yatim, kent, and clarkson, 2006, governance structures, ethnicity, and audit fees of malaysian...
TRANSCRIPT
Governance structures, ethnicity,and audit fees of Malaysian
listed firmsPuan Yatim
School of Business Management, Universiti Kebangsaan Malaysia, Bangi,Bangi, Malaysia
Pamela KentFaculty of Business, Bond University, Robina, Australia, and
Peter ClarksonUQ Business School, The University of Queensland, St Lucia, Australia
Abstract
Purpose – The purpose of this study is to examine the association between external audit fees, andboard and audit committee characteristics of 736 Malaysian listed firms. It is hypothesised that goodcorporate governance practices reduce auditors’ risk assessments, resulting in lower audit fees.Drawing on the existence of a clearly identifiable ethnic domination of board membership andownership of Malaysian listed firms, the study also posits that Bumiputera-controlled firms pay higheraudit fees because of their weaker governance practices.
Design/methodology/approach – This study employs a cross-sectional analysis of 736 firms listedon the Bursa Malaysia for the financial year ending in 2003. Multiple regression analysis is used toestimate the relationships proposed in the hypotheses.
Findings – Overall, the results of this study reveal that external audit fees are positively andsignificantly related to board independence, audit committee expertise, and the frequency of auditcommittee meetings. The study also finds a strong negative association between external audit feesand Bumiputera-owned firms. An additional analysis into the internal governance structures of firmsin the sample show that Bumiputera firms practice more favourable corporate governance practicescompared to their non-Bumiputera counterparts.
Originality/value – This study is a unique contribution in that it provides data on corporategovernance practices in Malaysia for a large sample in the period after the corporate governance reformstaken by Malaysian capital market regulators and participants. Previous studies have shown thatBumiputera-controlled firms pay higher audit fees than non-Bumiputera-controlled firms. These studieshave not tested theoretical explanations for this fee differential. A theoretical explanation provided in thecurrent study is that Bumiputera-controlled firms pay higher audit fees than non-Bumiputera-controlledfirms partially because of differences in corporate governance practices. The study finds conflictingresults with previous research suggesting that corporate governance practices have changed inMalaysia since the amendments of Bursa Malaysia Listing Requirements, 2001.
Keywords Corporate governance, Boards of directors, Audit committees, Auditor’s fees, Malaysia
Paper type Research paper
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/0268-6902.htm
The authors are grateful for useful comments of Allen Craswell, Marion Hutchinson, Jim Psaros,Carolyn Windsor, participants at seminars at Griffith University, UQ Business School 2005Research Student’s Colloquium, Emerging Scholars’ Colloquium of the Fourth Asia PacificInterdisciplinary Research in Accounting (APIRA) Conference, Singapore, 2004, and at the 2ndAACF International Accounting Conference, Turkey, 2005. Funding from UQ Business School,the University of Queensland is also gratefully acknowledged.
Governancestructures,
ethnicity
757
Managerial Auditing JournalVol. 21 No. 7, 2006
pp. 757-782q Emerald Group Publishing Limited
0268-6902DOI 10.1108/02686900610680530
IntroductionPrior studies predominantly using data from the USA, the UK, and Australia haveinvestigated the association between governance mechanisms and the financialreporting process. For instance, using US-based data, Beasley (1996), Dechowet al.(1996) and Jiambalvo (1996) document a negative relationship between effectivegovernance mechanisms and financial reporting decisions that are in violation ofgenerally accepted accounting principles (GAAP). Recently, Carcello et al.(2002) andAbbott et al.(2003) show that firms with stronger internal governance structuresdemand higher audit quality and thus pay higher audit fees. Using the UK data,Peasnell et al.(2000) confirm the findings previously reported by Dechow et al.(1996)that earnings management is negatively related to the independence of the board ofdirectors. Similar to findings reported by Peasnell et al.(2000), Davidson et al. (2005)also find that greater board independence lowers the likelihood of earningsmanagement in Australian firms.
Our study extends this line of research and examines the relation between externalaudit fees and internal corporate governance structures in emerging markets such asMalaysia. The examination of the association between internal governance structuresand audit fees in Malaysia is motivated by two factors. First, Malaysia is an appealingcase to study because it is likely that corporate governance practices used byMalaysian listed firms are different from those practiced in developed markets. Whilethe empirical results provide evidence from a strong and sophisticated capital marketenvironment (Carcello et al., 2002 (US); Goodwin-Stewart and Kent, n.d. (Australia)),very little research has been conducted in countries where capital markets are lessdeveloped and where corporate governance mechanisms are still evolving.
Further, institutional differences exist between developing capital markets suchas Malaysia and other developed markets. These institutional differences include aweak market for corporate control (Zhuang, 1999; Lins, 2003; Gibson, 2003), moreconcentrated stock ownership (Shleifer and Vishny, 1997; Claessens et al., 2000), andsignificant government ownership in listed firms (Shleifer and Vishny, 1997;Claessens et al., 2000; Lemmon and Lins, 2001; Mak and Li, 2001). At the corporatelevel, these institutional differences may influence how the managers and the boardof directors govern their firms. Although the audit fee literature is quite developed,no prior studies have examined the link between internal governance structures andexternal audit fees in emerging markets such as Malaysia. Therefore, this studyaddresses the scarcity of research in this environment and its influence on externalaudit fees in Malaysia.
Second, the Malaysian corporate environment offers clearly identifiable capitalsegments divided along ethnic lines. This division can clearly be observed in the listedfirms whose board membership and share ownership are predominantly controlled bytwo main ethnic groups, namely the Bumiputera Malays and the Chinese[1]. Althoughthere are other ethnic groups such as Indians, the distinction between the two mainethnic groups (i.e. Chinese and Bumiputeras) dominates much of the socio-economicactivities and political policy making decisions. Given this unique corporateenvironment, this study extends the audit pricing and governance literature byexamining proxies such as political and ethnic favouritisms (Gomez and Jomo, 1997) inthe audit fee models and how this dimension of inherent risk influences external auditfees in Malaysia.
MAJ21,7
758
Further, Eichenseher (1995) and CheAhmad and Houghton (2001) also document asignificant relation between ethnicity and audit fees in the Malaysian market.Together, these two studies show that Bumiputera-controlled firms pay higher auditfees compared to non-Bumiputera-controlled firms. They, however, do not providetheoretical explanations for the variations in audit fees. This current study isconducted after the incorporation of the Malaysian Code on Corporate Governance intothe Listing Requirements of Bursa Malaysia 2001. Since, then, the listed companiesincluding Bumiputera-controlled companies have improved their corporate governanceenvironment (KLSE-PricewaterhouseCoopers, 2002). This study, therefore, offersalternative explanations for the fee variations from a corporate governance perspectivegiven that there is a documented empirical association between ethnicity and externalaudit fees in previous studies. The study tests whether Bumiputera-controlled firmshave weaker corporate governance practices than non-Bumiputera-controlled firms.
The remainder of this paper is structured as follows. The next section brieflyexplores corporate governance in Malaysia. This is followed by a discussion on thebackground of the study and development of our hypotheses. The third sectiondescribes the research design. The results of our study are reported in the fourthsection while in the final section conclusions are drawn and the implications of thestudy are discussed.
Malaysian corporate governance environmentThe revised Listing Requirements of Bursa Malaysia in 2001 provides greater obligationfor public listed companies to enhance Malaysia’s corporate governance regime[2].Specifically, the amended Listing Requirements of 2001 outline the requirements forfinancial reporting disclosure on corporate governance matters and continuing listingobligations. Other than the audit committee, which has been mandatory since 1993, theMalaysian Code on Corporate Governance also recommends that the board of directorsappoint remuneration and nomination committees. The establishment of othercommittees such as a risk management committee and a corporate governancecommittee are also recommended but are less frequently set up by listed firms.
The Malaysian Code on Corporate Governance strongly recommends the separationof responsibilities between the board chair and the CEO even though the BursaMalaysia Listing Requirements (2001) does not require the segregation of thesepositions. The Malaysian Code on Corporate Governance also states as principle thatthe board of directors should maintain a sound system of internal control. This led tothe issuance by the exchange of A Guide on Statement of Internal Control in May 2000.This guideline explains the key areas that directors must pay attention before theypresent A Statement of Internal Control in their companies’ annual reports. A listedfirm is required to address in their annual reports the Principle and Best Practices inthe Malaysian Code on Corporate Governance relating to internal control such asidentifying principal risks and ensuring implementation of appropriate systems tomanage risks.
In addition, directors appointed to the board of directors of a public listed companyare required under the Listing Requirements to attend a director’s training programknown as the mandatory accreditation program. The scope of this program coverstopics such as Companies Act 1965, the listing requirements, risk management andinternal control and relevant securities laws. Self-regulatory initiatives also continue to
Governancestructures,
ethnicity
759
be developed by various industry and professional bodies aiming at promotingknowledge and awareness of corporate governance best practices in Malaysia.
Although the Malaysian Code on Corporate Governance and the Bursa MalaysiaListing Requirements (2001) provide the recommendations and rules to restoreinvestor confidence and to improve standards of corporate accountability, theefficacy of the recommended and mandatory governance practices in influencingexternal audit fees has not been empirically tested. Therefore, our study exploreswhether the various best practices and recommendations have influenced externalaudit fees in Malaysia.
Theoretical background and hypothesesThe economic framework underlying this study is that audit fees reflect the economiccosts of efficient auditors (Simunic and Stein, 1996). These costs vary considerablywith variables such as the size, complexity, riskiness, and specific characteristics of theaudited client including various aspects of its internal governance structures (Simunicand Stein, 1996; Carcello et al., 2002; Cohen et al., 2002; Goodwin-Stewart and Kent,n.d.). This study posits that good corporate governance practices are likely to reduceboth inherent and control risks, leading to lower external audit fees. Specifically, fromthe auditor’s perspective, a stronger control and governance environment is likely toreduce the auditor’s assessment of control risk and the extent of audit procedures, thusreducing audit fees. Following this line of reasoning, the study tests hypotheses relatedto several governance variables.
The study characterises internal corporate governance structures as strong when afirm adopts best governance practices as prescribed by the existing empirical literatureand various recommendations and rules for governance reforms (Fama and Jensen,1983a,b; The Cadbury Committee, 1992; Malaysian Code on Corporate Governance,2001; New York Stock Exchange, 2002; NASDAQ Corporate Governance Proposals,2003). The strong governance practices considered in this study include greater boardindependence (measured by both a higher representation of independent non-executivedirectors on boards and the separation of the board chair and the CEO positions),smaller boards, diligent boards, and boards with a risk management committee.
In addition, the study also addresses audit committee characteristics identified by theBlue Ribbon Committee (BRC) on improving the effectiveness of Corporate AuditCommittee (Blue Ribbon Committee, 1999). These characteristics include audit committeeindependence, audit committee expertise, audit committee size, and audit committeediligence. The Blue Ribbon Committee (1999) contends that a more independent, diligent,expert, and larger audit committee is better able to objectively evaluate management’saccounting and reporting practices. These attributes of an audit committee are likely tocontribute to an improved governance environment within a firm. Focusing on arisk-based approach of audit services, these audit committee characteristics are likely toreduce a firm’s inherent and control risks associated with the firm’s financial statementsand disclosures (Beasley, 1996; Dechow et al., 1996; Anderson et al., 2004). Finally, thestudy also considers elements of ethnicity that have been found to influence audit fees ininstitutional settings such as Malaysia (Eichenseher, 1995; CheAhmad and Houghton,2001). This current study offers explanations for fee variations from a corporategovernance perspective in which it argues that governance practices in ethnic favouredfirms are likely to result in differential external audit fees.
MAJ21,7
760
In the following section, the study develops two main testable hypotheses on therelation between external audit fees and both board and audit committeecharacteristics. The third hypothesis tests the association between external auditfees and governance practices associated with ethnic domination of board membershipand share ownership.
Board characteristics, the financial reporting process, and external audit feesThe board of directors has a primary responsibility of overseeing the firm’s financialreporting process. They also must assess the quality of corporate governance withinthe organisation and ensure that the organisation has, for example, effectiveaccounting practices, internal control and risk management, and audit functions.Therefore, Boards of Directors are potentially viewed, in particular their structures, asimportant elements in delivering credible and relevant financial statements.
Board independence. Agency literature suggests that board independence frommanagement provides, among other things, the most effective monitoring andcontrolling of firm activities in reducing opportunistic managerial behaviours andexpropriation of firm resources (Fama and Jensen, 1983a,b; Brickley et al., 1994). Boardindependence is also likely to provide superior oversight of the financial reportingprocess, hence greater reliability and validity in accounting reports is expected(Beasley, 1996; Dechow et al., 1996). As a result, this reduces the auditor’s riskassessments and less audit efforts are required, leading to lower audit fees charged.
The literature and governance guidelines also show that a board’s ability to performits governance role is likely to weaken when the CEO is also the board chair (Fama andJensen, 1983a, b; Dechow et al., 1996; The Cadbury Committee, 1992; the MalaysianCode on Corporate Governance, 2001). For instance, Dechow et al.(1996) show thatfirms identified as earnings manipulators are more likely to have a CEO who alsoserves as a board chair. This suggests that the combination of these roles is likely tocompromise the board effectiveness in monitoring management and the financialaccounting process. As a result, the auditor needs to devote more audit efforts, hencecharging the firm higher fees.
Board size. Board size may play an important role in directors’ ability to monitorand control managers (Jensen, 1993; Lipton and Lorsch, 1992; The BusinessRoundtables, 1997). Lipton and Lorsch (1992) and Jensen (1993) argue that because ofdifficulties in organising and coordinating large groups of directors, board size isnegatively related to the board’s ability to advise and engage in long-term strategicplans. Beasley (1996) also finds that the size of the board of directors significantlyaffects the likelihood of financial statement frauds. His results indicate that as boardsize increases, the likelihood of financial statement fraud also increases. As such, boardsize is likely to affect the financial reporting process, hence audit process. If largerboards are less effective monitors of the financial reporting process (Beasley, 1996),then the firm’s external auditor assesses the control environment as weak, hence moreaudit hours are required, resulting in higher external audit fees.
Board diligence. The intensity of board activities is likely to contribute to theeffectiveness of its oversight functions particularly in matters concerning the financialreporting process. Lipton and Lorsch (1992) and Byrne (1996) argue that boards thatmeet frequently are more likely to perform their duties diligently and are beneficial toshareholders. Similarly, Conger et al.(1998) and Vafeas (1999) argue that board meeting
Governancestructures,
ethnicity
761
time can improve the effectiveness of a board. Therefore, a board that demonstrates agreater diligence in discharging its oversight responsibilities is likely to enhance level ofoversight of the financial reporting process. It is, therefore, expected that more diligentboards (measured by the number of board meetings held during the financial year) arelikely to be negatively associated with external audit fees.
The existence of a risk management mechanism. Consistent with a risk-basedapproach, a firm that establishes a risk management committee as one of the boardcommittees demonstrates a greater awareness of the importance of risk managementand control (The Committee of Sponsoring Organisations of the TreadwayCommission, 1992, 2004; Hermanson, 2003; Selim and McNamee, 1999). Asmonitoring by the boards of directors is heightened through a stringent riskmanagement procedure, one can argue that the financial reporting quality of the firm isgreatly enhanced. Hence, in the context of good corporate governance attributes, thisstudy examines association between external audit fees and the existence of a riskmanagement committee. The study posits that firms with better internal control, interms of risk management, are likely to reduce external monitoring from auditors,thereby lower audit fees are expected.
In sum, the preceding discussion argues that more independent and diligent boards,smaller boards, and boards with a risk management committee are likely to enhanceinternal governance and the control environment and reduce auditors’ riskassessments associated with the financial reporting process. As a result, the extentof audit procedures is also likely to be reduced, leading to lower external audit fees.The following hypothesis is, therefore, tested:
H1. Lower external audit fees are associated with boards of directors that aremore independent, have a smaller number of members, establish a riskmanagement committee, and meet more frequently.
Audit committee characteristics, the financial reporting process, and external audit feesThe Blue Ribbon Committee’s (1999) recommendations that address audit committeeindependence, financial literacy and expertise, as well as audit committee size andauthority are expected to result in a more effective audit committee oversight of thefinancial reporting process. Therefore, in addition to board characteristics, this studyalso examines whether firms having audit committee structures consistent with theBlue Ribbon Committee’s (1999) recommendations are likely to be associated withhigher quality of financial reporting, resulting in lower auditors’ assessment of controlrisks, hence lower external audit fees are expected.
Audit committee independence. Consistent with the risk-based approach, anindependent audit committee is likely to result in a more effective audit committeeoversight of the financial reporting process, thus reducing the incidence of financialreporting problems (Blue Ribbon Committee, 1999; Abbott et al., 2004; Dechow et al.,1996; McMullen, 1996)[3]. An independent audit committee (relative toinsider-dominated audit committees) is better able to protect the reliability of theaccounting process and promote objectivity on the part of the audit committee. Thishelps strengthen internal controls and should lead to a reduction in the levels of bothinherent and control risk. As a result, the firm requires less substantive testing, andhence lower external audit fees are expected.
MAJ21,7
762
Audit committee expertise and size. The effectiveness of an audit committee is furtherenhanced if members of audit committees possess accounting and financial expertise. Inits third recommendation, Blue Ribbon Committee (1999) recommends a minimum ofthree audit committee directors, each of whom is financially literate and at least one ofwhom has accounting or related accounting and financial management expertise.Further, in addition to a requirement that a majority of audit committee members areindependent directors, Bursa Malaysia also mandates that audit committees of its listedfirms be composed of at least three members and at least one audit committee membermust be a member of the Malaysian Institute of Accountants (MIA).
Empirical findings also support the assertion that an audit committee should atleast consist of one member with accounting and financial expertise. Audit committeeexpertise allows for a better understanding of auditing issues and risks, and the auditprocedures proposed to address and detect these issues and risks (DeZoort andSalterio, 2001). Knapp (1987) and Cohen et al.(2002) also find that auditors are lesslikely to refer a complex auditing issue to an audit committee that is perceived as notbeing knowledgeable about the technical auditing and financial reporting mattersinvolved. Collectively, empirical evidence and governance best practices suggest thataudit committee expertise reduces auditors’ risk assessments associated with thefinancial reporting process, resulting in lower external audit fees.
The audit committee size recommendation is consistent with the desire to increasethe organisational status of the audit committee (Braiotta, 2000). Similarly, Kalbers andFogarty (1993) suggest that larger audit committees are legitimised by a meaningfulorganisational support from the board of directors and are thus more likely to beacknowledged as an authoritative body by the external auditors as well as by internalaudit functions of firms. Therefore, consistent with the recommendation by the BlueRibbon Committee (1999) with respect to audit committee size, this study argues thatlarger audit committees are likely to enhance the quality of financial reporting,resulting in lower external audit fees.
Audit committee diligence. A number of studies and governance best practices alsocall for audit committees to be diligent in carrying out their duties (Abbott et al., 2004;Blue Ribbon Committee, 1999; Kalbers and Fogarty, 1993). Generally, studies in thisarea use the most common proxy for audit committee diligence, which is the number ofaudit committee meetings held annually. Prior research also suggests that an auditcommittee that meets frequently can reduce the incidence of financial reportingproblems. By meeting and communicating frequently, for instance, with the externalauditor, the audit committee can alert the auditor on a particular auditing issuerequiring greater attention from the auditor (Raghunandan et al., 1998). Consistent withthe risk-based approach, it is therefore expected that a more diligent audit committee islikely to reduce financial reporting problems, leading to lower external audit fees.
Overall, audit committee structures that are consistent with the Blue RibbonCommittee’s (1999) recommendations help strengthen audit committee effectiveness intheir oversight functions. This is likely to result in less substantive testing, leading tolower external audit fees. The following hypothesis is, therefore, tested:
H2. Lower external audit fees are associated with audit committees that are moreindependent, have more members, have greater accounting and financeexpertise, and meet more frequently.
Governancestructures,
ethnicity
763
Ethnicity and audit feesThe effects of cultural differences and various aspects of culture (i.e. ethnicity anddemography) have been shown to influence business practices, organisations,and accounting disclosure practices and audit services (Hofstede, 1980, 1991; Haniffaand Cooke, 2002; Eichenseher, 1995; CheAhmad and Houghton, 2001). However, priorresearch fails to explore the effect of ethnicity on monitoring differences (i.e.monitoring by internal governance mechanisms) despite recognition of its importance.
Malaysian capital market exhibits a unique corporate environment where itseconomy offers clearly identifiable capital segments divided along ethnic lines(Jesudason, 1989). Early economic development exhibited a clear distinction betweenforeign (mostly British) and local Chinese capital formation. After independence in1957 from the British, the government initiated the new economy policy, whichgradually adds another distinct component of ethnic groups, namely the Bumiputera orMalay shareholders, to the Malaysian capital market.
The presence of clearly identifiable ethnic domination of board membership andownership of Malaysian listed companies is likely to provide evidence of monitoringdifferences that may exist in these firms. These differences are also likely to influenceauditors’ risk assessments, which in turn may affect variations in audit fees. Johnsonand Mitton (2003) and Gomez and Jomo (1997) argue that Bumiputera-controlled firms(i.e. ethnically-favoured firms) and politically connected firms are perceived to havepoor corporate governance practices and greater agency problems. In addition, Gul(2003) also documents evidence of a positive association between audit fees and agencycosts of political affiliations using a sample of Malaysian listed firms. He argues thatauditors perceive politically affiliated firms as having greater audit risks, thuscharging them higher fees. Both Eichenseher (1995) and CheAhmad and Houghton(2001) suggest that Chinese business practices may influence differences in levels ofagency conflicts and risks associated with Chinese-controlled companies (i.e.non-Bumiputera companies), leading to lower external audit fees charged to thesefirms. Collectively, these studies indicate that an element of ethnicity or ethnicfavouritism can be considered as another dimension of inherent risk.
Our study extends the work of Eichenseher (1995) and CheAhmad and Houghton(2001) by seeking an explanation for audit fee variations from a corporate governanceperspective. Gomez and Jomo (1997) and Johnson and Mitton (2003) suggest thatBumiputera-controlled companies are likely to have a poor governance environmentand are given preferential treatments by the ruling government. This study providesrecent evidence on whether Bumiputera-controlled firms pay higher audit fees thannon-Bumiputera-controlled firms. Further analyses are also sought for an associationbetween Bumiputera-controlled firms and these internal governance structures.
If Bumiputera-controlled firms (by virtue of government ownership, and boardmembership and share ownership controlled by Bumiputeras) are associated with poorgovernance practices, then they are likely to have higher levels of inherent and controlrisks[4]. The poor governance environment is likely to adversely affect the financialreporting process, hence reducing the quality and reliability of financial reporting. Asmonitoring from internal governance mechanisms is weak and ineffective, auditors areexpected to do more substantive testing, hence higher audit fees are charged for thesefirms. The following hypothesis is, therefore, tested:
MAJ21,7
764
H3. Bumiputera-controlled firms pay higher external audit fees associated withpoorer corporate governance practices.
Research designData collection and sample selectionThe sample comprises the Bursa Malaysia non-financial public listed companieswhose annual reports are available in 2003. Finance-related firms are excluded due totheir unique characteristics and different compliance and regulatory environment. Thefirms in the sample are either listed on the Main Board or the Second Board of theBursa Malaysia[5]. Both the financial and corporate data of these firms are obtainedfrom their annual reports. These annual reports are available and downloadable fromthe web site of the exchange (http://announcements.bursamalaysia.com). Listed firmsare required to disclose both non-audit fees and audit fees (statutory audit fees) in theirannual reports.
The number of firms listed on the Bursa Malaysia at the end of 2003 is 874 firms(excluding firms listed on Malaysia Exchange of Securities Dealing and AutomatedQuotations Berhad (MESDAQ)). About 32 firms do not have their annual reportsavailable, while 61 firms are involved in finance-related activities. The study alsoexcludes 45 firms that are newly listed[6]. The final sample consists of 736 firms. Atotal of 496 firms are listed on the Main Board of the Bursa Malaysia and the remainingfirms are listed on the Second Board. Just over three quarters of the sample representthree major sectors classified by the Bursa Malaysia namely industrial products (237firms), consumer products (119 firms), and trade and services (163 firms). Theremaining firms in the sample are in sectors including property (86 firms),constructions (57 firms), technology (19 firms), plantations (40 firms), infrastructurecompanies (6 firms), hotel (6 firms), and mining (3 firms).
Variable measurements and model specificationConsistent with existing literature on determinants of audit fees, this study employsthe traditional audit fee model introduced by Simunic (1980) and modified by Craswellet al.(1995). In addition to using variables that are proxies for internal corporategovernance structures (i.e. board and audit committee characteristics), the study alsoincludes control variables identified by prior research that explain the cross-sectionalvariations in audit fees (Simunic and Stein, 1987; Francis and Simon, 1987; Craswellet al., 1995). These variables include client size, number of subsidiaries, audit quality,leverage, profitability, receivable and inventory intensity, and industry types. Thefollowing regression equation is used as the primary model to test the hypothesespreviously discussed. It is also important to note that this is not the only model used inthe study. There are several other alternative models employed to test the validity ofthe primary model.
LAF ¼ b0 þ b1SIZE þ b2SUBS þ b3LEV þ b4REC þ b5INV þ b6ROA þ b7BIG4
þ b8INDUSTRYDUMMIES þ b9BoardIndependence þ b10Dual þ b11BoardSize
þ b12RMC þ b13BoardMeet þ b14ACIndependence þ b15ACSize þ16 ACExpertise
þ b17ACMeet þ b18BumiOwned þ e
Governancestructures,
ethnicity
765
Where LAF, natural log of audit fees; SIZE, natural log of total assets; LEV, total bookvalue of debt to total assets; SUBS, square root of the number of direct subsidiaries; REC,the ratio of receivables to total assets; INV, the ratio of inventories to total assets; ROA,returns on assets (the ratio of earnings before interests and taxes to total assets); BIG4, adummy variable of 1 if financial statements audited by Big 4 audit firms, 0 otherwise;INDUSTRY DUMMIES, a dummy variable of 1 if a firm is in industrial product sector, orin consumer product sector, or in construction sector, or in technology sector, or in tradeand services sector, or in property sector, or in plantations sector, or in hotel sector, or ininfrastructure companies sector, or in mining sector, 0 if otherwise; BoardIndependence,the proportion of independent non-executive directors on boards; Dual , a dummyvariable of 1 if the role of the Board Chair and the CEO is separated, 0 if otherwise;BoardSize, total number of directors on boards; RMC, a dummy variable of 1 if a firmestablishes a risk management committee, 0 if otherwise; BoardMeet, the number ofboard meetings held during the financial year; ACIndependence, the proportion ofnon-executive directors on audit committees; ACSize, total number of directors on auditcommittees; ACExpertise, the proportion of audit committee members with accountingand finance qualifications; ACMeet, the number of audit committee meetings heldduring the financial year; BumiOwned, a dummy variable of 1 if a firm’s sharesoutstanding are substantially held by Bumiputera shareholders[7].
Dependent variable. The Companies Act 1965 of Malaysia requires companies todisclose their non-audit and audit fees in their notes to accounts. The figure was,therefore, collected from company annual reports. The fees are then transformed intonatural logarithm and are used as the dependent variable. The natural log is used tocontrol for the skewed nature of audit fees.
Hypothesised variables. Recall that experimental variables tested in the hypothesesare board and audit committee characteristics, and board ethnicity. Consistent withprior studies, this study tests the assertion that boards of directors that are moreindependent, have a smaller number of members, establish a risk managementcommittee, and meet more frequently pay lower external audit fees (H1). Twomeasures of board independence are used in this study namely the proportion ofindependent non-executive directors on boards and whether different individuals holdthe positions of the board chair and the CEO. An independent non-executive director isa director who is not employed to run the firm’s day-to-day business activities, andwhose role is to provide an outsider’s contribution and oversight to the board ofdirectors (Hanrahan et al., 2001).
The study also posits in H2 that firms with more independent, diligent, expert, andlarger audit committees are likely to pay lower external audit fees. Audit committeeindependence and expertise are measured by the proportion of non-executive directorson the committee and the proportion of audit committee members with accounting andfinance qualifications,, respectively, (Goodwin-Stewart and Kent, n.d.). The proxiesused to measure audit committee diligence and size are the frequency of auditcommittee meetings held during the financial year and the number of directors on theaudit committee, respectively. Finally, the study also predicts thatBumiputera-controlled firms pay higher levels of external audit fees associated withtheir poor internal governance structures (H3). The ethnicity variable is measuredusing a dichotomous variable set as 1 if a firm’s outstanding shares are substantiallyheld by Bumiputera shareholders, 0 if otherwise[8].
MAJ21,7
766
Control variables. We expect audit fees to be positively related to the natural log oftotal assets, leverage, receivable and inventory intensity, and square root of thenumber of subsidiaries. The effects of these variables are controlled in our analysis.For instance, profitability is controlled by including return on assets (ROA) and it isexpected to be negatively associated with audit fees. The study also includes a dummyvariable for Big Four accounting firms to control for differences in audit quality.Further, there are likely to be variations in audit fees across companies in differentindustries (Low et al., 1990). Therefore, in order to control for industry variations, tenindustry dummy variables according to sectors classified by the Bursa Malaysia areincluded in the equation. The sectors are industrial products, consumer products,technology, construction, trade and services, properties, plantations, hotel, mining, andinfrastructure companies.
ResultsDescriptive statisticsTables I and II provide descriptive statistics for the variables employed in the model.Panel A reports those for continuous variables and panel B presents those fordichotomous variables. In general, the distributions are similar to previous studiesundertaken in the Malaysian market (Eichenseher, 1995; CheAhmad and Houghton,2001). For instance, the average audit fees of RM191,975 in this sample is slightlyhigher than those documented by Eichenseher (1995) and CheAhmad and Houghton(2001) where the average audit fees in their samples are RM113,000 and RM137,780,respectively. The mean asset size of RM1,167,849,518 is lower than that reported byCheAhmad and Houghton (2001) but higher than that found by Eichenseher (1995)which are RM1,290,316,000 and RM615,000,000, respectively. There are 506 firms(68.8 percent) that use Big Four external auditors.
Of particular interest to the study are the corporate governance variables. TheListing Requirements of Bursa Malaysia 2001 require all public listed firms to haveaudit committees comprising at least three members, the majority of whom must beindependent. The size of audit committees ranges from two to seven, with a mean of3.49[9]. On average, the firms in the sample have their audit committees comprised ofat least three non-executive members (about 79 percent of audit committee membersare non-executive directors). Further, some 252 (one-third) firms in the sample haveformally established a risk management committee. The majority of firms in thesample also separate the positions of the board chair and the CEO (about 84 percent ofthe firms practice dual leadership).
The average number of non-executive directors on Malaysian boards is 4.74,ranging from a minimum of 1 to a maximum of 12 non-executive board members(63 percent of overall board members are non-executives). The number of directors onMalaysian boards is between 3 and 16 with an average board size of 7.51. There are 205firms whose boards of directors are predominantly Bumiputeras. Finally, there aresome 135 firms whose outstanding shares are substantially held by Bumiputerashareholders.
Table III reports the correlations between the variables used in the regressions. Thecorrelation matrix reveals that few variables are highly inter-correlated (above 0.5).Variables with high significant correlations include natural log of total assets andnatural log of audit fees (0.741), the ratio of current assets to total assets and the ratio of
Governancestructures,
ethnicity
767
Var
iab
leM
inim
um
Max
imu
mM
ean
Std
.d
evia
tion
Med
ian
Au
dit
fee
(RM
)5,
000
6,90
0,00
019
1,97
542
4,34
487
,919
Tot
alas
sets
(RM
)2,
280,
406
75,0
08,3
78,0
001,
167,
849,
518
4,50
9,98
8,88
425
4,03
8,94
6N
o.of
sub
sid
iari
es0
290
11.3
817
.22
7.00
L/t
erm
deb
tto
T/a
sset
s0.
000.
570.
120.
140.
07C
urr
ent
asse
t/as
sets
0.02
1.00
0.46
0.22
0.46
Rec
eiv
able
s/as
sets
0.00
0.63
0.19
0.14
0.15
Inv
ento
ries
/ass
ets
0.00
0.40
0.09
0.09
0.06
Ret
urn
onas
sets
(RO
A)
23.
093.
680.
020.
250.
04N
o.of
NE
D1
124.
741.
944.
00N
o.of
Ind
.D
ir0
82.
960.
943.
00N
o.of
NE
Don
AC
16
2.76
0.80
3.00
No.
ofb
um
id
irec
tors
012
3.08
2.19
2.50
Boa
rdsi
ze3
167.
511.
997.
00A
ud
itco
mm
itte
esi
ze2
73.
490.
733.
00
Table I.Descriptive statistics(N ¼ 736) (Panel A:continuous variables)
MAJ21,7
768
Variable Yes Percent No Percent
Big four auditors (BIG4) 506 68.8 230 31.3Bumiputera-dominated boards using big fourauditors 141 68.8 64 31.2Non-Bumiputera dominated boards using big fourauditors 365 68.7 166 31.3Dual leadership 616 83.7 120 16.3Risk management committee (RMC) 252 34.2 484 65.8Bumiputera-dominated boards (BUMI_BRD) 205 27.9 531 72.1Firms with substantial bumiputera shareholdings(BUMI_OWN) 135 18.3 601 81.7
Table II.Descriptive statistics
(N ¼ 736) (panel B:dichotomous variables)
AFee TAsset Leverage CA/TA Rec./TA Inv./TA ROAPercentNED Bumi_brd
AFee 1.000 0.741 * * 0.321 * * 20.145 * * 20.118 * * 20.061 0.040 0.053 0.118 * *
TAsset 1.000 0.400 * * 20.271 * * 20.292 * * 20.203 * * 0.139 * * 0.153 * * 0.172 * *
Leverage 1.000 20.360 * * 20.228 * * 20.152 * * 0.025 0.059 0.155 * *
CA/TA 1.000 0.574 * * 0.318 * * 0.063 20.093 * 20.064Rec./TA 1.000 0.176 * * 0.002 20.163 * * 20.053Inv./TA 1.000 0.049 20.177 * * 20.205 * *
ROA 1.000 0.034 20.072percentNED 1.000 0.351 * *
Bumi_brd 1.000BSIZEpercentNED_ACACSIZEpercentBUMIAC meetBoard Meet#SubspercentNon_Bumi
BSIZEPercentNED_AC ACSIZE
PercentBUMI ACMeet BoardMeet #Subs
Percentnon-Bumi
AFee 0.269 * 0.118 * * 0.148 * * 0.148 * * 0.251 * * 0.186 * * 0.698 * * 20.148 * *
TAsset 0.357 * * 0.208 * * 0.176 * * 0.188 * * 0.232 * * 0.186 * * 0.532 * * 20.188 * *
Leverage 0.108 * * 0.072 * 0.053 0.168 * * 0.172 * * 0.095 * 0.244 * * 20.168 * *
CA/TA 20.018 20.107 * * 20.018 20.075 * 20.089 * 20.070 20.153 * * 0.075 *
Rec./TA 20.047 20.177 * * 20.051 20.060 20.069 20.074 * 20.120 * * 0.060Inv./TA 0.059 20.130 * * 20.001 20.218 * * 20.073 * 20.086 * 20.142 * * 0.218 * *
ROA 0.108 * * 0.114 * * 0.047 20.068 0.023 20.045 20.015 0.068percentNED 0.043 0.569 * * 0.110 * * 0.380 * 0.116 * * 0.184 * * 0.009 20.380 * *
Bumi_brd 0.018 0.217 * * 0.076 * 0.857 * * 0.180 * * 0.206 * * * 0.105 * * 20.857 * *
BSIZE 1.000 0.117 * * 0.448 * * 0.011 0.146 * * 0.047 0.164 * * 20.011percentNED_AC 1.000 0.063 0.247 * * 0.092 * 0.150 * * 0.035 20.247 * *
ACSIZE 1.000 0.102 * * 0.077 * 0.040 0.112 * * 20.102 * *
percentBUMI 1.000 0.160 * * 0.201 * * 0.106 * * 21.000 * *
AC meet 1.000 0.503 * * 0.204 * * 20.160 * *
Board Meet 1.000 0.154 * * 20.201 * *
#Subs 1.000 20.106 * *
percentNon_Bumi 1.000
Notes: * * * Correlation is significant at the 0.05, 0.01 level (two-tailed), respectively
Table III.Pearson correlation
matrix of variables usedin the study (N ¼ 736)
Governancestructures,
ethnicity
769
receivables to total assets (0.574), and the proportion of non-executive directors onboards and the proportion of non-executive directors on audit committees (0.569).While a few governance variables are significantly correlated with each other, theircorrelations do not indicate that multicollinearity is a serious problem. As shown inTable III, the correlations range from 0.448 (between board size and audit committeesize) to 0.503 (between the frequency of board meetings and audit committeemeetings)[10].
Multivariate analysisRegression results of the association between external audit fees and control variables(traditional audit fee model). Table IV presents the ordinary least square regressionresults for testing the hypotheses. In testing the validity of models used in the study,the traditional audit fee model introduced by Simunic (1980) is employed whereby thenatural log of audit fees is regressed on the control variables only[11]. Results in model1 show the association between external audit fees and control variables derived fromthe extant literature (Francis and Simon, 1987; Simon and Francis, 1988; Francis andStokes, 1986) as independent variables[12]. The model is significant ( p , 0.01), with anadjusted R 2 of 69.6 percent.
Following prior research (Simunic, 1980; Francis and Simon, 1987; Simon andFrancis, 1988; Craswell et al., 1995), it is expected that audit fees are positivelyassociated with size, receivables and inventory intensity, leverage[13], the number ofsubsidiaries, and whether a firm uses Big Four audit firms. The coefficients for thesecontrol variables are in the predicted direction and significant ( p , 0.01) except for theuse of Big Four audit firms. The strong negative association between audit fees andROA is also consistent with findings previously documented (Simunic, 1980; Craswellet al., 1995; Gul and Tsui, 1998). The industry controls are also included (the regressionestimates for industry variables are not reported in Table IV) in all regressionspresented in Table IV. Overall results show that audit fees are not significantly relatedto all industry variables[14].
Regression results of the association between external audit fees and internalgovernance structures. Model 2, which regresses external audit fees on controlvariables and the test variables of interest, is significant ( p , 0.01) with an adjustedR 2 of 69.8 percent. The results reported in model 2 show that external audit fees arepositively and significantly associated with the proportion of independentnon-executive directors on boards ( p , 0.05), audit committee expertise ( p , 0.05),and audit committee meeting frequency ( p , 0.10). There is also a negative associationbetween external audit fees and Bumiputera-controlled firms ( p , 0.10).
A number of board characteristics and audit committee characteristics are highlycorrelated (i.e. board size and audit committee size, the proportion of independentdirectors on boards and the proportion of independent directors on audit committees,and the frequency of board and audit committee meetings), and potentially lead tomulticollinearity between variables. Therefore, two regressions are run separately totest the association between external audit fees and the board and audit committeecharacteristics. In addition, consistent with Beasley (1996) and Carcello et al.(2002),model 3 in Table IV focuses on the full board of directors characteristics, which alsoinclude the existence of a risk management committee [15]. Model 3, which regressesexternal audit fees on the control variables and on the five variables of interest, is
MAJ21,7
770
Var
iab
les
Ex
p.
sig
nM
odel
1tr
adit
ion
alA
/Fee
mod
elM
odel
2g
over
nan
cev
ars.
and
eth
nic
ity
Mod
el3
boa
rdch
arac
teri
stic
sM
odel
4A
/com
mit
tee
char
acte
rist
ics
Inte
rcep
t?
3.09
9a*
* (7.
368)
b0.
000c
2.60
8*
* (5.
631)
0.00
02.
932
** (
6.65
8)0.
000
2.88
3*
* (6.
703)
0.00
0Controlvariables
Siz
eþ
0.38
4*
* (17
.379
)0.
000
0.38
4*
* (15
.728
)0.
000
0.37
9*
* (15
.824
)0.
000
0.37
7*
* (16
.677
)0.
000
Nu
mb
erof
sub
sid
iari
esþ
0.23
7*
* (14
.517
)0.
000
0.23
1*
* (14
.258
)0.
000
0.23
4*
* (14
.119
)0.
000
0.23
5*
* (14
.419
)0.
000
Lev
erag
eþ
0.00
9*
* (2.
769)
0.00
60.
008
** (
2.92
6)0.
004
0.00
8*
* (2.
759)
0.00
60.
009
** (
2.88
6)0.
004
Rec
eiv
able
s/T
otal
asse
tsþ
0.54
1*
* (3.
251)
0.00
10.
578
** (
3.39
0)0.
001
0.56
0*
* (3.
372)
0.00
10.
554
** (
3.28
8)0.
001
Inv
ento
ries
/Tot
alas
sets
þ1.
078
** (
4.77
1)0.
000
1.08
9*
* (4.
746)
0.00
01.
135
** (
4.97
6)0.
000
1.08
1*
* (4.
738)
0.00
0R
etu
rnon
asse
ts(R
OA
)2
20.
122
* (2
1.74
6)0.
081
20.
134
* (2
2.12
5)0.
034
20.
134
* (2
2.00
3)0.
046
20.
118
* (2
1.79
9)0.
072
Big
fou
rau
dit
ors
þ0.
058
(1.4
36)
0.15
10.
062
(1.5
13)
0.13
10.
057
(1.4
14)
0.15
80.
065
(1.5
82)
0.11
4Board
andauditcommitteecharacteristics
andethnicityvariables
Per
cen
tIn
dep
end
ent
dir
ecto
rs2
0.39
3* (
1.93
3)0.
054
0.39
5* (
2.03
4)0.
042
Du
alle
ader
ship
20.
003
(0.0
49)
0.96
12
0.00
3(2
0.05
1)0.
959
Boa
rdsi
zeþ
0.00
3(0
.196
)0.
845
0.00
5(0
.409
)0.
683
Bu
mi
own
edþ
20.
099
**
*(2
1.66
3)0.
096
Ris
km
gm
tco
mm
itte
e2
0.04
1(0
.917
)0.
359
0.03
9(0
.889
)0.
374
Boa
rdm
eeti
ng
s2
0.00
3(0
.324
)0.
746
0.01
1(1
.065
)0.
287
AC
ind
epen
den
ce2
0.06
9(0
.457
)0.
648
0.06
3(0
.442
)0.
659
AC
size
20.
004
(0.1
58)
0.87
50.
014
(0.5
46)
0.58
5A
Cex
per
tise
20.
169
* (1.
757)
0.07
90.
170
* (1.
762)
0.07
9A
Cm
eeti
ng
s2
0.03
1*
**
(1.6
29)
0.10
30.
035
* (1.
915)
0.05
6A
dju
sted
R2
0.69
60.
698
0.69
60.
698
F-s
tati
stic
s(p
-val
ue)
240.
993
** (
0.00
0)10
1.12
6*
* (0.
000)
141.
441
** (
0.00
0)15
5.33
2*
* (0.
000)
Notes:
aC
oeffi
cien
t;b
fig
ure
sin
par
enth
eses
aret-
stat
isti
cs;
cal
lp-
val
ues
are
two-
tail
ed,
oth
erw
ise
sin
gle
-tai
led
wh
end
irec
tion
sar
ep
red
icte
d,
** p
,0.
01,
* p,
0.05
and
**
* p,
0.10
Table IV.Regression results foraudit fees on control,
governance and ethnicityvariables (N ¼ 736)
Governancestructures,
ethnicity
771
significant ( p , 0.01) with an adjusted R 2 of 69.6 percent. The results in model 3indicate that there is a strong positive association between external audit fees and thepercentage of independent non-executive directors on boards ( p , 0.05). The results inmodel 3 also show that external audit fees are not significantly related to dual leadership,board size, the frequency of board meetings, and the existence of a risk managementcommittee [16]. Recall thatH1 predicts that lower external audit fees are associated withboards that are more independent, have a smaller number of members, establish a riskmanagement committee, and meet more frequently. Using a risk-based perspective ofaudit services, it appears that the results do not support H1 as there is no negativesignificant association between external audit fees and those governance variables.
Hence, the findings do not support the view widely held in the financial reportingand governance literature (Jensen, 1993; Beasley et al., 2000; Dechow et al., 1996;Davidson et al., 2005) that greater board independence, smaller board size, and theexistence of an internal control mechanism (as per the existence of a risk managementcommittee) are likely to reduce risks associated with the financial reporting process.With respect to the significant association between external audit fees and theproportion of independent non-executive directors on boards, the finding is similar tothose of recent studies by Carcello et al.(2002), Abbott et al.(2003), andGoodwin-Stewart and Kent (n.d.). These studies report that board independence ispositively associated with audit fees, suggesting independent boards are likely todemand a higher quality audit from external auditors, resulting in higher audit fees.Based on the propositions that directors wish to protect their reputational capital(Fama, 1980; Fama and Jensen, 1983a, b; Gilson, 1990), to avoid legal liabilities (Gilson,1990), and to protect shareholder wealth, firms with a stronger control environmentmay seek differentially higher quality audit services. The auditors are likely toincorporate the costs of providing these differentiated and higher quality audit servicesinto the audit fees, resulting in higher audit fees.
Recall that a number of board and audit committee characteristics are highlycorrelated and this potentially leads to multicollinearity problems. Model 4 in Table IVaddresses this issue by regressing external audit fees on control variables and on thefour audit committee variables of interest. The model is significant ( p , 0.01) with anadjusted R 2 of 69.8 percent [17].
The results reported in model 4 show that external audit fees are significantlyassociated with audit committee expertise ( p , 0.05) and the frequency of auditcommittee meetings ( p , 0.05). External audit fees, however, are not significantlyrelated to audit committee independence and audit committee size. From a risk-basedperspective of audit services, the results with respect to the association betweenexternal audit fees and audit committee characteristics do not support H2.The resultsare consistent with the demand approach of audit services in that more expert anddiligent audit committees are likely to seek higher quality audits from externalauditors, resulting in higher audit fees (Carcello et al., 2002; Abbott et al., 2003). Oneinterpretation with respect to the positive significant association between externalaudit fees and audit committee expertise is that audit committee members who possessprofessional certifications (i.e. Certified Public Accountants and CharteredAccountants) may seek to protect their reputations as experts in accounting andfinancial matters and are likely to demand increased monitoring (Fama and Jensen,1983a, b; Gilson, 1990).
MAJ21,7
772
External audit fees and Bumiputera-controlled firms. Recall that the study also positsthat firms whose directors are predominantly Bumiputeras are expected to payrelatively higher audit fees associated with poorer corporate governance practices (H3).The results in model 2 in Table IV show that there is a significant negative associationbetween external audit fees and the ethnicity variable measured using a dichotomousvariable of whether a firm’s outstanding shares are substantially held by Bumiputerasor otherwise ( p , 0.10) [18]. This indicates that Bumiputera-controlled firms pay loweraudit fees than non-Bumiputera-controlled firms, which is the opposite to the predictionof H3. This result is not consistent with those documented by Eichenseher (1995) andCheAhmad and Houghton (2001) in which they report that Bumiputera-controlled firmspay higher external audit fees than their non-Bumiputera-controlled counterparts (i.e.ethnic Chinese-controlled firms). Two possible explanations for this difference infindings are the changes in the regulatory environment during late 1990s and in 2001and the differences in increased focus on board and audit committee oversightresponsibilities. These are likely to have strengthened the motivation of boards ofdirectors and audit committees to act in a manner inconsistent with our hypotheses.
Additional testing of H3 is undertaken by an investigation into internal governancestructures of both Bumiputera-controlled and non-Bumiputera-controlled firms. Anexamination of internal governance structures is likely to help explain the lower auditfees paid by the Bumiputera-controlled firms as reported in the previous section.
Additional tests of corporate governance structures of firms. The proportion ofBumiputera directors on boards is regressed against the set of control variables andgovernance variables used in the study. An independent sample t-test is also conductedto compare means of internal governance structures for Bumiputera-controlled andnon-Bumiputera-controlled firms.
T-tests for internal governance variables for firms grouped as Bumiputera firms(assigned value of 1) and those non-Bumiputera firms (assigned value of 0) are presented inTable V. With the exception of audit committee expertise, there is a significant (p , 0.05)difference in internal governance structures for Bumiputera-controlled andnon-Bumiputera-controlled firms. The t-test results are confirmed by non-parametricMann-Whitney U tests. x 2 tests are conducted for dichotomous variables (i.e. dualleadership structure and the existence of a risk management committee) and indicate thatthere is a statistically significant (p , 0.01) difference between Bumiputera-controlledand non-Bumiputera-controlled firms in their internal governance structures (dualleadership structure and the existence of a risk management committee).
Table VI presents the results of a regression model using the proportion of Bumiputeradirectors on boards as the dependent variable, and the model is significant ( p , 0.01) withan adjusted R 2 of 18.6 percent. The results in Table VI indicate that boards ofBumiputera-dominated firms are more independent and these firms are more likely toseparate the roles of the board chair and the CEO. This is evidenced by the strong positiveassociation between the proportion of Bumiputera directors on boards and these twointernal governance variables (p , 0.01). Bumiputera-controlled firms also appear tohave a better risk management environment as evidenced by the strong positive relationbetween the proportion of Bumiputera directors on boards and the existence of riskmanagement committees ( p , 0.01). Further, Bumiputera-controlled firms are morediligent as indicated by the strong positive association between the proportion ofBumiputera directors on boards and the frequency of board meetings ( p , 0.05).
Governancestructures,
ethnicity
773
Var
iab
les
Mea
nB
um
ipu
tera
firm
sN
on-b
um
ipu
tera
firm
sIn
dep
end
ent
sam
plet-
test
stat
isti
cs(s
ig.
two-
tail
ed)
Man
n-w
hit
ney
Ute
stst
atis
tics
(sig
.tw
o-ta
iled
)
Pan
elA
:T
-tes
tsan
dM
ann
-wh
itn
eyU
test
for
con
tin
uou
sv
aria
ble
sB
oard
ind
epen
den
ce0.
430.
394.
960
**
(0.0
00)
25.
219
**
(0.0
00)
Boa
rdm
eeti
ng
freq
uen
cy6.
195.
164.
817
**
(0.0
00)
24.
795
**
(0.0
00)
AC
ind
epen
den
ce0.
840.
775.
698
**
(0.0
00)
25.
454
**
(0.0
00)
AC
size
3.58
3.46
2.05
7*
(0.0
40)
22.
398
*(0
.016
)A
Cex
per
tise
0.46
0.47
20.
322
(0.7
48)
20.
667
(0.5
05)
AC
mee
tin
gfr
equ
ency
5.28
4.71
3.80
5*
*(0
.000
)2
4.03
0*
*(0
.000
)P
anel
B:x
2st
atis
tics
for
dic
hot
omou
sv
aria
ble
sV
aria
ble
sV
alu
efo
rp
ears
onx
2A
sym
p.
sig
.(t
wo-
sid
edD
ual
lead
ersh
ip10
.308
0.00
1R
isk
man
agem
ent
com
mit
tee
15.6
240.
000
Notes:
aG
rou
pin
gv
aria
ble
:B
um
i_B
oard
(av
alu
eof
1as
sig
ned
ifa
firm
’sb
oard
ofd
irec
tors
isp
red
omin
antl
yB
um
ipu
tera
dir
ecto
rs,
0if
oth
erw
ise;
Bu
mip
ute
rad
omin
ated
boa
rds,
205
firm
s;n
on-B
um
ipu
tera
dom
inat
edb
oard
s,53
1fi
rms)
;* p
,0.
05;
** p
,0.
01
Table V.T tests, x 2 test forindependence, and Mann-Whitney U test forinternal governancevariables betweenBumiputera- controlledfirms and non-Bumiputera- controlledfirms a
MAJ21,7
774
The results also show that audit committees of Bumiputera-controlled firms are moreindependent as indicated by a strong positive association between the proportion ofBumiputera directors on boards and audit committee independence ( p , 0.01). Theproportion of Bumiputera directors on boards are not significantly related to auditcommittee expertise, audit committee-meeting frequency, and audit committee size.Overall, these results do not support the view that Bumiputera-dominated firms haveweaker governance environments relative to non-Bumiputera-controlled firms aspreviously suggested by Gomez and Jomo (1997) and Johnson and Mitton (2003).
With respect to the association between non-Bumiputera-controlled firms and theirinternal governance structures, these results appear to be consistent with the view thatChinese-controlled firms tend to have more inside directors on their boards of directorsand are likely to have a unitary leadership structure (Weindebaum, 1996; Redding,1995). Chinese businesses do adopt some of the practices of western firms likerecruitment of professional managers and some degree of public ownership. Butessentially, the ownership stakes and control are still in the hands of family holdingcompanies and family members. Therefore, it is likely that the selection of personnelfor key positions such as board chair, CEO, and managing director, and board ofdirectors is another way through which control of the company is maintained [19].
The study also explores the internal governance practices of Bumiputera-dominatedboards using logistic regression analysis with the dependent variable set at 1 for firmswhose board members are predominantly Bumiputeras and 0 if otherwise. Unreportedresults obtained using the logistic regression method confirm those obtained using theordinary least square method indicating that firms whose boards are predominantlyBumiputeras practice improved internal corporate governance practices relative totheir non-Bumiputera counterparts.
The results obtained in this additional analysis are likely to help explain thenegative significant association between external audit fees and Bumiputera-controlledfirms. It appears that Bumiputera-controlled firms practice favourable corporategovernance practices relative to non-Bumiputera firms and thus are likely to contributeto lower auditors’ risk assessments, resulting in lower external audit fees.
ConclusionThe unique institutional environment in Malaysia coupled with the presence of a clearlyidentifiable ethnic domination of corporate boards and ownerships of Malaysian listedfirms potentially provides evidence of monitoring differences that exist in these firms.The study examines the link between external audit fees and internal governancestructures and the element of ethnicity. Overall, we find that external audit fees arepositively and significantly associated with board independence, audit committeeexpertise, and the frequency of audit committee meetings. In contrast to the findingsdocumented by Eichenseher (1995) and CheAhmad and Houghton (2001), we document asignificant negative relation between external audit fees and Bumiputera-owned firms.
An additional analysis also reveals that Bumiputera-controlled firms practiceimproved internal corporate governance practices compared to their non-Bumiputeracounterparts. This finding supports the result obtained earlier whereinBumiputera-controlled firms pay lower external audit fees because their internalgovernance structures are stronger relative to their non-Bumiputera counterparts. Withrespect to the association between external audit fees and governance variables, the
Governancestructures,
ethnicity
775
results do not support the risk-based perspective of audit services (i.e. good corporategovernance practices reduce external audit fees). The results, however, are consistent withthe demand approach of audit services, wherein firms with good corporate governanceattributes demand higher audit quality, resulting in higher external audit fees.
While our study makes an important contribution to the governance debate, there are anumber of limitations that are inherent in this current study. One of the limitations is thatthe results of this study cannot be generalised and should be interpreted in the context ofthe Malaysian corporate environment, particularly where ethnicity is concerned. Further,the financial and corporate data employed in this study and the findings thereafter maynot completely explain the link between governance variables and external audit fees asother variables such as detailed ownership structures are likely to better explain therelationship. In addition to archival data, data gathered using survey methods for instanceare also likely to provide better insights to findings of this study.
Notes
1. Bumiputera means in Malay“sons of the soil”. It refers to Malays and other indigenouspeople as distinct from Chinese, Indians, and other non-indigenous residents.
2. The Kuala Lumpur Stock Exchange (KLSE) became de-mutualised exchange and wasrenamed Bursa Malaysia in April 2004.
3. It is also important to note that recent studies have documented the link between auditcommittee member independence and audit fees from a demand approach of audit services.For instance, Abbott et al.(2003) find that audit committee independence and other auditcommittee characteristics (i.e. expertise and diligence) are positively related to audit fees.They suggest that an independent, expert, and diligent audit committee demands increasedaudit coverage or purchases a higher-quality audit, hence higher audit fees. Independentaudit committee directors are likely to view their service on an audit committee as a means of
Variables Exp. sign Coefficients t-statistics ( p-value)a
Intercept ? 20.625 23.302 * *(0.001)Size ? 0.023 2.283 *(0.023)Leverage ? 0.0001 0.037 (0.970)Receivables/total assets ? 0.096 1.420 (0.156)Inventories/total assets ? 20.425 24.417 * *(0.000)Returns on assets (ROA) ? 20.116 22.857 * *(0.004)Big four ? 20.047 22.343 *(0.019)Number of subsidiaries ? 20.006 21.058 (0.291)Internal governance variables: board and audit committee characteristicsPercent independent directors ? 0.460 4.792 * *(0.000)Dual leadership ? 0.083 3.595 * *(0.000)Risk management committee ? 0.055 2.764 * *(0.006)Number of board meetings ? 0.011 2.447 *(0.015)AC independence ? 0.298 4.489 * *(0.000)AC size ? 0.014 1.209 (0.227)AC expertise ? 0.009 0.187 (0.852)Number of AC meetings ? 0.008 1.041 (0.298)Adjusted R2 0.186F-statistics ( p-value) 12.196 * * (0.000)
Notes: a All p-values are 2-tailed; * * p , 0.01; *p , 0.05
Table VI.Regression results forrelationship between theproportion of Bumiputeradirectors on boards oncontrol variables and onboard and auditcommittee characteristics(N ¼ 736)
MAJ21,7
776
enhancing their reputational capital (Fama and Jensen, 1983a, b). The preservation ofreputational capital serves as one motivation for higher quality reporting.
4. The government-linked companies (GLCs) are included in the Bumiputera ownership categorybecause the government is staffed mainly by Bumiputeras (Malays) and these companies havehiring policies that give strong preferences to Bumiputeras. In addition, government-controlledcompanies rely heavily on Bumiputera suppliers and vendors. Outstanding shares of thesecompanies are substantially held by government-linked investment companies (GLICs). Thesegovernment-linked investment companies include Khazanah Nasional Berhad (the investmentarm of the Ministry of Finance), Permodalan Nasional Berhad (manages various national unittrusts), employees provident funds (EPF), pilgrimage board funds (Lembaga Tabung Haji), andthe military pension funds (Lembaga Tabung Angkatan Tentera).
5. The Main Board companies have a minimum paid-up capital of Ringgit Malaysia (RM) 60millions while the second board companies are those that have a minimum paid-up capital ofRM40 millions.
6. Following Abbott et al.(2003), firms in their first year of public trading are excluded because it isexpected that these firms require an initial period of time post initial public offerings to appointan audit committee and effectively transfer functions to the committee and for the committee tomeet. Some companies in the sample did not meet during the year as stipulated by the ListingRequirements of Bursa Malaysia because they were new listings. The Listing Requirements ofBursa Malaysia requires audit committees of its listed firms to meet at least four times a year.
7. In compliance with the Companies Act 1965, all listed firms disclose their substantialshareholders including their 30 largest shareholders in their annual reports. The Bumiputerashareholding percentage is based on the 30 largest shareholders. Therefore, aBumiputera-controlled company (by virtue of their largest shareholdings in a company) isone in which 50 percent or more of the 30 largest shareholdings is held by government andsemi-government agencies, Bumiputera individuals, Bumiputera-owned companies, andBumiputera trust agencies. The same categorisation procedure is used to identifynon-Bumiputera shareholdings.
8. Other measures of ethnicity used in this study include: the proportion of Bumiputera directorson boards; a dichotomous variable set as 1 if a Board of Directors is predominantlyBumiputeras, 0 if otherwise; and the proportion of Bumiputera substantial shareholdings. Theethnicity of the board of directors is determined by examining the names of directors. If thenames are of Chinese origin, for example, having surname of Tan, Chan, or Lee, the criterion issatisfied, that is, the director is assumed to be Chinese. A similar approach is also used fordirectors who are non-Bumiputeras (i.e. foreign directors or other ethnicities other thanBumiputeras or Malays). For Bumiputera names or Malay names, the same procedure is used.The board of directors is categorised as non-Bumiputera directors in the case of uncertaintiesbetween Bumiputera directors and non-Bumiputera directors (Eichenseher, 1995). Thiscategorization rule is based on the notion that government pressure is exerted to place localand particularly Bumiputera (Malay) directors on boards of directors to increase Bumiputera’sparticipation in listed companies as per New Economy Policy initiatives (Mehmet, 1988).
9. All companies have at least three members on their audit committees except Johan HoldingsBerhad, KPS Consortium Berhad, and UCP Resources Berhad which have two directors ontheir audit committees.
10. Two tests are performed to detect multicollinearity. First, variance inflation factor (VIF)scores reveal no problems with multicollinearity (all scores are less than 2). Second, perBesley et al.(1980), the condition indices are calculated. The condition index is less than 30indicating weak relations among the independent variables (Besley et al., 1980). Therefore, itappears that multicollinearity is not a problem.
Governancestructures,
ethnicity
777
11. The study performs a number of diagnostics on the regressions reported in Table IV, and inthose reported in subsequent analyses, including investigation of outliers for both control andgovernance variables. The variables with univariate outliers include size, leverage, receivablesand inventory intensity, frequency of both board and audit committee meetings, and board size.To test whether outliers alter the results, the outlying observations whose standardised z-scoresexceed ^ 3 are excluded and the regressions are re-run. The unreported regression results arequalitatively similar as those reported in Table IV. Tests are also conducted to detectheteroscedasticity. The Breusch-Pagan-Godfrey test indicates that a heteroscedascity problemexists. White’s (1980) heteroscedasticity consistent covariance matrix is performed for allregressions run in this study to help correct the problem.
12. To address the potential effect of listed board, a dichotomous variable (i.e. whether a firm islisted on the main board or the second board of the Bursa Malaysia) is included in theregressions. The unreported results show that external audit fees are positively andsignificantly associated with the BOARD variable (the coefficient ¼ 0.108;t-statistic ¼ 2.204; p-value ¼ 0.028). The unreported correlation indicates that thisvariable is highly correlated with the SIZE variable (0.511; significant at the 0.01 level;two-tailed). Because of high correlation between the BOARD and the SIZE variables, the studydrops the BOARD variable from the regressions. Regression results remain qualitatively andstatistically unchanged when the variable is included (or excluded) from the regressions.
13. As an alternative measure of leverage, the ratio of long-term debt to total assets(LTDebt/TA) is used in place of the ratio of total debt to total assets (a variable denoted asLeverage in the regressions throughout the study). The unreported results show that thereexists a positive association between external audit fees and the ratio of long-term debt tototal assets. The positive association, however, is not statistically significant (thecoefficient ¼ 0.252; t-statistic ¼ 1.497; p-value ¼ 0.135).
14. External audit fees are not significantly associated with industrial products( p-value ¼ 0.827), Consumer Products ( p-value ¼ 0.665), Constructions( p-value ¼ 0.852), Technology ( p-value ¼ 0.563), Trade and Services ( p-value ¼ 0.495),and Hotel ( p-value ¼ 0.638). There also exists a non-significant negative association betweenexternal audit fees and property (0.909), plantations ( p-value ¼ 0.985), and infrastructurecompanies ( p-value ¼ 0.529) sectors. In order to have parsimonious models and also due totheir non-significant associations with external audit fees, the industry variables are excludedand regressions are re-run. The results remain qualitatively and statistically unchanged afterexcluding the industry variables from the regressions. It is also important to note that theadjusted R 2 for the regression with industry variables is 70.6 percent, which is about 0.01percent higher than that without industry variables (see R 2 in model 1 in Table IV). Othercontrol variables also remain qualitatively and statistically unchanged.
15. The ethnicity variable is excluded because model 3 seeks to investigate the association betweenexternal audit fees and board characteristics, which are largely found in the existing literature.
16. The study also further examines whether board size is likely to be influenced by size ofcompanies measured by total assets. One would argue that larger companies are likely tohave more directors on their boards to reflect complexity of their operations anddiversifications of the company. Therefore, a greater number of directors with diverseexpertise and greater facilitation of task assignments are required. The study, therefore,scales board size by total assets. When this transformed variable is included in theregressions and the regressions are re-run, the results obtained containing this variable isstatistically similar to those of absolute value of board size, indicating that board size is notinfluenced by the size of the company.
17. An explanation for the adjusted R 2s being almost the same for all models is that controlvariables consistently explain more variation in external audit fees than governance variables.
MAJ21,7
778
Governance variables provide little incremental explanatory power. Notwithstanding that, thestrong association between external audit fees and governance variables appear to supportexisting empirical evidence from the demand approach of audit services.
18. There are some mixed results with respect to other measures of ethnicity. For instance, whenmeasured using the proportion of Bumiputera directors on boards, the proportion ofBumiputera substantial shareholdings, and a dichotomous variable of whether a firm’sboard of directors is predominantly Bumiputeras, the association between external audit feesand these variables are not statistically significant.
19. The majority of businesses in Malaysia are owned and operated by Chinese (Horii, 1991).Chinese businesses generally have some common characteristics. These characteristicsinclude centralised decision-making with heavy reliance on one dominant chief executive,family ownership and control, and most if not all top management positions are filled withfamily members.
References
Abbott, L.J., Parker, S. and Peters, G.F. (2004), “Audit committee characteristics andrestatements”, Auditing: A Journal of Practice & Theory, Vol. 23 No. 1, pp. 69-87.
Abbott, L.J., Parker, S., Peters, G.F. and Raghunandan, K. (2003), “The association between auditcommittee characteristics and audit fees”, Auditing: A Journal of Practice & Theory, Vol. 22No. 2, pp. 17-32.
Anderson, D., Mansi, S.A. and Reeb, D.M. (2004), “Board characteristics, accounting report integrity,and the cost of debt”, Journal of Accounting and Economics, Vol. 37 No. 3, pp. 315-42.
Beasley, M.S. (1996), “An empirical analysis of the relation between the boards of directorscomposition and financial statement fraud”, The Accounting Review, Vol. 71 No. 4,pp. 443-65.
Beasley, M.S., Carcello, J.V., Hermason, D.R. and Lapides, P.D. (2000), “Fraudulent financialreporting: consideration of industry traits and corporate governance mechanisms”, TheAccounting Horizons, Vol. 14 No. 4, pp. 441-54.
Besley, D.A., Kuh, E. and Welsch, R.E. (1980), Regression Diagnostics: Identifying Influential Dataand Sources of Collinearity, Wiley, New York, NY.
Blue Ribbon Committee (1999), Report and Recommendations on Improving the Effectiveness ofCorporate Audit Committees, The New York Stock Exchange and the NationalAssociation of Securities Dealers, New York, NY.
Braiotta, L. (2000), The Audit Committee Handbook, Wiley, New York, NY.
Brickley, J.A., Coles, J.L. and Terry, R.L. (1994), “Outside directors and the adoption of poisonpills”, Journal of Financial Economics, Vol. 35 No. 3, pp. 371-90.
Bursa Malaysia Listing Requirements (2001), Bursa Malaysia, Kuala Lumpur.
Business Roundtable (1997), “Statement on corporate governance”, a white paper from theBusiness Roundtable, available at: www.businessroundtable.org/pdf/11.pdf
Byrne, J. (1996), “And you thought CEOs were overpaid”, Business Week, August 26th, p. 34.
Cadbury Committee (1992), Report on the Financial Aspects of Corporate Governance, Gee andCompany Limited, London.
Carcello, J.V., Hermanson, D.R., Neal, T.L. and Riley, R.A. (2002), “Board characteristics and auditfees”, Contemporary Accounting Research, Vol. 19 No. 3, pp. 365-84.
CheAhmad, A. and Houghton, K.A. (2001), “The effect of ethnicity on audit pricing”, workingpaper, Universiti Utara, Malaysia and University of Melbourne, Melbourne.
Governancestructures,
ethnicity
779
Claessens, S., Djankov, S. and Lang, L.H.P. (2000), “The separation of ownership and control inEast Asian corporations”, Journal of Financial Economics, Vol. 58 Nos 1/2, pp. 81-112.
Cohen, J., Krishnamoorthy, G. and Wright, A.M. (2002), “Corporate governance and the auditprocess”, Contemporary Accounting Research, Vol. 19 No. 4, pp. 573-94.
Committee of Sponsoring Organisations of the Treadway Commission (1992), Internal Control –Integrated Framework, AICPA, Harborside, NJ.
Committee of Sponsoring Organisations of the Treadway Commission (2004), Enterprise RiskManagement – Integrated Framework, AICPA, New York, NY.
Conger, J.A., Finegold, D. and Lawler, E.E. III (1998), “Appraising boardroom performance”,Harvard Business Review, Vol. 76 No. 1, pp. 136-48.
Craswell, A., Francis, J. and Taylor, S. (1995), “Auditor brand name reputations and industryspecialisations”, Journal of Accounting and Economics, Vol. 20 No. 3, pp. 297-322.
Davidson, R., Goodwin, J. and Kent, P. (2005), “Internal governance structures and earningsManagement”, Accounting and Finance, Vol. 45 No. 2, pp. 241-67.
Dechow, P.M., Sloan, R.G. and Sweeney, A. (1996), “Causes and consequences of earningsmanipulation: an analysis of firms subject to enforcement actions by the SEC”,Contemporary Accounting Research, Vol. 13 No. 1, pp. 1-36.
DeZoort, F.T. and Salterio, S. (2001), “The effects of corporate governance experience, financialreporting, and audit knowledge on audit committee members’ judgements”, Auditing: AJournal of Practice & Theory, Vol. 20 No. 2, pp. 31-47.
Eichenseher, J.W. (1995), “Additional factors in audit pricing: new evidence from Malaysia”,Accounting & Business Review, Vol. 2 No. 1, pp. 1-26.
Fama, E.F. (1980), “Agency problems and the theory of the firm”, Journal of Political Economy,Vol. 88 No. 2, pp. 288-307.
Fama, E.F. and Jensen, M. (1983a), “Separation of ownership and control”, Journal of Law &Economics, Vol. 26 No. 2, pp. 301-26.
Fama, E.F. and Jensen, M. (1983b), “Agency problems and residual claims”, Journal of Law &Economics, Vol. 26 No. 2, pp. 327-49.
Francis, J. and Simon, D. (1987), “A test of audit pricing in the small-client segment of the auditUS audit market”, The Accounting Review, Vol. 62 No. 1, pp. 145-57.
Francis, J. and Stokes, D. (1986), “Audit prices, product differentiation and scale of economies:further evidence from the Australian market”, Journal of Accounting Research, Vol. 24No. 2, pp. 383-93.
Gibson, M.S. (2003), “Is corporate governance ineffective in emerging markets?”, Journal ofFinancial and Quantitative Analysis, Vol. 38 No. 1, pp. 231-50.
Gilson, S.C. (1990), “Bankruptcy, boards, banks, and block-holders: evidence on changes incorporate ownership and control when firms default”, Journal of Financial Economics,Vol. 27 No. 2, pp. 355-88.
Gomez, E.T. and Jomo, K.S. (1997), Malaysia’s Political Economy: Politics, Patronage and Profits,Cambridge University Press, Singapore.
Goodwin-Stewart, J. and Kent, P. (n.d.), “The relation between external audit fees, auditcommittee characteristics and internal audit”, Accounting and Finance(in press).
Gul, F.A. (2003), “Financial crisis, cronyism and audit risk in Malaysia: some evidence from auditpricing”, working paper, City University of Hong Kong.
MAJ21,7
780
Gul, F.A. and Tsui, J. (1998), “A test of the free cash flow and debt monitoring hypotheses:evidence from audit pricing”, Journal of Accounting and Economics, Vol. 24 No. 2,pp. 219-37.
Haniffa, R.M. and Cooke, T.E. (2002), “Culture, corporate governance and disclosure in Malaysiancorporations”, Abacus, Vol. 38 No. 3, pp. 317-49.
Hanrahan, P., Ramsay, I. and Stapledon, G. (2001), Commercial Applications of Company Law,CCH Australia Limited, Sydney.
Hermanson, D.R. (2003), “What else in corporate governance should be changed?”, InternalAuditing, Vol. 18 No. 1, pp. 44-5.
Hofstede, G.H. (1980), Culture’s Consequences: International Differences in Work Related Values,Sage, Beverly Hills, CA.
Hofstede, G.H. (1991), “Management in a multicultural society”, Malaysian Management Review,Vol. 26 No. 1, pp. 3-12.
Horii, K. (1991), “Disintegration of the colonial economic legacies and social restructuring inMalaysia”, The Developing Economies, Vol. 29 No. 4, pp. 281-313.
Jensen, M.C. (1993), “The modern industrial revolution, exit, and the failure of internal controlsystems”, Journal of Finance, Vol. 48 No. 3, pp. 831-80.
Jesudason, J.V. (1989), Ethnicity and the Economy: The State, Chinese Businesses, andMultinationals in Malaysia, Oxford University Press, Singapore.
Jiambalvo, J. (1996), “Discussion of causes and consequences of earnings manipulation: ananalysis of firms subject to enforcement by the SEC”, Contemporary Accounting Research,Vol. 13 No. 1, pp. 37-47.
Johnson, S. and Mitton, T. (2003), “Cronyism and capital controls: evidence from Malaysia”,Journal of Financial Economics, Vol. 67 No. 2, pp. 351-82.
Kalbers, L.P. and Fogarty, T.J. (1993), “Audit committee effectiveness: an empirical investigation ofthe contribution of power”, Auditing: A Journal of Practice & Theory, Vol. 12 No. 1, pp. 24-49.
KLSE-PricewaterhouseCoopers (2002), Corporate Governance: 2002 Survey of Public ListedCompanies in Malaysia, Kuala Lumpur Stock Exchange (KLSE), Malaysia.
Knapp, M. (1987), “An empirical study of audit committee support for auditors involved in technicaldisputes with client management”, The Accounting Review, Vol. 62 No. 3, pp. 578-88.
Lemmon, M. and Lins, K. (2001), “Ownership structures, corporate governance, and firm value:evidence from the East Asian Financial crisis”, William Davidson working paper, DavidEccles School of Business, University of Utah.
Lins, K.V. (2003), “Equity ownership and firm value in emerging markets”, Journal of Financialand Quantitative Analysis, Vol. 38 No. 1, pp. 159-84.
Lipton, M. and Lorsch, J. (1992), “A modest proposal for improved corporate governance”, TheBusiness Lawyer, Vol. 48 No. 1, pp. 59-77.
Low, L., Tan, P.H. and Koh, H. (1990), “The determination of audit fees: an analysis in theSingapore context”, Journal of Business Finance & Accounting, Vol. 17 No. 2, pp. 285-95.
McMullen, D.A. (1996), “Audit committee performance: an investigation of the consequencesassociated with audit committees”, Auditing: A Journal of Practice & Theory, Vol. 15 No. 1,pp. 87-103.
Malaysian Code on Corporate Governance (2001), Malaysian Institute of Corporate Governance(MICG), Malaysian Law Journal Sdn Bhd, Kuala Lumpur.
Mak, Y.T. and Li, Y. (2001), “Determinants of corporate ownership and board structure: evidencefrom Singapore”, Journal of Corporate Finance, Vol. 7 No. 3, pp. 235-56.
Governancestructures,
ethnicity
781
Mehmet, O. (1988), Development in Malaysia: Poverty, Wealth, and Trusteeship, The Institute ofSocial Analysis, Kuala Lumpur.
NASDAQ Corporate Governance Proposals (2003), “Summary of NASDAQ corporategovernance proposals”, available at: www.nasdaq.com/about/Corp_Gov_Summary.pdf
New York Stock Exchange (2002), Report of the NYSE Corporate Accountability and ListingStandards Committee, available at: www.nyse.com/pdfs/corp_gov_pro_b.pdf
Peasnell, K.V., Pope, P.F. and Yong, S. (2000), “Board composition and earnings management: dooutside directors constrain abnormal accruals?”, working paper, University of Lancaster.
Raghunandan, K.R., Rama, D.V. and Scarbrough, D.P. (1998), “Accounting and auditingknowledge level of Canadian Audit Committee: some empirical evidence”, Journal ofInternational Accounting, Auditing and Taxation, Vol. 7 No. 2, pp. 181-94.
Redding, G.S. (1995), “Overseas Chinese networks: understanding the enigma”, Long RangePlanning, Vol. 28 No. 1, pp. 61-9.
Selim, G. and McNamee, D. (1999), “Risk management and internal auditing: what are theessential building blocks for a successful paradigm change?”, International Journal ofAuditing, Vol. 3 No. 2, pp. 147-55.
Shleifer, A. and Vishny, R.W. (1997), “A survey of corporate governance”, Journal of Finance,Vol. 52 No. 2, pp. 737-83.
Simon, D.T. and Francis, J.R. (1988), “The effect of auditor on audit fees: test of price cutting andprice discovery”, The Accounting Review, Vol. 63 No. 2, pp. 255-69.
Simunic, D. (1980), “The pricing of audit services: theory and evidence”, Journal of AccountingResearch, Vol. 18 No. 1, pp. 161-90.
Simunic, D. and Stein, M.T. (1987), Product Differentiation in Auditing: Auditor Choice in theMarket for Unseasoned New Issues, Canadian Certified General Accountants’ ResearchFoundation, Vancouver, British Columbia.
Simunic, D. and Stein, M.T. (1996), “The impact of litigation risk on audit pricing: a review of theeconomics and the evidence”, Auditing: A Journal of Practice & Theory, Vol. 15 No. 2,pp. 145-9.
Vafeas, N. (1999), “Board meeting frequency and firm performance”, Journal of FinancialEconomics, Vol. 53 No. 1, pp. 113-42.
Weindebaum, M. (1996), “The Chinese family business enterprise”, California ManagementReview, Vol. 38 No. 4, pp. 141-56.
White, H.A. (1980), “Heteroscedasticity-consistent covariance matrix estimator and a direct testfor heteroscedasticity”, Econometrica, Vol. 48 No. 4, pp. 817-38.
Zhuang, J. (1999), “Some conceptual issues of corporate governance”, EDRC Briefing NotesNumber 13, Asian Development Bank (ADB).
Further reading
Securities and Exchange Commission (1999), Final Rule: Audit Committee Disclosure, SEC,Washington, DC, Release No. 34-42266.
Corresponding authorPuan Yatim can be contacted at: [email protected]
MAJ21,7
782
To purchase reprints of this article please e-mail: [email protected] visit our web site for further details: www.emeraldinsight.com/reprints