The impact of prohibited non-audit
services and mandatory rotation on
auditor independence
Supervisor: C. Clune
Version: 2.0
Name: Romee Kruiswijk Student number: 10254331
Program: BSc Accountancy & Control Field: Auditing
Inhoud
Abstract ... 3
Samenvatting ... 4
1 Introduction ... 5
2. Background Information ... 6
2.1 Role of the accountant ... 7
2.2 What is independence ... 8
2.3 Separation between audit and non-audit services ... 11
2.5 Mandatory rotation ... 12
3 Literature review ... 14
3.1 Analysis of the impact of non-audit services on independence ... 14
3.1.1 Evidence from the US ... 14
3.2 Analysis of the impact of mandatory audit rotation ... 23
3.2.1 Mandatory rotation has a negative effect on auditor independence ... 23
3.2.2 Mandatory rotation has a positive effect on auditor independence ... 25
5 Conclusion ... 28
6 References ... 30
Abstract
Since the fall of Enron several large corporate frauds were discovered all over the world. As a result of these frauds the independence of the auditor was questioned by various organizations and users of financial statements. In order to restore the confidence of the public in the auditor two important measures were implemented by the Sarbanes-Oxley Act (SOx): a ban on
various non-audit services was introduced and it was obliged to rotate audit partners after every five years. The question that is tried to be answered in this literature review is whether these measures have had the intended impact on auditor independence. This adds to the existing literature by looking at the combined effect of the ban on non-audit services and mandatory rotation because much has been written about both topics separately, but not many articles were published about the joint effects. The main findings of this literature review are that perceived independence has improved by the ban on non-audit services, but independence in fact is difficult to find. For mandatory rotation there are more opponents than proponents. Two main arguments that were put forward to support this outcome are the costs which are involved with rotation and the economic argument that auditors are independent because they need to protect their reputation mandatory rotation does not improve auditor independence.
Samenvatting
Sinds het faillissement van Enron, een grote energiemaatschappij in Amerika, werden er op wereldwijde schaal ontdekkingen gedaan van fraude in de financiële jaarverslaggeving. Als gevolg van deze fraudes kwam de onafhankelijkheid van de accountant in een negatief daglicht te staan bij verschillende onafhankelijke toezichthoudende organisaties en bij de gebruikers van financiële jaarverslagen zoals investeerders. In een poging om het vertrouwen van de samenleving in de accountant te herstellen is de Sarbanes-Oxley (SOx) wetgeving in Amerika als eerste ingevoerd. Hierna volgden ook andere landen met soortgelijke
wetgevingen, maatregelen en codes waar de accountant zich aan dient te houden. De SOx-wetgeving bevat onder andere twee belangrijke maatregelen met betrekking tot adviesdiensten en verplichte roulatie van accountant na iedere vijf jaar. In dit literatuuroverzicht wordt
getracht een antwoord te geven op de vraag of deze maatregelen het beoogde effect hebben gehad op de onafhankelijkheid van de accountant. Het onderzoek levert een bijdrage aan de al bestaande literatuur door te kijken naar het gecombineerde effect van het verbod op
adviesdiensten en verplichte roulatie in plaats van, zoals in veel gepubliceerde stukken, naar elk van deze maatregelen afzonderlijk. De voornaamste resultaten van het literatuuronderzoek zijn dat onafhankelijkheid in schijn daadwerkelijk verbeterd is door het invoeren van het verbod op adviesdiensten. Dit is aangetoond door onderzoeken in de VS, Duitsland, Engeland, Denemarken en Ierland waarin verschillende gebruikers van financiële jaarverslagen hebben geparticipeerd. Onderzoek op dit gebied naar onafhankelijkheid in wezen kan niet aantonen dat er ook op hier een verbetering waar te nemen is. Op dit moment zijn er slechts een aantal landen die gebruik maken van verplichte roulatie van
accountantskantoor om de zoveel jaar. Er is daarom vooral gekeken naar onderzoeken uit de VS waar verplichte roulatie van audit partners verplicht gesteld is. Vanwege de hoge kosten die verplichte roulatie met zich meebrengt en de economische belangen van de accountant om zijn reputatie in het maatschappelijk verkeer te beschermen wordt verplichte roulatie niet gezien als een maatregel die ervoor kan zorgen dat de onafhankelijkheid van de accountant zal verbeteren.
1 Introduction
In December 2001 a large utility company in the United States named Enron filed for bankruptcy. After this day a process of several years began which made this case one of the biggest and best known in the US history (Benston & Hartgraves, 2002). During this process it appeared that the large rise in stock prices over the past decade was caused by too optimistic accounting practices in the previous periods. Therefore restatements had to be made which resulted in a sharp decline in stock price which led to an investigation by the Securities and Exchange Commission (SEC), the organization that is responsible for the supervision of financial markets in the United States. When the investigation by the SEC revealed that the payments that Arthur Andersen received from Enron were 25 million for audit services and 27 million for non-audit services the question rose whether the auditor was sufficiently
independent and whether the trust in the financial statements was still increased by auditing (Benston & Hartgraves, 2002). In response to this fraud SEC chairman Pitt decided that it was time to create a new oversight body to control and regulate auditing firms. On the 25th of July 2002 the new Sarbanes-Oxley Act (SOx) became effective (Zhang, 2007). The ban on several non-audit services by the SOx was not the only measure that has been taken to improve auditor independence. Since 1985 Arthur Andersen has been the audit firm of Enron, which means that the relationship between the companies existed for more than fifteen years until the fraud was discovered. The longer the relationship the more familiar companies get with each other which can reduce the willingness to challenge mistakes in the financial statements (Benston & Hartgraves, 2002). Therefore mandatory rotation was also included in the SOx regulation to ensure auditor independence.
Motivated by the ongoing discussion of auditor independence the purpose of this thesis is to provide an answer to the question what the impact is of the separation between audit and non-audit services and mandatory non-audit firm rotation on non-auditor independence. In the past few years the changes in laws and regulations regarding the field of auditing have caused a lot of discussion and this topic is researched extensively. Several papers (Francis, 2004; Kinney, Palmrose & Scholz, 2004; Quick & Warming-Rasmussen, 2009) have been published about the separation between audit and non-audit services on auditor independence in various countries like the United States, Germany, Ireland etcetera. But the question rises whether these results and conclusions can be generalized across countries. Furthermore the measures have not been implemented the same way and at the same time everywhere.
In the Netherlands for example the law of mandatory audit rotation will be effective from 2016 on and effects on auditor independence are not sure but can be estimated by looking at results in other countries which already have implemented new legislation regarding
mandatory audit firm rotation. Therefore this thesis can make a contribution to the existing literature because there is not much written about the combined impact of the separation between audit and non-audit services and mandatory audit rotation, the two measures that have been taken with the intention to improve auditor independence. Also most previous published literature reviews compare the results of the studies based on a positive or negative effect on independence, whereas here results from the US and Europe will be compared to each other to see if there are any differences in perceptions due to country specific rules and regulations or the timing of the implementation. The main findings of this literature review are that it is difficult to provide evidence for improved auditor independence in fact, but independence in appearance on the other hand is confirmed by several articles (Dart, 2011; Quick & Warming-Rasmussen, 2009). For mandatory rotation these findings are different. Mandatory rotation is not seen as a solution to improve auditor independence because of the high costs that this measure entails. Furthermore it is argued that an auditor will ensure that he is independent to protect his reputation in order not to lose his clients.
The remainder of this thesis is organized as follows. In the next paragraph some background information about the measures that have been taken in response to the frauds will be given. Furthermore the factors that determine independence will be discussed in more detail and the role of the auditor is described. Section 3 will outline the existing literature by reviewing the conclusions in different countries. This section is then followed by a discussion and finally in paragraph 5 the conclusion can be found.
2. Background Information
In the following paragraph several important concepts will be explained because they are the fundament for the following literature review. First the role of the auditor in society will be discussed. Afterwards the concept of independence will be described and at the end of the paragraph a description of various audit and non-audit services will be given as well as an introduction about mandatory audit firm rotation.
2.1 Role of the accountant
Before the role of the accountant can be discussed it should be clear what the reason of existence of the auditing profession is. Quick and Warning-Rasmussen describe the need for audit work to be done as follows: “Management performance is not completely observable and thus managers could be tempted to maximize their own benefit and not that of the shareholders. To reduce this problem management has to prepare financial statements. To increase trust in these statements they have to be audited” (2009, p. 142). According to Dart (2011) the role of the auditor in society is to provide reasonable assurance that the financial statements of the company that has been audited give a true and fair view about the financial performance in a certain period for the users of financial statements. Reasonable assurance indicates that it is not possible for the auditor to guarantee that there are no mistakes in the financial statements and that they give an hundred percent true view. Goldman and Barlev (1974) add to this description the requirement of consistency and the requirement of conformity with the accounting rules and principles. The reason that auditors are hired by companies is, according to Quick and Warming-Rasmussen (2009), that management has different objectives than investors which leads to the risk that management acts in his own interest rather than that of the shareholders. Financial statements are prepared to reduce this problem and by having these statements audited by an external third party the reliability should be increased.
“A distinguishing mark of the accountancy profession is its acceptance of the responsibility to act in the public interest” states the International Ethics Standards Board for Accountants (IESBA), which is an organization that sets ethical standards and that has developed a code for professional accountants (IFAC, 2013, p.11). This means that the auditor does not only has to comply with the desires of his client, the company by which the auditor is hired, but also has to acknowledge the interests of the shareholders, stakeholders and potential other investors, which are called the public all together by the IESBA. The fact that the auditor has to take into account several different groups of people can create a conflict of interest which is described by Goldman and Barlev (1974). There are three different conflicts of interest. The first one is a conflict regarding the auditor and the firm that is being audited. In this situation the shareholders and the management of the firm have the same preferences, namely that the report presents the company results in a positive way. For the management of the company this will lead to less criticism and for the shareholders this means that the share price of the company either remains the same or will increase. But the auditor is the one that has to
prepare the report according to the standards that are generally accepted in his profession. Therefore the firm could influence the auditor for example by the fee that it pays. The second conflict of interest is the one between management and shareholders. Without shareholders the company would not exist and therefore it is in the interest of management to satisfy the shareholders by presenting good results. Management could again try to influence the auditor to give a favorable opinion about the financial statements. The last conflict has to do with the self interest of the auditor and on the other hand compliance with the rules and standards of his profession. If the auditor agrees not to comply with the standards he could obtain a higher fee, but thereby increasing the risk of losing his reputation when this is discovered. If he decides not to violate the standards he could lose a client and his revenues will be reduced. The figure below schematically shows these three conflicts.
Figure 1: Schematic overview of conflicts of interest faced by the auditor.
2.2 What is independence
In the previous paragraph the role of the auditor in society and the potential conflicts of interest were described. To make sure the auditor can fulfill this role and adds value by improving the quality of the financial statements, it is important that he remains independent. DeFond, Raghunandan and Subramanyam define auditor independence as: “the probability that the auditor will report a discovered breach in the financial reports” (2002, p.1250). Goldman and Barlev (1974) characterize this trait as the cornerstone of the accounting
profession. During the last decade the independence of the auditor has become more and more important, because of the changing environment. Since several scandals have put the
accounting profession in a negative light the debate around independence has flared up. In a reaction to this ongoing discussion the IESBA has developed a code in which several fundamental principles of independence are discussed. These principles are all part of the ‘independence in fact’ which is one of the two categories in which independence can be divided according to the AICPA (Beattie et al, 1999). The IFAC describes the two forms of independence as independence in mind and independence in appearance. The former category of independence defined by the IFAC corresponds to the independence in fact. This means that there are just two forms of independence, but in the literature the terms independence in fact and independence in mind are used interchangeably. In the following paragraphs the term independence in fact will be used.Both organizations define the independence in fact or in mind as a mental attitude of the auditor that enables him to give an opinion about the financial statements without being influenced by others and to work with integrity, objectivity and professional skepticism (Quick & Warming-Rasmussen, 2005). Independence in appearance on the other hand has to do with the view of observers of the auditor rather than the auditor himself. When the observer, which could be an investor or other stakeholder, feels that the auditor-client relationship does not lead to a conflict of interest, the independence in appearance is not impaired (Beattie et al., 1999).
In its ‘Code of Ethics for Professional Accountants’ the IESBA outlines several principles of auditor independence in fact. These are integrity, objectivity, professional competence and due care, confidentiality and professional behavior. These principles will now be discussed in further detail. The principle of integrity implies that professional accountants must be honest and unambiguous, not only to their clients but also to the public. This means that when an auditor knows that the financial statements contain material misstatements he should not give an unqualified opinion. The second principle, objectivity, charges the auditor with the
responsibility to remain free from biases and avoid conflicts of interest. This implies that when someone, for example the management of the audited company, tries to influence the opinion of the auditor he should withstand this pressure. With the term professional
competence and due care the IESBA means that the auditor should make sure that his knowledge of developments in his profession is kept to a level which assures he can provide his services to his clients on a high level. It is not only important here that the auditor takes
part in training to obtain and maintain the required skills and knowledge, but he also has to make sure that he acts in accordance with the technical and professional standards.
In performing an audit the auditor usually obtains specific client information. Taking care of this information is part of the principle of confidentiality. Unless there is a legal obligation to do so the auditor should not provide third parties with client information, nor should this information be used in the auditor’s own advantage. This could for example be the case when a relative of the auditor wants to invest in the company. The code of ethics prohibits the auditor from providing such investment information to others. The last principle in the handbook of the IESBA is professional behavior. This more general principle requires that auditors should take the reputation of their profession in mind, besides this they are required to adhere to the laws and regulations in accounting.
When the auditor is complying to the previous described principles his independence is insured, but when these principles are violated several threats exist. The threats that are described by the European Commission and the IFAC are self-interest threat, self-review threat, advocacy threat, familiarity or trust threat and intimidation threat (Quick and
Warming-Rasmussen, 2005). To start with the self-interest threat which can occur when the auditor provides non-audit services to client for a large percentage of its total client fees. Another situation in which such a threat could exist is when an auditor has financial interest in the client, for example when he or one of his relatives is a shareholder of the company (IFAC, 2013). The self-review threat can also occur when the auditor provides non-audit services to a client: in this case the auditor might be unwilling to detect and correct mistakes that result from advisory services (Quick and Warming-Rasmussen, 2005). The advocacy threat may exist when an audit firm is helping the client firm in resolving a litigation or dispute
(European Commission, 2004). The objectivity of the auditor is also impaired by this threat when shares of the client company are promoted (IFAC, 2013). The familiarity threat occurs when the audit firm maintains a close or long relationship with the client. As a result of a long relationship the auditor could be so familiar with the practices of the client that he is not willing to see the mistakes anymore. The IESBA also mentions accepting gifts from the client as an example of this threat. The last threat, the intimidation threat, has to do with pressure on the auditor. This pressure can be actual or perceived. Intimidation could for example take place when the client firm tells the auditor that they will end the agreement when the opinion is not positive (IFAC, 2013).
2.3 Separation between audit and non-audit services
The previous paragraph described that both the self-review threat and the self-interest threat can be caused by providing non-audit services. Quick and Warming-Rasmussen (2009) also mention the provision of non-audit services as one of the four typical reasons for decrease auditor independence, next to having personal relationships, financial interest and personal interaction in or with the client. These services were for a long time not seen as threat to independence until changes in the amount and nature of non-audit services took place. Canning and Gwilliam (1999, p.402) state: “audit firms are adopting new audit approaches which are seen to erode the boundary between audit and consulting to the extent that
‘auditing has consequently become consulting”, besides they also stress the scale of the non-audit services. Due to this changes the composition of the revenues of non-audit firms changed substantially in the period from the 1980’s to 2000 (Citron, 2003; Ianniello, 2012). At the beginning of this period revenues consisted mainly of audit fees but in the course of time the amount of non-audit fees increased and finally became higher than the audit fees, the core business according to Francis (2004). After several frauds in which high ratios of non-audit services were found, measures were taken to separate audit and non-audit services. The next section will outline both categories of these services in more detail.
2.4 Non- audit services
In the existing accounting literature not only the term non-audit services is used. When authors talk about management advisory services (MAS), consultancy services or advisory services they are also describing the same topic. As mentioned before the ratio of audit and non-audit services changed during the past decades. Citron (2003) and Ianniello (2012) both state some figures about the increased scale of non-audit fees. According to the former research results from England at three of the big accounting firms, Deloitte, KMPG and EY, show an increase of 40 percent from 1990 to 1995. The next five years this increase even is 130 percent. Ianniello (2012) has found that the percentage of audit fees was 70 percent of total revenues in 1976, but only 31 percent in 1998, a reduction of 39 percent over almost twenty years. Even though the providence of non-audit services could lead to various threats, as discussed before, and several organizations are set up in order to create rules and
regulations to make sure auditor independence is not decreased by providing these services, there are also positive aspects of the providence of non-audit services. For example a decrease in the client’s consulting risk is mentioned as well as a reduction in costs because the auditor
knows his client better (Quick & Warming-Rasmussen, 2005). This increased knowledge is also named by Canning and Gwilliam (1999) as a beneficial factor of providing non-audit services. Besides they argue that the threat of a company switch is reduced because the audited firm is more dependent on his auditor. Even though these and several other advantages seem to exist there are also various perceived disadvantages such as more
dependence of the auditor on the client leading to a self-interest threat and more unwillingness from the auditor to find and correct errors in the statements of its client (Ruddock, Taylor & Taylor, 2006). To protect auditor independence the SOx act was introduced in the US. This act, of which the official name was 'Public Company Accounting Reform and Investor Protection Act', was the combined product of the work of senator Paul Sarbanes and Michael Oxley. It prohibits several non-audit services such as bookkeeping, financial information systems and design, actuarial, legal and internal audit services (Quick &
Warming-Rasmussen, 2005). Kinney et al. (2004) add to this list the appraisal and valuation services, human resources, management functions and broker-dealer services.
2.5 Mandatory rotation
The debate on mandatory rotation has been going on for forty years (Johnson, Khurana & Reynolds, 2002). Within this period of time there were times when the debate was more severe such as the period surrounding the corporate frauds and the implementation of the SOx. Section 203 of this act requires audit partners to rotate after every five years of
providing audit services to a client (Gul, Jaggi & Krishnan, 2007). The reason for the flare up of this debate in recent years were, according to Davis, Soo and Trompeter (2009), two changes that led to decreased quality of financial reporting which concerned the SEC. The first change related to increased earnings management by companies, whereas a reduction in auditor independence formed the second change.
As mentioned earlier the SOx act requires audit partners to switch after five years, but Kaplan and Mauldin (2008) note that this does not apply to audit firms. Pott, Mock and Watrin (2009) emphasize that there is a difference between these two sorts of rotation. When only the audit partner is rotated after a certain period of years the audit firm will still able to obtain specific client knowledge and the so-called learning curve has effects on audit quality in this system. When mandatory audit firm rotation is required instead the firm has less possibilities to make use of these effects. Chi and Huang (2005) also stress the fact that when studies use partner rotation and firm rotation interchangeable measurement errors exist. There are not many
countries that make use of the last system in which the audit firm is rotated. Jackson et al. (2008) name Italy, which has a nine year rotation period, and Brazil, which has a shorter rotation period of five years. India, Singapore and South-Korea are added to this list by Pott et al. (2009). Countries where only audit partner rotation is mandatory include the US and Australia (Jackson et al., 2008). When discussing mandatory rotation a term that is often used in accounting literature is audit firm tenure. Johnson et al. define audit firm tenure as: “the length of the audit-firm–client relationship as of the fiscal year-end covered by the audited financial statements” (2002, p. 645).
As described in the paragraph about independence before, one of the threats to auditor independence is the familiarity threat that was explained by the IESBA as the situation in which the auditor has a too long or too close relationship with his clients which impairs his objectivity (IFAC, 2013). This is just one of the arguments put forward by the proponents of mandatory rotation, but in the accounting literature more of these arguments are stated. First of all Gul et al. (2007) bring to the fore that as audit tenure increases the economic bonding with the client also increases and due to this personal connections auditor independence could be decreased. Another advantage proposed by Jackson et al. (2008) is the enhanced
competition on the audit market. They argue that when a company is looking for a new auditor audit firms will have to compete more with each other to obtain new clients. A consequence of this increased competition is that audit firms will improve their services which eventually leads to improvements in audit quality. Kaplan and Mauldin (2008) explain clearly what happens when mandatory audit firm rotation is enforced. According to them the present value of the audit-client relationship is worth less than when audit rotation is not mandatory which reduces the incentive of the auditor to act less independent because losing the client is less costly. In this way auditor independence could be improved by mandatory audit firm rotation.
Even though some authors come up with the theory of increased independence as described above there are also many opponents of the mandatory audit rotation system. Myers, Myers and Omers (2003) expect that there would be more audit failures as a result of less knowledge about the client in the first years that the auditor has a new client. These failures are
accompanied by higher audit costs, an argument that is proposed by several auditors as well as by the accounting profession (Gosh & Moon, 2005; Jackson et al., 2008; Myers et al., 2003). Both Ruiz-Barbadillo et al (2009), Davis et al (2009) and Kaplan and Mauldin (2008)
suggest that the audit market provides incentives that are strong enough to ensure the
independence of the auditor and thereby making mandatory audit rotation unnecessary. These market based incentives are related to the protection of their reputation in order to prevent them from losing future business and also prevent them to legal liability.
3 Literature review
In the next section an analysis will be provided about various papers from the past few years, the majority are from the period after de collapse of Enron. Because the goal of this thesis is to provide an answer on a research question that includes two separate regulations, mandatory audit firm rotation and prohibited non-audit services, this section will be split up in two parts. The first part considers the impact of prohibited non-audit services, in which the results are broken down by a part about the U.S and another part with results from Europe. This separation has been made to investigate if there are any differences in results between the U.S., where the frauds started and the legislation was developed and implemented first, and Europe which followed later in taking measures around this topic. In the second part the focus is on mandatory rotation where results are divided in two categories: the ones that have found a positive impact on auditor independence and the ones that conclude that the opposite effect is true.
3.1 Analysis of the impact of non-audit services on independence
3.1.1 Evidence from the US
A considerable amount of literature on the impact of the separation between audit and non-audit services comes from the US, which could be the case because several frauds have taken place here and the SOx legislation was implemented in this country before other countries have taken measures to ensure the independence of the auditor and restore investor
confidence. Kinney, Palmrose and Scholz (2004) have researched if there exists a relationship between restatements of financial statements that are audited and the fees a company has received for non-audit services. Based on the presumption that there is a positive relation between providing non-audit services and restatements more than 600 US companies who made restatements were linked to a company in the same industry with comparable revenues and the same audit firm, that did not make restatements to see whether or not this relationship really exists. The non-audit services are categorized as financial information system design
and implementation (FISDI) services and internal audits, tax services, audit related services and unspecified services. According to Kinney et al. (2004) the usefulness of investigating this relationship is that the debate around these services focuses on the dependence of the auditor on their clients which could lead to giving unqualified opinions about the financial statements when the companies do not comply with all the prescribed rules, leading to lower audit quality, which ultimately leads to restatements. The results they obtain are not always consistent with the presumption that restatements are positively correlated with providing non-audit services. In the case of tax-services they even conclude that there is a negative relation, but at the time of this research tax services are not a category of prohibited non-audit services. The negative relation implies that the provision of these specific category of non-audit services could improve the reliability of the financial statements because when there need to be made less restatements if the auditor also provides tax services this means that the overall quality of financial statements has improved. The argument that supports this finding is that when an audit firm provides more services to its clients, it will obtain more knowledge about the client and his specific business risks and practices which can lead to a better audit. On the other hand the experiment also shows a negative relationship when it comes to the FISDI and internal services. The last category, which contains audit related services and unspecified non-audit services, is the only category of services where a positive relationship between restatements and non-audit services is found. Due to the division in only four broader categories and the last of these categories being made up of what the authors define as
‘unspecified non-audit services’ it is difficult to draw conclusions about the outcomes of this study. Furthermore the proxy that Kinney et al. have used to research whether or not non-audit services are a threat to independence only gathers indirect evidence.
Whereas Kinney et al. conclude that audit quality can be reduced when tax services would be prohibited, Francis (2004) suggests the opposite, namely that audit quality could be increased by prohibiting audit firms from providing all non-audit services. While the former argues that the enhanced quality of financial reporting, because of more knowledge about the firm that is being audited, offsets the problem of economic dependence on a client, the latter does not share this vision. Francis states that even when different researches obtain different outcomes and have different opinions about whether providing non-audit services has a negative of positive impact on auditor independence in fact “there will always be at least the suspicion that auditors are not completely objective if they provide significant other services that put the auditor in a commercial business relationship with the client” (2004, p.363). With this
sentence he wants to make clear that even when results do not show that independence in fact is decreased, independence in appearance can still be decreased. Which makes his conclusion valuable is that Francis’ research is based upon literature from the US over a long time period of 25 years. Furthermore he argues that audit firms should focus on their core business which, according to him, is the attestation of financial statements.
A possible explanation for the discrepancy between the outcomes of both studies lies in the period they are looking at. The experiment of Kinney et al (2004) was based upon companies that restated their earnings in the five year period from 1995 to 2000, which is the period before the large corporate frauds took place and SOx legislation was implemented, whereas Francis (2004) is looking at more recent studies done in the US. The latter also writes about the change in the position of non-audit services as a possible explanation of changed
perspectives on auditor independence. For several years these services where not seen as a problem to independence because audit firms should still be independent to remain their reputation which has underlying economic motives. Besides Francis (2004) has found that the percentage of audit fees per client is less than 0,1% of the total revenues which made him conclude that auditors should focus more on their core business.
The authors from the third and still more recent US-based study did make use of audit committees to examine whether they believed that auditor independence was reduced by the provision of non-audit services. The role of the audit committee is described by Gaynor, McDaniel and Neal (2006) as an organ being responsible for the approval of providing a non-audit service. In deciding about whether or not to approve such a service the committee must consider the tradeoff between auditor independence and audit quality. The committee
represents the interests of investors and should therefore decide if the benefits of increased audit quality outweigh the reduced perceived auditor independence. This makes it useful to take a look at the decisions of the audit committee. Even though the evidence for this survey has a more indirect character, because the audit committee perceives that the confidence of investors in auditor independence is reduced, the results are consistent with those of Francis in the way that auditor independence is negatively affected by providing non-audit services. More specifically research of Gaynor et al. (2006) shows that the committees were more conservative in allowing non-audit services to be provided, even when it was known that the quality of the audit increased by a specific non-audit service.
3.1.2 Evidence from Europe
Within this section results from various European countries like the UK, Germany, Denmark and Italy will be compared to see whether there are differences in the way these authors perceived auditor independence has improved after the separation between audit and non-audit services.
Just like the US Germany has a very strong economic position. Even though the US is still the number one when it comes to global trading of goods and services, Germany follows on the second place. According to Quick and Warming-Rasmussen (2009) they are even at a third place when it comes to the strength of their national economy. The country also has a central position within the European region because of the high amount of inhabitants and gross domestic product. This, and the assertion made by the authors that the German audit context has a lot of similarities with those of other European countries, makes it interesting to take a look at the findings that Quick and Warming-Rasmussen obtained from this country. In addition, the fact that it this is a more recently published study than the previous discussed articles also makes the research of Quick and Warming-Rasmussen valuable.
Whereas Kinney et al. and Francis make use of the broader concept of audit quality, Quick and Warming-Rasmussen investigate the relationship between non-audit services and auditor independence based on four different hypotheses. The advantage that Quick and Warming-Rasmussen have because they research more than one hypothesis is that this enables them to not only draw conclusions about whether or not non-audit services have a negative impact on auditor independence but the research findings also tell something about the possible
solutions to provide non-audit services without having a negative impact on independence. Besides the positive or negative effect of providing non-audit services they take a closer look at nineteen different services an audit firm could provide to its clients. An advantage of looking to the individual services is that it provides a more detailed view than the study of Kinney et al. because in this study services are grouped into only four different categories. One of the remarkable differences between the German experiment and the study by Kinney et al. is that the latter have found that so called FISDI services are a threat to auditor
independence whereas Quick and Warming-Rasmussen have found that accounting
information systems services were not considered as problematic by German investors, even though these kind of services are currently prohibited by the German law. Only one other
category was found that did not have a negative impact on auditor independence. The majority of the nineteen services that companies could provide to their clients were seen as problematic, while some of these are not prohibited, such as human resources. So the overall conclusion of this study is in line with Francis (2004): in general non-audit services reduce auditor independence. Another hypothesis that was tested here was whether investors believed that independence could be increased by the separation of auditing and consulting staff. German investors did show that this is a good solution to let audit firms provide non-audit services without decreasing independence too much.
Compared to the US study of Kinney et al. the results that are obtained by the German study are quite different in the way that the latter do perceive that auditor independence is decreased by providing most non-audit services while the former considers that the providence of
several of these services, such as tax advice, do not change auditor independence in a negative way. The reason for these different outcomes may be caused by the choice of the subjects used in both studies. Quick and Warming-Rasmussen have drawn their conclusions based on the perceptions of German investors whereas Kinney et al. used US companies. The interests of investors in the financial statements of a company differ from the use of the financial statements by the company itself. The financial statements are a main source of information for investors when making decisions about whether or not to invest in a company. Therefore investors may have different perceptions about auditor independence than the company itself. Another study that does also make use of investors to examine whether auditor independence is impaired by providing both audit as well as non-audit services to clients is the study of Eleanor Dart (2011). She argues that the reason that she uses investors as her respondents in her questionnaire-based experiment is because they are both the main users of the financial statements but also the actual owners of the company. For them it is highly important that they can trust the information that is presented in the financial statement to make judgments about company performance and investments decisions. The most important requirement to ensure this confidence is that the statements are audited by an independent auditor.
According to Dart (2011) there are three factors that could reduce auditor independence. These are the length of the relationship with the client, the combined provision of audit as well as non-audit services to that client and the economic dependence on the client. Based on the requirements in the ‘Ethical Standards for Auditors’, which is the English version of the US SOx that was implemented in 2004 by the Auditing Practices board (APB), she conducted
a research in which the investors were shown different statements about non-audit services. One of the these statements was whether non-audit services should be prohibited when they were provided by auditors that were also responsible for the financial statement audit of that company. Almost 60% of the respondents agreed to this statement but when employees other than auditors of the same company would provide non-audit services more than 80%
indicated that this would not cause any problems regarding their perceptions of auditor independence.
Besides looking at the broader question whether or not non-audit services should be banned separate categories of non-audit services were investigated by Dart (2011). Even though both the German study and the UK study use the same subject, investors, for their experiments the outcomes do differ in certain aspects. In contrast to findings of Quick and
Warming-Rasmussen (2009) the investors in the study by Dart (2011) consider human resources services as not to be problematic to auditor independence and also the UK investors were not concerned about providing expert and legal services as well as tax services and information systems design and implementation. The last two categories were also negatively related with restatements according to Kinney et al. (2004). However, German investors do consider these last two categories as a threat to auditor independence. The overall conclusion of both studies is the same: when a company provides non-audit services to their client company this is a threat to auditor independence, only the perceptions about different categories of non-audit services that do harm auditor independence are different.
While both of the previous studies that have been discussed prove that the joint provision of audit and non-audit services by the same company have a negative impact on auditor
independence, another UK study by Beattie, Brandt and Fearnly (1999) shows that there are also some positive aspects of the combined provision of audit and non-audit services. The authors argue that both auditors and their clients benefit when both services are provided because the auditor has more knowledge about his client. Beattie et al (1999) do admit that this is only the case for independence in fact and that perceived independence could be decreased. It can be questioned whether this outcome is still accurate nowadays because this article was published in 1999 when the corporate frauds did not take place yet and there was no legislation for the providence of non-audit services. But on the other hand the position of non-audit services has changed in this period already according to Citron (2003) and Ianniello (2012). To investigate whether perceived independence did indeed decline several financial
directors and audit partners were surveyed in this study. Even though this study was carried out in an earlier period in the UK the results are consistent with those of Dart (2011): the main threats to auditor independence come from economic dependence on the client and from the joint provision of audit and non-audit services.
Not only in the US large corporate frauds existed, Europe also has some examples. One of them was the Italian company Parmalat which filed for bankruptcy almost a year after Enron, in 2003. This European issue which can also be compared in size with the Enron case makes it interesting to involve the results from an Italian study in this research. Ianniello (2012) investigates if there is a relationship between giving an unqualified opinion by the auditor and the providence of non-audit services. First, the average percentage of non-audit fees at the more than two hundred Italian companies in this study was researched and results showed this was 24 percent, an acceptable level according to the investors in the German experiment of Quick and Warming-Rasmussen (2009). Ianniello (2012) expected to find a positive relation between the amount of non-audit services that the auditor provides to a client and the number of times that an unqualified opinion is given. An explanation for this positive relation could be that when non-audit services indeed impair auditor independence the auditor is aware of misstatements in the financial statements but is unwilling to report them because of the impact on the client relationship. This assumption can be compared to the assumption made by Kinney et al who presume that a positive relationship exists between providing non-audit services and financial statement restatements.
Eventually, this positive link is indeed found but the author stresses that these results cannot be interpreted as an indication that auditor independence is indeed reduced. The explanation Ianniello (2012) gives for this caution is because in the case that only the two possible opinions unqualified and qualified were studied then no connection could be discovered. But when a third opinion was added to in the regression model, an unqualified opinion with an emphasis of matter opinion, positive connection was retrieved.
With the UK, Italy and Germany the results from several relatively large countries within Europe have been discussed, but are these findings comparable with those from the smaller European countries such as Ireland and Denmark? The same authors that studied the perceptions of German investors did also investigate several financial statement interest groups within Denmark. Even though Denmark is one of the smaller countries studied in this
literature review, the research that Quick and Warming-Rasmussen carried out here was larger than studies of Dart (2011) in England and of Quick and Warming-Rasmussen (2009) in Germany. Compared to these last studies which had only about hundred respondents, in the Danish study data of more than 600 respondents is included which makes that results obtained more reliable.
Whereas Quick and Warming-Rasmussen (2005) go in depth by focusing on separate sorts of non-audit services in the German study, various hypotheses were tested in the Danish
experiment. Instead of using only investors they used five groups of people with different interests in financial statements to figure out whether different groups have different perceptions about the way auditor independence is influenced by providing non-audit
services. Furthermore it was investigated if this is true for several categories of these services and whether the respondents think it will help to provide non-audit services and audit services by separate divisions in a company. In contrast to both the results of Dart (2011) and the findings of Quick and Warming-Rasmussen (2009) in Germany all categories of non-audit services are perceived to impair auditor independence. One remarkable outcome of the questionnaire was that a large percentage, more than 85%, of the state authorized auditors did not perceive that auditor independence was decreased by providing non-audit services. Unlike the conclusions found in Dart’s article, Quick and Warming-Rasmussen (2005) state that only one of the researched groups, business journalist, believed auditor independence increased when a company used different employees to provide non-audit services to an audit client. As the potential cause for the discrepancies in the results between these studies Quick and
Warming-Rasmussen (2005) state that the Danish audit market differs in several aspects from for example the German or English audit market. One of the main explanations they give is the size of the marketplace. In a smaller market an auditor has to be more independent to protect its reputation because when there are less clients in the market this reputation is even more important. On the other hand the smaller amount of customers could also be a threat: the auditor becomes more dependent on a client and might also have a longer relationship with the client company.
Another aspect which makes the Danish market different, according to Quick and Warming-Rasmussen, is because of cultural norms and values. To explain this statement they use the theory of Hofstede that explains differences in culture based on the masculinity-femininity dimension. Scandinavian countries as well as the Netherlands are more feminine which means
they make more use of negotiation in contrast to the US, UK and Germany which are more masculine “meaning that they find it preferable to fight out conflicts with the mentality ‘let the best man win’” (Quick and Warming Rasmussen, 2005, p. 145). This way of acting in
business situations could also have an influence on the perceptions of respondents auditor independence because when it is more common to make compromises and negotiate with business partners the relation between auditor and client could be influenced. Therefore the results from the Danish study can make contributions to the picture of how the provision of non-audit services is perceived to impact auditor independence across different countries.
Another country that is also known to have a smaller audit market is Ireland. Canning and Gwilliam (1999) studied several financial statement interest groups here. Even though this research was done in an earlier time period, the Irish lenders, investment managers and financial analysts generally had the same vision as the Danish respondents about auditor independence. Research results showed that they also perceived a lower level of independence of the auditor when non-audit services were provided to the same client. Furthermore they stated that respondents felt that the small size of the Irish audit market contributed to auditor independence in a positive way. Their explanation for this perceived independence was, just like in the experiment of Quick and Warming-Rasmussen (2005), the enhanced prudence at audit firms to protect their good reputation.
Compared to other studies this study is somewhat different in the way that Canning and Gwilliam (1999) also investigated the possible solutions to let audit firms provide non-audit services without decreasing the independence from the viewpoint of the financial statement users. Several possible solutions came to the fore here amongst which a complete ban on non-audit services for smaller companies, allowing non-audit firms to provide non-non-audit services but only to non-audit clients, separation of non-audit employees and audit employees and a permission for all non-audit services when they are fully disclosed. Different from the conclusion that Francis (2004) draws for the US, the results that Canning and Gwilliam (1999) obtain clearly indicate that a total ban on non-audit services is not desirable. Permission to provide all non-audit services under full disclosure was also not seen as the solution to increased auditor independence by the respondents. Instead of these two alternatives they indicated that a separation in the employees of the audit firm was most helpful in assuring auditor independence, which also corresponds to the solution
recommended by Dart (2011). The question is whether these results are still valid because in 22
1999, when the article of Canning and Gwilliam was published, the perspective on auditors could be different because no large scandals did take place at that time.
3.2 Analysis of the impact of mandatory audit rotation
3.2.1 Mandatory rotation has a negative effect on auditor independence
Just like with the previous topic on the impact of non-audit services on auditor independence a lot of research into this subject has been done in the US. In this country the SOx legislation requires that after five years audit partners should rotate. Another point of similarity with the provision of non-audit services is the separation between independence in fact and
independence in appearance. One of the studies that is looking at independence in fact based on a sample with data from the US is carried out by Myers, Myers and Omers (2003). Consistent with other studies in this area Myers et al. used accruals as a proxy to draw conclusions about audit quality which enables them to say something about the impact on auditor independence . The relation between audit quality and auditor independence was described by Pott, Mock and Watrin as “the joint probability that an auditor will both detect and report material misstatements” (2009, p.234). Here, the detection probability depends
upon the expertise of the auditor and the probability of the willingness of reporting is related to the independence of the auditor. Audit quality can in its turn be determined by the amount of accruals used, according to Myers et al. (2003). They state that when auditors are willing to stand up against the usual practices of management, which can be reflected in abnormal accruals, and correct them, audit quality will be improved. Even though the authors assumed that there was a negative relation between the length of the relationship between the auditor and his client and his independence this assumption was not confirmed after the research was carried out. In contrast to the assumption results showed that longer tenure did not have a negative effect on audit quality. On the other hand Myers et al. could not provide enough evidence to prove that audit quality would increase when audit rotation became mandatory. The article of Myers et al. makes an important contribution to the debate on this subject by the scope of the investigation. Over a period of 13 years there are data of more than 3000 companies used, which makes the results of this study more valid than those of smaller studies.
A study that corresponds to the one of Myers et al. is the one carried out by Johnson, Khurana and Reynolds (2002). Not only did they use the same proxy to come to their conclusions, they
also used data from the same county, in the same time period. Besides independence in fact is researched here as well, but instead of looking directly at audit quality they investigated financial reporting quality, which could be seen as the product of audit quality because when audit quality is higher, more errors will be found and corrected and therefore financial reporting quality will be increased. Even though Johnson et al. made a different cutoff in the lengths of tenure of companies in their sample, they obtained quite similar results. Whereas Myers et al (2003) used a five year period to differentiate between long tenures and short tenures, Johnson et al (2002) used a small category of two to three years, a medium category in which four to eight year tenures were placed and the last category existed of companies who had a relation of longer than nine years with the same auditor. Consistent with the results obtained by Myers et al. the category of long tenures did not lead to a decreased level of financial reporting quality. The results of short tenures on the other hand indicated a decrease in financial reporting quality (Johnson et al., 2002). One of the possible explanations for this outcome is the familiarity of the auditor with the client: in the earlier years of the relationship the auditor is less familiar with the clients systems and business.
Both articles stress the fact the US had a system of voluntary rotation at the time the experiments were done and that this system could lead to different results than a system in which rotation is mandatory because the behavior of the auditor could be different when he already knows the length of the relationship. Therefore the article of Ruiz-Barbadillo, Gomez-Aguilar and Carrera (2009) could give some new insights on the impact of mandatory audit rotation because in Spain, the country that is investigated by this authors, rotation was mandatory between 1998 and 1995. According to Ruiz-Barbadillo et al. (2009) the two
incentives an auditor has are the protection of the audit firm’s reputation and obtaining profits. Auditor independence is closely related to this incentives because based on his decisions the auditor could either lose his good reputation or lose the client. In an article about the
Australian audit market Jackson, Moldrich and Roebuck (2009) explain the relation between audit opinions and auditor dependence on a client. In their vision management is trying to avoid qualified opinions because these signal management failures and therefore giving an qualified opinion is a threat to terminate the agreement. By analyzing firms in the period of mandatory rotation and after this period no evidence of enhanced auditor independence was found by Ruiz-Barbadillo et al. (2009) when investigating different auditor opinions. Jackson et al. (2008) combine the proxy’s of all previous studies and come to the same conclusion as Myers et al. (2008) when examining accruals, namely that audit quality is unaffected by audit
tenure. Another outcome of this study revealed that the probability to issue a going-concern opinion had a positive relation with audit tenure and therefore increased the quality of the audit. This finding is consistent with both US studies which, just like Australia, only has a mandatory audit partner rotation system and no audit firm rotation system as in Spain.
Whereas all authors of the previous articles in this paragraph that did find a negative relation between auditor independence and mandatory rotation have researched independence in fact, some authors have as well studied the impact on independence in appearance. A study by Gosh and Moon (2005) investigates the perceptions of US investors and information
intermediaries. Their main finding is that for these interest groups the impact of mandatory rotation on audit quality is negative and would only lead to extra costs. Non-professional investors are participating in the experiment of Kaplan and Mauldin (2008) to investigate whether perceived independence differs between a system of mandatory audit partner rotation and firm rotation. The General Accounting Office (GAO) believes that besides the audit partner rotation, audit firm rotation should also be mandatory to increase auditor
independence but the investors do not display any preferences for partner or firm rotation.
3.2.2 Mandatory rotation has a positive effect on auditor independence
Consistent with previous studies of Myers et al. (2003) and Johnson et al. (2002) Chi and Huang (2005) use accruals to draw conclusions about how earnings quality is affected by audit tenure. Even though they also use the same time period of five years as Myers et al., Chi and Huang come to a different conclusion. According to them familiarity can contribute to audit quality until a certain level. When the years of tenure increase, the familiarity between the auditor and his client increases as well. Chi and Huang (2005) have found that until a tenure of five years the effects of familiarity are positive but after this period of time mandatory rotation would increase earnings quality. A possible explanation for this discrepancy between the studies of Chi and Huang and Myers et al. could be the countries they studied. The former have researched Taiwanese companies, whereas the latter obtained evidence from the US. Still another cause of the different results could be because Chi and Huang investigated audit partner tenure and Myers et al. looked into audit firm rotation. As explained earlier these two forms of rotation could have a different impact on audit quality because when only the partner is rotated the audit firm is still able to gain more specific client information which could improve audit quality.
A study that researches auditor independence in fact and is also looking at audit firm tenure, like Myers et al., but comes to a very different conclusion is that of Davis et al. (2009). Whereas both studies have the same proxy, abnormal accruals, and look at the same country, the US, they yield opposite results. The sample of Davis et al. (2009) consists of data from 1988 to 2006, which is a longer time period than Myers et al. who look at the years 1988 until 2000. Even though there is a difference of only six years, adding these years could make an important contribution to their research because Davis et al. (2009) take into consideration the years after the implementation of the SOx legislation. In fact their findings show that there is a difference between the period before and after the SOx became effective. Not only did they find evidence that the level of earnings management was higher in the earlier years of the auditor-client relationship like several other studies, they did also find an increased use of accruals leading to higher levels of earnings management when the length of the relationship increased. When the experiment was repeated for the post-SOx data no such relation was found. This implies that rotation of audit partners, which is mandated by the SOx, can have a positive effect on earnings quality. But the authors stress that the increased use of accruals could just as well be the consequence of the lack of client knowledge in the earlier years of tenure.
The concept of retention in combination with rotation is an aspect that has not been investigated in many studies but Dopuch, King and Schwartz (2001) argue that when
mandatory rotation is the only measurement taken to improve auditor independence the audit firm still faces the risk of losing its client earlier than after the maximum tenure period. When the client threatens the auditor to terminate the agreement the auditors independence could be decreased if the auditor gives a different opinion or decides not to report errors in order to keep the client. This could also explain why for example Davis et al. (2009) have found an increased level of accruals in the earlier years of the engagement. The auditor could have let management make more use of accruals in order to reduce the threat of losing the client. To make inferences about the effect on auditor independence in fact Dopuch et al. (2001) use four different settings in their experiment from no retention or rotation to combined rotation and retention. Finally they come to the conclusion that systems in which either rotation or combined rotation and retention is mandatory can improve auditor independence because there is evidence that under these regulations auditors are less willing to issue a report that is biased in favor of the management of the client firm in some way.
4. Discussion
Previous paragraphs give an oversight of the current literature around the debate of auditor independence. Two of the measures that have been taken to improve the independence of the auditor, mandatory rotation and a ban on various non-audit services, have been separately analyzed because not much is written about the combined of effect of these measures. At the beginning of this literature review, before the analysis of various articles, it was assumed that a positive relation between the ban on non-audit services and auditor independence exists. When one gives advice and recommends to do things in a certain manner and then has to review the consequences of that advice and recommendations his opinion will likely be positive. This is also what happens when the audit firm provides non-audit services, or consultancy services as these are sometimes called, to the client company: they give advice, design systems and then give an opinion about the financial statements that are produced by including this advice. Furthermore the relation between mandatory rotation and increased auditor independence also seems plausible since relationships often get closer over time and thereby objectivity could be reduced which could be a threat to independence.
In contrast to the first assumption evidence about reduced independence in fact regarding to the provision of non-audit services was difficult to obtain. Various authors have researched this category of independence without finding significant evidence.Kinney et al. (2004) have found no indications that preventing audit firms from providing non-audit services will improve audit quality, in fact they conclude that the ban on tax-services would even reduce it. Also Ianniello (2012) was not able to find enough evidence to state that auditor independence was reduced by giving more unqualified opinions when non-audit services were also
provided. Francis (2004) acknowledges that it is difficult to find evidence for reduced independence in fact but still advocates to impose a ban on non-audit services because of the perceived decrease of independence. This independence in appearance on the other hand was confirmed in various articles. Both groups of investors studied by Dart (2011) and Quick and Warming-Rasmussen (2009) have shown that their perceptions of auditor independence were reduced when the audit firm provides non-audit services to the client. Users of financial statements in smaller European countries such as Denmark en Ireland also had the same view. These results confirm that it is useful to have a ban on non-audit services because investors are the ones making use of financial statements the most and therefore it is highly important that at least in their view auditor independence is increased by this measure.
About the second measure, mandatory rotation, many articles were published that claim mandatory rotation will not improve auditor independence but will only lead to extra costs. Results obtained by Myers et al (2003) showed that longer tenure did not have a negative effect on audit quality, on the other hand there was no evidence that suggested that audit quality could improve when rotation was obligated. A study by Johnson et al. (2002) provided insights that are comparable with those of Myers et al. regarding the outcome that longer tenure does not decrease financial reporting quality, but in the earlier years of the auditor-client relationship they found a decrease in financial reporting quality.
5 Conclusion
Since various corporate frauds have taken place the profession of the auditor became subject of the discussion about independence. After various large corporate frauds, such as Enron in the US, Parmalat in Italy and Ahold in the Netherlands some extra measures have been taken to restore the public’s confidence in the auditor. In the US the Sarbanes-Oxley Act was enforced to regulate the accounting profession whereas other countries adopted their own regulations following this US regulation. Ever since these laws were enforced much is written about this topic, but there still is no clear answer to the question whether these measures have had the foreseen impact on auditor independence. Therefore this literature review is trying to give an answer on the question what the impact of prohibited non-audit services and
mandatory rotation is on auditor independence. Finding an answer on this question adds value to the existing literature by investigating the combined effect of both measures, whereas most previously published articles have researched them separately. Besides the way the
information is presented is different from previous literature on this topic. Because a per country base is used, conclusions can be drawn about the differences amongst the VS, which introduced new legislation first, and other European countries.
After analyzing various articles about the effect of the prohibition on non-audit services a clear answer on if the impact was positive can only be given for independence in appearance. Here investors and other users of financial statements clearly indicated that they perceived that the auditor was less independent when he provided non-audit services as well as audit services to its client. The positive effect on independence in fact however was not confirmed. Several authors tried to find evidence in this field, but they all came to the conclusion that it was hard to present significant evidence.
On the topic of mandatory rotation most evidence that is found indicates that this measure will not improve auditor independence because of high costs that are incurred by hiring a new auditor every few years. Furthermore opponents of this measure argue that the auditor will ensure himself to be independent because he has to protect his reputation in order to keep his business. This offsets the economic dependence argument of the proponents who argue that a longer relationship between auditor and client will have a negative impact on the objectivity of the auditor which makes him less willing to stand up against the practices of management.
6 References
Beattie, V., Brandt, R., & Fearnley, S. (1999). Perceptions of auditor independence: UK evidence. Journal of international accounting, auditing and taxation, 8(1), 67-107. Benston, G. J., & Hartgraves, A. L. (2002). Enron: what happened and what we can learn
from it. Journal of Accounting and Public Policy, 21(2), 105-127.
Canning, M., & Gwilliam, D. (1999). Non-audit services and auditor independence: some evidence from Ireland. European accounting review, 8(3), 401-419.
Chi, W., & Huang, H. (2005). Discretionary accruals, audit-firm tenure and audit-partner tenure: Empirical evidence from Taiwan. Journal of Contemporary Accounting & Economics, 1(1), 65-92.
Citron, D. B. (2003). The UK’s framework approach to auditor independence and the
commercialization of the accounting profession. Accounting, Auditing & Accountability Journal, 16(2), 244-274.
Dart, E. (2011). UK investors’ perceptions of auditor independence. The British Accounting Review, 43(3), 173-185.
Davis, L. R., Soo, B. S., & Trompeter, G. M. (2009). Auditor Tenure and the Ability to Meet or Beat Earnings Forecasts*. Contemporary Accounting Research, 26(2), 517-548. DeFond, M. L., Raghunandan, K., & Subramanyam, K. R. (2002). Do non–audit service fees
impair auditor independence? Evidence from going concern audit opinions. Journal of accounting research, 40(4), 1247-1274.
Dopuch, N., King, R. R., & Schwartz, R. (2001). An experimental investigation of retention and rotation requirements. Journal of Accounting Research, 39(1), 93-117.
Francis, J. R. (2004). What do we know about audit quality?. The British accounting review, 36(4), 345-368.
Gaynor, L. M., McDaniel, L. S., & Neal, T. L. (2006). The effects of joint provision and disclosure of nonaudit services on audit committee members' decisions and investors' preferences. The Accounting Review, 81(4), 873-879.
Ghosh, A., & Moon, D. (2005). Auditor tenure and perceptions of audit quality. The Accounting Review, 80(2), 585-612.
Goldman, A., & Barlev, B. (1974). The auditor-firm conflict of interests: Its implications for independence. Accounting Review, 707-718.
Gul, F. A., Jaggi, B. L., & Krishnan, G. V. (2007). Auditor independence: evidence on the joint effects of auditor tenure and nonaudit fees. Auditing: A Journal of Practice & Theory, 26(2), 117-142.
Healy, P. M., & Palepu, K. G. (2003). The fall of Enron. Journal of Economic Perspectives, 3-26.
Ianniello, G. (2012). Non‐Audit Services and Auditor Independence in the 2007 Italian Regulatory Environment. International Journal of Auditing, 16(2), 147-164. International Federation of Accountants. (2013). IFAC code of ethics for professional
accountants.
Jackson, A. B., Moldrich, M., & Roebuck, P. (2008). Mandatory audit firm rotation and audit quality. Managerial Auditing Journal, 23(5), 420-437.
Johnson, V. E., Khurana, I. K., & Reynolds, J. K. (2002). Audit‐Firm Tenure and the Quality of Financial Reports*. Contemporary accounting research, 19(4), 637-660.
Kaplan, S. E., & Mauldin, E. G. (2008). Auditor rotation and the appearance of independence: Evidence from non-professional investors. Journal of Accounting and Public Policy, 27(2), 177-192.
