IFRS 15 disclosure quality: The influence of corporate governance mechanisms

(1)

IFRS 15 disclosure quality:

The influence of corporate governance mechanisms

Master thesis

Thimo Werkman S2934337

MSc Accountancy, University of Groningen, Faculty of Economics and Business

Supervisors: R (Roy) van Duuren MSc Prof. dr. R.L. (Ralph) ter Hoeven

June 24, 2019 Word count: 12 472

(2)

ABSTRACT

For annual reports from January 1, 2018, IFRS 15 ‘revenue from contracts with customers’ is mandatory for companies reporting under IFRS. This study investigates the influence of different corporate governance mechanisms on the disclosure quality of IFRS 15 among European listed companies. Accordingly, the sample consists of European data from 52 companies in 12 countries for the financial year 2018. The investigated corporate

governance mechanisms in this study are board independence, financial expertise on the audit committee and board gender diversity. Furthermore, the expected positive relation between board gender diversity and IFRS 15 disclosure quality is expected to be negatively influenced by the presence of a board gender quota. Using multiple linear regression analysis, this research was unable to provide significant results for three of the four hypotheses. However, the multiple linear regression analysis showed a remarkable significant negative relation between board gender diversity and IFRS 15 disclosure quality. Besides, several additional tests have been performed to strengthen the main results.

Keywords: IFRS 15, disclosure quality, board independence, financial expertise on the audit committee, gender diversity, board gender quota, agency theory, stakeholder theory, legitimacy theory, Upper Echelons theory

(3)

1. Introduction

ntil the beginning of this century, almost no one dared to dream of a common financial reporting language. The demand for a common financial reporting language has arisen due to the globalization of international financial markets (Jeanjean & Stolowy, 2008). With the introduction of International Financial Reporting Standards (IFRS) in over 100 countries, this demand has been met and, moreover, the

introduction of IFRS has been one of the most significant changes in the history of accounting (Daske, Hail, Leuz, & Verdi, 2008). Since January 1, 2005, listed companies in the European Union and many other countries have been mandated to report under IFRS (Jeanjean & Stolowy, 2008). According to Chen, Tang, Jiang, & Lin (2010), the accounting quality has been improved significantly since and as a result of the implementation of IFRS.

De George, Li, & Shivakumar (2016) stated that, after having summarized the literature regarding IFRS, the majority of companies which have adopted IFRS benefit in terms of (i) lower costs of capital, (ii) improved transparency, (iii) improved comparability of financial reports, (iv) better cross country investments, and (v) increased attention from foreign analysts. To further improve the quality of financial statements and the associated benefits of companies that use IFRS, the International Accounting Standards Board (IASB) keeps improving the standards. For annual reports from January 1, 2018, the two major standards IFRS 9 ‘Financial instruments’ and IFRS 15 ‘Revenue from Contracts with

Customers’ became effective. In this paper, the focus is on the disclosure quality of IFRS 15, which will be determined using a disclosure quality index. The focus on IFRS 15 is essential because this standard is about revenue, which is one of the most important indicators of the functioning of a company as it is generally the largest earnings component (Stubben, 2010; Aarab, Bisseur, & Ter Hoeven, 2015). With this new standard, the IASB wants to ensure more consistent revenue recognition, better comparability between companies and, ensure that companies provide more information regarding revenue from contracts with customers.

Philip Takken, partner at KPMG, said mid-2017 that: “IFRS 15 should be at the top of the agenda for every company in the coming months. Not only to prevent a possible crisis within the department that deals with the external reporting, but also to ensure that the company can benefit from possible opportunities that arise during the implementation” (KPMG: Bedrijven onderschatten gevolgen invoering IFRS 15, 2017). Accordingly, research by Roozen & Pronk (2018) shows that IFRS 15 is expected to have a material impact on 32% of the investigated FTSE 100 index companies, excluding financial institutions. These results are based on IAS 8 ‘Accounting policies, changes in accounting estimates and errors,’ which

U

(6)

requires companies to provide known or reasonably predictable information relevant to assessing the potential impact of new standards on first-year financial statements application (Roozen & Pronk, 2018). The impact of 32% is quite high considering that all industries are part of their research and the impact is expected to be larger in industries where the allocation of revenue from contracts with customers requires a professional interpretation (EY, n.d.). The impact of IFRS 15 is mainly because of (a) the more detailed guidelines on identifying performance obligations, (b) activating costs for obtaining contracts and (c) the accounting of payments to customers as a reduction in revenue (Roozen & Pronk, 2018). Telecom,

construction, technology, and utilities are some of the industries the impact of IFRS 15 is expected to be quite large, due to complex and large sales contracts (EY, n.d.). For example, Vodafone, operating in the telecom sector, reported a revenue decline of 3.4% (389 million) in the third quarter of 2018 due to the transition from IAS 18 to IFRS 15 (Vodafone Group Plc, 2019). Although this study only looks at the first mandatory application of IFRS 15, this example of Vodafone shows how big the influence of the new standard can be.

The relevance of this paper is supported by the research of the Financial Reporting Council (2018), who reviewed interim disclosures in the first year of application of UK-listed companies. In their report, they felt that some companies insufficiently explained the impact of adopting the new principles. Their key findings were that some disclosures could be improved, these disclosures are information about transition adjustments, the explanation of the changes made to accounting policies, information about performance obligations and the impact on the balance sheet (Financial Reporting Council, 2018). Although interim

disclosures require less explicit information in accordance with IAS 34, it shows that not all companies sufficiently anticipate the new standard. Therefore, the results of this study can be useful for regulatory bodies such as the Financial Reporting Council.

In modern corporations, the board of directors is seen as one of the most important governance mechanisms. One of the main principles of corporate governance is to enhance shareholder trust through business transparency. As IFRS 15 improves business transparency, corporate governance is expected to be positively related to IFRS 15 disclosure quality (Aras & Crowther, 2008). Three aspects of corporate governance that are frequently discussed are board independence, financial expertise on the audit committee and, board gender diversity. The board of directors is responsible for monitoring the management and providing them with strategic advice. Research has shown that independent board directors are better able to monitor the management and the financial reporting process (Nicolas, Belen, & Marta, 2016). Also, there is demand for financial experts in the audit committee, because they are expected

(7)

to be better able to monitor the company’s accounting controls and financial reporting

(Garcia-Sánchez, Martinez-Ferrero, & Garcia-Meca, 2017). Webb (2004) found that boards of directors of socially responsible companies have a relatively higher percentage of outside and female directors. In line with these results, Nicolas, Belen, & Marta (2016) concluded that board independence and board gender diversity have a positive influence on the board effectiveness, they subsequently found a significant positive relationship in their study between board effectiveness and boards protecting the interests of shareholders and

stakeholders. However board diversity is expected to result in a higher disclosure quality of IFRS 15, this effect might be mitigated if companies are forced to select women. Ahern & Dittmar (2012) found that board characteristics changed due to the relatively small pool of possible female board candidates, resulting in younger and less experienced boards. This change in board characteristics resulted in a lower accounting quality (Ahern & Dittmar, 2012). Therefore, it is expected that if companies are operating in a country with a board gender quota, this weakens the positive relationship between gender diversity and IFRS 15 disclosure quality.

This research investigates the influence of board independence, financial expertise on the audit committee, and board gender diversity on the disclosure quality of IFRS 15.

Furthermore, board gender diversity is controlled for the moderating effect of the presence of a board gender quota. This research is based on the results found in previous studies and is supported by the agency theory, stakeholder theory, legitimacy theory, and the Upper

Echelons theory. To get more relevant results, this paper will focus on companies operating in the industries in which IFRS 15 is expected to have the most impact. These are the telecom, construction, technology and utility sectors. The purpose of this paper is to make an early contribution to the disclosure quality of IFRS 15 and contribute to the existing corporate governance literature by examining the following research question:

What is the influence of different corporate governance mechanisms on the disclosure quality of IFRS 15

2. Research setting

The research setting of this research is further explained in this chapter. First a short introduction about how IFRS 15 was developed followed by the most important differences with previous standards, then a brief explanation of the corporate governance mechanisms is given, followed by the academic contribution of this research. Finally, the research structure is briefly elaborated.

(8)

2.1 IFRS 15 – Revenue from contracts with customers

IFRS 15 is the result of a conjunction project between US FASB and IASB (Mattei & Paolomi, 2018). This conjunction project was set up because both parties used a different standard which could compromise international comparability of financial statements due to a lack of high-quality comparable financial information (Mattei & Paolomi, 2018). Both parties signed in 2002 the Norwalk Agreement which was the start of setting up IFRS 15. The first discussion paper ‘Preliminary Views on Revenue Recognition in Contracts with Customers’ was published on 18 December 2008, followed by an exposure draft ‘Revenue from Contracts with Customers’ published on 24 June 2010 (Deloitte, 2014). After the comment period the IASB decided to re-expose the updated proposal, which was published on 14 November 2011. On all publications, the IASB received a lot of comments which they have taken into account and led to some important changes in the standard (Deloitte, 2014). Finally, on 28 May 2014, IFRS 15 ‘Revenue from Contracts with Customers’ was issued and initially effective for annual reports on or after January 1, 2017, but due to the difficulty of transition, it has been postponed to annual reports on or after January 1, 2018 (Deloitte, 2014).

The new standard has replaced all previous revenue requirements in IFRS and is applied to all revenues arising from customer contracts, except contracts that are in the scope of other IFRS, such as IFRS 16 ‘lease contracts’ (Greiner, 2018). The standards that are replaced by IFRS 15 are IAS 11 ‘construction contracts’, IAS 18 ‘Revenue’ and

interpretations related to these standards. With the introduction of IFRS 15, the IASB wants to ensure consistent revenue recognition, better comparability between companies and providing more information (Roozen & Pronk, 2018).

In Appendix I, a further explanation, the main features, and the application of IFRS 15 can be found. During the application of IFRS 15, professional judgment is required in various areas including the identification of performance obligations, variable considerations and when to recognize revenue. Furthermore, disclosure standards are more extensive in both quantitative and qualitative terms compared with previous standards (Corby & Maithripala, 2019).

2.2 Corporate governance mechanisms

Various corporate governance mechanisms are expected to influence the disclosure quality of IFRS 15 positively. The first mechanism is board independence. Prior research has shown that board independence is positively related to better reporting quality (Beasley, 1996; Klein, 2002). This might be due to a larger stakeholder orientation among external directors

(9)

than among internal directors. Therefore, a positive association can be expected between board independence positively and the disclosure quality of IFRS 15. Kusnadi, Leong, Suwardy, & Wang (2016) found that financial expertise at the audit committee results in better financial reporting quality. According to Sánchez, Martinez-Ferrero, & Garcia-Meca (2017), this relation can be explained due to financial experts being able to monitor the company’s accounting controls and financial reporting better. Therefore it is expected that financial expertise on the audit committee is positively associated with the disclosure quality of IFRS 15.

The third mechanism is gender diversity. Cabeza-García, Fernández-Gago, & Nieto (2018) found evidence for a positive influence of female directors on CSR disclosure quality, therefore it is expected that gender diversity also has a positive influence on IFRS 15

disclosure quality. The effect of gender diversity on IFRS 15 disclosure is expected to be mitigated by the presence of a gender quota based on research by Ahern & Dittmar (2012) who found that a gender quota results in a lower accounting quality. The findings of this research might be interesting for standard setters because the results can be used to find out how the accounting quality can be improved.

2.3 Academic contribution

Revenue is one of the most important indicators of the performance of a company. According to Wagenhofer (2014), the previous regulations for revenue recognition in the financial statements contained several fundamental defects. These previous regulations were mainly IAS 11 ‘construction contracts’ and IAS 18 ‘Revenue’ and are replaced by IFRS 15. It is expected that IFRS 15 entails significant changes, in particular for companies with a lot of revenue from long term contracts. With the new standard, the IASB wants to ensure more consistent revenue recognition, better comparability between companies and companies providing more information (Roozen & Pronk, 2018). At first, the results of this study show the disclosure quality of the first mandatory implementation of IFRS 15. The results will be of importance for the users of the annual statements, including shareholders, analysts, investors and financial institutions. Besides, this research is interesting for regulatory bodies such as the ESMA and standard setters as the IASB. Based on the results of this study these institutions can determine whether changes or possible support for the companies are required. Also, the disclosure index, which is used to determine the disclosure quality of IFRS 15, is provided in appendix III, which therefore can be used in future research. The disclosure index is based on the disclosure requirements of IFRS 15, which are further elaborated into clear criteria on which companies are assessed.

(10)

Furthermore, the unique setting of this research also ensures an academic contribution. Since annual reports from January 1, 2018, are the first for which the application of IFRS 15 is mandatory, this is a unique setting for conducting research. The setting is not only unique because it is the first mandatory year of application, but also because there is little peer-information available which could lead to a lot of practical challenges for companies to meet the disclosure requirements. In that perspective, the first application of IFRS 15 could be used as a reference for future years. Therefore, another important aspect of this research is that it contributes to the limited existing literature regarding the implementation of new financial reporting standards and the influence of different corporate governance mechanisms on this. Although it will be quite difficult to generalize the results to other settings, this research contributes to limited existing literature and function as a basis for possible future research regarding the implementation of new (financial) reporting standards.

2.4 Research structure

The remainder of this paper is as follows; chapter 3 explains the theoretical framework in which the most important concepts and theories are explained. Based on these theories and previous research, hypotheses are formed which are subsequently examined and answered. Chapter 4 describes the used research method. It also explains how the variables are measured and which statistical tests in chapter 5 are discussed. Finally, in chapter 6, the conclusion of this research is given. Which consists of the results, limitations, and recommendations for future research.

3. Theoretical framework

In this chapter first, the different theories will be elaborated upon. Followed by a brief explanation about the different board structures and how this research has dealt with it. Then the background literature is reviewed followed by the development of the hypotheses.

3.1 Theories

In this research four theories are used to establish and substantiate the developed hypothesis. These theories are the agency theory, the stakeholder theory, legitimacy theory, and the upper echelons theory. The first three theories will mainly contribute to the

representation of the role and influence that a board has within an organization and what their influence is on the disclosure of extra (voluntary) information. The Upper Echelons theory is used to explain why different board characteristics can lead to differences in decision making and reporting style between companies.

(11)

3.1.1 Agency Theory

The agency theory of Jensen & Meckling (1976) is one of the most used theories in corporate governance literature. This theory explains among others why companies would be willing to invest in improving their disclosure quality. Healy & Palepu (2001) state that information asymmetry and agency conflicts between firm managers and outside investors cause the demand for financial reporting and disclosure. The agency theory states that

problems and inefficiencies arise if there is information asymmetry between the managers and owners of the firm (Jensen & Meckling, 1976). Jensen & Meckling (1976) distinguish these two roles respectively as principal and agent. The information asymmetry arises because the managers are operational on a daily basis while the owners have little insight into daily business activities (Boucková, 2015). Agency theory provides protection against the two negative consequences of information asymmetry, namely adverse selection and moral hazard (Scott, 2014). Moral hazard occurs due to information asymmetry which enables managers to pursue self-interest (Scott, 2014). In case of adverse selection, the principal is not able to assess the effectiveness of the agents work, which may lead to the principal selecting an agent not capable for a certain project or position (Boucková, 2015). Both consequences of

information asymmetry result in agency costs. Boucková (2015) defines agency costs as all costs incurred by the agency problem, which may consist of bonding costs, monitoring costs and residual loss.

From the perspective of the agency theory, accounting information reduces the agency costs resulting from outside financing, namely the costs of equity and debt. In the case of debt financing, agency costs origin from managers who act in favor of shareholders which leads to inefficient actions for firm creditors (Jensen & Meckling, 1976). These inefficient actions might lead to claim dilution, asset substitution and debt overhang, which reduces contract efficiency (Christensen, Nikolaev, & Wittenberg-Moerman, 2016). Healy & Palepu (2001) state that financial reporting and disclosure are important means for firms to communicate their performance and governance to outside investors.

Mitigating information asymmetry and its consequences can be done by mandated information disclosure, such as IFRS (Brown, Goetzmann, Liang, & Schwarz, 2008). Mandatory disclosure is, according to Brown, Goetzmann, Liang & Schwarz (2008), an important tool for regulators that intends to allow market participants assessing managers risks without constraining their actions and so reduce the information asymmetry. In addition, Bushee & Noe (2000) found that institutional investors are more likely to invest in forms with

(12)

more forthcoming disclosure. Besides, agency costs can be reduced by increasing the amount of voluntary information disclosure (An, Davey, & Eggleton, 2011).

The role of the board is, according to the agency theory, to monitor whether the interests of shareholders are being pursued. From this point of view, a board would mainly insist on disclosing extra (financial) information. Research by Nicolas, Belen, & Marta (2016) shows that board independence and board gender diversity result in higher board

effectiveness, which subsequently results in better protecting shareholder interests. Besides, Abdul Rahman and Haneem Mohamed Ali (2006) argue that a correct composition of the audit committee improves the credibility and accountability of financial information, which reduces information asymmetry. As revenue is one of the most important indicators of the functioning of a company, it is expected that companies with good corporate governance will try to mitigate the agency problem by providing more relevant information to its shareholders. This subsequently would lead to a higher disclosure quality of IFRS 15.

3.1.2 Stakeholder theory

While the Agency theory only focuses on the relationship between directors and shareholders, the stakeholder theory recognizes the importance of the relationship between directors and other stakeholders (Corfield, 1998). Freeman (1984) describes a stakeholder as 'any group or individual who can affect or is affected by the achievement of the organization's objectives.' In the stakeholder theory, financial reporting quality is seen as an important tool in the relationship between the company and the stakeholders. According to the stakeholder theory, the continuity of a company depends on the support of its stakeholders such as

employees, suppliers, investors, government and customers (Freeman, 1984). Companies will therefore not only have a responsibility towards their shareholders but also to the other

stakeholders. In fact, the stakeholder theory is an extension of the agency theory. The role of the board is, according to the stakeholder theory, not only to monitor whether the interests of shareholders are being pursued, but should also ensure that the interests of other stakeholders are pursued. Because it is difficult for stakeholders, due to being outsiders, to determine whether a company is acting according to its interests, it is important for a company to publish relevant information to the stakeholders. From this point of view, a board would mainly insist on disclosing extra (voluntary) information. Corfield (1998) stated in his paper that long term profitability depends on the extent to which a company looks at the interests of the various stakeholders.The trust and support of the stakeholders can be increased by publishing additional relevant information (Reynolds, Schultz, & Hekman, 2006). Based on the stakeholder theory, it can also be expected that

(13)

better corporate governance leads to a higher disclosure quality of IFRS 15. This same expectation as with the agency theory is supported by Nicolas, Belen, & Marta (2016) who imply that boards which are effective in protecting shareholder value are also effective in protecting stakeholder value, which shows that the agency theory and stakeholder theory can be expected to have the same outcome.

3.1.3 Legitimacy theory

Another widely used theory is the legitimacy theory, which also has a lot of overlap with the agency theory and stakeholder theory. In the legitimacy theory, it is assumed that the survival of a company depends on the support of society (Dowling & Pfeffer, 1975). This theory assumes that a company owes its existence to the extent to which the company meets the norms and values of society. The legitimacy theory is built upon the notion that companies operate via a social contract where it has agreed to perform multiple socially desired actions in exchange for approval of its objectives, its survival and other rewards (Guthrie & Parker, 1989). Therefore, companies are constantly busy with meeting the changing demands of society (Deegan, Rankin, & Tobin, 2002).

According to De Waard (2014), companies need a ‘license to operate’ in order to survive and will therefore be inclined to meet the interests and demands of society.

Companies can exert influence on the environment through external reporting, but it is also influenced by pressure from the environment. By adopting the values and norms, the company receives respect in the form of acceptance and customers in exchange (Cormier & Gordon, 2001). External reporting is an important means of communication for companies in this regard.By issuing an extensive (financial) report, companies show society that they act according to the norms and values of society. Companies can influence information and communication by their choice of word and method of presentation of their (voluntary)

reporting (De Waard, 2014). Companies will, therefore, tend to manipulate the information by diverting attention from important issues and by shedding light on more positive points

(Deegan, Rankin, & Tobin, 2002). However, a board should operate from the point of view of society and supervise that correct and relevant information is published to maintain the

company’s legitimacy.

According to the legitimacy theory, the company communicates with society from a legitimate point of view and the company will do so by issuing a (financial) annual report. An extensive application of IFRS 15, measured as the disclosure quality of IFRS 15, can be seen as part of the means of communication to prove its legitimacy and to justify its actions.

(14)

3.1.4 Upper echelons theory

Previous studies have shown that board characteristics influence the strategic choices, decision making, innovation activity and performance of a company (Bamber, Jiang, & Wang, 2010). Hambrick & Mason (1984), one of the first researchers to visualize the aforementioned relationship, have developed the upper echelons theory. This theory states that strategies and the decision making of companies are significantly influenced by the characteristics and properties of the board. Differences in demographic characteristics

(including age, gender and origin) would lead to different preferences among board members (Hambrick & Mason, 1984). Bamber, Jiang & Wang (2010) investigated the relationship between board gender diversity, board independence and size of the board and the various financial voluntary disclosure styles of companies. Based on their results, they conclude that board members have a large influence on the chosen voluntary disclosure style: the unique disclosure styles of the board members are associated with observable demographic characteristics of their personal backgrounds. The upper echelons theory could, therefore, explain the difference in voluntary disclosure between companies because the reporting strategies of a company partly depend on the characteristics of the board. As the disclosure quality of IFRS 15 is partly voluntary, it can be expected that managers will try their best to improve the disclosure quality of IFRS 15.

3.2 Board structures

In this study, European listed firms are investigated which are established in 12 different countries with different regulations for board structures. It is therefore considered necessary to explain the different board structures before discussing the relevant literature and developing hypotheses. In Europe, Great Britain is a prominent country where a single board system is applied, consisting of executive and non-executive directors, countries like Ireland use the same one-tier approach (Jungmann, 2006). In other countries such as Germany, the Netherlands, Austria, Finland and Denmark, a two-tier board consisting of a management board and supervisory board is used (Jungmann, 2006). In addition, there are countries such as France, Belgium, and Spain where companies are free to choose which board structure they want to use.

Unless there are a lot of differences between both board structures, there are also some similarities. One of these similarities is that both board structures consist of executive and monitoring bodies. In one-tier boards, the non-executive directors represent the monitoring body; in two-tier boards. the supervisory board is the monitoring body (Jungmann, 2006). So, in one-tier boards, the executive directors participate in the daily business of the company as

(15)

the management board does in a two-tier system, non-executive directors and the supervisory board only participate during board meetings (Millet-Reyes & Zhao, 2010). Where it is not possible to be both employee and non-executive director in the one-tier system, it is possible as an employee to be part of the supervisory board in the two-tier system (Millet-Reyes & Zhao, 2010). This makes that the composition of both so-called monitoring bodies is not entirely the same.

Despite the fact that there are some differences between the composition of the two monitoring bodies, the study by Jungmann (2006) led to the conclusion that it is impossible to consider one of the two systems for corporate control superior to the other. Since neither leads to better monitoring results, it has been decided not to make any further distinction between one-tier and two-tier boards during this research. As this research focusses on the influence of the composition of the monitoring body on IFRS 15 disclosure quality, only non-executive directors or the supervisory board will be considered for measuring board independence and gender diversity.

3.3 Background literature and hypothesis development

In the following paragraphs, the literature is reviewed and eventually leads to the hypotheses that will be investigated in this paper. Because IFRS 15 is a new standard, there is little literature regarding IFRS 15 disclosure quality. Due to the absence of available literature regarding the relationship between the independent and moderating variables and the

disclosure quality of IFRS 15, the literature on comparable (financial) reporting has been used to support the hypothesis. IFRS 15 is a mandated standard, but due to the large and difficult nature of the standard, there is also a voluntary aspect in properly applying IFRS 15.

Therefore this paper uses voluntary (environmental) disclosure and financial reporting quality as equivalents for the disclosure quality of IFRS 15 to support the expected relationships. Since there is little literature regarding IFRS 15 disclosure quality and it results in significant changes in annual reports, it is important to conduct research about the new standard.

Moreover, because revenue is the largest earnings component for most companies, it is one of the most important indicators of the functioning of a company (Stubben, 2010; Aarab,

Bisseur, & Ter Hoeven, 2015). 3.3.1 Board independence

With the introduction of the Sarbanes-Oxley Act (SOX) companies in the United States of America are mandated to have a majority of independent directors on the board, aimed at improving accountability by directors in an attempt to protect the interests of shareholders (Jiraporn, Chintrakarn, Tong, & Treepongkaruna, 2017). In Europe there is no

(16)

regulation yet to have a mandated majority of independent directors on the board.

Nevertheless, there has been some discussion at the European Commission to follow this SOX requirement.

The board of directors task is to supervise corporate management and their actions and decisions (Rupley, Brown, & Marshall, 2012). They serve an important role in reducing the information asymmetry between the company and its stakeholders, including shareholders. The board is responsible for creating alignment between the interests of shareholders and the CEO (Fama & Jensen, 1983). One of the main reasons to place independent directors on boards is to provide outside perspectives to assist a company in achieving its strategic goals (Rupley, Brown, & Marshall, 2012). These outside perspectives could include a desire to provide additional and transparent information to stakeholders, which is in line with Wang and Dewhirst (1992), who state that the stakeholder orientation with external directors is larger than among internal directors. Besides, appointing independent board directors is essential for effectively monitoring and controlling managers (Fama & Jensen, 1983).

Though, there is still no consensus on the definition of ‘independence’. The majority of companies define their directors as ‘executives’ or ‘non-executives’ (Kang, Cheng, & Gray, 2007). The term board independence usually refers to non-executive directors (García-Meca & Sánchez-Ballesta, 2009). In this respect, board independence refers to the extent to which a board consists of non-executive directors who do not have a relationship with the company outside the role of director (Davidson, Goodwin-Stewart, & Kent, 2005)

Research has shown that board independence has a positive influence on reporting quality (Beasley, 1996 and Klein, 2002). In the research of Rupley, Brown & Marschall (2012) a positive relation is found between board independence and voluntary environmental disclosure quality. Besides, Petra (2007) found that a higher proportion of independent

directors is associated with higher earnings informativeness to the stock market. As the task of the board is among others to monitor the reporting/ disclosure quality and due to the

aforementioned positive relationship between the proportion of independent board directors and environmental disclosure, earnings informativeness and financial reporting quality, it is expected that a more independent board will positively affect IFRS 15 disclosure quality. This leads to the first hypothesis, which is formulated below:

(17)

3.3.2 Financial expertise on the audit committee

The audit committee is enabled by the board of directors to acquire expertise and resources to improve the quality of financial statements. The aim of the audit committee is to increase the credibility and visibility of information released by companies, which has become an important issue after a number of major and known fraud scandals (De Vlaminck & Sarens, 2015). The audit committees task is to evaluate the credibility and relevance of the information in the financial statements, which they do together with both internal and external auditors (Piot & Janin, 2007).

According to regulators, there is a need for more financial experts on the audit committee because they lead to better board oversight and think better about the interests of shareholders (Gütner, Malmendier, & Tate, 2008). Financial experts are expected to be better able to monitor the company’s accounting controls and financial reporting (Garcia-Sánchez, Martinez-Ferrero, & Garcia-Meca, 2017). Kusnadi, Leong, Suwardy and Wang (2016) found evidence for financial expertise at the audit committee enhancing financial reporting quality. Furthermore, the presence of at least one audit committee having financial expertise is

negatively related to aggressive earnings management (Bédard, Chtourou, & Courteau, 2004). In line with these findings, Krishnan (2005) found that firms with audit committees with financial expertise are less likely to experience internal control problems and Farber (2005) shows that fraud firms have fewer financial experts on the audit committee. Moreover,

Carcelo, Hollingsworth, Klein and Neal (2006) found evidence that abnormal accruals decline after financial expertise is added to the audit committee. They suggest on the basis of these results that, due to the salutary benefits of better monitoring by financial experts, this may directly lead to an improved financial reporting quality.

One of the roles of the audit committee is to show independence in communication and oversight during the audit process, including contact with the internal and external auditor (DeZoort & Salterio, 2001). As a financial expert generally has more knowledge and

experience with regard to the application of financial reporting standards, contact with the internal and external auditor is expected to be more effective with the presence of a financial expert in the audit committee. In addition, all companies in the sample have marked IFRS 15 or ‘revenue recognition’ as KAM, indicating that IFRS 15 is an important issue. Usually, the audit committee is closely involved in important issues such as the first mandatory application of a new high impact standard as IFRS 15. In addition, limited peer information is available for the first mandatory application, which causes companies to face practical challenges to meet the interpretation requirements.

(18)

Previous literature on the impact of the presence of a financial expert on the audit committee on general financial reporting quality, internal control problems and fraud firms, combined with the key role of the audit committee in implementing a new high impact standard leads to the second hypothesis. Which is elaborated below:

H2: Financial expertise at the audit committee is positively related to the disclosure quality of IFRS 15.

3.3.3 Gender diversity

One of the most used diversity characteristics is gender; gender diversity refers to the proportion of men and women on the board (Appuhami & Tashakor, 2017). A lot of women have been working full-time for decades. However, only 25% of board members of listed European companies in 2017 are women (Smith, 2018). This shows that there is a lot of work to do to get more women on the board, important for this is to understand the impact of more gender-diverse boards.

Since governments also see gender diversity as an important topic, some governments try to promote gender diversity through legislation in two different ways. The first way is to use a soft quota which can be defined as a comply or explain law as applied in the

Netherlands, the second way is a hard gender quota which might result in for example fines if the quota is not met. Norway and Germany are countries which have set a hard gender quota. Due to the differences, there are three different populations in the sample, the first consists of companies that do not have to comply with a gender quota, the second consists of companies that have to deal with a soft gender quota and the third consists of companies that have to comply with a hard quota. These three different populations might bias the results, which will be tested in the fourth hypothesis.

It is expected that gender diversity provides more important resources to the committees, which can be information, external networks, skills and constituencies that increase understanding of the marketplace, creativity and innovation, human capital, and therefore improve general performance, such as monitoring and reporting (Appuhami & Tashakor, 2017). In line with this, Adams & Ferreira (2009) found evidence that female directors put more efforts to monitoring which results in female directors providing better oversight over managers, these results are based on a sample of Standard & Poor’s (S&P) 500, S&P MidCaps and S&P SmallCap firms for the period 1996-2003. Bear, Rahman, & Post (2010) and Boulouta (2013) found evidence that gender diversity results in higher

(19)

because women usually have a different educational and professional background, and are more participatory and democratic in decision-making processes. Adams and Ferreira (2009) found that female directors provide better oversight over managers. Women of boards of the Fortune 500 (2001) also seem more inclined to take into account the demands and

expectations of society and are better able to meet the needs of society , which is in line with the aforementioned agency theory, stakeholder theory and legitimacy theory (Bernardi, Bean, & Weippert, 2002). Another important difference between men and women is the gender difference in risk-taking. Charness and Gneezy (2012) compared experiments with field data and found that women are much more risk-averse in their investment decisions than men. Srinidhi, Gul and Tsui (2011) found that firms with greater female participation on their boards exhibit higher earnings quality. One of the explanations for this finding is that women are more likely to act decisively in improving earnings quality because they are more risk-averse. With all these positive assumptions about gender diversity, it can be expected that this will also ensure a higher (financial) reporting quality.

In the research of Cabeza-Garcia, Fernández-Gago & Nieto (2018) a positive relation is found between the relative number of female directors and CSR disclosure in Spanish boards of firms listed on the IGBM. The study of Frias-Aceituno, Rodríguez-Ariza, & Garcia- Sánchez (2013) shows a positive relationship between the number of women on the board and the issuing of an integrated report. These results are in line with the results found by Post, Rahman, & Rubow (2011) who found a significant positive relationship between female directors and Environmental Corporate Social Responsibility. The positive influence of gender diversity on disclosing extra (financial) information combined will all positive assumptions about gender diverse boards leads to the expectation that gender diversity will have a positive influence on the disclosure quality of IFRS 15. Which leads to the following hypothesis:

H3: Gender diversity in the board is positively related to the disclosure quality of IFRS 15. 3.3.4 Board gender quota

Gender diversity is also widely discussed by regulators and politicians. In some countries, this discussion has led to setting a board gender quota for firms. A board gender quota requires that women make up a minimum share of the board (Hughes, Paxton, Clayton, & Zetterberg, 2019). In Europe, some countries have set a soft or a hard gender quota, these countries can be found in table 1. The European Commission proposed a law that requires a minimum of 40% of the underrepresented sex on boards of listed companies in the European

(20)

Union (European Commission, 2012). However, this proposal did not find enough support. Ahern & Dittmar (2012) found evidence that mandatory female representation differs from voluntary representation during their research using Norwegian firms. This is caused by the fact that Norwegian firms are not entirely free to choose their board members because they have to comply with a gender quota of 40%, which results in firms being forced to choose women with less experience over their male counterparts (Ahern & Dittmar, 2012). More evidence is found in the research of Farrel & Hersch (2005) who looked at new members of corporate boards and discovered that gender influences these additions, which is mainly caused by internal or external pressure in companies from Fortune 500 and Service 500 lists from 1990-1999. During their research, they found that the probability of a female being selected increases when a woman left the board. They also found that the probability of a female being selected decreases with the percentage of women on the board being higher (Farrel & Hersch, 2005).

According to Ahern & Dittmar (2012), board characteristics changed due to the relatively small pool of possible female board candidates. The average new female directors were younger and less experienced. This resulted in more growth at the firms, more

acquisitions and a lower accounting quality (Ahern & Dittmar, 2012). Because of the lower accounting quality, it is expected that a mandatory board gender quota has a mitigating effect on the expected positive effect of gender diversity on the disclosure quality of IFRS 15. This brings us to the last hypothesis:

H4: A mandatory board gender quota reduces the positive relationship between gender diversity and the disclosure quality of IFRS 15.

4. Research methodology

To investigate the influence of the different corporate governance mechanisms on the disclosure quality of IFRS 15, a disclosure index has been set up. This section consists of an explanation of the selected sample, the research method, the data collection procedure, the independent, dependent and the control variables. Most of the data used for the different variables are manually collected from annual reports.

4.1 Research population and data collection

To test the hypothesis formulated in the previous chapter, a sample with 52 Stoxx 600 listed firms is used. The selected firms are active in the telecom, construction, technology and utility sectors. Companies in these industries are expected to be affected most by IFRS 15 because in these industries there are more large and complex contracts with customers. To be

(21)

able to generalize the results of this study and compare the differences between annual reports, it is required that the independent auditor of the firm has reported IFRS 15 as key audit matter or reported a material impact due to IFRS 15. This requirement is necessary because key audit matters provide insights on significant estimates and risks reported in the financial statements (International Auditing and Assurance Standards Board, 2018).

Therefore, it is expected that marking IFRS 15 or revenue recognition as key audit matter ensures that IFRS 15 is a relevant topic for all companies included in the sample.

Furthermore, for each selected firm an annual report from as of January 1, 2018, has to be available because it is the first year of mandated adoption of IFRS 15. Table 1 below shows the frequency and ratio of the different industries.

Table 1 Industries

Industry Frequency Percentage

Construction 8 15 Technology 18 35 Telecom 15 29 Utilities 11 21 Total 52 100 4.2 Dependent variable

To determine the disclosure quality of IFRS 15 a disclosure index has been developed consisting of the different components of IFRS 15. The disclosure quality will be measured via a quality score derived from the disclosure index. Annual reports will be analyzed and judged based on 27 relevant evaluation criteria from IFRS 15, which together complete the disclosure index which can be found in Appendix III. These relevant criteria cover three different themes which are specified in IFRS 15: Contracts with customers, Significant judgments and changes in the judgments made in applying the standard and recognized assets from costs to obtain or fulfill a contract. In appendix II, the scores per theme and the total score for all different companies are presented. For each of these three themes, the best practice has been chosen, which can be found in Appendix IV. An organization can receive points for all disclosure requirements from IFRS 15.110 to IFRS 15.129. The index score is measured via an ordinal scale. If the range is from 0 to 1, there is only a check if the object discloses the specific point. If the range is from 0 to 2, in most cases the object is able to score

(22)

for both quantitative and qualitative disclosures, in other cases which consist of two (or more) points a firm has to include a description of multiple components. It is assumed that the more points a company scores, the better the disclosure quality of IFRS 15 (Beattie, McInnes, & Fearnley, 2004). IFRS 15 disclosure quality will be coded as ISCORE.

The use of a disclosure index is partially subjective, which is due to the

comprehensiveness and extent of the information in the annual reports to be evaluated. Besides, an index can have problems with its validity and reliability. Despite the subjective nature, previous research has proven that it is a reliable tool to measure disclosure quality (Eng & Mak, 2003; Huafang & Jianguo, 2007; Wang, Sewon, & Claiborne, 2008). In addition, indices are seen as a very suitable measuring instrument for researching annual reports (Marston & Shrives, 1991; Beattie, McInnes, & Fearnley, 2004).

Although previous research has shown indices are a great tool, some measures are taken to reduce the subjectivity and check the reliability of the disclosure index. First of all, multiple workshops have been organized to map out important aspects of developing a disclosure index, also possible subjectivity problems of using a disclosure index have been addressed, followed by team discussions. Furthermore, to check if the used index is a reliable measuring instrument, a pilot test has been performed on multiple firms, which have been peer reviewed. According to Marston & Shrives (1991), an index is considered to be reliable if other researchers would be able to replicate the results. The pilot test provided insight into the consistency of data collection by the different researchers. Because every researcher had to fill in the checklist for the annual reports of all three companies, the researchers were able to review each other and see if every researcher would come to the same index score. The pilot test showed that there were some small differences between the different index scores per company. These differences have been discussed and in the end, the differences in interpretation and allocation of points were eliminated. Therefore, the disclosure index is considered reliable. In addition, overlap has been created for the first three firms of every researcher to exclude remaining bias. The last measure to reduce subjectivity is that the index is free of weighting, which makes every component of equal importance and therefore less subjective.

4.3 Independent variables

As the information about the board composition in the financial year 2018 for the selected firms is not yet available in BoardEx, the data for determining board independence is hand-collected. Companies disclose information about board independence based on the corporate governance code applicable in the country of origin. If a company has a one-tier

(23)

board, only the independence of non-executive directors is included. In case a firm has a two-tier board, only the directors in the supervisory board are included. Following Klein (2002), %BDOUT looks at the percentage of outside directors.

The second independent variable is financial expertise on the audit committee.

Badolato, Donelson, & Edge (2014) define a financial expert as an individual with experience in creating, auditing, using or overseeing the creation of financial reports. The presence of a financial expert is measured using BoardEx and someone is called a financial expert if he or she has ‘Functional Experience’. Financial expertise in the audit committee will be measured by looking at the presence of at least one financial expert on the audit committee. FINEX will be given a 1 if one or more audit committee members has financial expertise and 0 if no one has.

For gender diversity, this research focusses on non-executive directors in case of a one-tier board and in case of two-tier boards this research focusses on the supervisory board. The reason for this is to get better results by looking at the monitoring bodies. Gender diversity will be measured by the proportion of women in relation to the total number of directors and is coded as %BDFEM.

4.4 Moderating variables

Three different reports were used to determine whether a board gender quota is in force in a country. The first report is the report by Senden & Kruisinga (2018), which contains information about the board gender quotas in all EU countries and has been prepared with the help of national experts in the field of board gender regulation. To make sure the data in this report is correct, it has been compared with the report of Jizi & Nehme (2017) who

summarized the developments regarding board gender quotas in Europe. To fulfill the triangulation of sources, these two reports are compared with a report by Deloitte (2016) which looked globally at gender quotas and other initiatives. Because the reports have been prepared by professionals and experts and the content is also consistent between the reports, the information is assumed to be correct. All companies in the selected sample meet the criteria to comply with the gender quota applicable in the country of origin. However, there are still some differences between the different gender quotas, below is explained how this problem is tackled.

In some countries, there is a so-called hard law and in some countries, there is a soft law. In case of a hard law firms get a penalty if they don’t comply with the law, this can be a fine or being excluded for some specific benefits. In countries with a soft law like the

(24)

quota only applies to non-executive directors (NED), in other countries the gender quota also applies to executive directors. For firms operating in a country with a hard gender quota GENQUO will be given a 1, for firms operating in a country with a soft gender quota, the code will be given a 0 and for firms operating in a country without a gender quota, the firm will be excluded. In table 2 the requirements per country are given based on the three different sources.

Table 2

Countries with a board gender quota

Country Gender quota Percentage Hard or Soft NED or both

Denmark No

Germany Yes 30 Hard NED

Finland Yes 40 Soft NED

France Yes 40 Soft NED

Ireland No

Italy Yes 33 Hard Both

Netherlands Yes 30 Soft Both

Norway Yes 40 Hard Both

Spain Yes 40 Soft NED

Sweden No

Switzerland No

United Kingdom No

Sources: Senden & Kruisinga (2018), Jizi & Nehme (2017) and Deloitte (2016)

4.5 Control variables

The first variable to control for is leverage. Goss & Roberts (2011) found that

companies are likely to raise their CSR disclosure to increase their chances of obtaining loans from lenders. Because firms with higher leverage are more likely to disclose more

information, a positive relationship is expected. To determine the leverage the total liabilities will be divided by the book value of the total assets and is called LEV.

The second variable to control for is firm size. In the paper of Prado-Lorenzo et al. (2009) a significant positive relation is found between firm size and the disclosure of greenhouse gas emissions. Therefore it is expected that firm size might have a positive

influence on the disclosure quality of IFRS 15. To determine firm size, total revenue for 2018 in Euros will be looked at. As revenues does not have a normal distribution, it is transformed into normal distribution by using the natural logarithm. The natural logarithm of revenues is coded as FSIZE.

(25)

Board size is the third variable to control for, which is measured as the number of non-executive directors. In prior literature, there are a number of different views on the influence of board size. Some researchers state that large boards contain more expertise than small ones (Dalton, Daily, Johnson, & Ellstrand, 1999), or that larger boards tend to be more effective in monitoring accruals (Xie, Davidson, & DaDalt, 2003). Other researchers claim that smaller boards are considered conductive to better monitor managers, due to smaller boards are associated with less free-riding between members, better exchange of ideas and lower coordination costs (Jensen M. C., 1993; Lipton & Lorsch, 1992). This claim is empirically supported by Vafeas (2000). Due to the above differences in prior literature, the influence of board size on IFRS 15 disclosure quality is considered indefinite.

The last variable to control for is industry type. In the paper of Roozen & Pronk (2018), it is shown that the impact of IFRS 15 differs per industry, besides they state that the focus of the different industries is on different components. Hence, industry type will be the last variable to control for in this research.

4.6 Data adjustments

Prior to the analysis and regressions, the data has been adjusted. First, revenues are subject to the process of winsorizing, to reduce the impact of outliers. For revenues, the lower and upper bound are calculated using the mean and plus/minus two times the standard

deviation. Four outliers that were higher than these bounds received the corresponding value. Furthermore, to calculate the moderating effect, gender diversity and the presence of a board gender quota are centered. According to Schielzeth (2010) centering these variables improves the interpretability of the regression coefficient, because without centering the main effects would be meaningless. Moreover, centralizing reduces the multicollinearity between the variables in the interaction.

4.7 Model and statistical analysis

To test if the different disclosure index data may be summed up, a Cronbach’s Alpha test will be done. Only if the score is above 0.7, it is allowed to sum up the different

components of the disclosure index. After this, a correlation test will be executed to test the cohesion between the different variables. This test also tests if there are any problems with multicollinearity. This means that there will be problems if some variables will be tested together because it is hard to determine which variable influences the dependent variable. With the following multiple linear regression model the hypotheses are tested:

(26)

ISCORE

= β0 + β1 * %BDOUT + β2 * FINEX + β3 * %BDFEM + β4 * GENQUO * %BDFEM + β5 * FSIZE+ β6 * LEV + β7 * BSIZE + β8 * INDUS

To determine if a relation is significant, three different significance levels are used. These are 0.10, 0.05 and 0.01. Because the sample used is quite small, it is allowed to use a P-value of 0.10 (Labovitz, 1968).

5. Results

In this chapter, the study results are presented. In the first paragraph, a reliability analysis on the disclosure index is performed. The second paragraph shows the descriptive statistics of the different variables. In the third paragraph, the different hypotheses are tested and explained. In the last paragraph, some additional tests are performed to test the robustness of the main results.

5.1 Disclosure index validity

To check the internal validity of the disclosure index the Cronbach’s Alpha has been calculated. Because some fields do not contain any values due to firms not being required to disclose about it, all missing fields are filled with the mean of the concerned subject to calculate the Cronbach’s Alpha. Normally, listwise deletion would be preferred when there are missing values, but with applying listwise deletion only 4 observations would remain. Therefore, missing values are replaced with the mean of the component. The Cronbach’s Alpha calculated for the checklist as a whole is presented in table 3.

Table 3

Reliability statistics – Cronbach’s Alpha

α N of items

Disclosure index 0.790 27

For any disclosure index, the Cronbach’s Alpha must be at least 0.7 before the various components may be added together (Cronbach, 1951). With a Cronbach's Alpha of 0.790, the disclosure index is internally valid, which means that conclusions drawn from this disclosure index can be considered reliable.

5.2 Descriptive statistics

Table 4 shows that the mean IFRS 15 disclosure quality score is 0.482. This means that companies achieve on average 48.2% of the points they can achieve. The minimum IFRS 15 disclosure quality score of 13.3% is scored by EDF and the maximum score of 75.5% is

(27)

scored by ATOS. With an average score of 48.2% on the disclosure index, this study shows that there is still room for improvement. By considering that variables that do not apply to a specific annual report are not included in the disclosure index score, an index score which is a lot higher would be expected. A possible explanation for these low scores could be that there are voluntary aspects in applying IFRS. In addition, qualitative and quantitative aspects are considered for many different components, while in many cases companies only choose for only one of the two possibilities. Furthermore, It is expected that companies will achieve a higher score in the coming years because they will become more familiar with the

requirements of the new standard.

The average amount of independent non-executive directors is 76.5% with a minimum of 12.5% and a maximum of a full independent non-executive board. The number of female non-executive directors varies from 0 to a majority of 67% with an average of 33% female directors. The fact that the average number of women in the supervisory board is already 1 in 3 shows that companies are improving their gender diversity, compared to the paper of Ferreira & Kirchmaier (2013) who found a female representation of 4.5% in 2000 increasing to an average of 8% in 2010 by looking at both executive and non-executive directors.

Table 4 Descriptive statistics

N Min Max Mean St. Dev.

ISCORE 52 0.133 0.755 0.482 0.152 %BDOUT 52 0.125 1 0.765 0.245 %BDFEM 52 0 0.667 0.333 0.126 LEV 52 0.112 0.843 0.612 0.159 FSIZE 52 19.76 24.78 22.79 1.408 BSIZE 52 3 20 10.37 4.044

This research includes three dummy variables, the presence of financial expertise in the audit committee, a hard or soft gender quota and industry type. An overview of the

classification and distribution of these dummy variables is given in table 5. The whole sample consists of 52 firms, for financial expertise data is only available for 38 firms. In 27 out of 38 audit committees, at least one person has financial expertise. In addition, 36 companies come from a country that has set either a soft or a hard gender quota. Of these 36 companies, 17 companies have to deal with a hard gender quota.

(28)

Table 5 Dummy variables

Variable Frequency Percentage

FINEX Valid 0 11 28.9 1 27 71.1 Total 38 100 Missing 14 Total 52 GENQUO Valid 0 19 52.8 1 17 47.2 Total 36 100 Missing 16 Total 52 INDUS Construction 8 15 Technology 18 35 Telecom 15 29 Utilities 11 21 Total 52 100 5.3 Correlation analysis

The correlation matrix is presented in table 6 on the next page. The correlation

statistics show whether there is a correlating relationship between variables. The results show that there is no significant relationship between the explanatory variables and IFRS 15

disclosure quality. For board gender diversity and leverage the correlation matrix shows a negative relationship where a positive relation is expected. For all the other explanatory variables the correlation is positive. However, as mentioned before both positive and negative relationships are not significant.

Furthermore, a correlation matrix shows the results of multicollinearity. Generally, multicollinearity between explanatory variables is indicated by a Pearson correlation score greater than 0.7. Multicollinearity is undesirable because it means that two explanatory variables show a strong correlation between one another and therefore it is hard to tell which variable explains the difference in the independent variable. As there is no Pearson correlation score over 0.7, based on the Pearson correlation score no extra regression models are required.

(29)

Table 6 - Correlation Matrix

1 2 3 4 5 6 7 8 9 10 11 12

1 ISCORE Pears. Cor 1

Sig. (2-tailed)

2 %BDOUT Pears. Cor .058 1

Sig. (2-tailed) .684

3 FINEX Pears. Cor .117 .263 1

Sig. (2-tailed) .483 .110

4 %BDFEM Pears. Cor -.164 -.019 -.246 1

Sig. (2-tailed) .244 .894 .136

5 GENQUO Pears. Cor .081 -.125 -.011 -.271 1

Sig. (2-tailed) .637 .467 .956 .110

6 LEV Pears. Cor -.105 .047 -.219 .212 -.013 1

Sig. (2-tailed) .457 .740 .185 .132 .938

7 FSIZE Pears. Cor .188 -.140 -.107 .364*** -.110 .419*** 1

Sig. (2-tailed) .183 .321 .522 .008 .524 .002

8 BSIZE Pears. Cor .035 -.510*** -.306* .246* .137 .417*** .622*** 1

Sig. (2-tailed) .807 .000 .062 .079 .427 .002 .000

9 IND_TECH Pears. Cor .080 .192 .152 -.196 -.016 -458*** -.375*** -.335** 1

Sig. (2-tailed) .571 .174 .364 .164 .926 .001 .006 .015

10 IND_TELE Pears. Cor .147 -.133 -.274* .069 .159 .040 -.028 .161 -.463*** 1

Sig. (2-tailed) .298 .346 .097 .627 .355 .777 .842 .255 .001

11 IND_CON Pears. Cor .022 -.059 .276 .168 -.157 .148 .162 -.006 -310** -.271* 1

Sig. (2-tailed) .880 .678 .093 .234 .359 .295 .252 .964 .025 .052

12 IND_UTI Pears. Cor -.276** -.023 -.097 .003 -.032 .358*** .326** .218 -.377*** -.330** -.221 1

Sig. (2-tailed) .048 .871 .561 .982 .852 .009 .018 .120 .006 .017 .116

*** Correlation is significant on 0.01 level (2-tailed), ** correlation is significant on 0.05 lever

(30)

5.4 Regression analysis

In table 7 the results of the performed multiple linear regression analysis are presented. The table consists of eight models, each representing the results of a different regression. The first model, model 0, shows the regression results from the control variables. Models 1 to 5 present the regression results testing all four hypotheses. Finally, in the last models, the regression analysis consisting of the variables together is shown. All results are explained in the last chapter, consisting of the summary and conclusion of this research.

The first model shows the regression results between the control variables and IFRS 15 disclosure quality. Leverage has a negative, but a non-significant relationship with IFRS 15 disclosure quality (β= -0.087, p>0.1). However, the results show a significant positive relationship between firm size and IFRS 15 disclosure quality (β= 0.488, p<0.05). The third controlling variable is board size and has a non-significant negative influence on IFRS 15 disclosure quality (β= -0.130, p>0.1). The industries telecom, construction and utilities are compared with technology. Technology is set as a reference because this industry has most observations and therefore correlates most with the other dummy variables. The comparison shows that there is a significant negative influence of utilities on IFRS 15 disclosure quality (β= -0.378, p<0.05), but for the other industries there is no significant influence. The adjusted R2 of model 0 is 0.087, which means that only 8.7% of the variation in IFRS 15 disclosure quality can be explained by the control variables. Despite not all control variables are significant and the quite low adjusted R2, no variables will be excluded from the subsequent regression analysis. Deleting one or more control variables would affect the designed model, which is based on the different theories and previous literature and therefore it is undesirable to delete one or more control variables.

5.4.1 Hypothesis 1 – Board independence

In the first hypothesis, a positive relationship between board independence and IFRS 15 disclosure quality is predicted. Board independence is measured as the percentage of total directors. And is tested in model 1, which shows a positive but non-significant relationship between the variables (β= 0.079, p>0.1). Therefore hypothesis 1 is rejected, which means that there is no significant relationship between board independence and IFRS 15 disclosure quality. With an adjusted R2 of 0.071 in model 1, 7.1% of the variation in IFRS 15 disclosure quality can be explained by the variation in the control variables and board independence. While the adjusted R2 in model 1 is lower than the adjusted R2 in model 0, board

IFRS 15 disclosure quality: The influence of corporate governance mechanisms