had a positive and a negative effect. The use of non-GAAP measures can have very positive effect for shareholders if companies use it as a correct and consistent method to publish certain results. It can make life a lot easier for investors, because it would be easier to understand the numbers of a company. The negative effect of the method starts to arise when companies start using non-GAAP measures as a way to hide certain results. Therefore, the IOSCO warned investors. Non-GAAP measures can be a very good method to publish results. However, it can also be very misleading in case of bad consistency or false numbers.

On November 4th 2002, a new disclosure regulation was applied. This new regulation was implemented so that investors would have access to well-balanced financial information. Unfortunately, the term “Pro Forma” was a term that the SEC used in multiple contexts. Therefore, the name non-GAAP originated (SEC, 2017; Castillo, 2017)

2.2.3 What are non-GAAP measures?

Regulation G and item 10(e) of Regulation S-K define a non-GAAP financial measure as a numerical measure of historical or future financial performance, financial position or cash flows, that (1) excludes amounts that are measured in accordance with GAAP or amounts that are measured in accordance with GAAP but that have been adjusted too much or (2) includes amounts that are artificially created (Castillo, 2017). This matches the definition of Entwistle et al. (2006). Next to that, the SEC determines that a non-GAAP financial measure is (1) a measure that is different from the ones that are measured and presented in accordance with GAAP in the financial statements, or (2) a measure of liquidity that is different from the cash flows computed in accordance with GAAP (SEC, 2018). So, I conclude that a non-GAAP financial measure is an artificially created measure or a financial measure

(17)

measured and presented in accordance with GAAP, but is adjusted. This is why non-GAAP financial measures are often addressed as “adjusted …” (for example: adjusted EBITDA). Other examples of non-GAAP measures are: adjusted revenue, adjusted net income, EBITDA, adjusted EBITDA, EBITDAR, EBIT, adjusted EPS, free cash flow, funds from operation, net debt, book to bill ratio and unbilled deferred revenue and many more. A company has flexibility in which non-GAAP financial measures it chooses to report. Which is why different firms do not report the same non-GAAP financial measures. Firms can decide themselves which non-GAAP financial measures they use (Bhattacharya et al., 2003). Bentley et al. (2017) also describes non-GAAP earnings as customized versions of GAAP earnings. Managers first determine the GAAP earnings and then decide which components seem less representative of the core performance and finally exclude these earnings. So, managers customize GAAP earnings, which leads to non-GAAP earnings.

Because managers can decide which exclusion is not representative (no guidelines) or not relevant, there are no standard non-GAAP measures. Managers can decide for themselves what they recognize as a non-GAAP measure. However, prior research suggests that managers and analysts do frequently make similar reporting choices (Heflin & Hsu, 2008), which implies that there are some common non-GAAP measures that are used more often.

2.2.4 Two examples of non-GAAP measures (Apple Inc. & Marin Software Inc.)

Apple Inc. is a high technology firm that uses non-GAAP measures. Appendix 1 shows an example of the GAAP results versus the non-GAAP results of Apple Inc. In the first column the net sales are reported in accordance with GAAP ($ 8,337 million). The second column shows the adjusted non-GAAP number (1,405). In the last column, the first first two columns are added together (8,337 + 1,405 = 9,742). This gives us an “adjusted net sales” of $ 9,742 million (Apple Inc., 2009). Apple Inc. (2009) states that “the Non-GAAP adjustment to net sales reflects (i) the

reversal of the current period's amortization or deferred revenue from Apple handsets and Apple TV units shipped in current and prior-art and (ii) at the time of sale related to iPhone handsets and Apple TV units shipped in the current period”. The

management of Apple Inc. decided to adjust the net sales with $ 1,405 million. They state that adjustment of net sales is necessary because the net sales in accordance with GAAP does not reflect the deferral of revenue and product costs for recognition in

(18)

later periods (Apple Inc., 2009). Apple uses more non-GAAP measures like, “Adjusted Cost of Sales,” “Adjusted Gross Margin,” “Adjusted Operating Margin,” “Adjusted Net Income” and “Adjusted Diluted Earnings per Share”.

Another example of non-GAAP measures is presented in Appendix 2. This is from a smaller technology company, Marin Software Inc. In this example, all of the company’s expenses are adjusted. First, they present their expenses according to GAAP. After this, they adjust it for stock-based compensation expense, the amortization of intangible assets, the capitalization of internally developed software, noncash expenses related to the issuance of warrants, the amortization of internally developed software and the non-recurring costs associated with acquisitions and restructurings (Marin Software Inc., 2015). This results in the non-GAAP expenses. The company states that it uses these non-GAAP financial measures internally for analyzing its financial results. Additionally, they believe that the non-GAAP financial measures are useful to investors for evaluating the company’s performance (Marin Software Inc., 2015). These examples show that companies indeed use different non-GAAP measures for different reasons.

2.2.5 Why do companies use non-GAAP financial measures?

As of now, there is no one single answer to this question. However, previous studies provide many insights on the use non-GAAP measures.

Typically, companies in the United States prepare their financial statements with financial measures that are measured and presented in accordance with GAAP (Deloitte, 2016). Although GAAP measures are used for decision-making, non-GAAP measures can serve as an addition to this (SEC, 2018). This matches the findings of Lev and Zarowin (1999) that claim that the usefulness of financial statements in accordance with GAAP decreased, which led to an increased need for extra financial information. Nowadays, Non-GAAP financial measures can be found in company's earnings releases, periodic reports, press releases and other means of communication to stakeholders (Deloitte, 2016). The research of Chen et al. (2001) finds that companies disclose non-GAAP information when future earnings are relatively more uncertain or when current earnings are less informative. This matches the findings of the study of Bowen et al., which states that managers are more likely to report or emphasize non-GAAP earnings when GAAP earnings are less informative.

(19)

In conclusion, when GAAP-earnings are less informative, managers of companies could decide to be more informative towards the investors by providing non-GAAP measures. Unfortunately, there is an ongoing debate whether non-GAAP measures are really informative for a company itself and its investors. That is why there has been a lot of research on this subject the past few years.

Marques & Isidro (2017) merge different insights from existing researches on the disclosure of non-GAAP earnings. Their conclusion is that non-GAAP earnings are more informative than GAAP earnings. They also claim that managers report non-GAAP measures to provide relevant information about the company’s performance to financial statements users. They show that this is in line with the researches of Brown & Sivakumar (2003) and Bhattacharya et al (2003). Bhattacharya et al. (2003) investigated whether investors experience non-GAAP earnings to be more informative than GAAP operating income by analyzing a sample of 1,149 actual pro forma press releases. They also conclude that investors experience non-GAAP information to be more informative in the United States. Next to this, there are more researches that conclude that investors find non-GAAP measures more informative than GAAP measures (Bradshow & Sloan, 2012; Bradshaw et al., 2017). In conclusion, the results of these researches show that non-GAAP measures are used to provide more information to investors

Additionally, the study of Leung and Veenman (2018) examined the incremental information in loss firms’ non-GAAP earnings disclosures relative to GAAP earnings. They find that loss firms’ non-GAAP numbers are significantly more predictive and less strategic than non-GAAP numbers of profitable firms. Therefore, it is actually helpful for the investors. The non-GAAP numbers of loss firm’s help investors even more in forecasting future results and valuations. This is because the investors can decide for themselves if these losses can be categorized as forecasting profits. This means that non-GAAP disclosures are more important for low performance (loss) firms than for high performance (profit) firms. Moreover, the research of Heflin & Hsu (2008) states that the performance of a firm can be improved by excluding more expenses in the non-GAAP disclosure. This implies that a company with low performance would more likely disclose non-GAAP information, to improve their performance, than a company with high performance. In conclusion, non-GAAP measures are found to be more informative for loss firms than GAAP measures.

(20)

In addition to that, Castillo’s research (2017) explains that non-GAAP financial measures can also enable management to create new insights in how they see the company’s financial condition or results, in a manner that GAAP results alone may not be able to create. In conclusion, this means that the use of non-GAAP measures by companies also benefits the company itself by creating better insights.

However, another reason for using non-GAAP measures could be the opportunistic behaviour of managers. For example, when managers decide to hide the (negative) financial results determined according to GAAP. This is called opportunistic behaviour (Walker & Louvari, 2003). Marques & Isidoro (2017) also found that in certain circumstances the publication of non-GAAP is due to opportunistic behaviour, which means that it could also mislead investors. This matches the findings of Doyle et al. (2013), which states that managers disclose non-GAAP earnings opportunistically. Luckily, Marques (2017) also states that corporate governance mechanisms can control managers’ opportunistic behaviour. Though, in conclusion, opportunistic behaviour also could be a reason for disclosing non-GAAP measures. Studies also found that the management of certain companies tend to use more non-GAAP measures. Zhang & Zheng (2011) concluded that 59,1% of the high technology firms uses non-GAAP measures. This could indicate that there are certain factors in this industry, which influence the use of non-GAAP information.

From these researches, I conclude four common reasons for companies to disclose non-GAAP measures. These are: to be more informative, to create better insights of the company’s financial condition, opportunistic behaviour and the type of industry can contribute to the use of non-GAAP information.

However, what are the risks of using non-GAAP measures for companies and for investors?

2.2.6 Risks of using non-GAAP financial measures

GAAP financial measures are standardized and a company has no flexibility in the measurement and/or presentation of the GAAP financial measures. In contrast to non-GAAP financial measures, where managers of firms can decide themselves whether to disclose non-GAAP financial measures and which non-GAAP financial measures are relevant for their company and their investors (Entwistle et al, 2006). This leads to differences in definitions, labelling, calculations, presentations and the fact that these

(21)

measures are not comparable to each other. That is why non-GAAP measures are susceptible for misunderstandings (Deloitte, 2016).

The management of a company is responsible for their non-GAAP disclosure. This means that auditors do not check the non-GAAP financial measures (Black et al., 2016). Because of this, it is important that reported non-GAAP measures are truly informative and reliable, especially for investors who depend heavily on their use; otherwise it could be misleading for the investors (Bhattacharya, 2003).

2.3 Technology companies in the United States

Keeble (1990) defines the technology industry as “those activities in which rapid

technological change and high inputs of scientific research and development expenditure and employment are producing new, innovative and technologically advanced products”. Nowadays, technology is no longer a jumble of wires and metal.

It is a living, evolving organism that has its own unconscious needs and tendencies and consists of applied scientific knowledge (Kelly, 2010). This means that technology needs innovation to develop itself.

The technology industry is the largest single segment of the market and this sector eclipses all other sectors. A few examples of technology companies are: Alphabet (Google), Amazon, Apple, Facebook and Microsoft (“Big-five” super-competitors based in the U.S.). The technology industry experiences many changes and innovations. Therefore, to keep dominating the market, technology companies continually have to reinvest their profit. The expenses of reinvestment are often labelled Research and Development (R&D) expenses. This also matches the statement of Lantz & Sahut (2005) that technology companies are forced to invest in research and development (R&D) and that the growth of technology companies is driven by the exploitation of innovative products and services. I conclude that technology companies invest heavily in R&D and that they are a huge investment opportunity. This is because reinvestments of a technology company leads to that the management of a technology company expense (R&D) a lot to get future benefits in return. This is why technology companies are very attractive for investors. UHHH..

Furthermore, technology companies often have a high percentage of intangible assets. This is because a company also requires organizational structure, employees,

(22)

patents and other factors, which are called “intangible assets” (Brynjolfsson et al., 2002). These assets can also play a role for a company in distinguishing themselves from competitors. Key intangible assets for a technology company are: copyrights, patents, critical employees, research and development et cetera.

Lastly, the findings of Lanz & Sahut (2005) indicate that technology companies are associated with growth and the finding of Child (1974) indicates that growth is associated with performance of a company. However, the research of Konstantindi & Pope (2016) finds that loss firms are associated with increased uncertainty about future earnings compared to profitable firms. According to Chen et al. (2002), companies disclose non-GAAP information when future earnings are more uncertain. From this, I derive that the performance of a technology company can also be an influenceable factor in the use of non-GAAP financial measures. The following sections will explain more about the terms: R&D expenses, intangible assets and performance, in combination with technology companies.

2.4 R&D expenses of technology companies

In this section, I will first explain what investments are. Secondly, I will explain what R&D expenses are. Lastly, I will describe why I chose R&D expenses as a variable for my research.

2.4.1 Investments by technology companies

Investments are things of value or assets that a company buys now in the hope they will be worth more later on and/or they will provide income in the form of cash or other valuable items (Graham, 2009). There are different kinds of investments. This paper will only focus on one kind of investment: R&D investments. The reason for this is that technology companies invest a lot in activities that generate future income (for example, a new product), create competitive advantages and improve/create products and services. Thus, a huge part of their investments can be categorized as Research & Development (R&D) expenses.

2.4.2 What are R&D expenses?

The expenses of Research and Development are called R&D expenses. These are costs that are made to stay ahead of competitors in the field of innovation. Kor (2006) states that reinvesting in R&D is necessary for a firm, especially for firms whom are

(23)

competing on the basis of innovation. It is a general finding that a firm’s investment in R&D is an important source of productivity growth (Kancs & Siliverstovs, 2012). “A greater commitment in R&D should result in greater internal development of new

discoveries as well as enhance the flow of new scientific information into the firm. The relative amount of expenditures on research and development has traditionally been used as an indicator of a firm’s innovative activity in many industries” (Scherer,

1980). Previous studies conclude that technology companies invest more in R&D than other industries and that R&D has a higher effect on technology companies (Hall & Mairesse, 1995; Wang, 2004). In conclusion, R&D expenses can be used as a measurement for innovation. Technology companies have to innovate to stay ahead of their competitors. But what is the relation between R&D expenses and non-GAAP financial measures?

2.4.3 Relation non-GAAP financial measures and R&D expenses

Firms in high technology industries tend to make large investments in R&D expenses. These investments can lead to future earnings. However, these future earnings are uncertain (Chen et al., 2002). R&D investments are risky projects for companies and investors.

This is because the results/benefits of R&D investments are difficult to forecast. This leads to the fact that these earnings from traditional accounting models are less informative (Collins et al., 1997; Francis and Schipper, 1999; Lev and Zarowin, 1999). Therefore, investors would like to have more information on the R&D investments of a company.

Next to that, investors also have to deal with asymmetric information problems (Singh and Flaircloth, 2005). Managers of respective companies have more information than their debt holders, which means a greater risk for the debt holders. A solution to this problem is non-GAAP disclosure. Non-GAAP disclosure gives the debt holders more information about the R&D expenses, which will increase their trust in the company. Because of this, it is beneficial for companies to report non-GAAP measures that are useful to investors in assessing the company’s operating performance.

A few examples of non-GAAP measures for investments are: Non-GAAP product development expense (product development expense, excluding (1) asset impairment and loss on disposal of assets, (2) headquarters relocation costs, and (3) stock-based

(24)

expense), Non-GAAP product development margin (non-GAAP product development expense as a percentage of non-GAAP total revenue) or Adjusted EBITDA.

2.4.4 Conclusion of the relation between R&D expenses and non-GAAP measures

Based on the above, I assume that if companies invest a lot in R&D, investors require more financial information, which means that companies are more likely to use non-GAAP financial measures to be more informative about their future earnings. This is the reason why the factor “R&D expenses” is included in this research.

2.5 Intangible assets of technology companies

This section will first explain what intangible assets are. After that, I will describe what the problem with intangible assets is. Lastly, I will explain the relationship between intangible assets and non-GAAP financial measures.

The International Accounting Standards (IAS 38) defines intangible assets as “identifiable non- monetary assets without physical substance”. Arrighetti et al. (2004) defines intangible assets as “ it includes internally produced resources

(designs, blueprints, brand equity, in-house software, and construction projects) and assets acquired externally (technology licenses, patents and copyrights, and the economic competencies acquired through purchases of management and consulting services)”. In the previous section, I explained that the R&D expenses of technology

companies are often high. These expenses are important for innovation, which will lead to competitive advantage. Intangible assets have a similar role for companies. Currently, these assets are inductors for success and an important factor for the financial statements and achieving competitive advantage (Ancuta et al., 2017).

2.5.1 What is the problem with intangible assets?

There are a few problems with intangible assets. Firstly, the definition itself forms a problem. In literature, intangible assets are defined differently. Secondly, the recognition of intangible assets is a problem. Many researchers do not know what kind of assets are actually categorized as intangible assets. This is because there is an economic- and an accounting perspective (Falk, 2013).

Lastly, when it is clear that an asset can be categorized as an intangible asset, it is still unclear what accounting treatment should be applied to value the intangible asset. This is because the accounting treatment is unclear and complex (Ancuta et al., 2017).

(25)

Due to the rising importance of intangible assets and their known problems, financial statements’ users would like to get more information about the intangible assets of the companies (Vodak, 2011).

2.5.2 Relation non-GAAP financial measures and intangible assets

There are two reasons why I assume that a company, which has a high percentage of intangible assets, will more likely use non-GAAP financial measures. The first reason is that because the definition, recognition and measurement of intangible assets are complex, the current earnings are less informative. The consequence of this is that the management of companies chooses to be more informative towards their stakeholders by disclosing non-GAAP information, because the current earnings are less informative (Chen et al., 2001). In addition, not all intangible assets appear on the balance sheet (SEC, 2018). Ancuta et al. (2017) states that “recognition of an

intangible asset in the balance sheet appears when it is probable that future economic benefits to flow to the entity and the asset has a cost or a value that can reliably measured”.

However, there are three valuation approaches for intangible assets, namely: market-, cost- and income approach (Madhani, 2008). With these approaches, companies can value their intangible assets. But this means that they also need to amortize over it and test for impairment, just like for acquired intangible assets. These expenses will reduce the profit/earnings/net income (Ancuta et al., 2017), which, again, means that it is useful for companies to report the non-GAAP measures to give investors and others a better understanding of the company’s operating performance. I assume that, if the management of a company does not decide to disclose non-GAAP financial measures, it could cause the problem that, for companies with much intangible assets, their balance sheet does not reflect any real value. If they do not provide additional information, like non-GAAP, the consequence of this is that it could be misleading for an outsider that is trying to gain an understanding of the value of a business by analyzing its financial statements.

2.5.3 Conclusion of the relationship between intangible assets and non-GAAP measures

Concluding, I assume that if a company has a high percentage of intangible assets, their current earnings are less informative. That is why investors require more financial information, which means that a company will more likely use non-GAAP

(26)

financial measures. Which is why I included the factor “intangible assets” in this research.

2.6 Performance of technology companies

This section will describe which method I will use to measure the performance of a company. In addition, I will explain why I choose the variable “performance” for this research. The performance of a company can be measured in multiple ways. There are four main areas to measure the health of a company: liquidity, solvency, profitability and operating efficiency. Because previous researches were focused on profit and loss, this research will focus on profitability. To measure the profitability of a company, I will use the Return On Assets (ROA) indicator. Barber & Lyon (1996) state that matching on ROA tends to be better than matching on other variables. However, a shortcoming of my research could be that I am solely focusing on profitability. Another reason for using ROA is that it is a direct indication of an manager’s performance.

2.6.1 What is Return On Assets (ROA)?

Return on assets is a profitability ratio that provides insight in how much profit a company is able to generate from its assets. It is an important profitability ratio, which shows the percentage of profit that a company earns in relation to its overall resources. It measures the amount of profit per dollar of a company's assets, so it includes all of the assets of a company (IFRS, 2018). The difference between the ROA and the other profitability ratios is that the ROA generates the profit before leverage. Other profitability ratios generate the profit by using the leverage. This makes the ROA less interesting for investors/shareholders, but for measuring performance it is a good indicator (Barber & Lyon, 1996).

In addition, the ROA is a direct indication of an manager’s performance. The manager’s performance and decisions also influence the performance of a company and so also the profits of a company (Davis & Martin, 2008).

2.6.2 Relation non-GAAP financial measures and ROA

The performance of a technology company is important for its future (Lanz & Sahut, 2005). If a technology company performs well, it means that the company is growing (Child, 1974). In conclusion, performance and growth are associated with each other.

(27)

However, the research of Konstantindi & Pope (2016) finds that loss firms are associated with increased uncertainty about future earnings compared to profitable firms. According to Chen et al. (2002), companies disclose non-GAAP information when future earnings are more uncertain. Next to this, the research of Heflin & Hsu (2008) states that the performance of a firm can be improved by excluding more expenses in the non-GAAP disclosure. Furthermore, the findings of Bowen et al. (2005) indicate that when firms experience negative earnings it is more likely to emphasize on non-GAAP measures. As a result, companies with negative earnings due to less informative expenses will use more non-GAAP measures to give more information about their performance and help their investors. Next to this, the study of Leung and Veenman (2018) examines the incremental information in loss firms’ GAAP earnings disclosures relative to GAAP earnings. They find that loss firms' non-GAAP earnings disclosures compensate the low informativeness of non-GAAP disclosures. Therefore, I conclude that if the performance of a firm is low, it will more likely use GAAP measures than when their performance is high, since the non-GAAP disclosure reflects a better performance and it is more informative about future earnings than the GAAP disclosure.

2.6.3 Conclusion of relation between ROA and non-GAAP measures

Taking in consideration the points made in the previous paragraph, I conclude that technology companies with a low performance, or a loss, are more likely to use non-GAAP measures than companies with a high performance. The reason for this is that it can improve their performance disclosed by GAAP measures and it can compensate the low informativeness of GAAP disclosures so it can help to be more informative towards the investors.

(28)

3 Hypothesis development

As mentioned before, this research includes three variables: investments, intangible assets and performance. Three corresponding hypotheses belong to these three variables. In this section, I will describe how I developed these hypotheses.

3.1 Hypothesis development for investments

The findings of Lanz & Sahut (2005) state that technology companies are forced to invest in research and development (R&D). In addition to that, they state that the growth of technology companies is driven by the exploitation of innovative products and services. Next to that, Kancs & Siliverstovs (2012) state that firm investments in R&D are an important source of productivity growth and innovation. I therefore conclude that innovative products drive technology firms and that R&D expenses create innovative products. This means that technology firms will invest a lot in research and development to be innovative and grow. The investments in R&D can lead to future earnings. However, these future earnings are uncertain (Chen et al., 2002). This leads to a situation, in which outsiders require more financial information to assess the company’s operating performance, which could results in companies using non-GAAP measures to be more informative about their uncertain earnings. Concluding, I assume that the more a company invests in R&D, the more likely it will disclose non-GAAP measures to be more informative.

Combined, these arguments lead to the following hypothesis:

H1: High investments in R&D are positively related to the use of non-GAAP

measures by technology companies.

The hypothesis above is stated in alternative form. The corresponding null hypothesis is that high investments in R&D are not related to the use of non-GAAP measures by technology companies.

3.2 Hypothesis development for intangible assets

According to Ancuta et al. (2017), intangible assets are inductors for success and an important factor for the financial statements and achieving competitive advantage. As with the R&D expenses, I assume that competitive advantage is an important factor for technology firms to gain customers and be successful.

(29)

Unfortunately, intangible assets have their known problems, such as its definition, recognition and accounting treatment that are still unclear for many. Despite this, the importance of intangible assets for technology companies is increasing. This has led to the fact that the current accounting models are less informative about the intangible assets, because the balance sheet does not reflect the real value (Francis & Schipper, 1999). According to Chen et al. (2002) this would suggest that a company would decide to disclose non-GAAP information, because the current earnings are less informative. This leads to the assumption that it is useful for technology companies to report non-GAAP measures to give investors and others a better understanding of the company’s operating performance. Therefore, I assume that the more intangible assets a technology company has, the more it uses non-GAAP measures to be more informative about them.

Combined, these arguments lead to the following hypothesis:

H2: A high percentage of intangible assets is positively related to the use of

non-GAAP measures by technology companies.

The hypothesis above is stated in alternative form. The corresponding null hypothesis is that the percentage of intangible assets is not related to the use of non-GAAP measures by tech-companies.

3.3 Hypothesis development for performance

Furthermore, the findings of Lanz & Sahut (2005) indicate that technology companies are associated with growth and the finding of Child (1974) indicates that growth is associated with performance of a company. However, the research of Konstantindi & Pope (2016) finds that loss firms are associated with increased uncertainty about future earnings compared to profitable firms. According to Chen et al. (2002), companies disclose non-GAAP information when future earnings are more uncertain. Concluding, this means that non-GAAP disclosures are more important for low performance (loss) firms than for high performance (profit) firms. I therefore assume that the lower the financial performance of a technology company, the more likely it will use non-GAAP measures to be more informative about their future earnings. This means that, combined, these arguments lead to the following hypothesis:

(30)

H3: Low financial performance is positively related to the use of non-GAAP

measures by technology companies.

The hypothesis above is stated in alternative form. The corresponding null hypothesis is that low performance is not related to the use of non-GAAP measures by technology companies.

3.4 Hypothesis development for all three variables

Because I am unsure whether the three variables separately have a positive effect on the use of non-GAAP measures, I will also test whether two of the three variables together have a more positive relationship than all three variables separately. This leads to the following three hypotheses:

H4a: High investments in R&D and a high percentage of intangible assets are

positively related to the use of non-GAAP measures by technology companies.

H4b: High investments in R&D and low financial performance are positively related

to the use of non-GAAP measures by technology companies.

H4c: A high percentage of intangible assets and low financial performance are

positively related to the use of non-GAAP measures by technology companies.

Once again, the hypotheses above are stated in alternative forms. The corresponding null hypotheses are:

● High investments in R&D and a high percentage of intangible assets are not related to the use of non-GAAP measures by technology companies.

● High investments in R&D and low financial performance are not related to the use of non-GAAP measures by technology companies.

● A high percentage of intangible assets and low financial performance are not related to the use of non-GAAP measures by technology companies.

If investing in R&D expenses, a high percentage of intangible assets and low performance are each separately positive related to the use of non-GAAP measures, I expect that all three variables together to have an even greater positive relationship than all three variables separately or two of the three variables together. This means that if I combine the above hypotheses it will lead to the following concluding hypothesis:

(31)

H5: High investments in R&D, a high percentage of intangible assets and low

financial performance are positively related to the use of non-GAAP measures by technology companies.

The hypothesis above is stated in alternative form. The corresponding null hypothesis is that high investments in R&D, a high percentage of intangible assets and low performance are not related to the use of non-GAAP measures by technology companies.

The above stated hypotheses will be tested with a correlation- and regression analysis in chapter 5. First, I will discuss in chapter 4 how I collected the required data for the analysis.

(32)

(33)

4 Research Methodology

4.1 Description of variables

The main question of this research is: Why do companies that belong to the technology industry in the United States use non-GAAP financial measures in their disclosures?

The predictive validity framework (“Libby boxes”) presented in Appendix 2 shows the relation between the independent variables and the dependent variable the controlled for the effects of other variables. The four variables will be measured via the following terms:

• Investments will be measured via the R&D expenses. Previous research has measured investments via R&D expenditures (Dechow and Sloan, 1991). The data is available for technology companies quarterly in WRDS.

• Intangible assets will be measured via the total intangible assets. The data is available for technology companies quarterly in WRDS.

• Performance will be measured via the Return on Assets (ROA). This is the net income divided by the total assets. Previous researches have consistently used one measure to assess financial performance by a company's profitability in relation to its total assets. Friedman (1970) used the net income of the companies to assess the financial performance. In that research it is a logical choice to choose net income as a measure, because this is the final amount of profit or loss after all expenses are deducted. This means that the net income include a variety of non-operational expenses, gains, and/or losses. Besides, by dividing the net income by the total assets, the ratios of different technology companies are comparable with each other. This is interesting for this research, because if a technology company thinks that their net income is low because of non-operational expenses, they can decide to use non-GAAP measures to exclude these expenses and report the “adjusted net income”. • Non-GAAP disclosure will be measured via the dataset provided by Bentley et

al. (2017). It contains data on whether a company disclosed non-GAAP information in their quarterly financial report.

(34)

4.2 Empirical models

In order to examine the relationship between the variables, I will conduct a correlation- and regression analysis. This will be done using the following empirical model. Please note that this empirical model includes all variables that I will use in my research. Later on, I will also conduct several regression analysis based on models that include only some of the variables. I will provide the corresponding empirical model with each regression analysis.

Empirical model (with all variables):

!"#!!""#_!" = ! !_!+!!_!!&!_!"#$%&$&_!"+!!_!!"#$"%&'()_!""#$"_!"+!!_!!"#_!" +!!_!!"#$%&'"(#))_!"#$%"&_!"+!!_!!"#$%"&'_!"#$%"&_!"

+!!_!!"#$%_!"#$%"& +!!_!" The dependent variable definitions:

• !"#!!""#!" = Whether company i disclosed non-GAAP expenses in quarter

t.

1 = Yes, 0 = No.

The independent variables definitions:

• !&!_!"#$%&$&!" = !&!!!"#!$%!%!_!"#$%!!""#$" = The proportion of Total assets invested in R&D by company i in quarter t.

• !"#$"%&'()_!!""#$"!" = !"#$"%&'()!!""#$"_!"#$%!!""#$" = The proportion of Total assets invested in Intangible assets by company i in quarter t.

• !"#!" = _!"#$%!!""#$"!"#!!"#$%& = The Return on Assets of company i in quarter t. The control variables definitions:

• !"#$%&'"(#))_!"#$%"&!" = Whether company i made a profit or a loss in

quarter t.

1 = Profit, 0 = Loss.

• !"#$%"&'_!"#$%"&!" = Whether company i had more total assets in quarter t than the median of total assets for all observations.

The use of non-GAAP measures by the technology industry