This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Recently, auditor rotation has obtained wider 'publicity'. The collapses of high profile companies, such as Enron, WorldCom and Tyco International, lead to a closer public scrutiny on the credibility of financial reports. These collapses have been attributed to the poor audit quality. In turn, the independence of auditors and the requirements set to obtain high level of audit independence have also been closely monitored. One of the requirements is audit partner rotation.
The ideal of audit partner rotation policy is that audit quality would improve with the periodic rotation of the audit engagement partner, normally from 5 to 7 years. Before the 1960s, mandatory auditor rotation did not gain much public acceptance (Hoyle, 1978). A series of events, after 1960s, marked better recognition of mandatory auditor rotation among regulators, auditors and academicians. In 1961, Mautz and Sharaf suggested that long period of audit-client relationship severely deteriorates the independence and the objectivity of auditors. Therefore, a mandatory auditor rotation regime would enhance the audit quality. The Metcalf report, in 1976, recommended mandatory auditor rotation to increase competition among accounting firms and to maintain auditor independence (Hoyle, 1978). In 2002, Sarbanes-Oxley Act (SOX) amended the previous policy of a 7-year rotation (Hoyle, 1978), to the rotation of lead audit partner (or coordinating audit partner) and reviewing partner for every 5 years  . Malaysia, on the other hand, has mandated the audit partner rotation policy slightly later than the United States. The Malaysia Institute of Accountants (MIA) adopted the International Standards on Quality Control (ISQC) 1 on July 2006. Section 27 of ISQC 1 requires the lead engagement partner to rotate at least every seven years for listed companies. All these events have directed to the question on the need of mandatory audit partner rotation, and its effect on audit quality.
This paper seeks to conduct a cross-sectional study, to explore the effect of audit partner rotation on audit quality in Malaysian financially-distressed companies. Auditor rotation will be assessed through the examination of audit partner tenure, while the auditor's propensity to issue going-concern modified opinion will be used as the proxy for audit quality.
There are several motivations for this study. Carey and Simnett (2006) mentioned that the analysis on the length of audit partner tenure on audit quality, is best performed 'in an environment and at a time where the policy is not mandatory and the audit partner can be identified". Malaysia suits the environment and time, as Malaysia did not mandate the policy until 2006 and the audit partner can be identified in the auditor report. Thus, a study on audit partner tenure in Malaysia will be performed. Besides, this study will examine whether the decision to mandate audit partner rotation in Malaysia is favourable. It is hoped that this study would improve the audit quality in Malaysia and in turn, strengthen the credibility of the financial reports.
Contribution to the Literature
This paper will contribute to the existing literature in a few ways. Firstly, most of the studies carried out on audit partner tenure were based on data obtained from the developed countries (Taiwan and Australia). There are limited empirical evidence from the developing countries (including Malaysia), which examines the effect of auditor rotation on the quality of auditor's report. Secondly, there are no studies on audit partner rotation in Malaysia. There is only one related Malaysian study, investigated using archival data (Shafie, Hussin, Yusof, & Hussain, 2009) and two other questionnaire-based studies  . They examined the relationship between audit firm rotation and audit quality. However, audit partner rotation and audit firm rotation is different as both are not perfectly correlated (Chi & Huang, 2005). Lastly, there is a lack of consensus in the literature on the effect of the length of audit partner tenure on audit quality, because different findings were derived from previous studies. Therefore, this study would contribute to the current literature.
Structure of the Paper
This paper is structured as follows. Section 2 outlines the audit partner tenure policy in a few selected countries and in Malaysia. Section 3 develops the theory in audit quality and in audit partner tenure. The hypothesis will be constructed in Section 4. Section 5 presents the research methods while Section 6 summarises the paper and discusses about the limitations.
Audit Partner Tenure Policy
Audit Partner Tenure Policy in Selected Countries
The audit partner tenure policy is, essentially, different internationally. The policy of audit partner rotation was first enacted in the United States. The American Institute of Certified Public Accountants (AICPA) mandated audit partner rotation for every seven years in the 1970s. Section 203 of SOX further refined the requirement by shortening the partner rotation to every five years, in attempt to reinforce the public confidence.
In the United Kingdom, audit partner was obliged to rotate every seven years for listed companies since 1992, after the Cadbury Committee (Committee on the Financial Aspects of Corporate Governance) found in favour of the policy (Carey & Simnett, 2006). In January 2003, a review on the audit and accountancy regulations performed by the Department of Trade and Industry and the U.K. Treasury, recommended a rotation period of every five years. In addition, the rotation of every seven years is required for other key audit partners (Carey & Simnett, 2006).
In 2001, the International Federation of Accountants (IFAC)'s Code of Ethics stated that the lead engagement partner should be rotated after a pre-defined period, normally no more than seven years. The partner should not resume the role of lead engagement partner until two years have elapsed. More than 80 countries use the Code of Ethics issued by IFAC (Carey & Simnett, 2006).
As for Australia, the earliest reference to partner rotation arose in 1994, when the Australian Society of Certified Practicing Accountants and the Institute of Chartered Accountants in Australia reported on the issue (Carey & Simnett, 2006). However, the suggestion on the audit partner rotation of every seven years was not adopted. In 1996, the Australian Commonwealth Government working party recommended partner rotation of every seven years. Nevertheless, it was not adopted again (Carey & Simnett, 2006). In 2001, Australia revised their profession's standard on independence in line with the IFAC Code of Ethics (Carey & Simnett, 2006; Fargher, Lee, & Mande, 2008). Auditor rotation policy was first established in May 2002 (Fargher, et al., 2008), through the Professional Statement F1 Professional Independence, by the Institute of Chartered Accountants in Australia (ICAA). It requires the audit partner of listed companies to rotate at least every five years by 2004 (Carey & Simnett, 2006; Fargher, et al., 2008). Similarly, in Taiwan, the Taiwan Stock Exchange Corporation (TWSE) and GreTai Securities Market (GTSM) mandated rules that essentially require five-year auditor partner rotation in 2003 (Chi, Huang, Liao, & Xie, 2009).
Audit Partner Tenure Policy in Malaysia
Malaysia is a developing country, which relies on foreign direct investment to fund for growth and development. The credibility of financial reports is essential, to ensure that foreign investors are confident about the securities market in Malaysia (Teoh & Lim, 1996). This demonstrates the importance of the audit quality in financial reports.
Nasser, Wahid, Nazri and Hudaib (2006) mentioned in their paper that the issue of auditor tenure was not clearly addressed in any of the Malaysian official documents, before 2006. In March 1999, MIA, which was led by Datuk Hanifah Noordin announced the intention to propose a three- or five-year audit partner rotation for public listed companies (Shafie, et al., 2009). Nonetheless, no formal policy was enacted. Following a series of scandals in the U.S., the Malaysia Securities Commission (SC) and the Bursa Malaysia  became more concerned with the auditor rotation issue. With that, MIA and Malaysia Institute of Certified Public Accountants  (MICPA) have established an MIA/MACPA Joint Taskforce on Auditor Independence in Malaysia to examine the auditor independent issue (Shafie, et al., 2009). Once again, the proposal of mandatory audit partner rotation was recommended by MIA for a period of not more than five years. The audit partner should not resume the role of lead engagement partner until two years have elapsed. In July 2006, MIA adopted ISQC 1, mandating the rotation of audit partner for at least every seven years.
Auditor's Independence and Audit Quality
Auditor plays a vital role in reducing the agency costs incurred among the management and the shareholders, in order to increase the value of the firm (Jensen & Meckling, 1976). According to Watts and Zimmerman (1983), auditor will reduce the agency costs only if auditors are perceived to be independent. The MIA's By-Laws on Professional Ethics, Conduct and Practice  (2007) stated that independence requires independence of mind and independence in appearance. Therefore, auditor should avoid threats to independence and follow the approaches set out in the By-Laws to deal with specific circumstance. With that, the audit process can better enhance the audit quality.
The maintenance of good audit quality improves the credibility of financial reports. A higher quality audit minimises the uncertainty associated with the financial reports prepared by managers. Users of financial reports would also have more confident in the financial reports. One of the components of audit quality is the likelihood of auditor to report the irregularities detected, which relies on the independence of auditor (DeAngelo, 1981). For example, an auditor, who is impaired, might decide not to issue going-concern modified audit report, when the audit client exerts pressure on the auditor (Geiger & Raghunandan, 2002). Failure to report going-concern opinion leads to poor audit quality, and hence, audit failure (Francis, 2004). Therefore, auditor's independence is highly regarded when the auditor tries to obtain good audit quality.
Audit Partner Tenure and the Perceived Audit Quality
Generally, the public perceives audit partner rotation positively. From the audit committee point-of-view, Knapp (1991) found that audit quality is positively-correlated with auditor's tenure, if the auditor's tenure is less than 5 years. However, if the auditor's tenure is more than 5 years, the view changes into negative. Once again, the General Accounting Office  (GAO)'s study reconfirms the earlier public perception. The study was carried out on the public accounting firms and Fortune 1000's publicly traded companies. Most respondents of its survey, which studied on mandatory firm rotation and its effect on audit quality, believe that audit partner rotation will be sufficient enough to achieve the intended benefits of mandatory auditor rotation. Likewise, a survey done in Malaysia by Sori and Karbhari (2006), further supports the perception that the public favours the rotation of audit partners. Nonetheless, the focus of this paper is not on the perceived audit quality. This paper tries to examine the real quality in the auditing process.
Potential Benefits of Limiting the Length of Audit Partner Tenure on Audit Quality
Proponents of audit partner rotation support the motion to set a limit on audit partner tenure, as it is considered as one of the steps to increase audit quality. Firstly, auditors might be more independent and objective, if they are required to rotate after a period of time (Carey & Simnett, 2006). Secondly, auditors might be better in detecting fraud and irregularities in financial reports.
One of the threats to auditor's independence is familiarity threats (Hoyle, 1978; Lee & Roder, 2007; McLaren, 1958; Winters, 1976). Long association between auditors and audit clients impairs the objectivity of auditor as the auditor becomes less sceptical and more sympathetic to the clients' interests (Accountants, 2009). After working for extended period with the same client, auditor tends to put too much trust into the client (Arruñada & Paz-Ares, 1997). Moreover, the control valid in the client's audit program is presumed to be the same as the previous years (Hoyle, 1978). Auditor will also anticipate results, instead of paying attention to the details of the audit procedures (Arruñada & Paz-Ares, 1997). Unknowingly, auditor might be influenced by client's interest. As shown by Shockley (1981), auditor tends to be more complacent, becomes less innovative, prepares less rigorous audit procedures and develops confidence in client after dealing with the same client for an extended period.
Besides, Shafie et al. (2009) mentioned that prolong auditor-client relationship can undermine auditor's independence, as auditor feels committed to the client. Auditor might also collude with the client, as he/she become more intimate with the clients (McLaren, 1958; Winters, 1976). Furthermore, auditor might be worried about losing client, if he/she does not agree with the client's preferences (Hemraj, 2002). Consequently, impaired auditor might easily follow client's interest. Dopuch, King and Schwartz (2001) found that the mandatory partner rotation decreases the auditor's willingness to issue a biased audit report. Therefore, the rotation of audit partner is good, in attempt to avoid these issues and to enhance auditor's independence.
Furthermore, rotation of audit partner brings "fresh eyes" to make audit judgement (Arruñada & Paz-Ares, 1997; Fargher, et al., 2008; McLaren, 1958). Auditor rotation integrates scepticism into auditors' works, as auditors are less likely to detect fraud and irregularities after working for too long. Auditors might also become less challenged and might use less innovative audit procedures (Deis & Giroux, 1992). A new partner will judge based from different angles and he/she might be able to detect errors and irregularities that were overlooked by the previous auditor (Arruñada & Paz-Ares, 1997; Carey & Simnett, 2006; McLaren, 1958). Moreover, subjective area which necessitates professional judgement can be reviewed through rotation of auditors (Catanach & Walker, 1999). With that, the errors can be minimised.
Following these arguments, proponents support the limit set on audit partner tenure. Not only does the periodic rotation bring about improved auditor independence, it heightens the ability of auditor to detect fraud and irregularities in the financial reports.
Potential Costs of Limiting the Length of Audit Partner Tenure on Audit Quality
In contrast to the benefits mentioned in the previous section, critics oppose to the mandatory rotation of auditors due to a number of reasons. One of the reasons being a company is compelled to change the auditor, albeit improved audit quality (Hoyle, 1978). Other than that, mandatory rotation might also undermine audit quality (Arruñada & Paz-Ares, 1997; Hoyle, 1978; Johnson, Khurana, & Reynolds, 2002; Lee & Roder, 2007; Myers, Myers, & Omer, 2003).
Audit quality reduces as the new auditor does not have the knowledge of client, business and environment. There is a learning curve for auditor to obtain the knowledge needed in auditing a particular company (Pierre & Anderson, 1984). The rotation of audit partner impedes the learning curve as knowledge and experience from different clients might not be perfect substitutes (Arruñada & Paz-Ares, 1997; Chi & Huang, 2005). Auditor would need to familiarise him/herself with the audit process to enable better understanding and to improve auditors' ability to identify risks (Chi & Huang, 2005; Pierre & Anderson, 1984). A new auditor might have a higher possibility of overlooking certain errors and irregularities.
Pierre and Anderson (1984) found that the new auditors would have a higher tendency to commit errors and result in higher litigation risk, as compared to auditor with three-year experience. In addition, Carcello and Neal (2000) found that the occurrence of fraudulent financial reporting is higher in the first three years of an audit. They did not find evidence of greater fraudulent activity by client after prolonged audit tenure. Besides, Petty and Cuganesan (1996) presented that auditor rotation will undermine auditor's incentive to invest in a specific industry because auditor would need to abandon the industry periodically. Hence, there would be lesser auditor who has the speciality in a specific industry, further weakening the quality of audit. Due to these disadvantages, critics reject the idea of limiting audit partner tenure.
The Relation between Audit Partner Tenure and Audit Quality
There are a huge amount of literature on the relationship between audit firm rotation and audit quality (Geiger & Raghunandan, 2002; Jackson, Moldrich, & Roebuck, 2008; Jenkins & Velury; Johnson, et al., 2002; Myers, et al., 2003; Vanstraelen, 2000). However, the studies on the length of audit partner tenure and audit quality are limited. To date, there are only six studies, taking evidence from Australia and Taiwan (Carey & Simnett, 2006; Chen, Lin, & Lin, 2008; Chi & Huang, 2005; Chi, et al., 2009; Fargher, et al., 2008; Hamilton, Ruddock, Stokes, & Taylor, 2005). The results of these studies have been inconsistent.
Two studies have discovered about the diminishing audit quality with longer partner tenure. Hamilton, Ruddock, Stokes and Taylor (2005) supported mandatory audit partner rotation as the unexpected accruals (used as the audit quality proxy) is lower with partner changes. Similarly, Carey and Simnett (2006) found that audit quality decreased for long audit partner tenure. Using three proxies for audit quality, Carey and Simnett reported that only two measures, which are the propensity to issue a going-concern audit opinion by auditor; and the just beaten (or just missed) key earnings benchmarks, are consistent with the drop in audit quality. The use of abnormal working capital accruals did not lead to significant result. However, their study generally supports the case for mandatory partner rotation.
In addition, two other studies came out with the conclusion that audit quality increases in the initial year of engagement and decreases after a period of time. Chi and Huang (2005) found that earnings quality increases at the beginning, but decreases as partner tenure exceeds 5 years, when they examine the effect of partner tenure on earnings quality. Consistently, Fargher, Lee and Mande (2008) indicated that new audit partner coming from the same firm as the outgoing audit partner would improve audit quality in the first few years. Nonetheless, when audit partner tenure becomes lengthy (more than 6 years), managers' accounting discretion also increases, thereby decreasing the audit quality. Therefore, longer association is beneficial in the auditing process to produce higher audit quality. However, when the length of audit partner tenure becomes excessive, the audit quality will deteriorate. These studies indicate that audit partner tenure should be limited after a period of time.
In contrast, the remaining two studies reject the idea of limiting the audit partner tenure. Chen, Lin and Lin (2008) found that earnings quality increase with the increase on partner tenure. Similarly, the most recent study done by Chi, Huang, Liao and Xie (2009) on auditor rotation found that there is no support for the belief that mandatory audit partner rotation enhances audit quality. In fact, Chi et al. used sample that directly investigate the effect of mandatory audit partner rotation on earning quality (audit quality). The authors used data after the partner rotation policy was mandated, rather than voluntary data. These two studies lead to the controversy on the need of audit partner rotation.
Conceptual Framework & Hypothesis Development
There are two-sided findings on the relationship between the audit partner tenure and the audit quality. As previous studies lack of consensus on the issue of audit partner tenure, the following question arises:
Is there any association between the length of audit partner tenure and audit quality?
Schematic Diagram of the Conceptual Framework
Audit Partner Tenure
Despite the contradicting findings from Chen et al. (2008) and Chi et al. (2009), this study suggests that the length of audit partner tenure is negatively associated with the audit quality. The reason is, both Chen et al. (2008) and Chi et al. (2009) used accrual-based proxy in their research. Accrual-based proxy was found to be noisy (Chi, et al., 2009). Prior study done by Carey and Simnett (2006) also depicts the fact that accrual-based results can be contradictory. Therefore, Chen et al. (2008) and Chi et al. (2009)'s studies might be biased.
In contrast, the auditor's probability to issue going-concern modified opinion will be used as the proxy for audit quality in this study, consistent with Carey and Simnett (2006), Hopwood, McKeown and Mutchler (1994) and DeFond, Raghunandan and Subramanyam (2002). DeFond et al. (2002) mentioned in their paper that the audit report has important role to inform about the going-concern problem of audit client. Thus, auditor must be able to be free from impairment and withstand the client pressure to issue a clean opinion. In other words, there is a positive correlation between the auditor's propensity to issue going concern opinion, with the level of independence and the audit quality. Moreover, according to Sinason, Jones and Shelton (2001), auditor tenure is longer for clients who received unqualified opinions. All these suggest that the use of going-concern opinion as the proxy for audit quality is viable.
In the early period of auditing process, literature suggests that auditor has a higher failure risk to detect material misstatements (Chi & Huang, 2005; Fargher, et al., 2008; Geiger & Raghunandan, 2002). After the familiarisation period, auditor would have the knowledge and skill to option high level of audit quality (Chi & Huang, 2005; Pierre & Anderson, 1984). Carey and Simnett (2006) viewed these years as between three to seven years. The auditor is expected to maintain their independence within this period (Carey & Simnett, 2006). After this period, it was expected that the audit quality will deteriorate (Carey & Simnett, 2006). Even though the length of the audit partner tenure period is not fixed, recent policy depicts a period of either five or seven years. Based on that, this hypothesis is derived (in the alternate form):
HA: There is a negative association between the length of audit partner tenure and audit quality, ceteris paribus.
Sample and Data
The sample will consist of all 1238 Malaysian listed companies found in the OSIRIS database in 2005. The year of 2005 is selected as it is the most recent year after International Standard on Auditing (ISA) 570  came into force and the year before ISQC 1 was adopted to limit the engagement partner tenure. With that, the data will not include partner change due to rotation requirement and it will measure the going-concern opinion. The data needed to measure the audit partner tenure and the audit quality, can be obtained from the annual reports located in OSIRIS, including the name of the audit partner and the audit firm.
The sample will be confined to financially-distressed companies and non-banking and insurance industry. Previous researchers found that it is important to control for the financial health of client when trying to examine the auditor's reporting behaviour (Hopwood, et al., 1994). Moreover, the going-concern modified opinion is more salient among financially-distressed companies (DeFond, et al., 2002) as non-distressed companies does not normally issue going-concern audit opinion for non-distress company that fails suddenly (Geiger & Raghunandan, 2002). Therefore, financially-distressed companies are chosen. To control for financially-distressed companies, the sample will only include companies which report negative earnings and/or operating cash flows for the current fiscal year (Carey & Simnett, 2006). Other than that, the banking and insurance industry would also be omitted as they have incomparable total asset base and financial structure (Carey & Simnett, 2006).
Model of Audit Opinion
A logistic regression will be carried out, to examine the relationship between the length of audit partner tenure, and the auditor's propensity to issue going-concern modified opinion. The model, consistent with the approach of Carey and Simnett (2006), is as below:
OPINION = indicator variable equal to 1 if the auditor issue going-concern modified opinion, else 0
TENURE >7 = indicator variable equal to 1 if client engages the same audit partner for more than seven years, else 0
PBANK = probability of bankruptcy calculated from Carcello, Hermanson and Huss's (1995) model
SIZE = natural logarithm of total assets of client at year-end
AGE = natural logarithm of the number of years since public listing
LEV = total liabilities divided by total assets at year-end
CLEV = change in LEV during the year
RETURN = market-adjusted return over the year
LLOSS = indicator variable equal to 1 if client reported a loss for precious year, else 0
INVEST = short- and long-term investment securities (measured as current assets less debtors and inventory) divided by total assets at year-end
AFIRM = indicator variable set to 1 if audited by Big Four firm, else 0
FEERATIO = ratio of non-audit fees to total fees paid to incumbent auditor
CFO = operating cash flow divided by total assets
TENURE2 = indicator variable equal to 1 if audit partner is engagement partner on a clean company for two years or less, else 0
The dependent variable of TENURE >7 is chosen because the ISQC 1 adopted in Malaysia stated that the partner tenure should not be longer than seven years. The choice of control variable is based on the model used by Carey and Simnett (2006). However, the variable used to control for the mining companies is omitted in this model, as it is not a major industry in Malaysia.
The control variables in the model are informed by previous literature (Carcello, et al., 1995; Chen & Church, 1992; DeFond, et al., 2002; Dopuch, Holthausen, & Leftwich, 1987; Geiger & Raghunandan, 2002; Mutchler, Hopwood, & McKeown, 1997) and aid in the prediction of a going-concern opinion. Prior research suggests that financial distress, default on loan and big audit firm (PBANK, LEV, CLEV, LLOSS and AFIRM) are positively associated with the propensity to issue a going-concern opinion. Higher PBANK  value indicates higher probability of bankruptcy. LEV measures risk connected with higher levels of debt, while CLEV captures movements in leverage that might lead to unsustainable level of debt. LLOSS indicates companies with persistent losses. AFIRM captures the differences in the Big Four audit firms and non Big Four firms as big audit firms were found to be more objective with bigger client base.
In contrast, SIZE, AGE, RETURN, INVEST, FEERATIO, CFO and TENURE2 are negatively associated with the likelihood to issue going-concern opinion. Large audit clients (SIZE) have greater negotiating power and are less likely to receive going-concern audit report from auditors. The "younger" companies (AGE) are more prone to encounter financial distress. RETURN measures the market-based risk and company's performance and it is predicted to be negatively associated with going-concern opinion. INVEST captures the ability to raise cash promptly in the event of financial difficulties. FEERATIO is associated with diminished audit quality. CFO is related with a higher likelihood of financial distress as the PBANK score does not include the cash flows measure. TENURE2 is set as the control because there is an initial period for auditor to familiarise with the audit process.
Summary and Limitation
The public is concerned about the audit partner tenure policy, after recent collapses of high profile companies. Therefore, this study aims to investigate the impact of the length of audit partner tenure on the audit quality for financially-distressed companies in Malaysia. Prior studies have contradictory findings on the relationship between audit partner tenure and audit quality. This study is going to test the hypothesis which proposes that the length of audit partner tenure is negatively associated with the audit quality. The hypothesis is based on the belief that auditor would be impaired, after long association with client. This study tests the hypothesis using Carey and Simnett's (2006) logistic regression model, with the propensity to issue going-concern modified opinion as the proxy for audit quality.
There are a few limitations in this study. Firstly, the finding of this study may only be applicable for Malaysia. Secondly, the sample of this study only includes publicly-listed and financially-distressed companies. This limits the generalisation of the finding to other private entities and other non distress companies (external validity issue). Thirdly, the sample of financially-distressed companies is expected to be small, thereby raising the external validity issue. Lastly, not all financially-distressed companies, which receive going-concern opinion, will fail subsequently. Therefore, the use of going-concern opinion might lead to Type I error (rejecting a null hypothesis when it is actually true).