MSc Accounting Thesis

(1)

MSc Accounting Thesis

The effect of partner rotation on audit quality: difference

between audit firm size

University of Groningen Faculty of Economics and Business

Department of Accounting August 2019 Mart Japin Thomas a Kempisplantsoen 14 3532 AH Utrecht +31623822515 m.japin@student.rug.nl student number: 2544083 word count: 9624

(2)

Abstract

Due to large accounting scandals, new guidelines and policies were introduced to prevent that such scandals will occur in the future. In the U.S., the SOX was introduced and one important topic was the mandatory partner rotation. This study examines the relation between audit partner tenure and audit quality. Moreover, this study examines the relation between audit firm size and audit quality and attempts to compare the difference in perceived audit quality after audit partner rotation between big4 and non-big4 audit firms. Discretionary accruals is the proxy for audit quality. Prior literature suggest that longer audit partner tenure is associated with higher audit quality, because longer audit partner tenure leads to more client specific knowledge. Moreover, prior literature suggest that larger audit firms produce higher audit quality, because they have better trained employees and more resources. Therefore, I predict that audit partner rotation leads to lower audit quality and that this rotation effects are smaller for larger audit firms. Particularly, I found evidence that longer audit partner tenure leads to higher audit quality. Moreover, I found no evidence that big4 audit firms produce higher audit quality than non-big4 audit firms. Concludingly, we found no evidence that the effects of partner rotation are smaller for big 4 audit firms, in terms of audit quality. This study is based on 2903 firm-year observations from German companies in the time period 1999-2010.

(3)

Content

1. Introduction 4

1.1. background 4

1.2. Academic contribution 5

1.3. Research question 5

2. Recent policy initiatives, literature review and hypotheses 6

2.1. Recent policy initiatives 6

2.2. Theoretical Background 6

2.2.1. Audit partner rotation, audit partner tenure and audit quality 6

2.2.2. Audit firm size and audit quality 7

2.2.3. The difference of the effect of audit partner rotation and tenure between big four and

non-big four audit firms 8

3. Research Methodology 9

3.1. Sample 9

3.2. Variables 9

3.2.1. 3.2.1 Dependent variable: audit quality 9

3.3. Independent variables 11

3.3.1. Partner tenure 11

3.3.2. Audit firm size 11

3.3.3. control variables 11

3.4. data analyses 12

4. Results 13

4.1. Descriptive results 13

4.2. Difference in mean tests 14

4.3. Correlations 15

4.4. Main results 16

4.5. Additional tests 19

4.5.1. Defond and Park model (2001) 19

4.5.2. Big3 audit firms 22

5. Discussion and conclusion 24

5.1. Findings 24

5.2. Theoretical and practical implications 25

5.3. Limitations and suggestions for further research 25

(4)

1. Introduction

1.1. background

At the beginning of the 21st century, a number of large accounting scandals like Enron and WorldCom

dominated the news. This resulted in debates about audit quality and the position of audit firms (Cullinan, 2004). These audit failures have prompted regulators to question about new regulations. One issue concerns about the tenure of audit partners, a long tenure can lead to the fact that the auditor accepts questionable accounting and reporting of his client, because of familiarity threats and/or in order to retain the client (Code of Ethics for professional accountants). The independence of auditors has been a discussion point for several years. Chi, Huang, Liao, and Xie (2009) examined the effect of mandatory audit partner rotation, using abnormal accruals to measure audit quality. Mandatory audit partner rotation is expensive for audit firms. The SEC adopted the Sarbanes-Oxley-Act in 2003 and imposed mandatory audit partner rotation. The SEC imposed that a partner should rotate every five years, instead of every seven years. Moreover, the cooling off period will be five years instead of two years. Mandatory rotation for the review partner is also required. Mandatory partner rotation is required in several counties nowadays. For example, France, Germany, The Netherlands, the United Kingdom, and the United States audit partners are dealing with mandatory rotation.

Although, mandatory partner rotation is required in many countries, little research has been done regarding the effects of the rotation on audit quality. The reason is that most countries do not require partners’ names to be disclosed. The major benefit of partner rotation is that the new partner will be more independent (Litt et al., 2014). The new partner will enhance the independence due to his ‘fresh look’ and therefore will increase audit quality. However, opponents argue that mandatory partner rotation will cause a loss in audit quality for several reasons. The first reason is that mandatory rotation will increase the costs for audit firms and these costs will eventually be paid by investors. Johnson et al., (2012) argue that a new partner may lack the client specific knowledge of risks, procedures, operations and financial reporting practices, which could lead to lower audit quality. Moreover, the audit profession argues that this loss of quality would be more pronounced during the initial years of the new client engagement, when the lack of client specific knowledge and information asymmetry is likely to be severe.

Recent studies on partner rotation and audit quality have argued that there is need for further research, in several countries including the U.S. it is not required to disclose audit partner names. In Germany it is required to disclose the names of both partners, the engagement and the review partner. In Germany, both partners ‘signatures must be visible and identifiable in the audit report and both partners are formally responsible for the audit (§ 322 HGB of the German Commercial Code). As such, German audit report data offers the unique opportunity to examine tenure and rotation of the engagement partner by means of the identification of their respective signatures in the audit report. The purpose of the review partner is to provide quality control for audit engagements and to serve as an evaluation of the performance of the audit engagement partner and team (Epps and Messier 2007). Studies in Taiwan and Australia for example have examined partner rotation by observing the

signature changes on audit reports. Chi et al.,(2009), Carey and Simnett (2006) and Manry et

al.,(2008) argue that audit quality is higher for firms that were under mandatory rotation. On the other hand, Daugherty et al.(2012) and Bedard and Johnstone(2010) argue that partner rotation may

improve the independence of the auditor, but that mandatory rotation can cause a decrease in audit quality, because of the loss of experience.

(5)

1.2. Academic contribution

Several researchers have conducted studies about factors that might influence audit quality. This study contributes to the existing literature about mandatory auditor rotation in several ways. First it gives more insights due to more recent data on mandatory auditor rotation. It complements the study of Gold et al. (2012) by adding firm size as a variable to compare the effects of audit partner rotation on audit quality of big four versus non-big four audit firms. Second, this study is an extension of the research conducted by Chi and Huang (2005), and Carey and Simmnett (2006), because our study provides new empirical evidence by using financial reports on partner rotation and tenure data in a highly developed audit environment, i.e., Germany. Moreover, as Bamber and Bamber(2009) recommend, we isolate the effect of partner tenure from firm tenure. This helps us to investigate the expertise and independence issue. To conclude, I believe that this study is helpful for audit professions and legislators in countries where research on partner rotation is scarce, because of the lack of data available to study partner rotation, for example in the United States.

1.3. Research question

My expectation is that engagement partner rotation influences audit quality and that there is a difference in the perceived audit quality between big-four audit firms and non-big-four audit firms. The main research question is:

What is the effect of engagement partner rotation on audit quality, and is there a difference in perceived audit quality between big-four vs. non-big-four audit firms?

The remainder of this research is organized as follows. Section 2 describes the recent regulatory developments regarding partner rotation, and prior literature regarding partner rotation and ends with hypotheses development. Section 3 provides the research design, section 4 discusses the result and section 5 concludes.

(6)

2. Recent policy initiatives, literature review and hypotheses

2.1. Recent policy initiatives

Nowadays, in countries as Germany, the Netherlands and the U.S. for example, it is required to rotate the audit partner, but the maximum amount that an audit partner can serve a client differs. The U.S. was the first country that required audit partner rotation in the 1970s. The American Institute of Certified Public Accountants (AICPA) required that audit partners in charge of Securities and Exchange Commission (SEC) audits have to be rotated at least every seven years. In 2002, the

Sarbanes-Oxley Act (SOX) was implemented and reduced the rotation period from seven to five years which is applicable for the audits of public companies. Besides the reduced rotation period, the SOX also implies that both the engagement partner and the review partner have to be rotated. In 2006, the

European Commission reviewed the 8th directive requirements and in June 2008 all member states of

the European Union were required to implement this directive in their national law. One important aspect of this directive concerns the rotation of the key audit partner, but there was no clear maximum length of the tenure. Besides that, Germany had developed its own Commercial Code in 1998, which resulted in the implementation of mandatory rotation in 2002. As a result, an audit partner had to rotate if he/she had signed the audit report six times in a ten year relationship with the client. In 2004, Germany made an extension on the rotation of the partner and from that point the audit partner was mandated to rotate every seven years. This law is considered as a direct response to the SOX in 2002 and the recommendation of the European Commission.

2.2. Theoretical Background

2.2.1. Audit partner rotation, audit partner tenure and audit quality

The objective of an audit is to express an opinion about the reliability of the financial statements (ISA600). The auditor should plan the audit so that he can provide reasonable assurance that there are no material misstatements in the audit. D’Angelo (1981) defines audit quality as the probability that an auditor will discover a material misstatement and report that misstatement. According to this definition of audit quality, we can separate audit quality in two dimensions. The first is the ability of the auditor to detect material misstatements (expertise) and the second is the auditor’s ability to report those detected misstatements (independence). Prior literature explains the effects of auditor tenure (rotation) on audit quality. Somehow, there are two conflicting issues about the effect of auditor rotation on audit quality; the issue of expertise and the issue of independence (Iyer and Rama 2004; Shockley 1981). The issue of independence argues that longer auditor tenure is associated with lower audit quality. The first explanation for the associated lower audit quality is the threat of familiarity which can have an effect on the independence of the auditor (IFAC 2008). The IFAC states that a familiarity threat occurs when a member of the assurance team becomes too ‘close’ with the interests of the client. This could result in the fact that an audit partner is less objective, so his ability to judge the company’s performance and reporting decisions will decrease.

The second explanation for the lower audit quality is the threat of routine. It is likely that an audit partner will apply creative approaches of audit testing in the first years of his tenure (Hoyle, 1978). More knowledge of the client’s system and control procedures may result in more routine audit programs. Moreover, Shockley (1981) argues that a developed confidence in the client may arise after a long tenure. Wright (1988) argues that auditors often look back at the prior years when planning the audit for the current year. The main concern is that auditors will anticipate and compare the current

(7)

results with the years before instead of carefully and objectively reviewing the financial data for material misstatements. In conclusion, the independence theory suggest that audit partner rotation will lead to higher audit quality.

On the other hand, the expertise issue (Johnson et al., 2002) argues that audit quality will decrease due to partner rotation. This loss in audit quality is explained by the fact that there is a loss of knowledge about the client and its industry as the serving partner rotates. This explanation relates to information asymmetry. The longer the partner serves the client, the less there is information asymmetry, because the auditor acquires more knowledge of the client and the client’s industry over time. Having more client-specific and industry-specific knowledge improves the detection of material misstatements. When an audit partner rotates, the incoming partner would have less knowledge which implies that the quality of the audit will be lower (Litt et al., 2104; Myers et al., 2003). So, according to the expertise issue, auditor rotation would have a negative effect on audit quality.

Prior research has investigated the relationship between audit partner rotation and audit quality. Stanley and DeZoort (2007) investigated whether auditor rotation leads to higher audit quality. Moreover, Johnson et al., (2002) investigated whether auditor rotation leads to lower audit quality. Recent studies examined audit partner rotation in Taiwan, Australia and Germany. Carey and Simnett (2006) studied Australian companies .Their results showed that longer partner tenure is associated with higher audit quality. Chen et al. (2009) showed that the longer an audit partner servers a client the higher the audit quality. So, their conclusion is that audit partner rotation is associated with lower audit quality. Other researchers concluded that audit partner rotation is associated with lower audit quality (Litt et al., 2014; Gold et al., 2012). Regulators, on the other hand, argue that audit partner rotation is associated with higher audit quality due to the ‘fresh look’ that an incoming partner has on the client. Based on prior literature I developed the following hypothesis regarding the effect of audit partner rotation on audit quality.

Hypothesis 1: Audit partner rotation (short tenure) is associated with lower audit quality.

2.2.2. Audit firm size and audit quality

Prior literature explains the effect of audit firm size on audit quality. Prior research suggest that there are two different views regarding audit firms and audit quality. The first is that larger audit firms produce higher audit quality than smaller audit firms (Sundgren & Svanström,2013). Researchers that support this opinion, will be called opponents in this research. In the contrary, proponents assume that there is no difference in the perceived audit quality between larger and smaller audit firms (Arnett & Daros, 1979). In their research, Choi et al.(2010), found evidence that larger audit firms produce higher audit quality and that this is incorporated in the higher price.

In the contrary, Arnett & Daros, (1979) argue that it is unfair to make a distinction between larger and smaller audit firms, as long as professional standards and qualifications are maintained. On the other hand, D’Angelo (1981) argues that there is a difference between larger and smaller audit firms and that larger audit firms produce higher audit quality. This is because larger audit firms are less dependent on their clients as smaller firms, which can be expressed in quasi-rents. In Sweden,

Sundgren & Svanström(2013) also argue that larger audit firms produce higher audit quality and their results show that there is a difference in audit quality between top six audit firms and non-top six audit firms. Germany provides us an unique opportunity to investigate whether there is a an effect of partner rotation on audit quality and to compare the effects for big four and non- big four audit firms. In Germany it is unique, because both the engagement partner as the review partner need to sign the report. Moreover, in Germany there is less dominance of the big four in comparison with other

(8)

countries. In the Netherlands for example, 95 percent of the public companies are audited by big four audit firms. In Germany it is around 60 percent. Based on the prior literature about audit firm size and audit quality we have developed the following hypothesis: According to Francis (2006), there are two groups of audit firms, the big audit firms and the smaller audit firms. The big audit firms have more resources to train their employees, available facilities and technological capabilities than smaller audit firms. Arthur Anderson has left the market since the case of Enron, which was a huge accounting scandal, and led to an enormous reputational loss for this firm. Since that moment, the market consists of big four audit firms and smaller audit firms. The big four are the current four largest internationally accounting firms, consisting of PwC, Deloitte, KPMG and Ernst & Young. Their purpose is to deliver qualitative audit services. Smaller audit firms are all other audit firms. Based on the prior literature about audit partner rotation and firm size I have developed the following hypotheses.

Hypothesis 2: Big4 audit firms produce higher audit quality than non-big4 audit firms.

2.2.3. The difference of the effect of audit partner rotation and tenure between big four and non-big four audit firms

According to regulators the purpose of mandatory audit partner rotation is to increase the perceived audit quality. As prior empirical literature suggest, mandatory audit partner rotation is associated with lower audit quality for several reasons. One of the reasons is the fact that the new incoming partner has less experience and knowledge about the client’s systems and operations and the industry in which it operates (Gold et al., 2012). It is generally known that bigger audit firms have more resources to train their employees, which in turn could give them competitive advantages. The relationship between firm size and audit quality have been an issue for several years. Larger audit firms do not specifically produce better audit quality than smaller audit firms.

Francis et al. (1999) concluded that a lower level of discretionary accruals can be observed when firms are audited by larger audit firms and that bigger audit firms have more to lose when an audit failure occurs. Brand reputation is an example of what they have to lose (Klein and Leffler, 1981) and larger audit firms may therefore be less tolerant regarding the level of discretionary accruals adopted by firms than smaller audit firms. The chance that an auditor discovers a breach depends on the

auditor competence.Chaney et al. (2004) argued that auditor competence is determined by the

technological capabilities, training and the available facilities. Holm and Zaman (2012) conclude that the lower audit quality of the smaller audit firms is due to their lack of expertise. Germany provides us an unique opportunity to investigate whether there is a an effect of partner rotation on audit quality and to compare the effects for big four and non- big four audit firms. In Germany it is unique, because both the engagement partner as the review partner need to sign the report. Moreover, in Germany there is less dominance of the big four in comparison with other countries. In the Netherlands for example, 95 percent of the public companies are audited by big four audit firms. In Germany it is around 60 percent. Based on the prior literature about audit partner rotation and audit firms size related to audit quality we formulated the following hypothesis:

Hypothesis 3: Smaller audit firms have a higher loss of quality due to audit partner rotation than larger audit firms.

(9)

3. Research Methodology

3.1. Sample

The data used for this study were obtained from two different sources. The data for the accruals and other financial related issues were obtained from WorldScope and data related to partner rotation were obtained by hand collection. Our initial sample consisted of 5930 firm-year observations of German listed firms, covering the period 1998-2010. These firms are relevant to this research, because in Germany, mandatory partner rotation was required since 1998. I eliminated 3023 observations with missing accrual or partner rotation items. The final sample consists of 2907 firm years.

3.2. Variables

3.2.1. 3.2.1 Dependent variable: audit quality

As mentioned before audit quality is defined as the probability than an audit will discover a breach and report that breach. There are several proxies for audit quality in archival research. Vanstraelen (2000) argues that the propensity to issue unqualified audit opinions is a proxy for audit quality. Other researchers also argue that materiality judgements are a proxy for audit quality (Kemp et al., 1983). Earnings management is a recent topic that has been popular among researchers and which is

associated with audit quality. Recent studies have examined the firm’s accrual accounting behavior as a proxy for audit quality (Carey and Simmnet 2006; Myers et al, 2003). Higher accrual levels are associated with a higher level of earnings management and lower auditor conservatism and indicates a lower level of audit quality. Logically, lower accrual levels are associated with a lower level of

earnings management and higher auditor conservatism and indicates a higher level of audit quality. Chi et al., (2009) used the modified jones model to estimate earnings management or abnormal accruals. The accruals of a firm can be divided into discretionary and non-discretionary accruals. In line with Dechow et al.,(1995) I use the modified jones model to estimate the accruals. I use the following formula for the accruals:

TAACt = ΔCAt – Δcasht – ΔCLt + ΔDCLt – DEPt (1) In this formula,

TAACt = total accruals in year t

ΔCAt = change in current assets in year t

ΔCasht = change in cash and cash equivalents in year t

ΔCLt = the change in current liabilities in year t

ΔDCLt = the change in short term debt include in current liabilities in year t DEPt = the deprecation and amortization costs in year t

After the computation of the total accruals in year t, a regression analysis is used to estimate the alpha’s for every industry based in the SIC codes. The different industries are classified as in Barth, Beaver and Landsman(1998).

(10)

The step to estimate the modified jones model is as follows:

TAACt/TAt-1 = α1 * 1/TAt-1 + α2 *(ΔREVt – ΔRECt) / TAt-1 + α3 * PPEt/TAt-1 + εt (2)

Where:

TAACt/TAt-1 = the total accruals in year t divided by total assets in year t-1

ΔREVt = the change in net sales, calculated by revenues in year t less revenues year t-1 ΔRECt = delta receivables, receivables year t minus receivables year t-1

PPEt = gross, property, plant and equipment in year t

TAt-1 = total assets in year t-1

α1, α2 and α3 = the alpha parameters that have to be estimated

εt = the residuals in year t

I choose to make use of the modified jones model instead of the jones accruals model, because the modified version of this model contains the revenues received and therefore there is no space for manipulation.

After the total accruals and the specific parameters are calculated, the calculation of the non-discretionary accruals is as follows:

NDAACt/TAt-1 = α1 * 1/TAt-1 +α2 * (ΔREVt – ΔRECt)/TAt-1 + α3 *PPE/TAt-1 (3)

Where:

NDAACt/TAt-1 = non-discretionary accruals divided by assets in year t-1

The other components fit with the description as stated above in the TAAC formula.

The discretionary accruals will be calculated by the following formula: DAACt = TAACt – NDACCt (4)

Where:

DAACt = the discretionary accruals TAACt = total accruals

(11)

All variables are scaled by total assets to control for size effects and to reduce the possibility of heteroscedasticity.

3.3. Independent variables 3.3.1. Partner tenure

The first independent variable is audit partner tenure, especially engagement partner tenure. Audit partner tenure is defined as the number of years that an audit partner is responsible for the audit services for a specific client (Johnson et al., 2002; Myers et al., 2003). The tenures of the partners are divided into two groups, short tenure and long tenure. Short tenure is coded with 0 and contains tenures less than two years and long tenure is coded with 1 and contains tenure of two years and more.

3.3.2. Audit firm size

The second independent variable is audit firm size. In line with Myers et al., (2013) audit firms can be categorized in two group: big4 and non-big4 audit firms. Big4 audit firms are PwC, Deloitte, KMPG and EY and non-big4 are all other audit firms. Dummies are created for audit firms size, where 1 is a big4 audit firm and 0 is a non-big4 audit firm.

3.3.3. control variables

Previous research shows different other factors that could influence audit quality. Therefore we include several control variables to mitigate for this effect. These variables are size, ROA, leverage and growth.

The first control variable is the size of the client, measured as the natural logarithm of total sales. The size of firm influences its earnings management behavior (Watts and Zimmerman, 1983) in the sense that larger companies face higher political costs and are more dependent on their reputation.

Moreover, larger firms tend to report lower and more stable accruals(Dechow and Dichev, 2002).

Lang and Lungholm (1993) argue that larger firms face higher litigation risk. Smaller firms more often use earnings management to avoid losses (Lee and Choi, 2002). The second control variable is return on assets (ROA), which is measured by net income divided by total assets (Sloan, 1996). In their research, Kothari et al., (2005) found evidence that return on assets and the number of accruals are positively associated. Therefore, ROA is included as a control variable in this research.

The next control variable is the financial leverage of the firm, which is measured by debt divided by total assets(Johnson et al., 2002). Companies with more debt are likely to use their flexibility to engage in earnings management that lead to income increasing accruals (Defond & Jiambalvo, 1994). Moreover, companies with more debt are more likely to engage in earnings management that lead to loss decreasing accruals (Defond & Jiambalvo, 1994). Our last control variable is growth. Growth is measured as the sales in the current year minus the sales in the previous year, divided by the sales in the previous year(Johnson et al., 2002). Growth could influence the values of the accruals, in the sense that the growth of a company’s sales is likely to increase the absolute value of the accruals (Gold et al., 2012).

(12)

3.4. data analyses

The extreme outliers of the continuous variables are winsorized, to prevent for non-normal

distribution of the variables. The regression analysis is conducted to find significant or no-significant effects between the dependent variable, independent variables and control variables. To test the hypotheses I examine the impact of engagement partner tenure (Tenure), audit firm size (Big4) and the interaction between the independent variables on audit quality we use the following regression models :

Model 1:

DACC= β0 +β1Tenure + β2Growth + β3Size + β4ROA + β5Lev + year fixed effects + ε

Where:

DACC = discretionary accruals by Jones model measured in absolute values Tenure = dummy variable equal to 1 if the tenure is long and 0 otherwise

Big4 = dummy variable equal to 1 if the auditor is from a big-four audit firm and 0 otherwise

Growth = calculated as the sales in year t minus the sales in year t-1, divided by sales in year t-1

Size = natural logarithm of total sales

ROA = net income in year t, divided by total assets in year t Lev = debt in year t divided by total assets in year t

Model 2:

DACC= β0 +β1Big4 + β2Growth + β3Size + β4ROA + β5Lev + year fixed effects + ε

Where:

Big4 = dummy variable equal to 1 if the auditor is from a big-four audit firm and 0 otherwise

(13)

Model 3:

DACC= β0 +β1Tenure + β2Big4 + β3Tenure*Big4 + β4Growth + β5Size +β6ROA +β7Lev

+ year fixed effects + ε Where:

Tenure*Big4 = interaction between Tenure and Big4 All the other variables are the same as described above.

4. Results

4.1. Descriptive results

Table 1 provides the descriptive statistics of all variables. The variables are winsorized at a 5 percent level, to prevent outliers . The dependent variable, discretionary accuals (DACC), has a mean of 0.109, a minimum of 0.081 and a maximum of 0.369. The audit quality of a company with

discretionary accruals of 0.081 means that they tend to have high audit quality, while a company with discretionary accruals of 0.369 tend to have low audit quality. The mean of the first independent variable, whether there is a short or long tenure (Tenure), has a mean of 0.269 and a standard deviation of 0.444. Thus, in the sample 781 firm years are audited by an auditor which has a long tenure, and the remaining firm years by an auditor with a short tenure.

The second independent variable, whether the firm is audited by a big4 or non-big4 audit firm (BIG4), shows a mean of 0.604 and a standard deviation of 0.486. So, in the sample about 60 percent of the firms are audited by a big4 audit firm, and 40 percent otherwise. This is not surprising, because in Germany the big4 audit firms have a market share around 60 percent. Further, the mean of the first control variable, the sales growth (Growth), is 0.151 with a minimum of 0.008 and a maximum of 0.556. The second control variable, return on assets (ROA), shows a mean of 0.021, a minimum of -0.045 and a maximum of 0.614. When firms have a positive ROA, it means that they have a positive return on their assets. In this sample, the average firm has a positive return their assets. The average leverage in the sample is 0.205, which indicates that on average a firm has 1/5 of its assets in debt. The last control variable is the size of the firm (Size), measured as the natural logarithm of the sales. In this sample, the mean of the size is 5.221 with a standard deviation of 1.970.

(14)

4.2. Difference in mean tests

Table 2 provides difference in mean values for our independent variables. Two-sample t-tests were performed for the different partner tenure and audit firm size combinations. The table shows that 2121 observations have a short engagement partner tenure and 782 observations have a long engagement partner tenure. The mean DACC of the observations with a short engagement partner are significantly higher than the mean DACC of the observations with a long engagement partner tenure. No

conclusions can be based on the tests performed, but it indicates that longer partner tenure is associated with higher audit quality. Further, table 2 shows that 1271 observations have a short engagement partner tenure at a big4 audit firm and 483 observations have a long engagement partner tenure at a big4 audit firm. In this case, long engagement partner is also associated with higher audit quality. Moreover, 850 observations have a short engagement partner tenure at a non-big4 audit firm and 299 observations have a long engagement partner tenure at a non-big4 audit firm. At non-big4 audit firms, long engagement partner tenure is associated with higher audit quality.

Based on the above mentioned associations, I compared the difference of both the tenures between big4 and non-big 4 audit firms. Table 2 shows that 1271 observations have a short engagement tenure at a big4 audit firm and 850 observations have a short engagement tenure at a non-big4 audit firm. The mean DACC of short engagement tenure at a big4 audit firm is 0.106 and the mean DACC of short engagement partner tenure at a non-big4 audit firm is 0.124. The mean DACC of short engagement partner tenure at a non-big4 audit firm is significantly higher than same case at big4 audit firms, which indicates that short engagement partner tenure at a non-big4 audit firms is

associated with lower audit quality. At last, 783 observations have a long engagement partner tenure at a big4 audit firm and 299 observations have a long engagement partner tenure at a non-big4 audit firm. The mean DACC of long engagement partner tenure at a big4 audit firm (0.092) is significantly lower than the mean DACC of long engagement partner tenure at a non-big4 audit firm (0.108), which indicates that long engagement partner tenure at a big4 audit firm is associated with higher audit quality. No conclusions can be made based on the t-tests performed, so final conclusions will be made after the regression analysis is conducted.

Table 1: Descriptive statistics

Variable observation Mean Median Standard Minimum Maximum

Deviation DACC 2903 0.109 0.081 0.095 0.008 0.369 Growth 2903 0.151 0.097 0.150 0.008 0.556 Size 2903 5.221 4.958 1.970 2.107 9.418 Roa 2903 0.021 0.227 0.203 -0.458 0.614 Lev 2903 0.205 0.181 0.174 0.000 0.562 BIG4 2903 0,604 1.000 0. 489 0.000 1.000 Tenure 2903 0.269 0.000 0. 444 0.000 1.000

(15)

Table 2: Test in differences in mean values

variable observation Mean DACC T-statistics P-value

ST all 2121 0.114 LT all 782 0.098 4,003 0.000*** ST big4 1271 0.108 LT big4 483 0.092 3,301 0.001*** ST non-big4 850 0.124 LT non-big4 299 0.108 2,251 0.025** ST big-4 1271 0.106 ST non-big4 850 0.124 2,493 0.013** LT big4 783 0.092 LT non-big4 299 0.108 3,717 0.000*** 4.3. Correlations

Table 3 shows the correlations between all variables. The Pearson correlation matrix shows the pairwise correlations among all variables. The correlations are shown with significance at a 1%, 5% and 10% level. Results show that there is a positive correlation between the sales growth (Growth) of a firm and our dependent variable (DACC ), at a significance level of 1%. This indicates that the sales growth of a firm has a negative impact on audit quality. Besides, the results show that there is a significant negative correlation between the size of a firm (Size) and the dependent variable, which indicates that bigger firms are associated with higher audit quality. The correlation is significant at a 1% level. The correlation between the return on assets (ROA) and the discretionary accruals (DACC) shows a significant negative effect, which indicates that firms with a higher return on assets ratio, are associated with higher audit quality. Firms that are audited by Big4 audit firms tend to report better audit quality.

The correlation between Big4 and the dependent variable shows that there is a significant negative correlation (-0.083) between Big4 and DACC, the proxy for audit quality. My second independent variable, partner tenure (Tenure), shows that there is a negative correlation between partner tenure and DACC, which indicates that firms that are audited by auditors with a long tenure are associated with higher audit quality. The significance is at a 1% level. To ensure that there is no multicollinearity, the coefficients of the correlations must have a value between -0.7 and 0.7. In this model all correlation coefficients are between -0.7 and 0.7, which indicates that there is no multicollinearity.

(16)

4.4. Main results

To test the hypothesis, a multiple regression analysis is conducted. The independent variables are dummy variables. The hypotheses below are tested by the regression analysis:

H1: Audit engagement partner rotation (short tenure) has a negative effect on audit quality. H2: There is a difference in audit quality between big4 and non-big4 audit firms.

H3: There is no difference in the perceived effect of audit engagement partner rotation between big4 and non-big4 audit firms.

I performed an OLS-regression to investigate the relationship between engagement partner rotation and audit quality. Therefore, I test the relationship between the discretionary accruals (DACC), the proxy for audit quality, and the engagement partner rotation (Tenure). The first regression tests

hypothesis 1, according to which I expect a negative relationship between engagement partner rotation and audit quality. The results show a significant negative coefficient of -0.009 at a 1% significance level. This indicates that when the engagement audit partner has a long tenure, the audit quality increases. Especially, when the audit engagement partner has a long tenure, the discretionary accruals decrease by 0.009. The results provide evidence to accept hypotheses 1 and reject hypotheses 2. The results are in line with prior research, suggesting that long engagement partner tenure is associated with higher audit quality (Chen et al.,2009). The F-value of the regression is significant. The adjusted r-square is 0.117, which means that 11.7% of the changes of the dependent variable are explained through the independent variable.

The Size and ROA, show a significant negative relationship with the discretionary accruals, which suggests that larger firms and a higher return on assets is related to higher audit quality. The

coefficients are respectively -0.013 and -0.035, both coefficients are significant at a 1% level. These results are in line with other studies, that larger firms and a higher return on assets lead to higher audit quality (Dechow and Dichev, 2002; Lang and Lungholm, 1993; Lee and Choi, 2002; Kothari et al.,). The Growth and Lev, show a significant positive relationship with the discretionary accruals, which suggest that the sales growth of a firm and a higher debt is related to higher discretionary accruals,

Table 3: Pearson Correlations matrix

Variables A B C D E F G A. DAACR 1.000 B. Growth 0.179*** 1.000 C. Size -0.310*** -0.222*** 1.000 D. ROA -0.172*** -0.140*** 0.265*** 1.000 E. LEV 0.002 0.086*** 0.147*** -0.119*** 1.000 F. Big4 -0.083** -0.071*** 0.320*** 0.057*** -0.020 1.000 G. Tenure -0.074*** -0.053*** 0.054*** 0.105*** -0.021 0.016 1.0000 *** indicates the 1% significance, ** 5% and *** 10 %

(17)

thus lower audit quality. The coefficients are respectively 0.069 (1% significance) and 0.022(5% significance). These findings are in line with prior research, which suggests that the sales growth and the financial leverage of a firm are associated with lower audit quality (Gold et al., 2012; Defond & Jiambalvo, 1994).

The second hypotheses tested, is to investigate the relationship between audit firm size and audit quality. Therefore, I test the relationship between the discretionary accruals (DACC) and audit firm size (Big4). I expect a negative relationship between audit firm size and the discretionary accruals. The results showed an insignificant positive coefficient of 0.003. This indicates that when the audit is done by a big4 audit firm, the discretionary accruals increase. But, due to the fact that the coefficient is insignificant, no further conclusions can be made. I have to reject hypothesis 2. The F-value is significant and the adjusted r-square is the same as in the first regression. The results are not in line with prior research, suggesting that there is no difference in perceived audit quality between big4 and non-big4 audit firms (Francis et al.,199; Sundgren & Svanström,2013; D’Angelo, 1981). The control variables in the second regression model have the same value as in the first regression model.

The last hypothesis tested, is to investigate whether there is difference in perceived audit quality, after partner rotation between big4 and non-big4 audit firms. Therefore, I test the interaction between Tenure and Big4 on audit quality. I expect a negative effect between the interaction and the

discretionary accruals. The results showed an insignificant positive coefficient of 0.02. I can’t draw any conclusions, so we have to reject hypothesis 3. The F-value is significant and the adjusted r-squared is a little higher than in the previous regressions. I can’t compare our results with previous literature, because to my best knowledge, this is the first study that investigates this effect.

(18)

Table 4: Multiple regression analysis H1 H2 H3 DACC Tenure -0.009** --- -0.011* (0.011) (0.069) BIG4 --- 0.003 0.003 (0.342) (0.494) Tenure* BIG4 --- --- 0.002 (0.762) Growth 0.069*** 0.070*** 0.069*** (0.000) (0.000) (0.000) Size -0.013*** -0.013*** -0.013*** (0.000) (0.000) (0.000) Roa -0.038*** -0.036*** -0.037*** (0.000) (0.000) (0.000) Lev 0.022** 0.023** 0.023** (0.022) (0.018) (0.020) _cons 0.164*** 0.161*** 0.164*** (0.000) (0.000) (0.000)

Industry effects included included included Year effects included included included R-Square 0.119 0.118 0.131 Adjusted R-Square 0.117 0.116 0.1269 F-value 36.25*** 36.22*** 31.13*** Observations 2903 2903 2903 * = Coefficent is significant at level 0.1 (Two-way)

** = Coefficient is significant at level 0.05 (Two-way) *** = Coefficient is significant at level 0.01 (Two-way)

(19)

4.5. Additional tests

Additional test are performed to test for robustness and to gain more evidence-based results. The first additional test is the model of Defond and Park (2001), since they use another proxy for audit quality. The second additional test performed is to replace big4 by big3, since in Germany there is actually a big3 instead of a big4.

4.5.1. Defond and Park model (2001)

The modified Jones model is used for measuring earnings management as a proxy for audit quality. There are other proxies to measure earnings management. One of these proxies is the Defond and Park model (Defond & Park, 2001). To test for robustness, I used this model to measure earnings

management. Using this model helps to indicate the sensitiveness of the results. In this model the working capital accruals are the proxy for earnings management. The formula below shows how these accruals are calculated:

AWCA/TAt-1= WCt – ((WCt-1 / REVt-1) * REVt) / TAt-1

Where:

AWCA = abnormal working capital accruals in year t

WCt = net working capital in year t (current assets – cash and cash equivalents) – (current liabilities – short term debt)

WCt-1 = net working capital in year t-1 REVt = revenues in year t

REVt-1 = revenues in year t -1

TAt-1 = total active in year t-1

To test our hypotheses the following regression model is used:

AWCA = β0 +β1Tenure +β2Big4 + β3Growth + β4Size + β5ROA + β6Lev + year fixed effects + ε

Where:

AWCA = the absolute value of abnormal working capital accruals in year t

Table 5 shows the results of the robustness test. The findings of this research are tested on robustness by measuring audit quality with a different proxy. In the model, audit quality was measured by discretionary accruals and in the robustness test, audit quality is measured by the abnormal working capital accruals. After conducting the regression analysis, I see no difference between the dependent variables and the independent variable The results show a negative significant coefficient (Tenure) of -0.007 at a 5% significance level. This indicates that long partner tenure is associated with higher audit quality. Further, the results show an insignificant positive coefficient of 0.002, which indicates

(20)

that there is no difference in audit quality between big4 and non-big4 audit firms. It suggests that audits by a big4 audit firms decreases audit quality, but due to the insignificant correlation, no conclusions can be made.

In line with the model, the Size and ROA, show a significant negative relationship with the discretionary accruals and growth shows a significant positive relationship with the discretionary accruals. Moreover, the robustness tests shows that the financial leverage of a firm is negatively associated with the discretionary accruals, but that the relationship is not significant. This is not in line with our model, which suggests that the financial leverage of a firm has a significant negative

(21)

Table 5: Robustness test DeFond & Park model

H1 H2 H3 AWCA Tenure -0.007** --- -0.006* (0.042) (0.085) BIG4 --- 0.002 0.003 (0.468) (0.687) Tenure* BIG4 --- --- -0.001 (0.514) Growth 0.131*** 0.131*** 0.131*** (0.000) (0.000) (0.000) Size -0.007*** -0.007*** -0.006*** (0.000) (0.000) (0.002) Roa -0.022*** -0.022*** -0.024** (0.007) (0.008) (0.012) Lev -0.011** -0.011** 0.011** (0.025) (0.225) (0.028) _cons 0.092*** 0.092*** 0.092*** (0.000) (0.000) (0.000)

(22)

4.5.2. Big3 audit firms

As mentioned before, in Germany there is less dominance of big4 audit firms in comparison with other countries. In the Netherlands, for example, 95 percent of the public companies are audited by big4 audit firms. In Germany, it is around 60 percent. Actually, in Germany there no dominance of big4 audit firms. Instead, there are three audit firms in Germany that have the largest market share. These audit firms are PwC, KPMG and Ernst & Young, the big3 audit firms. To test for robustness, I conducted regression analysis with Big3 as our dependent variable, instead of big4. The following regression model is used:

DACC= β0 +β1Tenure +β2Big3 + β3Growth + β4Size + β5ROA + β6Lev + year fixed effects + ε

Where:

Big3 = dummy variable equal to 1 if the auditor is from a big3 audit firm and 0 otherwise

Table 6 shows the results for the robustness test. The findings of this research are tested for robustness by changing the independent variable big4 into big3, since in Germany there is actually a big3 instead of a big4. After the regression analysis is conducted, I see no difference in the relationship between tenure and discretionary accruals, suggesting that longer partner tenure is associated with higher audit quality. The relationship between audit firm size (big3) and discretionary accruals is still not

significant, which means that we can’t make any conclusions. But, the direction of the correlation has changed, in the model it showed a positive relationship and in the robustness test it shows a negative relationship. The directions of the control variables growth, size and ROA remain the same, but the direction of the control variable Lev changed in a negative direction, but the relationship is not significant. With the results of the robustness test, stronger evidence-based conclusions can be made.

(23)

Table 6: Robustness test Big3 H1 H2 H3 DACC Tenure -0.009** --- -0.009* (0.011) (0.097) BIG3 --- 0.005 0.005 (0.342) (0.175) Tenure* BIG3 --- --- -0.001 (0.920) Growth 0.069*** 0.070*** 0.069*** (0.000) (0.000) (0.000) Size -0.013*** -0.013*** -0.013*** (0.000) (0.000) (0.000) Roa -0.038*** -0.039*** -0.037*** (0.000) (0.000) (0.000) Lev 0.022** 0.023** 0.023** (0.022) (0.017) (0.019) _cons 0.164*** 0.161*** 0.163*** (0.000) (0.000) (0.000)

(24)

5. Discussion and conclusion

This chapter compares the results with the earlier mentioned expectations and hypotheses. Further, this chapter describes the conclusions that are based on the results and to answer the main research

question. The theoretical and practical implications will be discussed and finally the limitations and suggestions for further research of this study will be discussed.

5.1. Findings

In the financial and audit markets, the effect of audit partner tenure and audit firm size on audit quality is a topic that causes heavy discussions. Therefore this research examined the relationship between audit partner rotation (tenure) and audit quality. The study used the discretionary accruals to measure audit quality and short/long engagement partner tenure to measure audit partner rotation. Based on previous literature, it was expected that longer partner tenure influences the discretionary accruals negatively, which implies that there is a higher level of audit quality

Proponents of longer audit partner tenure argue that longer audit partner tenure increases audit quality, because longer tenure requires to generate specific knowledge about the client. Previous research has indicated that longer partner tenure is associated with higher audit quality (Chen et al. 2008; Manry et al. 2008). Johnson et al.,(2002) argued that in the first years after a partner rotation, the audit quality is lower. Their explanation is that the new incoming partner has to figure out the client specific

knowledge of risks, procedures, operations and financial reporting practices. Opponents of longer audit partner tenure believe that longer audit partner tenure decreases audit quality and that partner rotation leads to an increase in audit quality. They argue that audit partner rotation leads to an increase in audit quality, because the rotation could impair the independence of the auditor. Prior research has indicated that audit partner rotation is associated with higher audit quality (Chen et al. 2008; Manry et al. 2008). The results of this research show that longer audit partner tenure is associated with a lower level of discretionary accruals. H1 can be accepted, because significant evidence was found after conducting the regression analysis and robustness tests. So, in accordance with my expectations and several prior studies (Chen et al. 2008; Manry et al. 2008; Johnson et al.2002), this research shows that longer audit engagement partner tenure relates to higher audit quality. Additionally, this study examined the

influence of audit firm size on audit quality. Previous literature suggests that audit firm size has a positive effect on audit quality (D’Angelo, 1981; Sundgren & Svanström,2013). They argue that larger audit firms have more resources to provide higher audit quality. Therefore, this study predicted that size of an audit firm influences the audit quality positively. However, after conducting the

regression and robustness tests, I found no significant evidence to support my expectations. At last, this study examined if there is a difference in perceived audit quality after partner rotation between big4 and non-big4 audit firms. Based on the previous literature about audit partner rotation and audit firm size, I predicted that the effect of audit partner rotation is smaller for larger audit firms. After conducting the regression analysis and robustness tests, no evidence was found to confirm our expectations.

(25)

5.2. Theoretical and practical implications

This study contributes to the existing literature about audit quality regarding partner tenure and audit firm size. Prior studies found evidence that longer audit engagement partner tenure has a positive effect on audit quality (Chen et al. 2008; Manry et al. 2008; Johnson et al.2002). This study

contributes to the existing literature, by examining the effects of engagement partner tenure in a highly developed audit environment, Germany. Moreover, this study contributes to the literature

investigating the effect of audit firm size on audit quality. Prior studies (D’Angelo, 1981; Sundgren & Svanström, 2013) found that larger audit firms tend to deliver higher audit quality than smaller audit firms. But, other studies found that there is no difference in the perceived audit quality between larger and smaller audit firms (Arnet & Daros, 1979). This is the first study, to the best of my knowledge, that investigates the effect of audit partner rotation and audit firms size on audit quality.

In practice, the finding of this study can be relevant for regulators, audit firms, clients, auditors and other stakeholders. Several practical recommendations can be made, based on our research. My first recommendation is about the tenure of audit partners and their rotation. Since the SOX, it is

mandatory to rotate the partner every five years, because otherwise it could result in lower audit quality. The results showed that longer audit partner tenure provides higher audit quality. My second recommendation is to reconsider mandatory partner rotation and investigate other issues that could result in a loss of audit quality. Moreover, the results showed no significant evidence that larger audit firms provide higher audit quality. In practice, firms try to be audited by a big4 audit firm, because they suggest that big4 audit firms deliver higher audit quality. Therefore, my third recommendation is that firms that need to be audited, have to investigate which audit firm suits them best, based on their capabilities and not based on their reputation.

5.3. Limitations and suggestions for further research

Several caveats could be considered when interpreting the findings of this study. First, the findings are not represent for current companies, since the data used is covers the years 1998-2010. I suggest for future research to cover recent data, for example the years 2009-2019. Moreover, in this research I only examined the effects on German companies. Examining the effect in one country limits the generalizability, because other counties imply different laws and regulations. Therefore, I suggest for future research to include several countries to make the findings more generalizable. Moreover, I suggest to conduct a similar research in the U.S., since the discussion about audit partner rotation has taken its appearance after entering the SOX. The third limitation in this research is that I did not make a distinguish between mandatory and voluntary partner rotation. There could be a difference in the audit quality between mandatory and voluntary partner rotation. Therefore, I suggest for future research to make a distinguish between mandatory and voluntary partner rotation. The last drawback of this study is that the accruals measures are an indirect proxy for audit quality. I used discretionary accruals and abnormal working capital accruals as the proxy for audit quality, but there are several other proxies to measure audit quality.

(26)

6. References

Bamber, E. M., & Bamber, L. S. (2009). Discussion of" Mandatory Audit Partner Rotation, Audit Quality, and Market Perception: Evidence from Taiwan". Contemporary Accounting Research, 26(2), 393-402.

Barth, M. E., Beaver, W. H., & Landsman, W. R. (1998). Relative valuation roles of equity book value and net income as a function of financial health. Journal of Accounting and

Economics, 25(1), 1-34

Bedard, J. C., & Johnstone, K. M. (2010). Audit partner tenure and audit planning and pricing. Auditing: A Journal of Practice & Theory, 29(2), 45-70

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

Chen, C. J., Su, X., & Wu, X. (2009). Forced audit firm change, continued partner-client relationship, and financial reporting quality. Auditing: A Journal of Practice & Theory, 28(2), 227-246.

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. Chi, W. and Huang, H., 2005. Discretionary accruals, audit-firm tenure and audit-partner tenure:

Empirical evidence from Taiwan. Journal of Contemporary Accounting & Economics, 1(1), pp.65-92

Choi, B., & Lee, H. (2002). Knowledge management strategy and its link to knowledge creation process. Expert Systems with applications, 23(3), 173-187.

Cullinan, C. (2004). Enron as a symptom of audit process breakdown: can the Sarbanes-Oxley Act cure the disease?. Critical perspectives on Accounting, 15(6-7), 853-864.

Daugherty, B. E., Dickins, D., Hatfield, R. C., & Higgs, J. L. (2012). An examination of partner perceptions of partner rotation: Direct and indirect consequences to audit quality. Auditing: A Journal of Practice & Theory, 31(1), 97-114.

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

Dechow, P. M., Sloan, R. G., & Sweeney, A. P. (1995). Detecting earnings management. Accounting review, 193-225

DeFond, M. L., & Jiambalvo, J. (1994). Debt covenant violation and manipulation of accruals. Journal of accounting and economics, 17(1-2), 145-176.

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.

Epps, K. K., & Messier Jr, W. F. (2007). Engagement quality reviews: A comparison of audit firm practices. Auditing: A Journal of Practice & Theory, 26(2), 167-181.

(27)

Francis, J.R. and Krishnan, J., 1999. Accounting accruals and auditor reporting conservatism. Contemporary accounting research, 16(1), pp.135-165.

Francis, J.R. and Yu, M.D., 2009. Big Four office size and audit quality. The Accounting Review, 84(5), pp.1521-1552.

Francis, J.R. and Wang, D., 2008. The joint effect of investor protection and Big Four audits on earnings quality around the world. Contemporary accounting research, 25(1), pp.157-191. Francis, J.R., 2004. What do we know about audit quality?. The British accounting review, 36(4),

pp.345-368.

German commercial code

Gold, A., Lindscheid, F., Pott, C. and Watrin, C., 2012. The effect of engagement and review partner tenure and rotation on audit quality: Evidence from Germany.

Holm, C., & Zaman, M. (2012, March). Regulating audit quality: Restoring trust and legitimacy. In Accounting forum(Vol. 36, No. 1, pp. 51-61). Taylor & Francis.

Hoyle, J. (1978). Mandatory auditor rotation: The arguments and an alternative. Journal of Accountancy (pre-1986), 145(000005), 69.

Iyer, V. M., and D. V. Rama. 2004. Clients' expectations on audit judgments: A note. Behavioral Research in Accounting 16 (1): 63-74.

IFAC Code of Ethics for Professional Accountants

ISA 600. Objective and General Principles Governing an Audit of Financial Statements. IFAC 2016 Johnson, V. E., Khurana, I. K., & Reynolds, J. K. (2002). Audit‐firm tenure and the quality of

financial reports. Contemporary accounting research, 19(4), 637-660.

Klein, B., & Leffler, K. B. (1981). The role of market forces in assuring contractual performance. Journal of political Economy, 89(4), 615-641.

Kothari, S. P., Leone, A. J., & Wasley, C. E. (2005). Performance matched discretionary accrual measures. Journal of accounting and economics, 39(1), 163-197.

Lang, M., & Lundholm, R. (1993). Cross-sectional determinants of analyst ratings of corporate disclosures. Journal of accounting research, 31(2), 246-271.

Litt, B., Sharma, D. S., Simpson, T., & Tanyi, P. N. (2014). Audit partner rotation and financial reporting quality. Auditing: A Journal of Practice & Theory, 33(3), 59-86.

Manry, D. L., Mock, T. J., & Turner, J. L. (2008). Does increased audit partner tenure reduce audit quality?. Journal of Accounting, Auditing & Finance, 23(4), 553-572.

Mgbame, C. O., Eragbhe, E., & Osazuwa, N. P. (2012). Audit partner tenure and audit quality: An empirical analysis. European Journal of Business and Management, 4(7), 154-162.

(28)

Myers, J. N., L. A. Myers, and T. C. Omer. 2003. Exploring the term of the auditor-client relationship and the quality of earnings: A case for mandatory auditor rotation? The Accounting Review 78 (3): 779-799.

Stanley, J. D., & DeZoort, F. T. (2007). Audit firm tenure and financial restatements: An analysis of industry specialization and fee effects. Journal of Accounting and Public Policy, 26(2), 131-159. Sarbanes-Oxley Act (SOX). 2002. One hundred seventh congress of the United States of America.

Available at: http://www.sec.gov/about/laws/soa2002.pdf

Shockley, R. A. 1981. Perceptions of auditors' independence: An empirical analysis. The Accounting Review 56 (4): 785-800.

Sundgren, S., & Svanström, T. (2013). Audit office size, audit quality and audit pricing: evidence from small-and medium-sized enterprises. Accounting and Business Research, 43(1), 31-55. Vanstraelen, A. (2000). Impact of renewable long-term audit mandates on audit quality. European

Accounting Review, 9(3), 419-442.

Watts, R. L., & Zimmerman, J. L. (1983). Agency problems, auditing, and the theory of the firm: Some evidence. The Journal of Law and Economics, 26(3), 613-633.

Wright, A. 1988. The impact of prior working papers on auditor evidential planning judgments. Accounting, Organizations and Society 13 (6): 595–605.

Wright, W. F. (1988). Audit judgment consensus and experience. Behavioral accounting research: A critical analysis. Publishing Horizons.