Audit firm tenure and audit quality.

(1)

Audit firm tenure and audit quality

Student name: Joost Bakker Student number: 6189733

Supervisor: Dr. G. Georgakopoulos

MSc Accountancy and Control, track Accountancy Amsterdam Business School

Faculty of Economics and Business, University of Amsterdam

(2)

Abstract

Mandatory rotation is considered by many government’s en regulators to be an important measure to enhance auditors independence. Prior literature does not show consistent findings about the relation between audit quality and mandatory rotation.

This study examines the relationship between audit quality and audit firm tenure. Accruals are used as a to measure audit quality. Two different methods to measure accruals are used. First the modified-Jones (1991) discretionary acrrulas (DA) model is used. Dechow (1995) found that the modified Jones model had the highest statistical power in detecting earnings management. Second the abnormal working capital accruals (AWCA) model of DeFond and Park (2001) is used.

The sample includes non-financial US corporations from the COMPUSTAT global fundamentals annual database from a period between 2002-2012. The results indicate for both the DA of the modified Jones model as well as the AWCA of DeFond and Park there is no significant relation between the medium and long term tenure periods and the accruals. Suggesting the results do not show that longer tenure is related to higher audit quality nor is there a negative effect associated with long tenure. This result is consistent with the papers of Carcello and Nagy (2004) and Johnson et al (2002) who also found no significant relation between a longer auditor-client relationship and audit quality.

The absence of evidence that mandatory audit firm rotation would enhance audit quality might be an important message for regulators in the United States and Europe.

(3)

2.4 Prior literature 11 3. Hypothesis 13 4. Research Design 4.1 Sample 14 4.2 Regression models 4.2.1 Dependent variable 14 4.2.2 Explanatory variables 16 4.2.3 Control variables 17 4.2.4 Regression model 17 5. Descriptive statistics 19 6. Results 6.1 Discretionary accruals 21

6.2 Abnormal working capital accruals 23

7. Conclusion 25

8. References 27

(4)

1 Introduction

In recent years there have been major failures in corporate financial reporting, including the collapse of Enron and WorldCom and breakdowns of other major corporations that led to restatement of financial statements and bankruptcy adversely affected thousands of shareholders and employees (U.S. Governmental Accounting Office (GAO) 2003). It is believed that the impairment of auditor independence is the main contributing factor underpinning these failures (Monsouri, Pirayesh, & Salehi, 2009).

As a result of these major failures in corporate financial reporting, the Sarbanes-Oxley Act of 2002 was enacted to protect investors by improving the accuracy and reliability of corporate disclosures (U.S. Governmental Accounting Office (GAO) 2003). The act’s requirements included reforms to strengthen corporate responsibility for financial reports and auditor independence. It has been revealed that an individual and his interpretation of information can unconsciously be influenced by certain features of his mind, which are augmented by the length of relationships (Bazerman, Loewenstein & Moore, 2002). The more familiar an auditor gets with his client, the more subjective his judgment might become, even though he may not be aware of it.

The Sarbanes-Oxley Act contains significant reforms aimed at enhancing auditor independence. A clear policy objective of the rotation requirements is to promote auditor independence. They ensure that auditors do not remain with the audited body for significant periods that may result in inappropriate relationships developing between the auditor and the audited body, compromising the independence of the audit function. Auditor independence enhances the reliability and credibility of financial reports (ASIC, Regulatory guide 187, 2007). The Sarbanes-Oxley act mandates a five-year rotation for the engagement partner (Chi et al., 2009). Also the Sarbanes-Oxley Act stated that in considering reforms to increase auditor independence, some argue that to maintain the objectivity of audits, mandatory audit firm rotation of auditing firms was necessary, other argue that the costs of mandatory audit firm rotation would outweigh the benefits and that audit firm rotation could be distorting to the public company (U.S. Governmental Accounting Office (GAO) 2003).

Some fierce resistance to the mandatory auditor rotation idea was encountered by the PCAOB encountered. It required U.S. public companies to change auditors every few years. In 2013 even the House of Representatives interfered with the act, the came up with a bill that would have amended the Sarbanex-Oxley Act to prohibit the PCAOB from requiring

(5)

companies to use specific auditors or require the use of different auditors on a rotating basis (Tysiac, 2013).

Opponents of mandatory audit firm rotation argue that it might heighten the risk of audit failure since auditors are unable to develop in-depth client-specific knowledge , it might prevent the development of an effective working relationship between auditor and management and it does not pass the cost/ benefit analyses, it would result in an increase in costs, such as set-up costs.

Proponents argue that audit firm rotation would bring to bear skepticism and a fresh perspective that the incumbent auditor may lack and it promotes objectivity because new firms are not tied down by judgments, compromises, and personal relationships of the past. A new audit firm brings a fresh set of eyes, and has the opportunity to raise issues that have been overlooked or settled in the past.

The total effect of mandatory audit firm rotation on audit quality is uncertain. Audit firm rotation can help to bring skepticism and a fresh state of mind and enhance independence but it can also deteriorate the competence of the auditor due to a lack of client-specific knowledge. The opposing results and the interference of the House of Representatives in the discussion whether or not to introduce the audit firm rotation is the motivation to perform this study. The research question this study aims to answer is the following: What is the effect of mandatory audit firm rotation on audit quality?

This study contributes to the literature on the global debate about mandatory audit firm rotation and the effect of this measure on audit quality. It contributes by providing evidence that there is no significant relationship between a longer auditor-client relationship and the level of discretionary accruals or abnormal working capital accruals. Indicating no influence on audit quality. Subsequently, mandatory firm rotation is not the measure to enhance audit quality.

The sample includes all of the non-financial US corporations on the COMPUSTAT global fundamentals annual database. The sample to calculate the accruals spans the years 2011-2012. To get a timely and relevant measurement of accruals and because these were the most recent years available with the needed data the sample spans these years. Audit firm tenure is measured as the consecutive number of years a company is audited by the same audit firm. To determine the consecutive years a company is audited by one firm data between 2002-2012 is used. Audit quality is proxied by discretionary accruals (DA) the modified-Jones (1991) and by abnormal working capital accruals as developed by De Fond and Park (2001).

(6)

The thesis is organized as follows. In the next section the theoretical background of this research is explained. Audit rotation, regulation, audit quality and the findings of prior-studies regarding audit tenure are explained. In the third section the hypotheses are discussed. In the fourth section the research design is explained. In the fifth section the results are discussed. Finally, the sixth section contains concluding remarks and limitations of the study.

(7)

2 Theoretical background

In this chapter the main concepts of this research will be explained. First the definitions of audit quality and auditor rotation are defined. Secondly, the advantages and disadvantages of auditor rotation are discussed. Lastly, the results of prior literature about the relation between auditor rotation and audit quality are discussed.

2.1 Audit quality

Audit services are demanded as monitoring devices because of the potential conflicts of interest between owners and managers as well as those among different classes of security holders (DeAngelo 1981). The quality of audit services is defined by DeAngelo (1981) as: The quality of audit services is defined to be the market-assessed joint probability that a given auditor will both (a) discover a breach in the client's accounting system, and (b) report the breach. The probability that a given auditor will discover a breach depends on the auditor's technological capabilities the audit procedures employed on a given audit, the extent of sampling, etc. The conditional probability of reporting a discovered breach is a measure of an auditor's independence from a given client

Auditor tenure can have a negative impact on either of the components mentioned by DeAngelo. Knechel (2007) states:

Long auditor tenure may increase auditor competence as the auditor can base audit decisions on extensive client knowledge that has developed over time, or it may undermine auditor independence as lengthy tenure fosters closeness between management and the auditor. Short auditor tenure may undermine auditor competence since the auditor knows less about a company in the early years of an audit, but it may also undermine auditor independence since auditors will wish to retain a new client long enough to recoup the costs of the initial audit setup or a lowball fee.

That is, impairment in audit quality in a short auditor-client relationship may be due to a loss of independence or a lack of competence, while a loss in quality in a long auditor-client relationship is most likely subject to a loss of independence (Knechel and Vanstrealen, 2007).

Following Francis (2004):

audit quality can be conceptualized as a theoretical continuum ranging from very low to very high audit quality, and outright failures (i.e., GAAP failure or audit report failure) occur on the extreme low end of quality. Given that known audit failures with material consequences are relatively infrequent.

This raises the question of how one can measure audit quality on the rest of the continuum. There is a large body of literature providing evidence of differential audit quality above and

(8)

beyond the legal and professional minimum along a number of dimensions including ﬁrm size, industry specialization, ofﬁce characteristics, and cross-country differences in legal systems and auditor liability exposure. The most commonly used outcome measures to infer audit quality are auditor litigation, audit reports, investigations by security regulators, earnings restatements, and earnings quality measures (Maijoor and Vanstraelen (2012).

In prior literature various measures are used to examine the relationship of auditor tenure with audit quality. Some of these measures are: the extent of earnings management as measured by accruals; audit failures; and the likelihood of issuing a modiﬁed or qualiﬁed opinion (Knechel and Vanstrealen, 2007).

To identify whether there is a negative association between long auditor tenure and audit quality, the level of discretionary accruals (DA) are measured as a proxy for audit quality. Jackson et al (2008) state:

DA are accruals that do not relate to normal operating activities, and so a higher level of these accruals may indicate that management has been able to exert its power over the auditor by being able to report on terms favourable to management. As this second measurement of audit quality is a continuous variable, an OLS regression is sufficient.

2.2 Auditor rotation

2.2.1 Definition of auditor rotation

Auditor rotation in general means that a company switches from her auditor to another. Auditor rotation can be accomplished by changing either the audit firm or the audit partner (Bamber, 2009). A system of mandatory audit firm rotation would require companies to rotate their independent audit firm periodically (Jackson, 2008). With audit partner rotation the audit firm remains the same and should the engagement partner and the individual responsible for the engagement quality control review be rotated after a pre-defined period (Code of Ethics for Professional Accountants, 2012, p.153). The effect on audit quality is expected to be different with the two levels of auditor rotation, so it is important to differentiate between the two.

Mandatory audit-firm rotation, which set a limit on the period of years an audit firm control a firm, are often proposed as a means to enhance auditor independence and audit quality. These proposals are based on the assumption that extended audit-firm tenure can impair auditor independence and thus setting a limit on audit-firm tenure would improve audit quality (Chi et al., 2009) (Gates et al, 2007, p.12).. There is also an expected negative side to

(9)

audit firm rotation: There is a loss in client specific knowledge, which impairs the audit quality (Geiger and Raghunandan, 2002).

2.2.2 Regulation related to audit firm rotation

For effective functioning of the capital and credit markets in the United States financial information of public companies need to be full, fair, and accurate. Publicly owned companies are required by federal securities laws and regulations to disclose financial information in a way that precisely and free from error show the results of a company’s activities and require that the companies’ financial statements be audited by an independent public accountant (U.S. Governmental Accounting Office (GAO) 2003). To enhance the public confidence in the reliability and relevance of the company’s financial statements and to decrease the agency problem, publicly traded companies need to enhance the audit process and have an independent auditor who’s integrity is not questioned (U.S. Governmental Accounting Office (GAO) 2003). To protect investors by improving the accuracy and reliability of corporate disclosures, the Sarbanes-Oxley Act of 2002 was enacted. The act’s requirements included reforms to strengthen corporate responsibility for financial reports and auditor independence and created the Public Company Accounting Oversight Board (PCAOB). It is the responsibility of the PCAOB to register and inspect public accounting firms that audit public companies, and the PCAOB has the authority to investigate and impose appropriate sanctions to the accounting firms (Tysiac, 2013).

The Sarbanes-Oxley Act contains significant reforms aimed at enhancing auditor independence. The act mandates a five-year rotation for the engagement partner (Chi et al., 2009). Also the Sarbanes-Oxley Act stated that in considering reforms to increase auditor independence, some argue that to maintain the objectivity of audits, mandatory audit firm rotation of auditing firms was necessary, other argue that the costs of mandatory audit firm rotation would outweigh the benefits and that audit firm rotation could be distorting to the public company. It was decided by the Congress that more study was necessary to decide about mandatory audit firm rotation (U.S. Governmental Accounting Office (GAO) 2003). Audit firm rotation is not mandatory in the US so we speak of a voluntary regime of audit firm rotation.

(10)

2.3 Advantages and disadvantages of auditor rotation

Now we know the concept of auditor rotation and the origin of the regulation in the US, the advantages and disadvantages of implementing mandatory auditor rotation will be discussed.

2.3.1 Advantages of auditor rotation

The reduction of audit quality due to long auditor tenure is one of the primary arguments for mandatory audit firm rotation. Below are two reasons given for a reduction in audit quality due to long auditor tenure.

First, a long auditor-client relationship may lead, consciously or not, to self-satisfaction among the audit team. Auditors or audit teams may be influenced by the reputation of some clients and may believe these clients have strong financial reporting controls, accurate financial statements, and there is no questioning mind among the audit team about the integrity and independence of these clients. If audit teams believe these characteristics hold for the upcoming years it can reduce the skepticism and vigor with which the audit team performs the audit. A new audit firm would bring to bear skepticism and a fresh perspective that the incumbent auditor may lack (Carcello and Nagy, 2004). The deterioration in the audit partner’s capacity to effect critical appraisal is known as the familiarity threat. According to the code of ethics for professional accountants the use of the same senior personnel on an assurance engagement over a long period of time may create this familiarity threat. The ICAEW states that the familiarity threat occurs when, by virtue of a close relationship with an audit client, its directors, officers or employees, an audit firm or a member of the audit team becomes too sympathetic to the client’s interests (ICAEW, guidance for audit committees, 2003)

Second, if audit firms have longstanding clients these can be perceived as a source of never ending revenue. It is argued by DeAngelo (1981) that an existing audit client provides the auditor with client specific future quasi-rents, this can be explained as all the future revenue in excess of the costs the audit firm expects to receive form this specific client during the auditor client relationship. Viewing the client as the source of some sort of never ending revenue may influence the auditors’ independence and questioning mind. A definition of auditor independence can be found in AICPA’s Principles of Code of Professional Conduct:

“Independence of mind is the state of mind that permits the performance of an attest

service without being affected by influences that compromise professional judgment, thereby allowing an individual to act with integrity and exercise objectivity and

professional scepticism”. “Independence in appearance is the avoidance of circumstances

(11)

that would cause a reasonable and informed third party, having knowledge of all relevant information, including safeguards applied, to reasonably conclude that the integrity, objectivity, or professional scepticism of a firm or a member of the attest engagement team had been comprised”.

This potential threat to auditor independence is best captured by the Commission on Public Trust and Private Enterprise:

Rotation of auditors would also reduce any financial incentives for external auditors to compromise their judgment on borderline accounting issues. In disagreeing with management, auditors would no longer be risking a stream of revenues that they believed would continue in perpetuity since the audit engagement would no longer be perceived as permanent (2003).

A new audit firm brings a fresh set of eyes and has a questioning mind, issues that have been overlooked or were settled in the past may so come to the surface and can be prevented. Mandatory audit firm rotation promotes objectivity because a new audit firm is not tied down by subjective features such as judgments, compromises, and personal relationships (Ball et al).

While a fresh set of eyes and a questioning mind look at issues that might have been overlooked by the existing auditor, mandatory audit firm rotation also has implications that might lower the quality of the audit These will be discussed in the next paragraph.

2.3.2 Disadvantages of auditor rotation

While regulatory agencies in the European Union and the U.S hope to achieve the ideal of independence through audit firm rotation, the mandatory rotation is likely to have other unintended and undesirable consequences.

First there is no room to develop in depth client-specific knowledge if rotation is mandatory, which will increase of the risk of audit failure (Ewelt et al. 2012). To develop in-depth client-specific knowledge a long auditor-client relationship is required. The audit quality will be reduced due to insufficient knowledge of the company’s ongoing day to day operations, the IT systems and financial statement and reporting practices. The issue of audit quality was specifically addressed in an AICPA Statement of Position (1992), which discussed the substantial learning curve that is needed for auditors to become familiar with the corporation’s operating environment, risks, and technical accounting policies and procedures. For an auditor to fully understand the company processes and specifics of an intricate client at

(12)

least two to three years of an auditor-client relationship are necessary (Raiborn et al., 2006. AICPA, 1992). So to the extent that a new auditor may be less familiar with the client, fraudulent financial reporting may be more likely (Geiger and Raghunandan 2002).

The second argument against mandatory rotation is that it will impair the development of an effective working relationship between auditor and management due to the shorter engagement period. Since auditors constantly need to recover start-up costs, they might low-ball, they might be more flexible toward management and less objective in judging the work of the client, as compared to a longer auditor-client relationship (Ewelt et al, 2012).

Third, according to the PCAOB (2011) there is a wide agreement that mandatory audit firm rotation would result in an increase in costs. For auditors to understand company’s processes, the organizational structure and other client specific knowledge there exists a learning curve which increases the start-up costs of the auditor. Also in the past, the audit fee in the first years might have been below marginal costs. Auditors expect to recover from this with future long term revenue in later years of a long auditor client relationship. Costs of creating familiarity between auditors and new clients and its company procedures are spread over fewer years with mandatory rotation. To overcome the spread over a shorter period audit fees might increase (Raiborn et al., 2006).

To sum up, the disadvantages of mandatory audit firm rotation are; a bigger risk of audit failure due to a lack of client specific knowledge; an ineffective working relationship and an increase in cost such as start-up costs and the learning curve.

2.4 Prior literature

A substantial amount of research has been done between audit firm tenure and audit quality, generating varied results.

Geiger and Raghunandan (2002) document in their study the relation between earnings quality and auditor tenure by using accruals as proxies for earnings quality. They assert that earnings quality can be used to draw inferences about audit quality. Their results provide evidence inconsistent with the claim that earnings quality deteriorates with extended auditor tenure under a voluntary rotation system. Further, they find some evidence that longer auditor tenure is associated with less extreme income increasing accruals. This suggests that as the relationship lengthens, auditors limit management's ability to use accruals to increase current period earnings. They also find evidence that longer auditor tenure is associated with less extreme income decreasing accruals. This suggests that as the relationship lengthens, auditors

(13)

limit management's ability to create reserves to manage future earnings. These results imply that long tenure does not reduce the audit quality.

According to Carcello and Nagy (2004) the issue of mandatory audit firm rotation is quite controversial. They provide evidence on the relation between auditor tenure and audit quality by examining the relation between audit firm tenure and fraudulent financial reporting. Their results indicate a higher frequency of fraudulent financial reporting in the early years of the auditor-client relationship. They generally fail to find support for a higher frequency of fraud when there is a long auditor-client relationship. The results of Carcello and Nagy (2004) suggest that in the early years of the auditor-client relationship audit quality may be impaired, and that in general there is no evidence that audit quality is impaired when the auditor-client relationship is long (Carcello and Nagy, 2004).

Johnson et al (2002) examines whether short and long audit tenures are associated with the issuance of lower-quality financial reports. This is done by using two proxies for financial reporting quality: first, the use of absolute value of unexpected accruals to proxy for the extent of management interventions in reported earnings numbers. Second, the extent to which current accruals persist into earnings in the subsequent year as a proxy for the quality of accruals reported in current period earnings. Their results suggest that short audit firm client relationships are associated with lower quality financial reports compared to medium audit firm relationships. In the short audit firm relationship there is greater evidence of management intervention in the reported earnings and the quality of the accruals is lower for short audit firm tenures. In contrast to the short audit firm tenure they were not able to provide evidence that a long auditor-client relationship is associated with a reduced quality of financial reporting compared to medium auditor-client relationships. No statistically differences were found between management interventions in reported earnings or accrual quality for long audit firm tenures.

Myers et al (2003) document evidence on the relation between auditor tenure and earnings quality using the dispersion and sign of both absolute Jones-model abnormal accruals and absolute current accruals as proxies for earnings quality. Results show that the negative values of both accruals measures are less extreme when auditor tenure is greater, and longer auditor tenure is generally associated with less dispersion in the distributions of accruals, which suggest higher earnings quality with longer auditor tenure. These results are interpreted as auditors with longer tenure on average place greater constraints on extreme management decisions in the reporting of financial performance.

(14)

Carey and Simnett (2006) investigate whether there is a decrease in audit quality associated with long partner tenure (in Australia) by using three common measures of audit quality: the auditor’s propensity to issue a going-concern audit opinion; an examination of the signed and absolute amount of abnormal working capital accruals; and an analysis of the extent to which key earnings targets are just beaten. They find that for longer partner tenure there is a lower propensity to issue a going concern. There is no evidence of an association of long tenure with the abnormal working accruals. So they find mixed signals. The lower propensity of issuing a going concern is a deterioration of audit quality while the lack of evidence for an association with abnormal accruals does not impair audit tenure and audit quality (Carey and Simnett, 2006).

3 Hypotheses

Proponents of mandatory audit firm rotation argue that it might heighten the risk of audit failure since auditors are unable to develop in-depth client-specific knowledge, it might prevent the development of an effective working relationship between auditor and management and it would result in an increase in costs, such as set-up costs.

Opponents argue that audit firm rotation would bring to bear skepticism and a fresh perspective that the incumbent auditor may lack and it promotes objectivity because new firms are not tied down by judgments, compromises, and personal relationships of the past. A new audit firm brings a fresh set of eyes, and has the opportunity to raise issues that have been overlooked or settled in the past.

Given the arguments for and against mandatory audit firm rotation and the requirements of regulators concerning auditor independence and audit quality, more assessments are needed to examine the effectiveness of mandatory audit firm rotation, as a measure to improve auditor independence, increase audit quality and to restore public trust.

I suggest that longer auditor tenure will result in a higher audit quality, compared to shorter auditor tenure. This view is consistent with prior research by Geiger and Raghunandan (2002), Carcello and Nagy (2004), Johnson et al (2002) and Myers et al (2003) as mentioned before. The above discussion leads to the following hypotheses

H1: Auditor tenure has a positive effect on Discretionary accruals

H2: Auditor tenure has a positive effect on Abnormal working capital accruals

(15)

.

4 Research design

4.1 Sample

The initial sample includes all of the US corporations on the COMPUSTAT global fundamentals annual database. Individual firm year observations were deleted if there were any data considerations or auditor considerations. Observations were deleted if financial data was not available; the auditor could not be determined and at least nine consecutive years had to be available. Furthermore financial companies were excluded from the sample, because their financial reporting environments differ from those of non-financial firms, they have high leverage which more likely indicate distress and might influence the level of earnings management and they have fundamentally different accrual processes that are not captured well by accruals models used in this study (Peasnell et al, 2000). To reduce the influence of outliers for the estimation of the models and thus creating a more robust estimation the data is winsorized at the 1st and 99th percentile respectively. The sample to calculate the accruals spans the years 2011-2012 as these were the most recent years with available information. To determine the consecutive years a company is audited by one single firm data between 2002-2012 is used.

SIC codes between 100 and 8742 are used. The largest group represented is the group in the between 2000 and 3000, which is the division of manufacturing (i.e. computers, cars). Numbers in the division finance, between 6000 and 6799 were excluded if applicable.

After excluding financial companies, deleting incomplete data and winsorizing to reduce the influence of outliers, there are 1124 firm year observations left.

To reduce heteroscedasticity with DA the variables β1, β2 and β3 are scaled by total lagged assets (total assets t-1) and with AWCA all variables are scaled by average total assets (Myers et al., 2003).

4.2 The regression models

4.2.1 Dependent variable

The dependent variable is the audit quality, which is influenced by the explanatory variable auditor tenure. It is difficult to measure audit quality with a direct measure, the reason for this

(16)

the limited data and/ or observability. As a result several more available indirect measures of audit quality have been developed in prior research. Indirect measures of audit quality are: Auditor size; industry expertise; auditor tenure; whether or not a company is audited by a big 4 audit firm. Audit quality for this research is determined by looking at the discretionary accruals. Some of the other indirect measures will be used as control variables.

To determine whether the financial statements are fairly presented are one of the main purpose of audit work. When audit quality is poor, the financial statements are more likely to contain items that do not represent fairly the company's real operating results and financial condition. The quality of audit work is thus reflected by the quality of reported earnings. Furthermore, poor quality earnings numbers are more likely to result in audit failures and auditor litigation, and large accruals are found to be positively associated with subsequent audit failures and auditor litigation (Geiger and Raghunandan 2002). Discretionary accruals are used as a proxy for earnings quality and investigate the relation between auditor tenure and audit quality. Following prior literature it is assumed that higher estimated discretionary accruals reflect lower earnings quality and thus lower audit quality.

To measure discretionary accruals (DA) two different models are used:

- the modified-Jones (1991) DA model. Dechow (1995) found that the modified Jones model had the highest statistical power in detecting earnings management.

- The abnormal working capital accruals model of DeFond and Park (2001)

Discretionary accruals according to the modified jones model are estimated by:

DA = TA - [β0 + β1 (∆SALES - ∆REC) + β2 PPE + β3 ROA + β4 GROWTH]

TA = Total accruals, calculated as operating net profit after tax – cash flows from operations

∆SALES = The change in sales for the year

∆REC = The change in accounts receivable during the year PPE = Gross property plant and equipment

ROA = Return on assets

GROWTH = The ratio of current year’s sales to prior years

ε = The residual from the regression ε, is the measure of DA. The higher the level of the residual indicates a lower level of accrual quality.

(17)

The coefficients β1, β2, β3, β4 and β5 are the parameters from estimating the following cross sectional version of the modified jones (1991) model:

TA = β0 + β1 (∆SALES - ∆REC) + β2 PPE + + β3 ROA + β4 GROWTH + ε The variables β1, β2 and β3 are scaled by total lagged assets (total assets t-1).

Next to the cross-sectional estimation of discretionary accruals by the Jones model, a firm-specific measure of abnormal working capital accruals was developed by De Fond and Park (2001). The abnormal working capital accruals model of DeFond and Park is based on the deviation of the current year’s working capital from an expected level of working capital required to sustain sales. As DeFond and Park argue, their model is superior to industry-wide estimates because it takes into account the firm’s size, age and accounting choices. The De Fond and Park measure of audit quality (AWCA) is also used in the Carey and Simnett (2006) study on Australian auditors. The model is useful as a measure of audit quality, as it shows the difference of realized working capital above the expected level of working capital needed to support a current sales level. The empirical proxy of abnormal accruals is as follows:

AWCA t = WC t – [(WC t–1 /S t–1 ) * S t ]

WC = non-cash working capital in year t computed as (current assets - cash & equivalents - current liabilities + current portion of long-term debt.

S = Net sales or revenue

All variables are scaled by average total assets to reduce heteroscedasticity (Myers et al., 2003).

4.2.2 Explanatory variables

Audit firm tenure is measured as the consecutive number of years a company is audited by the same audit firm. Similarly to Jonhson (2002) the number of years a company is audited by one audit firm is divided into three different periods. Johnson distinguishes between short (1-3 years), medium (4-8 years) and long tenure periods (more than 9 years). To determine short, medium and long term tenure at least nine consecutive years had to be available

(18)

4.2.3 Control variables

A number of control variables derived from prior literature which have an impact on discretionary accruals are examined (Carcello and Nagy, 2004; Myers et al., 2005).

Cash flow from operations (CFO) is negatively correlated to discretionary accruals (Dechow et al., 1995).

Leverage is also known to be significantly negatively related to discretionary accruals because high leverage might be a sign of bad financial condition (Watts and Zimmerman, 1990).

Becker et al.(1998) ﬁnd that Big 4 clients report lower absolute discretionary accruals than non-Big 4 clients, representing a higher audit quality. The size is of importance as DeAngelo’s (1981) states: Large auditors have more wealth at risk from litigation and thus have more incentives to issue accurate reports. Also Myers et al. (2005) found that smaller companies are more prone to fraudulent reporting in comparison to larger companies. Thus these two studies suggest that larger companies deliver higher audit quality.

Myers et al. (2005) also found that younger firms have a greater incentive to misstate financial statements, suggesting that the age of the firm is of importance.

According to Carey and Simnett (2006) firms with high growth opportunities are likely to be associated with earnings management.

4.2.4 Regression model

The accrual models described above are estimated by use of OLS regression. Once the dependent variable audit quality is available, its relationship with the explanatory variable is tested, while controlling for other influencing factors. To measure the hypotheses the following equations will be used.

Accruals = β0 + β1 MEDIUM + β2 LONG + β3CFO + β4LEV+ β5BIG4 + β6GROWTH + β7SIZE + ε

Accruals denote audit quality, measured by discretionary accruals or abnormal working capital accruals.

MEDIUM = Equals the number of consecutive years the client has retained the audit firm, whereas medium is between 4-8 years

(19)

LONG = Equals the number of consecutive years the client has retained the audit firm, whereas long is 9 years or more

CFO = Equals net cash flow from operations scaled lagged assets. LEV = Ratio of total liabilities to total assets

BIG4 = A dummy variable equal to one if the auditor is from a big 4 audit firm and equal to zero otherwise

GROWTH = Equals the growth rate of total assets over the previous years SIZE = the natural log of total assets

(20)

5 Descriptive statistics

In this chapter the results about the effect of auditor tenure on audit quality are discussed. The effects will be discussed in two different ways. First, by measuring the effect of auditor tenure on audit quality by using the discretionary accruals method. Secondly, by measuring the effect of auditor tenure on audit quality by using abnormal working capital accruals.

In table 1 I have included the descriptive statistics from this study. The sample consist of 1124 firm-year observations for the year 2012 after the initial sample size has been reduced due to data availability constraints and as the result of outliers. To reduce the influence of outliers for the estimation of the models and thus creating a more robust estimation the data is winsorized at the 1st and 99th percentile respectively. Also firm-year observations without the necessary data to measure the models were excluded. To determine short, medium and long term tenure at least nine consecutive years had to be available prior to 2012.

Table 1

N Mean Std. Deviation Minimum Maximum

DA 1124 -0,082 0,122 -0,635 0,939 AWCA 1124 0,002 0,086 -0,547 0,586 CFO 1124 0,034 0,278 -3,818 1,174 LEV 1124 0,653 1,196 0,000 31,858 BIG4 1124 0,527 0,500 0 1 Growth 1124 0,038 0,349 -1,000 4,599 Size 1124 15,508 2,622 -2,797 20,611 Period 1 1124 0,116 0,320 0 1 Period 2 1124 0,206 0,404 0 1 Period 3 1124 0,679 0,467 0 1

Based on these statistics we can conclude that the distribution of accruals by period of tenure is as follows: 11,6% of the sample observations belong to the short term tenure period of 1-3 years; 20,6% belongs to the medium tenure period of 4-8 years; and 67,9% belongs to the long term tenure of 9 years or more. One explanation for a large amount of long term tenure might be the currently low switching rates in the US, and also previously done analysis that found that for publicly listed companies the average auditor tenure is 21 years (European commission, 2011). Twenty large companies in the US are even been audited by the same firm for the last 80 to 120 years (European commission, 2011). The long tenure of auditors at 19

(21)

(large) companies is evidence of the lack of competition and choice in the market for the provision of audit services. Another explanation might be an increase in costs, such as set-up costs of the new auditors to understand the client’s business model and organizational structure.

Having a closer look at the control variable BIG4 is essential. 52, 7% of the firm-year observations are audited by one of the Big4 auditors. The remaining 47,3% are divided among other smaller audit firms. This is below the average market share of the Big 4 firms, which is 90% for publicly listed firms. The high concentration level of Big 4 firms might be explained by the intense M&A activity of the Big 4 audit firms until 2005, followed by a similar trend among mid‐tier and smaller audit firms; and by significant barriers to entry and development of mid‐tier firms in the audit market of listed companies. These main barriers are in particular lack of size or insufficient capacity in terms of number of auditors, limited geographical reach, a strong preference among large companies to choose Big 4 audit firms because of their reputation, resistance among companies and the absence of incentives to change the audit firm (European commission, 2011). Becker et al. (1998) found that income increasing DA for the clients of non-big auditing firms are higher than big auditing firms, suggesting that Big 4 audit firms have a higher quality of earnings. While there is almost no difference in the level of modified Jones accruals, scaled abnormal working capital accruals are lower for clients of Big4 auditors with a mean of 0.0021 compared to 0.0016 for non-Big4 clients. Thus, companies audited by a Big4 audit firm display lower levels of accruals, indicating that the audit quality of Big4 firms is higher compared to non-Big 4 auditors

Although there is a slight decreasing trend of AWCA from short to long term tenure this difference is not significant. Suggesting there is no difference in the level of AWCA for the different tenure periods. This is consistent with the Carey and Simnett study (2006) that found no association of abnormal working capital accruals with longer auditor tenure.

(22)

6 Results

The results of the regressions testing the hypothesis are provided in the following sections. Section 6.1 describes the results for the effect of auditor tenure on audit quality by using discretionary accruals. Section 6.2 shows the results for the effect of auditor tenure on audit quality by using the abnormal working capital accruals method.

6.1 Discretionary accruals

Table 2 is estimated by using the following equation:

Accruals = β0 + β1 MEDIUM + β2 LONG + β3 CFO + β4 LEV+ β5 BIG4 + β6 Growth + β7 SIZE + ε

Where accruals is measured by discretionary accruals as described by Jones (1991).

Table 2

Coefficient T- stat P-value

Constant 0,015782454 0,852 0,3943 Medium 0,00283683 0,297 0,7669 Long 0,007997665 0,949 0,3430 CFO -0,249187119 -23,949 0,0000*** LEV 0,01130417 5,062 0,0000*** BIG 4 -0,007133805 -1,194 0,2326 growth -0,052147426 -6,934 0,0000*** Size -0,006279968 -5,221 0,0000*** F-value 0,000000 R-squared 0,498917237 N 1124 *=10%, **= 5%, ***=1%

There is no significant relation between the medium and long term tenure periods and the discretionary accruals. Suggesting the results do not show that longer tenure is related to higher audit quality nor is there a negative effect associated with long tenure. This is consistent with the findings of Jackson (2008) who found there is neither an increase nor a decrease in audit quality conditional on the length of auditor tenure when using the level of discretionary accruals to measure audit quality. The F test value shows the regression in its entirety is significant. The control variables CFO, Growth and Size all have a significant

(23)

negative coefficient. While LEV has a significant positive coefficient with discretionary accruals. Consequently, firms with higher leverage have higher discretionary levels, which means that firms in bad financial condition show more earnings management. On the other hand, large companies with higher cash flow from operations and sales growth show lower discretionary accruals. This is consistent with the findings of Myers et al. (2005) who found that smaller companies are more prone to fraudulent reporting in comparison to larger companies. The Big 4 control variable is not significant.

By replacing long term tenure for short term tenure the model is re-estimated to enhance robustness.

Table 3

Coefficient T-stat P-value

(constant) 0,02378012 1,320 0,187 Short -0,007997665 -0,949 0,343 Medium -0,005160835 -0,776 0,438 CFO -0,249187119 -23,949 0,000*** LEV 0,01130417 5,062 0,000*** BIG 4 -0,007133805 -1,194 0,233 growth -0,052147426 -6,934 0,000*** Size -0,006279968 -5,221 0,000*** F-value 0,000000 R-squared 0,498917237 N 1124 *=10%, **= 5%, ***=1%

The results of table 3 are consistent with those of table 2. There is no significant relation between tenure and discretionary accruals. The regression in its entirety is significant. The control variables CFO, LEV, growth and Size are significant. This result shows there is no difference in the level of discretionary accruals during the different levels of audit tenure.

(24)

6.2 Abnormal working capital accruals

Accruals = β0 + β1 MEDIUM + β2 LONG + β3 CFO + β4 LEV+ β5 BIG4 + β6 Growth + β7 SIZE + ε

Where accruals is measured by the abnormal working capital accruals, as described by De Fond and Park (2001).

Table 4

AWCA Coefficient T- stat P-value

(constant) 0,026502439 1,447 0,148 Medium -0,007468087 -0,789 0,430 Long -0,008275473 -0,993 0,321 CFO 0,021408443 2,081 0,038** LEV -0,005838491 -2,644 0,008*** BIG 4 0,007081506 1,199 0,231 growth -0,015791158 -2,123 0,034** Size -0,001150087 -0,967 0,334 F-value 0,007695223 R-squared 0,016985887 N 1124 *=10%, **= 5%, ***=1%

There is no significant relation between the medium and long term tenure periods and the abnormal working capital accruals. Suggesting the results do not show that longer tenure is related to higher audit quality nor is there a negative effect associated with long tenure. Furthermore there is a slight decreasing trend in the mean of abnormal working capital accruals from short to long term tenure, this difference though is not significant. Suggesting there is no difference in the level of abnormal working capital accruals for the different tenure periods. This is consistent with the Carey and Simnett study (2006) that found no association of abnormal working capital accruals with longer auditor tenure. The F test value shows the regression in its entirety is significant. The control variables LEV and Growth have a significant negative coefficient. While CFO has a significant positive coefficient with abnormal working capital accruals. Consequently, firms with higher Cash flow from

(25)

operations have higher abnormal working capital accruals. On the other hand, large companies with higher LEV and sales growth show lower abnormal working capital accruals. The Big 4 and the Size variable control variable are not significant.

By replacing long term tenure for short term tenure the model re-estimated to enhance robustness.

Accruals = β0 + β1 SHORT + β2 MEDIUM + β3 CFO + β4 LEV+ β5 BIG4 + β6 Growth + β7 SIZE + ε

Table 5

Coefficient T- stat P-value

(constant) 0,018226966 1,023 0,307 Short 0,008275473 0,993 0,321 Medium 0,000807386 0,123 0,902 CFO 0,021408443 2,081 0,038** LEV -0,005838491 -2,644 0,008*** BIG 4 0,007081506 1,199 0,231 growth -0,015791158 -2,123 0,034** Size -0,001150087 -0,967 0,334 F-value 0,007695 R-squared 0,016985887 N 1124 *=10%, **= 5%, ***=1%

The results of table 5 are consistent with those of table 4. There is no significant relation between tenure and abnormal working capital accruals. The regression in its entirety is significant. The control variables LEV, growth and Size are significant. This result shows there is no difference in the level of abnormal working capital accruals during the different levels of audit tenure.

(26)

7. Conclusion

Would a fresh pair of eyes (a new audit firm) or clients specific knowledge without a questioning mind about the ins and outs of a complex company make for more effective audits? This question reflects an ongoing discussion after the 2008 financial crisis. Audit firm rotation is not a new concept, it already gained some legislative support in Europe, in 2013 the European Parliament’s Legal Affairs Committee approved a draft law that would require public-interest entities to rotate audit firms every 14 years. Regulators continue to pursue rules and legislation as the best means of enhancing auditor independence, objectivity and professional skepticism. August 2011 the PCAOB released a concept where they sought comment on whether mandatory audit firm rotation would significantly enhance auditors’ objectivity and ability and willingness to resist management pressure. During subsequent years the PCAOB received comments that mandatory audit firm rotation faces numerous hurdles and that there would be struggles with mandating rotation for public companies.

Proponents of mandatory audit firm rotation argue that it might heighten the risk of audit failure due to a lack of in-depth client-specific knowledge and it would result in an increase in costs, such as set-up costs. Whereas opponents argue that audit firm rotation would bring to bear skepticism and a fresh perspective and it promotes objectivity. A crucial question remains whether mandatory audit firm rotation would significantly enhance auditors’ objectivity and ability and willingness to resist management pressure. Therefore this study investigates whether a longer auditor-client relationship impairs the audit quality. Where there would be less objectivity and less resistance to management pressure if audit quality is low. The sample consists of 1124 firm year observations of US non-financial companies after excluding financial organizations and companies with incomplete data, gathered from the COMPUSTAT database. The period is 2011-2012. Two different measures of abnormal accruals (discretionary accruals as described by the Jones model and abnormal working capital accruals as described by De Fond and Park) are used as a proxy for audit quality. The hypotheses predicted, that a longer auditor-client relationship would have a positive effect on the audit quality. No significant relationship is found between a longer auditor-client relationship and the level of discretionary accruals or abnormal working capital accruals. The results do not show that longer tenure is related to higher audit quality nor is there a negative effect associated with long tenure. The hypothesis can therefore not be supported.

(27)

This result is consistent with the papers of Carcello and Nagy (2004) and Johnson et al (2002) who also found no significant relation between a longer auditor-client relationship and audit quality.

Overall, no significant effect is found for audit firm rotation, which implies that the US regulators should not implement mandatory audit firm rotation. The absence of evidence that mandatory audit firm rotation would enhance audit quality is a clear message for regulators in the United States and Europe. To enhance the objectivity and independence of the audit mandatory firm rotation is not the right measure, and for now the time has come to end the debate on audit firm rotation. Us regulators

As to this date not many countries have enforced mandatory auditor firm rotation. It is therefore difficult to generalize findings on audit firm rotation, making it hard to research the question if longer auditor tenure decreases audit quality in a mandatory regime.

Although mandatory firm rotation might not be the best way, with future research in countries with a mandatory regime the effect of auditor firm rotation on audit quality and auditor independence might show a different outcome. In this thesis only the effect of a longer auditor-client relationship is examined, for future research it might be interesting to perform a research on the influence of a longer audit-client relationship on the perception of audit quality by end users of the financial statements.

(28)

References

American Institute of Certified Public Accountants (AICPA). (1992). Statement of position regarding mandatory rotation of audit firms of publicly held companies. New York. Australian Securities and Investment Commission (ASIC). (2007). Regulatory guide 187 Auditor

rotation

Ball, T.C., Glover, J., Jamal, K., Kassam, R., Kouri, K., Paterson, D.B., Radhakrishnan, S. and Sunder, S. Audit Firm Rotation: A Joint Academic and Practitioner Perspective.

Bamber, E. M., and L. S. Bamber. 2009. Discussion of Mandatory audit partner rotation, audit quality, and market perception: Evidence from Taiwan. Contemporary Accounting Research 26(2): 393-402.

Bazerman, M. H., Loewenstein, G. & Moore, D. A. (2002). Why good accountants do bad audits. Harvard Business Review, 80 (11), 97-102.

Becker, C., M. DeFond, J. Jiambalvo, and K. R. Subramanyam. 1998. The effect of audit quality on earnings management. Contemporary Accounting Research 15 (1): 1–24.

Carcello, J.V. and A.L. Nagy (2004). Audit firm tenure and fraudulent financial reporting. Auditing: A Journal of Practice & Theory, 23, (2), 55-69.

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

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

Code of Ethics for Professional Accountants (2012 )

http://www.ifac.org/sites/default/files/publications/files/2012-IESBA-Handbook.pdf Commission on Public Trust and Private Enterprise. 2003. Findings and Recommendations

Part 2: Corporate Governance Part 3: Audit and Accounting. New York, NY: The Conference Board.

DeAngelo, L. 1981. Auditor independence, low balling, and disclosure regulation. Journal of Accounting and Economics 3: 113-127.

Dechow, P.M., R.G. Sloan and A.P. Sweeney (1995). Detecting Earnings Management. The Accounting Review, 70, (2), 93-225.

Defond, M.L., and C.W. Park (2001). The Reversal of Abnormal Accruals and the Market Valuation of Earnings Surprises. The Accounting Review, 76, (3), 375-404.

Durkin,P. King,A. (2012) ASIC threatens auditors with mandatory rotation, The Australian financial review.

Ewelt-Knauer, C., Gold, A. and Pott, C. (2012). What do we know about audit firm rotation. The institute of chartered accountants of Scotland.

(29)

Gates, S.K., D. Jordan Lowe, and Reckers, P.M.J. (2007). Restoring public confidence in capital markets through auditor rotation. Managerial Auditing Journal, 22, (1), 5-17.

Geiger, M.A., and K. Raghunandan, (2002). Audit tenure and audit reporting failures. Auditing: A Journal of Practice & Theory, 21, (1), 67-78.

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

Johnson, V., Khurana, I., Reynolds, J. (2002). Audit-firm tenure and the quality of financial reports, Contemporary Accounting Research 19 (4), 637-660

Maijoor, S.and A. Vanstraelen (2012). Research Opportunities in Auditing in the EU, Revisited, Auditing: A Journal of Practice Theory,31, (1), 115-126.

Monsouri, A., Pirayesh, R., & Salehi, M. (2009). Audit competence and audit quality: Case in emerging economy. International Journal of Business and Management, 4(2), 17-25. 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.

PCAOB (2011), ‘Concept release on auditor independence and audit firm rotation, PCAOB Release No. 2011-006, August 16, 2011.

Peasnell, K. V., Pope, P. F., & Young, S. (2000). Detecting earnings management using cross- sectional abnormal accruals models. Accounting and Business Research, 30(4), 313-326.

Raiborn, C., Schorg, C.A., Massoud, M., 2006. Should auditor rotation be mandatory? The Journal of Corporate Accounting and Finance 17 (4), 37–49.

Tysiac, K. (2013). Bill prohibiting mandatory audit firm rotation passes U.S. House. April 29 2014,

http://www.journalofaccountancy.com/News/20138294

U.S. Governmental Accounting Office (GAO). 2003. Report to the Senate Committee on Banking, Housing, and Urban Affairs and the House Committee on Financial Services, Public

accounting firms: Required Study on the Potential Effects of Mandatory Audit Firm Rotation. Washington, DC: GAO.

Audit firm tenure and audit quality.