In which ways does the mandatory audit rotation influence the
materiality and the key audit matters of the new auditor’s
report?
23 February 2018 Word count:11996
Student name: Melchior Hensen Student number: s1629514
University: Rijksuniversiteit Groningen Academic year: 2017-2018
Master: MSc Accountancy & Controlling Faculty: Economics and Business Course: Master's Thesis Accountancy Supervisor: prof. dr. De Waard RA MA
In this paper is being researched how the rotation of external auditors influences the materiality, number of key audit matters and motivation of these key audit matters in the new audit report for listed companies. This research finds empirical evidence that there is
a positive significant correlation between rotation and key audit matters. There is no correlation found between rotation and the motivation of key audit matters or between
Table of contents 1 Introduction ... 5 1.1 Motivation ... 5 1.2 Problem statement ... 7 1.3 Relevance ... 7 Scientific relevance ... 7 Social relevance ... 8
1.4 Structure of this report ... 8
2 Literature review ... 9
2.1 Introduction ... 9
2.2 The auditor’s report ... 9
2.3 Expectations gap... 10
2.4 Agency theory in auditing ... 11
2.5 Auditor independence and rotation ... 12
2.6 Measuring influence of auditor rotation on the New Audit Report ... 14
Materiality ... 14
Key audit matters... 15
Audit rotation ... 16
Audit fees ... 17
3 Methodology ... 19
3.1 Sample design and data collection ... 19
3.2 Model and variables ... 19
3.3 Dependent variable ... 20
Key audit matters... 20
Materiality ... 20
3.4 Independent variables ... 21 Rotate... 21 Base changes ... 21 3.5 Control variables ... 21 Audit fees ... 21 Size ... 21 Years ... 22 4 Empirical results ... 23 4.1 Dataset ... 23 4.2 Descriptive statistics ... 23
4.3 Testing for parametric assumptions ... 24
4.4 Results of regression analyses ... 25
4.5 Hypothesis testing ... 27
5 Discussion and Conclusion ... 29
5.1 Conclusion ... 29
5.2 Limitations... 30
5.3 Further research ... 31
1 Introduction
“External audit is promoted as a trust engendering technology to persuade the public that capitalist corporations and management are not corrupt and that companies and their directors are made accountable” (Power, 1999). After a series of audit scandals in the Netherlands (Econcern, DSB Bank, investment bank Van der Hoop and Imtech) the Dutch accounting regulator Autoriteit Financiële Markten (AFM) publicly stated that 45% of the examined audits were inadequate. This leads to questioning the trustworthiness of the extern audit by public opinion (AFM, 2014).
1.1 Motivation
“Audits should be a reassurance to all who have a financial interest in companies” (Committee on the Financial Aspects of Corporate Governance, 1992). But during the
financial crisis the reassurance of an audit represented a very limited meaning. It was striking that within months after many financial enterprises that received an unqualified audit opinion sought state support or even went bankrupt (Sikka, 2009). These occurrences have raised questions about the value of company audits, auditor independence and quality of audit work (Sikka, 2009).
In response, politicians and regulators have been looking to win back the trust of the external audit, which they see as important to the functioning of financial markets. They have come up with new laws and regulations for auditors. The European Commission published a Green Paper in 2010 which contains measures to enhance auditor’s independence and reliability to amend the quality of the audit. The main issue was to reduce the entrenchment of audit teams with clients.
In 2013, as a first step, the UK accounting regulator (the Financial Reporting Council) introduced a new audit report, this reform should increase the communicative value of the audit report. Stakeholders were dissatisfied with the informativeness of the old audit report (Blake et al. 2011; Amiram 2015) as it contained a limited amount of specific information to base an informed opinion on.
Most unqualified audit reports contained a boilerplate set of statements, which were essentially the same for all listed company’s (Lennox, 2005). These uniform audit reports barely include company-specific information (PCAOB, 2013), have only symbolic value and transfer little communicative value (Church et al., 2008). The auditor’s opinion is valued but should be more informative and relevant (ISAAB, 2015).
The new audit report should improve the dialogue between companies, auditors and the users of financial statements by introducing a qualitative dimension into the judgement of the auditor. The new audit report should lead to more transparent and accessible reporting of audit findings, more granular descriptions of risk and the disclosure of materiality. Investors deem the communicative value of the enhanced information a new audit report provides as high (FRC, 2016).
In 2013, the NBA (Nederlandse Beroepsorganisatie van Accountants) introduced the new audit report in the Netherlands as the second country in the world after the UK. At the end of 2014, the new report became mandatory for public interest entities (PIEs). The NBA stated that users of audit statements are not only interested in the auditor's conclusion but moreover in the information on key issues which appeared during the audit and the way in which the auditor performed his work (NBA, 2014).
In the new audit report the height of materiality and the determination thereof, are issues which get more attention and are comprehensively clarified. This should give the user of the new audit reports a better understanding of the interpretation of the facts and the realisation of the report.
After being introduced in The UK and the Netherlands the new audit report became
mandatory for firms of PIEs with financial years ending on or after December 15, 2016 (for countries that adopted the Independent Auditor’s Report on Financial Statements standards). The Independent Auditor’s Report on Financial Statements (ISAAB) has revised auditor reporting standards including International Standards on Auditing (ISAs) 700: Forming an Opinion and Reporting on Financial Statements and 701: Communicating Key Audit Matters in the Independent Auditor’s Report.
Another step, in order to influence and improve the quality of content of the audit report, was to create a (more) independent (professional/less familiar) relationship between the audit firm and the client. After the financial crisis and the audit scandals in the Netherlands, the Dutch politicians found the existing solutions to safeguard the auditor independence was not sufficient. The existing rule on partner rotation (changing the controlling auditor within the audit firm, every 7 years) was not adequate to ensure enough independence of the audit firm. A major change was that PIEs had to change from audit firm, which should have a significant effect on an independent client relation. The new legislation aims to avoid a too narrow (trusted) relationship between auditor and client (Roos, 2012). The European Commission assumes that mandatory rotation of audit firms will result in better and higher quality audits (European Commission, 2011).
This legislation came effective at a European level in June 2016 and regulated the mandatory rotation of audit every 10 years (Dijsselbloem, 2015). The Dutch government made a show of virtue to change the legislation in anticipation of the changing EU laws. The mandatory rotation period of 8 years, came into force on January 1, 2016 and was effective until the European legislation on 17 June 2016 became active. The accelerated implementation in the Netherlands is especially interesting for this research, because the rotation period was retroactively. The effect of the Dutch legislation was that companies of public interest were required to change auditing firm before the materialisation on January 1, 2016. As a result, PIEs in the Netherlands started to rotate audit firms a lot sooner than in the rest of the EU. Parallel to the developments in the Netherlands the UK also sped up the rotation
requirements, resulting in a higher level of audit firm rotations in the period 2013-2016 for the FTSE100 and FTSE250 companies.
These developments are in contrast with the EU regulations stipulating that audit
engagements with a tenure of more than 20 years can wait to rotate until 17 June 2020 and with a tenure of 11-20 years can wait to rotate until 17 June 2023.
The new regulations have led to more competition in the audit market because public interest companies have to put out new tenders at least every 10 years. A side effect was, that new clients take a heavy toll on audit firms; a first-year audit requires significantly more inspection hours than continuing audits (PwC, 2016).
1.2 Problem statement
The purpose of this paper is to research the impact of mandatory audit rotation on the new audit report. In particular it researches the influence on the reporting of materiality and the key audit matters. Will the change of auditor become visible in the new audit report by a modified materiality or changed number of key audit matters? For this research the audit reports are analysed of 329 listed companies listed on the FTSE 100, FTSE 250, AEX, AMX and ASsX in the years 2013, 2014, 2015 and 2016.
The goal of this research is translated to the following research question:
How does the rotation of external auditors influence the materiality, number of key audit matters and motivation of these key audit matters in the new audit report for listed companies?
1.3 Relevance Scientific relevance
This paper contributes to the existing literature on audit rotation and the new audit report. Due to the fact that the legislation is recently introduced, the majority of current research about the new audit report is experimental. Christensen et al. (2014), in their experimental research, find that the key audit matters paragraph in the new audit report change the investment decision of investors compared with the standard audit report. Doxey’s (2014) experimental research endorses this with his finding that additional disclosures are value-relevant for investors decisions. Sirois’ (2017) experiment illustrated that users of financial statements focus more to the key audit matters in the new audit report, therefore are less thoroughly on other parts.
International auditor rotation has been researched quite often, but research in the Dutch context is limited. This research adds empirical testing to the current body of knowledge on the effect of audit rotation. This is mainly possible due to the availability of data after the moment of auditor rotation in the Netherlands.
Social relevance
While substantial revisions to auditor and audit committee reporting are being discussed internationally, including the United States, the impact of these reforms on audit quality is still largely unknown.
The new audit report is first introduced in the UK and subsequently in the Netherlands and has been replicated. Since December 15, 2017 even the American accounting regulator Public Company Accounting Oversight Board (PCAOB) has adopted a new auditor’s report standard (PCAOB, 2017).
The UK and the USA have a lot in common on regulation and culture, this research of the impact of the ISAAB reporting requirements, is for the PCAOB a unique opportunity to determine what the impact will be in the US (Reid et al., 2017).
This research helps to show causal relations between audit rotation and predictors for audit quality and could help identify how specifically a rotation helps the transparency and informativeness required to regain the trust in the external audit.
1.4 Structure of this report
Section 2 consists of a literature review. Section 3 describes the methodology. Section 4 presents the empirical results. Section 5 concludes with a discussion of the findings, limitations and ideas for further research.
2
Literature review
In this chapter, an overview is given of relevant theory and insights of prior research in the effect of regulation on audit quality.
2.1 Introduction
At global level regulators and standard setters are looking for opportunities to strengthen the position of the auditor. Many of the issues we face today are as old as the profession itself. At the beginning of the 19th century the auditor independence, the audit report, the role and scope of the audit, competition between auditors and the lack of skills were a few of the primary problems that caused concern (Chandler & Edwards, 1996). This shows that this problem will not easily be solved permanently.
Nevertheless, every effort is being made to develop improvements and revolutionary changes are made. Ahead of the rest of the world the Netherlands and the UK have introduced new regulations to enhance the relevance of the auditor’s report and to improve the independence of the auditor. However, it remains unclear of the auditor rotation will improve the audit quality (Cameran, 2016). Especially considering the existing agency problem and the disadvantages of rotation (Lennox, 2014). What is the explanatory value of the literature about the improved independence of the auditor?
2.2 The auditor’s report
The auditor’s report is the one and only proof of a performed audit. But how the audit was produced, and the working of the associated processes and organisational principles is a big unknown (Sikka, 2009). The usual auditor’s report consisted of a three-paragraph statement that was almost identical for the large majority of public enterprises (Blake et al., 2011). Originally the auditor’s report was based on a standardized approach, with as major benefit the comparability between enterprises. The auditing profession took the view that the standard approach was preferred above an entity-specific approach (Simnett & Huggins, 2014). The concept behind the standardized approach, is the principle that management is accountable for the financial statements, and the auditor verifies whether a company’s financial statements present a “true and fair” view in accordance with the accounting standards, without adding their findings or insights (IAS 700).
In the past, audits used to offer only a binary opinion. The method of reporting was brief and uninformative: “Audited and certified”, “Audited and found correct” or simply “Audited” was standard practice in 19th century. (Monkhouse, 1890, in Chandler & Edwards, 1996). These methods were insufficient for the public interest and led to discussions. Even in that time frame, there was not only a demand for a detailed auditor’s report including a specification of the audit but also on entity-specific information on the performance and situation of the company (Robertson 1897, in Chandler & Edwards, 1996).
More than a century later, the NBA noted that Dutch stakeholders, politicians and also the accounting profession itself, demanded a more extensive auditor's report (NBA, 2014). The users of financial reports found the audit report not particularly useful, nor informative (ISAAB, 2011). The auditor’s report needed a qualitative explanation, a clear explanation of the procedure followed and of the findings (FRC, 2013).
In 2013, the FRC introduced the new auditor’s report which focuses on improving the communication value by presenting more of the acquired content during the audit (FRC, 2013a). The final aim is to achieve a more relevant auditor’s report for the users. The structure and content were changed to achieve this goal.
The new standard concerning the new audit report for Dutch auditor is described in NVCos 702N as follows:
The auditor's objectives are1:
a. to structure the statement in such a way that the most important issues are mentioned first; b. to provide more insight in the process, by mentioning the key audit matters, the materiality and scope of a group check in addition to the requirements regarding the audit statement as required by Standard 700;
c. to make the audit report entity-specific.
The new announced materiality and the key audit matters are particularly important for this research and are distinctive for the old report.
Simultaneously, the IAASB proposed that there would be some elements of the auditor’s report that would be commonly required in all auditors’ reports, while others could be tailored to meet national requirements (IAASB, 2012). This complies with the approach in paragraph 43 of IAS 700. The new audit report as proposed by the FRC and NBA seems to meet the information requirements of stakeholders, while barely infringing on the stance of the profession on the audit report, specified in IAS700.
2.3 Expectations gap
A review of the FRC shows that investor appreciated the enhanced value of the new audit report (FRC, 2016). The purpose of the extended auditor's report is to increase the
information value with transparent and accessible reporting of audit findings (FRC, 2016). This more relevant auditor’s report could reduce the expectation gap. There is an expectations gap between the users of the annual report and the auditors concerning the responsibilities of auditors (Humphrey, Moizer, & Turley, 1992). According to Porter it would be the reason of many lawsuits against accountants (Porter, 1993). The new audit report has led to more debate
in the UK between the audit committee, investors and auditors before publication of the annual report and has provided more confidence to the investors (PCAOB, 2014).
2.4 Agency theory in auditing
The agency theory clarifies the need and demand for an audit. Audits are obliged by law, dependent on the company’s size, assets, number of employees or legal structure. The principal-agent problem defined by Jensen and Meckling (1976) explains the economic reasoning for the existence of an audit. Listed companies exists by the virtue of an agency relationship. The principal-agent relationship contains a moral hazard problem that in turn creates agency costs.
There are two sides to agency problems. Firstly, the goals of the principal and the agent do not match. Second there is the risk of adverse selection, which makes it complicated and costly for the principal to verify what the agent is actually doing to achieve his goals (Eisenhardt, 1989).
Alignment of the interest between the shareholder (principal in this case) and management (agent) or a monitoring mechanism is necessary to reduce the information asymmetries and potential opportunistic behaviour of management.
This alignment can be achieved by realizing common incentives by making the agent a shareholder or give option bonuses. An independent audit of the financial statement reduces the identified agency cost by providing the reassurance about the reliability of the financial statements and therefore giving the principal a measure to determine if their goals are met (Dopuch & Simunic, 1982). This illustrated by figure 1.
Although an external audit is seen as a solution for the existing agency problems between owners and managers, the auditors themselves are also caught in an agency problem with their client (Audit Quality Forum, 2005). To a certain degree the auditor serves as an agent for the benefit of the principal. It creates the similar distrust as in the management-shareholder relationship and ensures questions about the auditor’s independence.
Figure 1: Shareholder-Management-Relationship including the Auditor (Messier, 2003)
Auditor independence from management is essential to deliver a reliable and trustful audit. Independence is difficult to reach considering that good partnership between the auditor and management is required for an effective audit. Appointing an audit committee contributes some distance between the auditor and management, but eventually the audit committee recruits and shareholders foot the bill for the audit. For auditors that is a trigger to act in favour of the management rather than in the favour of the shareholder or public interest. The rationale behind forcing auditor rotation is to limit the possible strong relationship between management and auditor that would serve management in maintaining its
information asymmetry from the shareholder. This would lower agency cost for shareholders, in terms of more reliable disclosure of financial statements. It is argued that the rotation itself represents an agency cost, as both the auditor and client will have to acquaint themselves with a new situation.
Adding entity-specific information and context to the auditor report also lowers possible information asymmetry for principals in this relation as they receive more information to interpret the financial statements.
The extension of the new audit statement can also be seen as an example of the application of information theory. Information theory explains the need for auditing from the information risk. The risk of inaccurate information, specifically by PIEs (Majoor & Kollenburg, 2011). The theory is based on the risk that the company provides unreliable information. A lower information risk results in a lower compensation for risk and more security.
2.5 Auditor independence and rotation
As stated in IAS700, for an assurance engagement a three-party relationship is required involving a professional accountant in public practice, a responsible party (management), and intended users (shareholder). Independence is an essential requirement that auditors can perform their professional judgment objectively without any interference (Tritschler, 2014). For listed companies it is important that the audit is performed by someone independent to reduce the incentive problem that exist because the agency problems (Watts & Zimmerman, 1983).
Audit quality is difficult to define, after more than 20 years of research there is still no consensus how to measure audit quality (Knechel, 2012). One of the most quoted definition of audit quality is from DeAngelo’s (1981, p. 186) “The quality of audit services is defined to be the market-assessed joint probability that a given auditor will both (a) discover a breach in the client's accounting system, and (b) report the breach.” Independency is an important factor by this definition.
When an auditor is dependent, the chance that the auditor will report uncertainties is very unsure, what will have a negative effect on the audit quality (Tepalagul & Lin, 2015).
Independent audits are essential for adding credibility to the audit (Watts & Zimmerman, 1986).
Independency is guaranteed by the following fundamentals of the principles objectivity, integrity, confidentiality and professionalism (International Ethics Standards Board of Accountants (IESBA) Handbook of the Code of Ethics for Professional Accountants, 2016). An auditor is considered not only independent by fact, but also independent in appearance. An auditor is responsible to act in public interest. If the auditor loses the trust of the public
interest, it is a lost cause.
Independence is essential for an auditor, in section 290.6 of “Code of Ethics”, dependence is comprised as:
Perfect independency as auditor from the client is implausible (Watts and Zimmerman, 1981). The European Commission also takes this into account in its definition of independence, according the European Commission (2002) it is impossible to pursue absolute independence, it is unrealistic to demand that an auditor does not hold an economic, financial or other
relationship in fact, everyone is more or less dependent of one another.
The code of ethics describes the following threats that could compromise the independence: a self-interest threat, a self-review threat, an advocacy threat, an intimidation threat and the familiarity threat (Code of Ethics, 2016, Section 100.13).
The familiarity threat warns “that due to a long or close relationship with a client or
employer, a professional accountant will be too sympathetic to their interests or too accepting of their work” (Code of Ethics, 2016, Section 100.13D). An example is a long connection
between a senior auditor and the assurance client. The more the auditor is familiar with the client, the more the auditor could lose his sharpness and eventually will go along with clients' accounting policies. When a long-term relationship between auditor and client undermines independence, a mandatory auditor rotation could be mitigated this risk.
There is also a threat from economic pressure. According to the economic theory, there is an economical incentive for an auditor to maintain the relationship and keep earning their wages. Over time the auditor could get susceptible for reduced vigilance (Kaplan & Mauldin, 2008).
a) Independence of Mind
The state of mind that permits the expression of a conclusion without being affected by influences that compromise professional judgment, thereby allowing an
individual to act with integrity and exercise objectivity and professional skepticism.
b) Independence in Appearance
The avoidance of facts and circumstances that are so significant that a reasonable and informed third party would be likely to conclude, weighing all the specific facts and circumstances, that a firm’s, or a member of the audit team’s, integrity,
The familiarity threat and economic pressure exists within a long-term relationship between auditor and client. It undermines the independence in mind and appearance. A mandatory auditor rotation could mitigate this risk considerably.
2.6 Measuring influence of auditor rotation on the New Audit Report
When measuring the effect of the auditor rotation on the quality of the audit, this concept should be operationalised. Audit quality is a complex concept and is not easily caught in a single variable. When connecting this to the new audit report, operationalisation becomes clearer. The measurements in the new audit report are the materiality described, the number of key audit matters noted, and the amount of words used to describe these matters in relation to the complete text of the audit report.
Materiality
The purpose of an audit of a financial statement is
A misstatement is material if it influences economic decision of the user's decision on the basis of the financial statement (IAS 320). Since 2009 a distinction is made between performing materiality (IAS 450) and planning materiality (IAS 320). In the new auditor's report, planning materiality and how materiality is determined are reported. Performing materiality is applied to determine the risks of material misstatement and the nature, timing and scale of more audit proceedings. Performing materiality involves carefulness, the professional judgement is built of the auditor’s estimation of risk of the company which is created to reduce the risk of undiscovered material misstatement to a suitable level (FRC, 2016).
The definition of performance materiality:
(a) To obtain reasonable assurance about whether the financial statements as a whole are free from material misstatement, whether due to fraud or error, thereby enabling the auditor to express an opinion on whether the financial statements are prepared, in all material respects, in accordance with an applicable financial reporting framework; and
(b) To report on the financial statements, and communicate as required by the IASs, in accordance with the auditor’s findings.
For purposes of the IASs, performance materiality means the amount or amounts set by the auditor at less than materiality for the financial statements as a whole to reduce to an appropriately low level the probability that the aggregate of
uncorrected and undetected misstatements exceeds materiality for the financial statements as a whole. If applicable, performance materiality also refers to the amount or amounts set by the auditor at less than the materiality level or levels for particular classes of transactions, account balances or disclosures.
During an audit the actual materiality used is lower than the declared materiality (Gray et al., 2011). The aim of the auditor is to guarantee the minimum use of the declared materiality. The risk of incorrect assumptions and estimates is also minimized by this caution of using a lower materiality. The literature review done by Messier, Martinov-Bennie, and Eilifsen (2015) shows when a materiality threshold is established, the threshold has a significant impact on audit judgment and decision-making whether or not to report a misstatement (Legoria et al., 2013).
For the auditor materiality has two areas, the auditor has to pick the suitable base to calculate materiality and he has to select the correct percentage applied the base (Steinbert, 1987). According to Steinbert (1987) these decisions are affected by the auditor’s impression of the needs of the user of the annual report and specificities of the client (the complexities of the business of their internal controls system). Most firms (in a sample by Citigroup 78%) reports the materiality base as profit before tax, which is around 5% but occasionally as high as 10% (Citi Research, 2014).
According to Amiram et al. (2015) when the materiality was not yet published in the new audit report, the user could not determine the reliability factor or the riskiness-persistence factor. The reliability factor indicates to users of the annual report how carefully the
statements have been prepared. The riskiness-persistence factor means that investors cannot tell the difference between riskiness or future persistence of the earnings (Amiram et al. 2015).
Explanation of the determination of materiality may be important for a good understanding of the audit and its limitations. The materiality serves as a measurement for classifying a
deviation as significant or not. A lower materiality leads to a higher audit quality (Minutti-Meza, 2015).
Key audit matters
In July 2013 the ISAAB introduced the key audit matters in the new audit report. In 2015 the ISAAB came with new standard IAS 701 “Communicating Key Audit Matters in the
Independent Auditor’s Report” which will be effective on December 15, 2016.
IAS 701.2 about the purpose of key audit matters:
The key audit matters add a lot of value to the new audit report, beforehand the areas of
The purpose of communicating key audit matters is to enhance the communicative value of the auditor’s report by providing greater transparency about the audit that was performed. Communicating key audit matters provides additional information to intended users of the financial statements (“intended users”) to assist them in understanding those matters that, in the auditor’s professional judgment, were of most significance in the audit of the financial statements of the current period. Communicating key audit matters may also assist intended users in understanding the entity and areas of significant management judgment in the audited financial statements. (Ref: Para. A1–A4)
higher assessed risk of material misstatement were not always reported which was also the case with the key audit maters specific to the organization (NBA 2014). The research by Simnett & Huggins (2014) reveals that users of financial statements have a significant need for information about key audit matters. According to DeZoort (2009) investors have more confidence in financial reporting if it contains internal audit information. This would add to the effectiveness of the investors. The purpose of adding a critical audit matter paragraph is to communicate the auditor's insights into difficult audit issues (Christensen et al., 2014). From the research by Hijink et al. (2018) on "De effecten van (verplichte) roulatie van accountantskantoor in de Nederlandse praktijk" in 23% of cases, the first-year audit (indirect auditor rotation) was considered by the auditor as a key audit matter in the statement. It also shows that, in 40% of cases, the key audit matter of the first-year audit is presented more prominent than the other key audit matters.
Audit rotation
The long-term relationship between an audit firm and the client could essentially affect the auditor's independence (Roos, 2012). Research of independency in appearance reveals that compulsory rotation has a positive effect (Gold et al., 2013). As the relationship between auditor and client lasts longer, incentives can arise that threaten and or impair the auditor's independence and hence the professionally critical institution of the auditor (Bazerman et al., 1997). Rotation can reduce these issues (APB, 2010). The general knowledge of the audited individual firms remains the same. Client-specific knowledge largely disappears in rotation (Nelson, 2009). Rotation prevents the accountant from getting outdated, the control program may become static over time (GAO, 2003). An audit by a new independent auditor would increase the quality of auditing (FRC, 2010).
Lennox (2014) provides a brief summary of the advantages and disadvantages of the routing Arguments in favour of mandatory rotation:
• A reduced threat of economic dependence • A fresh perspective
• Avoidance of close personal relationships and misplaced trust • Greater competition
Arguments against mandatory rotation:
• Diminished incentives to acquire client-specific knowledge • A loss of knowledge at the time of rotation
• Costly auditor switching
• Wrong incentives when the end of the rotation approaches.
The relation between audit tenure and quality is difficult to demonstrate. If determined in the literature the relation is positive. According to a Korean research of Kim, Song, and Tsui (2013) longer auditor-client relationship are associated with higher audit quality. Bowlin et al. (2015) find that rotation will lead to lower quality because of the loss of the experience and extra cost in the first year of the rotation. According to Knapp (1991) an auditor has the best chance to find the material errors if the tenure is between 5 and 20 years.
For the formation of the hypothesis, the emphasis is placed on the improved independency by rotation.
H1 Auditor rotation leads to a lower materiality measured in the audit report H2 Auditor rotation leads to higher number of reported key audit matters
H3 Auditor rotation leads to a higher fraction of text dedicated to key audit matters in the new audit report
Audit fees
All this new legalization provides extra complexity which could lead to higher audit fees. According to Hay, Knechel, and Wong (2006) audit fees are positively associated with client size, risk, and complexity, as well as with auditor litigation risk. Both the processes and the accounting rules on which the accountant reviews have increased in complexity (Kim, Liu & Zheng, 2012). Minutti-Meza et al. (2015) find that audit fees rise by approximately five percent of the new audit report. Minutti-Meza et al. (2015) check in the UK the audits fees for listed companies who voluntarily used IAS 700.
Research shows accordingly that there is a significant connection between auditor effort and audit fees (Blokdijk, Drieenhuizen, Simunic, & Stein, 2006). For this research only, the audit fee for audit work is included, the non-audit fees have not been taken into account. This
makes audit fees a proxy for auditor effort more reliable (Hay, Knechel, & Wong, 2006). The audit fee should reflect the number of work hours in the audit.
The audit rotation competition has increased, this has put prices under pressure, and often it is also a tender that does not include the actual costs. There is more noise for audit fees as a proxy for audit effort. Large fees provide more work effort, but abnormal fees are seen as bribes which decrease the audit quality (Kinney and Libby, 2002).
The research by Batenburg & Schaik shows that the determination of materiality plays a crucial role in finding a balance between audit risks and audit costs. Higher cost should lead to more audit effort and lower materiality. Minutti-Meza et al. (2015) find that lower
materiality is associated with higher audit quality. Minutti-Meza et al. (2015) also find that higher number of key audit matters is associated with a higher audit fees.
This leads to the following hypotheses:
H4 Higher audit fees have a negative influence on the height of the materiality level reported in the audit report
H5 Higher audit fees will have a positive effect on the number of reported key audit matters in the audit report
H6 Higher audit fees will have a positive effect on the fraction of words dedicated to key audit matters in the audit report
3
Methodology
The next step to answer the main question, will be to test the hypotheses from the literature review using a regression analysis. This chapter describes how the research is performed, the sample is defined, and how the data is collected. Furthermore, the variables in the model will be quantified and is determined which test will been used.
3.1 Sample design and data collection
This research intended to find a correlation between the audit rotation and materiality and the key audit matters. IAS 700 is essential for this purpose. In IAS 700, the standard is
established for the new auditor’s report and guidance is provided on the form and content (IAS 700). The new audit report differs from previous reports by the mandatory naming of the used materiality and the key audit matters (FRC, 2013).
In 2013 the new audit report is introduced in the UK and in 2014 in the Netherlands. The new auditor’s report is since 2013 applied in the UK and (voluntarily) in the Netherlands. In the Netherlands, the NBA imposed the NVCos 702N for annual reports of PIEs since 2014 (NBA, 2014). In the UK the FRC compels (revised) ISA (UK and Ireland) 700 for company listed on the he FTSE 100, FTSE 250 and FTSE Small Cap indices since 2014 (FRC, 2013). Use of the report is mandatory since in 2014 for the companies in this sample.
Given that the auditor’s report is implemented since 2013, this sample starts with the financial years 2013 until 2016, with listed companies of FTSE 100, FTSE 250, AEX, AMX and ASsX. Only companies with data available of the new auditor report were added to the sample. The sample consist of 326 auditor’s reports whereof 311 used the new auditor’s report. All the companies in the sample published the materiality and the key audit maters. The data collection is composed from different sources. Most of the data, the key audit matters, materiality, the base of the materiality, auditor, assets were hand collected from the annual reports of the companies in the sample. The team for collection was trained
beforehand in collecting and reporting the data. The team consisted of seven master students, responsible for adding the data for 2016 to the existing database. The complete dataset was then checked by an independent employee on gaps and misstatements.
The audit fees however were added from the Compustat database using an excel add-on. The data was matched on ISIN code for each of the years. The missing audit fees were added by hand from annual reports. Of the existing fees in the database, 5% was checked by hand to determine the reliability of the reported fees.
3.2 Model and variables
The variables have been measured following the research of Amriram et al. (2015) and Minutti-Meza et al. (2015). The regression model used, is based on the work of these researchers. For each of the dependent variables three models were tested:
𝑌𝑖𝑡= 𝛽0+ 𝛽1𝑅𝑜𝑡𝑎𝑡𝑒𝑖𝑡+ 𝛽2𝐵𝑎𝑠𝑒𝑐ℎ𝑎𝑛𝑔𝑒𝑖𝑡+ 𝛽3𝐴𝑢𝑑𝑖𝑡𝑓𝑒𝑒𝑠𝑖𝑡 + 𝛽4𝑆𝑖𝑧𝑒𝑖𝑡 + ɛ𝑖 (1)
𝑌𝑖𝑡 = 𝛽0+ 𝛽1𝑅𝑜𝑡𝑎𝑡𝑒𝑖𝑡+ 𝛽2𝐵𝑎𝑠𝑒𝑐ℎ𝑎𝑛𝑔𝑒𝑖𝑡+ 𝛽3𝐴𝑢𝑑𝑖𝑡𝑓𝑒𝑒𝑠𝑖𝑡 + 𝛽4𝑆𝑖𝑧𝑒𝑖𝑡+ 𝛽5𝑦𝑒𝑎𝑟 𝑑𝑢𝑚𝑚𝑦𝑖𝑡+ ɛ𝑖 (2)
𝑌𝑖𝑡+1 = 𝛽0+ 𝛽1𝑅𝑜𝑡𝑎𝑡𝑒𝑖𝑡+ 𝛽2𝐵𝑎𝑠𝑒𝑐ℎ𝑎𝑛𝑔𝑒𝑖𝑡+ 𝛽3𝐴𝑢𝑑𝑖𝑡𝑓𝑒𝑒𝑠𝑖𝑡 + 𝛽4𝑆𝑖𝑧𝑒𝑖𝑡 + ɛ𝑖 (3)
Model 1is a time series regression measuring the relationship between the dependent and predictor variables
Model 2 is a time series regression in which year fixed effects were added to control for changes in response to the changes in regulation from 2013 to 2016 for the whole sample. Model 3 is model 1 with a lagged dependent factor to determine the effect of the rotation in a following year.
For each dependent variable, the same predictive models were employed.
3.3 Dependent variable
The following variables were used in the model as dependent variables.
Key audit matters
The dependent variable in this research consists of the total of key audit matters mentioned in the new auditor's report of company 'i' in year 't' (KAM i,t). KAM are the most significance matters in the audit according to the auditor’s judgement. The KAM are described in the new auditor’s report (ISAAB, 2015). In the paper by Lennox et al. (2015) the total number of KAMs is used (among other variables) to measure the informative level of the new audit report (Lennox et al. 2015).
Materiality
Materiality is a dependent variable in this research measured the materiality threshold mentioned in the new audit report divided by total assets, the materiality is expressed as a percentage of total assets (Amiram et al., 2015). Materiality is shown of company 'i' in year 't' (Materiality i,t). A different way of quantifying the materiality is to pick only the base “profit before lost”, this is generally used. But a change of auditor is often associated with a change of base used to determine materiality (FRC, 2006), with this calculation method these company get not excluded.
KAM Fraction
KAM Fraction is the total number of words of the key audit matters section divided by the total number of words in the auditor’s report. KAM fraction is shown of company 'i' in year 't' (KAM Fraction i,t). KAM Fraction gives more information than only the number of Key Audit matters. It shows how much attention in proportion has been paid to the key audit matters section.
3.4 Independent variables
To measure the effect on the dependent variables the following predicting variables have been identified.
Rotate
Rotate (Rotate j,t) shows if the auditor is rotated in year ‘t’ for company ‘j’. The auditor of the company in 2013 will be compared with the auditor in the next years. This is done with a dummy. The auditor is mentioned in the auditor report. Since the period of this research is four years no account is taken with a second change of auditor.
Base changes
Base change (Base change j,t) measured if there is a change in the base to calculate materiality in year ‘t’ for company ‘j’. For the auditor to choose of the materiality has two dimensions, to pick the suitable base to calculate materiality and select the correct percentage applied the base (Steinbert, 1987). In review of experience of new auditor report from FRC indicated a link between change of auditor and a change in the base used to determine
materiality (FRC, 2016). The sample shows that the change in auditor and base does not have to change simultaneously. Most firms (in a sample of Citigroup 78%) report the materiality base, profit before tax which around 5% but occasionally as high as 10% (Citi Research, 2014). When selecting cases on this requirement, most of the rotations would be excluded from the research.
3.5 Control variables
This research investigated a link between auditor rotation and number of key audit matters and the materiality mentioned in the new auditor’s report. There are more variables that affect the key audit matters and the materiality. The following control variables are included: the audit fees (auditfees j,t) , the size of the company (size j,t) and year dummies (years).
Audit fees
Audit fees is the log of total fees paid to the auditor for audit services in each year (Minutti-Meza, 2015). Hope et al. (2012) used audit fees as a proxy for audit effort. Effort should lead to a higher number of identified Key Audit Matters and more dedicated to the KAMs in the audit report. The relationship with materiality is slightly more complex, as there is a
reciprocal effect of effort on materiality. A lower materiality should lead to more work, which would imply a greater effort. On the other hand, more effort could increase the reliability and therefore lower the materiality.
Size
The size is measured by the log of the total assets of a company (Simunic, 1980). Large companies publish more and more complex annual accounts (Li, 2008), which can affect the materiality and key audit matters. Assets is already used in the dependent variable materiality, but there are no signs that this leads to a problem. The value of R-squared is high but not extreme, there is no evidence of a too strong correlation between size and materiality.
Years
It is checked whether the financial years influence the materiality and the number of key audit matters of the auditor’s report, in view of the fact that 2013 was a year audit. In a first-year audit many processes are new, and the work pressure is high. It is plausible that there is a learning curve. Year dummies are used to determine if the adoption of regulation has effect on the dependent variables for the whole sample.
4
Empirical results
This chapter reports the results of the statistical analysis, shows the descriptive statistics and finally, a multiple regression analysis is used for accepting or rejecting the hypotheses.
4.1 Dataset
In order to properly perform the regression analysis, outliers in the variables have to be
minimized. Outliers weaken the connection between the independent and dependent variables. With variable materiality, audit fees and size, winsorizing has been applied to correct the outliers. The variables were winsorized at level 1% and 99% levels for each year. Stata 13 has been used for the tests.
4.2 Descriptive statistics
The descriptive statistics of the variables are presented in Table 1. The low standard deviation of the key audit matters shows that there is little variation between the observation of the number of key audit matters. The minimum of key audit matters report is zero, the maximum is nine and the mean is 3,2.
The number of 8 reported key audit matters by Rolls Royce in 2013 was very extensive, could it have been a signal for the discovered bribery scandal (Ft., 2015). The key audit matters, on average, make up 8% of the new audit report.
Table 1: Shows the descriptive statistics of all variables in the models.
VARIABLES N mean sd min max var skewness kurtosis
KAM 1,259 3.295 1.963 0 9 3.852 -0.0161 2.673 frac_KAM 1,259 0.0867 0.0604 0 0.523 0.00364 1.616 10.51 Materiality 1,053 0.00749 0.0148 0.000156 0.130 0.000220 7.011 55.87 rotate 1,298 0.125 0.331 0 1 0.109 2.270 6.155 base_change 1,298 0.258 0.438 0 1 0.192 1.106 2.222 Auditfee size 1,237 1,257 3.102e+06 2.01 5.918e+06 1.947 5.918e+06 13.34 5.100e+07 28.61 3.503e+13 3.792 4.337 0.410 27.10 4.514 Year 1,298 2014 1.118 2013 2016 1.251 0.00221 1.640 Number of id 311 311 311 311 311 311 311 311
Materiality expressed as percentage of total assets has a mean of 0,0749 with a big spread. The size of the materiality changes considerably. An explanation can be found in the different types of companies. As materiality is measured as a fraction of assets, the size of a company is of influence.
The mean of rotation is 12,5% which means that 1 out of 8 companies changed auditor in the period 2013-2016. In table 2, below, is an overview in which year the companies have rotated auditor. 2013 was taken as reference point, so the rotation of that year is not included. In the Netherlands companies have already proportionally more rotated than in the UK because of the intended law of mandatory auditor rotation.
Table 2: An overview whether the companies in the sample rotate of auditor. 2013 2014 2015 2016 Netherlands no rotation 73 56 48 46 rotation 17 25 27 UK no rotation 250 223 224 214 rotation 29 27 37 Total companies 325 325 324 324
The mean of the base change is higher than the mean of the rotation. There is more often a base change than a rotation of the auditor. A base change implies that the calculation base for materiality has changed in a way that significantly changes the judgement of the auditor This could be a signal from the accountant that the company is an uncertain environment, or the original base has little value for the judgment.
4.3 Testing for parametric assumptions
To make sure the selected regression models were appropriate for the data, the parametric assumptions have been tested. The data was tested on linearity, independence of observations, normalcy and equality of variance.
Linearity was tested by determining patterns in scatter plots between the dependent and independent variable and by checking scatter plots of the predicted values and the residuals. No proof of non-linear patterns was found (Field, 2013). Independence of observations is limited in time series. This is corrected by using a regression model for panel data. To make sure that predictor variables do not negatively influence each other, these were tested using a correlation matrix. Field (2013) states that when a correlation is higher than 0.7 t at a 5% level, this indicates a multicollinearity problem. The highest measured value is 0.627 between audit fee and size, indicating multicollinearity is not a concern in this research.
Normalcy was tested by checking the data for skewness and kurtosis. The results of these tests are reported in table 1. Variables with a high value were winsorized. Equality of variance was tested after the regression models were run with a modified Wald test in Stata. As a result, the models were tested with robust standard errors, as there was heteroskadicty in the data. Table 3 Pearson’s correlation analysis
Materialit y keyaudit matters frac_KA M rotate base_cha nge size_w Auditfee Materiality 1.000 keyauditmatters -0.142 1.000 frac_KAM 0.144 -0.553 1.000 rotate 0.007 0.006 -0.012 1.000 base_change 0.013 -0.092 0.034 0.189 1.000 size_w -0.379 0.375 -0.188 -0.022 -0.146 1.000 Auditfee -0.103 0.326 -0.190 -0.008 -0.088 0.627 1.000
4.4 Results of regression analyses
In the tables below the results of the regression analysis are presented
Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
(1) (2) (3) (4) (5) (6) (7) (8) (9)
VARIABLES KAM1 KAM2 KAM3 Materiality1 Materiality2 Materiality3 Frac KAM1 Frac KAM2 Frac KAM3
Mat_ass_w 2.069 4.253 6.977 0.201 0.0580 0.181 (3.442) (3.397) (7.218) (0.157) (0.139) (0.204) rotate 0.223** 0.249** -0.0425 0.000753 0.000171 -0.00159 -0.000338 -0.00446 0.00768* (0.108) (0.107) (0.153) (0.000732) (0.000725) (0.00109) (0.00397) (0.00394) (0.00430) base_change 0.0828 -0.0274 -1.275*** -0.000499 0.000144 -0.00181 -0.00734** -0.00784** -0.0288*** (0.0752) (0.0826) (0.155) (0.000590) (0.000676) (0.00157) (0.00285) (0.00307) (0.00383) size_w -0.0512 0.00206 0.0774 -0.0200*** -0.0203*** 0.00189 0.00962** 0.00345 0.00171 (0.103) (0.108) (0.201) (0.00192) (0.00192) (0.00169) (0.00395) (0.00389) (0.00602) Auditfee -5.72e-08 -5.27e-08 -1.80e-08 4.39e-10 3.41e-10 -1.31e-10 1.61e-10 -3.36e-10 3.64e-10
(4.12e-08) (4.30e-08) (3.77e-08) (3.77e-10) (3.58e-10) (1.16e-10) (5.93e-10) (5.59e-10) (8.12e-10)
2014.Year 0.169 0.000577 0.0219*** (0.136) (0.000537) (0.00319) 2015.Year -0.00208 1.261*** 0.00133** -0.00226* 0.0284*** 0.0376*** (0.135) (0.144) (0.000652) (0.00118) (0.00349) (0.00295) 2016.Year -0.0851 1.225*** 0.00356*** -0.00294** 0.0262*** 0.0504*** (0.142) (0.142) (0.00107) (0.00115) (0.00357) (0.00365) Constant 5.124** 3.928 1.251 0.448*** 0.453*** -0.0318 -0.111 0.00644 0.0238 (2.295) (2.389) (4.451) (0.0421) (0.0419) (0.0364) (0.0881) (0.0864) (0.134) Observations 1,031 1,031 875 1,031 1,031 735 1,031 1,031 875 R-squared 0.021 0.033 0.375 0.635 0.650 0.049 0.022 0.129 0.443 Number of id 311 311 311 311 311 306 311 311 311 Company FE Y Y Y Y Y Y Y Y Y Yeat FE N Y Y N Y Y N Y Y
The predictive value of the models ranges from 2% to 65%, Adding a lagged dependent
Analysis of the main findings
The predictive value of the models ranges from 2% to 65%, adding a lagged dependent variable has a positive effect on the predictive value of the model, increasing the strongly, except for the models for Materiality, where the R2 is reduced from .650 to .049.
Due to the time series nature of the data a fixed effects model was used. The model takes into account the time series of 311 companies. To control the changes in legislation, impacting all companies in the sample, time fixed effects were used.
The first-year rotation has a positive association to the number of key audit matters. If this association is correct for time effects the positive association holds. The lagged model shows that a year later the positive association is gone.
Rotation has zero effect on the materiality. Also, the lagged and time effects model shows no significant association.
Fraction KAM, in the lagged model, has a positive association the year after the rotation. In the first year of the engagement, key audit maters are found, the next year the auditor provides the motivation to the key audit matters.
Audit fees have no association with materiality, KAM and KAM fraction. Audit fees, in this research, seem not to be a predictor for the quality of audit.
Additional analysis
There are a number of things that have not been hypothesized but are striking.
The base change has, in the lagged model, a negative association with the KAMs. Hence, if a base change is implemented, the following year the number of key audit matters will
decrease. Base change has also a negative association with Fraction KAM.
Size has a negative association on materiality, for this sample applies: the bigger the company the lower the materiality. But in the lagged model size has no significant association with materiality. Size has no association with the number KAMs, but size has a positive association with Fraction KAM.
The Year-effects has in the lagged model a positive association with Fraction KAM and KAM, de years-effects are useful. So as the years go by (regardless of how many rotations) the number of key audit matters is increased. Non-rotating companies are also paying more attention to the KAMs. The same applies to the Fraction KAM, every year the ratio increases, more attention is given to the key audit matters and are described more comprehensive.
The years effects have an association with the materiality, apart from the rotation is the effect of the new audit report clearly visible in the reporting. It is fascinating, that the materiality increases in the same year and is lower year later. The auditor has less time to spend on auditing because they are busy with the new regulations and understanding of the processes. In the second year, the materiality is getting smaller, so there is a learning effect. The auditors can also go deeper into the matter, because they understand the processes better.
Base change has a negative effect on the number of KAM’s that will be reported in the following year. The effect on the lagged dependent variables is less strong, which shows that the key audit matters in the report will have been addressed. In the second year and following year the number of reported audit matters declines, while the amount of text dedicated to the KAMs slightly increases, indicating that those matters still requiring reporting, need more text to define.
Rotation did not have a significant effect, the new audit report itself does have a significant effect on the issues that have to be reported.
4.5 Hypothesis testing
Based on the regression analysis of the three models the following can be concluded with regard to the hypotheses:
H1 Auditor rotation leads to a lower Materiality measured in the audit report
No proof for a statistically significant relation between rotation and materiality was found. Hypothesis 1 is therefore rejected.
H2 Auditor rotation leads to higher number of reported key audit matters
The number of key audit matters is positively influenced by auditor rotation in the year of change. Hypothesis 2 is hereby supported.
H3 Auditor rotation leads to a higher fraction of text dedicated to key audit matters in the new audit report
Auditor rotation does not immediately positively influence the amount of words dedicated to the key audit matters in the New Audit Report. In the next year there is a significant positive effect, indicating a delay in the effectiveness of rotation on the amount of attention for the key audit matters. This supports Hypothesis 3.
H4 Higher audit fees have a negative influence on the height of the materiality level reported in the audit report
No proof for a statistically significant relation between audit fees and materiality was found. Hypothesis 4 is therefore rejected.
H5 Higher Audit fees will have a positive effect on the number of reported key audit matters in the audit report.
No proof for a statistically significant relation between the audit fees and the key audit matters was found. Hypothesis 5 is therefore rejected.
H6 Higher Audit fees will have a positive effect on the fraction of words dedicated to key audit matters in the audit report.
No proof for a statistically significant relation between the audit fees and fraction KAM was found. Hypothesis 6 is therefore rejected.
5
Discussion and Conclusion
This chapter includes the conclusion, the limitations and further research
5.1 Conclusion
In these revolutionary times, this research has been conducted under the influence of two recent major changes: the new audit report and auditor rotation. The purpose of this paper is to research the impact of the mandatory audit rotation on the new audit report. In particular it researches the influence on the reporting of materiality and key audit matters. Will the change of auditor be visible in the new audit report by a modified materiality or changed number of key audit matters? To find out, the main question of this research is:
How does the rotation of external auditors influence the materiality, number of key audit matters and motivation of these key audit matters in the new audit report for listed companies?
This question has been answered with empirical research. This research finds empirical evidence that there is a positive significant relation between auditor rotation and the number of reported key audit matters. From the literature this was expected, rotation would lead to a more independent auditor, reporting more issues that need attention in the annual report, limiting the information asymmetry for shareholders. Independency leads to a higher audit quality and with a higher audit quality more key audit matters will surface.
Within the sample there is no correlation between rotation and materiality. From the literature, it was expected that improved independence would lead to better audit quality, resulting in a lower reported materiality. A high audit quality is correlated with low
materiality. But the influence of the departed knowledge by changing auditor seems to be of more influence than expected.
There is no relation between rotation and the motivation of key audit matters in the new audit report. The new, and more complex legation, costs more time (than expected) to the auditor, certainly in the first year. It seems that there is not enough time and specific knowledge to motivate the key audit matters and the processing properly. There seems a delayed effect, as rotation does have a positive effect on the fraction of words used to describe the key audit matters in the second year of the new engagement.
There is no significant relation found between audit fees, materiality, number of key audit matters and motivation of key issues in the new audit report. This shows that audit fees are not a good proxy for audit effort. Due to rotation, many prices are determined by a tender and are even below cost price in the first few years. The additional cost of the first year audit is not passed on to the client. This initial loss must be redeemed to make the engagement worthwhile. As new regulations and scrutiny from legislators diminish the change to become more efficient, the best bet is to raise audit fees as the years go on.
Recent research by Hijink (2018) about the rotation of audit firms in the Netherlands states that there is no reason to conclude that office rotation leads to more independency of the auditor or to better quality audits. Hijink argues that being unknown with the organisation makes the auditor more prone to misrepresentation, missing key audit matters, and being less dependent on the information provided by management of the client.
This research shows that the new audit report has generally lead to more reported key audit measures. The rotation adds a new incentive to report more key audit measures in the first year of the engagement. From 2014 to 2016 a learning curve was noticeable, in lowering he materiality, adding more reported key audit matters and devoting more text to these matters. The rotation of auditors only had a limited impact on these factors.
Carefully it can be stated that the new audit report seems to help lower the information asymmetry between shareholders and clients. Limited proof is found that the rotation in the early phase has led to enhanced independence and higher audit quality. On new engagements the auditor has limited knowledge of the inner workings of the organization and will therefore hedge its bets in materiality and will chose to describe key audit matters not too specifically. As confidence grows with the relation, so will the quality of the audit.
5.2 Limitations
There are a lot of changes in the world of auditing, as well from without as from within. This makes it more difficult to measure the ceteris paribus effect of rotation. This research has modelled the regression analysis as a time series regression, the effect of unobserved factors on the variance of the dependent variables is around 0.55, indicating that other factors than those in the model influence the dependent variable.
The sample consists only of listed companies, which limits the generalizability of this research. Listed companies are only a fraction of the companies in the market and although they represent a large fraction of the clients for the Big Four, they hardly matter for other firms.
Is the independency of an audit firm still maintained when a whole ‘client-team’ rotates with the client? There is anecdotal evidence of auditors and even complete client teams being hired by firms winning a tender to continue the audit for a new firm. This leads to the question, in which cases is this research indeed seeing the effect of a completely new audit team as is proposed in the regulations? In this dataset no information on the audit team was present. As both firms and their clients have opposed the strict rotation requirements in the
Netherlands, it has become hard to enforce a strict independence of client and auditor as the benefits of a good relationship are shared between management and audit team and the side effects come at a cost to society.
5.3 Further research
It will be worthwhile to do the research when the rotation cycle of ten years is completed. This will give insights in the long-term effect of the regulation, and will possible help dispel the arguments on both sides of the regulations on the effect of rotation.
It would be useful to do a thorough case study, in order to have a better view of what is really happening to the audit and how the new audit team sets up an audit for a new client compared to how a seasoned team would revisit a longstanding client. How would a fresh take on the situation compare to the experience of been able to analyse patterns of behaviour and learn form past mistakes?
At this moment, most rotations are not older then two years, so presumably there is still a lack of knowledge which factors influence the audit quality. Further in the process it would be worthwhile to see if the knowledge of the audit team has increased and whether the auditor will be better equipped to outperform compared to the old situation. Doing an event study on the auditor fees before the and after the rotation might give insight if experience gives way to effort.
References
Amiram, Dan & Chircop, Justin & Landsman, Wayne & Peasnell, Ken. (2015). Mandatorily Disclosed Materiality Thresholds, their Determinants, and their Association with Earnings Multiples. SSRN Electronic Journal. . 10.2139/ssrn.2631876.
Audit Quality Forum. (2015). Agency theory and the role of audit. AuditQuality.
Autoriteit Financiële Markten. (2014). Uitkomsten onderzoek kwaliteit wettelijke controles Big 4-accountantsorganisaties. Amsterdam: Autoriteit Financiële Markten.
Bailey III, K. E., Bylinski, J. H. & Shields,, M. D., (1983). Effects of Audit Report Wording Changes on the Perceived Message. Journal Of Accounting Research, 21(2), pp. 355-370. Blake, K., Carcello, J.V., Harrison, N.J., Head, M.J., Roper, B.E., Simpson, A., Sondhi, T., Tarola, R.M., Turner, L.E., Williams, M. en Yerger, A. (2011). Response of a subgroup of the PCAOB’s investor advisory group to the PCAOB’s concept release concerning reports on audited financial statements and related amendments to PCAOB standards. Current Issues in Auditing, 5(2): 21-50.
Blokdijk, H., F. Drieenhuizen, D. A. Simunic, and M. T. Stein. (2006). An analysis of cross-sectional in Big and Non-Big public accounting firms’ programs. Auditing: A Journal of Practice and Theory 25 (1): 27–48.
Cameran, Mara & Trombetta, Marco. (2015). Mandatory Audit Firm Rotation and Audit Quality. European Accounting Review. na. . 10.1080/09638180.2014.921446.
Chandler, R., EdwardsJ. R., (1996),"Recurring issues in auditing: back to the future?", Accounting, Auditing & Accountability Journal, Vol. 9 Iss 2 pp. 4 – 29.
Christensen, B., S. Glover, M. Steven, and C. Wolfe. (2014). Do critical audit matter
paragraphs in the audit report change nonprofessional investors’ decision to invest? Auditing: A Journal of Practice & Theory 33 (4): 71–93.
Christensen, B.E., Glover, S.M., Omer, T.C., & Shelley, M.K. (2015). Understanding audit quality: insights from audit professionals and investors. Contemporary Accounting Research. Advance online publication. DOI: 10.1111/1911-3846.12212.
Christensen, Brant E. and Glover, Steven M. and Wolfe, Christopher J., Do Critical Audit Matter Paragraphs in the Audit Report Change Nonprofessional Investors’ Decision to Invest? (2014). Auditing: A Journal of Practice & Theory, Forthcoming. Available at SSRN:
https://ssrn.com/abstract=2318590 or http://dx.doi.org/10.2139/ssrn.2318590.
Church, B.K., Davis, S.M., McCracken, S.A. (2008). The Auditor’s Reporting Model: A Literature Synthesis and Overview. Accounting Horizons vol. 22, No 1: 69 – 90.
Citigroup, (2014). “New UK Auditor’s Reports: A Review of the New Information,” Citigroup Equity Research, London.
Commission on Auditors' Responsibilities: Report, Conclusions, and Recommendations (1978). New York: AICPA.
Committee on the Financial Aspects of Corporate Governance (1992). The financial aspects of corporate governance (Cadbury report). London: Gee.
Davis, S. M., (2007). Market response to auditor reports: A reexamination of auditor materiality thresholds. s.l.:Working Paper, Emory University.
DeAngelo, L. (1981). Auditor Size and Audit Quality. Journal of Accounting and Economics vol. 01: 183 – 199.
DeFond, M.L., & Zhang, J. (2014). A review of archival auditing research. Journal of Accounting and Economics, 58 (2): 275-326.
Dijsselbloem, (2015). Beantwoording Kamervragen over de roulatieplicht van OOB- accountantsorganisaties in de EU-Verordening, Retrieved from
https://www.accountant.nl/contentassets/b57e2ba30a324df386cb9067b5550113/antwoorden-kamervragen-roulatieplicht-accountantsorganisaties.pdf.
Dopuch, N., and D. Simunic. (1982). Competition in auditing: An assessment. In Symposium on Auditing Research IV, 401–50. Urbana: University of Illinois.
Doxey, M., (2014) The Effects of Auditor Disclosures Regarding Management Estimates on Financial Statement Users’ Perceptions and Investments (2014). Available at SSRN:
https://ssrn.com/abstract=2181624 or http://dx.doi.org/10.2139/ssrn.2181624 Eisenhardt, K. M. (1989). Agency theory: An assessment and review. Academy of management review, 14(1), 57-74.
European Commission (2002), Commission Recommendation of May 16, 2002 (notified under document number C(2002), 1873, 2002/590/EC.
European Commission (EC) (2010) Green Paper. Audit policy: Lessons from the Crisis, Commission Staff working paper, COM(2010) 561 final (October 13, 2010). Brussels: EC. European Commission. (2011). Proposal for a Regulation of the European Parliament and of the council on specific requirements regarding statutory audits of public-interest entities. November 30, 2011. Retrieved from
http://ec.europa.eu/internal_market/auditing/docs/reform/regulation_en.pdf.
Europese Commissie (2011). Voorstel voor een Verordening van het Europees Parlement en De Raad betreffende specifieke eisen voor de wettelijke controles van financiële overzichten van organisaties van openbaar belang, Gedownload op 29 november 2015,
http://ec.europa.eu/ internal_market/auditing/docs/reform/ regulation_nl.pdf. Field, A. (2013). Discovering statistics using IMB SPSS Statistics. Sage.
