The impact of non-audit services on auditor independence

(1)

1

The impact of non-audit

services on auditor

independence

Bachelor thesis

Accountancy & control

Fawzia Amir

10590242

Supervisor:

Dhr. A.T.A Koet

Final draft

26 juni 2018

Words: 11274

(2)

2

Statement of originality

This thesis is written by Fawzia Amir 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)

3

Abstract

During the accounting scandals companies as WorldCom and Enron have been collapsed in the period 2000 - 2002. The scandal of Enron was not disclosed by their auditors, which increased public criticism about the auditing profession. Based on this issue this paper provides a literature review on whether the provision of non-audit services do compromise auditor independence, comparing the different researches that use different measurements to proof that the audit is not performed correctly. These measurements are: discretionary accruals and restatements. As a result this study concludes that the results and interpretation are statistically fragile and sensitive to changes in research design, measurement of variables, and assumption of a population. The research question is remained unresolved. There is no unambiguous answer to this question. The main contributions to prior literature are the suggestions and a proposition for future research, stated at the end of the paper.

(4)

4

Summary (Samenvatting)

In de jaren 2000 – 2001 is de accountant betrokken geraakt bij verschillende schandalen zoals de Worldcom en Enron. Er is in de media veel negatieve aandacht voor het accountantsberoep geweest. Ook was er een dramatische groei van het aantal non-audit services (NAS) aan audit-klanten en de omvang van de vergoedingen die worden betaald voor niet-auditdiensten. Deze gebeurtenissen vormde een belangrijke aanleiding voor vele onderzoekers om te onderzoeken of er een relatie bestaat tussen het verlenen van NAS en onafhankelijkheid van de accountant. Accountants worden geacht onafhankelijk op te treden tijdens de jaarrekeningcontrole. Gebruikers moeten kunnen vertrouwen op de informatie uit de jaarrekening. Omdat onafhankelijkheid niet eenvoudig te meten is, hebben de onderzoekers gebruik gemaakt van onder andere de volgende maatstaven voor het aantonen dat een audit niet goed is uitgevoerd: discretionay accruals en restatements.

In deze scriptie is er een literatuuronderzoek gedaan naar het effect van het verlenen van non-audit diensten op de onafhankelijkheid van externe accountants. Onderzoekers hebben de kwaliteit van wat auditors doen afgemeten van het al of niet voorkomen van veel te hoge discretionary accruals. Dat is een teken dat er sprake is van fraudulent financial reporting. Als een accountant een

goedkeurende controleverklaring afgeeft bij een verantwoording waarin sprake is van cooking the books dan heeft hij geen goede kwaliteit geleverd. Als er dan ook nog sprake was van hoge

opbrengsten uit het verlenen van non auditservices dan was hij blijkbaar niet onafhankelijk genoeg.

Uit de onderzoeken blijkt er geen relatie bestaat tussen bedrijven die non-auditdiensten afnemen en de onafhankelijkheid van de accountant. Daarnaast indien een jaarrekening fouten bevat, dan moet restatements plaatsvinden. Omdat fouten vaker voorkomen, wanneer de accountants non-audit diensten afnemen, dan kan er hierdoor de restatements dienen als indicator voor de onafhankelijkheid. Om de invloed van non-auditdiensten op de

onafhankelijkheid aan te tonen, wordt in het literatuuronderzoek gekeken of restatements vaker voorkomt bij bedrijven die relatief veel non-auditdiensten afnemen. Uit de onderzoeken is naar voren gekomen dat non-auditdiensten niet leiden tot meer restatements.

Deze scriptie geeft aan dat er geen eenduidig antwoord kan worden gegeven op de vraag of de onafhankelijkheid van de accountant verslechtert als er non-auditdiensten worden verleend, omdat de resultaten velen afhankelijk zijn van verschillende omstandigheden, waaronder het researchdesign en omdat de beoordelingscriteria verschillende uitkomsten geven. Dit

literatuuronderzoek beperkt zich slechts tot twee beoordelingscriteria met name; discretionary accruals en restatements. Een toekomstig onderzoek kan zich richten op meerdere

beoordelingscriteria en door middel van empirisch onderzoek kan er beter antwoord worden gegeven op de onderzoeksvraag.

(5)

5

1. Introduction

During the accounting scandals companies as WorldCom and Enron have been collapsed in the period 2000 - 2002. After this collapse the stock prices of these companies fell with nearly 40 %. The scandal of Enron was not disclosed by their auditors, which increased public criticism about the auditing profession and raised amongst others questions about the independence of the auditor in charge. Regulators, academics and investors are still questioning auditor independence. One of these question was that auditor independence could have been impaired by providing non-audit services (NAS) (Hope and Langli, 2010, p.574). The reason for this question was the fact that Arthur Anderson played a role in the fraudulent financial reporting (Sinha and Hunt 111, 2013, p.302) and provided a lot of non-audit services to Enron. The audit reports that Arthur Anderson made for Waste

Management were materially false and the results were misleading with an inflation of over $1 billion dollars (Stinson, 2004, p.1). These scandals has led to doubts about the independence of the auditor. This has made me curious to investigate about the independence of auditors with respect to non-audit services.

In response to the accounting scandals the question was raised about whether the auditors were in an independent position during their audits. It is important for the auditor to be independent when performing the audit. The auditor is responsible for detecting material misstatements in the financial statements (NBA, handreiking, 2018). From the year 1960 on the discussion started about the effect of NAS on auditor independence.Regulators of Europe, U.S.A. were concerned about the auditor independence when providing non-audit services. Auditors get a lot of revenues from these non-audit services. So it could be very important to satisfy the client when providing services to him. This causes the auditor’s economic dependence on the revenues of those services and clients that also lead to bias and a loss of objectivity and integrity, because of the relation between the client and auditor (Gul et al., 2007, p118).

It is important that auditors remain independent from the firm who’s financial statements that they are auditing, because of the fact that regulators and other financial statement’s users depend on auditor’s opinion of financial statement (Ashbaugh et al., 2003, p.143).

The Securities and Exchange Commission (SEC, 2000) mentioned that society and regulators of financial markets are concerned that the provision of NAS will compromise auditor independence, because the fees from NAS makes auditors financially dependent on their clients and the nature of these services can sometimes lead auditors in managerial roles. Also, it can lead in reducing audit quality (Habib and Islam, 2007, p.447). More specific, the SEC (2000) states that auditors find that benefits of keeping clients who buys non-audit services, are exceeding the costs of sacrificing auditor independence.

(7)

7

The contribution of this research is as mentioned above to see whether the concerns of the regulators are correct. Also, it contributes to prior literature by setting up suggestions and a propositions for future research.

Motivated by the ongoing debate of auditor independence there is still a concern about the impairment of the independence of the auditor by the fees for NAS.

So the purpose of this thesis is to conduct a literature review which leads to the following research question:

Do non-audit services compromise auditor independence?

This research question will be answered through a literature study. The next chapter of this paper provides background information of important theoretical concepts regarding this subject. It will give an explanation of what is an auditor, what is his role, what is assurance, what are non-audit services, what is auditor independence, what is independence of mind and independence in

appearance. In chapter 3 the discussion of the different outcomes of the researches will be compared. Based on this comparison, an answer to the research question will be given. Chapter 4 describes an interview with two accountants. Lastly, in chapter 5 the conclusion of this study is presented.

(8)

8

2. Background information

2.1 Introduction

In the following paragraph of this chapter some important concepts will be explained. First, the role of auditors. Second, the definition of assurance and non-audit service will be given. In the last paragraph auditor independence will be discussed and the two types of independence: independence of mind and independence in appearance.

2.2 The role of auditor’s

The role of auditor’s according to Arens et al. (2014, p.164) is to provide reasonable assurance that the financial statements as a whole are free from material misstatements in a certain period. For example, when an auditor concludes after his research that the financial statements are misleading and the client is unable/unwilling to correct the misstatements, then the auditor is required to give a modified opinion. In that specific case the auditor will conclude that the financial statements were not prepared in conformity with Generally Accepted Accounting Principles (GAAP) (Arens et al., 2014, p.84). It is the role of the external auditor to express an opinion about whether the financial

statements are presented fairly in accordance with an applicable financial reporting framework.

It is important that auditors have an independent mental attitude so that the society has trust and faith in their work. Confidence is an important aspect for the audit profession. When the financial statement is presented fairly, investors get more correct and accurate information. Hereby investors are more willing to invest. When the society does not perceive the auditor to be independent, the auditor’s opinion is considered useless and the credibility of the financial statement will reduce (Arens et al., 2014, p.25). The auditor can express several opinions.

An unqualified opinion, when misstatement in the financial statements exists, and users’ decisions are unlikely to be affected. For example when auditors determine that there is a substantial doubt about the company’s ability to continue as a going concern then the auditor is required to give an unqualified opinion with an explanatory paragraph in which he expresses his concern about the continuity of the organization. (Arens et al., 2014, p.76). Francis (2004, p.346) add to this that a good audit is designed to provide reasonable assurance that financial statements are free from material misstatements. Misstatements are considered material if the combined uncorrected errors/fraud in the financial statements would influence the decisions of a reasonable individual. Reasonable assurance is a measure of the level of certainty that the accountant has obtained at the end of the audit. In other words, an audit provides assurance this however is not a guarantee (Arens et al.,2014, p.164). Goldman and Barlev (1974, p.707) add to this definition the requirement of consistency and conformity with the accounting principles. According to them the reason that auditors are hired by firms is, that management has different objectives than investors. This can lead to the risk that

(9)

9

management acts in his own interest rather than in the interest of the shareholders. An auditor is also required to give a qualified opinion when he is unable to gather all evidence required by auditing standards. The qualified opinion is considered as the least severe type of departure from an

unqualified report. A qualified report can be a qualification of both the scope and the opinion or the opinion alone. A scope and opinion qualification can be issued when the auditor is unable to

accumulate all evidence required by auditing standards. Whereas the use of an opinion alone is restricted to situations in which the financial statements are not stated in accordance with GAAP (Arens et al., 2014, p.79).

When an auditor issues a qualified report, he must use the term except for in the opinion. This implies that the auditor is satisfied that the overall financial statements are correctly stated “except for” a specific aspect of them (Arens et al., 2014, p.79).

Another opinion that auditors are required to express is an adverse or disclaimer opinion. An adverse opinion can be used when the auditor believes that the overall financial statements are material misstated and do not fairly represent the financial position or results of company’s and the cash flows in conformity with GAAP. Whereas a disclaimer opinion is issued when the auditor is unable to satisfy himself that the overall financial statements are fairly presented. This opinion is necessary because of a severe limitation on the scope of audit or a non-independent relationship between the auditor and its client. Both adverse and disclaimer opinion are used when the condition is highly material (Arens et al., 2014, p.79).

Besides these several opinions to express by auditors, they fulfills several responsibilities. The first one is that they are responsible for planning an audit and supervise assistants. The second one is that auditors are responsible for determining and applying materiality level(s) through the audit. Third one is that they are responsible for assessing risk that financial statements contain material misstatements. Further, they must perform audit procedures to look whether the material misstatements exist. To assess the risk, the auditor must understand the client’s business and industry. The last one is that auditors are responsible for obtaining sufficient appropriate evidence for the assertions of the management in the financial statements by designing and implementing responses to audit-risk (Arens et al., 2014, p.57).

DeFond et al. (2002, p.1248) adds to this by defining the role of an auditor as objectively examining a firms financial performance and resisting pressure from the managers of the company. On the other hand Bingham et al. (2002, p.8) define auditor’s responsibility as to express an opinion with a reasonable assurance.

It is essential that the auditor is independent when giving an objective judgement for an assurance engagement. When users of services do not have confidence in the opinions, or the

(10)

10

Certified Public Accountant (CPASs), the ability of auditors to serve clients is reduced or lost (Arens et

al., 2014, p.131-134).

It can be concluded that the role of auditor’s according to Arens et al. (2014, p.164) is to provide reasonable assurance that the financial statements as a whole are free from material misstatements.

2.3 What are assurance services?

The term assurance services is defined as independent professional services by the auditor in which he gives an opinion about the quality of information. These services are valued because the

assurance provider is independent and has perceived unbiased information. Assurance services improve the reliability and relevance of the information for making decisions. These assurance services can be done by CPAs or several other professionals (Arens et al., 2014, p.28). For example Consumers Union, which is a non-profit organization to tests several products used by consumers. In the Consumer Reports they reports their evaluations of the quality products that is tested. The organization provides these information to the customers to help in making choices for buying the products they select. The purpose of this assurance is to improve the quality of information.

Auditors provided many assurance services with regard to historical financial statement information. Assurance services can be done by several professions.

One category of assurance services is the attestation services that is provided by the CPAs. This is a type of assurance in which the CPA issues a report about a subject matter that is made by the other party (Arens et al., 2014, p.28-29). He gives his opinion about the reliability of information. It is important that the CPA is independent when providing attestation services. The attestation services have five categories namely:

1. Audit of historical financial statements

2. Audit of internal control over financial reporting’ 3. Review of historical financial statements

4. Attestation services on information technology

5. Other attestation services applied to a range of subject

First, the audit of historical financial statements, management suggests that the statements are fairly presented in accordance with the GAAP. This form of attestation services are the most common services provided by the auditors. Second, the audit of internal control over financial reporting, management suggests that the internal controls have been developed and implemented by following the established criteria. Third, a review of historical financial statements, management suggests the same as in category one. The CPA provides a low level of assurance for the reviews of financial

(11)

11

statements. Fourth, attestation on information technology, management makes several assertions about the reliability of information. Other attestation services are natural extensions of the audit of historical financial statements (Arens et al., 2014, p.29-30).

Some types of attestation services are restricted by regulation to licensed CPASs. But other forms of attestation and assurance services are not restricted to non-CPA competitors. For example, CPAs compete with the market research firm to help clients in preparing clients surveys and evaluate the reliability and relevance of information.

Kinney et al. (2004, p.586) defined that assurance services are services for planning and conducting the audit on the financial statements. These services also include discussing the audit related issues with the audit committee and consultation on audit issues which arise during an audit.

It can be concluded that assurance is an independent professional services which improves the quality of information to decision makers.

Auditors also provide other services that differs from the definition of assurance services which is the non-audit services. The definition will be explained in the next paragraph.

2.4 What are non-audit services?

Non-audit services are any professional services provided by CPAs during the period of an audit engagement which are not related to an audit or review of an institution’s financial statements (Brandon et al., 2004, p.90). This services could involve accounting, but is not related to assurance services. There are nine non-audit services which are prohibited (Arens et al., 2014, p.135). These are:

1. Bookkeeping and other accounting services

2. Financial information systems design and implementation 3. Appraisal or valuation services

4. Actuarial services

5. Internal audit outsourcing

6. Management or human resource functions

7. Broker or dealer or investment adviser or investment banker services 8. Legal and expert services unrelated to the audit

9. Any other service that the PCAOB determines by regulation is impermissible

The three specific examples of non-audit services provided by auditors are: bookkeeping and accounting services, tax services, and management consulting services (Arens et al., 2014, p.47).

(12)

12

• Bookkeeping and accounting services. Several small clients with restricted accounting staff rely on auditors to prepare their financial statements. Some of these client lack personnel or expertise to use accounting software. Thus, auditors used these services to meet the needs of these clients.

• Tax services. Auditors are preparing corporate and individual tax returns for both audit and nonaudit clients.

• Management consulting services. Auditors provides these services which enables their clients to run their business more effectively. These services vary from simple suggestions to improve the clients system to advices on risk management, information technology and e-commerce design, business valuation, mergers and acquisitions, and actuarial benefit consulting. The SEC and

Sarbanes-Oxley Act (SOX) restrict auditors from providing these consulting services to public

firms that are audit clients, whereas it is not restricted to private and public firms that are not audit clients.

A problem with the nine non-audit services is, that CPA firms have incentives to keep a client for which they provide audit services, because they receive revenues from non-audit services from this client. The provision of non-audit services could lead to several threats, namely self-interest threats, self-review threats, advocacy threats, familiarity threats, and intimidation threats. Also it could lead to several organizations that are set up to create rules and regulations to be sure that the

independence is not compromised. In contrary to this negative effect of providing NAS, the positive effect is that, a reduce in the client’s consulting risk is mentioned as a reduce in costs, because the auditor understands his client. Also, the audited firm is more dependent in his auditor when the threat of a firm changeover is reduced (Arens et al., 2014, p.135).

According to Ashbaugh et al., 2003 (cited as in Anandarajan, 2012, p.105-106) NAS can create a strong economic bond between the auditor and its client. This is a bond that could impair auditor’s independence. Thereby it is said that it reduces audit quality.

It can be concluded that non-audit services are any services provided by external auditors to their audit clients during the period of an audit engagement which are not related to an audit or review of an institution’s financial statements. In this paper the focus will be on non-audit services.

(13)

13

2.5 What is auditor independence and the two types of independence?

In the previous paragraph the role of the auditor, assurance services and non-audit services were described. It is important that the auditor is independent when performing a financial audit. Independence is an abstract concept and difficult to define. It is the first rule of conduct of the

American Institute of Certified Public Accountants (AICPA) and Securities and Exchange Commission (SEC). There are multiple definitions of auditor independence.

According to Antle (1984, p.1) the American Institute of Certified Public Accountants (AICPA) and the Securities Exchange Commission (SEC) had no precise definition of auditor independence. The AICPA and SEC rules on auditor independence are long and subject to a constant

reinterpretation. Both of them have attempted to provide a precise definition, but it seems hard to give a precise explanation of auditor independence. According to Antle (1984, p.1) there is no ambiguous definition for auditor independence.

DeAngelo (1981, p.116) has a definition of auditor independence in which he said that this is the conditional probability of auditors reporting an error when it has been discovered. Whether an auditor will find an error depends on the technical skills of the auditor. DeAngelo et al. (1981, p.116) suggest that auditor independence is important because it has an impact on audit quality. If auditors are not independent then they will be less likely to report irregularities, which can lead to reduced audit quality.

The AICPA Code of Professional Conduct and IESBA Code of Ethics for Professional Conduct (and the NBA) requires auditors to be independent both in fact (independence of mind) and in appearance. The International Federation of Accountants (IFAC) identify also these two types of independence.

Independence in mind (fact) is also referred as independence in fact. The professional accounting literature provides that auditors should be independent both in fact and in appearance (Campa and Donnelly, 2016, p.423).

Independence in mind (fact) reflects the auditor’s state of mind which permits the audit to be

performed with an unbiased attitude. This refers to actions of auditor based on actual independence, and can be described as the ‘actual’ independent behavior of the auditor (Arens et al., 2014, p.135). Although the profession claims that auditors are able to maintain this mental attitude, some critics argue that auditors will lose the confidence of third parties performing his function as auditor, because the mental attitude is immaterial (Hylton, 1964 as cited in Wines, 1994 p.77). Independence in mind exists when auditors are able to act with objectivity, integrity, professional competence and free from any conflict of interest.

Independence in fact is the foundation of the independence requirement which is not observable and difficult to regulate (Scheutze, 1994 cited as in Dopuch et al., 2003, p.84). Specially,

(14)

14

according to SEC a reasonable investor conclude that an auditor cannot act without bias. Auditor independence is about investors’ confidence in the numbers (Turner 2000 as cited in Dopuch et al., 2003, p.84).

DeFond et al. (2002, p.1248) also defines independence in fact as a mental state of

objectivity. They add to this the auditor’s ability to resist any pressure from the company’s managers. Independence in appearance is the result of others’ interpretations of the independence in fact (Arens et al., 2014, p.135): the auditor should be perceived by others to be independent. Beattie et al. (1999, p.68) describes that independence in appearance is the perception by a reasonable observer that the auditor has no relationship with an audit client which would suggest a conflict of interest. According to regulations, a violation of independence in appearance occurs if auditors provide material amounts of certain kinds of non-audit services to audit clients (Dopuch et al., 2003, p.85).

According to Pott et al. (2009, p.212) independence in appearance is described as the avoidance of significant facts that would cause an informed third party to conclude that a firm’s integrity, objectivity or professional skepticism had been compromised. The result of evidence of auditor failure leads to a loss of trust in the audit process. Hereby it reduces the possibility for an auditor to act independently.

According to Frankel et al. (2002, p.73) independence in appearance is a public perception that auditors are objective when conducting an audit. The SEC give a few examples of a relationship which indicate that an auditor is not independent, such as relationship between auditor and clients, and relationship where an auditor is providing certain non-audit services.

In its ‘Code of Ethics for Professional Accountants’ the IESBA outlines several principles of auditor independence in fact when providing professional service to clients and employers.

- Responsibilities: When auditors are carrying out their responsibilities as professionals, members should exercise professional and moral judgements in all their activities.

- Public interest. Members should accept the obligation to act in a way that will serve, respect the public interest.

- Integrity. To maintain and broaden public confidence, members should perform professional responsibilities with the highest sense of integrity.

- Objectivity. A member should maintain objectivity and must be free of conflicts of interest in discharging professional responsibilities.

- Due care. A member should observe profession’s technical and ethical standards, strive to improve competence and quality of services, and discharge professional responsibility. Integrity, objectivity, and due care are the most important ethical principles when providing

(15)

15

assurance services. Therefore it is important to reveal to the public when an engagement has been carried out with these ethical principles.

It can be concluded that auditor independence is the conditional probability of auditors reporting errors when it has been discovered. Independence of mind is the auditor’s state of mind which permits the audit to be performed with an unbiased attitude. While independence in appearance is the result of others’ interpretations of dependence is.

(16)

16

3. Literature review

3.1 Introduction

During the research many ways were find out to measure the audit quality and to determine that the auditor is independent. By researching the relationship between providing non-audit services and auditor independence the following measurements are used to proof that the audit is not performed correctly. These measurements are: discretionary accruals (the auditor allows that results are

manipulated) and restatements (the financial statement is incorrect). In the research design each measurement is described followed by the discussion, in which the different outcomes are

compared. At the end it will be enclosed with an overall conclusion. The next paragraph will describe the role of discretionary accruals on audit quality and what it means for the independence in mind.

3.2 First measurement: Discretionary accruals

Several researchers find that the quality of what auditor do can be measured, whether discretionary accruals occurred or not. This gives an indication that there is fraudulent financial reporting. For example, when an auditor gives an unqualified opinion and there is sign of a material fraud, the auditor has not performed his role properly. And when it appears that the provision of NAS has yielded in high revenues, then the auditor was apparently not independent.

Discretionary accruals is defined as a form of biased financial reporting which reflects the fact that the books are cooked. The discretionary accruals will be used as a proxy for biased financial reporting, to gain evidence that auditors did not perform a proper job and violated their

independence as a result of clients paying high fees for non-audit services.

Frankel et al. (2002, p.83-97) used discretionary accruals to identify earnings management.

Discretionary accruals is an indicator of earnings managements. Earnings management is a strategy that management of a firm can use to manipulate the profits. If the indicators of the discretionary accruals are low then the audit quality is ok. Frankel et al. (2003, p.71) examine the relation between non-audit services and accruals, because of the concerns about the effect of income from non-audit services on independence of auditors. The argument behind this is, when providing NAS the auditor was not independent because he did not corrected the financial report where there was fraudulent financial reporting (cooking the books).

In their research they use the ratio of non-audit fees to total fees associated with abnormal accruals as a measure of auditors’ incentive in impairing their independence (p.82). They used a sample of 3,074 proxy statements for their research (p.76). Frankel et al.(2003, p.82) used three specifications of auditor fees. The first is the ratio of non-audit fees to total fees (FEERATIO). Second, is the rank of non-audit fees (RANKNON) and audit fees (RANKAUD). And the third, is the rank of total

(17)

17

fees (RANKTOT). Frankel et al.(2003, p.85) suggest that their findings implies that organizations with a high FEERATIO have higher cash flows and profitability, lower total accruals and stock performance, higher litigation risk, growth, which are more likely to engage in financing and acquisition activities. The results of Frankel et al. (2003, p.85) indicate that FEERATIO is positively correlated with

RANKNON and RANKTOT. This means that in the case of providing non-audit services to audit clients

the auditors have more incentives to acquiesce to client pressure to manage earnings. And FEERATIO is negatively correlated with RANKAUD. This means that the provision of non-audit services has no effect on earnings management.

Srinidhi and Gul (2007, p.598-605) use in their research a regression model to measure the quality of the accruals. Hereby the total current accruals is the dependent variable and the cash flows the independent variable. These variables helps to improve the expectation of current accruals. They use the absolute value as the dependent variable to determine the effect of economic bonding (p.600). Fees from non-audit services could also have an effect on auditor’s independence because it creates an economic bond. Srinidhi and Gul (2007, p.602) imply to separate audit and non-audit fees. They use size as a control variable in the estimation of economic bonding in accrual quality and conduct a robustness check by conducting small and large firms. Srinidhi and Gul (2007, p.605) use a sample of 4,282 firms.

Srinidhi and Gul (2007, p.598) examined the effect of audit fees and non-audit fees on the quality of accruals. The quality of an audit can be measured by how much effort auditors put into reducing error in accruals. They suggest that audit quality is affected by auditor independence when reducing the estimation errors in accruals. Hereby the independent auditor require management to correct his estimates and change the accounting methods to improve accrual quality. Furthermore, when auditors are competent and independent it will discourage managers from making intentional errors and motivate them to focus more in reducing unintentional errors. Even when the auditor is independent to detect errors, economic bonding can reduces accrual quality.

Larcker and Richardson (2004, p.629-630) examined the relation between audit fees and accruals by using a single regression model with a restricted sample of 5,815 firms from the years 2000 and 2001. Larcker and Richardson (2004, p.633) did the same test as Frankel et al. (2002). They used discretionary accruals to estimate earnings management. Discretionary accruals is according to them the difference between cash flows and net income. Discretionary accruals are easier to manipulate than cash flows. In many companies, cash flows and revenues are not identic to each other. There are two types of accruals namely discretionary and non-discretionary. This difference is based on

(18)

18

whether the management can affect its magnitude. Thus, to use accruals as a measure it is useful to examine the quality of financial reports.

Larcker and Richardson (2004, p.627) specifically looked at abnormal/ discretionary accruals. They suggest that as the amount of discretionary accruals increases, earnings quality decreases since earnings are less reflective of the firm’s performance and are more reflective of the amount of earnings management. Larcker and Richardson (2004, p. 627) found a positive association between fees paid and unexpected accruals, when they used the ratio of non-audit fees to total fees paid. However these findings only applies to 8,5% of their sample, but when Larcker and Richardson (2004, p. 627) use the ratio of dollar fees paid scaled by total fees revenue and abnormal audit fees based on the expectation models developed by Simunic (1984) and Crashwell et al (1995), they found a negative association between fees paid and unexpected accruals. According to Larcker and

Richardson (2003, p.626) the relationship between discretionary accruals and auditor independence is negative. This means that auditor independence is not compromised by providing non-audit services.

Another research is Ashbaugh et al. (2003, p.612). He also examined discretionary accruals to get evidence of auditor independence violations. They suggest that total fees captures economic bond between the audit firm and client. The fee ratio captures the relative value of audit versus non-audit services provided by the audit firm to a client.. Ashbaugh et al.(2003, p.614) has a critique on this ratio. In their research design, they investigate whether client auditor bonding is associated with evidence that financial statements reflect a biased view of the client’s financial condition. They use discretionary accruals and the likelihood of meeting earnings benchmarks which can be viewed as evidence of auditor independence violations.

Ashbaugh et al. (2003, p.615) use a sample of 3,170 firms and four audit fee metrics in their tests: AUDIT and NONAUDIT, TOTAL and FEERATIO. They investigate the robustness of Frankel et el.(2003) accrual tests by using two alternative measures of accruals in the regression model. First measure is Portfolio Performance Adjusted Discretionary Current Accruals (PADCA), which controls firm performance through a portfolio technique. Second measure is ROA in Estimation Discretionary Current Accruals (REDCA), which controls firm performance (p.621). In their test they do not include total accruals or absolute value of total accruals, because these variables derived accruals which can result in overstatements (p.624-625).

They found no significant relation between FEERATIO and positive PADCA or positive REDCA. They also found no significant relation between TOTAL and positive PADCA or positive REDCA. This means that there is no relation between non-audit services and auditor independence.

(19)

19

Chung and Kallpur (2003, p.932) examined the relationship between abnormal accruals and auditor independence. They use abnormal accruals as a measure of earnings, and therefore, audit quality. They use a sample of 1,871 firms. According to Chung and Kallapur (2003, p.939) the cross-sectional Jones-model abnormal accruals are associated with audit qualifications. They estimate the following model:

ACC = aₒ 1/TA ₋ 1 + a ₁ (∆ SALES - ∆ AR)/TA ₋ 1 + a₂ PPE/TA ₋ 1 + e,

where:

ACC = total accruals deflated by TA ₋ 1; they define accruals as the difference between earnings (before extraordinary items and dis- continued operations) and cash from operations;

TA ₋ 1 = total assets at the beginning of the year; SALES = sales;

AR = accounts receivable; and

PPE = property, plant, and equipment.

They estimate the model separately for each industry (two-digit SIC code). They calculate abnormal accruals, DACC, as the residual from the above model. For each industry, they estimate Model (1) using all firms belonging to that industry (including sample and non-sample firms). Because Levitt (1998) expresses concerns about "cookie-jar accounting," implying that firms manage earnings downward as well as upward, they use the absolute value of abnormal accruals in subsequent tests (p.939).

They use the ratio as a measure of auditor independence impairment would imply that, the amount of non-audit fees in relation to audit fees is perceived to impair independence only by those who do not understand the process of auditing.

One interpretation that could support the use of the non-audit to audit fee ratio as a measure of auditor incentives to compromise independence is as follows. It could be that there are only certain specific types of non-audit services that impair an auditor’s independence -this is the logic underpinning regulatory and legislative actions to ban specific non-audit services as do the SEC Final Rule (SEC 2000) and the Sarbanes-Oxley Act of 2002. As the ratio of non-audit to audit fees increases, it could become more likely that the provision of non-audit services do not impair the independence of auditors (p.949).

(20)

20

3.2.1 Discussion

As mentioned before in paragraph 3.2 we discussed the role of discretionary accruals in the quality of the audit- and the independence of auditor. These were five researches and they all conclude

different.

The academic researches discussed above provides different results on whether providing non-audit services compromise auditor independence. They use discretionary accrual to measure the quality of the audit and the independence of the auditor. The idea is that an auditor who is not independent because he provides a lot of non-audit services will not perform a good audit, which is reflected in the fact that the auditor accepted earnings management. According to Ashbaugh et al. (2003, p.634) the difference in results is due to the different types of research design. The

researchers of Frankel et al. (2002) and Srinidhi and Gul (2007) use all accruals as a measurement and came to the same conclusion that providing NAS do compromise auditor independence. Larcker and Richardson (2004), Ashbaugh et al. (2003), and Chung and Kallapur (2003) also use accruals as a measurement and conclude that there is no evidence that providing NAS do compromise auditor independence.

Frankel et al.(2002, p.89) found a positive relation between NAS fees and the amount of discretionary accruals. This means that Frankel et al. (2002, p.74) provides evidence that the provision of non-audit services do compromise auditor independence. The reason that they give for their result is that if the auditor faces the possibility to loose non-audit services fees, he will be less resistant to management pressure. The study of Frankel et al. (2002, p.89) also found that providing NAS (which was indicated by the ratio between non-audit fees and total fees) has a relation with: 1) the likelihood that

reported earnings are equal or higher than the expectations of analysts and 2) the amount of abnormal accruals. The researchers say this is evidence that the independence of an auditor is decreased when clients have to pay a relatively high non-audit fee (compared to total fees). The study of Srinidhi and Gul (2007, p.596) shows that there is a relationship between feerelated variables and the quality of accruals, which means that they find evidence that the provision of non-audit services compromise auditors’ independence by examining the effect of both audit and non-audit fees on accrual quality. They find that reduced quality of accruals reflects the negative effects of an economic bond between the auditor and the client. This bond reflects a reduction in the independence of the auditor, which indicates that the provision of non-audit services compromise auditors’ independence.

The study of Larcker and Richardson (2004, p.627) shows that it is not yet known what the role of fees for audit and non-audit services is regarding the choices an auditor makes. Their findings indicate that the provision of non-audit services does not compromise auditor independence when

(21)

21

the ratio of dollar fees paid scaled by total fees revenue and abnormal audit fees is used. They found a positive relation between fees and unexpected accruals, but only when they measured the ratio of non-audit fees to total fees, and when unexpected accruals were made non-directional (absolute).

The study of Ashbaugh et al.(2003, p.634) looked at the results of Frankel, which indicate that there is a positive relation between non-audit fees and biased financial reports. They used the same tests as Frankel to see if the results were influenced by the research choices they made. The key difference is that Ashbaugh et al.(2003, p.634) found no relation between positive discretionary accruals and any of the measures of auditor fee (when performance of the firm is controlled for).So they are actually concluding that the findings of Frankel et al (2002) are very doubtful and that there is no evidence that indicate that the provision of non-audit services compromise auditor

independence.

The study of Chung and Kallapur (2003, p.951) looked at absolute, unexpected accruals as a proxy for earnings quality (and therefore also audit quality). They didn’t find a significant relationship. This could be due to the fact that their tests had low power. The model they used was based on the theory of auditor independence. The researchers suggest a way of splitting up the sample to create samples in which the independence of auditors is more likely to be reduced. When they split up the samples on the basis of size, client opportunities or incentives to do earnings management, they didn’t find a significant relation between the ratio of client importance and abnormal accruals. All in all there was no evidence that the reduction of an auditor’s independence is influenced by different client fee ratios, which concludes that the provision of NAS do not compromise auditor

independence.

3.2.2 Conclusion

Based on the five researchers as discussed above it can be concluded that the researchers using discretionary accruals as a measurement are more divided with regard to do NAS compromise auditors independence. Three researchers conclude there is no relationship; two researchers namely Frankel et al. (2002) and Srinidhi and Gul (2007) conclude there is a relationship between non-audit services and auditor independence. The main argument of Frankel et al.(2002) was that when the probability of providing NAS is higher, the discretionary accruals will be higher, that causes audit quality to reduce. And the argument of Srinidhi and Gul (2007) is that high amount of non-audit fees decreases accrual quality, which indicates a reflection of the adverse effect of economic bonding of auditors that result in a loss of independence. Based on what is discussed above, it can be concluded that the research question of do non-audit services compromise auditor independence can not be answered. The researchers that use discretionary accruals show the most contrasting outcomes, but as said above, these researchers namely Frankel et al.(2002) and Srinidhi and Gul (2007) are given

(22)

22

less weight. Besides, for all the exceptions is found an explanation. The exceptions are caused by the culture of the corresponding country or the period.

This table below is given a summary of the researchers as mentioned above.

Author Do non-audit service

compromise auditor independence?

Main argument

Frankel et al. (2002) Yes The higher the probability of providing NAS, the higher de discretionary accruals, that causes audit quality to reduce. Srinidhi and Gul (2007) Yes High amount of non-audit fees decreases

accrual quality, which indicates a reflection of the adverse effect of economic bonding of auditors that result in a loss of independence.

Larcker and Richardson (2004)

No There is a negative relation between auditor independence and accruals. Auditor’s are less likely to allow abnormal accrual choices for firms where they have the greatest financial interest or

dependence on the audit client.

Ashbaugh et al. (2003) No The results are sensitive to research design choices and very doubtful indicating that independence will not impaired.

Chung and Kallapur (2003) No Their analysis and tests cast doubt and there is no relation between

independence and accruals. Table 1: Overview of relation between auditor independence and discretionary accruals.

On the basis of these five researches using accruals as measurement, all of them came with different results. The results and interpretation of these researches are statistically fragile and sensitive to changes in research design, measurement of variables, and assumption of a population. The research

(23)

23

question of do providing non-audit services compromise auditor independence is remained unresolved using accrual as a measurement. There is no unambiguous answer to this question.

(24)

24

3.3 Second measurement: Restatements

Researchers find that to comply the laws and regulations, auditors have to adjust errors in the financial statements by using the measurement restatements. Restatements are adjustments made in the financial statements of a previous year. It shows directly the way of financial reporting was violating the accounting principles. The principle for using this measurement is that an independent auditor perform less audit quality and indicate that the auditor was not independent, whereby financial statements often contains errors. In this research I want to find out whether the restatements often occur in firms that purchases non-audit services.

Kinney et al.(2004, p.563-564) looked at whether prices for non-audit services (NAS) change when financial statements have to be restated. The reason for this use is because audits are more intensive to carry more legal and regulatory responsibilities (p.569). Kinney et al. (2004, p.569) used

restatements to measure bad financial reporting. It is about financial statements that were audited or reviewed by an independent firm. The firm found that the numbers were wrong and against the GAAP. This is why they were restated.

The study of Kinney et al. (2004, p.578-579) had a sample of 617 companies that issued financial statements. In the study, RESTATE stood for the company which restated the financial statement. ACQUIS stood for acquisitions. Acquisitions can make restatement more likely. Acquisitions can also make it necessary to get tax advice, systems advice, business advice, or

additional investigation. This is because an acquisition can create new problems, or make accounting different or more.

Kinney et al.(2004, p.) findings result are that there is a relationship between RESTATE and: acquisition, audit fees, taxes, and other services. The relationship with taxes is negative. For smaller companies, two factors may therefore cancel each other out. These factors are: 1) economic

dependence and 2) factors that improve the quality of tax-related financial reporting The relationship between tax fees and financial restatements is not significant for smaller companies. For larger companies, however, the relationship is significant and negative.

Another research is of Schmidt (2012). This study examines the initiation and resolution of audit restatement-related litigation to provide evidence on whether NAS or another source of fee dependence) impairs perceived auditor independence. Because the initiation of audit litigation requires an expressed allegation of an audit failure, auditor independence perceptions can be directly observed in the legal arguments presented for why a substandard audit occurred (p.1037). Schmidt (2012, p.1045)) use a sample of 1,543 observations and restatements to measure the proportion of an auditor's revenue that was obtained through NAS, because the NAS fee ratio is the

(25)

25

metric, most often cited in the popular press as a measure of auditor independence (Finance Week (p.1040). He suggest that a positive and significant coefficient on NAS Impair would provide evidence. The results of Schmidt (2012, p.1045) is, restatements that result in audit litigation involve clients that are more economically significant to an auditor's office portfolio and clients that are billed a greater proportion of NAS fees. Restatements that result in audit litigation also have significantly more GAAP failures.

Schmidt (2012, p.1055) findings result is that this sensitivity test indicate that, when fees are higher (especially fees from NAS), the likelihood of restatement-related audit litigation is greater. However, the severity of a restatement plays a larger role in litigation initiation than the relative magnitude of the fees.

In contrast with Kinney et al.(2004) and Schmidt (2012), Raghunandan et al. (2003, p. 225) looked at the relation between firms that do restatements and the fees they charge for NAS. They looked at auditors during periods in which they had to issue restatements.

The sample was 3,591 financial restatements. The argument that non-audit fees reduce audit quality may apply to any fees that the auditor charges. In other words, if the economic bond that is created by charging non-audit fees reduces the quality of audits, this should apply to all fees, including the actual audit fee (p. 227-228).

It was expected that there is a positive relationship between non-audit fees, non-audit fee ratio, and total fees on the one hand, and (1) firm size, (2) growth, (3) complexity, (4) using a Big 5 auditor, and (5) special situations on the other hand. Special situations include mergers, issuing debt or equity and other unusual situations. This was indeed what they found. Also, there is a negative relation between each of these three factors (non-audit fees, fee ratio and total fees) and

profitability.

The results show that firms which do restatements probably won’t have non-audit fees, fee ratios or total fees that are higher than usual. The study does not support the idea that restatements are more likely in firms that pay higher non-audit fees or total fees (p. 231).

Another research is of Agrawal and Chadha (2005, p.377). They examined the effect of NAS fees on restatement probability using a sample of 318 firms of which 159 firms had restated earnings. They stated that with the revelation of accounting problems potential conflict of interest generated by the lack of auditor independence have received research from the media. The auditors have long resisted calls to refrain from providing non-audit services to their audit clients. They argue that providing NAS to audit clients do increases their knowledge and understanding of the client’s business, which result in better audit quality.

(26)

26

The variables that is used are: independence of the board/audit committee, financial expertise of the board/audit committee, CEO’s influence in the board, auditors’ fees. The data was collected from the last proxy statement that was issued before the restatement. This is because firms may make changes to their board/audit committee/CEO after a restatement. (p.386)

Agrawal and Chadha (2005, p.392) findings result are that the likelihood of restatement (RESTATE) is lower in firms that have an independent director who has financial expertise on the board (IDFE) or on the audit committee (IDFEAUD). The number of independent directors on the board (PID) is positively related to the proportion of independent directors on the audit committee (PIDAUD). It is higher if the CEO is part of the family which founded the company.These relation are robust to alternative specifications.

Agrawal and Chadha (2003, p.376) suggest that auditors with no experience in accounting are less likely to detect problems in financial reporting. On the other hand it is not clear that auditors with expertise can detect accounting irregularities. Alternatively, an auditor with expertise can lead other auditors to become less alert.

The last research is of Bloomfield and Shackman (2008, p.128-129). This study looked at the relation between charging NAS fees on the one hand and earnings-related restatements on the other. Restatements in this study were a proxy variable for earnings management. This study looked at the relation between specialized and non-specialized auditors on the one hand, and restatements on the other. They found a negative relationship (p.128-129).

Lawmakers and the public believe that NAS fees reduce the independence of auditors. According to the Sarbanes-Oxley Act, NAS fees are particularly dangerous. This is because they can create a conflict of interest, because they can give the client extra power over the auditor when negotiating. Therefore, the researchers expected a positive relationship between NAS and the likelihood of restatement.

Their sample consisted of 250 restatements of financial statements. The study directly compared the characteristics of auditors with the non- audit services fees they charged. The researchers thought that these two factors would influence the likelihood of restatements. They found a significant relationship between total fees and the likelihood of restatements. Also, they found a negative relationship between restatements and the specialization of an audit firm. The conclusion is that the characteristics of an auditor have much more influence on the likelihood of restatements than non-audit services fees.

(27)

27

3.3.1 Discussion

As mentioned before in paragraph 3.3 we discussed the role of restatements in auditor’s quality together with the independence of auditor. These five researches has different findings with different conclusion that will be defined.

Kinney et al. (2004, p.585) find some evidence for a significant positive relationship between audit fees, audit-related fees, and unspecified NAS fees/restatements. Unspecified NAS fees have a positive relationship with restatements, which indicates that the provision of non-audit services impairs auditors’ independence. Some unspecified NAS may create an economic dependence. This economic dependence leads to more restatements. Unspecified NAS/restatements also leads to a lack of factors that improve the quality of financial reporting.

Schmidt (2012, p.1059-1060) find evidence of a positive association between the amount of auditor fees and in particular, unspecified NAS fees, and the likelihood of restatement-related audit litigation. This study also finds that more severe restatements and restatements involving

revenuerecognition failures are more likely to result in audit litigation. This finding is important because it suggests that, despite the regulatory regime shift that occurred in 2002 and the recent surge in the number of restatements, the primary drivers of audit restatement-related litigation remain largely constant across time. The major exception to this conclusion is that NAS fees are positively associated with the initiation of audit litigation and NAS fees and client importance are associated with the resolution of audit litigation in the post-2000 time period. The positive association refers to that providing NAS will decrease audit quality, thereby impairing auditor independence.

According to Raghunandan et al.(2003, p.232), the SEC, lawmakers and others speculate that if auditors charge fees for non-audit services, restatements are more likely. This study looked at the relation between firms that did restatements and 1) the amount of non-audit fees they charged, 2) the fee ratio and 3) the total fees they charged. They used the values for non-audit fees, fee ratio and total fees to calculate unexpected values for each firm. They found no significant differences

between the two groups regarding any of the three variables. The study therefore does not conclude that non-audit fees reduce the quality of the audit, which would mean the audit has to be restated. This indicates that providing NAS do not impair auditor independence (p.232).

According to Agrawal and Chadha (2003, p.403) they found that non-audit services have no relation to the likelihood of restatements. These mechanisms are: independence of board/audit committee, and how much auditors provide NAS. Restatements are significantly less likely if a

(28)

28

They are significantly more likely if the CEO is part of the family which started the company. There is no strong association between NAS and restatements and auditor independence.

Bloomfield and Shackman (2008, p.137-138) study showed that the characteristics of an auditor are more important than the fees charged by an auditor, regarding the likelihood of

restatements. In this study, restatements were a proxy variable for earnings management. However, the total fees do have a small, significant effect on the likelihood of restatements. The result of the study states that when providing NAS auditor independence will not be compromised.

3.3.2 Conclusion

The purpose of this study was to conduct a literature review which leads to the following research question: Do non-audit services compromise auditor independence? To answer this question several researchers discussed above used restatements as a measurement. Three of the five researchers shows no relationship between non-audit services and auditor independence. Only the research of Kinney et al. (2004) and Schmidt (2012) shows a small relationship. The main argument of Kinney et al. (2004) is that purchasing non-audit services increases the probability of restatements of audited or reviewed financial statements, which indicates that independence is reduced, due to the

economical dependence of the auditor to his client. And the argument of Schmidt (2012) is that more severe restatements and restatements involving revenue-recognition failures are more likely to result in audit litigation. This means that NAS fees are positively associated with the initiation of audit litigation and NAS fees that causes in reducing auditor independence. So based on the measurement restatements there is no direct relationship between non-audit services, audit quality and auditor independence.

This table below is given a summary of the researchers as mentioned in paragraph 3.3.1

Author Do non-audit service compromise auditor independence?

Main argument

Kinney et al.(2004) Yes Purchased non-audit services increase the probability of restatement of audited or reviewed financial statements, which indicates that independence is reduced, due to the economical dependence of the auditor to his client.

Schmidt (2012) Yes More severe restatements and restatements involving revenue-recognition failures are more likely to result

(29)

29

in audit litigation. This means that NAS fees are positively associated with the initiation of audit litigation and NAS fees, that causes in reducing auditor independence.

Raghunandan et al.(2003) No Firms paying higher fees for non-audit services are less likely to restate their financial statements that lead in increasing auditor independence.

Agrawal and Chadha (2003)

No A restatement is significantly lower in companies whose boards or audit committees include an independent financial expert. This means that the independence of the auditor is not reduced. Bloomfield and Shackman

(2008)

No Characteristics of the auditor are more important predictors of earnings management as measured by restatements. NAS fees do not have strong

explanatory power, because higher NAS fees are less likely to restate earnings, thereby casting doubt on the public perception, that reduce auditor independence.

Table 2: Overview of relation between auditor independence and restatements.

On the basis of the five researches as mentioned in the previous paragraph their results are different and that non-audit services do not lead to more restatements. There is no relationship or a strong relationship between non-audit services and audit quality and auditor independence.

(30)

30

3.4 Conclusion: Quality of the measurements

The amount of discretionary accruals and restatements are all different measurements for auditor independence. Which of these measurements has an impact on auditor independence? To compare these two measurements it is possible to give outcomes of the researchers of one category of measurement to look which of these weight more than the other categories of measurement. First discretionary accruals. This is less direct way for measuring earnings management. The researchers Frankel et al. (2002), Srinidhi and Gul (2007), Larcker and Richardson (2004), Ashbaugh et al.(2003), and Chung and Kallapur (2003) use all accruals as a measurement. Discretionary accrual is a noisy measure for earnings management. When a study finds that there is a relation between nonaudit services and accruals, then is does not have to mean that there is a relation between NAS and auditor independence and audit quality.

Second restatements. This is a very direct measurement of the failure of an auditor’s independence. The researchers Kinney et al. (2004), Schmidt (2012), Raghunandan et al.(2003), Agrawal and Chadha (2003), and Bloomfield and Shackman (2008) use all restatements as a measurement. It shows directly that the way of financial reporting was violating the accounting principles and thus that the auditor was not critical or enough. This could indicate that the auditor was not independent. It is stated here that there is a positive relationship between the amount of non-audit services and the amount of restated financial statements and auditor independence. It can be concluded that the measurement restatements are stronger than discretionary accruals. Based on the discussion of the several researchers of each measurements, it can be

concluded that there is no relationship between non-audit services and auditor indepenendence. The researches that use discretionary accruals and restatements shows the most contrasting outcomes to this conclusion. Besides, for all the exceptions is found an explanation. The exceptions are caused by the culture of the corresponding country or the period.

(31)

31

4. Interviews with accountants

Building on my research question: do non-audit services compromise auditor independence? I also want to do a little pilot study, for which I interviewed 2 accountants. The main focus of the interviews was the accountants’ degree of independence. The interviews consisted of 3 questions aimed at obtaining an answer to my research question. I will now briefly describe the 2 interviews, referring to the accountants as follows:

- WH (initials only), referred to as ‘Z’ - RR (initials only), referred to as ‘B’

4.1 Interview with accountant Z

Z has worked for PwC for two years as an Associate. In the interview, he indicated that his company regularly emphasizes the importance of professional principles, including professionality, integrity, objectivity and independence. Furthermore, he receives ongoing training through online courses as well as training-focus group seminars which take place four times a year. In these meetings, he and his colleagues share experiences and talk about how to best put these principles in practice.

The company values independence in particular. Accountant Z carefully reflects on his own independence during each planning stage. When we discuss how much independence he feels he has in different aspects of his work, he notes that the topic of independence is becoming increasingly important.

He also described ways in which his independence might be compromised when providing nonaudit services. Z believes that as it currently stands, the independence of accountants is not yet compromised. At PwC, each field of expertise is made up of various departments (Assurance, Tax and Consultancy). Other departments can only gain access to Assurance-related files if permission is obtained from the RAs. In his personal opinion, accountant Z believes that a more stringent measure that has been proposed for the future – namely that different departments should be separated entirely and function as independent companies – is much too drastic.

Finally, he notes that loss of independence as a result of non-audit services has occurred more than once. He documents this with every step he takes. Nowadays, audit working papers are no longer used the way they once were, but function more like logbooks. In response to this

development, management continually reminds him from higher up that he should strive to preserve his independence.

(32)

32

4.2 Interview with accountant B

Accountant B has worked as a Senior Audit & Assurance Manager for AREP Accountants & Tax Advisors B.V. for twelve years. He believes that the fundamental principles of his profession are embedded in the so-called VGBA (‘verordening gedrags- en beroepsregels accountants’; English: regulations and code of conduct for accountants), particularly professionality, integrity, objectivity, competence, diligence and confidentiality. Professionality applies to all of the accountants’ actions and negligence’s (including private affairs). Independence is included in the so-called VIO

(‘verordening inzake onafhankelijkheid’, English: independence regulations). In his capacity as a professional, the accountant is required to comply with the regulations and codes of conduct listed in the VGBA and VIO. AREP Accountants & Tax Advisors B.V. employs office regulations designed to promote these principles. In doing so, they take have taken sufficient steps to mitigate threats and perform their services adequately.

Regarding the degree to which non-audit services might hamper accountants’ independence, B believes that providing combined assurance and non-assurance services to another entity may naturally affect the accountant’s independence if the accountant provides a non-assurance service that has a material effect on the assurance object. One example is self-assessment. In such a case, the accountant may not accept the commission if the non-assurance service in question is subjective or non-routine. This is because it could constitute a threat if there is conflict of interest, or if the assurance team casts doubt on the suitability of a particular method of processing the assurance object. The commercial interest of the accountant remains an important dilemma in daily practice. The accountant will always have to weigh his options carefully.

To the abovementioned example, B adds that if the accountants options to offer an entity both assurance and non-assurance services anyway, he/she needs to record his considerations in the file. Additionally, the compliance officer may have to issue a statement on whether sufficient

measures have been taken to preserve independence. B’s company does not offer non-assurance services that have material consequences for the assurance object. It is, however, possible to offer a combination of assurance and non-assurance to an entity simultaneously, but in such cases, the nonassurance services in question have no material consequences for the assurance object. Moreover, non-assurance services are carried out by other (non-audit) departments in accountant B’s network. In its capacity as an accountancy firm, AREP Accountants & Tax Advisors B.V. does not perform assessments for the entity, nor does it adopt any particular stance on behalf of the entity regarding non-assurance services. The case is always proposed to their in-house compliance officer.

The impact of non-audit services on auditor independence