THE EFFECT OF AUDIT FIRM ROTATION ON THE AUDIT QUALITY AND THE MODERATING ROLE OF INDUSTRY SPECIALIZATION

(1)

33

THE EFFECT OF AUDIT FIRM ROTATION ON THE

AUDIT QUALITY AND THE MODERATING ROLE OF

INDUSTRY SPECIALIZATION

(2)

THE EFFECT OF AUDIT FIRM ROTATION ON THE AUDIT QUALITY AND THE MODERATING ROLE OF INDUSTRY SPECIALIZATION

Master Thesis , MSc Accountancy

University of Groningen, Faculty of Economics and Business Supervisor: dr. C.A. Huijgen

June 19, 2019

MEYKE TERPSTRA S2750813

m.w.terpstra@student.rug.nl

(3)

ABSTRACT:

Audit quality is an important concept, but difficult to define. Prior research often examined the quality of the audit and the determinants of audit quality. In this research paper, I study the impact of the short tenured audit firm offices on the quality of the audit. Furthermore, this paper will examine the moderating role of industry specialization on this short tenure of the audit firm office. The impact will be measured with hand collected data from annual reports of German firms. I find that short tenured audit firm offices are negatively associated with the quality of the audit. However, an audit firm office that is qualified as an industry specialist has no moderating effect on the quality of short tenured audit firm offices. The findings are important, because audit firm rotation is mandatory in the European Union since 2014. This regulation was implemented to improve the quality of the audit.

KEYWORDS: AUDIT QUALITY, AUDIT FIRM ROTATION, TENURE, OFFICE INDUSTRY SPECIALIZATION, LISTED FIRMS, GERMANY

(4)

ACKNOWLEDGEMENTS:

At Deloitte Amsterdam, I have written this master thesis to graduate for the Master Accountancy at the University of Groningen. I am grateful for the helpful feedback and advice I received from my supervisor dr. Carel Huijgen. Furthermore, I would like to thank co-students for the discussions and support during the time I was writing this thesis.

(5)

I. INTRODUCTION

In September 2008, Jeffrey Skilling was released from prison on the basis of good behavior after less than 12 years (Financial Times, 2018). The former CEO of Enron was convicted of conspiracy, securities fraud, insider trading and lying to auditors. In 2001 Enron collapsed, while in 2000, Enron exceeded a market capitalization of $60 billion, 70 times earnings and 6 times book value (Healy and Palepu, 2003). The company obscured billions of dollars of debt due to failed projects and deals. The bankruptcy of Enron is one of the most discussed accounting scandals in the world. At the time, one of the world’s biggest audit firms, Arthur Anderson, turned a blind eye during the crimes. Due to the failure of preventing this accounting scandal, the society lost trust in audit firms.

As a reaction to this scandal the Sarbanes Oxley-act1 (hereafter SOX) was introduced in the United States. This legislation had a crucial role in restoring the trust of the public in the nation’s capital markets and strengthening corporate accounting controls (Act, 2002). The SOX aimed to prevent deceptive accounting and management misbehavior by requiring more supervision, dealing with potential conflicts of interest and imposing severe sanctions for managerial misconduct (Zhang, 2007). The implementation of SOX required the establishment of the Public Company Oversight Accounting Board (PCOAB), which oversees the audit of public companies (McClelland and Stanton, 2004). One can say, the first step to improve the quality of the audit was taken here.

Together with the creation of the PCAOB, the SOX implemented several other regulations to improve audit quality. One of the regulations is concerned with mandatory audit rotation. According to Healey and Kim (2003), audit rotation helps to restore confidence in the audit profession. Rotation can be performed in two ways, namely audit firm rotation and audit partner rotation. SOX introduced legislation at partner level to improve the quality of the audit. SOX 203 states: “It shall be unlawful for a registered public accounting firm to provide audit services to

an issuer if the lead (or coordinating) audit partner (having primary responsibility for the audit), or the audit partner responsible for reviewing the audit, has performed audit services for that

1 The Sarbanes Oxley Act is formulated by the American senators Paul Sarbanes (democrat) and later Michael Oxley (republican)

(7)

issuer in each of the 5 previous fiscal years of that issuer”. In other words, the partner of the

audit is allowed to audit the same client for a maximum of 5 consecutive years. The implementation of SOX 203 safeguards the objectivity and independence of an auditor (Sanders, Steward and Bridges, 2009). According to ISA 2002, the independence of an auditor reinforces the ability to work with integrity, to maintain objectivity and to approach the audit with professional skepticism. Independence contributes to a higher quality of the audit.

The other way of rotation is audit firm rotation. Firm rotation prevents auditor firms from becoming too close with the manager of the client (Arel, Brody and Pany, 2005) and avoids large-scale corporate collapses (Jackson, Moldrich and Roebuck, 2008). Researchers conclude that audit firm rotation increases the independence of the auditor as well (Arel et al., 2005 and Jennings, Pany and Reckers, 2006).

Nonetheless, several studies found a downside of the mandatory rotation. The General Accounting Office (2003) concluded that rotation means a loss of client-specific knowledge. Other researchers argue that rotation leads to lower audit quality due to the lack of knowledge the new auditor has on the client in the first years after rotation (Corbella, Florio, Gotti and Mastrolia, 2015). Stated differently, longer tenured auditors increase audit quality. According to Lim and Tan (2010), auditors with longer tenure develop a better understanding of the client and the industry. There are several studies about other determinants of audit quality. Liu, Xie, Chang and Forgione (2017) argue that auditors with industry-specific knowledge improve the audit quality through “their experience serving other clients in the same industry as well as through

learning and sharing best practices across the industry”. Auditors will protect their reputation

and loss of clients in this industry by performing an audit of high quality (Krishnan, 2003). Industry-specific knowledge has a positive effect on the audit quality. In this research, the focus will be on the effect of firm rotation and the moderating role of industry specialization on audit quality.

This research paper contributes to the existing literature regarding mandatory firm rotation and audit quality. Prior research mostly investigated audit partner rotation, because of the

2

(8)

implementation of SOX 202. This paper will focus on audit firm rotation. In 2014, the European Union provided regulation about mandatory audit firm rotation due to common debates about the quality of the audit. The European Union is one of the first that introduced mandatory audit firm rotation contrary to the rest of the world.

While most studies investigate the link between mandatory rotation and audit quality with a viewpoint of fraudulent reporting (Wilson, McNellis, and Latham, 2018) or the single relation between mandatory rotation and audit quality (Daugherty, Dickins, Hatfield and Higgs, 2013), this paper will deepen the role of the industry specialization.

There is limited research concerning the European Union. Most research is conducted in Asian countries (Chen, Lin and Lin, 2004; Chi, Huang, Liao and Xie, 2009) or in Australia (Jackson, Moldrich and Roebuck, 2008). Chen et al. (2004) found a significant increase of audit quality with a longer audit firm tenure. Chi et al. (2009) found that audit quality of companies, after mandatory rotation, is lower under new audit partners and they found that companies did not find significant differences in audit quality with voluntary rotation. Lastly, Jackson et al. (2008) found minimal benefits of mandatory audit firm rotation. The reason that most studies were conducted in these countries, is because of the obligation of mentioning the name of the engagement partner in the audit report. In this research I will focus on German companies. All audit firms will be included, the BIG 4 and smaller firms, in order to obtain more reliable outcomes.

Furthermore, audit quality will be defined from the three main parties of the assurance, the user, preparer and the auditor. The user group will be divided in the main users of the audit report: lenders and investors (Ojala, Niskanen, Collis and Pajunen, 2014). The perspective of regulators towards audit quality will be added.

The research question of this paper is:

“To what degree does mandatory audit firm rotation influence audit quality and what is the effect of industry specific-knowledge on this relationship?”

(9)

II. LITERATURE AND HYPOTHESES DEVELOPMENT

In this chapter the theory underlining the research will be discussed. Furthermore, the hypotheses will be introduced.

II.I. Relevant theory

The agency theory is often used to describe the role of the auditor in general. The theory distinguishes two parties, the principals (stakeholders) and agents (management). Agency problems occur when there is information asymmetry between these two parties (Jensen and Meckling, 1976). Information asymmetry exists when the agent has an information advantage over the principal (An, Davey and Eggleton, 2011).Stakeholders do not have all the information to make a deliberate decision, since the manager has the opportunity to act on his own behalf. This separation of ownership and control turns out in a conflict of interest between stakeholders and the management (Chi et al., 2009). A third party, the auditor, is introduced to mitigate this information asymmetry and to reduce the agency costs. The auditor ensures the protection of the stakeholders' interests (Niskanen, Karjalainen and Niskanen, 2011). This protection is guaranteed by auditing the financial statements prepared by the management. However, the value of the auditing depends on the quality of the audit (Chi et al., 2009).

The quality of the audit is difficult to define. According to the agency theory, the audit is particularly important for the stakeholders. It provides information to the stakeholder to make a well-considered decision. However, even when the auditor abides by the rules, the stakeholders will still have criticism on auditors. They blame the auditors for not being independent towards the client, which makes that the audit is not performed correctly. This is also known as the audit expectation gap. Liggio (1974) introduced the expectation gap by referring to the difference between the expectations as understood by the auditor and the society’s expectations of the auditor. Others, for instance Guy and Sullivan (1988), refer to the audit expectation gap as the difference between the responsibilities of the auditors according the society and the responsibilities auditors themselves believe to take. The audit expectation gap can be divided into two other gaps, namely the “reasonableness” gap and the “performance” gap (Porter, 1993). This is shown in figure 1.

(10)

Following Porter (1993), the reasonableness gap is the gap between the duties what can reasonably be expected from auditors and what the society expects auditors to achieve. This gap is derived from unreasonable expectations of society. The performance gap is the gap between the performance of auditors and the duties what can reasonably be expected from auditors. The performance gap is divided into deficient performance (the gap between the auditor’s existing duties as expected by standards and the perceived performance of auditors by society) and deficient standards (the gap between the duties what can reasonably be expected from auditors and auditors’ existing duties defined by standards).

The definition of audit quality will be explained by the audit expectation gap based on the different stakeholders, who are concerned with the audit of the financial statements.

FIGURE 1

The audit expectation gap (Porter, 1993)

II.II. Audit quality

Notwithstanding the contribution of auditors to the stability of the capital markets, stakeholders continue the debate about audit quality to find a proper definition, composition and measurement (DeFond and Zhang, 2014; Knechel, Krishnan, Pevzner, Shefchik and Velury, 2013). This is due to the difficulty in observing audit quality directly (Francis, 2004) and to the different expectations and perspectives of the stakeholders of the audit report (Watkins, Hillison and Morecroft, 2004). One of the first definitions of audit quality, as argued by DeAngelo’s (1981) is: “the joint probability that an auditor will detect a material misstatement in the financial

(11)

and independence (objectivity) of the auditor, to define audit quality. Over the years, audit quality has become more important and several definitions are identified. Using the audit expectation gap as discussed earlier, the definitions of audit quality described by the different stakeholders are as follows.

Based on Gonthier-Besacier, Hottegindre and Fine-Falcy (2016), the stakeholders will be divided in three parties; the external auditor, the preparer and the user of financial statements. The regulator will be added as a fourth stakeholder.

The first stakeholder is the external auditor. Professional auditors define audit quality with respect to the compliance with professional auditing standards (Christensen, Glover, Omer and Shelley, 2016). According to the PCAOB, the professional auditing standards require auditors to plan and perform audits to obtain reasonable assurance about whether the financial statements give a fair view and if the financial statements are free of material misstatements. Besides the compliance with standards, audit quality is also associated with a combination of the on-the-job experience and formal training (Gonthier-Besacier et al., 2016). Job experience is related to the behavioral characteristics of auditors. Previous research found a positive relation between job satisfaction and the performance of work, which leads to higher audit quality (Shore and Martin, 1989).

The preparers of the financial statements are the second group of stakeholders. The management has as objective to communicate information about the financial position, performance and the cash flows of the company (Surup ceanu, 2011). The audit client expects, in return for the fees paid, a high quality of the audit (Hoopes, Merkley, Pacelli and Schroeder, 2018).

The third stakeholder is the user. The audit gives the users valuable accounting information, which helps in the economic decision-making process. This accounting information helps to reduce the information asymmetry. The user has certain expectations of the work delivered by the auditor. This expectation is not always justified, because of the expectation gap. Auditors are not required to perform all the work that users expect. Overall, the users of the audit must trust that the work of the auditor fulfills a suitable standard and that a high quality audit is performed (IAASB, 2011), which means the absence of material misstatements (Knechel et al, 2013).

(12)

According to Ojala et al. (2014), the main users of financial reporting are investors and lenders. Investors emphasize the individual characteristics of the audit team (Christensen et al., 2016). The audit is useful for the investors to estimate future cash flows. Lenders receive more guarantee that the audit client can fulfill its payment obligations. Even for the negotiation position when contracting with lenders, it is favorable for the audit client. The audit client can ensure the quality of the financial statements and could receive low-interest loans.

Regulators make an effort to stimulate auditors to communicate about the findings and progress of their audit to the client (Gonthier-Besacier et al., 2016). In that sense, regulators are a fourth group of stakeholders. The independent standard-setting body that serves the public interest by providing high quality standards, such as the International Auditing and Assurance Standards Board (IAASB), acknowledged the different perspectives and sets the quality of the information provided to the users above the quantity. It strives to achieve global financial stability.

II.III. Mandatory firm rotation and audit quality

Over the last decades, the discussion about the importance of mandatory audit firm rotation increased. In 2011, the PCAOB investigated the considerations to implement mandatory audit firm rotation. After three years, the outcome was that the advantages did not outweigh the disadvantages, no further action was taken. However, the European Union endorsed regulation about audit firm rotation in the same year. This law became effective in 2016 and it applies to the 28 member nations. Each EU member state has the right to determine the rotation terms independently as long as the maximum term would not be exceeded. The maximum was set on ten years, when exceptions are not met. This also applies to the German audit firms. According to commissioner Michel Barnier, the law needs to reduce the risk of excessive trust between the auditors and the audit clients, stimulate fresh thinking and mitigate joint interests (Wilson et al., 2018). These reasons are most common among proponents which would lead to a positive effect on the audit quality. However, opponents argue that rotation decreases the quality of the audit due to the loss of specific client knowledge, just after rotation takes place.

(13)

II.III.I Proponents of mandatory audit firm rotation

Previous research mention two main reasons in favor of mandatory audit firm rotation. The first is the threat of independence of the auditor. The second is the fresh perspective the rotation brings to the audit.

Independence is an important concept to provide high audit quality. Junaidi, Apriyanto, Nurdiono and Suwardi (2015) define independency as: “the attitude of someone who is

characterized by integrity and objectivity professional duties”. A lack of independence is,

according Jennings et al. (2006), a lack of objectivity in collecting audit evidence to reach an informed opinion about the financial statements of the client. The independency will be impeded as the auditor is too aligned with the management. This relates to the familiarity threat introduced by AICPA Code of Professional Conduct (2015). This code describes it as the threat that a member will become too sympathetic to joint interests or accepting the person’s work too much (AICPA, 2015). This effect is related to the tenure of the audit firm; long tenured auditors can develop a close relationship with the client. An increased tenure indicates an increase in trust between the auditor and client. Trust between the auditor and client means a high level of confidence in each other’s honesty and reliability (Cote and Latham, 2006). As a result, the professional skepticism an auditor should have to stay objective, weakens (Rose, 2007).This can lead to accepting more aggressive accounting choices, which impairs the ability to remain independent (Hoyle, 1978).

The other reason supporting mandatory audit firm rotation is linked to the short tenure of the auditor. Rotation will lead to a fresh perspective on the audit. According to Cameran, Marra, Pettinicchio and Francis (2015), rotation will lead to more professional skepticism, improved objectivity and higher audit quality. A fresh perspective could bring light to issues that had been unnoticed before. The unnoticed issues can be due to the friendly relationship between the long tenured auditor and the client. To solve this problem, rotation gives the opportunity to reset the auditor’s professional skepticism. The auditor is not familiar with the client, which increases objectivity (Jackson et al, 2008). Moreover, the client cannot reengage the auditor which means that the independence of the auditor is less likely to be jeopardized (Copley and Doucet, 1993).

(14)

II.III.II. Opponents of mandatory audit firm rotation

Several researchers (Wilson et al.,2018; Cameran et al., 2013) argue that auditors are associated with an increase in audit risk during the first years of a new engagement. This is due to the loss of specific client knowledge about the business, risks and processes. As a result, there is a higher chance of financial reporting failures. Arel et al. (2005) find an increase in audit failures in the early years of the engagement. With new clients, the auditor has not all the knowledge and information about the business and problems are more likely to occur (Stanley and DeZoort, 2007) Thus, there is a higher litigation risk with short tenured auditors (Stice, 1991). Furthermore, rotation brings high start-up costs and additional work for the new audit firm (Jackson et al., 2008; Daniels and Booker, 2011). Further, Ruiz-Barbadillo, Gomez-Aguilar and Carrera (2009) think that the market for audit services strives to prevent potential loss of future business and the risk of conspiracy between the auditor and client. They argue that mandatory audit firm rotation is unnecessary.

To conclude, mandatory audit firm rotation will increase independency, which is needed to remain objective and work with professional skepticism. Further, rotation offers a fresh look towards the client’s business and risks. However, start-up costs and loss of client-specific knowledge cannot be preserved which impacts the quality in the first years after rotation. Taking the before mentioned views into account, this study draws on the argument that the implications of audit firm rotation outweigh the advantages and therefore the following hypothesis is proposed,

Hypothesis 1 (H1): A short tenure of the audit firm has a negative impact on audit quality.

II.IV. Industry specialization and audit quality

The definition of industry specialist, given by Nagy (2014), is: “the individual personal beliefs,

experience and values and is not easily articulated.” Auditors who are industry specialists are

better informed about industry-specific accounting-related issues (Francis, Reichelt and Wang, 2005). According to the PCAOB (2009), the expertise of the auditor has a positive influence on audit quality. Auditors with industry specialization have more knowledge and background about detecting risks and errors within their industry. Trained and well experienced auditors in specialized industries outperform the auditors without training and experience. Specialized

(15)

auditors perform better risk assessments and, therefore, make better decisions that secures the quality of the audit planning (Low, 2004). This research will not focus on audit firms at a national level, but on office level. This because audit engagements contracts are administered by an office-based audit partner, who is located in the same city as the client (Reynolds and Francis, 2000). Further, decisions regarding the audit of the financial statements are made here (Francis, 2004). The audit offices stand closer to the client and are expected to be better informed about the industry risks and opportunities. As stated in hypotheses 1, audit quality is negatively affected by the rotation in the first years, due to a lack of client specific knowledge of the auditor. In this research, I expect that this negative association is mitigated by the industry specialization of the audit office firm. This leads to the following hypothesis,

Hypothesis 2 (H2): Industry specialization mitigates the negative association between short term tenure and audit quality.

(16)

III. METHODOLOGY

In this section the methodology that will be used to perform this research will be explained. First, the data used will be analyzed. Second, the measurement of the dependent variable will be clarified. Third, the independent variables and control variables will be discussed.

III.I. Data collection and sample selection

For this research paper, the Worldscope database is used for the accounting variables. The data about audit information is hand collected from annual reports of 840 firms in Germany for the period 1999-2009. In total, the database includes 5520 observations. The information from the audit firms includes the name of the engagement partner, the name of the review partner, the audit firm and the audit office.

The data contains 5520 observations, but not all observations are included in the used sample. To measure the impact of the audit firm tenure, firms can only be included as the first switch of an audit firm has taken place or when a company is audited by the same audit firm for the two years before 2001. The tenure is at least 2 years from that point. After eliminating the data before the first rotation, the remaining sample consists of 3554 observations. This is shown in table 1.

TABLE 1 Sample selection

Description Firm years

Initial sample 5520

Eliminated firm years (1966)

Final sample 3554

III.II. Dependent variable

In this research paper the dependent variable is the audit quality. Audit quality is complex to measure. For this variable, the measure according to prior research will be used, namely abnormal accruals. As Chi et al. (2009), I use the modified Jones model. This model is primarily used for detecting earnings management. Audit quality and earnings management are closely related with each other. Higher discretionary accruals suggest aggressive earnings management

(17)

The discretionary accruals are calculated by subtracting the non-discretionary accruals from the total accruals. First, total accruals are determined as follows:

(1)

TACCt = ΔCAt – ΔCash – ΔCLt + ΔDCLt - DEPt

TACCt stands for total accruals in year t, ΔCAt for the change in current assets in year t, ΔCash for the change in cash and cash equivalents in year t, ΔCLt for the change in current liabilities in year t, ΔDCLt for the change in short term debt included in the current liabilities in year t and DEPt the depreciation and amortization expense in year t. Next, the modified Jones model is used in the following equation:

(2)

At-1 stands for the total assets in year t - 1, ΔREVt for the change in revenues in year t, ΔRECt for the change in receivables in year t - 1 and PPEt for the gross property, plant and equipment in year t. The parameters ɑ1, ɑ2 and ɑ3 will be estimated and εt are the residuals in year t. The parameters are estimated by using an ordinary least squares regression. After the parameters are known, the discretionary accruals can be calculated as follows:

(3)

DACCt stands for the total discretionary accruals in year t and NDACCt for the total non-discretionary accruals in year t. These non-non-discretionary accruals can be calculated by using the following equation: (4)

(18)

III.III. Independent variable

In this research, the independent variable is audit firm tenure focusing on the short term tenure. The audit firm needs to rotate every ten years. In this paper I use the first two years as the short term tenure (Carey and Simnett, 2006).

III.IV. Moderating variable

According to Audousset-Coulier, Jeny and Jiang (2016), there are different ways to measure industry expertise. The most common approach is market share. In this research the industry specialism of an audit firm will be measured as the proportion of sales per year and per SIC (Barth) code. The market share will be calculated as the proportion of the log sales of the audit office’s clients in an industry out of the total sales of all clients in the same industry. From each client the log sales will be calculated. This number will be allocated to the different audit firm offices. Subsequently, the sales of the clients per audit firm office will be computed.

III.V. Control variables

There are several control variables influencing discretionary accruals. First, client size has an influence on abnormal accruals. The larger the firm size, the lower the abnormal accruals compared to smaller companies (Behn, Choi and Kang, 2008). The next control variable is auditor type. Large auditors are more independent and professional competent (Huguet and and a, 2016), which negatively impacts the amount of abnormal accruals. Large audit firms face greater economical- and reputational losses than non BIG4 firms (Khuruna and Raman, 2004). In this case, large auditors are BIG4 auditors. Further, firms with a high debt leverage will be more likely to boost or manipulate the financial statements to prevent violating debt covenants (Watts and Zimmerman, 1986). Sales growth of the client also effects discretionary accruals. According to Penman and Zhang (2002) the growth of the firm, calculated by the market-to-book ratio (Dechow, Ge and Schrand, 2010), indicates a positive association with earnings management, so abnormal accruals will be higher. The model will also take the return on assets into account, to control for the nondiscretionary part of the abnormal accruals (Corbella et al., 2015). Lastly, the effect of losses in the previous year of the audit client will be included. Companies suffered from a loss in the previous year have higher incentive to use earnings management to avoid violating debt covenants or prevent negative reactions of shareholders

(19)

(Behn et al, 2008; Corbella et al., 2015). This leads to the following equation, including the moderating variable and the control variables.

DACCt = β0 + β1 (AF TENUREt) + β2 (IND SPt) + β3 (AF TENUREt)(IND SPt) + β4 (SIZEt) + β5

(BIG4t) + β6 (DEBTt) + β7 (GROWTHt) +β8 (ROAt) + β9 (LOSSt) + β10 (YEARFE) + β11

(INDUSTRYFE) + ε

(20)

IV. RESULTS

IV.I. Descriptive statistics

The sample consists of 3554 observations from 652 German listed-firms. Not all variables contain the same number of observations due to missing data for the calculation. The absolute discretionary accruals (DACC) vary from 0 to 0,614 and are on average 0,093. Almost a quarter of the observations have short term tenures up to two years and 1,4% is the average industry specialism measured by market share. Furthermore, the discretionary accruals are ‘winsorized’ because the minimum and maximum exceed three times the standard deviation from the mean of the dependent variable. Winsorizing is also applied for the control variables: size, debt, growth and ROA. The most remarkable control variable is the BIG 4, only 57% of the observations are done by a BIG 4 audit firm. This is relatively low compared to other countries. The descriptives are presented in table 2.

TABLE 2 Descriptive Statistics

Variable N Mean Median Standard deviation Minimum Maximum

DACC 3554 0,093 0,060 0,105 0 0,614

AF TENURE 3554 0,238 0 0,426 0 1

IND SP 3531 0,014 0,028 0,135 -2,257 0,160

AF TENURE * IND SP (moderator) 3550 0,005 0 0,053 -1,568 0,156

SIZE 3554 2,240 2,122 0,925 0,589 5,057 BIG 4 3554 0,568 1 0,495 0 1 DEBT 3554 0,589 0,604 0,236 -0,231 1,419 GROWTH 3551 2,252 1,548 3,917 -28,635 33,471 ROA 3554 -0,000 0,025 0,143 -0,637 0,623 LOSS 3553 0,283 0 0,451 0 1

IV.II. Differences in means analysis

I start with a differences in means analysis. Therefore, I divide the audit offices into two groups depending on: tenure up to two years and longer than two years. The descriptives are shown in table 3.

(21)

As can be seen from table 3, the results of the analysis are consistent with my expectations. The mean discretionary accrual of short tenured audit firms is higher than long tenured audit firms (0,113 > 0,087), which indicates a lower audit quality. This is in accordance with the first hypothesis. The t-test shows a statistically significant difference in means at 1% level.

TABLE 3

Descriptive statistics – tenure and discretionary accruals

Variable N Mean Standard

deviation

Minimum Maximum

Short tenure 846 0,113 0,125 0,000 0,614

Long tenure 2708 0,087 0,097 0,000 0,614

IV.III. Correlation analysis

The pair-wise Pearson correlation coefficients are calculated, to determine to what extent the variables are correlated. When multicollinearity occurs, a linear relation exists between two variables. The variables fully correlate when the coefficient is 1 or -1, this indicates a positive or a negative association. The highest coefficient is 0,368 (between BIG 4 and size) and the lowest, -0,418 (between loss and ROA). This is shown in table 4. The positive coefficient (0,368) means that bigger audit firms have larger clients. The lowest coefficient (-0,418) is also understandable since losses influence the numerator of the ROA. However, these correlations do not indicate serious problems. Furthermore, the table indicates independency between the variables, meaning that the variables do not measure the same effect or are seriously related to each other.

(22)

TABLE 4

Correlation matrix Pearson

DACC AF TENURE IND SP

IND SP x AF TENURE

(moderator)

SIZE BIG4 DEBT GROWTH ROA LOSS

DACC 1,000

AF TENURE 0,106*** 1,000

IND SP -0,008 0,003 1,000

IND SP x AF TENURE (moderator) 0,011 -0,022 0,071*** 1,000

SIZE -0,185*** -0,159*** -0,276*** -0,179*** 1,000 BIG 4 -0,048*** -0,028 -0,145*** -0,093*** 0,368*** 1,000 DEBT 0,066*** 0,019 -0,017 0,005 0,172*** 0,060*** 1,000 GROWTH 0,073*** 0,027 0,089*** 0,092*** -0,060*** -0,051*** -0,001 1,000 ROA -0,116*** -0,110*** -0,114*** -0,073*** 0,218*** 0,076*** -0,254*** 0,042** 1,000 LOSS 0,134*** 0,156*** 0,098*** 0,082*** -0,292*** -0,093*** 0,186*** 0,006 -0,418*** 1,000

(23)

IV.IV. Regression analyses

To test the different hypotheses, this research applies linear regression analyses. The results are shown in table 5. In the first column the expected sign between the independent variables, the control variables and the amount of discretionary accruals is given. In the first model only the effect of the control variables on the dependent variable is shown. The second model investigates the relationship between audit firm rotation and audit quality including the control variables according to hypothesis 1. Model 3 tests the effect of industry specialization on discretionary accruals together with the control variables. The fourth model tests both hypotheses including the control variables together. The F-value indicates the overall significance level of the independent variables. The R-squared explains the percentage of the variance in the dependent variable explained by the independent variables. Lastly, the adjusted R-squared represents the level of variation of the independent variables that affect the dependent variable.

As shown in table 5, the control variables have the predicted sign and are mostly significant. This also applies to the independent variable audit firm tenure. Although industry specialization is not significantly related to the amount of discretionary accruals, the coefficient has the expected sign in the regression analyses.

(24)

TABLE 5

Results linear regression analyses

Expected sign Model 1 Model 2 Model 3 Model 4

Constant 0,068*** 0,062** 0,068*** 0,062**

Independent variables

AF TENURE (H1) + 0,014*** 0,014***

IND SP - -0,007 -0,008

AF TENURE x IND SP (H2) (moderator) - 0,046

Control variables SIZE - -0,018*** -0,017*** -0,018*** -0,017*** BIG 4 - 0,002 0,002 0,003 0,002 DEBT + 0,049*** 0,049*** 0,050*** 0,050*** GROWTH + 0,001*** 0,001*** 0,002*** 0,002*** ROA - -0,018 -0,016 -0,019* -0,017 LOSS + 0,009** 0,007* 0,008* 0,007

Year fixed effects included Yes Yes Yes Yes

Industry fixed effects included Yes Yes Yes Yes

F-value 11,71*** 11,78*** 11,18*** 10,95***

R-squared 0,077 0,080 0,077 0,081

Adj R-squared 0,070 0,073 0,070 0,073

***, ** and * coefficients are statistically significant at 1, 5 and 10 percent

IV.IV.I. Mandatory audit firm rotation

In this research, the discretionary accruals are used as a proxy for audit quality. The results from model 2 indicate that short term audit office tenures have a positive and significant (p < 0,01) effect on the discretionary accruals. A short tenure of the audit office is associated with lower audit quality. This is also shown in model 4. The previous analysis in table 3 strengthens this outcome. The mean of the discretionary accruals of the short tenured audit firm in comparison of the long tenured audit firm is higher (0,113 > 0,087) and significant. To conclude, hypothesis 1 is supported. This is consistent with previous studies that found a negative relation between a short tenure and audit quality (Corbella et al., 2015; Cameran et al., 2013; Arel et al., 2005; Wilson et al., 2018).

(25)

IV.IV.II. Industry specialization

The outcome of the regression analysis concerning the effect of industry specialization is shown in model 3. The independent variable shows the expected sign (-), but is not significantly associated with the amount of discretionary accruals (p > 0,1). This means that there is no evidence that industry specialization increases audit quality.

Further, the results concerning hypothesis 2 are shown in model 4. The moderating role of industry specialization is not significantly (p > 0,1) associated with the discretionary accruals, and has the unexpected sign. This means that audit quality will not improve by the industry specialization of an audit firm during the first years of the audit engagement. Thus, hypothesis 2 is not supported by the regression analysis.

IV.IV.III. Control variables

Most control variables are statistically significant and have the expected sign, as shown in table 5. First, client size is negatively and significantly associated with the amount of discretionary accruals. This result is in accordance with Behn et al. (2008). The BIG 4 audit firms were previously negatively related with the amount of discretionary accruals, based on the correlation matrix in table 4. According to table 5, there is no significant relation between these variables anymore. This can be due to the fragmented market for audit services in Germany. Abedalqader Al-Thuneibat et al. (2011) also concluded that audit firm size has no effect on audit quality. The debt leverage of the client is positively and significant related with the discretionary accruals, which is in accordance with previous research (Watts and Zimmerman, 1986). Hereby, earnings management is used to prevent violation of the debt covenants. Furthermore, the results indicate a significant and positive association between growth, measured by market-to-book-ratio, and the amount of discretionary accruals (Penman and Zhang, 2002). The control variable return on assets is not significantly related with the discretionary accruals. Next, losses incurred in the previous year are positively and significantly related with the amount of discretionary accruals. This is in line with previous research (Behn et al., 2008; Corbella et al., 2015), to prevent violating debt covenants and negative reactions of shareholders. Lastly, there is no serious multicollinearity between the control variables and the independent variables, since the level of significance and the effect of the control variables are relative constant.

(26)

IV.IV.IV. Robustness tests

This research will conduct two additional tests to determine if the conclusions change when the assumptions change. These robustness tests add more reliability. The first robustness test is a different accrual model of the dependent variable audit quality. Instead of the modified Jones model, I performed the linear regression with another common used approach, the discretionary working capital accruals (hereafter DWCA) as defined by deFond and Park (2001). The equation is shown below. (5)

DWCAt stands for discretionary working capital accruals in year t and WCt for the noncash working capital in year t computed as (current assets – cash and short-term investments) – (current liabilities – short-term debt).

This model focuses on the short term accrual effect in comparison with the modified Jones model. First of all, the average DWCA after winsorizing is 0,087 (table 6) and the test will be conducted with the DWCA after winsorizing. This because the minimum and maximum exceed three times the standard deviation from the mean of the dependent variable.

TABLE 6

Descriptive analysis DWCA

Variable N Mean Median Standard

deviation Min Max

DWCA 3554 0,087 0,046 0,147 0 2,742

The results of the linear regression analyses are presented in table 7. The short tenure, the moderating role of industry specialization and the single relation of industry specialization are not significantly related to the amount of DWCA. However, the coefficient of the standalone independent variables, short tenure and industry specialization, have the predicted sign.

(27)

TABLE 7

Results linear regression analyses - DWCA

Model 1 Model 2 Model 3

Constant 0,083** 0,083** 0,084**

AF TENURE (H1) -0,004 -0,004

IND SP -0,007 -0,007

AF TENURE x IND SP (H2) (moderator) 0,007

Control variables included Yes Yes Yes

Year fixed effects included Yes Yes Yes

Industry fixed effects included Yes Yes Yes

F-value 2,98*** 3,01*** 2,82***

R-squared 0,022 0,022 0,022

Adj R-squared 0,014 0,015 0,014

The second robustness test concerns the independent variable, industry specialization, which is defined differently. For this test, I define industry specialization as the client market share of each audit firm. This is calculated by the amount of clients per audit office divided by the overall clients of all audit offices for each year and each SIC (Barth) code. The results of the regression are shown in table 8. The findings support the conclusions made about the standalone role of industry specialization according to table 5. Industry specialization is not significant (p > 0,1) related with the quality of the audit. Further, the moderating effect of industry specialization has the expected negative sign, but is not significant. The significance and effect of the control variables also remain the same. To conclude, industry specialization measured by client market share does not lead to different outcomes.

(28)

TABLE 8

Results linear regression analyses – client market share

Model 1 Model 2 Model 3

Constant 0,062** 0,071*** 0,065**

AF TENURE (H1) 0,014*** 0,015***

IND SP -0,065 -0,074

AF TENURE x IND SP (H2) (moderator) -0,002

Control variables included Yes Yes Yes

Year fixed effects included Yes Yes Yes

Industry fixed effects included Yes Yes Yes

F-value 11,78*** 11,30*** 11,02***

R-squared 0,080 0,077 0,081

Adj R-squared 0,073 0,070 0,073

IV.IV.V. Additional test

To obtain a better understanding of the sign of discretionary accruals, I performed an additional test where the dependent variable is divided into positive and negative discretionary accruals. Positive discretionary accruals indicate a client that aggressively manage its earnings upwards, where negative discretionary accruals indicate conservative earnings management. Table 9 shows the linear regression analyses of the two different discretionary accruals.

In models 1 and 4 the first hypothesis is tested. There is a negative and significant relation between the short tenured audit firm and the amount of positive discretionary accruals (model 1). This result indicates that the audit firm is not capable to restrain aggressive earnings management during its first years’ tenure. Further, there is no effect of tenure on constraining earnings decreasing behavior of clients. The other models show no significant association between the independent variables and positive or negative discretionary accruals.

(29)

TABLE 9

Results linear regression analysis – positive/ negative DACC

Positive DACCR Negative DACCR

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6

Constant 0,070** 0,063** 0,072** -0,79** -0,077* -0,079**

AF TENURE (H1) -0,017* -0,016* 0,007 0,007

IND SP 0,017 0,018 -0,025 -0,025

AF TENURE x IND SP (H2) (moderator) 0,191 -0,063

Control variables included Yes Yes Yes Yes Yes Yes

Year fixed effects included Yes Yes Yes Yes Yes Yes

Industry fixed effects included Yes Yes Yes Yes Yes Yes

F-value 2,45*** 2,23*** 2,26*** 1,67** 1,65** 1,53*

R-squared 0,030 0,028 0,031 0,014 0,014 0,014

(30)

V. CONCLUSION & DISCUSSION

In this chapter the findings of this research will be summarized, followed by the implications of these results. Furthermore, the limitations of this study will be addressed. Finally, recommendations for future research will be suggested.

V.I. Findings

The aim of this study was to examine the impact of audit firm rotation on the audit quality and the moderating role of industry specialization for the German audit market. The results are based on a final sample of 3554 observations, during the period 2000-2009. Linear regression analyses were used to test the hypotheses.

This study began with the requirements introduced by SOX due to the failure of preventing accounting scandals. To improve audit quality, one of the regulation is about mandatory audit rotation. SOX introduced the audit partner rotation in 2002. Since 2014, audit firm rotation is mandated in the European Union. There are several opinions about the effect of audit firm rotation on audit quality. Supporters of rotation claim that firm rotation will increase the independence of the auditor and the fresh perspective at the beginning of the engagement, which will lead to higher audit quality. However, opponents argue that the loss of client-specific knowledge decreases the quality of the audit after firm rotation. This study adds the effect of industry specialization on the audit quality with short tenured audit firms.

The first hypothesis focuses on the causal relation between the short tenure of an audit firm and audit quality. The regression analysis shows a positive and significant relation between the short tenured audit firm and the amount of discretionary accruals, hence hypothesis 1 is supported. This finding is in line with previous studies. The difference in means analysis supports this result. Thus, the amount of discretionary accruals is higher for short tenured audit firms than for the long tenured audit firms. However, with an alternative measure of the amount of discretionary working capital accruals there is no significant relation between the short tenure of the audit firm and audit quality. A reason for this could be that this model assumes that the previous year’s accrual is normal while companies may manage their earnings in consecutive years.

(31)

The second hypothesis focuses on the moderating role of industry specialism on the relation between the short tenure of an audit firm and audit quality. The results show no significant effect of the moderator, thus hypothesis 2 is not supported. The amount of discretionary accruals will not decrease in the first two years of the engagement due to the industry specialism of an audit firm. There is either no significant effect between industry specialism and the amount of discretionary accruals in general. When using the alternative approach, discretionary working capital accruals, there is still no significant effect. For measuring industry specialism, I conducted an alternative approach, client market share. This did not have a different impact on the (moderating) role of industry specialism. An explanation can be that industry specialism has only an effect with a cofounding variable, which gives the incentive to provide higher audit quality. Another reason is the highly fragmented audit market in Germany. The sample consists of 266 different audit firms. Audit firms cannot easily become an industry specialist.

In sum, the research question “To what degree does mandatory audit firm rotation influence

audit quality and what is the effect of industry specific-knowledge on this relationship?” can be

answered as follows. The results show a significant and positive effect of short tenured audit firm offices on the amount of discretionary accruals, which indicates a lower audit quality. Furthermore, there is no significant effect of the (moderating) role of industry specialism on audit quality.

V.II. Implications

This research found evidence that audit firms provide low audit quality in the first two years of the audit engagement. With this finding, the EU regulation that mandates audit firms to rotate every ten years to increase audit quality, is not sensible, although I did not examine long-term tenures specifically. I also found evidence that audit firms who are industry specialists have no mitigating effect on the quality of the audit. Regulators should consider cofounding incentives that strengthens audit quality in the first years of the engagement. These findings could also be relevant for investors to obtain better estimates of the risks of reporting. Another important issue for investors is the incapability of the audit firm to restrain aggressive earnings management in the first years of the engagement.

(32)

V.III. Limitations

This study has several limitations. First, the data contains only voluntary audit firm rotation. This is because the data is from the years 1999-2009, while audit firms are mandated to rotate in the European Union since 2014. The findings of this research cannot be directly assigned to audit firms which are mandated to rotate after ten years. More recent data will increase the reliability of the results. Further, as mentioned before, audit quality has different proxies. The modified Jones model is one of the most common measurement, although another proxy could give different outcomes. This is shown with the robustness test, where discretionary working capital accruals is used as proxy.

V.IV. Recommendations future research

In this research together with previous research the impact of the short tenured audit firm is examined. Most studies, including this research, conclude that short tenured audit firms provide low audit quality. Short tenured is mostly defined as the first two years of the engagement. For future research I recommend to examine the tip-over point of the short tenure, thus the maximum first years that can be allocated to ‘short tenure’. However, long tenured audit firms can develop a too close relationship with the client, which will be at the expense of the quality of the audit as well. Little research has been conducted concerning the effect of long tenured audit firms on audit quality. The European Union has set the maximum tenure on ten years, but the maximum tenure that is most beneficial for audit quality has never been studied before.

Furthermore, the short tenured audit firms are a sub-population of the overall tenures. This outcome does not tell us the insignificance of industry specialization for long tenured audit firms. It is possible that in the first years of the engagement industry specialization can make a difference in the audit quality, but in the following years industry specialization can be insignificant when the specialization is combined with the engagement experience of a longer tenured audit firm.

(33)

(34)

VI. REFERENCES

Abedalqader Al-Thuneibat, A., Tawfiq Ibrahim Al Issa, R., & Ata Baker, R. (2011). Do audit tenure and firm size contribute to audit quality? Managerial Auditing Journal, 26(4), 317-334. Act, S. O. (2002). Sarbanes-Oxley Act. Washington DC.

An, Y., Davey, H. & Eggleton, I.R. (2011). Towards a comprehensive theoretical framework for voluntary IC disclosure. Journal of Intellectual Capital, 12(4), 571-585.

Arel, B., Brody, R. G., & Pany, K. (2005). Audit firm rotation and audit quality. The CPA

Journal, 75(1), 36.

Armstrong, R. (2018). As Skilling leaves prison, Enron’s old businesses are thriving. The

Financial Times.

Audousset-Coulier, S., Jeny, A., & Jiang, L. (2016). The validity of auditor industry specialization measures. Auditing: A Journal of Practice and Theory, 35, 139–161.

Behn, B., Choi, J., & Kang, T. (2008). Audit quality and properties of analyst earnings forecasts.

Accounting Review, 83(2), 327-350.

Cameran, M., Marra, A., Pettinicchio, A., & Francis, J. (2015). Are there adverse consequences of mandatory auditor rotation? evidence from the Italian experience. Auditing, 34(1), 1-24. Carey, P., & Simnett, R. (2006). Audit partner tenure and audit quality. The Accounting Review, 81(3), 653-676.

Chen, C., Lin, C., & Lin, Y. (2008). Audit partner tenure, audit firm tenure, and discretionary accruals: Does long auditor tenure impair earnings quality?*. Contemporary Accounting

Research, 25(2), 415-445.

Chi, W., Huang, H., Liao, Y., & Xie, H. (2009). Mandatory audit partner rotation, audit quality, and market perception: Evidence from Taiwan. Contemporary Accounting Research, 26(2), 359-391.

(35)

Christensen, B., Glover, S., Omer, T., & Shelley, M. (2016). Understanding audit quality: Insights from audit professionals and investors. Contemporary Accounting Research, 33(4), 1648-1684.

Copley, P. A., & Doucet, M. S. (1993). Auditor tenure, fixed fee contracts, and the supply of substandard single audits. Public Budgeting & Finance, 13(3), 23–35.

Corbella, S., Florio, C., Gotti, G., & Mastrolia, S. A. (2015). Audit firm rotation, audit fees and audit quality: The experience of Italian public companies. Journal of International Accounting,

Auditing and Taxation, 25, 46-66.

Cote, J., & Latham, C. K. (2006). Trust and commitment: Intangible drivers of interorganizational performance. Advances in Management Accounting, 15, 293-325.

Daniels, B., & Booker, Q. (2011). The effects of audit firm rotation on perceived auditor independence and audit quality. Research in Accounting Regulation, 23(1), 78-82.

Daugherty, B., Dickins, D., Hatfield, R., & Higgs, J. (2013). Mandatory audit partner rotation: Perceptions of audit quality consequences. Current Issues in Auditing, 7(1), 35.

Dechow, P., Ge, W., & Schrand, C. (2010). Understanding earnings quality: A review of the proxies, their determinants and their consequences. Journal of Accounting and Economics, 50(2-3), 344-401.

DeFond, M. L., & Park, C. W. (2001). The reversal of abnormal accruals and the market valuation of earnings surprises. The Accounting Review, 76(3), 375-404.

DeFond, M. L., & Zhang, J. (2014). A review of archival auditing research. Journal of

Accounting and Economics, 58(2–3), 275-326.

Dunn, K. A., & Mayhew, B. W. (2004). Audit firm industry specialization and client disclosure quality. Review of Accounting Studies, 9, 35–58.

Francis, J. (2004). What do we know about audit quality? British Accounting Review, 36(3), 345-68.

(36)

Francis, J., Reichelt, K. & Wang, D. (2005). The pricing of national and city-specific reputations for industry expertise in the US audit market. The Accounting Review, 80(1), 113-136.

Gonthier-Besacier, N., Hottegindre, G., & Fine-Falcy, S. (2016). Audit quality perception: Beyond the ‘role-perception gap’. International Journal of Auditing, 20(2), 186-201.

Government Accounting Office (GAO). (2003). Report to the senate committee on banking, housing, and urban affairs and the house committee on financial services. In Public accounting firms: Required study on the potential effects of mandatory audit firm rotation. GAO.

Gul, F. A., Wu, D., & Yang, Z. (2013). Do individual auditors affect audit quality? Evidence from archival data. The Accounting Review, 88(6), 1993-2023.

Healy, P. M., & Palepu, K. G. (2003). The fall of Enron. Journal of economic perspectives, 17(2), 3-26.

Hoopes, J., Merkley, K., Pacelli, J., & Schroeder, J. (2018). Audit personnel salaries and audit quality. Review of Accounting Studies, 23(3), 1096-1136.

Hoyle, J. (1978). Mandatory auditor rotation: The arguments and an alternative. Journal of

Accountancy, 145, 69–78.

Huguet, ., and a, J. (2016). Audit and earnings management in Spanish smes. Brq Business

Research Quarterly, 19(3), 171-187.

International Auditing and Assurance Standard Board (IAASB). (2011). Framework for Audit Quality: Key Elements that Create an Environment for Audit Quality. IFAC.

Jackson, A. B., Moldrich, M., & Roebuck, P. (2008). Mandatory audit firm rotation and audit quality. Managerial Auditing Journal, 23(5), 420-437.

Jennings, M. M., Pany, K. J., & Reckers, P. M. (2006). Strong corporate governance and audit firm rotation: Effects on judges' independence perceptions and litigation judgments. Accounting

Horizons, 20(3), 253-270.

Jensen, M. & Meckling, W. (1976). Theory of the firm: Managerial behavior, agency costs, and ownership structure. Journal of Financial Economics, 3(4), 305-360.

(37)

Junaidi, J., Apriyanto, H., Nurdiono, N., & Suwardi, E. (2015). The effect of audit firm tenure in artificial rotation on audit quality. Journal of Economics, Business & Accountancy Ventura, 17(3), 439-439.

Khurana, I. K., Raman, K. K. (2004). Litigation risk and the financial reporting credibility of Big 4 versus non-Big 4 audits: evidence from Anglo-American countries. Account. Rev., 79, 473-495.

Knechel, W. (2016). Audit quality and regulation. International Journal of Auditing, 20(3), 215-223.

Knechel, W. R., Krishnan, G. V., Pevzner, M. B., Shefchik, L., & Velury, U. (2013). Audit quality: Insights from the academic literature. Auditing: A Journal of Practice and Theory, 32(1), 385–421.

Krishnan, G. V. (2003). Does Big 6 auditor industry expertise constrain earnings management?

Accounting horizons, 17, 1-16.

L. Nagy, A. (2014). Audit partner specialization and audit fees. Managerial Auditing Journal, 29(6), 513-526.

Liggio, C.D. (1974), The expectation gap: the accountants Waterloo. Journal of Contemporary

Business, 3(3), 27-44.

Lim, C. Y., & Tan, H. T. (2010). Does auditor tenure improve audit quality? Moderating effects of industry specialization and fee dependence. Contemporary Accounting Research, 27(3), 923-957.

Liu, L., Xie, X., Chang, Y., & Forgione, D. (2017). New clients, audit quality, and audit partner industry expertise: Evidence from taiwan. International Journal of Auditing, 21(3), 288-303. Low, K. (2004). The effects of industry specialization on audit risk assessments and audit planning decisions. The Accounting Review, 79(1), 201–219.

(38)

Niskanen, M., Karjalainen, J. and Niskanen, J. (2011), “ emand for audit quality in small private firms: evidence on ownership effects”, International Journal of Auditing, 15(4), 43-65.

Ojala, H., Niskanen, M., Collis, J., & Pajunen, K. (2014). Audit quality and decision-making in small companies. Managerial Auditing Journal, 29(9), 800-817.

Penman, S., Zhang, X. (2002). Accounting conservatism, the quality of earnings, and stock returns. The Accounting Review, 77, 237–264.

Porter, B. (1993). An empirical study of the audit expectation-performance gap. Accounting &

Business Research, 23(93), 49-68.

Public Company Accounting Oversight Board (PCAOB). (2010). Auditing Standard No 8, Audit Risk. PCAOB, 4.

Reynolds, J., & Francis, J. (2000). Does size matter? the influence of large clients on office-level auditor reporting decisions. Journal of Accounting and Economics, 30(3), 375-400.

Rose, J. M. (2007). Attention to evidence of aggressive financial reporting and intentional misstatement judgments: Effects of experience and trust. Behavioral Research in Accounting, 19(1), 215–229.

Ruiz-Barbadillo, E., Gomez-Aguilar, N., & Carrera, N. (2009). Does mandatory audit firm rotation enhance auditor independence? evidence from Spain. Auditing, 28(1), 113-135.

Sanders, C. B., Steward, M. D., & Bridges, S. (2009). Facilitating knowledge transfer during SOX-mandated audit partner rotation. Business Horizons, 52(6), 573-582.

Shore, L., & Martin, H. (1989). Job satisfaction and organizational commitment in relation to work performance and turnover intentions. Human Relations, 42(7), 625-638.

Shu, P., & Chiang, S. (2014). Firm size, timing, and earnings management of seasoned equity offerings. International Review of Economics and Finance, 29, 177-194.

Stanley, J. D., & F. T. DeZoort. (2007). Audit firm tenure and financial restatements: An analysis of industry specialization and fee effects. Journal of Accounting and Public Policy, 26, 131–159.

(39)

Stice, J. D. (1991). Using financial and market information to identify pre-engagement factors associated with lawsuits against auditors. The Accounting Review, 66(3), 516–533.

Surup ceanu, E. (2011). The objective of financial statements.company’s financial position.

nalele niversit ii onstantin r ncu i in T rgu iu : Seria Economie, 1(4), 181-186.

Watkins, A.L., Hillison, W. & Morecroft, S.E. (2004). Audit quality: a synthesis of theory and empirical evidence. Journal of Accounting Literature, 23, 153-93.

Watts, R., Zimmerman, J. (1986). In: Positive Accounting Theory. Prentice-Hall Inc.

Wilson, A., McNellis, C., & Latham, C. (2018). Audit firm tenure, auditor familiarity, and trust: Effect on auditee whistleblowing reporting intentions. International Journal of Auditing, 22(2), 113-130.

Zhang, I. X. (2007). Economic consequences of the Sarbanes–Oxley Act of 2002. Journal of

(40)

VII. APPENDICES

1. The definitions and measurements of the variables.

Independent variable DACC - Audit quality Absolute number of

discretionary accruals

Dependent variables AF ROT - Audit firm rotation 1 if the audit firm is in the

first two years of the engagement, otherwise 0

Moderating variable IND SP - Industry specialization Continue variable of market

share.

Control variable SIZE - Client size Absolute number calculated

by the LOG of the total assets

BIG4 - Auditor’s type 1 if the client is audited by the BIG4, otherwise 0 DEBT - Debt leverage of the client Absolute number calculated

by total liabilities divided by total assets

GROWTH - Growth of the client Market-to-book ratio, Market value of equity divided by book value of equity

ROA - Return on assets Net income divided by the total assets

LOSS - Losses of the client 1 if the client incurred a loss in the previous year, otherwise 0

YEAR FE Year fixed effect

THE EFFECT OF AUDIT FIRM ROTATION ON THE AUDIT QUALITY AND THE MODERATING ROLE OF INDUSTRY SPECIALIZATION