Perceived influence of a governmental auditor on
audit quality
Qualitative research in the Dutch audit practice
Name: Kharina Tromp Student number: 11304529
Thesis supervisor: Georgios Georgakopoulos Date: 21-06-2018
Word count: 25.663
MSc Accountancy & Control, specialization Accountancy Faculty of Economics and Business, University of Amsterdam
2 Statement of Originality
This document is written by student Kharina Tromp who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
Abstract
This paper aims to provide insight into the concept of audit quality and the perceived influence of a governmental auditor on audit quality. Audit quality has been an important topic in the audit field since recent scandals and negative oversight reports. Independence of the audit firm has been mentioned in past research as an important contributor to audit quality. Most past research has been focussed on data analytics in which the notion of audit quality is tried to be captured through the use of proxies. By conducting a case study and performing interviews I have provided additional insight into the concept of audit quality through another perspective. Municipalities in the Netherlands are able to install their own audit company within the
organization and therefore bypass instalment of a public, independent audit firm to perform the financial audit. This might influence the perceived independence of the auditor. By conducting eleven interviews with individuals involved with the audit of a local government in the
Netherlands, I have researched the notion of audit quality and the perceived influence of the instalment of a governmental auditor on audit quality. Through this study I have identified influential factors of a governmental auditor on audit quality to include independence, audit hours, audit budget, specialization, intrinsic motivation and non-commercialism. The governmental auditor is overall described as a positive contributor to audit quality. For an organization to be able to perform their services they need to obtain and maintain legitimacy. With the use of different legitimacy theories, I have illustrated the strategies undertaken by the governmental auditor that have contributed to them being able to remain appointed. This research complements past studies on providing insight from a novel perspective and observing the influence of governmental auditors on the wide-researched topic audit quality.
4
C
ontents
1 Introduction ... 6
1.1 Audit independence, audit quality and government auditors ... 7
1.2 Qualitative research in audit quality ... 8
1.3 Legitimacy theory as a framework ... 9
2 Literature review ... 10
2.1 Audit quality ... 10
2.1.1 Audit quality definition from an academic perspective ... 10
2.1.2 Factors associated with audit quality ... 12
2.1.3 Audit quality from a standard setters’ perspective ... 13
2.1.4 Summary of audit quality ... 15
2.2 Local governmental auditors ... 16
2.2.1 Current governmental auditing rules and regulations in the Netherlands ... 17
2.2.2 Differences public and government auditors ... 18
2.2.3 Summary of governmental audit organization... 19
2.3 Research question ... 19
3 Theoretical Framework ... 21
3.1 Defining the legitimacy theory ... 21
3.2 Social contract and legitimacy theory ... 22
3.3 Types of legitimacy ... 23
3.4 Legitimacy strategies ... 24
3.5 Auditing as a social construct ... 25
3.6 Use of legitimacy theory ... 26
4 Research methodology ... 29
5
4.2 Research method interviews ... 31
4.3 Research method documents ... 32
4.4 Research method data analysis ... 32
5 Findings ... 34
5.1 The audit process and audit quality ... 34
5.2 Governmental auditor influential factors ... 37
5.2.1 Independence ... 37
5.2.2 Audit budget and hours ... 43
5.2.3 Specialization and government knowledge ... 46
5.2.4 Intrinsic motivation, non-commercialism and other influential factors ... 48
5.2.5 Governmental auditor influential factors conclusion ... 50
5.3 Legitimacy theories applied... 51
5.3.1 Pragmatic legitimacy ... 51 5.3.2 Moral legitimacy ... 51 5.3.3 Cognitive legitimacy ... 53 5.3.4 Legitimacy strategies ... 54 5.3.5 Legitimacy conclusion ... 56 6 Conclusion ... 57
6.1 Summary, discussion and conclusion ... 57
References ... 62
6
1 Introduction
This thesis will focus on what the perceived influence on the audit quality of a local government in the Netherlands is when this audit is performed by a governmental auditor. In the last years there have been multiple developments in the auditing of local governments that have led to media and parliamentary attention, questioning the way the audit is conducted (AFM, 2014) and sharpening the rules and regulations placed on auditing, including governmental auditing.
An effect of these developments was visible for the audits of the 2016 financial year, where there were several smaller municipalities who were not able to appoint an accountant yet in December 2016 (FD, 2016). Their previous auditor contracts were disbanded or not extended, leading to multiple municipalities suing their former accountant. Their previous auditor stated that they cancelled the audit agreement due to the developments of the rules and regulations and staff shortages, which made it financially not attractive enough to retain the clients1. Having an
audit of their financial information is mandatory for local governments in the Netherlands. Absence of the statutory audit on the financial information can lead to a discount on the governmental budget of the municipality.
Next to this event about halve of the municipalities has not been able to get an unqualified audit statement for their financial statement of 2015 (BMC, 2016). Plasterk, the Dutch minister of Finance, stated in an informative letter to the Dutch Parliament that the qualified opinions were due to errors and uncertainties in the financial statements that arose due to changes in the social domain (Ministerie BZK, 2016). In 2015 a decentralization from the government to municipalities took place for a range of task in the social domain, like juvenile law, increasing work and accountability duties of the municipalities. This led to more work for both the municipalities and accountants, while the additional charge for the audit was already under pressure in many municipalities2.
All these developments and events lead to questions about the way audit of local government is conducted. Several researches and reports were performed (AFM, 2014; BMC, 2016). Parliamentary questions were raised. The Nederlandse Beroepsorganisatie van Accountants (hereafter: NBA), the professional body for accountants in the Netherlands and the
1 https://www.binnenlandsbestuur.nl/financien/nieuws/grote-accountants-keren-kleine-gemeenten-rug-toe.9552722.lynkx
7 Vereniging Nederlandse Gemeenten (hereafter: VNG), the Dutch association for municipalities, have had multiple meetings in which representatives consulted with each other about improving financial reporting and the auditing process. A result from these meetings was the establishment of a commission by the VNG that formulated recommendations for the financial audit, the commission vernieuwing Besluit Begroting en Verantwoording (hereafter: commission BBV). These recommendations are stated in an advisory report in which questions are raised about the requirements of accountants and municipalities and monitoring needs (BBV, 2015). Although there have been a lot of questions, regulations haven’t changed (yet) and local governmental auditing continues to be discussed.
1.1 Audit independence, audit quality and government auditors
Christensen et al. (2016) state that the degree to which the users of financial statements can relate on audit opinions depends on the quality of the audit performed (p. 1649). The definition of audit quality has been an ongoing debate in existing literature (DeFond and Zhang 2014; Francis 2011).
An important attribute often associated with audit quality is the independence of the involved auditor. When a governmental auditor is established the kind of relationship between the municipality and the auditor differs from the relationship when a public auditor is involved. It is therefore interesting to question whether a governmental auditor is still able to perform an audit with the desired audit quality. Andon et al. (2014) stated in their research for legitimacy of audit performance of cap auditors that the independence of the auditor might be not as crucial as it is looked upon, this might be the case for governmental auditors of municipalities.
In this paper I will therefore conduct a qualitative research answering the research question what the perceived influence on audit quality is when a governmental auditor is involved.
Answering the research question is important because it adds information to the debate whether the current system of local governmental audit is effective or if, and how, the audit regulations for local government should be altered. Next, it can give insight in whether the quality of the financial audit is perceived differently for aligned audit firms compared to public audit firms, adding information to the debate whether there should be a governmental auditing firm responsible for all the local governmental audits, as there used to be till 1987 (BMC, 2016).
8 It will provide additional insight in the interaction between independence and auditor specialization and their relationship with audit quality.
1.2 Qualitative research in audit quality
Audit quality is an intensively researched topic but most research is from a quantitative perspective. The direct outcome of an audit such as the quality of the audit documentation, is mostly unobservable (Defond and Zhang, 2014). Past research has mainly focussed on the aspects of audit quality that can be observed, such as going-concern opinions and discretionary accruals. To understand more about the influence of the aspects of being a governmental auditor on the audit quality, factors which can probably not all be derived from the financial report, I conducted an interpretive case study. Governments annual statements differ from private-sector statements. The nature of the entity operations and their accounting and financial reporting is different (Samelson et al. 2006). This makes quantitative research regarding the often-used proxies for audit quality harder, as not all proxies are relevant or available. Researching the influence of the governmental auditor on audit quality is a novel area, because earlier research regarding audit quality performed at private firms cannot be generalized. As Samelson et al. (2006) stated: ‘The technical requirements of governmental auditing are sufficiently specialized that private-sector audit quality research findings may not generalize to the government audit sector’ (p.143).
Auditing and its aspects can be studied in many ways. A thorough understanding of an object of study can never be achieved by a single perspective (Chapman, 2012; Gendron, 2013). Audit quality research can be approached through different archival, experimental and fieldwork research (Gendron and Power, 2015). To my knowledge no field work has been performed in the governmental audit field in regard to audit quality. Performing this research from a qualitative approach will provide insight from a different perspective on the notion of audit quality and will hopefully lead to more understanding of the real-life settings in which assurance and auditing practices take place. Qualitative research is helpful for gaining more insight into the contextual information (Guba and Lincoln, 1994)
The Monitoring Commission Accountancy (hereafter: MCA) has recommended the accountancy sector to put more emphasize on the qualitative definition of audit quality next to the quantitative definitions (MCA, 2016, p.21). A case study into the perceived influence of a governmental audit organization on audit quality will contribute to this recommendation.
9 Previous literature discusses the reasons why a municipality might prefer a public accountant over a governmental accountant, but to my knowledge there is no earlier in-depth research upon this topic. Pursuing a qualitative research provides the opportunity to gain understanding about the reasoning behind assigning a governmental auditor while literature on this topic is inconclusive about the positive aspects of governmental auditors and leans towards preference of public auditors. As Malsch and Saterio (2016) noted “By definition, the concrete deployment of political influence and power relationships is difficult to capture through archival proxies or experimental settings”; I think therefore a qualitative research can provide us with more insight into this topic than performing another quantitative analysis on governmental auditors’ effect on audit quality would.
1.3 Legitimacy theory as a framework
To understand an analyse the results of the research field work, I will use the legitimacy theory. Embracing the legitimacy theory hopefully leads to more understanding of the motivation behind the actions pursued by the governmental audit organization. Legitimacy theory has previously been applied in numeral accounting studies to explain corporate social and environmental disclosure motivations, but legitimacy can be found in all actions undertaken by an organization. The base of the legitimacy theory can be found in the social contract between an organization and its environment. The framework and its application for this research will be further discussed in chapter 3.
10
2 Literature review
To answer the research question in this chapter relevant existing literature on audit quality and governmental auditing is provided. To gain insight into the meaning of audit quality and the aspects that are related to this subject a literature review is provided in paragraph 2.1. Auditing a municipality is a whole different ballgame from what can be seen as a tailored audit process for most private companies (Boxmeer, 2011). To understand the nature and structure of governmental auditing an introduction into the Dutch governmental system will be described in paragraph 2.2 followed by an explanation of the current applicable rules and regulations for governmental auditors. In paragraph 2.3 the research question will be presented.
2.1 Audit quality
The concept of audit quality is an in-depth studied topic (Francis, 2011; DeFond and Zhang, 2014;) and often appears in both academic research and the audit practice itself (e.g. van Buuren, 2015; PWC, 2017). There is however not one generally accepted definition. The complexity of the concept is frequently emphasized (Francis, 2011) and the definitions used in frameworks issued by industry standard setters such as the International Auditing and Assurance Standards Board (hereafter: IAASB) and Public Company Accounting Oversight Board (hereafter: PCOAB) often seem to take a different approach than academic papers when it comes to the proxies used to describe different factors that influence the audit quality. To gain a better understanding of audit quality and the factors that associated with the concept, in this chapter several well-known academic papers and definitions issued by the audit industry regulators will be analysed and compared. In subchapter 2.1.1 academic key papers that have provided different definitions for audit quality will be described. In subchapter 2.1.2 factors that influence audit quality will be described and analysed. In subchapter 2.1.3 frameworks including the influential factors provided by some of the industry standard setters will be explained and compared and lastly subchapter 2.1.4 provides a summary of the different viewpoints and details of audit quality that will reappear throughout the rest of this thesis.
2.1.1 Audit quality definition from an academic perspective
An often-cited paper and arguably one of the most common used definitions of audit quality is provided by DeAngelo (1981). She states audit quality as “the market assessed joint probability that a given auditor will both (a) discover a breach in the client’s accounting system, and (b)
11 report the breach” (p. 186). DeAngelo argues that audit quality is independent of different factors, including audit firm size: “the larger the auditor as measured by the number of current clients and the smaller the client as a fraction of the auditor’s total quasi-rents, the less incentive the auditor has to behave opportunistically, and the higher the perceived quality of the audit’’ (1981, p. 197). Audit firms with less clients are more reluctant to keep a client and this incentive might influence the auditor report and therefore the audit quality. DeAngelo reasons that when auditors have an economic interest or lack independence this influences the audit quality. In her paper she recognizes the influence of the used control procedure, auditor independence and auditor competence, three elements which are not directly measurable in real life (Knechel, 2009). Audit quality in DeAngelo’s definition is a binary concept (DeFond and Zhang, 2014) in which the output is either sufficient or insufficient.
A second definition of audit quality is described by Francis (2011). In his paper he states that audit quality is not a binary concept but should be interpreted as a continuum from low- to high quality audits (p. 127, 2011) something which is affected by different elements. These elements he included in a framework and the impact of these factors on audit quality are described. The factors Francis included in this framework are separated in six different categories, specifically: audit inputs, audit process, accounting firms, audit industry and audit market, institutions and lastly, economic consequences of audit outcomes (p.126, 2011). According to Francis (2011) audit quality is affected by all six drivers and research should be conducted for each unit of analysis. Several of these factors are expected to reappear in the research conducted for this thesis and their perceived influence shall be discussed in chapter 5 which provides the findings.
A third definition of audit quality can be found in the paper of DeFond and Zhang (2014). They describe higher audit quality as ‘greater assurance that the financial statements faithfully reflect the firm’s underlying economics’ (p. 280) Something which is also conditioned on its financial reporting system and innate characteristics. They describe the importance of recognition of the relationship between audit quality and financial reporting and emphasize that audit quality should be seen as ‘a continuous construct that assures financial reporting quality’ (p. 276).
12 2.1.2 Factors associated with audit quality
The supply of audit quality is a matured subject in auditing research (DeFond and Zhang, 2014) and different contributing factors are described in academic papers. In research the link between observable outcomes of the audit, such as auditor size and audit specialization and indicators of financial reporting quality such as issuance of a going concern opinion has often been examined. There is however, in line with a definition of the term audit quality, no consensus over which measures are the best indicators of audit quality. To give an impression of indicators that might reappear during the analysis of I will describe several of them as discussed in academic literature concerning audit quality, such as auditor knowledge, firm tenure and independence.
In their research, Defond and Zhang (2014) separate the factors that influence quality into client-incentive; factors which are driven by regulatory intervention and a clients’ pursuit to reduce agency costs (Jensen and Meckling, 1976), and auditor-incentive; factors that are a function of the auditor’s incentives for independence and competence (Defond and Zhang, 2014, p. 296). A Factor that drive the client-demand side for audit quality are not always directly observable and often research is limited to input measures such as auditor size, specialization and fees. Research focussed on auditor-incentives is often limited towards reputation and litigation risk and big N as a determinative factor. Another division made by DeFond and Zhang (2014) into two categories of the factors in into input-based and output-based quality measures. They describe how different input-based proxies such as auditor characteristics and auditor-client contracting futures have been used predominantly in archival studies. Output-based structures, such as the material misstatements and going-concern opinion, are constrained by innate characteristics and mostly appeared in experimental researches.
DeFond and Zhang (2014) emphasize the importance of including several proxies in quantitative research to understand the concept of audit quality. Proxies are needed because audit quality is a multidimensional subject, no definition is agreed upon and subsequently there is no way to directly measure it. Quantifying operational variables is an essential part of quantitative research and many researchers have used several proxies and other indicators to attempt to measure their influence on audit quality. However, not all influences on audit quality are directly measurable.
Auditor independence is described as a foundation for audit quality (Grasso and Sharkansky, 2001; DeAngelo, 1981). Grasso and Sharkansky (2001) stated: “The auditor’s
13 opinion on the financial condition of a corporation derived not just from technical expertise, but also from the fact that the auditor stood outside the corporation itself.” (p.2). This importance of auditor independence is underwritten by López and Peters (2010). They investigated audit report differences between governmental auditors and certified public accountant firms. They performed a research in which they conclude that private sector accountants are more likely to issue auditor reports that identify internal control concerns, complementing other research on perceived audit quality differences. This research conducts to the belief that independence leads to a higher audit quality.
Deis and Giroux (1992) state that reputation and power conflict are two of the variations in audit quality that explain a difference in audit quality level. These variations are expected to impact auditor independence for governmental auditors as will be described in this thesis.
Audit quality is also possibly affected by audit rotation. Elder et al. (2015) describe in their paper the effects of audit rotation on municipalities, stating that rotation policies are associated with higher audit quality. For the municipalities audited by governmental audit organizations, seemingly no audit rotation policies are installed. For municipalities in the Netherlands at large there is no law on obligated audit rotation.
2.1.3 Audit quality from a standard setters’ perspective
Several of the definitions and factors provided by academic research reappear in the frameworks provided by standard setters such as the IAASB and the PCOAB. Auditor independence is also described as an important factor in the accountant code of ethics. The audit quality definitions issued by these standard setters influence the way organizations and accountant firms deal with the concept (van Buuren, 2015). Gaining an understanding of their frameworks can therefore help into understanding the perspectives given in the case study.
The IAASB is an independent standard setting body that has been issuing since 1978. They issue standards on auditing, quality control and other services to support the auditing of financial statements. The IAASB issued a framework for audit quality in 2014. Audit quality is defined in their framework as a ‘an environment which maximizes the likelihood that quality audits are performed on a consistent basis’ (IAASB 2014, p.4). This quality audit is influenced by factors such as sufficient knowledge, useful and timely reports and appropriate interaction with relevant stakeholders. In their framework the different factors are grouped into four categories: inputs, outputs, process, interaction and contextual factors.
14 Inputs are the values, ethics and attitude of an auditor in combination with the
knowledge, skill and performance. The different input factors are further organized at
engagement, audit firm and national level. Process is characterized by the strictness of the audit process and control procedures. Several quality attributes are given in the report and include things such as use of a proper methodology, review of audit work and use of information technology. The factors are again organized at the engagement, audit firm and national level. Outputs include the reports and information that are presented and output from the audit that will be used internally in the audited firm. These factors are often determined by the context and legislative requirements. Key interaction entails that the way in which the stakeholders from an audit interact can have an impact on the audit quality. These communications can be both formal and informal and are mostly expected between management, regulators, auditors, users and those charged with governance. The last factor included in the framework is contextual factors, the environmental factors that include audit quality such as laws and regulations, corporate governance and the place where the audit takes place (IAASB 2014).
The NBA is the Dutch professional organization of accountants. They are responsible for promoting right professional practice by accountants by establishing professional regulations, promoting the common interests of accountants, ensuring the honorary state of the profession and taking care of practical training based on the Dutch law on the accounting profession (Wab, art. 33). In 2017 they published a green paper to promote the discussion of the term audit quality.
In this paper they discuss the term from a multi-stakeholder perspective. This perspective argues in favour of a broader conceptualization of audit quality, taking into consideration four perspectives: societal, supervisor, the audit client and the audit team (Bik, 2016, p. 371).
From a societal perspective the importance of the signalling function of auditing is emphasized. Auditing should not only provide assurance over the statements but assist in timely identification of important topics (NBA, 2017 p.7) The supervisory perspective emphasizes the importance for accountants to carry out their responsibility to explain how an audit was conducted and substantiated by professional judgement. From a client perspective added value is described, which entails a thorough process, no surprises and low fees. Accountants can add value to the comprehensive auditors’ report by providing recommendations based on the audit (NBA, 2017, p.7) Lastly, audit quality is defined by the input from the accountant and the
15 accountant organization. This is impacted by how accountant set up and carry out the audit as well as tooling available to support the chosen methodology. Factors included are partner involvement, team discussions, process management and thorough risk analysis (NBA, 2017, p.7)
The International Federation of Accountants (hereafter: IFAC) is a global organization that focuses on improving the accountancy profession. They support developments in standard-setting by supporting four standard-standard-setting boards, including the aforementioned IAASB, and issuing guidance reports. In 2010 an independent standard setting board was established under IFAC, the International Ethics Standards Board for Accountants (hereafter: IESBA). IFAC created a code of ethics in 2005 and revised standards have been issued by IFAC and IESBA since. An important part of their ethics code are the auditor independence requirements. By the IESBA Codes of Ethics auditor independence gets separated into two aspects: independence in mind
and independence in appearance. The definitions given by the code (2016, p. 46) are:
Independence in mind: The state of mind that permits the expression of a conclusion without being affected by influences that compromise professional judgment, allowing an individual to act with integrity, and exercise objectivity and professional scepticism
Independence in appearance: The state of mind that permits the expression of a conclusion without being affected by influences that compromise professional judgment, allowing an individual to act with integrity, and exercise objectivity and professional scepticism.
Independence is therefore acknowledged as an important factor in audit ethics and as an influential factor of audit quality by standard setters.
2.1.4 Summary of audit quality
There is no generally accepted definition of audit quality and there are many factors that might influence the concept. Audit quality has been described both as a binary concept and a continuum. Several factors that are believed to influence audit quality have been described by both standard setters and academic researchers include factors analysed from an auditor’s perspective, such as independence, audit tenure, audit size and audit specialization, and factors analysed from a client’s perspective, such as added value and fee. Audit quality can, and should, be interpreted from different angles.
16 2.2 Local governmental auditors
A governmental auditor is defined by its relationship with the State, a municipality or other governmental service that it is serving. Most of the audits are performed by either independent external auditors or internal auditors (Hayes et al., 2014, p.15). Governmental auditors take both the function of internal and external auditor (Hayes et al., 2014, p. 33). They are a department of their client but issue the independent accountant statement of the financial report.
The Netherlands has had an history of a governmental auditors for municipalities. The oldest accounts from local councillors that are still known and intact date from 1308. Audits in those ages were performed by accountants that worked for the municipality. Municipalities have been appointing related governmental audit organizations since at least 1890 (GAD, 1991) but this has changed since 1987 when acquisition of VB Group, the national leader on governmental auditors, by Deloitte took place. Since then the market share of the Big-4 auditors in governmental auditing has increasing, covering a market share of 70% in 2015 (BMC, 2016). In 2006 the Municipality Law (Dutch: Gemeentewet) was implemented. This law still leaves option for the appointment of a governmental auditor in article 213:
Accountants as referred to in the second paragraph can be appointed in a municipal service and in that case are appointed, suspended and dismissed by the Board.
Having the option to appoint a governmental auditor is not something that is going unnoticed in the Dutch newspapers and outlets due to the controversy. A governmental auditor is discussed by some to be less independent due to their relationship with the municipality, being an internal department (Boxmeer, 2011).
Currently there are only two municipalities left where a governmental accountant organization is appointed for the auditing of the financial statements; Amsterdam and the Hague. Assigning a governmental auditor might be a solution for the smaller municipalities to appoint the necessary auditor, especially since the Big-4 market share in governmental auditing is decreasing (FD, 2016) and the complexity of financial audits at local governments in the Netherlands is ascending with the delegation of more tasks (Ministerie van BZK, 2016). Smaller municipalities have trouble finding an accountant organization that is willing to perform the audit (BMC, 2016).
Governments differ substantially from private sector companies in certain aspects (Samelson et al. 2006, p. 140). The complexity of local governments is often attributed to the
17 different laws and regulations to which a municipality needs to apply. The Municipality acts states that Dutch municipalities every year need to compose budget and accounting documents. Municipalities do not act in accordance with common laws and regulations for financial statements but are assigned a different law, the Act Budget and Accountability (BBV).
2.2.1 Current governmental auditing rules and regulations in the Netherlands
At present there are 380 municipalities in the Netherlands of which two have a related governmental audit institution, the municipalities of Amsterdam and The Hague. Appointing and establishing a governmental auditor instead of appointing an independent public auditor has always been an option in the Netherlands. The Municipality law states that it is mandatory for local governments in the Netherlands to assign an auditor who conducts an audit on the financial statements, but the option in accountant organization has been left free, restricted to guidelines that don’t require the accountant organization to be external. Not conforming to a yearly audit of the financial statements leads to a discount on the governmental budget of the municipality which is assigned on a yearly basis by the Dutch government, or a reduction in other municipal allowances.
Based on the Dutch supervision act ‘Wet Toezicht Accountantscontroles’ (hereafter: Wta) there is independent oversight board that supervises all the auditors who perform legally required audits. For public audit firms this supervision role is performed by the Dutch Authority for the Financial Markets (hereafter: AFM). The AFM does not supervise municipalities who have a governmental auditor assigned. In the case of the governmental auditors involved with the audit of the municipalities of Amsterdam and The Hague the supervision is assigned to the local audit councils. This is in accordance with the law stated in the Gemeentewet and Wta.
Municipalities and the governmental audit organizations are governed by laws and regulations. For assurance services all auditors in the Netherlands have to comply with the Wta and Regulation on the independence of auditors at assurance assignments (Dutch: Verordening inzake de onafhankelijkheid van accountants bij assurance-opdrachten, hereafter: ViO). The actual audit should be performed in accordance with the Dutch audit standards ‘Nadere voorschriften Controle- en overige standaarden’ (hereafter: NV COS) issued by the NBA. For the financial statements the municipalities are obligated to apply to the rules and regulations stated by the Budget and Accountability Legislation (Dutch: Besluit Begroting en Verantwoording, herafter: BBV) and the Decree on the audit of decentralized authorities (Dutch:
18 Besluit accountantscontrole decentrale overheden, hereafter: Bado). The governmental accountant audits whether the financial statements meet these requirements.
2.2.2 Differences public and government auditors
There is a difference in the perception of auditor independence for public and government auditors, as researched by Karns et al. (1983). They state that private auditors were preferred in a survey they conducted for American households, because private sector auditors are perceived as being more independent. Aligning information is given by Gendron et al. (2001) who state that concerns have been raised about the independence of state auditors in Canada (p. 279). The reasons for these concerns are the risks involved with reviewing effectiveness by the state auditor. In their reviews they would inevitable encounter problems that are not administration, but politics related, and challenging political strategies is a risk due to the needed neutrality and independence in reviewing.
The study of Cagle and Pridgen (2015) states that a significant difference is presented in the number of audit findings reported upon by public sector and private sector accountants on governmental audit engagements (p.91), they elaborate that this might be due to a difference in reputational concerns and client pressure. Client pressure for a governmental auditor is different; a governmental auditor is assigned to audit the financial statements of the related local government and this is their main concern, public auditors operate on a greater level serving multiple customers with different backgrounds.
Rubin (1992) states that the existence of a governmental auditor leads to additional issues. He discusses the differences between audit fees charged by state and private-sector accountants.
In contrast to the information given above, López and Peters (2010) state there is anecdotal and empirical evidence that indicates that governmental audits performed by governmental accountants are of higher quality due to the unique challenges and incentive mechanisms associated with governmental auditors (p. 483). They gain more in-depth insight and expertise due to the specialist nature of the audit.
Samelson et al. (2006) discusses some of the determinants for audit quality at local governments. They performed a survey and discovered how perceived audit quality at local governments is influenced by auditor expertise client responsiveness, professionalism and understanding of client systems.
19 2.2.3 Summary of governmental audit organization
Governmental audit organizations are different from public audit firms. The difference can be found in the rules and regulations the firms must apply to as well as the rules their clients have to apply to, the relationship between the accounting firm and the client and the oversight board that is responsible for overseeing the accounting organization. Not everything is different, like public audit firms, governmental audit firms have to apply to independency rules and regulations stated in the Wta and ViO.
2.3 Research question
The possible different perceptions in audit quality and the recent development of the rules and regulations in the Netherlands regarding governmental auditing motivate the development of the research question for this thesis:
What is the perceived influence on the audit quality of the financial audit reporting at a local government in the Netherlands when a governmental auditor is appointed?
Given the existing literature upon audit quality and public and governmental auditing there is a reason to suspect that the perceived audit quality might be different when a governmental auditor is assigned in comparison with a private accountant. A governmental auditor is adjunct to the client they serve and therefore can have independency issues. This belief would be at par with the report of the commission BBV, which states that it depends on the municipality whether they choose between a public or governmental auditor and that this choice should not have impact on the quality (p.26). There are different municipalities who have shown interest in the establishment of a governmental auditor due to the developments in the public auditing firms and governmental auditing (VNG, 2014). The establishment of these auditors would solve the issue that some municipalities are not able to find an accountant willing to do the audit (BMC, 2016).
Gendron et al. (2001) express their worry about the lack of research about state auditor independence: “In spite of the importance of these concerns with regard to the legitimacy of the state auditors' mandate, there are few empirical studies that have examined the extent to which state auditors behave in a way that is consistent with the socially accepted boundaries of independence” (p. 279). Research of Lowensohn et al. (2007) suggests that local governments might be well served by engaging specialized auditors. Although governmental audit
20 organizations might be perceived as less independent, they are often associated with high levels of knowledge and specialization (Gendron, 2001).
Concerns about the independence of state auditors is not unique to the Netherlands. In earlier research Power (1997, p.51) also expresses this, stating that state auditors inevitably encounter issues in their claim of neutrality from policy-making. By performing a case study at a Dutch municipality, I hope to gain more insight into this research topic by observing the
perceived influence of being a governmental auditor on topics such as independence.
I will perform a qualitative research in which the perceived influence on audit quality is analysed for a municipality that has a governmental auditor. Answering this research question can lead to insight in whether the perceived audit quality of a governmental auditor differs and what factors are contributing to this. It will add insight in whether the quality of the finical audit is perceived differently for aligned audit firms or public audit firms and adds to the information about the interaction between independence and auditor specialization and their relationship with audit quality.
21
3 Theoretical Framework
To understand whether the audits of a governmental auditor are perceived as of sufficient quality and whether this differs from the perceived audit quality of public auditors the legitimacy theory will be used as a theoretical framework for the case analysis. In research as of lately the legitimacy theory has been used mostly to describe the extension of auditing into new assurance fields (Power, 2003, p. 387; Deegan, 2007) or to explain corporate motivation behind disclosing social and environmental information (Deegan, 2014, p. 248). However, based on the theoretical underpinning of the method I think it will provide a useful framework in this research. Legitimacy is not only applicable for new audit fields but can be seen as a condition or status which should be obtained throughout the existence of an organization (Deegan, 2007, p.128). The legitimacy theory finds its origin in sociology. Another reason to apply the legitimacy theory is the under-documented and under researched role of auditing in providing legitimacy (Power, 2003, p. 393). Before applying the legitimacy theory to the case study I will describe the legitimacy theory in this chapter. I will do this by describing a definition for legitimacy, the underlying construct, the different types of legitimacy and the ways in which it can be obtained. Lastly, I will illustrate how the legitimacy theory will be used to analyse the case study.
3.1 Defining the legitimacy theory
Legitimacy is a relative concept, it depends on the social system in which an entity operates and is subjective to the time and place in which it takes place (Dowling and Pfeffer, 1975; Deegan, 2007). A broad-based definition of legitimacy that incorporates this was adopted by Suchman (1995, p. 574):
“legitimacy is a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions”.
Suchman describes how legitimacy research often can be distinct into two groups, strategic and institutional research. In his paper he explains: “Work in the strategic tradition, adopts a managerial perspective and emphasizes the ways in which organizations instrumentally manipulate and deploy evocative symbols in order to garner societal support. In contrast, work in the institutional tradition adopts a more detached stance and emphasizes the ways in which sector-wide structuration dynamics generate cultural pressures that transcend any single organization’s purposive control” (Suchman, 1995, p.572). This distinction in strategic and
22 institutional legitimacy is backed up by Kaplan and Ruland. They remark about strategic legitimacy that “Underlying organizational legitimacy is a process, legitimation, by which an organization seeks approval (or avoidance of sanction) from groups in society” (Kaplan and Ruland, 1991, p. 370).
Suchman (1995) emphasizes how these approaches are a matter of perspective and how it is important to consider both viewpoints. Legitimacy acts both as a resource and a taken-for-granted belief system (Swidler, 1986). For research, it is important to consider the environment and timeframe in which a research takes place, however most legitimacy research is informed by the strategic (or managerial or operational) perspective (Deegan, 2007).
Organizational legitimacy can be seen as a resource that is necessary to survive for an organization, as Deegan and Unerman (2006, p. 271) formulate:
Legitimacy theory asserts that organizations continually seek to ensure that they are perceived as operating within the bounds and norms of their respective societies, that is, they attempt to ensure that their activities are perceived by outside parties as being legitimate.
A distinction is made by Lindblom (1994) between legitimacy, which she considers to be a status or condition, and legitimation, which she describes as the process that leads to an organization being found legitimate. Legitimation processes aim to justify the work that is undertaken and how it is undertaken (O’Dwyer et al., 2011) The legitimacy theory has been used by researchers in attempts to explain the acts undertaken by an organization in order to be deemed legitimate.
3.2 Social contract and legitimacy theory
It is not the actual conduct of an organization that is important in order for it to be perceived as legitimate, but what society knows or perceives about an organization’s conducts (Deegan, 2014, p.249) The theoretical construct known as the social contract is a central premise of the legitimacy theory. Social contract is used as a concept to represent the multitude of implicit and explicit expectations society has about how an organization should operate (Deegan and Unerman, 2006, p. 271). An organization can only maintain operations to the extent they are supported by the community (Deegan, 2007, p.133) because society is most likely to supply resources to organizations that appear desirable, proper or appropriate (Parsons, 1960). The societal expectations are considered to form a social contract. The social contract is theoretical construct, and perceptions about its attributes vary. Organizations will undertake various actions
23 to establish congruence between the values associated with the social contract and the values associated with or implied by their actions (Dowling and Pfeffer, 1975, p. 122). An important factor of the social contract is that is it not necessary for he organization to act in alignment with the social norms, as long as their behaviours are perceived as converging with the societal norms. As Suchman stated: “Legitimacy is dependent on a collective audience, yet independent of particular observers.” Legitimacy is not an abstract measure but rather a measure of societal perceptions (Suchman, 1995) and can range from highly legitimate to highly illegitimate (Nasi et al., 1997, p.300).
3.3 Types of legitimacy
Suchman (1995) proposes three types of legitimacy: pragmatic, moral and cognitive (p. 577). Which type of legitimacy is desirable depends on the social construct. The three types co-exist in and tend to merge in real life settings (Brinkerhoff, 2005). Each type rests on a somewhat different behavioural dynamic (Suchman, 1995, p.577). A schematic overview of the types composed by O’Dwyer et al. (2011) that will be discussed hereafter is given in appendix 5.
Pragmatic legitimacy is subordinated to the perceived self-interest of an organization’s most immediate audience (Suchman, 1995; O’Dwyer et al. 2011). This type of legitimacy can take the form of exchange legitimacy, a form in which support for an organizational policy is based on the value of the policy to a particular set of constituents (Dowling and Pfeffer, 1975). A related type is influence legitimacy, in which constituents support organizational policies because they are deemed to be responsive to the constituent’s interest at large, in oppose to their own interests (O’Dwyer et al. 2011). The last variant of pragmatic legitimacy is dispositional legitimacy, which is secured by prompting the constituents that the organization has their best interest at heart.
Moral legitimacy rests on the constituents’ judgement whether something is the right thing to do. This is defined by the audiences constructed value system (Suchman, 1995). It can take four forms. The first is consequential legitimacy, which involved judgement based on the accomplishments, the output and consequences of an action. The second is procedural legitimacy, in which organizations are judged based on social status of their adopted techniques and procedures. This type of legitimacy becomes most significant in the absence of clear outcome measures (Suchman, 1995, p. 580). The third form is personal legitimacy, one that rests on the charisma and status of an individual organizational leaders. The last form is structural
24 legitimacy, a form in which organizations are deemed worthy because their structural characteristics relate to a social constructed favoured taxonomic category (Suchman, 1995, p. 581).
The last type of legitimacy, cognitive legitimacy, is obtained when organizational policies are taken for granted as being appropriate, proper and desirable (O’Dwyer et al., 2011). It operates mainly at the subconscious level. Cognitive legitimacy can be divided into two variants, comprehensibility and taken-for-granted. Comprehensibility is dependent on popular available explanations for the existence of practices being available. In the presence of such explanations organizational practices will be seen as predictable and meaningful. Taken-for-granted legitimacy represents the most subtle and powerful source of legitimacy (Suchman, 1995, p.583; O’Dwyer et al. 2011, p. 37) It depicts actions as being given, deeming alternatives to these actions unthinkable. Taken-for-granted status generally lies beyond the reach of most but has been obtained by certain technologies and policies.
3.4 Legitimacy strategies
To gain legitimacy and remain legitimate an organisation can undertake several actions. Dowling and Pfeffer (1975) have outlined the resources organizations can use to legitimate their activities. They can adapt to conform to the prevailing definitions of legitimacy, attempt to alter the definition so that the organizational behaviour conforms to the social standard or they can attempt to become identified to the social norm through communication (p. 127).
In line with the strategies described by Dowling and Pfeffer, Lindblom (1994) describes the four strategies which an organisation can undertake in order to gain legitimacy. The first strategy is to change its actual performance and activities to conform to the expectations of the key stakeholders and educate and inform this audience about this change; the second strategy is to not change its behaviour but change the perception of the key stakeholders in order to convince them that there has been a change matching their expectations; the third strategy is to not change its behaviour and not inform but deflecting the attention of the key stakeholders to other activities, a form of manipulation; and lastly the fourth strategy is trying to change the key stakeholders expectations in order to match the organizations current behaviour. The strategies proposed by Lindblom overlap with those defined by Dowling and Pfeffer but a manipulation strategy which works as a distraction from the actual organizational activities was added.
25 Suchman (1995) described how legitimacy management relies highly on communication. He distinguishes three challenges of legitimation: gaining legitimacy, maintaining legitimacy and repairing legitimacy (p. 586) and provides three extensive strategies to confront these challenges at the given different levels of legitimacy. O’Dwyer et al. (2011) noted how the strategies described by Suchman aimed at securing the legitimacy fall along a continuum and can be seen as ‘relatively passive conformity’ to ‘active manipulation’ (p.36). According to Suchman an organisation might conform to the environment, select its environment or manipulate its environment. The favoured action changes according to the type of legitimacy an organization is trying to establish.
The first strategy provided by Suchman (1995) is the conform strategy. The easiest example of a conform strategy is when an organization seeks to position themselves within the existing social standard. Conform strategy to gain cognitive legitimacy can be pursued by conforming to well established standards (O’Dwyer et al., 2011, p.37). The second strategy Suchman (1995) describes is the selection strategy. In this strategy an organization wishes to avoid changing their actual actions to gain legitimacy. The simplest way to do this is by selecting an environment that will grand the organization legitimacy (Suchman, 1995, p.589). O’Dwyer et al. (2011) noted “most practices will attain legitimacy through carefully conforming and selecting their environments.”. There is however, a third strategy: the manipulation strategy. When conforming and selection do not suffice, organizations must actively promote alternative explanations of social reality. Again, pragmatic legitimacy is the easiest form to obtain with this tactic. Examples of the manipulation strategy are new product advertising or showing a record of success in hopes of spill over effect on other moral dynamics (Suchman, 1995, p. 592). A schematic overview of the different legitimacy strategies, prepared by O’Dwyer et al. (2011), is given in appendix 5.
3.5 Auditing as a social construct
An important contribution to our understanding of the social construct of the audit process is given by Pentland (1993). By describing a field observation of two audit engagements he tries argues how the micro level interactions of an engagement team contribute to the macro level role of audit. He does this by describing the interaction rituals embedded in auditing. According to Pentland this leads to understanding of the social construct of auditing, including things as professionalism, independence and trust (1993, p. 606). Power observes how Pentlands’ research
26 proves the importance of the role of image management, an example of a legitimation method, in the audit process (2003, p. 386). He argues that micro-productions of legitimacy such as usage of a certain language, are an intimate and powerful component of auditing. Based on these papers it can be argued that legitimacy can tried to be obtained in many ways. The legitimacy theory will be used to explain and understand the acts undertaken.
3.6 Use of legitimacy theory
Aspects of the legitimacy theory, such as the types of legitimacy and the strategies used to gain legitimacy, will be used to analyse and understand which strategies the governmental auditor undertakes to remain legitimate for the activities involved with the financial audit. This framework is a useful instrument to examine the processes by which the auditor tries to gain and obtain his legitimacy and whether and how he delivers on the requested audit quality.
A theoretical framework forms an abstraction of reality and cannot be expected to provide a full account of description of particular behaviour (Deegan and Unerman 2006, p.268). The legitimacy theory is a positive theory, which is used to explain why certain actions are taken, not what actions are most desirable (Deegan, 2007) The legitimacy theory can be identified as a system orientated theory. In accordance with Deegan (2007, p. 129) this can be explained as: Within a system-based perspective, the entity is assumed to be influenced by, and in turn to have influence upon the society in which it operates.
With the use of a system-oriented perspective we can focus on the role of information and disclosure in the relationship between the involved organizations, as stated by Gray et al. (1996).
Legitimacy theory is not without its weaknesses. The concept of legitimacy is often broadly defined (Free et al. 2009). This broad nature makes it possible to claim many actions undertaken by management as being part of their legitimacy strategy. This lack of refinement and adequate definition (Suchman, 1995) has put up legitimacy theory as a theoretical structure that has been used as ‘a blind man’s hammer (Hybels (1995, p. 241).
Legitimacy theory cannot be easily distinguished from other well-known theories such as the Stakeholder theory, Institutional Theory and the Positive Accounting Theory (Deegan and Unerman, 2006, p. 284; Deegan, 2014, p. 252). Compared to the Positive accounting theory, organizations according to the legitimacy theory rely more on the social contract with society and not all actions are driven by individual self-interest. When it comes to the Stakeholder theory,
27 some academics discussed that the legitimacy theory and Stakeholder theory should not be treated as two individual theories. Gray et al. (1995) stated that Stakeholder and Legitimacy theory should be seen as two overlapping perspectives of the issue which are set in within a framework of assumptions about political economy (p. 52). Stakeholder theory considers a more refined resolution by referring to particular groups within society as different as opposite to legitimacy theory where society is often defined at large (Deegan and Blomquist, 2006). The Institutional theory provides a complementary view to the legitimacy theory in understanding how organizations respond to changing social and institutional pressures and expectations (Deegan, 2014, p. 386). Arena et al. (2006) identified three external forces that impact both individuals and organizations: laws and regulations, choices of other organizations and consultation of professional bodies. Institutional theory has been adopted often in internal audit studies (Endaya and Hanefah, 2013). Legitimacy theory provides a more focused perspective when compared to the richness of institutional theory but might be losing in popularity in upcoming research compared to institutional theory (Deegan, 2014).
As Malsch and Salterio (2016) noted: “Importantly, no single theoretical perspective can account for the complexity and the subjective perceptions of people involved in the construction of their social environment.” (p. 27). The use of legitimacy theory should be seen as a framework to gain a more in-depth understanding but is not the only theoretical viewpoint that can be adapted to interpret and understand this case, and the other theories should be considered.
O’Dwyer et al. (2011) provide a more refined concept of legitimacy in their paper on sustainability reporting. In this paper they take the three types of legitimacy; pragmatic, moral and cognitive legitimacy and the three strategies for establishing legitimacy; conform, selection and manipulation strategy as provided by Suchman (1995) and add to this theory by describing how different strategies in which legitimation for new practices can be undertaken to gain legitimacy differ with the different audiences. They identify three key audiences: the outside world, also noted as clients, the potential report users, also noted as non-client external world and the inside Risk department, noted as the inside world (p.40). Hybels (1995) also argued how good models in legitimacy must examine the different stakeholders involved (p. 243). Developing a comparable distinction in the key audience might be helpful in the case analysis and combines aspects of the stakeholder theory with the legitimacy theory. To achieve this, I will conduct interviews with people from different backgrounds to gain a multi-perspective insight.
28 I will draw upon the frameworks as described above to examine and analyse the governmental auditor developments and process undertaken to provide sufficient audit quality in their role as the municipality auditor. Legitimation processes are used to understand both the process of activities and their outcome (O’Dwyer et al. 2011).Before presenting this examination an outline of the research method will be given.
29
4 Research methodology
In order to answer the research question an interpretive case study was performed. To gain access to the right people and documents I contacted a governmental auditor commissioned with the role of financial auditor for a municipality in the Netherlands. This governmental audit organization was selected because they are one of the two governmental audit organizations who still perform the audits for a municipality in the Netherlands. During the writing of this thesis I conducted an internship at their firm for four months, to gain access to the right people and get a better insight of the work they provide for the municipality. Through the internship I approached employees from the municipality working at several departments, members of the audit council and the civil council member who are involved with financing and the performance of the financial audit. A series of semi structured interviews was held with people are involved in the governmental audit field. For this research eleven interviews were conducted, and various sources of relevant information such as papers, reports and audit documentary have been analysed and compared. Next to the conducted open-interviews during the months that I spend doing the internship at the audit firm I also spoke with employees of both the municipality and the auditor on an informal base, to gain a better understanding of the culture and work they conducted. The research is supplemented by documentary evidence to understand, support and contextualize the findings.
4.1 Case context
The Netherlands has a rich institutional history and currently remains as a constitutional monarchy since 1815. Most contact between the public and the government is on a local level, at one of the 380 (2007: 388) municipalities. There are three levels of governmental organization in the Netherlands, municipalities being the primary level of government most citizens have contact with (VNG, 2007). They have a certain amount of autonomy in creating and implementing policies, taking decisions about local subjects such as public housing and transportation, and thus maintaining a central position in the everyday life of Dutch citizens. In every municipality there is a municipal council representing the public and supervising the municipal executives. They are chosen every four years by elections. At the municipalities, public officials carry out the daily tasks as ordered by municipal executives and the major. They perform the work that must be done but are not a part of the political debate.
30 The financing of local government consists of different sources of income, of which the most significant remains the grants of the national fund received yearly by all municipalities. This consists of an Earmarked fund and a General Grant. To receive these funds, it is mandatory for local governments to perform a financial audit yearly for which an auditor report has to be obtained. This audit report is mostly performed by external private accountant organizations, with Deloitte being the leading one in conducting governmental audits, but throughout the history there have also been governmental accountant organizations appointed to perform this task.
Currently only Amsterdam and the Hague have a governmental auditor, both cities play an important role in Dutch politics, one being the capital of the Netherlands and the other being the residence of the Head of State. Because there are currently only two governmental audit organizations, selection of one of these organizations was inevitable. For the research conducted, further descriptions of the governmental accounting firm are code-named for anonymity purposes. The governmental audit organization at which this research was performed will be described as GAX for the rest of this research. The municipality for which this organization is performing the audit will be coded XAX.
GAX was founded in 1961 (GAD, 1991). They have been appointed as the independent auditor to conduct the auditing of the financial statements of the municipality since then. Appointing the auditor is not a yearly process but is dependent on the discussions brought up in the city council. GAX has been appointed as the current signing auditor at the municipality that fulfills the task of the accountant of the municipality financial statements lastly in 2014 (Council decision).
Within GAX there are three job levels: audit manager, auditor, audit contributor; and multiple levels of seniority. Within the firm several teams are constructed to perform the audit on the different organizational flows within the municipality. Next to the external GAX provides, they also have internal projects, such as the team working on the creation of a knowledge management vision. In 2014 there has been a reorganization of GAX as the result of a change in the financial procedures conducted by the municipality. Based on Dutch law the local audit council RAX is assigned with supervising GAX for the execution of the statutory audit.
31 4.2 Research method interviews
To gain proper insight into the different notions of influence of governmental auditor on audit quality I conducted eleven interviews with twelve persons from the municipality and the governmental audit organization. There were ten one-on-one meetings conducted and one group interview on request of the interviewees, due to time and their complementary points of view. The purpose of the interviews is to gain a deeper insight into the perceived influence on audit quality of governmental auditors. There are about 50 employees at GAX auditors and numerous persons involved at the municipality XAX, including the local audit council RAX consisting of 15 persons, 45 city councillors, and many public officials organized into different departments. The interviews consisted of semi-structured open-ended questioning, to leave room for related viewpoints, questions or to pursue a topic into greater detail. A list of examples of the themes that have been discussed during the interviews and that served as a guideline for the interviews is included in the appendix 4.
The interviewees were selected based on their involvement in the auditing process. For the interviews I have spoken with people from four different perspectives to observe the influence on audit quality from different perceptions. Audit quality should not only be discussed by accountants involved as the NBA Green paper (2017) pointed out, but also from other perspectives. In the end the group of participants consisted of three members of the municipality city council, two members of the oversight board, four accountants from the governmental audit organization and three employees/civil servants of the services that are audited and are part of the municipality services. The number of participants was allocated and selected based on the amount of meaningful insight gained during previous interviews. When I gained the perception that performing more interviews in a specific group would not lead to more knowledge and understanding of their perspective another group was selected. The participants were approached by email, and interviewees were selected due to their involvement in the audit process. Selection of the interviewees was ad random. For triangulation, persons from different backgrounds and job level were selected.
The interviews were all conducted in person and a recording has been made on approval of the interviewee of the interview. All interviewees were informed that the recordings were not to be shared with another party. There was no interviewee who objected to recording of the interview. Locations for the interviews were selected to be free from distractions at places and times that were most suitable to the interviewees. All interviews were held indoors behind closed
32 doors. During the interview field annotations of the information have been taken. This assisted in further development of interviewing techniques, new subjects and was of assistance in the data analysis process. After the interviews all the information discussed was transcribed verbatim with the use of the recording. The transcripts of the interviews and field notes were shared with the related interviewee to protect against and mitigate bias and interpretation differences and to possibly validate the data. Possible modifications and/or changes due to remarks by the interviewee on the notes have been processed before analysing the information.
During one of the interviews the audio recording was corrupted which resulted in a missed recording of approximately 15 minutes of the interview, something which was discovered directly after the interview. Because of planning difficulties, the interview was not retaken. Instead the recording that was accessible has been transcribed as described earlier, and the last approximately 15 minutes that went missing have been reconstructed using the notes taken during the interview. This process took place directly after the interview. The result of this transcript and notes has been shared and discussed with the interviewee and based on this information enough material was available to include the data in the research.
4.3 Research method documents
Public council reports can be reached on the website of all Dutch municipalities. For the municipality XAX an online search engine is available at which the reports and related documents are provided. I have researched the documents by looking at all available information for the years 2015, 2016, 2017 and 2018 using several keywords related to the performance of a financial audit by the governmental auditor. Access to other relevant information and documentation has been gained through the internship. Information from the publications issued by XAX and GAX which support and contextualize the research is described in the findings. Not all documents are included in the literature overview due to their influence on the anonymity of the audit organization and municipality and their confidentiality.
4.4 Research method data analysis
After the interviews the obtained data was analysed in a structural manner. In order to understand and analyse the information gained from the interview transcripts and field notes, observation coding has been used to label themes and categorize the data into topics that are discussed throughout the multiple interviews and other documents, in accordance with Saldaña
33 (2013). Coding is a method that enables you to organize and group similarly coded data into categories or “families” because they share some characteristic (Saldaña, 2013, p. 9)
Coding was performed during and after the collection of data (Miles and Huberman, 1994) This analysis assisted in developing topics to discuss in additional interviews. The initial themes were established based on the interview guide but through thorough subsequent reading of the transcripts and accompanied notes there was refinement in the identification of key themes. This was done using data reduction, data display and conclusion drawing/verification (Miles and Huberman, 1994; O’Dwyer 2004). The coding method used to identify the themes was descriptive coding (cf. Miles and Huberman, 1994; Saldaña, 2013). Themes with broader patterns of meaning were grouped and categorized through mind mapping and merging initial themes. This helped in identifying the broad topics into different segments.
Subcoding was later used to gain a more enriched insight, by categorizing the data into different segments based on the position of the interviewee. Subcode is a second-order tag assigned after a primary code to detail or enrich the entry, depending on the volume of data you have or specificity you may need for categorization and data analysis (Miles & Huberman, 1994, p. 61).
During the coding attention was paid to contradictive opinions among interviewees (Patton, 2002) A summary table has been prepared where the coding of each transcript has been displayed. This summary table can be found in appendix 3. Based on this analyzation of the data, by data interpretation (O’Dwyer 2004) answers to the research question has been formulated in the next chapter.
