The Research of Audit Committee Characteristics and BY Chong …libproject.hkbu.edu.hk/trsimage/hp/10020195.pdf · 2012-04-29 · significant relationship on above characteristics

The Research of Audit Committee Characteristics and

Changes of Auditor in Hong Kong

BY

Chong Sai Wai

10020195

Accounting Concentration

An Honours Degree Project Submitted to the

School of Business in Partial Fulfillment

Of the Graduation Requirement of the Degree of

Bachelor of Business Administration (Honours)

Hong Kong Baptist University

Hong Kong

April 2012

1

Acknowledgement

I wish to express my genuine appreciation to my supervisors, Professor Raymond Chan. I

would like to thank for his intellectual advice, valuable comments and helping me to

overcome the problems and provide guidance. Thanks for his kind support.

1. Abstract

This research would like to study whether the audit committee nature are related to

external auditor changes in Hong Kong. Even though audit committee was required to be

established under the board of directors and the basic functions and nature of audit committee

was regulated under the Listing Rules for the listed companies, it can still worthwhile to

perform a research to find out whether there are still have significant relationship under

current situation. In this research, five qualities of audit committee including independence,

financial expertise, and frequency of audit committee meetings, governance expertise and

firm specific knowledge were widely supported and practiced by regulators and identified for

this study. In order to have a clear result, we examine auditor change firms for only four

reasons and others would be excluded, the reasons include accounting disagreements, auditor

resignation, audit fee disagreements and issuance of audit qualified opinion. By applying the

logistic regression model, the result was found that it may not have any statistically

significant relationship on above characteristics with the auditor change. Besides, the research

also have studied whether each Big 4 firm have a slightly difference between each other on

the possibility of dismissed by their clients related to audit committee characteristics. This

research is quite different from the other research. Since the previous research has stated that

the higher independence of committee member, more knowledge on the firm and the more

financial expertise in the firm would be facilitate for the reduction of the chance of auditor

switches. However, it may still a valuable study because effectiveness of audit committee may

not be the critical factor of the auditor change activities in Hong Kong.

2

Table of Contents

1. Abstract 1

2. Introduction 3

3. Literature review 7

4. Statement of hypothesis 15

5. Methodology 17

6. Results 20

7. Discussion 26

8. Conclusion 29

9. Reference and citations 32

10. Appendices 36

List of table and figures

Table 1A: Number of Auditor Change Firms from 2004 to 2011 36

Table 1B: Selection of Auditor Change Firms from 2004 to 2011 36

Table 1C: Descriptive Details for sample and control 37

Table 2: Descriptive Statistics Univarite Results 38

Table 3: Correlation Matrix: Independent Variables 38

Table 4: Logistics Regression Results Auditor Change Model 39

Table 5: Logistic regression results with Modified Audit Variable 40

Fig. 1: Probability on changing external auditor related to independence 41

Fig. 2: Probability on changing external auditor related to financial expertise 41

Fig. 3: Probability on changing external auditor related to number of meeting 42

Fig. 4: Probability on changing external auditor related to board served 42

Fig. 5: Probability on changing external auditor related to years serving in board 43

3

2. Introduction

2.1.Statement of problem

The separate of ownership and the control was occurred and practiced from many

decades ago. The mangers operate and control the corporation and the shareholders act as the

owner and investor of company. Board of directors acts for the highest responsible person and

body to manage and direct the whole corporation on behalf of the company’s shareholders

(Miller 1998). Their duties include preparing reports to shareholder under their stewardship

and monitor the manager’s operations. Financial report with audited by an external auditor is

one of the reports providing for all shareholders.

Audit committee is a kind of board of directors’ sub-committees which is formulated by

the members of board of directors and supposing they are outside directors who are only

insignificant dependence on the management (Carcello and Neal 1997). In U.S., audit

committee is mandated by regulators such as New York Stock Exchange. In 2002, the U.S.

Congress had passed the law, Sarbanes-Oxley Act, requiring the audit committee is fully

independent and contains with financial expert (Lee, Mande, and Ortman, 2004). However,

since Hong Kong was previously the colony of U.K. for over 150 years. Many laws, practices

and principles follow the U.K. standard including the corporate governance frameworks and

audit committee formulation. In 1999, the Stock Exchange of Hong Kong (SEHK) had

revised the Code of Best Practice, which is the code that originally based on the code of the

London Stock Exchange. In this revised Code, audit committees in the listed corporations are

recommended. At the end of 2000, most listed firms formed the audit committee. It leads to

the criticism of market regulators such as Securities & Futures Commission and poor

reputation in public. (Firth and Rui 2006). Nowadays, the Listing Rule was be revised in order

to to comply the updated requirement for corporate governance, audit committee is enforced

to be established as the company’s compulsory subcommittee.

4

According to research of Robinson and Owens-Jackson (2009), Blue Ribbon Committee

in U.S. distinguishes five fundamental characteristics of effective audit committee which are

the independence of audit committee members, committee member’s responsibilities

commitments, financial expertise existence, firm specific knowledge and corporate

governance expertise. These qualities of auditor committee members would positively affect

the effectiveness of audit committee under prior studies. Also the desired financial reporting

quality would be increased and improved under the mentioned key characteristics (Robinson

and Owens-Jackson 2009).

Under previous researches, there are several reasons for the auditor switches. Firstly are

the audit fee arguments, which means auditors may charge the firm a premium; Secondly is

the accounting disagreement between management and external auditor, which may lead to

the inappropriately statement and improper disclosures on financial reporting; Thirdly is the

public release of qualified opinion with going concern statement to the financial statement,

which the management is seeking for the favorable audit opinion by change the new auditor;

And fourthly is the auditor resignation, which can affect the change of audit fees payment and

lead the audit quality fluctuations (Robinson and Owens-Jackson 2009).

In these years, some Hong Kong listed companies always change the auditors for

preparing the external audit report, for example, Lam Soon (HK) and Tianyi Fruit. List

companies change auditors under different reasons and circumstances, for example, fees

charged on the audit and non-audit services, auditor’s voluntary resignation and existing

auditor was dismissed by existing audit committee.

Thus, this research would try to analyze and establish the relationship of whether the five

qualities of the audit committee affecting the auditor changes under the above four resignation

reasons by applying the similar model and tests under Robinson and Owens-Jackson. Also,

5

this research would try to link up with the previous research around the auditor resignation

and observe whether this research can obtain the similar results with prior research or not.

2.2.Objectives of study

Nowadays many auditor switches happened in Hong Kong due to many reasons, for

example, audit fee dispute or different opinion with the management and external auditor.

Audit committee is usually to be elected by the board of directors and usually come from

existing company’s board of directors. No matter the members of audit committee are

working according to the rules, procedures and guidelines set up within the company or under

their favorite and self-interest, they have the rights to choose or change the auditor for

providing the audited financial reports service, and then recommend to all shareholders in the

annual or extraordinary general meeting. Although according the company ordinance in Hong

Kong, the auditor switches should be finally approved by the voting of all shareholders in the

general meeting within a certain period. This action implies that the switch of auditor is the

final collaboratively decision of all shareholders. However, if the company is majorly

controlled by one or several shareholders such as family business, which is family members

or founders of companies, they may tend to protect their interest at a priority. One of possible

situations is that, the powerful members of companies would like to put themselves as the

members of board of directors, with some other “amicable and trustable” person into the

board as the non-executive director and member of audit committee. The company daily

operation although is done by the managers, however, the manager positions may be tightly

controlled by the board of directors or the managers are originally following with these

influential shareholders before the company was listed in stock exchange. When the company

management has some accounting disagreement on change of accounting policies, treatment

of assets and earnings for the year etc, the board and audit committee may decide on auditor

switch in order to produce their acceptable audited financial reports to the public, especially

6

the bondholders, lenders and minority shareholders of company. This action may try to protect

the major shareholders’ interest, for example, the share price may be prevented from any

fluctuation. However, once some possible problems exist and this will lead all shareholders

and bondholders and the stakeholders suffers in the loss. This study on audit committee and

auditor change relationship is expected to help the investors and stakeholders like minor

shareholders and bondholders and public in Hong Kong to know more on the relationship

between audit committee, which may appointed or recommended by the board of directors,

and the auditors, and also let them to pay the proper attention on the auditor switches in the

Hong Kong listed firm to protect their financial interest.

The study would like to know whether the research of Robinson and Owens-Jackson

(2009) is validly applying in Hong Kong in current situation. Since their research was

performing in the U.S., with under the U.S. situation and regulation in the U.S. and analysis

of the U.S. stock market like New York Stock Exchange (NYSE). Before 2005, audit

committee was not required for the listed firm. Nowadays, Hong Kong Listing rule required

all listed firm must establish the audit committee as the sub-committee of the board of

directors. With adopting same audit committee characteristics mentioned by Robinson and

Owens-Jackson, this research is purposely attempting to understand the conclusion made in

their paper can be simply applied in Hong Kong with a consistently similar result or a

substantially different outcome with the prior research..

This research may also benefit to other stakeholders like boards of directors and

management as well, since it can provide the clear picture on the relationship to the boards.

The board can choose the best person to perform the member of audit committee and this will

help to encounter the special problem of the financial reporting in each firms. (Robinson and

Owens-Jackson 2009)

7

Accounting professional and external auditor can also use this research to be the

reference as they can aware on the qualities of the audit committee in each audit client They

can take the consideration of audit committee as the a part of assessing the audit risk before

the acceptance of audit engagement. (Robinson and Owens-Jackson 2009)

3. Literature review

The studies included some studies of experiential research and academic literature about

the properties of board of director and audit committee and reasons of auditors’ resignation.

In Hong Kong situation, according to Listing Rule of equity security SEHK Chapter 3,

audit committee was required to establish under the board of directors, it stated the

preliminary requirement of the members. In the Listing Rules Appendix 14, it stated that the

major functions of audit committee with transparent and formal policies should be

well-developed. They have also recommended for guidelines, and functions of the audit

committee. It included full minute of meeting should be taken and be kept, keeping

relationship with firms external auditor. The members have the responsibility to review the

financial information, internal control and financial reporting systems of the firms. Also, the

committee needs to make recommendation on choices of external auditors, and appointment,

resignation or dismissal of external auditor. The committee should be ensured that sufficient

resources should be provided for exercise their authority and deal with its responsibility.

Hong Kong Institute of Certified Public Accountants (HKICPA), which is the recognized

accounting professional body in Hong Kong, also released the guidelines for the effective

audit committee (2002). They have established more detailed guidelines for the high quality

audit committee. HKICPA is not just inform the responsibility and function of audit

committee, but also suggested some membership qualifications, the operation, conducts and

procedures required for the committee. For instance, they have suggested the size of each

committee, frequency of meetings and positions and their specific responsibility needed for

8

committee, who can be attended to the committee meetings. Also, they have established the

guideline for the communication with the board and the details for disclosure of the audit

committee to the shareholders. HKICPA have also issue the other guideline to the purpose of

improving the understanding, communication and interaction between board members, audit

committees and external auditors. It provides the aid to audit committees to have effectively

cooperated with external auditor with better understanding on the role, responsibility and

scope of external audit and the consideration of audit quality factors. This guideline suggested

the external auditor should inform the audit committee for 5 qualities of audit, including the

audit firm’s culture, audit partners and staffs’ skills and personal qualities, audit process

effectiveness, audit reporting reliability and usefulness, factors outside the control of auditors.

With understanding and familiar more on nature and responsibility of external auditor, this

reduces the possibility of change auditor because of misunderstanding and misconception of

the view towards external auditor by the audit committee.

For the independence of the committee, since the companies are be required appointing

the independent non-executive director, this can ensure the independence of audit committee.

According to the Hong Kong Institute of Directors (HKIOD), they have drafted up the

Guideline for Independent Non-executive Directors. Besides they stated that some qualities of

independent non-executive directors must have, like independence, devoting sufficient time

on the board, familiar with the company’s business, the matters confidentiality and disclosure

of its interest, they also suggesting that if these independent non-executive directors are

requested for joining the audit committee. Director should reconsider again that whether they

have sufficient time and skills to apply in the work of audit committee. HKIOD suggested that

audit committee member have required heavy workload with apply much accounting

specialized knowledge and skills. Thus, since being the committee member would also be

required for exercising due care and skills as the board member, these directors should reject

9

the offer of committee members if they are unable to exercise the duties.

As mentioned before, Robinson and Owens-Jackson (2009) state that there are five

properties on the audit committee and they found out the relationship between them and the

switched of auditor. They reported that three out of five properties of audit committees are

substantially affecting to the external auditor changes. They are including independence,

financial specialist and distinctive knowledge to the firm which these are conversely

connected with auditor change, when these three factors increase, the auditor change would be

theoretically decreased. Although the remaining two factors, diligence and governance

expertise, do not have the significant relationship with auditor change, this is still important

because this research outcome is match with the suggestion of prior research, such as change

of auditor can cause to the reduction of audit quality and agrees with the suggestion of the

Sarbanes-Oxley Act in U.S. 2002. The result directs the attention of independence and

financial expertise which would benefit to audit committee effectiveness (Robinson and

Owens-Jackson, 2009)

Also, according to Carcello and Neal (2000, 2003), state that the lower chances of

dismissal of existing auditor after issuing the report with the paragraph stating the going

concern problem under the situation of highly independence member of audit committee who

have more experience on the related position on board of directors or management at the same

time. Their papers established that some audit committees’ characteristics affect audit

committee effectiveness in positively way and reduce chance of switches auditor.

Lee, Mande and Ortman (2004) also report that the independence of audit committee and

the board of directors would diminish the possibility of auditor resignation. They found that

they have greater incentive to serve for the firm’s external auditor and reduce the hidden audit

risk of auditors. And the successor auditor would likely reduce the audit quality than before

under the situation of non- fully independent audit committee.

10

According to the research of Biedma, Ruiz-Barbadillo, and Gomez Aguilar (2010), they

have also found that the importance of audit committee independence to auditor change might

be less happened under more independent audit committee even though the company has

received the qualified opinion on audit report. Also, more independent audit committee may

choose to dismiss auditor even though they issue the clean opinion. In other words, the

independent audit committee would choose to dismiss the auditor under the condition that not

to weaken their potential independence of the audit committee Besides, more independent

committee will tend to find the auditor that have lower economic dependence to the company,

which implies that they tend to recommend the auditor for more professional and

independence on the external audit. This reflects more independent audit committee would

tend to support the auditor opinion choices. Also, as per Lennox (2002), it stated that if the

management understands that the existing auditor likely to issue the qualified opinion, they

would skillfully dismiss the existing auditor in advance and appoint the new auditor, and

some cases of auditor dismissal are happened with the absences of discussion and without

consents of audit committee. It reflects that effective audit committee would lead to reduce

the auditor changes.

Klein (2002) stated that inverse relationship between independent audit committee and

earning manipulation is demonstrated in their research, also they found that more structures

and independence board of director would be more effective in supervising the financial

accounting procedures in the company. This would help to analyze the reason of auditor

resignation as their personal reasons. Krishnan and Krishnan (1997) reported that the auditor

would reduce the audit risk by several ways, for example, increase audit fee to perform more

audit procedures and ensure the audit quality, providing more improved audit quality, better

planning, larger chance to issuance of audit modified opinion and going concern opinion,

increase the client portfolio by selecting the client and thus quitting the high risk job. The

11

potential legal risk will lead to withdrawal of the audit engagement. This study concludes that

the ligation risk is highly associated with the auditor dismissal and resignation. They also

summarize that auditor would choose resignation rather than waiting for the dismissal by the

client due to high financial stress, lower auditor independence, long seniority to performing

for the same job client, high fluctuated and receipt the modified opinion. The existence of

financial professional is important for audit committee. Krishnan, and Lee (2009), stated that

more financial expertise can reduce the ligation risk once they can establish the good

corporate governance. Also, Rich (2009) demonstrated that more financial expertise in the

audit committee would improve the financial statement and reporting quality and enhance the

corporate governance, since they have involved in the financial monitoring.

Audit committee is not only established for adopt the Listing Rule requirement, but also

need to practice its functions. It is difficult to determine its diligence and only can measure

that as the number of meetings. Raghunandan and Rama (2007) demonstrated that more audit

committee meeting would be existed that more members appointed, more meetings on board

of directors, higher non-insider holdings on the shares and the firms’ industry that with high

risk of legal actions. Also they also demonstrated that more financial experts in the committee

would be tend to hold more meetings and it may connect to a better results on the firm’s

financial reporting and auditing.

Abbott, Parker and Peters (2003) found that there are the existence of relationship

between the audit fee charged and the properties of audit committee. This paper suggests that

an audit fee is related the independence of audit committee and the existence of financial

experts in the audit committee positively in a large extent but except the relationship between

audit fee and the frequency of audit committee meeting in a financial year. This paper can

assist for this current research in Hong Kong since audit fee is one of criteria of auditor

change for considering the sample which should include in the test.

12

DeZoort and Salterio (2001) found that audit committee members who have more

governance experience that can make judgments and decisions in a consistent way and the

independent directors would tend to put a comparatively greater support for external auditors

in the dispute between auditor and the listed firm.

Knowledge on firms is also an important element; Hermalin and Weisbach (2001)

mentioned that the board composition would affect the firm performance. With the

independent directors from outside and having longer tenure, more firm-specific knowledge

could be accumulated and improve the firm’s profitability. Thus, more knowledge on the firm

may reasonable predicted that also benefit to audit committee performance if they are elected

as the members of audit committee.

However, some prior research has discovered that the audit committee. Chan, Lau and

Ng (2010) reported that the satisfactory compliance on appointing member with independence,

and financial expertise would not add the firm’s value. They suggest that insufficient

resources allocated and lack of meetings were the reasons for the ineffectiveness of audit

committee. Nevertheless, audit committee would work in effective way only base on the

combination of interacting of important characteristics and the actual practices of the board.

Beasley (1996) states his findings that the outside director characteristics and proportion

would negatively related to the fraud in the financial reporting. This suggests that the outside

auditor can contribute to the audit committee to fulfill their sight and responsibilities on

financial reporting.

Lin, Xiao and Tang (2008) found that the group of stakeholders of Chinese listed

corporations such as investors, creditors, external auditors and the company’s important

position to investigate that their degree on generally acceptance on the formal capacity and

responsibilities and audit committee. These stakeholders agree that this can improve the

auditor and the board of director communications and reduce both sides’ conflicts, and obtain

13

a better the corporate governance images. But the stakeholders are not familiar on the specific

duties of audit committee such as regulation fulfillment, better internal control maintaining

good financial reporting of companies. The outcome of this paper may not directly related to

auditor resignation since, in my opinion, the research of Lin, Xiao and Tang on the audit

committee in China would still be benefit and can assist for this research, since nowadays

many Chinese incorporated companies, including stated-owned enterprises listed and

proposed to list in Hong Kong.

For the information about auditor change, it is not just the effect of the audit committee

characteristics but also may have other reasons, for example, as the research done by Ahmed

(2010), he suggests that the increase of audit fee due to structural change of law regulations

cause the auditor switches. The smaller firms may tend to switches the auditor from the big

auditor to the non-big auditors due to reduction the growth rate of audit fee expenses rather

than the largest firms. Also, the failure of auditor business may cause the voluntary and

involuntary auditor changes. Tanyi, Raghunandan and Barua (2010) mentioned that the firm

following the original partner of original audit firm (involuntary changes) can have a shorter

time lag on preparation of audit report than the clients change to auditor and not appoint the

staff of previous firm for auditing service. This kind of auditor change may affect the choice

of auditor which it is suggested that choices of auditor either choose the completely new

auditor as an auditor rotation or choose auditor who can familiar with the company’s business.

Also, Abu Thahir, Emelin, and Hudaib et. al. (2006) suggested that the chance to audit change

depends in the tenure of the auditor. Long period audit engagement with the same auditor on a

specific client appears as auditor independence threats. Tenure of serving the client who the

client chooses to switch from small to larger auditor is slightly shorter than the auditor change

from large to smaller. Big4 auditors are able to remain longer time to serving the client than

14

the small auditors. Moreover, large size companies have tended to remain the existing auditor

more than the small size companies.

According to the research of Whisenant (2003), auditor change and removal were mostly

initiated by the clients and relatively lower proportion was due to the auditor’s decline to

stand for reelection of new engagement. The research discovered that many kinds of the

reasons of auditor switches, and they are generally divided into structural changes,

client-initiated changes (including fee dispute), auditor-client frictions changes (including

accounting disagreement and issuance of qualified opinion) and unclassified items. In our

study, we have deal with the reasons except unclassified items which is not related to audit

committee characteristics.

Finally, Robinson and Owens-Jackson (2009) had applied the same model to each of the Big 4

firm to see whether there is different result. Also, Gabre stated that if the audit committee

members are not responsible or being rejected the recommendation on the audit firm, the

firm’s management would comparatively higher dissatisfy Big 4 firms rather than local firms.

As mentioned, many China firms have listed and arrange to list in Hong Kong, change of

auditor would imply corporate governance problem. Liu (2007) stated that firms’ switching to

a larger auditor is the sign of good quality on earnings while switching to small auditor is the

negative effect on the abnormal earnings. Also, firms with larger controlling owners, and in

which the positions of the Chief Executive Officer and the board chairman would less likely

to appoint large auditors. With the change of auditor, quality of corporate governance and

audit committee would be indicated that as declining. Nevertheless, Chen and Zhou (2007)

stated that the companies which present a more active and bigger audit committees and more

board independence prefer to choose a Big 4 as new auditor.

In this research the model applied will be consistent with the paper of Robinson and

Owens-Jackson (2009), which is consistence with Beasley (1996) and Carcello and Neal

15

(2000, 2003) and other prior research like Robinson and Owens-Jackson (2009) mentioned.

4. Statement of hypothesis

The hypothesis of this research would be applying the hypothesis of Robinson and

Owens-Jackson (2009); they had produced five hypotheses between the nature of audit

committee and the resignation of auditor which they are all negatively related. These

hypotheses would be applied in this research same as Robinson and Owens-Jackson (2009).

Hypothesis 1: Higher possibility of auditor changes when the firms have a lower

percentage of independent audit committee members. Carcello and Neal (2003) found that

decreasing probability of auditor change if the committee standing as independent and not to

stand as the same side as management. Since the audit committee members should have a

more objective sight on the management and thus they can exercise a better internal control

and supervise the activities of management. If they are independent, they can have a lesser

economies and social ties to the firm and fairly judge on disputes and problems (Robinson

and Owens-Jackson 2009). This would consider that the percentage of the independent

member sitting in the committee, which independent means who is not a current and former

staff and officer of the firm, related parties employee, management relatives and professional

advisor to the firm, significant suppliers or customers officer, interlocking director, and one

who has no significant transaction with the company. (Robinson and Owens-Jackson 2009)

Hypothesis 2: Higher possibility of auditor changes when the firms have a lower

percentage of financial expertise as the members audit committee. Abott et al. (2004) stated

that material negative association between the financial restatements and the financial

specialist’s occurrence in the audit committee; this is expected also to have the same

relationship in the auditor changes. For purpose of this study, either CPAs or CFO experience

would be treated as the financial professionals, which is also the requirement under Main

16

Board Listing Rule of equity securities in Hong Kong (Chapter 3). (Robinson and

Owens-Jackson 2009)

Hypothesis 3: Higher possibility of auditor changes when the firms have a fewer audit

committee meetings. According to hypothesis of Robinson and Owens-Jackson (2009), more

audit committee meeting can reflect the greater promises towards to the firm and reflect their

interest on monitoring their company. Also, the risk of the financial restatement and legal

action taken with the external auditor would be reduced. This is expected to exercise the

effective control towards the company and effectively exercise their diligence. Abott et al.

(2004) also presented that more meetings of audit committee would stronger the audit

commitment and eventually reduce the chance of restatement of financial reports.

Hypothesis 4: Higher possibility of auditor changes when the firms’ audit committee

members have worked for fewer number of the board of directors. Carcello and Neal (2003)

reported that more directors with more board experience in the board would reduce the

occurrence of auditor dismissal. According to Robinson and Owens-Jackson (2009), since

audit committee members may need to deal with the financial reporting trouble with hardship,

if they have experience on serving in board of director in the previous company, this can help

the company to better dealing with the troubles and problems in a more experienced way. This

would be measured as average number of board served before of all audit committee

members.

Hypothesis 5: Higher possibility of auditor changes when there are shorter and decrease

on the average years of serving as the company’s audit committee member who is working on

the existing board of director or with the existing company. According to previous research,

when they have a higher seniority in the board, the members can have adequate ability to

accumulate more specific firm-based knowledge. This would lead the better earning ability,

with the effect on increase the quality of financial reporting and exercising a better and

17

effective supervision with company. This would be calculated for the average tenure of

members working in the company. (Robinson and Owens-Jackson 2009)

5. Methodology

5.1. Data Collection

Sample would be collected that appeared in the auditors change listed firms in the

website of Stock Exchange of Hong Kong (SEHK), Hong Kong Exchanges and Clearing

Limited (HKEX). The auditor change of all listed firm are published under the

“Announcements Concerning Changing Auditors of Listed Issuers” in HKEXNEWS website,

which is the public sources for searching the documents of announcement and notice from

2004. In the research, the data sample would be expected to collect through from 2004 to

2011. All samples would be tried to be identified by four reasons which were mentioned by

Robinson and Owens-Jackson (2009), this includes accounting disagreements between

external auditors and company’s management; audit fee disputes; auditor resignation due to

independence issues and person reasons; the external issuing the qualified opinion to the

listed clients.

According to Robinson and Owens-Jackson, in order to avoid making results in the

complex way, no firms of auditor change with other reasons, which is not fall in the above

mentioned reasons, like effect of fraud cases and financial restatements would be included in

the sample. In this research, some unclear reasons were also excluded like serving for the

interest of whole group and firm, streamline the audit process by changing auditor. However,

auditor resignation due to the restructuring, merger and reorganization of auditor itself would

be counted as auditor resignation, since the personnel serve for auditing, procedure and

quality would be changed and this may affect the audit committee choices of auditors.

In this research, there are the 752 samples are suitable for tested and only 130 sample

firms would be randomly selected and be tested

18

5.2.Selection of control firms

This research would like to apply the tests written by Robinson and Owens-Jackson

(2009). In this research, the control firms would be match with at least two factors with the

sample. One is the time period and the other factor is either the firm size or industry. Firm’s

Industry would be classified as the Hang Seng Industry Classification System. Unlike the

prior research, the information of firm size would be measured on the asset value rather than

net sales generated for the year of altering auditor of the corresponding testing firm, since it

can reduce the difficulty of defining the meaning of sales turnover and calculating the net

sales. Also, some prior research stated that management tends to manage and play tricks on

reported earnings if the actual growth of earnings was less than expectations, which the

managers were willing to manage the earnings in order to impress the stock markets and

persuade the regulators. (Dechow and skinner 2000 and Skinner and Sloan 1999) All firms

with earnings restatement and fraud cases, which are similar with test variables, plus auditor

substitutions for any reason, were being excluded. Also, these control firms selected should

not have any incidence of auditor switches occurred beyond the change year of auditor of both

control and test firms. Since there may have time lag of auditor change and the ending date of

financial statement. This would require for the examination on two years beyond the year of

auditor switch for confidence. (Robinson and Owens-Jackson 2009) Also, the number of

control firm is 130 which is same number as tested firms. The total number of firms for

research is 260.

5.3. Research design

Statistical model: This research would be applying in the research model used by Robinson

and Owens-Jackson, which is consistent with those prior researches mentioned in their paper.

The attached logistic regression model would be used for testing the mentioned hypothesis

studying the connections between the auditor switches and audit committee qualities.

19

𝐴𝐶 = 𝛽0 + β1𝐼𝑁𝐷𝑃 + 𝛽2𝑇𝐸𝐶𝐻 + 𝛽3𝑀𝐸𝐸𝑇 + 𝛽4𝐺𝐸𝑋𝑃 + 𝛽5𝐾𝑁𝑂𝑊 + 𝛽6𝐺𝑅𝑂𝑊𝑇𝐻

+ 𝛽7𝑆𝐼𝑍𝐸 + 𝛽8𝐴𝑈𝐷𝐼𝑇 + 𝛽9𝐹𝐸𝐸

According to Robinson and Owens- Jackson (2009), all test and control variables will be

included. In this research, the research model; dependent, test and control variables of their

paper would be applied in this study as much as possible.

Dependent variable (AC) equals one when any auditor changes was existed due to accounting

disagreements; audit fee disagreements; auditor resignation and qualified audit opinion issued

and coded zero once there is no auditor change.

Test variable: Proxy documents, annual reports and other relevant documents filed in the

SEHK and these documents are appeared publicly with no charges. They supply information

for the audit committee activity and committee members’ information

Independence (INDP), which is recorded at the percentage of independent members out

of all members on audit committees.

Financial expertise (TECH), which is proportion of members in audit committee who

have previous and current experience in CPA or CFO position.

Diligence (MEET), which is the frequency of audit committee meetings were held within

the year of auditor switches.

Governance expertise (GEXP), which is the average number of boards of directors which

each audit committee member have performing or previously performed for as at the

year of the change of auditor.

Knowledge of firm (KNOW). It included averaging number of years of audit committee

members serving on the existing board or as management in current firm.

Control variables: The control variables will be totally four, three from under the Robinson

and Owens-Jackson designs which includes growth, company size and audit firm size and

plus one additional control variable which is factor of audit fee.

20

Growth (GROWTH) is measured as the percentage change in total assets for two years

ending before the occurrence of auditor change corresponding to testing firms, which is

following the Robinson and Owens-Jackson approach (2009) and Beasley (1996). Firms

without correction on earnings was presented a comparatively a lower growth rate than

the firms that having restatement of earnings (Defond and Jiambalvo 1991). If using

growth in earnings approach, since management may always try to gain acceptance on

their decisions from the external auditor by auditor fees, the data generated may not be

reliable than the figures on growth of assets.

Firm size (SIZE) is determined as the natural logarithm of book value of total assets,

since size of listed firm will either lead the management more power incentive to make

change in the auditor or the reduction of incentive change in external auditor since firms

have more resources for better internal control (Robinson and Owens-Jackson 2009).

Auditor size (AUDIT) will put a value of either a large audit firm (which is Big 4

nowadays) or the other audit firms. The size of auditor may affect the financial reporting

quality, however, supplying the non-audit engagement services like consultation services

may also threatened the auditor independence.

Audit fee (FEE) is the final control variable this is the percentage change of the auditors’

remuneration. Fee dispute and fee arrangement would affect the choice of auditors, for

example, reducing cost on audit service by changing to small audit firm, this is related to

the interest of the shareholders and the company.

6. Results

6.1. Descriptive Statistics

Table 2 in Appendix is the comparison of means and median of each independent variable

for both test and control firms samples. Some of the audit committee characteristics are

comparatively little different between the test and the control firms. Also, control firms which

21

are no switches of auditor have a significant difference on number of meeting of audit

committee (MEET) and the firm-specific knowledge (KNOW) aspect. It implies the slightly

higher in the knowledge specifically on the firm of control firm than the test firm. Perhaps it

may surprise, there are slightly fewer audit committee meeting in the control firms comparing

with the sample firms. The remaining three variables, including the independence of audit

committee (INDP), financial expertise members (TECH) and the number of board previously

served (GEXP) were not significant. For the control variables, only AUDIT component,

which is the size of auditors, is significant effect in both set of data, while the remaining three

control variables are not significant enough. It reflected that the control listed firms tend to

choose larger size of auditor for performing audit than the test firms.

Table 3 in Appendix shows the result of correlation coefficients for the independent

variables. As Robinson and Owens-Jackson (2009) mentioned, the presence of high

correlations, which the magnitudes are +/-0.80 or greater, would be problematic on the

research. Based on Table 3, the highest correlation is -0.299. This suggests that the variables

are not multi-collinear, which it causes no matter.

Correlations for all independent variables, including GROWTH, SIZE, AUDIT and FEE

with all the audit committee variables are mixed, which different variable would be positively

or negatively related with varied audit committee characteristics

For GROWTH, the lower growth of the listed firm, implies more likely to have more

independent members, with financial expertise and board experiences. Oppositely, higher

growth rate companies may tend to have directors with and number of meeting and firm

specific knowledge.

For SIZE, larger size of listed firms may tend to appoint more independent, with

financial expertise directors and they are more willing to hold more audit committee

22

meeting. However, smaller firm size tends to have the members who have more board

experiences and higher seniority on serving the board.

For AUDIT, the large external auditor would tend to connect with audit committee with

more higher independence and more audit committee meetings while the smaller

external audit firms would appear to associate with the committee with more financial

expertise, with members of more boards served before and more members with firm-

specific knowledge.

Finally, for the FEE, the audit committee with holding more meetings, and the members

with more board experience and more knowledge towards to the firm, they are willing to

accept the higher auditor remuneration expenses. On the contrary, the audit committee

with more independence and contains more financial expertise may tend to choose to

concern on control of audit fee expenditure.

6.2.Regression analysis

Table 4 in Appendix shows the result generated from SPSS by applying logistic

regression model. This model is significant at the p<0.01 level (Two tail). The goodness of fit

(𝑅2 or pseudo-𝑅2) figure is 0.292 and the concordant pairs’ measure is 0.65. The figure of

goodness of fit is similar to the research of Robinson and Owens-Jackson (2009) and other

audit committee studies like Carcello and Neal (2000, 2003). However, unlike the prior

research, this study shows only insignificant support to overall hypothesis that reduce the

auditor switches incidents with more of the satisfactory characteristics of audit committee.

Since no test variables, which are the audit committee characteristics, reflect as significant

explanatory variables since they have a high p-value. The comparatively significant variables

are the number of meetings of audit committee (MEET, p=0.123) and the firm-specific

information by committee members (KNOW, p=0.259). The remaining three audit committee

nature were unlikely significant. Interestingly, two audit committee variables, number of

23

boards each committee member served (GEXP) and firm-specific information by board

members (KNOW) are not consistent to the hypothesis and prior research, although they are

both not significant. For the control variables, firms’ growth, firms’ size and external auditor

size are shown as significant, while only audit fee are not be classified as significant.

Although all audit committee characteristics are not significant, this model for studying

auditor switches is a logistic model, the probability of the change of auditor can be displayed

by the different levels of various audit committee nature. For each of following figures, all

remaining variables in model would be set at average value and only one specific variable of

the audit committee nature. All graphs are attached in the appendix.

Figure 1 is the testing of independence members of audit committee with the predicted

probability of the auditor change of listed firms, although the result is insignificant. The

proportion of independence increases from 0.3 to 1, the probability changes in the range of

0.36 to 0.58. At the lowest proportion, 0.3 and the highest proportion, 1.0 has the same

probability of change auditor which is 0.5. The shows the independence characteristics may

not have a direct relationship with auditor change.

Figure 2 is the testing of proportion of members of audit committee with financial

professional and predicted probability of the auditor change of listed firms which the result is

also insignificant. When the proportion of financial expertise in the committee increases from

0 to 1, the probability would be varies in the range of 0.15 to 0.6. But generally it shows the

reducing trend on probability of auditor change when the proportion of financial expertise

increases, when the proportion is 0 to 1, the probability is 0.6 and 0.45 respectively. Inverse

relationship of audit committee member with financial expert and auditor change was

reflected and it reflects the importance on existence of financial expertise in audit committee.

Figure 3 shows the testing of number of audit committee meetings during the fiscal year

and predicted probability of the auditor change of listed firms which this is the significance of

24

this variable is p=0.123, which is still not statistically significant. This evidence also shows

the inverse relationship between the probability of change auditor and the number of audit

committee meetings held in the year. The range of probability is 0.2 to 0.55 as the number of

meetings increase from 0 to 7. The probability is reduced from 0.5 to 0.2 under the number of

meeting increases from 0 to 7, and this reflects the importance to increase of meeting times in

order to reduce the likelihood of changing the auditor.

Figure 4 displays that the relationship between the governance expertise in audit

committee and the predicted probability of the switches of auditor, which the statistically

insignificance were observed. The range of probability is between 0.18 and 0.75, under the

condition of the number of the board served increases from 0 to 12. However, no general

trend can be identified since the graph shown the large fluctuation. This reflects the

independence characteristics may probably have weak relationship with auditor changes. The

switches of auditor due to the reason of different board experience of members were varied by

different companies’ size, growth or industries etc.

Figure 5 presents the relationship of the year of service on the current board and work for

the existing company, and the predicted probability of the switches of auditor, which it is

again statistically insignificant. The range of probability is between 0.25 and 0.8, under the

condition of the number of year serving the current companies increases from 0 to 15. It also

shows the direct relationship generally since there are the increasing trends from the 0.4 to 0.8

at 0 years of experience and 15 years of experience. This finding is not consist with the prior

research since the finding of Robinson and Owens-Jackson (2009) reported that there are

negative relationship between these two factors, where the chance of changing auditor should

be lower when the members’ year of service on existing firms increases. Oppositely, in this

research, the increase of firm’s specific knowledge by staying in the existing board of the

companies for longer period can increase the likelihood to change the auditor.

25

Firm growth, firm size and auditor size are the control variables that having the statistical

significance. Higher growth listed companies are comparatively less likely to change auditor

since the coefficient appears at negative value. Also, the larger firm size of listed companies

would less likely to change the existing external auditor. Interestingly, the auditors’ firm size

has negative relationship with the auditor change, reflecting that larger auditors are high

chance to be dismissed and retired by the audit committee. For the remaining control variable,

audit fee changes, it shows that the larger change audit fee may cause less likely to change of

auditor, however it is not statically significant, so this variable might be rejected.

6.3. Logistic regression results with Modified Audit Variable

In this study, performed as the study performed by Robinson and Owens-Jackson (2009),

the situations of specific large CPA firms, which are the ‘Big 4’, would be further

investigated in order to study whether these firms have greater and lesser opportunities of

being dismissed and switches by clients due to the different audit committee characteristics.

Accordingly, the additional tests were done repeatedly to obtain the specific result. The

objective is to determine whether auditor switches occur in listed firm that audited by these

specific audit firms. Existing logistic regression was re-performed with the same sample and

variables, except the auditor variable (AUDIT) would be replaces as modified auditor variable

called AUDITM. AUDITM would be coded as 1 for the Big 4 firms tested and 0 would be

coded for other audit firms. For example, if tested audit firm is KPMG, the modified auditor

variable would be coded as 1 if the sample is KPMG and other CPA firms including other

three Big4 firms would be coded as 0. This revised regression would be performed for all Big

4 firms and four regression results of these 4 firms were displayed in Table 5 in Appendix.

Among the 4 largest audit firms in the market, KPMG and PricewaterhouseCoopers were

shown that the statistically significant effect with the positive coefficients. It implies the

existence of the chances on switches of auditors have a comparatively higher risk to be

26

changed by the clients, while the chance of auditors were relatively lower under the other two

firms Deloitte and Ernst and Young. For the audit committee variables, all of these firms have

significant results on the number of meetings of audit committee. With the negative

coefficients on all CPA firms, it implies that more audit committee meetings would decrease

the possibility of change of auditor by the client dismissal. For the control variables, the listed

firm’s growth were displayed a significant results towards all Big 4 firms. Again, with the

negative coefficients appeared in all CPA firms, it implies that higher growth of client firm

would less likely to change of auditor and dismiss them.

7. Discussion

For the reasons of change auditors, including accounting disagreements, auditor

resignation, auditor remuneration dispute and receipt of qualified opinion, over a half is

falling to the fee disputes and only a few cases are related to the accounting disagreement and

audit qualified opinion. Perhaps the reason is the auditor firm may avoid being the

whistle-blower. Since the auditor may find some queries on the company’s behavior but may

not gather adequate evidence to prove them. They might choose to resign the audit

engagement and inform the public that because of other reasons arising at the same time

rather than disclose their doubt directly and simultaneously. This can remain audit firm’s

reputation in auditing industry and avoid lose clients by making clients frightened on the

sudden disclosure of the company business information.

Nowadays, according to the Listing Rule of equity securities Appendix 14 “Corporate

governance code and corporate governance report”, public listed companies are required to set

up the audit committee, and according to Listing Rule of equity securities Chapter 3, at least

one independent non-executive director should have appropriate accounting professional

qualifications or related financial management expertise and audit committee should consist

with non-executive directors only. As the result, during the collection of data, there are rare

27

cases for the large number of the directors without independence and sit in the audit

committee. Also, there are only few cases that with the absence of financial professional in

the audit committee, so it is probably the reason to explain the weak linkage between audit

committee independence and expertise and auditor change.

Moreover, although there are insignificant effects statistically on variable of governance

expertise and firm-specific knowledge, the positive coefficients existed in this research, which

is contradicted with the prior research. These positive signals imply some positive correlations.

The higher chances of existence of auditor change when the committee members have more

board experience and serving for current firms for longer period. Nevertheless, it has some

reasons behind these outcomes. Since the committee members have more experience on

performing the director duties, on the one hand, they may more concerning on the auditing

quality since they have experience on dealing with auditors. On the other hand, they may

know that what auditor would more concern on some aspects of financial statement, they

might have skills to hide up something and they might suggest for the change of external

auditor in order to avoiding any qualified opinion. If the members have served for a longer

period with the existing company’s board, they might understand more the financial position

and the operating situation of company with the updated information. On the negative side,

they might want to avoid any modified opinion on the auditors’ report and they might have

disagreement on the accounting and the internal control of the firm, however, on the bright

side, they might choose to change auditor because of the need of the streamline the auditing

process with the related parties like subsidiaries or parent company, and probably the need to

change of auditor in order to stimulate the new inspiration on the business operation and

internal control systems by cooperation.

Number of audit committee meeting is a negative correlation with the change of auditor.

Since if the audit committee can communicate with the auditors more frequently, like have

28

more regular meetings. Both sides can be more informed and audit committee can understand

more on the characteristics and the effectiveness of audit, the objective of audit and the

problems appears during the audit process. Auditor also can be ensured for adequate

opportunities to inform the committee for the material matters on their firm should be noticed

and discuss for the fee charged. Thus, more communication of the audit committee and

external auditor will lead them understand the needs of each other and enhance for the closer

cooperation and relationship and reduce the possibility to change of auditor.

For logistic regression results with modified audit variable on the specific audit firm,

KPMG and PricewaterhouseCoopers clients were shown that they have comparatively higher

possibility on the change of auditor. This reflects that they may have more disagreement

between them and their client than Deloitte and Ernst & Young. This situation can be

interpreted that Deloitte and Ernst & Young have a better customer relationship, provide the

more valuable services to their clients, and have charged fees for audit and non-audit service

that the clients may think that more reasonable. However, according to ideas of Robinson &

Owens-Jackson, it can be explained as what extent of standard has been practiced. In this case,

they may reflect that KPMG and PricewaterhouseCoopers may exercise the relatively more

strict standards on the auditing internally and lead to more clients would like to choose to

switch their auditors.

Finally, unlike prior research, no significance on the audit committee characteristics was

found in this research. The further research should be performed in the future and it should

explore more variables and more suitable measure for testing, it can be explained in two ways.

Firstly, the effectiveness of some audit committee could be doubted and at least have the room

for improvement, which means that some audit committee nature may not be identified but

still can affect the change of the auditor. This also implies that more requirements and some

natures that audit committee should have but are still not discovered. Secondly, on the

29

contrary, perhaps it is a good signal on the corporate governance in Hong Kong current

situation. Since audit committee of listed companies in Hong Kong were enforced to be

established, with consisting of independent non-executive directors and financial expertise as

the members and requiring meetings frequently. Most of the listed public companies can

adopt the audit committee characteristics in a relatively higher standard and thus increase the

degree of corporate governance. This implies the reasons of auditor change happened in these

years may not because of the weak audit committee and poor corporate governance but

because of pure commercial reasons or reasons that already discovered by audit committee.

8. Conclusion, limitation and Recommendations

To conclude, audit committee members were suggested that to have acquired the key

characteristics, for example, independence of members, professionals in the financial areas.

Those qualities of audit committee members are reflected as necessarily and effective factors

an audit committee. Also, accounting professions and regulators, which includes HKEX have

point out some requirements and critical natures for the success of audit committee and

increase its intrinsic value to the firms. In this research, five critical nature of audit committee

identified by Robinson & Owens-Jackson (2009) including independence of members,

financial expertise, audit committee meetings number, corporate governance experience and

seniority in the firm served were used for testing. This study is examining the relationship

between the mentioned five characteristics identified by the prior research whether this

relationship may still suitable in current situation of Hong Kong. However, this research

results told us that all characteristics are not significant associated with the auditor switches.

Indeed, this shows that only little effects and unclear relationship on the auditor changes with

five characteristics, and enhancing the good qualities of audit committee seems only little help

to reduce auditor changes.

30

Some limitations were consisted in this research that may affect the completeness and

accuracy of the research result. Firstly, this research is based on the list released in the Hong

Kong Exchange (HKEX) website announced by the listed companies Most of the auditor

change firms are the enterprises that comparatively small in size. The assets value of the

sample firms collected and reviewed are mostly below one billion Hong Kong dollars, and the

large size companies with one billion dollars assets value above are rarely change auditors. So

the research may not be comprehensive enough. Secondly, most of annual reports of listed

companies would disclose the bibliophiles of all directors and senior managements, however,

most of them only disclose the general information of the mentioned important figures. They

would not fully describe the previous board experience of the directors, for example, they do

not disclose the number of firm served before be appointed for the existing firm’s director.

Even though search of director can be made in HKEX website, it is still not sure that whether

the director has served the overseas companies and private companies. Nevertheless, the data

on previous board served has been ensured that have been collected as accurate as possible.

During the research, although all listed companies are required to set up the audit

committees for fulfilling the listing rules. First at all, the listed companies in the main board

should be enforced to released quarterly reports, which as same as the listed companies in

GEM Board requirement. (Listing rules of equity securities of GEM Board Chapter 18). Since

audit committee are required to preliminary review the result before the release to the public,

the number of audit committee meetings of GEM board companies would be relatively higher

than main board companies, so the research reflects more board meeting and increase the

probability of change auditor. Secondly, in this research and the prior research of Robinson &

Owens-Jackson, the research did not involve element of average attendance of the audit

committee meetings. Since I have found that some independent non-executive director did not

attend the audit committee meeting, in order to increase the corporate governance quality, the

31

regulators and HKEX should amend the listing rules to requiring the independent

non-executive director should attend at least 50% of meetings. Otherwise, the directors should

disclose the reason why they did not attend meetings and the board of directors should not

recommend this director to shareholder for the re-appointment for some period. Since the

independent non-executive director should consistently practice their professional duties and

they should put sufficient time on the company (Guideline of independent non-executive

director P.3-4), even though they are under pressure, if these directors did not attend the

meetings, there is the doubt that whether they have spending time, efforts and abilities on

managing and supervising the listed firm effectively.

Even though the research result may not be clear to show the relationship, it can be

considered that this result still expresses some remarkable signals. Firstly, this suggests that

the change of auditor may mainly because of normal and reasonable business reasons, for

example, reduce the audit fee for the company’s interest. And it shows that change of auditor

not mainly due to tricks of some dominant shareholders and powerful directors played in the

audit committee. The high quality of audit committee would provide the confidence and more

meaningful financial reporting to the investors, potential investors and the other related users.

However, since nowadays many listed companies would change their external auditor at the

time of financial situation variations, which may tend to be bad side, the regulators and

accounting professionals and other stakeholders, should concern and explore more necessarily

and qualitative characteristics of audit committee in order to polish and update the recent

standards on audit committee and increase effectiveness of audit committee. Regulators,

accounting professions and directors and audit committee members and other stakeholder

should keep on work hard and maintain high standard of quality of audit committees.

32

9. Reference and citation

Abbott, LJ, Parker, S, Peters, GF, Abbott, LJ (2004) "Audit committee characteristics and

restatements" Auditing: A Journal of Practice & Theory 23 1 69-87

Abu Thahir, A. N., Emelin, A. W., Sharifah Nazatul Faiza Syed,Mustapha Nazri, &

Hudaib, M. (2006). Auditor-client relationship: The case of audit tenure and auditor

switching in malaysia. Managerial Auditing Journal, 21(7), 724-737.

doi:10.1108/02686900610680512

Ahmed, E. (2010). Audit fee premium and auditor change: The effect of sarbanes-oxley

act. Managerial Auditing Journal, 25(2), 102-121. doi:10.1108/02686901011008936

Biedma López, E., Ruiz-Barbadillo, E., & Gomez Aguilar, N. (2010). Do independent

audit committees prevent auditor opinion shopping?. Rochester, Rochester:

http://search.proquest.com/docview/821919074?accountid=11440

Beasley, Mark S. (Oct 2006) an empirical analysis of the relation between the board of

director composition and financial statement fraud. The Accounting Review; 71, 4;

443-465

Carcello, Joseph V. and Neal, Terry L. (July 1997) Audit Committee Characteristics and

Auditor Reporting, SSRN Working Paper Series.

Carcello, Joseph V. and Neal, Terry L. Neal (2000), 'Audit Committee Composition and

Auditor Reporting', The Accounting Review, 75, 4: 453-67.

Carcello, Joseph V. and Neal, Terry L. (Jan 2003) "Audit committee characteristics and

auditor dismissals following "new" going-concern reports." The Accounting Review 78:1:

95-117

Chan, R. S. Y., Lau, C. K. S., & Ng, A. W. (2011). Compliance and value relevance of

audit committees: Evidence from hong kong. Journal of Financial Reporting and

Accounting, 9(1), 74-97. doi:10.1108/19852511111139813

33

Chen, K. Y., & Zhou, J. (2007). Audit committee, board characteristics, and auditor

switch decisions by andersens clients. Contemporary Accounting Research, 24(4),

2-1117. http://search.proquest.com/docview/194213662?accountid=11440

Dechow, P. M., & Skinner, D. J. (2000). Earnings management: Reconciling the views of

accounting academics, practitioners, and regulators. Accounting Horizons, 14(2),

235-250. http://search.proquest.com/docview/208922433?accountid=11440

DeFond, M. L., & Jiambalvo, J. (1991). Incidence and circumstances of accounting

errors. The Accounting Review, 66(3), 643-643.

http://search.proquest.com/docview/218546417?accountid=11440

DeZoort, F. T., & Salterio, S. E. (2001). The effects of corporate governance experience

and financial-reporting and audit knowledge on audit committee members judgments.

Auditing, 20(2), 31-47.

http://search.proquest.com/docview/216741209?accountid=11440

Firth, M., & Rui, O. M. (2007). Voluntary audit committee formation and agency costs.

Rochester: Retrieved from

http://search.proquest.com/docview/189877582?accountid=11440

Gabre, H. G. (2008). An empirical investigation of audit committee decision quality. The

University of Mississippi). ProQuest Dissertations and Theses, , n/a.

http://search.proquest.com/docview/304523848?accountid=11440

Hermalin, B. E., & Weisbach, M. S. (1991). The effects of board composition and direct

incentives on firm performance. Financial Management, 20(4), 101-101.

http://search.proquest.com/docview/208184242?accountid=11440

Hong Kong Exchange and Clearance Limited, Chapter 3, Main Board Listing Rules

updated at 30-03-2012

http://www.hkex.com.hk/eng/rulesreg/listrules/mbrules/documents/chapter_3.pdf

34

Hong Kong Exchange and Clearance Limited, Chapter 13, Main Board Listing Rules

updated at 30-03-2012

http://www.hkex.com.hk/eng/rulesreg/listrules/mbrules/documents/chapter_13.pdf

Hong Kong Exchange and Clearance Limited, Chapter 18, GEM Board Listing Rules,

updated at 30-03-2012

http://www.hkex.com.hk/eng/rulesreg/listrules/gemrules/documents/chapter_18.pdf

Hong Kong Institute of Certified Public Accountants, A Guide for effective Audit

committees, 2002, http://app1.hkicpa.org.hk/publications/audit/audite.pdf

Hong Kong Institute of Certified Public Accountants, Audit Committee and Audit

Quality, May 2010,

http://www.hkicpa.org.hk/file/media/section6_standards/quality_assurance/audit-quality/

booklet/acp.pdf

Hong Kong Institute of Directors. 2010, Guidelines of Independent Non-executive

Director, 4th

edition,

http://www.hkiod.com/document/INEDguide/Guide_INED_E_small.pdf

Klein, A. (2002) Audit committee, board of director characteristics, and earnings

management. Journal of Accounting and Economics 33 (3): 375-400.

Krishnan, Jagan and Krishnan, Jayanthi (Oct 1997) Litigation risk and auditor

resignations. The Accounting Review; 72, 4; 539-560

Krishnan, J., & Lee, J. E. (2009). Audit committee financial expertise, litigation risk, and

corporate governance. Auditing, 28(1), 241-261.

http://search.proquest.com/docview/216745193?accountid=1144

Lee, Ho Young, Mande, V., & Ortman, R. (2004). The effect of audit committee and

board of director independence on auditor resignation. Auditing, 23(2), 131-146.

35

Lennox, C. (2002). Opinion shopping, audit firm dismissals, and audit committees.

Rochester, Rochester: doi:10.2139/ssrn.299843

Lin, Z Jun, Xiao, Jason Z and Tang, Qingliang. (2008) the roles, responsibilities and

characteristics of audit committee in China. Accounting, Auditing & Accountability

Journal; 21. 5, 721-751.

Liu, M. (2007). Corporate governance, auditor choice and auditor switch: Evidence from

china. Hong Kong Baptist University (Hong Kong)). ProQuest Dissertations and Theses, ,

n/a. http://search.proquest.com/docview/304898041?accountid=11440

Miller, Gary (1998) Corporate Governance in Hong Kong: Ethical Considerations, Legal

Responsibilities and Board Structure. Allied Academies International Conference

Academy of Accounting and Financial Studies. Proceedings 3. 2. 159-172.

Raghunandan, K., & Rama, D. V. (2007). Determinants of audit committee diligence.

Accounting Horizons, 21(3), 265-279.

http://search.proquest.com/docview/208910600?accountid=11440

Rich, K. T. (2009). Audit committee accounting expertise and changes in financial

reporting quality. University of Oregon). ProQuest Dissertations and Theses, , n/a.

http://search.proquest.com/docview/304978598?accountid=11440

Robinson, D. R., & Owens-Jackson, L. (2009) Audit committee characteristics and

auditor changes. Academy of Accounting and Financial Studies Journal, 13(10963685),

117-132.

Skinner, D. J., & Sloan, R. G. (1999). Earnings surprises, growth expectations, and stock

returns: Dont let an earnings torpedo sink your portfolio.

36

Tanyi, P., Raghunandan, K., & Barua, A. (2010). Audit report lags after voluntary and

involuntary auditor changes. Accounting Horizons, 24(4), 671-688.

http://search.proquest.com/docview/822243649?accountid=11440

Whisenant, S. (2000). Evidence on the auditor and client relationship: What can be

learned from reasons reported by managers for changing auditors?. Rochester, Rochester:

doi:10.2139/ssrn.205094

10. Appendices

Table 1A: Number of Auditor Change Firms from 2004 to 2011

Number of notice about auditor change firm listed at Hong Kong Stock

Exchange website 1234

Less: number of the duplicate notice (with describing the same audited

change incidents) 186

1048

Less: number of firms with no clear reasons 68

Less: number of firms with other reasons than accounting disagreement,

auditor resignation audit fee disputes and receipt of qualified opinion 158

226

Total number of firms with reasons mentioned

822

Table 1B: Selection of Auditor Change Firms from 2004 to 2011

Accounting

Disagreement

Auditor

resignation Fee dispute

Audit Qualified

opinion Total

Auditor change firms 2004-2011 18 293 503 8 822

Less:

Firms with not enough audit committee data 0 5 35 1 41

Firms with not enough financial data 1 15 13 0 29

Total number of firms suitable for research 17 273 455 7 752

Total number of firms randomly chosen

for the sample 6 35 86 3 130

37

Table 1C: Descriptive Deaitls for sample and control (con'd)

Industry

Sector

Sample

firm

Control

firm Total

00 Energy 001 Oil & Gas 5 7 12

002 Coal 5 2 7

05 Materials

051 Gold & Precious Metals 0 0 0

052 Diversified Metals & Minerals 3 3 6

053 Basic Materials 6 2 8

10 Industrial Goods 100 Industrial Goods 12 10 22

20 Consumer Goods

201 Automobiles 1 2 3

202 Household Goods & Electronics 3 8 11

203 Textiles, Clothing & Accessories 7 10 17

204 Food & Beverages 1 6 7

205 Health & Personal Care 16 8 24

206 Agricultural Products 0 0 0

30 Services

301 Retailers 2 3 5

302 Hotels, Casinos & Leisure Facilities 4 6 10

303 Media & Entertainment 4 9 13

304 Transportation 2 2 4

305 Support Services 5 8 13

35 Telecommunications 350 Telecommunications 1 1 2

40 Utilities 400 Utilities 3 3 6

50 Financials

501 Banks 0 0 0

502 Insurance 0 0 0

503 Other Financials 10 9 19

60 Properties & Construction 601 Properties 11 14 25

602 Construction 3 4 7

70 Information Technology

701 IT Hardware 10 4 14

702 Software & Services 9 8 17

703 Semiconductors 0 0 0

80 Conglomerates 800 Conglomerates 3 1 4

Unclassified 4 0 4

Total 130 130 260

38

Table 1C: Descriptive Details for sample and control

Sample firm Control firm Total

Listed in Main Board 98 106 204

Listed in GEM Board 32 24 56

130 130 260

Sample firm Control firm

Asset value (HKD in Million) - Mean 4075 4907

Asset value (HKD in Million) - Median 578 539

Table 2 : Descriptive Statistics Univarite Results

Variable

Auditor change Firm (n=130) Control Firm (n=130)

Mean Median Std.

Deviation

Mean Median Std.

Deviation

Diff in

Means t-test

INDP 0.949 1.000 0.113 0.939 1.000 0.139 0.010 -0.610

TECH 0.351 0.333 0.167 0.336 0.333 0.178 0.014 -0.673

MEET 3.077 3.000 1.350 2.808 2.000 1.020 0.269 -1.813*

GEXP 1.943 1.333 1.962 2.101 1.333 2.224 -0.158 0.609

KNOW 3.416 3.000 2.311 4.010 3.333 2.910 -0.594 1.823*

GROWTH 0.572 0.151 1.512 0.229 0.082 0.718 0.343 -2.335**

SIZE 8.740 8.762 0.800 8.738 8.732 0.836 0.002 -0.018

AUDIT 0.346 0.000 0.478 0.217 1.000 0.482 0.129 4.910***

FEE 0.291 -0.001 0.903 0.217 0.071 0.804 0.074 -0.694

* = p-value <0.10 , ** = p-value <0.05, *** = p-value <0.01

variable definition

INDP = the independent audit committee members on the audit committee proportion

TECH = the audit committee members on the audit committee with financial expertise (CPA / CFO Experience)

MEET = the number of audit committee meetings held during the firm's reporting year

GEXP = the average number of boards of directors which the audit committee members have served on

KNOW = the average number of years audit committee members have served with the current particular firm

GROWTH = the average percentage change in total assets for two years ending before the occurrence of auditor change

SIZE = The natural logarithm of the book value of total assets at the year of auditor change

AUDIT = The external auditors' firm size where an audit done by BIG4 =1 and other small audit firm =0

FEE = The change of audit fee at the year of auditor change with the previous year

39

Table 3: Correlation Matrix: Independent Variables

INDP TECH MEET GEXP KNOW GROWTH SIZE AUDIT FEE

INDP 1.000 -.023 -.014 -.049 .069 -.039 .151 .059 -.003

TECH 1.000 -.023 -.152 -.045 -.075 .101 -.008 -.019

MEET 1.000 -.014 .012 .074 .153 .060 .133

GEXP 1.000 -.065 -.040 -.129 -.036 .027

KNOW 1.000 .067 -.165 -.038 .145

GROWTH 1.000 -.160 .054 -.149

SIZE 1.000 -.299 .049

AUDIT 1.000 -.196

FEE 1.000

Table 4: Logistics Regression Results Auditor Change Model

Variable Predicted Sign Estimated Coefficients Wald Statistic

INTERCEPT None 2.571 1.502

INDP - -0.029 0.001

TECH - -0.441 0.313

MEET - -0.180 2.383

GEXP + 0.033 0.253

KNOW + 0.061 1.276

GROWTH - -0.243* 2.705*

SIZE - -0.304* 2.674*

AUDIT + 1.285*** 19.575***

FEE - -0.169 0.999

Number of Observation 260.000

Goodness of fit 0.292

Concordant Pairs 65%

*** Statistically significant at less than the .01 level

** Statistically significant at less than the .05 level

* Statistically significant at less than the .10 level

40

Table 5: Logistic regression results with Modified Audit Variable^

Deloitte Touche

Tohmatsu Ernst & Young KPMG PricewaterhouseCoopers

variables

Estimated

Coefficient

Wald

Statistics

Estimated

Coefficient

Wald

Statistics

Estimated

Coefficient

Wald

Statistics

Estimated

Coefficient

Wald

Statistics

INDP -.272 .068 -.485 .216 .033 .001 -.255 .058

TECH -.581 .572 -.369 .235 -.283 .139 -.383 .253

MEET -.209 3.445* -.200 3.162* -.249 4.746** -.251 4.852**

GEXP .046 .525 .048 .571 .036 .325 .037 .339

KNOW .074 2.081 .075 2.065 .056 1.060 .076 2.177

GROWTH -.300 4.224* -.284 3.769* -.285 3.767* -.285 3.690*

SIZE -.086 .257 -.128 .553 -.132 .591 -.163 .862

AUDITM^ .339 .956 .655 2.460 1.696 4.312** .764 3.556*

FEE -.057 .124 -.042 .071 -.061 .144 -.044 .077

Constant 1.516 .578 1.952 .951 1.720 .736 2.194 1.158

*** Statistically significant at less than the .01 level

** Statistically significant at less than the .05 level

* Statistically significant at less than the .10 level

^In the logistic Regression, AUDITM = 1 if the external auditor is specific Big 4 audit firm and otherwise is 0

41

42

43