A tense coexistence: how independence and client knowledge interrelate for a higher audit quality

(1)

A tense coexistence: how independence and client

knowledge interrelate for a higher audit quality

A paper on audit quality threatening behaviour

Master thesis Msc Accountancy & Controlling

University of Groningen, Faculty of Economics and Business

Date: 21-6-2020

Rick Jan van der Schuit

S3146707

It Trekskip 15

9035GH Dronrijp

Tel +31 6 14059260

Email: R.j.van.der.schuit@student.rug.nl

Word count (excluding references): 8500

Supervisor

(2)

1

Abstract:

Auditors are expected to carry out their audit work with a skeptical mindset towards their clients. On the one hand auditors need to have an in-depth understanding of their clients, on the other hand auditors also need independence of their clients to provide the appropriate audit quality. In this regard, both positive and negative views on auditor client tenure and their effects on audit quality exist. Some research highlights the importance of independence, which may decrease as auditors serve the same client for a long time. Other research stresses the beneficial effects of having in-depth client knowledge, which also increases as auditor tenure increases. This leaves space for more research when exactly auditor client tenure increases or reduces the quality of auditor’s work. In this regard I introduce client complexity as a key moderator for this relationship. Using a sample of 163 Dutch auditors, this paper demonstrates that for highly complex clients, a long audit-client tenure is positively related to an auditor’s professional skepticism towards that client, resulting in a decrease of audit quality threatening behaviours. Auditor client characteristics represent an important influence in the auditors work and this paper contributes to the literature by showing that client complexity can significantly influence the behaviours as exhibited by auditors in their work.

Keywords: Auditors, Professional skepticism, Audit quality threatening behaviours, Audit-client tenure, Client complexity, Independence, Learning.

(3)

2

I. Introduction

The financial crisis of 2008 and the role of the auditor is a highly debated subject, with fundamental questions about the role of the auditor and the expected audit quality (Sikka, 2009). This crisis was characterized by “excessive risk taking through complex financial instruments” and raised questions about auditor independence, knowledge, professional skepticism and ultimately audit quality (Sikka, 2009). Audit quality is a broad concept with definitions reliant on differing stakeholder perceptions, ranging from the absence of material misstatements to “satisfactorily completing all tasks required by the firm’s audit methodology” (Knechel et al., 2013). Because of the numerous scandals (for instance, Enron or ING1_{) audit quality became highly regulated and the drivers of audit}

quality researched, given the pivotal role of audit quality in supporting financial stability (Beattie et al., 2013; IAASB, 2016). Although the importance of high-quality audits is

recognized, the quality of the audit is ultimately determined by individual auditors carrying out their work. At this moment, research on the drivers of audit quality has progressed, but many factors related to individual auditors and the interrelations between these factors remain obscured (Knechel et al., 2013).

Professional skepticism however, is one of the cornerstones of the audit profession and is regarded as a main driver of the audit quality provided by individual auditors, given that the lack of professional skepticism is regarded as an importance cause of audit failure (Knechel et al., 2013; PCAOB, 2008; PCAOB 2010; Hurtt et al., 2013). The ICAEW, or the Institute of chartered accountants in England and Wales, even notes that “professional skepticism is at the heart what auditors do, without it the audit has no value” (p. 2). The PCAOB, the oversight board of public company accounting, defines professional skepticism as “an attitude that includes a questioning mind and a critical assessment of audit

evidence”. Continuing, “the auditor should conduct the engagement with a mindset that recognizes the possibility that a material misstatement due to fraud could be present, regardless of any past experience with the entity and regardless of the auditor's belief about management's honesty and integrity” (PCAOB, AU316.13). This definition implies that

1_{ING experienced global backlash after not controlling enough for cash laundering through its accounts.}

Fake holdings and off-the-books accounting practices led to the bankruptcy of Enron and its auditor Arthur Andersen.

(5)

4 professional skepticism needs to be integrated throughout the entire audit and confirms the importance of individual auditors in defining audit quality.

Given its previously described fundamental role, many studies seek to understand what drives professional skepticism (Hurtt et al., 2013; Peytcheva, 2014). Extant work recognizes the multi-dimensional nature of professional skepticism (Hurtt, 2010). This is characterized by the distinction between trait professional skepticism, being individual attributes of an auditor and states, in which skepticism is aroused by external factors (Olsen, 2017; Shaub, 1996; Hurtt, 2010). State professional skepticism is dependent on situational factors, explaining the fact that auditors with the same level of trait skepticism still exhibit different levels of skepticism given certain circumstances (Hurtt et al., 2012; Robinson et al., 2018). In addition to this, states are less stable than traits and therefore need more timely response than for instance training in trait professional skepticism. Research into state influencing factors is scarce, but highly relevant in explaining situational changes in

professional skepticism and thus audit quality (Robinson et al., 2018, Hurtt et al., 2013). In their research into audit quality, Knechel et al. (2013) stress the importance of changing client characteristics in determining the delivered audit quality. In state professional

skepticism literature, client characteristics are also in high regard of determining situational skepticism (Robinson et al., 2018; Hurtt et al., 2013).

Given a broad range of varying client characteristics, it is necessary for auditors to have client-specific knowledge. This is positively related to audit efficiency and effectivity as multiple papers find a negative relation to audit rotation and audit quality due to less client knowledge (Bowlin, et al., 2015). Having audit tenure is part of the well-known and long-standing tension between having client-specific knowledge and the appropriate

independence of the client (Simunic, 1984). This tension is reflected in for instance long audit tenure due to the effects of increased familiarity and less distrust, which threatens auditor independence and thereby audit quality (Tepalagul & Lin, 2015). Being a

fundamental building block of the audit profession, independence is also regarded as a core driver of professional skepticism (Ye et al., 2011; Hurtt et al., 2013). Both independence and

(6)

5 client-specific knowledge are recognized drivers of state professional skepticism given their changing effect between different clients (Hurtt et al., 2013) 2_.

Having discussed client-specific knowledge and independence as important drivers of professional skepticism and ultimately audit quality, the situational factors influencing when which effect is most influential remain obscured. An expected, but scarcely researched and of growing importance characteristic of a client is the complexity of the client (Hurtt et al., 2013; Nelson, 2009; Rasso, 2015). Client complexity is defined as a client with “a more diverse organisational structure and harder to review transactions” (Al-Harshani, 2008). For instance, complex estimates require auditors to make multiple assumptions or

computational models, which are of increasing importance to the financial statements (Griffith et al. (1), 2015). In response of the upcoming complex needs of auditors, the PCAOB issued the “Critical audit matters”. These describe matters that prove difficult for auditor judgement and require the reporting of information of auditor independence and client tenure (PCAOB, 2013; DeFond & Zhang, 2014). Since it’s the purpose of auditors to enhance reliability, complex estimates and the reliance on biased management assertions for

explanations “jointly imply that audit quality in this area is critical to the reliability of

financial reports” (Griffith et al. (1), 2015). This research therefore answers the call by Hurtt et al., (2013) for more research into the relationship of impaired skeptical judgements for complex clients.

Present research treats professional skepticism mostly as a trait, thereby missing the situational factors that influence an auditor’s judgement process (Robinson et al., 2018). This research will contribute by primarily looking into the well-known tension of

independence and client-specific knowledge. Both are prominent effects in literature, however, when and why which effect is most influential is scarcely researched. This paper tries to pinpoint this effect on the complexity of clients as defining factor in this tense relationship in order to generate new consensus in two well-known factors in auditing. The goal of this research will be to refine our current knowledge on situational factors

influencing state professional skepticism as well as how client-audit tenure relates to this.

2_{I refer to Appendix A, which shows table 1 of the paper of Hurtt et al. (2013). This shows the different}

(7)

6 Next the theoretical background and method will be discussed, which leads to the results and discussion section. The guiding research question for this paper will be:

When and why does client-audit tenure impact professional skepticism and indirectly audit quality?

II. Theoretical background and hypothesis development

Audit Quality Threatening Behaviour

Throughout auditing literature audit quality is a prominent subject, which is

described as a broad and difficult to grasp concept (Knechel et al., 2013). However, widely used is the definition of DeAngelo, which consists of two elements; first, audit quality is determined by the likelihood of discovering misstatements and second how the auditor responds to this evidence (Knechel et al., 2013; DeAngelo, 1981). Where the first element relates to the competence of auditors in collecting and evaluating evidence, the second relates to the objectivity and independence needed to act on the found evidence (Knechel, 2013). Corresponding with this view, audit quality is affected by many factors related to audit testing and the engagement team (Francis, 2011). Summarizing, audit quality is primarily affected by auditors’ individual characteristics in relation to how the auditor responds to his judgements in differing situations (Knechel et al., 2013; Nelson, 2009).

In order to conceptualize this broad subject many attempts have been made to capture the essence of audit quality. When it comes to individual auditors, audit quality threatening behaviours, (hereafter AQTB) lie at the heart of high-quality audit work (Bedard et al., 2008). These behaviours summarize a broad array of quality diminishing behaviours “such as failure to pursue questionable items, false signoff on audit steps, performing a superficial review of audit evidence, or suppressing questionable audit evidence”

(Peytcheva & Gillet, 2012). With the current focus on audit quality control, great emphasis is put on the identification and the causes of AQTB in order to achieve higher audit quality (Khan et al., 2013). Although the behaviours itself receive much research attention, the factors influencing the decision to engage in these quality diminishing behaviours remain unspecified (Sweeny et al., 2013). To bridge that gap, judgement and decision-making research attempts to map the thought process before engaging in AQTB (Pietsch & Messier, 2017). Many skills in auditing are learned through education, however AQTB’s have a rather

(8)

7 social fundament, which makes this concept a more individual and psychological contract (Herrbach, 2001). This fundament of psychological mindset and auditor attributes illustrate the shared fundament of professional skepticism and AQTB, which will be explained in the next section.

Professional skepticism in the audit

The relationship between audit quality and professional skepticism takes a

prominent place in literature as well as in auditing standards. Many standards advocate the need of professional skepticism in the collection and evaluation of audit evidence (Lee, et al., 2013). Glover & Prawitt (2014) describe professional skepticism as one of the most important means of enhancing audit quality. A reason for this intuitive relationship is the perfect correspondence between the two elements of audit quality and the applicability of professional skepticism. The element of discovering misstatements is influenced by a higher audit effort (Knechel et al., 2013), high quality hypotheses development and evaluation (Nelson, 2009), higher focus on evidence (Harding et al., 2017) and a higher likelihood of assigning the right conclusion to fraud signals (Messier, 2013). In addition to this, an important state of mind is presumptive doubt of skeptical auditors as opposed to presumptive trust, since situational distrust is an important contributor to the state of professional skepticism (Lee et al., 2013; Harding et al., 2017). Then, when the right amount of evidence is collected and evaluated, the auditor must act in the appropriate way on the information (Knechel et al., 2013). This is the second element of audit quality (Knechel et al., 2013).

Research into state professional skepticism however is scarce, as many researchers relate trait professional skepticism to audit quality and other factors (Robinson et al., 2018; Hurtt et al., 2013). State skepticism relates to variables that are “more malleable, influenced by a particular situation or context” (Robinson et al., 2018; Steyer et al., 1999). In the

literature synthesis of Hurtt et al. (2013) on professional skepticism, factors influencing state skepticism take a prominent place, with client factors being the most influential in

explaining audit quality. This leads to the following hypothesis:

(9)

8 Audit client-tenure

Next to skepticism, independence is an important determinant of audit quality (Tepalagul & Lin, 2014). To aid auditors and after concerns expressed by regulators,

safeguards and standards impose rules to maintain an appropriate level of independence in their audits (Church et al., 2015). One important safeguard is the mandatory firm rotation, needed to disturb bonds of familiarity and affiliation with the client (Tepalagul & Lin, 2014; Church et al., 2015). A prominent effect of client affiliation is excessive trust in the client’s statements and explanations (Church et al., 2015). Through excessive trust, the judgement of the auditor becomes biased and leads to different actions on the collected evidence, since auditors have actions more aligned with client preferences (Church et al., 2015). Independence of the mind permits the auditor to express a conclusion without influences (Chiang, 2016). The relationship between longer audit tenure and an erosion in

independence thus leads to lower audit quality (Ye et al., 2011). Skepticism is also

influenced by independence, since auditors may be unwilling to act to conserve their own self-interest (Chiang, 2016). Therefore, a lack of independence can lead to an auditor not displaying the appropriate amount of professional skepticism in his work (Chiang, 2016).

Closely related to the independence issue is the term client familiarity, where familiarity relates to employees becoming too acquainted with the client (Svanberg & Öhman, 2015). This results in transactions taken for granted and not appropriately verified (Ebimobowei & Keretu, 2011). An example for this is the finding of Carey & Simnett (2006) that audit partners with longer tenure tend to have a lower propensity to issue a modified going-concern opinion. Next to familiarity, client identification in combination with longer audit tenure has negative effects on audit quality, again because of an impairment in objectivity (Svanberg & Öhman, 2015). In summary, long audit tenure creates a learned trust in the client, jeopardizing the independence and objectiveness of auditors, leading to a loss of motive to perform the audit with due professional care (Al-Thuneibat et al., 2010). This leads to the following hypothesis:

H2a: There is a negative association between audit client-tenure and professional skepticism.

(10)

9 However, as previously mentioned, independence is tensely related to the

acquisition of client-specific knowledge, which results in two opposing views around longer audit tenure; one is described earlier and holds a negative view point, the other however shows the positive effects of longer audit tenure on professional skepticism (Tepalagul & Lin, 2014). The positive view, as supported by multiple papers relate longer tenure to higher audit performance (Hung Chan & Wu, 2011), more audit expertise (Knechel et al., 2013) and a higher understanding of the client (Tepalagul & Lin, 2014). Understanding can be defined as knowledge acquired from client or industry specific experiences and provide the best explanation of auditor expertise (Hung Chan & Wu, 2011). Papers researching the relation between auditor tenure and audit quality find notable results: audit quality is lower for the short tenure auditor and as some find a deteriorating quality for longer tenure, most find no effect on quality of middle and short tenure (Tepalagul & Lin, 2014; Church et al., 2015; Carey & Simnett, 2006). The lower quality for short term auditors is related to their knowledge, since auditors need time to develop their understanding of the client at the early moments and are therefore more reliant on their client (Tepalagul & Lin, 2014; Gul et al., 2009). Another argument for this case is found in familiarity with the client, where knowledge is crucial to get familiar with for instance the client’s system and asset

management (Boone et al., 2008; Al-Thuneibat et al., 2011). The auditors with medium and long tenures benefit from their increased knowledge of clients, which effect is enlarged through client and industry specializations (Gul et al., 2009; Simunic & Stein, 1990;

DeAngelo, 1981; Hope & Langli, 2010; Knechel et al., 2013). These specializations counteract the effects of biases and help audit firms improve audit efficiency (Church et al., 2015; Habib, 2011; Arthur et al., 2017). This leads to the following hypothesis:

H2b: There is a positive association between audit client-tenure and professional skepticism.

The most prominent of these views is the independence view, with a negative view on long audit tenure on professional skepticism. Throughout literature, law and regulations, audit partner rotations and independence issues are of grave importance (PCAOB; Chan & Wu, 2008; Carey & Simnett, 2006). However, given the existence of the two opposing views alongside, another factor is expected to provide clearance in when which effect is most

(11)

10 prominent and influential. With the growing importance and decisive influence of client characteristics, the complexity of clients is expected to give insight in this relationship. This will be elaborated on next.

The moderating effect of client complexity

Research into audit client characteristics is widely recognized, where client complexity is regarded as a determinant in audit judgement, which connection is

established by DeFond & Zhang (2014) in their research on audit quality determinants. The underlying cause of this relationship was found already in 1999 where client complexity was related to a higher chance of unforeseen problems and more time usage in audits (Gist & Davidson, 1999) They related this effect to more time needed for auditors to learn about their new clients (Gist & Davidson, 1999). The audit judgement process of complex estimates is different from the usual process, because auditors are testing management’s process of generating the complex estimates, which makes them reliant on management (Griffith et al. (1), 2015; Griffith et al. (2), 2015). This reliance disturbs the judgement process of professional skepticism, because auditors may become overwhelmed with information provided by management (Hurtt et al., 2013). Complex information requires auditors to rely more on their professional skepticism and thus on self-collected evidence stemming from their client-specific knowledge. By doing so, they are willing to extend their information search as well as be open to analysing inconsistencies among the assumptions and assertions from management (Nolder & Kadous, 2008; Griffith et al. (2), 2015). In conclusion, in the face of complex clients, client specific knowledge is needed in order to maintain independency from management. Thus, it appears that for complex client’s high client tenure is less likely to undermine auditor professional skepticism.

A lack of expertise and client-specific knowledge leads to the possibility of a decline in the exhibited professional skepticism during the audit of complex clients (Hurtt et al., 2013). A deliberate mindset focussed on skepticism is needed to counter the effects of uncertainty and complexity, where the auditor notices and appreciates seeded errors and inconsistencies in management assertions (Griffith et al. (2), 2015). A different approach is therefore needed with respect to the judgement process, where auditors shift from testing management’s process to an approach where the auditor relies on developing independent expectations using client-specific knowledge (Griffith et al. (1), 2015). This is consistent with

(12)

11 Srinidhi et al. (2010), who find that, in general, unique client-specific knowledge is

moderated when considering the effect of complex clients. Considering the judgement process of professional skepticism, I anticipate that the relationship between audit-client tenure and professional skepticism is moderated client complexity. Where a high complexity calls for auditors to require more knowledge of the client to be able to exhibit the

appropriate level of independency and professional skepticism. This leads to the following hypothesis:

H3: The relationship between audit client-tenure and professional skepticism is moderated by client complexity, such that this relationship is more positive for complex clients.

By extension, because I also anticipated that professional skepticism is negatively related to AQTB (H1), and because I anticipate that the relationship between audit client-tenure and professional skepticism is moderated by client complexity (H3), I arrive at the following moderated-mediation hypothesis:

H4: The indirect relationship between audit client-tenure and AQTB through professional skepticism will be moderated by client complexity.

The research model underlying the three hypotheses is conceptualised in the next figure. Where the plus sign stands for a positive expected relationship and a minus for a negative expectation. Given the double-sided expectation for hypothesis 2 this relationship has a plus/minus sign.

(13)

12

III. Methods

Research setting

The study was conducted in the Netherlands. A total of 180 auditors completed the survey. The sample consists of Dutch external auditors from different ranks, audit firms and offices. Members of the research team contacted individual auditors who then agreed to fill in the survey, which was sent later. Survey is the chosen research method given the

subjective and psychological foundation of the constructs. Since the auditors were all Dutch, the survey and reminders were sent out in Dutch as well. The personally asked participants received the survey from the research supervisor as well as a first reminder. After that, participants were personally engaged by the research team to ensure a participation as high as possible. Given the nationality of the participants the measures, originally taken from English literature, have been translated to Dutch. To ensure validity the measures were translated and then back translated to control for accuracy issues. Participation was done on a voluntary basis. Auditors of all ranks could participate in the survey with the only requirement being that the participant has two years or more of service. This requirement was to secure enough experience in auditing to adequately fill out the survey.

The total sample existed of 163 completed surveys after correcting for missing values. The following two graphs explain the demographics of the sample. These graphs show that the majority of the participants has a tenure of 2-12 years and that the majority has a seniority of average in their firms.

(14)

13 Measures

Dependent variable: AQTB

The measure for audit quality adopted in this paper is AQTB, which will be measured using the questions of Herrbach (Herrbach, 2001). This 8-item survey instrument has been repeatedly used throughout literature and exists of the following items: 1. Reduction of work below what is considered reasonable, 2. Superficial review of a client document, 3. Acceptance of weak client explanation, 4. Failure to research an accounting principle, 5. Failure to pursue a questionable item, 6. Signed off an audit programme step without completing the work, 7. Greater than appropriate reliance on client work, 8. Personally engaged in underreporting of time (i.e., deliberately failing to record the full amount of time spent on a task) (Herrbach, 2001; Otley & Pierce, 1995). These items were presented on 7 item Likert scale with 1 = never and 7 = very often. The Cronbach alpha for the 8 items together was α=0.822.

Mediator: state professional skepticism

Audit client-tenure will be related to audit quality through professional skepticism. This relationship will be measured by using the state skeptical judgement survey questions of Robinson (Robinson et al., 2018). Using the abovementioned literature of Hurtt and Nelson they disentangle trait and state professional skepticism and so provide the first measure of state skeptical judgement. These questions are validated in their paper by means of different validity tests. The same dimensions of skeptical judgement have been used by Robinson in the creation of the questions, thus being questioning mind, search for knowledge and suspension of judgement. The formulated questions are different from Hurtt’s questions, because they are made more specific for client use as well as shortened per construct. Sample questions consist of for instance “Overall, I tended to question the statements that I read from Phil, the controller” and “While working on this case, I

frequently questioned things that I saw or read.” Above examples show the more specified questions formulated by Robinson. The measurement exists of 12 survey items. The

(15)

14 Independent variable: audit-client tenure

In their literature synthesis on professional skepticism, Hurtt et al. (2013) conclude that “the impact of experience on skeptical judgment (or lack thereof) is derived from a number of factors, such as the level of knowledge of the client’s business and industry, the number of years one works as an auditor, task-specific experience, and experience with more complex audit tasks“. Following this, these effects are combined in one variable of audit client tenure. This will be measured by the auditor’s client related tenure, i.e. the years of work performed for this particular client. Important to note here is that in order to be as specific as possible about clients the setup of this part of the survey has been framed with the same method used by Andiola & Bedard (2018) in their paper. Stemming from behavioural research asking about specific review cases (in this paper clients), yields situations which are better predictors of changed or created behaviours, while maintaining reliability (Morgeson et al., 2015; Mignonac & Herrbach, 2004; Andiola & Bedard, 2018). Moderator: client complexity

Client complexity is measured using the widely used questions of Semmer, which exists of 4 items (Semmer, 1982; Zapf, 1993). The items correspond to the definition of job complexity used by Morgeson and Humphrey (2006), that also reflects cognitive demands present in the crucial judgement process. However, to be more specific for auditing purpose adjustments have been made resulting in the following set of items: 1. To what extent does your work for this client consist of tasks that are particularly complex and difficult, 2. To what extent do you have to make very complicated decisions in your work for this client, 3. To what extend do you have to use all your knowledge and skills in your work for this client, 4. To what extent can you learn new things in your work for this client. To be more specific the client effect has been added as well as “to what extent” in order to make the answers quantifiable on a 7-point Likert scale. A last adaption was needed for questions 1 and 4 according to Li et al. (2017) to enhance the reliability of the items for specific usage according to the HILDA survey. The Cronbach’s alpha for the 4 questions is α=0.736.

(16)

15 Control variables

In this paper, we control for a number of variables on the audit firm and individual auditor level. Given the different effects size of the audit firm has on audit quality a dummy variable has been used to counter the effect of Big4/nonBig4 audit firms (DeAngelo, 1981; Chen et al., 2013; Becker et al., 1998). Next to this the gender of the auditor performing the audit judgement is affecting the audit quality delivered. Therefore, this effect will be

controlled for by means of a dummy variable (Garcia-Blandon et al., 2019). In addition to this, the effect of auditor experience will be controlled for. Audit experience has a positive effect on specific audit tasks and the engagement level overall, therefore this effect will be controlled (Cahan & Sun, 2015; Wang et al., 2015). This effect will be captured using two measures: firstly, the overall tenure in auditing as well as the seniority achieved in the audit firm. Since almost all auditors are from the Netherlands, no country dummy is needed to control for differences between countries (Choi et al., 2010). To further isolate the effects of audit-client tenure alone, the effects of client hours will be controlled for, given the

research on audit quality stating the effect of audit hours on quality (Kwon et al., 2010). Furthermore, hours spent on a client is positively related to professional skepticism, where more hours relate to a higher exhibited professional skepticism. For instance, time pressures are negated for skeptic auditors still willing to invest more hours on their client to keep their skepticism at the same level in spite of external factors (Hurtt et al., 2013). In other

literature on audit quality the client hours or fee are taken as proxies for audit quality. The assumption there is that “the more hours budgeted, the more professional skepticism is shown” (Quadackers et al., 2014). To control for these effects on professional skepticism, client hours will be added as a control variable.

Biases of items

Given the subjective nature of the asked survey questions, the problem of common source bias needs to be addressed. This bias can possibly inflate the researched

relationships and so disturb results (Meier & O’Toole, 2012). Perceptual bias can be present in both the client complexity and state professional skepticism items, given their subjective character and self-assessment of the respondents. To control for these biases, an

exploratory Harman single factor test was done. This gives insight in whether common source bias could be present. The results of this test (25.82%) were below the threshold of 50%, so this reduces the potential common source bias in this sample. Next to this effect,

(17)

16 the dependent variable of AQTB is subject to conservative bias given the nature of the asked questions. By introducing confidentiality and anonymity to the students who receive the data, this effect is reduced, but the conservative bias can still be present. Another reason for biases is the social desirability of the construct AQTB, where AQTB are essentially negative work behaviours, auditors admit to lowering their quality in their work through these behaviours. Therefore, this must be considered when interpreting the results.

IV. Results

Descriptive statistics

Table 1 presents the percentage of frequencies of the used items of client-specific AQTB on a 7-point Likert scale. The overall mean of the items is not very high (M=2,09). However, table 1 shows that al the behaviours are being engaged in. Furthermore, table 1 indicates that especially three items of QTB are prominent, being: superficial review of client document, greater than appropriate reliance on client work and personally engaged in underreporting of time. Using Stata, a confirmatory factor analysis shows however that the eight QTB items load on one factor (𝜒2 (20) =56.45, p=0.000, CFI=0.945, TFI=0.923,

RMSEA=0.105, SRMR=0.045). Modification indices confirm this model as the optimal fit for measuring AQTB. Following Coram et al., (2008) some behaviours as for instance false sign-offs are perceived as the most threatening behaviour and not testing all sample items as much less of a threat, meaning that the gravity or the moral intensity underlying these behaviours varies. When interpreting the results of table 1, this view corresponds with the findings of Coram et al., (2008) where the behaviours perceived as less threatening are more present than the most threatening behaviours.

Table 2 presents the descriptive statistics for all variables as well as the correlations between the variables. This table indicates that 28% of the auditors in the sample are female and 72% male. Next to this, 84% of the sample consists of Big4 offices and 16% of other audit firms. The participants have approximately 7.8 years of experience and work for almost 3 years at the same client. Table 2 further shows that the correlations with the independent variable are statistically significant, but below the threshold of 0.6, which indicates that no problems of multicollinearity should arise. Furthermore table 2 shows that

(18)

17 the correlations between the models and the control variables are present and statistically significant. Given the format of table 2, this table has been allocated behind the references.

Table 1 – Frequencies, mean and standard deviation of each AQTB (N=163) 3

QTB 1 QTB 2 QTB 3 QTB 4 QTB 5 QTB 6 QTB 7 QTB 8 1 "never" 32.3 27.5 28.1 35.9 53.9 47.9 25.1 34.1 2 39.5 41.9 46.1 38.9 34.7 29.3 44.3 32.9 3 18.0 17.4 16.8 18.6 6.6 8.4 16.8 10.8 4 4.8 7.2 5.4 3.0 3.6 6.0 7.8 7.2 5 5.4 5.4 3.6 2.4 1.2 7.8 5.4 9.6 6 0 0.6 0 1.2 0 0.6 0.6 5.4 7 "Very often" 0 0 0 0 0 0 0 0 Mean 2.11 2.23 2.10 2.01 1.63 1.98 2.26 2.41 Hypothesis testing

Table 3 shows the results for models 1 and 2 reliant on QTB. The first part of the model states the control variables and model 2 shows the main results of the relation between professional skepticism and QTB the total dependent variable of this paper. Model 2 of table 3 presents the results on hypothesis 1, showing that the coefficient of

professional skepticism is negative and significant (B=-0.34, p-value<0.001). This result is in line with hypothesis 1, which predicted that a higher score of professional skepticism would be negatively related to QTB in audit work.

3_Notes:

AQTB 1: “Reduction of work below what is considered reasonable.” AQTB 2: “Superficial review of client document.”

AQTB 3: “Accepted weak client explanations.” AQTB 4: “Failure to research an accounting principle.” AQTB 5: “Failure to pursue a questionable item.”

AQTB 6: “Signed off on an audit programme step without completing the work.” AQTB 7:” Greater than appropriate reliance on client work.”

AQTB 8: “Personally engaged in underreporting of time (i.e., deliberately failing to record the full amount of time spent on a task).”

(19)

18

Note: One-tailed test for hypothesized variables, two-tailed test for control variables. Coefficient standard error in parentheses. * p<0.1, ** p< 0.05, *** p <0.01.

Table 4 shows the main results reliant on professional skepticism. The first part of the model, further described as model 1, centers the control variables on professional skepticism. Model 2 describes the main relation of hypothesis 2, audit-client tenure on professional skepticism. This result is positive although insignificant meaning no direct relationship giving no direct result for hypothesis 2. In spite of these results, model 3 shows the moderation effect of client complexity on the relationship between audit-client tenure and skepticism. Model 3 of table 4 shows the main results. Column 1 shows that the coefficient of the moderator-term Audit-client tenure x Client complexity is positive and significant (B= 0.12, p-value<0.05). This result is in line with hypothesis 2b and 3. First, hypothesis 2 states an either positive or negative relationship between audit-client tenure and professional skepticism. Then hypothesis 3 further expresses the moderating effect client complexity has in defining this relationship. Model 3 shows the negative relationship

Table 3: Regression analyses for Audit Quality Threatening behaviours

(1) (2)

Variables _Controls Main effect

Constant _3.30*** 4.69*** (0.31) (0.47) Big4 _-0.45** -0.25 (0.18) (0.18) Gender _-0.16 -0.15 (0.14) (0.13) Experience _-0.01 0.00 (0.01) (0.01) Seniority _-0.20*** -0.19** (0.08) (0.07) Client-hour _0.00 0.00 (0.00) (0.00) Mediator Professional skepticism -0.34*** (0.09) R2 0.09 0.16

(20)

19 of hypothesis 2b and 3 predicting that the negative relationship between client tenure and professional skepticism is less negative for more complex clients.

Table 4: Regression analyses for Professional skepticism

(1) (2) (3)

Variables Controls Main effect Moderator

Constant 4.00*** 3.95*** 4.21*** (0.27) (0.30) (0.28) Big4 0.60*** 0.62*** 0.53*** (0.16) (0.16) (0.16) Gender 0.02 0.02 0.02 (0.12) (0.12) (0.14) Experience 0.01 0.01 0.02 (0.01) (0.01) (0.01) Seniority 0.05 0.06 0.03 (0.07) (0.07) (0.07) Client-hour 0.00 0.00 0.00 (0.00) (0.00) (0.00) Stage 1: main relationship

Audit client tenure 0.01

(0.04) Stage 2: Moderator Audit client-tenure 0.01 (0.05) Client complexity 0.14** (0.06) Audit client-tenure x Client complexity 0.12**

(0.05)

R2 0.13 0.13 0.17

To gain further insight into this effect, the relationship has been plotted. Figure 3 represents the resulting graph showing the above stated effects. Most notable is the effect of a high complexity client on the professional skepticism. When high client tenure is combined with a highly complex client the professional skepticism is positively influenced. This result shows that for low complexity clients professional skepticism remains the same,

(21)

20 however when with a highly complex client the auditor benefits from a longer client tenure. This is shown through the increase in professional skepticism for complex clients with high tenure.

Figure 3: The relationship between Audit-client tenure and Professional skepticism at different levels of Client complexity

Furthermore, hypothesis 4 predicts a mediating relationship, so that audit-client tenure and QTB are indirectly related through professional skepticism. To test the

significance of these indirect effects the bootstrapping procedures of Hayes (2018) are used. The applied method was the PROCESS macro application for SPSS with 5000 bootstraps. This method is used to be able to better assess the indirect effects of the combined moderated mediation relationship. Results for this moderated mediation effect show a significant indirect relationship between audit-client tenure and QTB, when client complexity is high. The bootstrapped 95% confidence intervals include zero for mean (-0.01, 0.07) and low complexity at one standard deviation below the mean (-0.03, 0.02), but not for high

complexity at one standard deviation above the mean (-0.08, -0.01). These results show that a high client complexity indirectly influences QTB through professional skepticism. This is in

4 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5

Low Client tenure High Client tenure

Pr of essi on al sk ep tic is m Low complexity High complexity

(22)

21 line with the predictions of hypothesis 4, stating a moderated mediating relationship

between client tenure and QTB through professional skepticism, moderated by client complexity. This largely supports hypothesis 4.

Table 4: Indirect conditional effects of the moderating relationship on audit quality threatening behaviours through Professional skepticism

Indirect conditional effects at client complexity

Mean ± SD4

Client complexity Boot indirect effect Bootstrap SE Bootstrap 90% confidence interval Lower bound Upper bound -1 SD (4.42) 0.03 0.02 -0.01 0.07 Mean (5.28) -0.00 0.01 -0.03 0.02 +1 SD (6.14) -0.03 0.02 -0.08 -0.01

n=163 auditors. Bootstrap sample size is 5000.

Additional testing

Next to the main models, additional tests have been done to enhance robustness of the sought relationships. Through these tests, variables of the main models have been changed or new ways of categorising have been used. The next paragraphs explain the different robustness tests.

Client hours

Although the results of audit-client tenure were as predicted and significant, further investigation was done into audit-client hours. Similarities between the two concepts can be drawn and as client-hours are controlled for, the main hypotheses are not influenced. A prominent difference between tenure and hours might be that an auditor has a low tenure of for instance two years but worked a high number of hours per year for that client. The explanatory power of hours might therefore be higher than audit-client tenure. However, the relationship of hypotheses 2 and 3 are not significant and thus not supported with client hours. Therefore, reinforcing the current method of using client tenure as the independent variable, being moderated by client complexity and controlling for differing spent hours on clients. Data on client hours shows a substantial difference between client hours of Big4 auditors and non-Big4. However, as a Big4 control variable dummy has also been used, this

(23)

22 effect is tested by either controlling for Big4 or leaving it out of the model. The results of client hours remain insignificant, whereas the main model with audit-client tenure remains robust and significant at the 5% level (p-value<0.05). The overall conclusion therefore is that even when controlling for different effects the explanatory power of audit-client hours is low and thus provides robustness to the used audit-client tenure in the main models. Professional skepticism subcategories

The chosen measure of state professional skepticism is based on subcategories elaborated by Hurtt (2010). By means of additional testing, the moderating relationship has also been tested according to these three subcategories, being Suspension of judgement, a questioning mind and search for knowledge. Table 4 model 3 has been repeated, replacing the measure of state professional skepticism for the three new constructs measuring the subcategories. Regression results on these categories show only significant results for the moderating relationship on the last subcategory search for knowledge. However, as the main model consists of state professional skepticism as a whole these additional tests add to robustness and further insight into the relationship with professional skepticism. Mediating role of professional skepticism

Testing further on the professional skepticism variable, the mediating role might be questioned. As the similarities between audit quality and professional skepticism are

intuitive the mediating role might be unnecessary in explaining the moderating role of client complexity. To investigate this possibility, table 4 model 3 has been altered to show the moderating role directly on AQTB. The result of this additional test is negative and weakly statistic (B=-0.08, p<0.1)5_{. In addition to this the R squared is significantly lower than the}

original model (0.10 as opposed to the original 0.17) This additional test adds to the robustness of the original model where the indirect effect of audit-client tenure on QTB is mediated through professional skepticism.

(24)

23 Big4 vs non Big4

As a final test, the difference between Big4 offices and non Big4 will be enlarged in additional test. Descriptive statistics of this dataset show an additional reason for this test, as client hours reported are significantly higher for Big4. Where in the main model a dummy variable was used to control for these differences, now model 2 and 3 have been redone to elaborate on these differences. The results of these additional tests show no significant results for the hypothesized relationships of 2 and 3 for the only Big4 sample. When further inspecting the results of the non-Big4 sample, a highly significant moderating relationships becomes evident (B=0.51, p-value<0.001, n=27). These results could indicate a stronger effect for non-Big4 auditors with complex clients. As explained in the control variable section, notable differences exist between these two samples. Therefore, it is important to be aware of these differences and might provide avenues for future research into more differences in state effects between Big4 and non Big4. As this research warrants more investigation elaborations will be made in the discussion section.

V. Discussion and conclusion

Discussion and contribution

Auditor client tenure is a highly debated subject with advantages and disadvantages made explicit in laws and regulation, but also directives and guidelines from auditing instances6_{. This paper has sought to shed light on this tense relationship by looking at the}

positive and negative effects and whether client complexity could play an influencing role in which effect is more prominent. The complexity of a client directly influences the work an auditor has to perform. Consequently, this might affect auditor behaviours by which audit quality is negatively influenced. The results of this paper show two main results. First, audit-client tenure has a slight positive effect on professional skepticism, but this effect is

enlarged for complex clients. Second, through professional skepticism, this moderation effect is linked to audit quality threatening behaviour, so that long audit tenure for complex clients has a negative effect on auditors engaging in QTB.

6_{See also various guidelines by the PCAOB or auditing directives, most prominent in PCAOB AS1005 as well as}

(25)

24 The first result brings a new viewpoint on long auditor tenure and the moderating effect of client complexity. Stemming from regulation, current focus is on the negative effects of longer audit-client tenure, thus highlighting the independence of auditors. In addition, literature describes this negative viewpoint as also an opposing view where client knowledge has a positive effect on auditor judgement. This paper provides new insights into these positive effects of longer tenure in combination with more complex clients. The moderating relationship of complexity became evident as the key determinant, which to my knowledge has not yet been explored in current literature. Where several papers mention client characteristics, the effects of client complexity remained a theoretical concept, with not proven effects in this setting (Gist & Davidson, 1999; Hurtt et al., 2013). This paper provides this new perspective by observing the effects of client complexity on this tense relationship and shows that complex clients have a positive effect on professional skepticism exhibited by auditors.

The second result is the relationship linking the found moderating relationship not only to professional skepticism, but also to audit quality. This results in the complete and new perspective of this paper, being that client tenure when moderated by complexity and mediated by professional skepticism ultimately leads to a higher audit quality. Because of this mediating effect the connection between audit quality and the moderation becomes evident, providing more practical solutions than just state professional skepticism as dependent variable. As mentioned, literature shows the explanatory power of professional skepticism as a determinant of audit quality as well as an outcome of auditor judgement. Going back to the start of this paper, state professional skepticism as a determinant of audit quality is an under researched topic, which could lead to new insights. This direct

relationship between state professional skepticism and client specific quality threatening behaviours has been observed as a significant positive influence. This paper thus adds to the proven effects on the theory of state professional skepticism, since state professional

skepticism alone had not been connected to AQTB. This paper also adds to literature by showing that state professional skepticism is a significant determinant of audit quality in itself, but also when mediating the relationship of audit-client tenure and client complexity.

(26)

25 Future research directions

As discussed, this paper provides a new viewpoint of state professional skepticism as a determinant and outcome. This result provides a foundation from which new papers can be started, with state professional skepticism as fertile ground of research. Research into state factors influencing auditors is scarce, but as this paper shows has applicable

implications for auditors. The complexity of a client combined with client tenure is a predictor of professional skepticism and ultimately audit quality. From this basis new avenues on state and client factors can be explored in future research.

Since the direct relationship between audit-client tenure could not be found for this sample, this leaves space for more research into this relationship with state professional skepticism as dependent variable. This can be done finding different measures of audit-client tenure. This paper used tenure in years and additional testing used audit-client hours as measures of audit-client tenure. However, other avenues measuring the time spend on a client or the underlying aspects of a threat to independence could provide different results. In addition, the differing effects between Big4 and non Big4 might prove a future direction as significant difference between them have been observed.

Where this paper provides evidence for the positive viewpoint on audit-client tenure, the relationship could benefit from further investigation into the reasons why for complex clients longer audit tenure is beneficial. The question here is whether this effect is due to learning effects of audit-clients or whether other effects influence this relationship. This can be researched by delving deeper into the found relationship of tenure and

complexity. A more qualitative study using interviews over multiple years can be used to isolate learning effects on longer client tenure from other influencing variables, such as for instance overall experience and training. Where these effects are controlled for in the main models, a more qualitative stand could provide new insights in this new-found relationship. Finally, where this paper focusses on Dutch auditors a more generalizable sample of

different countries could shed light on the found relationships in a broader perspective. Practical implications

Next to the theoretical contributions this paper also demonstrates more practical implications for auditors and the firms they operate in. Following the model again, first the state professional skepticism and its positive effect on auditor quality have to be noted. The

(27)

26 difference between trait and state professional skepticism is regarding practical implications an important distinction, whereas traits can be learned through audit training, state

skepticism is reliant on outside factors such as client characteristics. Given the positive effect on quality as well as the differing effects between clients, state professional skepticism needs to be more encouraged when the auditor is in the client environment, instead of more common formal training, since the client has a determining impact on state professional skepticism. States are less stable than traits and therefore need more timely response and training than trait professional skepticism.

In addition, given the influencing role of audit-client complexity, client characteristics as determinant of audit quality can be further incorporated in team composition. For

complex clients, longer client tenure results in a sharp increase of professional skepticism and audit quality. Based on this result, a recommendation could be that a measure of complexity is assigned to clients, resulting in a complexity score. With this measure in mind longer client tenure can be created and maintained for complex clients. Keeping

independence and rotation rules in mind, ultimately a more efficient, due to potential learning effects and higher quality audit can be delivered for complex clients.

Limitations

As with every study, this paper has limitations which need to be recognized. Firstly, the sample consists only of Dutch auditors, which might impose problems of

generalizability. An auditing climate can have different effects on auditors, resulting in different work ethics and behaviours (Shafer et al., 2013). Therefore, this only Dutch sample needs to be considered as a limitation of this paper. As already mentioned, a more

generalizable sample spanning different countries can be a future direction for research. A second limitation as already mentioned is the potentially biased measurement of AQTB, related to the social desirability and self-reflective character of the construct. One possible research solution might be to replace the measure of AQTB with an accounting measure such as abnormal accruals (Francis, 2004) or the earlier mentioned audit fee (Quadackers et al., 2014). Finally, the subjective nature of client complexity has to be considered as a limitation. Future research might also consider an accounting related measure for client complexity such as using industry SIC codes the firm operates in (see also Carcello & Nagy, 2004). The used measure of client complexity does involve subjectivity as participants

(28)

27 measure it as they perceive it to be complex or not. By using for instance SIC-codes a more objective measure of client complexity could provide more robust findings on the effects of complex clients on auditors.

Conclusion

To conclude, the connection has to be made between the results of this paper and the current quality reports of the MCA and CTA aimed at improving the Dutch auditing climate. Their findings bear correspondence with the results, which is visible in their notion of professional skepticism, being an important driver of audit quality. In their report, the MCA refers to professional judgement of auditors being impaired by a focus on compliance. Their results show that a prominent effect hereof is less learning on the job, which leads to a lower audit quality. This result is in line with my results that auditors benefit from longer tenure when they face complex clients, which might be due to learning effects. In line with the MCA report, an encouragement of on the job learning might result in the wanted

improvement of audit quality. This then leads to the general conclusion of this paper, where audit quality is a broad concept, partly but significantly determined by just single auditors doing their job. They do this, while trying to maintain a skeptical mindset and influenced by clients, their environments and other factors yet to be researched.

(29)

28

References

Al‐Harshani, M. O. (2008). The pricing of audit services: evidence from Kuwait. Managerial Auditing Journal, 23, 958-696.

Al‐Thuneibat, A. A., Al Issa, R. T. I., & Baker, R. A. A. (2011). Do audit tenure and firm size contribute to audit quality? Managerial Auditing Journal, 26, 317-334.

Andiola, L. M., & Bedard, J. C. (2018). Delivering the “tough message”: Moderators of subordinate auditors’ reactions to feedback. Accounting, Organizations and Society, 70, 52-68.

Arthur, N., Endrawes, M., & Ho, S. (2017). Impact of partner change on audit quality: An analysis of partner and firm specialisation effects. Australian Accounting Review, 27(4), 368-381.

Beattie, V., Fearnley, S., & Hines, T. (2013). Perceptions of factors affecting audit quality in the post-SOX UK regulatory environment. Accounting and Business Research, 43(1), 56-81.

Becker, C. L., DeFond, M. L., Jiambalvo, J., & Subramanyam, K. R. (1998). The effect of audit quality on earnings management. Contemporary accounting research, 15(1), 1-24.

Bedard, J. C., Deis, D. R., Curtis, M. B., & Jenkins, J. G. (2008). Risk monitoring and control in audit firms: A research synthesis. Auditing: A Journal of Practice & Theory, 27(1), 187-218.

Boone, J., Khurana, I. and Raman, K. (2008), Audit tenure and the equity risk premium. Journal of Accounting, Auditing, and Finance, 19, 5-23

Bowlin, K. O., Hobson, J. L., & Piercey, M. D. (2015). The effects of auditor rotation, professional skepticism, and interactions with managers on audit quality. The Accounting Review, 90(4), 1363-1393.

(30)

29 Cahan, S. F., & Sun, J. (2015). The effect of audit experience on audit fees and audit

quality. Journal of Accounting, Auditing & Finance, 30(1), 78-100.

Carcello, J. V., & Nagy, A. L. (2004). Client size, auditor specialization and fraudulent financial reporting. Managerial Auditing Journal, 19, 651-668

Carey, P., & Simnett, R. (2006). Audit partner tenure and audit quality. The accounting review, 81(3), 653-676.

Chen, Y. S., Hsu, J., Huang, M. T., & Yang, P. S. (2013). Quality, size, and performance of audit firms. The International Journal of Business and Finance Research, 7(5), 89-105.

Chiang, C. (2016). Conceptualising the linkage between professional scepticism and auditor independence. Pacific Accounting Review, 28, 180-200

Choi, J. H., Kim, C., Kim, J. B., & Zang, Y. (2010). Audit office size, audit quality, and audit pricing. Auditing: A Journal of practice & theory, 29(1), 73-97.

Church, B. K., Jenkins, J. G., McCracken, S. A., Roush, P. B., & Stanley, J. D. (2015). Auditor independence in fact: Research, regulatory, and practice implications drawn from

experimental and archival research. Accounting Horizons, 29(1), 217-238.

Coram, P., Glavovic, A., Ng, J., & Woodliff, D. R. (2008). The moral intensity of reduced audit quality acts. Auditing: A Journal of Practice & Theory, 27(1), 127-149.

DeAngelo, L. E. (1981). Auditor size and audit quality. Journal of accounting and economics, 3(3), 183-199.

DeFond, M., & Zhang, J. (2014). A review of archival auditing research. Journal of accounting and economics, 58(2-3), 275-326.

(31)

30 Ebimobowei, A., & Keretu, O. J. (2011). Mandatory rotation of auditors on audit quality, costs and independence in South-South, Nigeria. International business management, 5(3), 166-172.

Francis, J. R. (2004). What do we know about audit quality?. The British accounting review, 36(4), 345-368.

Francis, J. R. (2011). A framework for understanding and researching audit quality. Auditing: A journal of practice & theory, 30(2), 125-152.

Garcia-Blandon, J., Argilés-Bosch, J. M., & Ravenda, D. (2019). Is there a gender effect on the quality of audit services?. Journal of Business Research, 96, 238-249.

Gist, W. E., & Davidson, R. A. (1999). An exploratory study of the influence of client factors on audit time budget variances. Auditing: A Journal of Practice & Theory, 18(1), 101-116.

Glover, S. M., & Prawitt, D. F. (2014). Enhancing auditor professional skepticism: The professional skepticism continuum. Current Issues in Auditing, 8(2), 1-10.

Griffith, E. E., Hammersley, J. S., & Kadous, K. (1) (2015). Audits of complex estimates as verification of management numbers: How institutional pressures shape

practice. Contemporary Accounting Research, 32(3), 833-863.

Griffith, E. E., Hammersley, J. S., Kadous, K., & Young, D. (2) (2015). Auditor mindsets and audits of complex estimates. Journal of Accounting Research, 53(1), 49-77.

Gul, F. A., Fung, S. Y. K., & Jaggi, B. (2009). Earnings quality: Some evidence on the role of auditor tenure and auditors’ industry expertise. Journal of accounting and Economics, 47(3), 265-287.

(32)

31 Habib, A. (2011). Audit firm industry specialization and audit outcomes: Insights from

academic literature. Research in Accounting Regulation, 23(2), 114-129.

Harding, N., Azim, M.I., Jidin, R. & Muir, J. P. (2017). Auditor Professional skepticism: impact of trust and distrust A synthesis of research findings. SSRN electronic journal.

Hayes, A. F. (2017). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. Guilford publications.

Herrbach, O. (2001). Audit quality, auditor behaviour and the psychological contract. European Accounting Review, 10(4), 787-802.

Hope, O. K., & Langli, J. C. (2010). Auditor independence in a private firm and low litigation risk setting. The Accounting Review, 85(2), 573-605.

Hung Chan, K., & Wu, D. (2011). Aggregate quasi rents and auditor independence: Evidence from audit firm mergers in China. Contemporary Accounting Research, 28(1), 175-213.

Hurtt, R. K. (2010). Development of a scale to measure professional skepticism. Auditing: A Journal of Practice & Theory, 29(1), 149-171.

Hurtt, R. K., Eining, M., & Plumlee, D. (2012). Linking professional skepticism to auditors' behaviors. Unpublished Working Paper.

Hurtt, R. K., Brown-Liburd, H., Earley, C. E., & Krishnamoorthy, G. (2013). Research on auditor professional skepticism: Literature synthesis and opportunities for future research. Auditing: A Journal of Practice & Theory, 32(sp1), 45-97.

Institute of Chartered Accountants in England and Wales (ICAEW). 2018. Skepticism, the practitioners take.

(33)

32 International Auditing and Assurance Standards Board (IAASB). 2016. Invitation to

Comment: Enhancing Audit Quality in the Public Interest: A Focus on Professional Skepticism, Quality Control and Group Audits.

Khan, S., Panatik, S., Saat, M., & Perveen, H. (2013). Auditors’ behavioral intention towards dysfunctional audit behavior applying theory of reasoned action. Jurnal Teknologi, 64(3), 153-158.

Knechel, W. R., Krishnan, G. V., Pevzner, M., Shefchik, L. B., & Velury, U. K. (2013). Audit quality: Insights from the academic literature. Auditing: A Journal of Practice &

Theory, 32(sp1), 385-421.

Kwon, S. Y., Lim, Y. D., & Simnett, R. (2010). Mandatory audit firm rotation and audit quality: Evidence from the Korean audit market. Unpublished paper, 1-61.

Lee, C. C., Welker, R. B., & Wang, T. W. (2013). An experimental investigation of professional skepticism in audit interviews. International Journal of Auditing, 17(2), 213-226.

Li, J., Burch, T. C., & Lee, T. W. (2017). Intra‐individual variability in job complexity over time: Examining the effect of job complexity trajectory on employee job strain. Journal of

Organizational Behavior, 38(5), 671-691.

Meier, K. J., & O’Toole, L. J. (2012). Subjective organizational performance and

measurement error: Common source bias and spurious relationships. Journal of Public Administration Research and Theory, 23(2), 429-456.

Messier, W. F., C. A. Simon, and J. L. Smith. 2013. Two decades of behavioral research on analytical procedures: What have we learned? Auditing: A Journal of Practice & Theory 32 (1), 139-181.

(34)

33 Mignonac, K., & Herrbach, O. (2004). Linking work events, affective states, and attitudes: An empirical study of managers' emotions. Journal of Business and Psychology, 19(2), 221-240.

Monitoring Commissie Accountancy. 2020. Eindrapport Spiegel voor de Accountancy sector.

Morgeson, F. P., & Humphrey, S. E. (2006). The Work Design Questionnaire (WDQ):

developing and validating a comprehensive measure for assessing job design and the nature of work. Journal of applied psychology, 91(6), 1321-39.

Morgeson, F. P., Mitchell, T. R., & Liu, D. (2015). Event system theory: An event-oriented approach to the organizational sciences. Academy of Management Review, 40(4), 515-537.

Nelson, M. W. (2009). A model and literature review of professional skepticism in auditing. Auditing: A Journal of Practice & Theory, 28(2), 1-34.

Nolder, C. J., & Kadous, K. (2018). Grounding the professional skepticism construct in mindset and attitude theory: A way forward. Accounting, Organizations and Society, 67, 1-14.

Olsen, C. (2017). A study of professional skepticism. Springer International Publishing.

Otley, D. T., & Pierce, B. J. (1995). The control problem in public accounting firms: An empirical study of the impact of leadership style. Accounting, organizations and society, 20(5), 405-420.

Peytcheva, M. (2014). Professional skepticism and auditor cognitive performance in a hypothesis-testing task. Managerial Auditing Journal, 29(1), 27-49.

Peytcheva, M., & Gillett, P. R. (2012). Auditor perceptions of prior involvement and reputation threats as antecedents of quality threatening audit behavior. Managerial Auditing Journal, 27(9), 796-820.

(35)

34 Pietsch, C. P., & Messier Jr, W. F. (2017). The effects of time pressure on belief revision in accounting: A review of relevant literature within a pressure-arousal-effort-performance framework. Behavioral Research in Accounting, 29(2), 51-71.

Professional Skepticism – meeting public expectations. (2018). IESBA

Public Company Accounting Oversight Board (PCAOB). 2008. Report on the PCAOB’s 2004, 2005, 2006, and 2007 Inspections of Domestic Annually Inspected Firms. Release No. 2008-008. Washington, DC: PCAOB.

Public Company Accounting Oversight Board (PCAOB). 2010. Report on Observations of PCAOB Inspectors Related to Audit Risk Areas Affected by the Economic Crisis. Release No. 2010-006. Washington, DC: PCAOB.

Public Company Accounting Oversight Board (PCAOB). 2013. Proposed auditing standards: The auditor's report on an audit of financial statements when the auditor expresses an unqualified opinion. August 13, 2013.

Quadackers, L., Groot, T., & Wright, A. (2014). Auditors’ professional skepticism: Neutrality versus presumptive doubt. Contemporary accounting research, 31(3), 639-657.

Rasso, J. T. (2015). Construal instructions and professional skepticism in evaluating complex estimates. Accounting, Organizations and Society, 46, 44-55.

Robinson, S. N., Curtis, M. B., & Robertson, J. C. (2018). Disentangling the trait and state components of professional skepticism: Specifying a process for state scale

development. Auditing: A Journal of Practice & Theory, 37(1), 215-235.

Semmer, N. (1982). Stress at work, stress in private life and psychological well-being. Mental load and stress in activity: European approaches, 42-55.

(36)

35 Shafer, W. E., Poon, M. C., & Tjosvold, D. (2013). Ethical climate, goal interdependence, and commitment among Asian auditors. Managerial Auditing Journal, 28(3), 217-244.

Shaub, M. K. (1996). Trust and suspicion; the effects of situational and dispositional factors on auditors' trust of clients. Behavioral Research in Accounting, 8, 154-174.

Sikka, P. (2009). Financial crisis and the silence of the auditors. Accounting, Organizations and Society, 34(6-7), 868-873.

Simunic, D. A. (1984). Auditing, consulting, and auditor independence. Journal of Accounting research, 679-702.

Simunic, D. A., & Stein, M. T. (1990). Audit risk in a client portfolio context. Contemporary Accounting Research, 6(2), 329-343.

Srinidhi, B., Leung, S., & Gul, F. A. (2010). Auditor tenure and audit quality: The role of the demand for unique client specific knowledge. SSRN Electronic Journal.

Steyer, R., M. Schmitt, and M. Eid. 1999. Latent state-trait theory and research in personality and individual differences. European Journal of Personality 13 (5): 389–408.

Svanberg, J., & Öhman, P. (2015). Auditors' identification with their clients: Effects on audit quality. The British Accounting Review, 47(4), 395-408.

Sweeney, B., Pierce, B., & Arnold Sr, D. F. (2013). The impact of perceived ethical intensity on audit-quality-threatening behaviours. Accounting and Business Research, 43(2), 112-137.

Tepalagul, N., & Lin, L. (2015). Auditor independence and audit quality: A literature review. Journal of Accounting, Auditing & Finance, 30(1), 101-121.

(37)

36 Tucker, R. R., Matsumura, E. M., & Subramanyam, K. R. (2003). Going-concern judgments: An experimental test of the self-fulfilling prophecy and forecast accuracy. Journal of Accounting and Public Policy, 22(5), 401-432.

Wang, X., Wang, Y., Yu, L., Zhao, Y., & Zhang, Z. (2015). Engagement audit partner experience and audit quality. China Journal of Accounting Studies, 3(3), 230-253.

Ye, P., Carson, E., & Simnett, R. (2011). Threats to auditor independence: The impact of relationship and economic bonds. Auditing: A Journal of Practice & Theory, 30(1), 121-148.

Zapf, D. (1993). Stress-oriented job analysis of computerized office work. The European Work and Organizational Psychologist, 3, 85–100.

A tense coexistence: how independence and client knowledge interrelate for a higher audit quality