The influence of tone at the top and incentives on the professional skepticism of an auditor

(1)

1 Amsterdam Business School

The influence of tone at the top and incentives on the professional

skepticism of an auditor

Name: Yafet Josef

Student number: 10885137 Thesis supervisor: dr. Peter Kroos

Date: 14 August 2016

Word count: 17402

MSc Accountancy & Control, specialization Accountancy Faculty of Economics and Business, University of Amsterdam

(2)

2 Statement of Originality

This document is written by student Yafet Josef who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

3

Abstract

The past years the professional skepticism of auditors has gained more attention in the media due to the audit failures that were highly visible which led to several large accounting scandals like Enron and WorldCom. From the analyses of fraud-related restatements by the U.S. Securities and Exchange Commission (SEC), it is concluded that the reason that auditors failed to detect material

misstatements is that auditors had a lack of sufficient professional skepticism during the audit. This thesis examines two potential variables that may foster professional skepticism. First, if tone at the top within an audit firm improves the professional skepticism level of auditors. Secondly, the influence of the incentive compensation at the client firm on the professional skepticism level of auditors is examined, as well as the interaction effect of the tone at the top within the audit firm and incentive compensation at the client firm. The findings of the first measurement of skepticism in time for the case-based experimental research based on a case showed that neither the tone at the top nor the type of incentive compensation at the client firm improves the professional skepticism level of auditors. Additionally, there appears to be no interaction effect between tone at the top and incentive compensation. However, the subsequent measurement of skepticism found that skepticism of auditors improved when the client firm had monetary incentives for the senior management. These findings contribute to the literature by examining two determinants, namely tone at the top within the audit firm and client firm incentive compensation of one specific auditor competence namely professional skepticism. Second, to the stream of literature that argue that greater monetary incentives of senior management increase the likelihood of accounting irregularities. Finally, to the capital market that assurance provided does not significantly change by the tone at the top and when there are great monetary incentives at the client firm.

(4)

4

Acknowledgements

I would like to take the opportunity to thank my thesis supervisor dr. Peter Kroos for his guidance throughout the process, motivation, patience and knowledge. This thesis would never have been accomplished without his helpful advice and feedback. The door to dr. Kroos was always open whenever I ran into a trouble or had a question about my research. I could not have imagined having a better supervisor for my thesis.

I also want to thank my supervisor Peter van Geest for his help, motivation and advice. He kept me focused by asking critical questions about the operationalization of my research. As well for Peter van Geest his door was also always open whenever I had questions or ran into a trouble spot.

Finally, I would like to thank the partners and executive directors of Ernst & Young for the assistance with approaching participants for my experiment and also the participants of my experiment.

(5)

5

1. Introduction

This paper investigates whether tone at the top within an audit firm and incentive compensation of the audit client influences the professional skepticism of an auditor.

1.1 Background

Professional skepticism has different definitions in the literature. According to Quadacker (2014) there is no universally accepted definition of professional skepticism, he considers professional skepticism as an essential element of the audit. Nelson (2009) defines professional skepticism as: “auditor judgment and decisions that reflect a heightened assessment of the risk that that an assertion is incorrect”. The AICPA (2002) defines the professional skepticism as “an attitude of the auditor that include a

questioning mind”. Glover and Prawitt (2013) define professional skepticism as “an attitude that not only include a questioning mind but also a careful observation and looking beyond the obvious”. They also mention that

an auditor need to be skeptic during situations where complex business transactions, varying

incentives and motives, and evidence collection and evaluation are playing a role (Glover & Prawitt, 2013).

A final definition to be discussed is that of Hayes et al. (2014), they defined professional skepticism as “an attitude that includes a questioning mind, being alert to conditions which may indicate possible misstatements

due to error or fraud and critical assessment of audit evidence’’. This assumes that auditors need to adopt

skeptical behavior when they are conducting an audit procedure (Noviyanti & Winata, 2015). This also implies that the auditors should be not easily satisfied with the audit evidence that they receive from the client.

A subsequent question is why some auditors display an insufficient amount of professional

skepticism. Prior research by Hurtt (2010) indicates that questioning mind, suspension of judgment, search for knowledge, interpersonal understanding, self-esteem and autonomy are important

determinants of the degree in which auditors exhibit skepticism, little research looked into the tone at the top, which means to what extent the senior managers at the audit firm set the right example. The scarce attention that tone at the top received is supported given that standards setters suggest that the proper tone at the top must be set by the partners to increase the likelihood that auditors will be professional skeptic (AICPA, 2003).

Furthermore, prior evidence on audit failures and breaches in auditor independence suggest that incentives for manipulation play an important role, both at the audit firm and the client firm. For example Tepalagul and Ling (2015) shows that the increasing levels of monetary incentives at the

(8)

8 client firm are associated with an increased likelihood of accounting restatements. This also suggest that auditors should be vigilant when reviewing evidence if management, that is ultimately

responsible for the production of accounting information, have large executive compensation packages. So the importance of professional skepticism is that it is necessary for the critical

assessment of audit evidence. The auditor should not be satisfied with less-than persuasive evidence because of the belief that management is honest.

In this research, I will investigate if professional skepticism of an auditor is influenced by the tone at the top within the audit firm and by incentive compensation at the client.

1.2 Research question

Even though professional skepticism is defined as an important characteristic there is only limited research on factors that may foster professional skepticism. Therefore it can be interesting to examine the following research question:

Does tone at the top within the audit firm and incentive compensation at the client firm influence the professional skepticism of an auditor?

1.3 Relevance of the question

In the auditing literature it is known that the concept of professional skepticism is important, however there is very little research on what professional skepticism is, and what could have an influence on the professional skepticism (Hurtt, 2010). The purpose of this thesis is to investigate two potential variables that may foster professional skepticism, specifically whether tone at the top within an audit firm and incentive compensation at the client firm influences the professional skepticism of an auditor. Prior audit research emphasis the importance of auditor independence and auditor competence in determining audit quality. Most of the prior literature examining audit quality focused on auditor independence. I contribute to this literature by examining one specific dimension of auditor competence, namely professional skepticism.

Second, the literature on skepticism is rather limited where prior work by Noviyanti and Winata (2015) identified knowledge of fraud and skeptical attitude as the main determinants of professional skepticism. I contribute to this research by examining two potential determinant of professional skepticism, namely tone-at-the-top and client firm incentive compensation.

(9)

9 Third, most literature on the relation between incentive compensation and accounting irregularities document a positive relation. This research documents that greater monetary incentives of senior management is associated with an increased likelihood of accounting irregularities. I contribute to this stream of literature by examining a potential countervail power. That is, do stronger monetary incentive compensations prompt a more vigilant approach by auditors.

Finally, this thesis also has a societal contribution. The assurance provided by auditors is important for a proper functioning of capital markets. Professional skepticism by auditors in this respects is an important component. So my research will also generate important findings to infer how professional skepticism is influenced so capital markets participants can infer whether auditors can play their assigned role.

1.4 Structure of the thesis

In the second chapter the prior literature and the hypothesis development will be discussed. In the third chapter the research methodology will be discussed. In the fourth chapter the results of the experiment will be presented. In the fifth chapter the conclusion and the limitations of this research will be presented.

(10)

10

2. Literature review and hypothesis development

2.1 Financial reporting

2.1.1 The need for financial reporting

An organization needs to provide accounting information in a market based economy (Beyer, Cohen, Lys, & Walther, 2010). This demand for accounting information arises from the information asymmetry and conflicting incentives between management and shareholders (Healy & Palepu, 2001). The need for accounting information is necessary for the shareholders to evaluate the return of the investment the shareholders made in an organization. Because of the separation of ownership and control, investors are at a significant distance of the firm. Alternatively formulated the managers of the firm possess private information that can be used at the expense of the shareholders.

To solve the problem, the information asymmetry needs to be reduced. Information asymmetry lead to two main problems namely moral hazard and adverse selection.

Models of moral hazard suggest that managers want to maximize their own welfare this can mean that managers can make investments that are not always in the best interest of the shareholders (Biddle, Hilary, & Verdi, 2009). The problem that moral hazard refers to is that agents has private information about effort choices. This misalignment of interest between principal and agent will cause and imbalance of power.

On the other hand models of adverse selection suggest a situation where there is lack of symmetric information between managers and investors. The models of adverse selection suggest that managers are better informed than investors about a firm’s prospects. Due to this information advantage manager will try to time the issue of capital to sell securities that are overpriced (Biddle, Hilary, & Verdi, 2009). Akerlof (1970) states in his research that the need for information arises from the interaction between quality heterogeneity and information asymmetry. The problem that arises according to Akerlof (1970) is that in the capital markets bad investments claimed to be as good as the good investments. This will finally result in undervaluation of the good investments and overvaluation of the bad investments (Healy & Palepu, 2001). This is also called the lemons problem (Akerlof, 1970). So both models emphasize the information asymmetry between the so called principal (shareholders) and the agent (e.g., CEO or management). To decrease the information asymmetry between the managers and the investors and shareholders of an organization, firms periodically disclose financial reports.

(11)

11 According to the Financial Accounting Standards Board (2010) the purpose of financial reporting is “to provide financial information about the reporting entity that is useful to existing and potential investors, lenders, and

other creditors in making decisions about providing resources to the entity”. Decisions that are made could involve

buying, selling or holding equity and debt instruments and providing or settling loans and other forms or credit. The reason why financial reporting is introduced is to deliver financial information to the investors and shareholder of a business. Investors and shareholders have a demand for financial information because of the existence of the previous discussed problem of information asymmetry between managers (agent) and the shareholder (principal) of an organization. Healy and Palepu (2001) suggest that there is a demand for information intermediaries who are engaged in the production of private information. For an organization there are two reasons why there is a demand for financial disclosure. The first reason is to solve the moral hazard problem. Shareholders want to verify if the manager has handled in the best interest of the shareholders. The second reason is to solve the adverse selection problem. As an investor or shareholder there is a demand for high quality information about the organization, to make informed decisions about the buy or sell of stocks. So due to the financial disclosure there will be more precise information for the shareholders and investors, which will assist among other investors in predicting firm value and enhance the speed with which the stock market responds to the information content of financial disclosures (Scott, 2009). This finding is confirmed by prior study by Leuz and Verrecchia (2000) who stated that information asymmetry that causes moral hazard and adverse selection could be mitigated by a higher quality of financial reporting. The effect that the higher quality of financial reporting will have is that it will enhance the investment efficiency. So to solve the information asymmetry an organization needs to disclose financial information periodically.

2.1.2 Role of the accountant/auditor

In general there are two types of auditors namely the internal auditor and the independent external auditor. One of the roles that the internal auditor fulfill is providing assistance to the management in the day to day operations. Furthermore the internal auditor provides also assistance in the determination of long range policies (Rampy, 1952). On the other hand the external auditor fulfill nowadays the role of independent assurance provider over the credibility of accounting information (Defond & Zhang, 2014; Knechel et al., 2006). There are two reasons why the providing of assurance is important. The first reason is that the reader of the financial reports put a considerable reliance on

(12)

12 the accuracy of the audited financial statements. The second reason is that the assurance that the auditor provides fulfill a certain security role for the principal (e.g. shareholders)(Watts & Zimmerman, 1986). The principal has an interest in monitoring financial reporting systems if they produce timely, relevant and credible information about the firm’s current- and future cash flow and the risk the organization is exposed to. Investors rely on the financial statements for their integrity (Bazerman, Morgan, & Loewenstein, 1997). To give the investors, shareholders and creditors a certain amount of confidence in the financial statements, they expect that the audit is performed with a certain independence of the auditor. According to Watts and Zimmerman (1986) the importance of providing assurance of the accounting system is therefore to reduce the agency conflicts.

2.1.3 Whence consist the role of the auditor?

The role of the auditor can be divided in three groups namely: (1) assurance-, (2) consulting- and (3) participative role (Huibers, 2013). There are three types of audits that can be classified in (1) audits of financial statements, (2) operational audits and (3) compliance audits. The key role of external auditor consists of (1) auditing the financial statements and other financial reporting, (2) attesting internal control statements and (3) review or attest corporate governance statements. To obtain reasonable assurance that the financial statements and other financial reporting are free of material misstatements the auditor needs to gather sufficient appropriate audit evidence (Hayes, Dassen, Schilder, & Wallage, 2014). The reason why the auditor needs to gather sufficient audit evidence is to reduce the audit risk to an acceptable low level which will result that the auditor can draw reasonable assurance which will be the base of the opinion the auditor will give. Based on the findings of the auditors they can provide one of the four types of opinions that the auditor got.

The first opinion the auditor can give is the qualified opinion if there are material misstatement in the financial report but these are not pervasive. With this opinion the audit states that the financial statement is prepared in accordance with the applicable financial reporting standards and that there are no pervasive material misstatements in the financial report. This opinion also gives the signal that the auditor had sufficient independence, professional care and was proficiency. Also that the field work has been properly planned and supervised which has resulted in an sufficiency understanding of the environment and internal control system and sufficient audit evidence. The second opinion is the qualified opinion with emphasis of matter paragraph or other matter paragraph. This opinion draw the users attention to an important matter in the financial statement. It indicate that there is uncertainty

(13)

13 which may affect the financial statement. So there is a doubt about the going concern assumption. The third opinion is the modified opinion. There are three forms of the modified opinion. The first is (a) qualified opinion this conclude that the financial statement is not free from material misstatements but not pervasive and that the auditor is unable to obtain sufficient appropriate evidence. The second (b) is the adverse opinion the auditor conclude that misstatements are material and pervasive. The third (c) is the disclaimer of opinion when the auditor is unable to obtain sufficient appropriate audit evidence and that the undetected misstatements could be material and pervasive (Hayes, Dassen, Schilder, & Wallage, 2014). So the auditors are charged with providing assurance whether their client have been prepared their financial reports in accordance with IFRS or GAAP and issue on base of their findings their opinion (Moore, Tetlock, & Tanlu, 2006).

When providing assurance it is important that the accountant keeps in mind the code of ethics. The importance of the role of the auditor is because a reader may place a considerable reliance on the accuracy of audited financial statements. The users of the financial statements expect a certain independence of the accountants. Besides independence, professional skepticism is also an important characteristic of the auditor. The auditor is required to comply with ethical requirements that are set out by ISA 200 which relates to the audit of financial statements (Hayes, Dassen, Schilder, & Wallage, 2014). The auditor shall plan and perform an auditor with professional skepticism. So the auditor needs to be competent in recognizing circumstances where there may exist material misstatements in the financial statements. In other words this will imply that the auditor cannot assume that the information provided by the client is trustful. This is because there can be a difference of interests between the managers of an organization and the shareholders. The manager could manage some earnings to achieve the objectives that are set for the organization to receive an incentive compensation in the form of an bonus or stocks (Hayes, Dassen, Schilder, & Wallage, 2014). Therefore the auditor needs to obtain a certain amount of professional skepticism. Professional skepticism will lead that the auditor ask questions about any clue to collect enough audit evidence.

2.2 Professional skepticism

2.2.1 Professional skepticism definition

The importance of professional skepticism is that it will lead an auditor to question about any clue to collect enough audit evidence. During the audit auditors are required to adopt the attitude of professional skepticism (ACCA, 2015). Professional skepticism is therefore defined as an important

(14)

14 characteristics that the auditors needs to apply. It is also a concept that is widely accepted but there seems to be a lack in the literature of a clear understanding of what constitutes professional skepticism (Hurtt, 2010). The understanding of what professional skepticism is can be broadly interpreted. As a result of the inconsistency in defining professional skepticism there is no universal definition of professional skepticism in the literature. Nelson (2009) has found during the review of the definitions that professional skepticism was interpreted in two ways namely neutral and presumptive doubt.

Hurtt (2010) defines professional skepticism in six characteristics of skepticism namely questioning mind, suspension of judgment, search for knowledge, interpersonal understanding, self-esteem and autonomy. McMillan and White (1993) defined professional skepticism as conservatism bias

in audit judgments, referred to as a hypothesis frame. Shaub (1996) defines professional skepticism as “one who instinctively or habitually doubts, question or disagrees with assertion or generally accepted conclusions’’. Nelson’s

(2009) defines professional skepticism as: ‘’an heightened assessment of the risk that an assertion is incorrect

conditional on the information available to the auditor’’.

Nelson (2009) has found during the review of the definitions of professional skepticism that the definition of Hurtt has a neutral perspective and that the definition of Shaub has a presumptive doubt interpretation. According to the definitions they share the idea that professional skepticism is an attitude of the auditor to have an questioning mind, critical assessment of audit evidence.

To measure professional skepticism Hurtt (2010) developed and tested a scale that measures whether an auditor is skeptic or not. The Hurtt scale is based on six proxies that are characteristics of skepticism and 30 items which has to be filled in on a scale from 1 strongly disagree to 6 strongly agree to determine whether an auditor is skeptic or not. The range of the score differ from 30 to 180. The characteristics that Hurtt bases her scale on are based on the literature and auditing standards. She defines and describes the characteristics as follows: (1) questioning mind, (2) suspension of judgment, (3) search for knowledge, (4) interpersonal understanding, (5) confidence and (6) self-determination. Before the model of Hurtt (2010) professional skepticism in the accounting research was usually measured by using scales which was intended to measure constructs as trust (Shaub, 1996), independence (Shaub, 1996), or suspicion (Shaub, 1996). However, Shaub found no significant relation between the scores of trustworthiness and independence on the professional skepticism of auditors.

2.2.2 Previous studies professional skepticism

(15)

15 Knapp and Knapp (2001) stated that it is hard to detect fraud when there is a lack of professional skepticism or no professional skepticism. The reason that the detection of fraud is harder is because the perpetrators conceal the fraud. Therefore the professional skepticism of an auditor is an

important behavioral characteristic to detect fraud or any material misstatements. One of the first studies in the field of professional skepticism which had an focus on external auditors and

information search was the research of McMillan and White (1993). They investigated whether confirmation bias and professional skepticism would be countervailing forces in the search for information. Their results were that there was a difference on the reaction of the auditors depending upon the frame of the hypothesis they favored and their belief extremity. In detail auditors who favored the frame of error had a more strong reaction to confirming and disconfirming evidence in contrast to those who favored the environmental frame. They also found that the differences in

behavior that are attributable to the professional skepticism of auditors is dependent on how the data is scaled.

Fullerton and Durtschi (2004) examined whether high levels of professional skepticism are correlated with behavior that might enable internal auditors to better detect fraud. They found that the auditors who had reported a higher skepticism had a desire for more information than those who rated themselves with low skepticism. This would imply that an auditor who is skeptic will analyze more information to give a funded opinion about the financial statements.

The researchers Hurtt, Eining and Plumlee (2008) made a distinction between auditor behavior in their research. They distinguishes behavior in (1) evidence assessment and (2) generation of alternative explanation. In their research they examined if there were any associations between professional skepticism and differences in the behavior of the auditor. They found that skeptical auditors found more contradictions and generate more alternative explanations, however they detected fewer mechanical errors and examine relatively less substantive tests than less skeptical auditors. Their finding is consistent with viewing professional skepticism as a trait. This means that auditors who has a higher level of skeptical behavior behave differently than those who are less skeptical even when the engagement is not seen as skepticism inducing. So they concluded that auditors with a higher level of professional skepticism behaved systematically different than less skeptical auditors.

Shaub and Lawrence (1996) argue that staff auditors demonstrate a level of professional skepticism in thoughts and behavior higher than seniors, managers and partners. In their research

(16)

16 they found that there were three significant factors in predicting the level of professional skepticism namely: (1) client situational factors, (2) higher concern with professional ethics and (3) experience. Asare and McDaniel (1996) investigated the effects of familiarity with the preparer and task complexity on reviewer’s effectiveness at detecting errors. They found that a high familiarity between reviewer and preparer can reduce professional skepticism and that a low familiarity can increase the professional skeptic behavior.

Shelton et al. (2001) investigated whether auditors of the former Big five audit firms adjusted professional skepticism to increased risk factors. They found that the professional skepticism of the auditors responded to the increased risk factors. But they did not indicate in their research whether any of the firms had an actual instruction what the adjustment could entail.

Robinson, Curtis and Robertson (2013) investigated the influence of the trait and state components of professional skepticism on auditor behavior. They also investigated whether the factors time pressure and goal frame would have an effect on the professional skepticism of auditors. They found that auditors higher in state professional skepticism display a higher level of skeptical behavior. Also the increase in state skepticism has a greater effect on auditors with a low trait skepticism than high trait skepticism. Second they found a positive relationship between time and professional skepticism.

In the research of Rittenberg (2012) is found that the professional skepticism was very high when students left college but as time passed by and the students moved from staff auditors to audit managers the professional skepticism decreased.

Nelson (2009) found in his study that there are three factors that have an influence on the skeptical behavior of an auditor. Nelson (2009) mentioned characteristics, knowledge and incentives as factors that influence the skeptical behavior. Overall, each of the discussed studies provides insight in professional skepticism. However, it is difficult to draw a conclusion about professional skepticism because it is not clear what the cause is of lack of professional skepticism (Curtis, 2014).

2.2.3 Positive versus negative determinants of professional skepticism

According to academic research there are several positive and negative determinants of professional skepticism. Like described earlier Hurtt (2011) mentioned that there are six determinants that could have a positive impact on the professional skepticism of an auditor if they are implemented well. These determinants are (1) questioning mind, (2) suspension of judgment, (3) search for knowledge, (4)

(17)

17 interpersonal understanding, (5) self-confidence and (6) self-determination. If these determinants are not available in the character of an auditor than according to Hurtt the auditor is less skeptic. Additionally, Hurtt et al. (2008) mentioned that there are skepticism inducing circumstances. Previous academic audit research also found that there is an interaction between professional skepticism and audit circumstances (Hurtt et al., 2008). Like discussed previously Shaub and Lawrence (1996) found that risk factors like for e.g. financially stressed clients could have an positive effect on the professional skepticism. Similarly, Quadacker (2007) found that there was an interaction between interpersonal trust and the management control philosophy.

According to Nelson (2009) there can be two determinants that increases and decreases professional skepticism. In his model Nelson (2009) describes that the influence of the engagement partner can be seen as an incentive to influence the skepticism of auditors. Nelson (2009) states also that an partner who emphasize efficiency can reduce professional skepticism and that engagement partner who emphasize effectiveness of the audit can increase the professional skepticism of an auditor. Additionally, Nelson’s determinants are in line with the determinant according to the standard SAS No.99. This standard emphasize the importance of professional skepticism. The standard requires that before an audit the auditors need to evaluate what the possibility is that there is fraud within a client firm. During the evaluation auditors are also required to note the risks which will help in maintaining the professional skepticism of auditors (AICPA, 2002). In accordance with SAS No.99 Carpenter and Reimer (2011) stated that the professional skepticism of an auditor is dependent on the fraud risk assessments.

Additionally, Glover and Prawitt (2013) argue that there are professional skepticism inducing circumstances like (1) professional licensing and continuing education requirements, (2) there must be supervision, mentoring, review and inspection of work and performance evaluations, (3) effective planning and audit programs, (4) performance metrics that reward people for high quality work, (5) stringent recruitment requirement, (6) effective engagement partner and leadership messaging and (7) training on core competencies and professional judgment. However, Glover and Prawitt (2013) argue that there are several potential threats to the professional skepticism of auditors. They made a separation of threats for 4 engagements. The first engagement is for all general engagements. They acknowledge 4 threats for this engagement namely: (1) judgment traps and biases, lack of knowledge and expectations, (2) deadline pressure, inherited preferences and expectations, (3) auditor character,

(18)

18 and personal and cultural attributes and (4) performance and compensation metrics and incentives that do appropriately encourage professional skepticism.

Consequences for the auditor if s/he is not skeptic is that the audit then usually is considered as an audit of low quality. An non skeptic auditor could have neglected unusual circumstances. Skepticism is also how strict the auditor sets his materiality. The auditor assess the materiality related to specific account balances and decide how the accounts are being examined. The consequence of not being skeptic is that the auditor will set in the planning phase the audit risk very low which will result in that the materiality will be set high. This could results in giving an audit opinion that is not suitable for the current situation of the firm that is being audited. By giving the wrong opinion shareholders could sue the auditor for negligence if they suffer a form of loss (ACCA, 2015) which will result in a costly litigation or losses to the reputation of the accountant firm.

2.3 Tone at the top and professional skepticism

2.3.1 Tone at the top

Tone at the top within the organization is probably the most important lever within an organization to set the share values to the employees, to create an ethical atmosphere and to realize the objectives of an organization. The ethical atmosphere that prevails within the organization finds its origin at the top level management. A strong tone at the top is therefore necessary to create and maintain and internal control system that is effective and helps in preventing fraud and other behavior that are not ethical at all levels within organizations (Staicu, Tatomir, & Linca, 2013). This can be confirmed by the finding of Trevino et al. (1999) who found that if the tone don’t set the right example, ethical behavior will not be adopted by the rest of the organization.

In the academic literature several studies have described a definition of tone at the top. Cunningham (2005) describes tone at the top as: “a set of share values that are propagated by the most senior

staff, the tone at the top is the reflection of the actions of these seniors”. Roth (1997) defines tone at the top as:

‘’management’s philosophy and operating style; management’s understanding and management of risk; communication’’. Sklar (2009) defines tone at the top as: “the visible willingness by top management to prioritize corporate values

above other values in decisions making and to expect all others in the organization to do the same’’. According to

D’Aquila and Bean (2003) the tone at the top could have an influence on the behaviors and decisions of auditors.

(19)

19 According to the report of the AFM (2013) a proper tone have eye for what is needed to achieve a good quality. A good tone will also take the measures that are necessary to achieve the objectives. If employees sees the manager bending the rules they will follow the set example and will bend the rules as well (Cutler, 2004). Tone at the top is important because it has an contribution in realizing the objectives of organizations it refers also to the ethical atmosphere or organization culture that is created by the head of the organization (D'Aquila & Bean, 2003). According to the definitions that are discussed tone at the top can be described as the mind-set and acting of managers to act in the most desirable in the organization because it sets the tone for all the others in the organization.

2.3.2 Previous studies tone at the top and hypotheses development

Like discussed previously the tone is set by top level management. In the academic research there are several researchers who investigated the influence of tone at the top. D’Aquila (1998) provided evidence that there is an association between a high tone at the top and a higher quality for the financial reports. This implies that a high tone at the top has a positive effect in improving financial reporting and that it can also impact the operating performance. In another study D’Aquila and Bean (2003) found that an unethical top promotes an unethical decision making for employees. This implies that the tone at the top is important in creating an ethical environment to promote ethical decisions. Ethical decisions would also imply that managers improve the financial reporting quality. Additionally, Margheim and label (1990) found that the integrity of the management has an influence on the reliability of the external auditors work performed.

Berson et al. (2008) found that the CEO values were indirectly associated with the financial performance of an organization. Their first finding implies that the core values like characteristics of the CEO will translate to the organizational culture and is associated with subjective as objective aspects of performance. Additionally, Wang (2015) found that internal auditors are less concerned about management control issues that could arise when there is a poor tone at the top. Her finding suggest that corporate ethics and the audit committee quality has an important role promoting the objectivity of the auditor. This implies that the quality of the work of the internal auditor could increase by simply improving the tone at the top and the quality of the audit committee.

Asare et al. (2008) found that internal auditors are consistent sensitive to any variation in the management’s performance incentives to misreport financial information. Auditors are also sensitive to an increased budgeted work hours when the management has a high incentive to misreport. This

(20)

20 implies that internal auditors are conscious about factors that may have an impact on the tone that is set in the organization. Based on the factors they recognize the internal auditors adjust their audit plan. Asare et al. (2008) have not considered explicit how tone at the top is related to internal auditing.

Carpenter and Reimers (2011) investigated the effect of a partner’s emphasis on professional skepticism and the presence of fraud on auditors identification of fraud risk factors, auditors fraud risk assessment and their selection of audit procedures. In their research they made an distinction between high- and low tone at the top. To operationalize the high and low tone at the top they made use of Nelson (2009) statements. He stated that an partner who emphasize efficiency reduces the professional skepticism of an auditor. The second statement Nelson made was that an partner who emphasized effectiveness would increase the professional skepticism of an auditor. So the low tone at the top is operationalized as an partner who emphasize efficiency and the high tone at the top is an partner who emphasize effectiveness. The findings of their study was that the auditors who worked for the partner who emphasized effectiveness provided a higher fraud risk assessment than the auditors who worked for the partner who emphasized efficiency. Even when there was no fraud the auditors in the high tone at the top provided a higher risk assessment than those who worked in the low tone at the top. Similarly, Dickins et al. (2008) provided evidence for that tone at the top plays an important role in determining the riskiness of an client and that external auditor would charge more for clients that are typed as risky. Other research that is conducted in the field of tone at the top is the research of Noviyanti and Winata (2015) they investigated the influence of skeptical attitude, tone at the top and knowledge of fraud on auditors professional skeptical behavior. They found that an high tone at the top played an important role in maintain- and improve skeptical behavior for auditors who had a weak skeptical attitude. So they state that the skeptical attitude of an auditor could improve if there is a high tone at top.

According to the report of the AFM (2013) a good tone at the top is necessary and has a significant influence on the company culture of the audit firm. However the behavior of auditors more specially the behavior of the tone at the top is faced with two dilemma’s. The first dilemma they face is that the auditor need to be objective and skeptic during to the audit to provide a high quality audit. But on the other hand auditors wants to satisfy the client, in order to retain or obtain the engagement. However Peecher (1996) and Turner (2001) have investigated the influence of the partner emphasis on the professional skepticism of auditors. They both found that the emphasis of the partner on professional skepticism did not had an influence on the auditor’s judgment. An important characteristic

(21)

21 of a high tone at the top at the audit firm is that the top should emphasize the professional skepticism of an auditor (Carpenter & Reimers, 2011). Therefore based on the discussion above I hypothesize that a high tone at the top is positively associated with professional skepticism. Specifically, I form the following hypothesis:

H1: Tone at the top is positively associated with professional skepticism for auditors.

2.4 Client incentive compensation within the client firm

2.4.1 Previous relations incentive compensation

Within an organization the top receives a fixed salary, short-term incentive and a long-term incentive. Incentive compensation within an organization is usually an compensation that is highly dependent and based on the performance of an entity. Organizations provide incentive compensation to attract- and retain key employees. It is also used to motivate appropriate managerial behaviors and to influences organizational performance (Han Ming Chng et al., 2012). According to Gilson and Vetsuypens (1993) incentive compensations is an important part of the turnaround strategy of many declining organizations. However, managers can apply earnings management to achieve their targets and to receive or increase their incentive compensation. When auditing an organization with incentive compensation, auditors are the most effective when they understand the motivation behind earnings management. Earnings management is usually concealed for the auditor because there can be a possibility that the auditors gives an opinion based on this events that is unfavorable for the client.

Prior research by Otto (2014) found that optimistic CEOs received a smaller incentive compensation than their peers. Additionally, Finkelstein et al. (2009) found that equity ownership is important for resolving problems between principal and agent. Their study found also that if the variable compensations of managers is highly dependent by the value of the company the manager has a great incentives to perform the task that are desired from him. Contrary to this is that the variable compensation of managers whose compensation is not dependent by the performance of the organization have little incentives to perform the task that are desired from the manager.

Schneider (2010) found that internal auditors were reluctant to extend audit procedures. Even incentive compensation and stock ownership affected the decisions of the auditors and the planning of the audit. Schneider (2010) confirms his prior finding who argued in an prior paper (2003) that earnings based incentive compensation and stock owner do not affect the internal auditors reporting

(22)

22 about violations of generally accepted accounting principles. This is because auditors have their own standards they have to comply with. However, Oyer (2004) highlights a different side of incentive compensation namely why firms implement forms of incentive compensation like stock option or other pay instruments that reward “luck”. In his research he derived an model that provides reconciliation of agency theory with that employees are often rewarded or punished for things they cannot control. This model assumes that it is costly to make adjustments to the compensation scheme of employees. In his research he also shows that variability in an employee reservation utility could lead that the firm wants to transfer the risk to the worker as an mean of insuring participation during various states of the economy (Oyer, 2004).

2.4.2 Relation between incentive compensation and earnings management

In the academic literature several researchers provide evidence that when there are incentive compensation within an organization, this may increase the likelihood of (opportunistic) earnings management. Earnings management occurs according to Healy and Wahlen (1999) “when managers use

judgment in financial reporting and in structuring transactions to alter financial reports to either mislead some stakeholders about the underlying economic performance of the company, or to influence contractual outcomes that depend on reported accounting numbers”. Earnings management occurs according to Dye (1998) “when there is the existence of information asymmetry between the management and shareholders”. Also Schipper (1989) acknowledged the

occurrence of earnings management when there is persistence information asymmetry. Another condition for the occurrence of earnings management is the existence of agency cost. According to Jensen and Meckling (1976) there could be conflicting interest between the agency and the principal therefore principals use contracting to motivate agents. Watts and Zimmerman (1986) argues that bonus plans, debt covenants and other factors provides incentives to make strategic accounting choices. If the management wants to maximize their bonus the management can make use of income increasing accruals (Healy, 1985). Additionally, Burgstahler and Dichev (1997) found evidence for the link between earnings management with income smoothing and incentive compensation. These authors also showed that organizations with incentive compensation avoid negative earnings.

According to Healy (1985) managers are related to the nonlinear incentives in their bonus contract. Healy (1985) states also that managers are more likely to select an accounting procedure and accruals to increase their incentives. According to the study of Healy (1985) there can be concluded that there is a strong incentives for managers to report a higher earnings to increase their cash

(23)

23 compensation. However, other research by Holthausen et al. (1995) and Gaver et al. (1995) reexamined the issue of short term incentives and earnings management. They found contrary to the research of Healy (1985) no evidence for income decreasing accruals by managers when there earnings are below on target to receive an bonus. Additionally, Guidry et al. (1999) found that managers manipulate earnings during a period when they have incentives. An example of a period when the company or the managers have incentives is when shares are sold to the capital market. Additionally, Francis et al. (1999) found also that manager with large accruals have incentives to choose high quality auditors to credibly manipulate the financial reporting process.

Schrand and Zechman (2010) examined if overconfidence as an manager trait is associated with fraud. The authors have measured the amount of earnings management by comparing the reported earnings against unmanaged earnings. They found that the amount of earnings management increased over time. They, amongst others, found that managers that started out inflating earnings by tampering with the accruals over time resort to more dramatic measures (with an increased likelihood of fraud) to sustain the higher reported accounting numbers.

2.4.3 Client incentives and professional skepticism

For example stock options could give the management incentives to boost the stock price at the expense of the long term viability. Prior literature has documented that firms often engage in accounting manipulation to inflate reported accounting numbers at the expense of the credibility of the numbers in the financial statements. Auditors need to be independent when reviewing a firm financial report. Levitt and Dwyer (2002) have found that auditors who audit organizations with incentives are complicit with the firm management. Additionally, Moore et al. (2006) states that accountant firms avoid negative audit opinions to the managers that hire the accountant firm and pay their audit fees.

Hirst (1994) investigated if auditors are sensitive to management incentives to manage earnings. He conducted an experiment where he tested whether bonus-induced incentives would affect the auditors judgment of the probability that there would exist an material misstatement. The auditors which he included in his experiment agreed that bonus plans create incentives for the management to manage earnings. However, his experiment found that the auditors judgment was unaffected by whether the unexpected difference and management incentives were congruent.

(24)

24 Based on the research of Hirst (1994), I will test whether incentive compensation within an organization will affect the professional skepticism of the auditor. Theoretically, auditors assess the client environment when providing assurance on the financial statements. Lavish executive compensation packages that may increase the likelihood that managers may engage in accounting manipulation to increase the proceeds from their compensation plans at the expense of outside investors can represent a warning signal for auditors. Increased professional skepticism may be on the auditors safeguards to protect their reputation in client environments characterized by an increased likelihood of accounting irregularities. For example, the behavior of income smoothing for managers whose bonus is linked to the targets and earnings an organization achieves makes sense and thus auditors need to increase their skepticism. So, I formulate the following hypothesis:

H2: Incentive compensation at the client firm is positively associated with professional skepticism for auditors. Finally, the question is how tone at the top and the incentive compensation at the client firm interact in explaining professional skepticism. That is, I will examine whether incentive compensation at the client firm will influence the relationship between tone at the tone at the top and auditors’ professional skepticism levels. I expect that the returns of a high tone at the top in terms of an increased professional skepticism are higher when auditors, indeed, face potential hazards for the successful realization of an effective audit engagement. I expect a complementary relation between on the one hand tone at the top within the audit office and on the other hand potential risk factors such as excessive executive compensation packages contingent on accounting numbers. In short, the expectation is that the positive relation between tone at the top and skepticism is stronger (more positive) when there is more executive compensation at the client firm. Specifically, I form the following hypothesis:

H3: Incentive compensation at the client firm will positively influence the positive relationship between tone at

(25)

25

3. Research method

3.1 Research design

For this research I conduct a case-based experiment. The participants of this experiment had to complete an evaluation task wherein the participant indicates how much time they would spend with the audit. The participants fulfilled the role of the auditor that audits a Dutch company that manufactures- and sells mobile phones, the company’s results are under pressure and the firm has taken some potentially controversial accounting decisions. One of the task of the auditors was to evaluate how much time they would spend with the audit of this client relative to a client firm with similar general characteristics (the scope and complexity of the client firm). Second, the auditors had to fill in a questionnaire on professional skepticism with 10-items to what extent they disagree or agree with the item on a six point scale differing from strongly disagree to strongly agree.

From the point of view that an auditor who is professional skeptic spent more time on the audit, I use time as a measure for professional skepticism. As a validation of the time measure, I use a second measure that I modified based on the Hurtt scale to measure professional skepticism. From the Hurtt scale I have chosen 10-items that are context-specific as they might be influenced by the provided case information. The removed items referred to how skeptic they were as an individual. This would provide information on their innate skepticism but should not be related to the specific manipulations.

The participants received information about the events during the fiscal year 2015, incentive system and about the audit. The information of events during the fiscal year 2015 was representing the situation the company was in. It included information about the profit of 2014, the expectations for 2015, the sales compared to previous years, the cash flow from operating activities in comparison to previous years, the inventory position, the provision for bad debts, the impairments and the preliminary accounting numbers for 2015. This information of the events during the fiscal year is for all the conditions the same. In all experiments I manipulate between participants: (1) tone at the top

(high tone at the top which emphasis the effectiveness of the audit or low tone at the top which emphasis the cost and time efficiency) and (2) the presence of incentive compensation within the client firm (incentive compensation or no incentive compensation).

This experiment employs a 2 × 2 factorial design (between subjects) to test the tone at the top influence and incentive compensation at the client on the professional skepticism of an auditor. I will

(26)

26 use the ANOVA to compare 4 groups in combination with an independent variable. The dependent variable in this research is the professional skepticism of auditors. Table 1 summarizes the manipulations of the 4 conditions. For the collecting of data for this research I have randomly assigned one of the 4 cases to the participants in my experiment with the Qualtrics software because this software offers the function to randomize the cases within the participants. Finally, the four case descriptions can be found in appendix A.

TABLE 1 Conditions

1) Incentives 2) No incentives

1) High tone at the top Condition 1(group11) Condition 2 (group 12) 2) Low tone at the top Condition 3 (group 21) Condition 4 (group 22)

3.2 Sample

This study is an experimental research. The total number of auditor participants were 124 whereof 110 have completed and 75 have passed the manipulation check. Therefore, the three hypotheses are tested by using data from 75 auditors working within Ernst & Young. The auditors were invited by a direct mail by their partner, in which they were asked to participate voluntarily in the experiment by clicking on the survey link added in the e-mail. The participants that are used for this experiment are auditors who were able to remember what the concern was of the engagement partner and if there were strong variable bonuses or highly fixed fees. Furthermore, for this experiment the choice is made to involve the following participants: (1) trainee, (2) staff, (3) senior staff, (4) managers, (5) senior managers and (6) directors. The trainee, staff and senior staff are directly related to the audit because they are the practitioners of the audit. They also deal with what the tone expect from them, because their supervisors -who received the expectations from the engagement partner- pass it on to them. The fourth and fifth group consisting of managers are also directly related to the audit because they are the reviewers of the performed audit work by the staff. Based on the orders and expectations the managers receive from the top they will review the audit keeping in mind what the engagement partner wants. Another reason why managers are appropriate is because they have experience in assessing the potential risk that could occur at the client firm. This is also confirmed by Carpenter (2007) who found that managers are effective in assessing risk. Therefore, for this experiment the data of 75 auditors

(27)

27 comprised 59 males (78,7%) and 16 females (21,3%) from trainee, staff, senior staff, managers to directors working within Ernst & Young that have participated in this experiment is used for the hypotheses testing. In table 2 an overview is summarized of the staff level of the participants with the number and expression of the percentage of the total participants.

TABLE 2 Participants

Staff level N Percent

Trainee 2 2,7 Staff 31 41,3 Senior staff 28 37,3 Manager 7 9,3 Senior manager 5 6,7 Director 2 2,7 Total 75 100 3.3 Experimental task

The 4 cases consisted of the same introduction and the same events during the fiscal year 2015. The task for the participants was also the same for the 4 cases. Participants were asked to fulfill the role of the auditor of the Dutch company that is active and specialized in the manufacturing and sales of mobile phones. This company got 68 locations over the world with over 52.000 employees and is at risk of becoming obsolete of the inventory.

The information of events during the fiscal year 2015 was representing the situation the company was in. The participants received the information that in 2014 there was a low profit. The company believed that it would be better for 2015. In 2015, the operational results were less than hoped for by directors and managers. Total sales were a bit less compared to 2014 and the proportion of sales on credit increased. The cash flow from operating activities has been lower than previous years. The inventory position has increased. The provision for bad debts did not change relative to prior years. The preliminary accounting numbers for 2015 suggest that the profit has increased slightly relative to 2014. For 2015, the firm recorded less impairments on company assets. The goal of this information was to trigger the skepticism of the auditors by providing information that could indicate that management of the Dutch company was managing the earnings. This information was the same for all cases. Furthermore, the participants received information about the incentive system of which the company made use of and information about the audit wherein the concern of the engagement

(28)

28 partner was included. This information was manipulated. How this information was manipulated will be elaborated in paragraph 3.4. Finally, after reading the case carefully each auditor was asked to indicate how much time they would spend with the audit for the year 2015 of this client in comparison with a client of the same general characteristics on a scale from 1 to 10.

3.4 Independent variables

The independent variables tone at the top and incentive compensation at the client firm is given to the participants in a description. The information under audit indicated the tone at the top and in the information under the incentive system indicated the type of incentive compensation. I manipulate the tone at the top by providing information to the participants showing that the engagement partner have expressed his concern with the effectiveness of the audit and that the audit quality and the reputation of the audit are key (high tone at the top condition). To make the manipulation less obvious an sentence is added wherein included that the audit firm adopted a program that should improve the cost-awareness. Furthermore, I also manipulate tone at the top for the low tone at the top condition by providing information that the engagement partner expressed his concern with the efficiency of the audit and the cost awareness and the time efficiency of the audit firm are key. To make this manipulation also less obvious an sentence is added wherein is included that the firm recently adopted a program that should improve the quality of the audit engagements. This manipulation is partly based on the manipulation that Carpenter and Reimers (2011) have used.

Second, I manipulate incentive compensation at the client firm by providing information that the rewards the executive management of the client firm receives are strong variable bonuses and that the bonuses are highly dependent on the reported profit numbers. To make this manipulation less obvious an sentence is added wherein is included that the rewards are for a minor part based on customer satisfaction (incentives condition). Furthermore, for the absence of incentives at the client firm condition, I manipulate incentive compensation by providing information that the rewards for the executive management are high fixed salaries and that the management receives no bonuses.

3.5 Manipulation checks

To determine whether the auditors interpreted the operationalization of the tone at the top and the incentive compensation within the client firm conditions well, several manipulation checks were included in the case material. After filling in how much time the auditor would spend on the audit of

(29)

29 the client firm, the auditor was asked three manipulation checks (1) if the engagement partner expressed his concern with the effectiveness or efficiency of the audit, (2) if the rewards were strong variable bonuses or if the rewards were high fixed salaries and (3) how skeptic they were with the assignment. If the auditors answered consistently with the manipulation, then they passed the manipulation check and the data was used. Final the auditors were asked to fill in their age, gender, work experience in years and staff level. A summary of the variables that are used in my research, with their descriptive statistics are included in table 3.

TABLE 3 Variables used

Category and name Theoretical values Range Mean SD

Dependent variable Time Scale: 1-10 6 6,75 1,253 Subsequent measurement skepticism Score: 10-60 30 41,58 4,988 Independent variables

Tone at the top 0 = Low tone at the top; - - -

1 = High tone at the top - - -

Incentive compensation 0 = Fixed fees; - - -

1 = Strong incentive

compensation - - -

Manipulation check

Concern of the engagement partner 1 = Effectiveness - - -

2 = Efficiency - - -

3 = I don’t know - - -

Rewards client firm 1 = High fixed salaries - - -

2 = Strong variable bonus - - -

3 = I don’t know - - - Control variables Gender 0 = Female 1 0,79 - 1 = Male Age - 22 27,82 4,8 Work experience - 18 5,22 4,287

Staff level 1 = Trainee Staff

2 = Senior Staff 3 = Manager 4 = Senior manager 5 = Director

(30)

30

4. Results

4.1 Descriptive statistics

In table 4 the descriptive statistics is summarized for the first time measure of skepticism. Table 4 reports the number of participants as well as the average skepticism for each of the four groups and the totals for each of the four conditions. With respect to the first two hypotheses, one can see that the average time is almost similar for the high- vs. low tone at the top (7,00 vs. 6,54). The difference in skepticism is higher for the presence vs. absence of incentive compensation at the client firm. That is, consistent with my expectations, skepticism is somewhat higher when incentive compensation is present at the client firm. However, it remains the question whether the difference is significant. With respect to the third hypothesis, I expect the difference between high- vs. low tone at the top on skepticism to bigger when there is incentive compensation present at the client firm. The descriptive results seem to indicate the opposite. That is, the difference between high- vs low tone at the top (7,00 vs. 6,19 is larger for no incentives compared to the difference between high- vs low tone at the top (7,00 vs. 6,90) for incentives.

TABLE 4

Descriptive statistics skepticism in time

Condition 1) Incentives 2) No incentives Total

1) High tone at the top N = 17 N = 17 N = 34

Mean = 7 Mean = 7 Mean = 7

2) Low tone at the top N = 20 N = 21 N = 41

Mean = 6,9 Mean = 6,19 Mean = 6,54

Total N = 37 N = 38

(31)

31

4.2 Results of hypotheses test (measurement in time)

Hypothesis 1

The first hypothesis predicts that tone at the top is positively associated with professional skepticism for auditors. This first hypothesis is supported when the mean of the conditions 111_{and 12}2_{is higher}

than the mean of the conditions 213_{and 22}4_{, in this case this is the situation, because the mean for}

condition 11 and 12 is 7 and the mean for 21 and 22 is 6,54. However, to accept H1 this difference must be statistically significant (P < .05). Since, H1 tests the positively association with professional skepticism a one-way ANOVA is conducted.

Before conducting an ANOVA there are 6 assumptions5_{where the data has to meet. For this}

research I only focus on the fifth and sixth assumption, that is that the dependent variable should be approximately normally distributed for each category of the independent variable and that there need to be Homogeneity of Variances. According to table 5 the assumption for Homogeneity is violated and therefore the Welch and Brown-Forsythe tests (presented in table 6) are performed to check if the assumption can be met. The Welch and Brown-Forsythe are both not significant. Because the assumption of equal variances is violated, therefore I need to conduct a non-parametric test, in this case I will use the Kruskal- Wallis test.

TABLE 5

Test of Homogeneity of Variances

Levene Statistic df1 df2 Sig.

6,705 1 73 0,012

TABLE 6

Robust Test of Equality of Means

Statistic df1 df2 Sig.

Welch 2,754 1 71,832 0,101

Brown- Forsythe 2,754 1 71,832 0,101

1_{High tone with incentives} 2_{High tone no incentives} 3_{Low tone with incentives} 4_{Low tone no incentives}

5_{First assumption is that the dependent variable should be measured at the interval or ratio level. Second assumption is}

that the independent variables should consist of two or more categorical groups. Third assumption is that there should be independence of observations and the fourth assumption is that there should be no significant outliers.

(32)

32 TABLE 7

Test statistics Kruskal-Wallis Test

Task (Time)

Chi-Square 2,058

df 1

Asymp. Sig 0,151

Monte Carlo Sig Sig. 0,153c

99% Confidence Interval Lower Bound 0,144

Upper Bound 0,163

A Kruskal-Wallis test was conducted to compare the effects of tone at the top on the professional skepticism levels of auditors. The Kruskal-Wallis results show that there was not a significant effect of tone at the top on the professional skepticism levels of auditors. The effect of tone at the top yielded an p-value of 0.151 (see table 7), therefore I am not able to support hypothesis 1. Because the results were not significant there is no need to perform a post hoc tests for the Kruskal-Wallis test, because this test shows between which groups the significance difference exist.

Hypothesis 2

The second hypothesis predicts that incentive compensation at the client firm is positively associated with professional skepticism for auditors. The second hypothesis is supported when the means of 11 and 21 is higher than the means of 12 and 22, in this case this is the situation as the means for the presence of incentives (6,95) is higher than the mean for the absence of incentives (6,55). However, to accept H2 this difference must be statistically significant (P<.05). Since, H2 test the positively association with professional skepticism an one-way ANOVA is conducted.

To conduct an ANOVA the assumption of homogeneity is met (P = 0.298) see table 8 for the test results of homogeneity of variances. Therefore, we can conduct the way ANOVA. The one-way ANOVA was conducted to compare the effects of incentive compensation on the professional skepticism level of auditors.

a. Kruskal-Wallis Test

b. Grouping variable: Tone at the top

The influence of tone at the top and incentives on the professional skepticism of an auditor