The effect of Integrated Reporting on audit fees : an empirical study focused on companies based in the European Union.

(1)

University of Amsterdam

The effect of Integrated Reporting on Audit fees

An empirical study focused on companies based in the European Union.

MSc Accountancy & Control, specialization Accountancy

Faculty of Economics and Business, University of Amsterdam

Name: Rachelle Sarianamual

Student number: 11393742

Supervisor: Dhr. Dr. A. Sikalidis

Date: 26 June 2017

(2)

Statement of Originality

This document is written by student Rachelle Sarianamual 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 of the work, not for the contents.

(3)

What is the effect of Integrated Reporting on Audit fees of European listed-companies?

By

Rachelle Sarianamual

THESIS

Submitted in partial fulfilment of the requirements

for the degree of Master of Science in Accountancy in the Graduate College of the

University of Amsterdam 2017

Haarlem, The Netherlands

Adviser:

(4)

Abstract:

The purpose of this paper is to examine the relationship between integrated reporting <IR> and Audit fees within European listed-companies. This relationship exists through the risk identification character of IR. The International Integrated Reporting Council (IIRC) suggest that integrated reporting helps firms to identify new types of risks and opportunity’s. Firms that use IR are able to develop an improved and comprehensive risk management and internal control system. Prior literature argues that internal audits and external audits are substitutes. Therefor it is assumable that IR can result in lower audit fees because of the increased internal audit which may result in less external audit effort. I utilize on the audit fee model of Goncharov et al. (2013). However, the results show a significant positive relationship between IR and audit fees. I found evidence that suggests that the moderating variable ‘total assets’ doesn’t have an influence on the just mentioned relationship, which indicate that the size of the client doesn’t explain the significant positive relationship of IR and audit fees. A clarifying explanation could be the assurance of IR which is emphasized often by researchers in this relatively new field of study.

(5)

Abbreviations

COSO

Committee of Sponsoring Organizations of the Treadway Commission

ERM

Enterprise Risk Management

FASB

Financial Accounting Standards Board

<IR>

Integrated Reporting

IASB

International Accounting Standards Board

IIRC

International Integrated Reporting Council

IFRS

International Financial Reporting Standards

NBA

Nederlandse Beroepsorganisatie van Accountants

NGO

Non-government Organization

ROA

Return on Assets

(6)

Introduction

1.1 Background

In a world of uncertainty, there is an ever-increasing need for information, transparency and accountability (Moolman & Oberholzer, 2016, p. 601). Information asymmetry and agency problems hinder the efficient allocation of resources in the capital market. Therefor disclosure of information to investors is needed for the functioning of an efficient capital market (Healy & Palepu, 2001; Dumay, Bernardi, Gurthrie & Demartini, 2015). The International Accounting Standards Board (IASB) emphasise the importance of the primary purpose of external reporting being the ability for inventors, lenders and other creditors to use financial information for estimating the value of the firm (IASB, 2010). Adams, Fries and Simnett (2011) discuss their growing concern of traditional external reporting, which don’t seem to meet the information needs among different stakeholders. The valuation purposes of the content of financial statements is a concern, meaning that only a small piece of the relevant information is used in the accounting process (Dopuch, 2017). When the crisis hit the capital market, the decision-makers realised that the financial information from external reporting was opulent. There was a shortage of faithful and relevant non-financial information. Non-non-financial information could have provided more details and context additional to the financial information (EY, 2017). For these reasons, over the past years companies began with the disclosure of non-financial reports in addition to financial information financial reports.

Sustainability reporting helps organizations to disclose the non-financial information by setting targets, measure performance and manage changes. In sustainability reports, positive and negative impacts of the operations on the environment, society and economy are analysed (GRI, 2013). According to Adams (2004) sustainability (non-financial information) reporting improves the completeness of external reporting. It reduces the gap between the expectation stakeholders have of the external reports and the actual disclosure of information by firms. But a dominant concern of adding sustainability disclosures as an afterthought to financial disclosure is the generation of more information instead of useful and more reliable information. This is called the tick-box approach; stakeholders getting an information overload (EY, 2017). 70% of 85 investment professionals mentioned in interviews with PwC that they need a clearer link between financial results and the business model, risk and strategy information in the annual report (IIRC, 2014b). Investors realized integration of financial and non-financial information is necessary to define a holistic picture of the risk and opportunities of the organization (EY, 2017).

In order to create a clear link between non-financial and financial information the IIRC introduced the integrated reporting framework. Integrated reporting (IR) integrates financial and non-financial information to one integrated report. De Villiers, Rinaldi and Unerman (2014) describe sustainability reports as an attempt to connect with the concept of sustainability while in reality nothing is changed. In

(9)

contrast integrated reporting is expected to make a real change in organisations. According to the IIRC (2013) the more integrated reporting is applied in the entity, the more smoothly relevant information will flow into the accounting process and will improve the information system. IR is becoming important since increasingly more entities see IR as a considerable replacement for their traditional financial reporting. However, Criticism of IR find contradictory results and suggest that companies won’t benefit from changing their traditional external reporting to integrated reporting (Maniora, 2015). Proponents emphasise that instead of Traditional financial reporting, IR enables a more comprehensive understanding of the firm’s risks (EY, 2017). This leads to a better and more comprehensive risk management system (Moolman et al., 2016).

If a firm is able to enhance their understanding and identification of inherent risks, the management can use this information to provide an improved working internal control system. As the quality of an internal control system increases, there will remain less additional work for the external auditor to provide

reasonable assurance of the financial statements (Simunic, 1980; Wallace, 1984). The theory about internal control and external control being substitutes has been confirmed by multiple studies (Felix & Gramling, 2001; Raghunandan & Rama, 2006; Hogan & Bedard, 2008; PCAOB, 2010). They find control risk is an important indicator for changes in audit fees. Control risk is the risk an auditor doesn’t detect material misstatements during the audit procedures. However, other researches claim internal control and external complement each other (Anderson & Zeghal, 1994; Gerrard, Houghton & Woodliff, 1994; Hay, Knechel & Ling, 2008). Anderson and Zeghal (1994) found that characteristics of firms, which are not about size or complexity ensures that both internal and external control is needed. In this research I expected a positive relationship between internal control and audit fees. Prior research finds evidence for both ways of correlations. However, since a practical view: if an organization has a accurate working internal control system in place, which results in a descent amount of inherent risks it is not assumable more external audit services are needed.

1.2 Research question and contribution

In the previous part the background of IR and the association with internal and external control is provided. Villiers et al (2014) proposes some areas for further research in the new audit space of IR. They recommend that the research area about the role of external auditors in IR needs more attention.

Examining the relationship between IR and audit fees could provide some information about this research area. By the best of my knowledge, no research has emphasised on the association between IR and audit fees. According to Hay et al. (2006) research in audit fees models could be interesting to examine issues of contracting and independence related to the audit process, such as the quality of IR. Therefor the purpose of this study is to empirically analyse whether IR reduces audit fees. This leads to the following research questions: “Does Integrated reporting lead to lower audit fees for EU-listed companies?” . This study

(10)

contributes to the existing literature by providing insight in the effect of IR on audit fees in EU-listed companies. Prior research is mainly focussing on the relationship between internal and external controls and are not looking how information is disclosed. The relationship between IR and audit fees could be useful for all parties in the context of cost-efficiency, which is the practical contribution.

To test the research question, the sample used in this research consists of European listed-firms, which integrate their financial and non-financial reports in the period 2011-2016. I will use an adjusted version of the audit fee model of Goncharov, Riedl and Selhorn (2014). The timeline and adjustments made to the model will be discussed later in this paper. Those mentioned before examine the effect of fair value reporting on audit fees for real estate firms in Europe, the UK and the US (Goncharov, Riedl, & Sellhorn, 2014). This study differs in three important ways. First this research will concentrate on the effect of IR on audit fees, where they focus on the effect fair value reporting on audit fees. Second, this research is in European context. According to Hay, Knechel and Wong (2006) most studies in the field of IR are done in the US and the UK and little research is done in the EU. They find only two researches performed in Belgium. Another reason for the European sample is about regulatory concerning external reporting, IR, internal control and risk management, which are for all European firms the same. Third, I will focus on all industries in contrast to Goncharov et al. who focus only on real estate firms and choose for a smaller sample size. Goncharov et al. choose for one industry to hold other factors which are industry specific constant. I will use a dummy variable for the different industries to get the same effect. The evidence shows a significant positive relationship between IR and audit fees. The conclusion is drawn from a regressive analyse and robustness test. I repeated the analyses with a moderating variable, which reveals that the size of the client doesn’t explain the significant positive relationship between IR and audit fees. The remainder of this paper is structured as follows. The next chapter provides the theoretical framework and literature review on the IR framework, risk management, internal control, external control and audit fees. I will continue with the formulation of my hypothesis. The following next chapter will discuss the results from the different analyses used to test the hypothesis. This paper will conclude with the findings and limitations and suggestions for further research.

(11)

2 Literature Review

As described in the introduction, IR is a response to the changing world economy. Traditional external reporting is changing towards voluntary integrated reporting, which integrate financial and non-financial and information to increase useful information, transparency and accountability (Dumay et al., 2015). This chapter provides an overview of the literature and theory about IR. First the IR Framework will be discussed, followed by the Guiding principles and Content elements. Then there will be attention for the assurance of IR. In the second paragraph theory about risk management will be explored. I will describe how risk management interacts with IR and internal control systems. The third paragraph will emphasize the importance of internal control systems and the chapter will end with information about external audit and audit fees.

2.1 The <IR> Framework

The International Integrated Reporting Council (IIRC) introduced in December 2013 the <IR>

Framework (IIRC, 2014). The IIRC consists out of regulators, standard setters, companies, investors, the accounting profession and NGO’s. The IIRC is heavily populated and funded by the big four auditors and other important global audit networks (Cheng et al., 2014). Their common purpose is to improve

communication about value creation in financial reporting. The IIRC defines IR as follows:

An integrated report is a concise communication about how an organization’s strategy, governance, performance and prospects, in the context of its external environment, lead to the creation of value over the short, medium and long term (IIRC, 2013, p.8).

2.1.1 the fundamental concepts

The capitals and the value creation process are the underlying fundamental concepts that form the basis for integrated reporting (IIRC, 2013, p. 5). According to the IIRC (2013, p. 6) an integrated report benefits all parties that are interested in the organization’s ability to create value over time, including employees, customers, suppliers, business partners, local communities, legislators, regulators and policy-makers. The < IR> framework fulfils this purpose and set a foundation for the future by creating a globally accepted integrated reporting framework for the short, medium and long term (Moolman et al., 2016). The emergence of IR provide a high quality of research papers about integrated reports; de Villers et al. (2014); Adams (2014); Cheng et al. (2014); Dumay et al. (2016), Deloitte (2015) and Reimsbach, Hahn & Gurturk (2017). They argue the <IR> framework helps entities by improving management and

(12)

reporting by tailor its business model and strategy to respond to the risks it faces, which ultimately results in better financial performance. The more IR is embedded into the organization, the better the

communication and information within the organization (IIRC, 2013).

Figure 2: The value creation process (IIRC, p. 13, 2013)

Figure 2. shows the six capitals of the <IR> framework; financial, manufactured, intellectual, human, social and relationship and natural capital. The capitals help organizations to make better decisions about the allocation of capital and they provide more insights in risks and opportunities (IIRC, 2014b). The IR framework is principle-based and establishes guiding principles and ensures elements that drive the content of integrated reporting. The six capitals go’s into the normal business model, presented in the middle of figure 2. Around the business model the content elements are presented and they interact with the business model. On the left, the six capitals come out the model; during this process there is value creation (below). Another clarifying image of the relationship of IR is the flow chart of Abeysekera (2013) shown in figure 3. The image shows how the four capitals are interacting with each other in order to achieve the organisations mission in a day-to-day mission. The organisation can use the different capitals to realise their goals (Abeysekera, 2013). The principle-based nature is to ensure that different firms, situations and circumstances can apply the IR and still ensure comparability across firms. In the next paragraphs the guiding principles and content elements will be clarified.

(13)

Figure 3: The Flow chart of integrated reporting (Abeysekera, 2013, p.233)

2.1.2 Guiding principles

The capacity and presentation of the external reports that integrate the financial and non-financial information are informed by the guiding principles, which are presented below (Cheng et al., 2014). The IIRC (2013, p. 6) summed the guiding principles in their <IR> framework as follows:

Strategic focus and future orientation - An integrated report should provide insight into the organization’s strategy, and how this relates to the organization’s ability to create value in the short, medium and long term, and to its use of and effects on the capitals

Connectivity of information - An integrated report should show a holistic picture of the combination,

interrelatedness and dependencies between the factors that affect the organization’s ability to create value over time

Stakeholder relationships - An integrated report should provide insight into the nature and quality of the organization’s relationships with its key stakeholders, including how and to what extent the organization understands, takes into account and responds to their legitimate needs and interests

Materiality - An integrated report should disclose information about matters that substantively affect the organization’s ability to create value over the short, medium and long term

(14)

Conciseness - An integrated report should be concise

Reliability and completeness - An integrated report should include all material matters, both positive and negative, in a balanced way and without material error

Consistency and comparability - The information in an integrated report should be presented: (a) on a basis that is consistent over time; and (b) in a way that enables comparison with other organizations to the extent it is material to the organization’s own ability to create value over time (IIRC, p. 6, 2013).

2.1.3 Content elements

To ensure every firm from their own unique value creation process the content elements below are important during the preparation of the integrated report (Cheng et al., 2014). Again, the IIRC (2013, p. 6) made a summary of the content elements in their <IR> framework as follows:

Organizational overview and external environment - What does the organization do and what are the circumstances under which it operates?

Governance - How does the organization’s governance structure support its ability to create value in the short, medium and long term?

Business model - What is the organization’s business model?

Risks and opportunities - What are the specific risks and opportunities that affect the organization’s ability to create value over the short, medium and long term, and how is the organization dealing with them? Strategy and resource allocation - Where does the organization want to go and how does it intend to get there? Performance - To what extent has the organization achieved its strategic objectives for the period and what are its outcomes in terms of effects on the capitals?

Outlook - What challenges and uncertainties is the organization likely to encounter in pursuing its strategy, and what are the potential implications for its business model and future performance?

Basis of presentation - How does the organization determine what matters to include in the integrated report and how are such matters quantified or evaluated (IIRC, p. 6, 2013)?

(15)

2.1.4 Assurance of integrated reporting

Prior research emphasise the importance of assurance of integrated reports (Reimsbach, Hahn & Gurturk, 2016). Assurance of integrated reports is necessary to enhance the credibility of the information disclosed. Users of the integrated report need this assurance to relay on the information and therefor it improves decision-making (IIRC, 2014b). It concerns the development of the accounting core and hence the relevance of the accounting profession in the future (NBA, 2013). During the assurance of integrated report the following issues should be included to give a reliable picture of the company:

Materiality – Using a materiality level or threshold to guide judgements in planning and performing an assurance engagement involves such issues as defining a material error or omission, applying qualitative considerations and assessing aggregated misstatements.

Reporting Boundary – If information from outside the entity is included in the financial reporting, determining what constitutes sufficient appropriate evidence with respect to such information and designing procedures to obtain that evidence may present challenges.

Connectivity – Issues include the nature and extent of procedures to determine whether an integrated report demonstrates sufficient connectivity, and the evidence required to support an organization’s assertions about the course of certain connections.

Completeness – Assessing the completeness of an integrated report is likely to be a significant concern to assurance practitioners, including, the interplay of completeness with:

(a) the concept of conciseness

(b) exclusions allowed for by the framework regarding the unavailability of reliable information, specific legal prohibitions and information that would cause significant competitive harm.

Narrative reporting and future-oriented information – Disclosures in an integrated report might include soft narrative or future-oriented information that require the assurance practitioner to exercise a high degree of professional judgement and scepticism; in some cases it may be difficult to obtain sufficient appropriate evidence to support such disclosures (IIRC, 2014, p.8).

The above described issues are provided by the IIRC and are an citation of the <IR> framework (IIRC, 2014, p.8).

(16)

2.2 Risk Management

This section will start with explaining the process of risk management. The second paragraph describes the relationship between risk management and internal reporting. The relationship mentioned is the first component of the conceptual model, which is explained later in this research. Finally, the link between risk management and internal control is provided.

2.2.1 Risk Management process

The financial crisis and scandals of the last decade ensured firms to broaden their risk management. Organizations now identify risk in social and economies situations and look further than only financial risks (Power, 2009). Most countries respond to these matters by setting specific requirements for

corporate governance, internal control and risk management. In addition, several organizations introduced frameworks for business risk; the Enterprise Risk Management (ERM) Framework of COSO in 2004, the association of Insurance and Risk Managers published a risk management standard in 2002 (Kajüter, Woods, & Linsley, 2007, p.2). The regulation concerning to risk management varies among countries, therefor the type and degree of implementation of risk management in countries is also different (Kajüter et al., 2007). Risk management is identifying and managing the risks that are concerned with achieving the organizations objectives. When the risks are identified, adjustments and adequate control systems are necessary to manage the risk (Power, 2009). Instead of shareholders, managers are more able to identify and manage risk in the organization because of the agency and asymmetry information problems. (Kajüter et al., 2007).

The crisis and scandals didn’t only change the scope of risk management; they also enabled a change in the different kinds of risks and their weighting. Risk experts become more accountable for failure in risk management, therefor secondary risks are becoming increasingly important. These are the risks that risks experts fail in their job of identification and managing risks (Power, 2004). During the risk management process real risks or inherent risks are identified and managed and after that they are captured by the internal control systems.

2.2.2 Risk Management and integrated reporting

Since risk management has moved to a broader scope, it is interesting to research the role of IR during this process (Kotze, Vermaak, Kirsten, 2015). As mentioned before IIRC defined the relevance of an integrated report as follows: IR identifies the key risks and opportunities that are unique to the organization, including those that relate to the organization’s effects on, and the continued availability, quality and affordability of, relevant capitals in the short, medium and long term (IIRC, 2013, p.6). A firm, that has implemented IR and integrated their risk management system, is able to identify specific sources

(17)

of risks (internal, external, or both) and circumstances that would cause certain risks. The firm can consequently develop steps to mitigate and control key risks and enhance their risk management process (ACCA, 2012; IIRC, 2013; Moolman et al., 2016). This has been confirmed by multiple studies; the research by EY also found that 45% of the respondents also uses environmental, social and governance information by determining risks (EY, 2015). Moolman et al. (2016) discusses that 82% of the

respondents of their research found that IR improved the integration of risks into overall strategic objectives. According to EY (2015) companies who disclose non-financial information to demonstrate management of risk has grown from 29% in 2014 to 42% in 2015. Another example from practice is 63% of the investors take integrated reports into account by making decisions to manage risks (CFA institute, 2015). Moolman et al. (2016) sums the information required about risks disclosed in IR:

• The specific risks and opportunities that affect the organisation’s ability to create value • Risks and opportunities that affect the availability, quality and affordability of relevant capitals • Specific external sources of risk

• Specific internal sources of risk

• The organisation’s assessment of the likelihood that the risk or opportunity will come to fruition and the magnitude of its effect if it does

• Steps being taken to mitigate or manage key risks, and

• The organisation’s approach to any real risks that are fundamental to the ongoing ability of the organisation to create value and that could have extreme consequences (Moolman et al., 2016, p. 604). The above describes disclosures of risks could also be helpful during the risk identification procedure of the external audit because external auditors are more informed about the risks the firm faces.

2.2.3 Risk Management and Internal control

The internal control systems are adept to the risks identified during the risk management process. According to Power (2004), risk management has become incorporated within a lager accountability and less sophisticated internal control framework. The quality of operations of firm’s increases when they have a broad and formal risk management system (Power, 2004). Internal control becomes more risk-based and therefor risk management now focusses more on the whole organization.

2.3 Internal control

In this component, I will provide theory about the integrated control framework. Followed by a short explanation of the link between internal control and IR. Finally, the relationship between internal and

(18)

external control will be discussed. This relationship is the last section of the conceptual model of chapter four.

2.3.1 Internal control framework

Similar at risk management, internal control received a lot of attention as result of the financial crisis and scandals. For this reason, a lot of research is conducted over the past years in the concept of internal control (Vaassen & Meewissen, 2005). The importance of internal control has become clearer during these occasions. Managers benefit from internal controls because they secure the quality of their operations. Next to this, auditors benefit from internal controls because it gives security of the reliability of information (Vaassen & Meewissen, 2005, p.29). An influential framework for internal control is the COSO framework.

COSO (1992) defined internal control as follows:

Internal control is a process, affected by an entity’s board of directors, management and other personnel, designed to provide reasonable assurance regarding the achievement of objectives in the following categories:

• Efficiency and effectiveness of operations; • Reliability of financial reporting;

• Compliance with applicable laws and regulations.

Important elements of an internal control system are: the audit committee, codes of conduct and the internal audit. Creating an internal control system can be based on the typology of an organization or the transaction process of an organization. Every organization consist out of numbers of transactions and processes, so the outcome of those two processes should not differ much (Vaassen & Meewissen, 2005).

As mentioned before, Power (2004) suggests internal control became more risk-based and changed into risk management. He identifies three reasons for this trend: first, the increasing attention for secondary risk, as mentioned in paragraph 2.2. Second, insurance; organizations try to control secondary risks by transferring them to third parties (external auditor). Third, organizations try to prevent themselves against scandals and crises by creating new risk structures to remain public and private trust (Power, 2004).

Internal control has the interest of external auditors because it tells them something about the risks already captured by the internal control system. They identify whether a enhanced and reliable internal control system is in place and then analyse the additional work they have to perform during the external audit.

(19)

2.3.2 Internal control and integrated reporting

The introduction mentioned how IR enables firms to make better decisions based on financial and non-financial information and improves internal processes like internal control systems (Stubbs and Higgings, 2014). Both IR and internal control have a risk-based nature. IR can improve internal control by better risk assessment where risk can be identified in the six capitals of IR and therefore will be more

comprehensive. The internal controls will be adjusted to the identified risks, which lead to a better internal control system.

2.3.2 Internal control and External control

As mentioned before, a lot of research is conducted to examine the relationship between audit fees and internal control risk (Hogan & Wilkings, 2008; Hoitash & Bedard, 2008; Raghunandan & Rama, 2006). These researches find that there is a negative relationship between internal control risk and audit fees. Internal control risk is the risk that the internal control system don’t get grip on material misstatements. They find that if good internal controls aren’t in place, the audit fees are higher. Wallace (1984) & Simunic (1980) found that internal and external audit are substitutes. The consistent findings suggest that auditors have to perform more effort to manage control risks in these circumstances which will result in higher audit fees (Jiang et al. 2015).

2.4. External control

These section starts with provide information about external controls. The auditor received audit fees for the audit effort they provide by clients. I will discuss the audit fees and their key drivers: Audit effort, complexity, firm risks, extreme negative performance and the auditor.

2.4.1 External control framework

External audit are meant to provide reasonable assurance to the users of the external reports. Reasonable assurance means that the external reports are relevant and faithful represent the situation of the

organisation, which indicate that they are complete, neutral and free from errors (Abeysekera, 2013). According to IAASB (2011) faithful representation means that the information is clear and accurately described, neutral indicates that the disclosed information is free from manipulation and free from errors means that the information is free from misstatements or omissions. The errors are material if they are inherent risks or control risks (Figure 3.) During the audit the auditor first examines which risks are filtered out by the internal control system and consequently analyse which additional audit procedures are needed to capture this remaining control risks. The misstatements that auditor’s don’t detect during the external audit are the detection risks. As mentioned earlier, during the crisis and several scandals the auditor is held responsible for the possibility of material misstatements in the external reporting to

(20)

stakeholders. In the situation of IR the internal control covers more inherent risks in contrast to the traditional way of reporting. The auditor therefor has less control risks to catch during the external audit, which means less audit effort.

Figure 3. Audit Risk model

Inherent risks Control risks Detection risks

2.4.2 Audit fees

Hay, Knechel and Wong (2006) evaluate and summarize the research conducted about the determinants of audit fees. They found that most researches has followed the original seminal work by Simunic (1980) which investigated the client and auditor characteristics that are associated with an increase or a decrease in audit fees.

Simunic (1980) defined the term audit fee as the product of unit price and the quantity services demanded by the management of the auditee. He found evidence that some variables cause more or less audit effort and result in different amounts of audit fees (Simunic, 1980). The amount of effort needed for an auditor in order to perform the audit so that it can give reasonable assurance that the financial statements are free from errors and give a reliable and relevant picture of the organization is largely dependent of the

characteristics of the client (Caramanis & Lennox, 2008). Characteristics of the client that should be represented in the audit fee model are: size, complexity, firm risks, extreme negative performance and industry. Also characteristics of the auditor should be taken into account, like reputations effects and business season (Goncharov et al., 2014).O'Keefe, Simunic & Stein (1994) reveals that size, complexity and risk pronounce eighty percent of the effort spend by auditors during the external control. According to Simunic (1980) auditors charge a premium for clients who have higher risks, this is called risk premium.

Internal control

External audit

<IR> Internal control

(21)

Larger client risks result in business risks which can be stated as the probability an auditor suffers a loss (for example: litigation and impaired reputation) because of the client risks (Niemi, 2002). The risk premium is to compensate for more effort during the audit and the business risk. However, research brings evidence based on audit fees and hours that the client risks are only reflected in effort and not in risks premiums (Niemi, 2002). Therefor in this research risk premiums are ignored.

Audit effort

According Simunic (1980) size is the most dominant characteristic of audit fees. For bigger organizations size pronounces for seventy precent of audit fees, this doesn’t hold for smaller organizations (Bell, Knechel, Willingham, 1994). Hay et al. (2006) evaluate the existing research about audit fees. They show that eighty-seven studies are using assets as control variable. Another slightly less indicator used is revenue. I will use the logarithm of total assets as indicator for size.

Complexity

Organisations that are more complex and diverse automatically face more and more varied risks from their environment (Hay, Knechel, & Ling, 2008). The more difficult the practices of the client are, the more difficult the audit procedures will be. Complexity is measured in many different ways by researches (Hay te al. 2006, p.169). I will measure complexity with the variables percentage of foreign assets outside the EU and the number of operating segments. Firms who operate outside the EU have to deal with different cultures, valuta and regulations. Also the operating segments say something about complexity; the more segments a client operates in, the more processes and risks the client has.

Firm Risks

Certain assets are riskier to the external audit then others. These risks can lead to potential financial distress, failure to detect client errors by the external audit and the quality of client’s internal control system(Hay et al., 2008). When the auditor notice that these risks are high, they increase their control practices and the audit fees to reduce the change of litigation or damaged reputation of the audit firm (Chan, Ezzamel, William, 1993). Firm risks are a very broad concept, therefor I will use four variables as indicator for firm’s risks. The variables are: return on assets (ROA), loss, percentage of receivables, percentage of inventory and the leverage ratio. These variables will be discussed later in this paper.

Extreme negative performance

Simunic (1980) discussed how some firms are more risky than others, Firms firms that are in financial distress, in the case of bankruptcy or with an unprofitable performance are examples of risky firms. These firms are more risky as client for auditors, because they have more change on fraud, which may damage

(22)

negative book value of equity, qualified audit opinion and standard deviation of monthly stock returns are included.

Auditor

The above described components of the audit fee model are all characteristics of the client. Also auditor characteristics should be taken into account. An important and dominant characteristic of the auditor is the reputation effects. Reputation effect means the advantages big four (EY, PWC, KPMG, Deloitte) auditors has in audit pricing with regard to non-big four auditors(Goncharov et al., 2014).

How the above-discussed variables are measured will be further explained later in this research.

(23)

3 Hypothesis Development

As mentioned before, an auditor’s client risk level is an indicator for audit fees. Earlier researches suggest that business risk leads to higher audit effort and higher audit fees. This significant relationship between internal audit and external audit is called the substitution view. However, other researchers found the opposite relationship; that an increase in internal audit will lead to an increase in external audit. On basis of the arguments of previous research and taking into account the practice context, wherein auditors adjust their audit procedures to the internal control system, a negative relationship between the internal control and audit fees is predicted in this study. When an auditor’s client has implemented IR they should be able to better identify and control their risk. If the client has a good internal control system in place, the auditor has less additional work to perform which lowers the audit effort and audit fee. Therefore I predict that IR can result in lower audit fees. If the hypothesis holds, IR lead to less audit effort, which lead to lower audit fees which will consequently lead to an negative relationship between the two. If the relationship between IR and audit fees is positive the hypothesis doesn’t hold and other explanations must be found for this result.

Hypothesis 1. Integrated reporting leads to lower audit fees because the audit effort is lower.

Prior research shows firm size has a significant positive effect on audit fees (Hay et al., 2006; Simunic, 1980). To exclude the effect of the firm size of the client on the effect of integrated reporting on audit fees, the following hypothesis is provided:

Hypothesis 2. The negative relationship between integrated reporting and audit fees is less pronounced for large firms.

(24)

4. Research Design

This chapter will describe the methodology of this research used to examine if IR has an effect on audit fees. First the model used to test the hypothesis will be explained. The model will contain nineteen control variables of which seven represent the industry dummies. This chapter will end with attention for the data and sample collection.

4.1 The Model

I will start with examining whether audit fees change when firms use more integrated reporting for external reporting practises. The following conceptual model is used to clarify the relationship between the independent variable <IR> and the dependent variable audit fees (Figure 1.).

Figure 1. Conceptual model

+

The above conceptual model shows through which processes the relationship exists between IR and audit fees. IR increase risk management and internal control, (IIRC, 2013; Power, 2004). Internal control has an impact on the external audit effort (Wallace, 1984). Audit fees are a product of the price and quantity audit effort spend (Simunic, 1980). I will compare changes in audit fees over time from the period 2011 to 2016 with the percentage of integrated reporting used by firms in the same period. In December 2013 the IIRC introduced the IR framework as guideline for firms who are interested in disclosure of integrated reports (IIRC, 2013). As mentioned before, I will use the audit fee model of Goncharov et al. (2013). I will change the independent variable fair value reporting into IR and I added the control variable Inventoryit. This variable explains the inventory divided by total assets of firm i for year t. According to Jiang et al. (2015) this is an important variable to capture inherent risks. An important assumption of the audit fee model is that both the auditor and the client are risk neutral and seek to maximize their own expected profits each period (Simunic, 1980, p. 163). I will implement the regression analyses using the following model:

LogFees

ir

Σ

Risk managem ent Internal Control

<IR>

External

(25)

Where the vector CONTROL captures two firm-level controls: the characteristics of the audit client (LogAssetstotalit, Foreignit, NSegmit, ROAit, Lossit, Receivablesit, Inventoryit, Levit, Distressit, Qualifiedit,

Volatilityit & Industryit) and those of the audit firm (Yearendit) (Goncharov et al., 2014). An overview of all

variables is given in Appendix 1.

4.2 The variables

In this paragraph the dependent, independent, control and moderating variables will be discussed. The expected signs for the coefficients are presented in table 1.

Dependent Variable

Goncharov et al. (2014) expressed the variable for audit fees on basis of research by Hay et al. (2006). Consistent with those researches, I will express the dependent variable in log form to mitigate the effects of nonlinear relations (Goncharov et al., 2014). The logarithm of audit fees for firm i in year t (LogFeesit) is the dependent variable.

Independent Variable

The independent variable (irit) expresses the rate of how much external reports are integrated for firm i in year t. It reflects a balance view of a company’s quality of disclosure of integrating performance in the areas: economic, environmental, social and corporate governance, which according to Sarafeim (2015) is an appropriate indicator for the quality of IR. The predicted sign is negative.

Control Variables

As mentioned earlier in this paper, the control variables present characteristics of both the auditor and the client. I will start with describing the characteristics of the client. The size will be measured with the variable ‘total assets’. According to Simunic (1980), size is the most dominant determinant of audit fees. The size measure is usually transformed by taking the natural logarithm of the raw data in order to improve the linear relationship with audit fees (Hay et al., 2006, p. 169). Therefor I measure size as the logarithm of total assets (LogAssetstotalit). An increase of firm size is expected to increase the audit fees.

To capture audit complexity this research includes the percentage of international assets

(excluding EU) and the number of operating segments (Goncharov et al., 2013). If a firm operates outside Europe, they have to deal with another economic and political environment and foreign currencies. According to Goncharov et al. (2013) audit effort is higher for firms with more international complexity. Foreign assets are calculated by the percentage of internal assets (excludes European countries) of total

(26)

assets of firm i for year t (Forgeinit). DataStream represents internal assets as the total or identifiable assets

of foreign operations before adjustments and eliminations. Another criterion for complexity is the number of operating segments because it ensures complex operations which increases audit effort. This variable shows the number of segments organisation i operates in, in year t (Nsegit). The expected sign for both

variables is positive.

Goncharov et al. (2013) include four control variables as indicator for firm risk. They include two income statement-based measures of firm performance: Return on assets and loss. Return on assets represents the continuous variable and loss is an indicator for distress (Goncharov et al., 2013. p. 218). ROA is measured by net income divided by total assets for firm i in year t (ROAit); a negative coefficient

is predicted here. Loss is represented by 1 if the firm makes a loss i in year t and 0 otherwise (Lossit);

predicted positive sign. They also include two balance sheet-based constructs, percentage of receivables and the leverage ratio. The percentage of receivables is included because the opportunity for errors is higher in the presence of receivables; predicted sign is positive. The receivable variable is the outcome of receivables divided by total assets of firm i in year t (Receivableit). The leverage ratio is included because

leveraged firms experience greater financial constraints (Goncharov et al., 2013. p. 218). The leverage ratio shows how much capital comes in the form of debt and is calculated by total debt divided by total equity. The distress variable shows the leverage of firm i in year t (Levit). The predicted relationship of the distress

variable and audit fees is expected to be positive. Goncharov et al. (2013) don’t have to deal with inventory risks because they focussed only on the real estate market. I will focus on all the industries, which includes firms that holds inventory. Prior research frequently suggest that the most difficult accounts to audit are inventory and receivables (Hay et al., 2006; Newton and Ashton, 1989; Simunic, 1980). Firms with physical inventory face greater risks of errors and fraud, therefor I include the variable percentage of inventory. The variable for inventory shows the percentage of inventory divided by total assets for firm i in year t (Inventoryit); expected sign is positive.

The two following variables are indicators for extreme negative performance. Both variables are predicted to have a positive relationship with audit fees. Firms with negative equity or qualified audit opinions are more likely in distress; therefor the indicators for extreme negative performance are negative book value of equity and the incidence of a qualified audit opinion (Goncharov et al., 2013. p. 218; Hay et al., 2006). The variable (Distressit), represents the negative book value of equity for firm i in year t, if there

is a negative value it will present 1 and 0 otherwise. The qualified audit opinion variable will present 1 if qualified for firm i in year t and 0 otherwise (Qualifiedit).

The last control variable captured in the model of Goncharov et al. that represents the

characteristics of the client is the standard deviation of monthly stock returns. This variable is a market-based measure of risks: as more volatile stock returns reflect riskier firms (Goncharov et al., 2013). The variable shows the standard deviation of monthly stock returns for firm i in year t (Volatilityit). The

predicted sign is positive. Consistent with Goncharov et al. (2013) the variable market-to-book ratio and negative stock returns, which represent fundamental performance, aren’t included in this research.

(27)

Goncharov et al. (2013) suggest that the results are unchanged by including the two additional control variables.

Instead of Goncharov et al. (2016) this study will focus on the entire industry. Therefor I included the dummy industry to capture industry specific risks. Prior research commonly made the suggestion that some industries are more difficult to audit than others; industries with inventory, receivables and

knowledge-based assets for example (Hay et al., 2006; Simunic, 1980; Turpen, 1990). The industry variable consist out of: Mining and Construction (MiningandConstructionit), Manufacturing (Manufacturingit),

Transportation & Public utilities (TransportationPublicutilitiesit), Wholesale trade and retail trade

(Wholesaletraderetailtradeit), Finance, insurance, real estate (Financeinsurancerealestateit), Services

(Servicesit) and Public administration (Publicadministrationit). It shows 1 if it contains the right industry

for firm i in year t and 0 otherwise. Consistent with prior research; transportation & public utilities, finance, insurance, real estate are expected to have a negative relationship with audit fees. Manufacturing, wholesale trade and retail trade and services are expected to have a positive sign. The others have no specific expectations, since they don’t have remarkable characteristics associated with audit fees.

Goncharov et al. (2013, p. 218) includes two variables, which represents the characteristics of the auditor; Big four and Yearend. Prior research finds that Big four has a significant positive relationship with audit fees (Goncharov et al., 2013; Hay et al., 2006; Simunic, 1980). However, data about the auditor for the period of 2011-2016 for European firms is not available in the databases that are in my reach. Therefore, BigN will not be included in the model of this research. In this model size is included, indirect this says something about Big N auditors However, the significant results are not up to date and date back from the 1990s. Recent research don’t find a significant relationship between BigN and audit fees (Hay et al., 2006). Yearend captures the audit fees charged when audits occurred during periods of constrained auditor resources (Goncharov et al., 2013, p.218). The variable (Yearendit) will be 1 if the

audit takes place in March till June for firm i in year t and otherwise 0. The expected sign is positive. Another proxy that consistent with Goncharov et al. (2013) which isn’t included in this research is non-audit services. There are contradictory results revealed about this variable. Prior research emphasises on the importance of non-audit services in the determining of audit fees (Simunic, 1980; Turpen, 1990). However, other results suggest that because of cross-subsidization of fees between audit and non-audit services can be associate with lower audit fees (Hay et al., 2006). This research focuses on the relationship of the quality of disclosure IR on audit fees, which is associated with audit services.

Moderating variable

The moderating variable change the effect that the independent variable (irit) has on the dependent

variable (Logauditfeesit) (see figure 4). I Added the moderating variable (irLogAssetstotalit) to test the

influence of total assets on the relationship between irit and Logauditfeesit for firm i in year t. I expect the

(28)

Figure 4: Moderating variable

4.3 Data construction

The sample used to give an answer to the research question and hypothesis will be one of European companies. These companies are based in the same economic and political environment, which means they all experience the same influences like culture or regulatory requirements. They face the same factors that have an influence on the relationship of IR and audit fees. The European companies are mandatory to report under the same accounting standard: the International Financial Reporting Standards (IFRS). Further, Hay et al. (2006) examine earlier research in the field of audit fees and found that most research is conducted in the US and the UK. Therefor it is interesting to see how audit fees are accomplished in Europe. This research will focus on all industries which will give a comprehensive view of the situation. Notice, in which industry a firm operates could influence both IR and audit fees. I will use a dummy variable to indicate different industries. The industries are chosen on behave of Standard Industrial Classification (SIC) to identify the major industry group. This research focuses on the time period of 2011-2016. In the middle of this timeline, the IIRC introduced the IR framework for guidance for companies who want to integrate their financial and non-financial reports (IIRC, 2013).

This paper combines the databases Thompson Reuters ASSET4, to find the integrated rating and audit fees for each firm in year 2011-2016 and COMPUSTAT, which provides data about financial, statistics and market information, therefore all other variables are found with the use of this database. Since 2000 the Securities and Exchange Commission (SEC) adopted new requirements about audit fees. Since then disclosure of information about audit fees are mandatory.

ir

it

Logauditfees

it

irLogAssetstotal

it

(29)

5. Empirical results

First I will provide a summary of the data sample with the use of descriptive statistics. Followed by the results of the Pearson correlation matrix. Finally, I will discuss the outcomes of the OLS regression analyse among with the results of the Robustness test.

5.1 Descriptive statistics

The final sample present 1485 observations of 247 unique European firms for the period 2011-2016. Table 1 presents the mean and medians of this sample, excluded are the variables with values of 0 or 1. The mean for audit fees for European companies is 14.082. This is not consistent with the research of Goncharov et al. (2013) since they find a mean of 4.915. The explanation here for is the difference in industries. The mean for; foreign, ROA and inventory is positive, since they present million euros, which indicate they all have a mean below the million euros. The independent variable for disclosure quality of IR has a mean of 67.131% and a remarkable standard deviation of 27.358%. The standard variation indicates a significant variation in the practice of IR. Table 1 further reveals an average of 866.6 of total assets in millions, foreign (excluding EU) assets is 2.3% of total assets, 2.968 number of segments, 0.052 return on assets, receivables are 11.1% of assets, inventory is 9.5% of assets, a leverage ratio of 369.4% and 7.507 standard deviation of monthly stock returns.

Table 2. Descriptive statistics

Variable Observations Mean Std. Dev. Min Max

LogAuditfees 1532 14.08219 1.59894 9.5518 18.95131 ir 1536 67.13062 27.35811 2.93 96.64 LogAssetstotal 1485 8.66571 2.033166 4.359397 15.06403 Foreign 1536 .0226965 .2906145 0 7.141342 Nsegm 1518 2.968379 2.312227 0 10 ROA 1536 .052108 .0774883 -.2739513 1.048831 Receivable 1536 .1109784 .1126181 0 .6841805 Inventory 1536 .0954815 .1422578 0 .9365994 Leverage 1536 3.693641 14.29132 -95.75573 331.2197 Volatility 1500 7.507327 26.92824 0 467.51

This table reports descriptive statics. All variables are defined in Appendix 1.

(30)

5.2 Pearson correlation matrix

Table 2 present the Pearson correlations for all proxies used in the model except the variables with value of 1 or 0. The significance is divided into three levels; *p<0.05, **p<0.01 and ***p<0.001, the lower the value, the more significance a correlation is. There is correlation between both variables if a certain linear relation exists, which are reflected in coefficients. In the situation two or more variables in the regression model are highly correlated (Coefficient > 0,8) there is a case of multi-collinearity (Gujarati, 1988).

Table 3. Pearson Correlation Analyses

Logaudit fees

ir LogAsset

stotal

Forgein Nsegm ROA Receivabl e

Inventory Lev Volati lity Logaudit Fees 1.0000 Ir 0.4233*** 0.0000 1.0000 LogAssets Total 0.7275*** 0.0000 0.2906*** 0.0000 1.0000 Forgein 0.0967*** 0.0002 0.0413 0.1060 0.1573*** 0.0000 1.0000 Nsegm 0.3365*** 0.0000 0.1478***0 0000 0.3376*** 0.0000 0.1215*** 0.0000 1.0000 ROA -0.1829*** 0.0000 -0.0142 0.5781 -0.2472*** 0.0000 -0.0440 0.0844 0.0845*** 0.0010 1.0000 Receivable 0.0743** 0.0036 0.1561*** 0.0000 -0.2446*** 0.0000 0.0532* 0.0372 0.0696** 0.0067 0.0571* 0.0252 1.0000 Inventory -0.1221*** 0.0000 0.1242*** 0.0000 -0.2180*** 0.0000 -0.0465 0.0687 -0.0065 0.7989 0.0917*** 0.0003 0.1710*** 0.0000 1.0000 Lev 0.1086*** 0.0000 0.0179 0.4828 0.2753*** 0.0000 0.1413*** 0.0000 0.1203*** 0.0000 0.1139*** 0.0000 -0.1125*** 0.0000 -0.0868*** 0.0007 1.0000 Volatility 0.0284 0.2725 -0.0410 0.1121 0.0018 0.9454 -0.0201 0.4373 0.0322 0.2156 0.0143 0.5790 -0.0046 0.8592 -0.0045 0.8624 -0.0326 0.2077 1.0000 This table reports descriptive statics. All variables are defined in Appendix 1. Significance level *p<0.05, **p<0,01, ***p<0,001

The independent variable quality of IR has a significant positive relationship with the dependent variable audit fees. If the quality of IR increases by 1, audit fees increases by 0.4233. The results are not consistent with the expectations of this research. The size of a firm, the percentage foreign assets, number of segments, receivables, leverage and volatility has a positive relationship with audit fees, which was expected and consistent with the research of Goncharov et al. (2013); Simunic (1980) and Hay et al. (2008). However the correlation of the variable volatility is not significant. ROA and inventory shows a significant negative correlation with audit fees, this is consistent with the expectations. Remarkable is the

(31)

significant correlation of size on both audit fees and the quality of IR. To exclude the effect that size has on the effect of the quality of IR on audit fees a moderating variable is included in the model. This will be further discussed later in this chapter. Overall, the correlations of the control variables are consistent with the predictions. However, the positive relationship between the quality of IR and audit fees was against the predictions of this study, but can be explained by other characteristics of IR. It can be concluded that the results are in line with the OLS assumptions and that there are problems expected with regard to multi-collinearity.

5.3 Regression analyses

Earlier research found that inferences are unchanged using all available observations or winsorizing at the 1st_{and 99}th_{percentiles of LogAuditfees (Goncharov et al. 2013, p. 219; Simunic, 1980). Therefor I will use}

all variables during the empirical analysis. Table 4 presents the empirical results based on the OLS regression analyses. For all regressions, the statistical inferences are based on the heteroskedasticity-consistent variance-covariance matrix (Jiang et al., 2014). The model used in this research explains 68,11% of the variation of audit fees (Adj. R-squered = 0,6811), which indicate that the model has a high

explanatory power. The paper examines the relationship between the disclosure quality of IR and audit fees and predict that the relationship is negative. The results suggest that audit fees are increasing by the independent variable quality of the IR (0.007***, t stat = 6.91), which indicates a significant positive relationship. However the results show a significant (sig. 0.000) positive relationship which means the hypothesis doesn’t hold.

Among the control variables I find that audit fees are increasing in total assets (0.609***, t stat = 37.41), which is in line with prior research (Goncharov et al.2013; Simunic, 1980, Hay et al. 2008; Che-Ahmad et al. 1996). Number of segments (0.039**, t stat = 3.38), return on assets (0.211, t stat = 0.56), loss (0.105, t stat=1.18), receivables (2.330***, t stat = 8.63), distress (1.954***, t stat = 7.26), qualified audit opinion (0.698, t stat = 0.77) and volatility (0.001, t stat = -2.02). Only the public administration industry has a significant positive coefficients (1.339*, t stat = 1.97). The following control variables show audit fees are decreasing in percentage of foreign assets (-0.008, t stat = -0.09), inventory (-0.391, t stat = - 1.79), leverage (0.006**, t stat = -3.42) and Yearend (-0.100*, t stat = -2.02).

The coefficients of the control variables show the predicted signs, except for: percentage of foreign assets, return on assets, percentage inventory of total assets, leverage and yearend. The negative relationship between foreign assets may be associated with the exclusion of European companies from the variable; the international assets are all outside the EU. Companies who have assets in 10 European countries and zero outside the EU should be more complex then companies with zero international assets in the EU and 1 outside the EU. The effect of ROA can be nonlinear on audit fees, because a decrease in ROA has a

(32)

different effect on audit fees when the company is already losing money instead of making profit (Hay et al., 2006).

Table 4. OLS regression analyse

Variable Expected sign Coefficient p value

ir - 0.0071 0.000 LogAssetstotal + 0.6094 0.000 Forgein + -0.0075 0.926 Nsegm + 0.0389 0.001 ROA - 0.2111 0.579 Loss + 0.1049 0.239 Receivable + 2.3297 0.000 Inventory + -0.3910 0.074 Lev + -0.0060 0.001 Distres + 1.9538 0.000 Qualified + 0.6987 0.444 Volatility + 0.0005 0.519 Yearend + -0.0995 0.043 MiningandConstruction + 0.4433 0.498 Manufacturing + 0.8943 0.165 TransportationPublicutilities + 0.2917 0.653 Wholesaletraderetailtrade + 0.4024 0.535 Financeinsurancerealestate + 0.0946 0.884 Services + 0.8505 0.192 Publicadministration + 1.3392 0.049 F Value 153.51 Adjusted R2 _0.6811 n 1429

A better proxy for inventory and receivables should be the combination of inventory and receivables, which according to Hay et al. (2006) is an important driver for audit fees. The negative relationship of inventory is remarkable since inventory increase audit effort. However, the relationship is not statistically significance. The association of leverage and yearend is significant, which mean it materially affect the audit fees. Leverage and yearend both are showing a negative coefficient while it was expected to be positive. Prior research suggests that the leverage ratio is more important in the US and the UK and less important in other countries, therefor no further attention is paid to the negative coefficient (Hay et al., 2006). The negative relationship of yearend can be explained by the period chosen for business season. I choose for the period March till June in contrast with the research of Hay et al. (2006), they follow business season in January and February. The control variables: Foreign, ROA, Leverage, Inventory and Yearend hold in prior research to the predicted sign (Goncharov et al. 2013; Hay et al. 2008; Simunic, 1980). In contrast to prior literature Qualified audit opinion shows the expected sign in this research (Goncharov et al. 2013; Jiang et al. 2015). The other coefficients of variables are consistent with prior literature on audit fees (Goncharov et al. 2013; Jiang et al. 2015; Simunic, 1980; Kim et al. 2011).

(33)

As mentioned before, total assets have a significant correlation with both the disclosure quality of IR and audit fees. Also the standard variation indicates a significant variation in the quality of disclosure of IR. Larger firms are more able to disclose higher quality of integrated reports and increased audit effort, which lead to higher audit fees, this are results of the Pearson correlation model. If the quality of IR is low and the client firm size is high the audit fees will be higher and the other way around; if the quality of IR is high of a smaller firm the audit fees also will be higher. I will therefor perform a second OLS regression model that includes the moderating variable irLogAssetstotal. In table 5 the results are presented from the OLS regression model with the moderating variable included. The coefficient of the moderating variable shows a significant negative variation (-2.980**, t stat = 0.001). The results show that total assets reduce the effect the quality of IR has on audit fees. This result indicates that the significant positive relationship of quality of IR on audit fees cannot be explained with the firm size of the client, which means hypothesis 2 doesn’t hold.

Table 5. OLS regression model (including moderating variable)

ir - 0.0074 0.000 LogAssetstotal + 0.6346 0.000 irLogAssetstotal + -2.9800 0.001 Forgein + -0.0117 0.886 Nsegm + 0.0431 0.000 ROA - 0.2333 0.538 Loss + 0.1211 0.174 Receivable + 2.3396 0.000 Inventory + -0.3428 0.117 Lev + -0.0057 0.001 Distres + 1.9210 0.000 Qualified + 0.6973 0.444 Volatility + 0.0005 0.561 Yearend + -0.0954 0.052 MiningandConstruction + 0.4434 0.519 Manufacturing + 0.9019 0.160 TransportationPublicutilities + 0.2856 0.658 Wholesaletraderetailtrade + 0.4247 0.511 Financeinsurancerealestate + 0.1558 0.810 Services + 0.8969 0.168 Publicadministration + 1.2932 0.056 F Value 147.64 Adjusted R2 _0.6832 n 1429

5.4 Robustness Check

(34)

The robustness check is used to test if the regression model consists out of outliers or influential observations. I perform a Cook-Weisberg test for heteroskedasticity and found that a test for robustness is needed (Prob > chi2 = 0.000). The results from table 4 and 5 are tested on their robustness. Table 6 present the robustness performed to check the regression presented in table 4. Overall, the results show that all the coefficients come closer to the significance level. The P>[t] change for variables; ROA, Loss, Inventory, Leverage, Qualified, Volatility and all industry dummies. The leading changes are found Inventory (from sig = 0.074 to sig. = 0.034), Qualified audit opinion (from sig =. 0.444 to sig. = 0.034) and industry dummies: Manufacturing (from sig. = 0.165 to sig. = 0.017) and services (from sig. = 0.192 to sig. = 0.027), which show now a significant coefficient. The significant relationship of Manufacturing and services can be explained by their inventory, receivables and knowledge-based assets. Overall, both the explanation power and the most important variables stay the same.

Table 6. Robustness check

ir - 0.0071 0.000 LogAssetstotal + 0.6094 0.000 Forgein + -0.0075 0.926 Nsegm + 0.0389 0.001 ROA - 0.2111 0.569 Loss + 0.1049 0.278 Receivable + 2.3297 0.000 Inventory + -0.3910 0.034 Lev + -0.0060 0.000 Distres + 1.9538 0.000 Qualified + 0.6987 0.000 Volatility + 0.0005 0.441 Yearend + -0.0995 0.046 MiningandConstruction + 0.4433 0.255 Manufacturing + 0.8943 0.017 TransportationPublicutilities + 0.2917 0.448 Wholesaletraderetailtrade + 0.4024 0.293 Financeinsurancerealestate + 0.0946 0.807 Services + 0.8505 0.027 Publicadministration + 1.3392 0.001 F Value . R-squared 0.6856 n 1429

Table 7 present the robustness check for the regression including the moderating variable. According the Cook-Weisbert test for heteroskedasticity a test for robustness was needed (Prob > chi2 = 0.000). The most important variable in table 7 is the moderating variable irLogAuditfees and this isn’t changed at all. Remarkable is that inventory in table 7 did not become significant like in table 6. Further there are no noteworthy observations.

(35)

Table 7. Robustness check (including moderating variable)

ir - 0.0074 0.000 LogAssetstotal + 0.6346 0.000 irLogAssetstotal + -2.9800 0.001 Forgein + -0.0117 0.886 Nsegm + 0.0431 0.000 ROA - 0.2333 0.538 Loss + 0.1211 0.174 Receivable + 2.3396 0.000 Inventory + -0.3427 0.117 Lev + -0.0057 0.001 Distres + 1.9210 0.000 Qualified + 0.6973 0.444 Volatility + 0.0005 0.561 Yearend + -0.0954 0.052 MiningandConstruction + 0.4197 0.519 Manufacturing + 0.9019 0.160 TransportationPublicutilities + 0.2856 0.658 Wholesaletraderetailtrade + 0.4247 0.511 Financeinsurancerealestate + 0.1558 0.810 Services + 0.8969 0.168 Publicadministration + 1.2932 0.056 F Value . Adjusted R2 _0.6879 n 1429

The effect of Integrated Reporting on audit fees : an empirical study focused on companies based in the European Union.

University of Amsterdam