Corporate Governance Mechanisms and Thei

Australian Journal of Management
http://aum.sagepub.com/
Corporate governance mechanisms and their impact on company performance: A

structural equation model analysis
Mohammad I Azim
Australian Journal of Management published online 2 August 2012
DOI: 10.1177/0312896212451032
The online version of this article can be found at:

http://aum.sagepub.com/content/early/2012/07/30/0312896212451032
Published by:
http://www.sagepublications.com
On behalf of:
Australian School of Business
Additional services and information for Australian Journal of Management can be found at:
Email Alerts: http://aum.sagepub.com/cgi/alerts
Subscriptions: http://aum.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
>> OnlineFirst Version of Record - Aug 2, 2012
What is This?
Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

0010.1177/0312896212451032AzimAustralian Journal of Management
2012
Article
Australian Journal of Management

1–25
Corporate governance mechanisms © The Author(s) 2012
Reprints and permission: sagepub.
and their impact on company co.uk/journalsPermissions.nav
DOI: 10.1177/0312896212451032
performance: A structural aum.sagepub.com
equation model analysis
Mohammad I Azim
Department of Accounting, Economics, Finance and Law, Swinburne University of Technology, Australia
Abstract
The purpose of this study is to use structural equation modelling (SEM) to investigate the extent to which
different monitoring mechanisms – the board and its committees, shareholders and independent auditors –
are complements (i.e. a positive covariance) or substitutes (a negative covariance) for each other. The
lack of consistent results in previous corporate governance research may be attributable to attention not
being paid to monitoring mechanisms’ substitution or complementary relationships. By using SEM, this study
concludes that complementary and substitution relationships among monitoring mechanisms are present.
Using data from the pre – and post – global financial crisis period, this study explains where such corporate
government impacts occurred, the inconsistencies that are evident in previous studies and provides insights
into corporate governance practices.
JEL Classification: M41, M42, M49
Keywords
complement, monitoring, performance, structural equation modelling, substitution
1. Introduction
In recent years governments’, regulators’ and researchers’ attention to corporate governance has
increased significantly. Such interest has largely been directed towards the monitoring of overall
corporate governance structures, financial controls and financial performance. This research
focuses on monitoring of overall corporate governance structures and the relationships between
those monitoring roles using structural equation modelling (SEM). SEM is a technique for analys-
ing data designed to assess relationships existing across both manifest (i.e. directly measured or
observed) and latent (i.e. the underlying theoretical construct) variables (Martens, 2005). When
Corresponding author:
Mohammad I Azim, Department of Accounting, Economics, Finance and Law, Swinburne University of Technology,
Hawthorn, Melbourne, VIC 3122, Australia.
Email: mazim@swin.edu.au

2 Australian Journal of Management
using statistical techniques, such as multiple regression or analysis of variance (ANOVA), the
researcher only conducts the analysis on variables that are directly measurable, which can be some-
what limiting when the individual is interested in testing alternate governance mechanisms and
including a list of the control variables in their analysis. Using various measures of financial per-
formance as a proxy for shareholders’ interests, this research examines whether selected monitor-
ing mechanisms affect company performance. A sample was taken from Australia’s top-500 listed
companies for the pre- and post-global financial crisis (GFC) period, in order to examine the extent
to which these mechanisms function as substitutes or complement each other.
SEM as a method emerged in the early 20th century, developed by Sewall Wright in 1916 (Bollen,
1990). The increasing complexity of research questions and the appearance of flexible user-friendly
computer software (Marcoulides and Hershberger, 1999) have increased interest in SEM (Hair et al.,
2009; Kelloway, 1998; Raykov and Marcoulides, 1999). SEM has become increasingly popular in
psychology, education, the social sciences (Fan et al., 1999), biology, economics, medicine, market-
ing (Raykov and Marcoulides, 1999), psychology (Bollen, 1990), demography and even genetics
(Hair et al., 2009). It has also been used in fields related to this research, including consumer behav-
iour (Jarratt, 2000) and ethical decision-making (Hair et al., 2009).
In corporate governance research, there has been a slow but steady increase in the use of SEM
(Daily et al., 1999; Lin, 2005; Sahnoun and Zarai, 2009; Tam and Tan, 2007; Zhang, 2010). For
example, Daily et al. (1999) used a structural equation model to measure board composition. Lin
(2005) studies the influence of board of directors and large external shareholders on controlling
CEO compensation using SEM. Tam and Tan (2007) examines the relationship between ownership
types and firm performance through SEM. Sahnoun and Zarai (2009) uses SEM to investigate the
impact of auditee business risk, audit risk and auditor business risk evaluation on auditor–auditee
negotiation outcomes. Zhang (2010) employs SEM to examine how corporate boards can create a
sustainable competitive advantage. While all these studies used SEM in a single monitoring setting
(i.e. board of directors, shareholders or auditors), attention has not been given to the combined
effect of monitoring mechanisms and the possibility that these mechanisms may have a substitu-
tion or complementary effect.
Again, the possibility of substitution among various monitoring mechanisms has been previ-
ously examined by using different statistical methods, except SEM (e.g. Boo and Sharma, 2008;
Coles et al., 2001; Fernández and Arrondo, 2005; Ward et al., 2009). Ward et al. (2009) proposed
that when companies are performing poorly, outside monitoring by institutional investors can com-
plement internal monitoring, a function usually done by boards of directors. Boo and Sharma
(2008) examined the relationship between internal governance, external audit monitoring and reg-
ulatory oversight. They concluded that regulatory oversight reduces information asymmetry;
hence, reducing the demand for costly monitoring by the external auditor. Fernández and Arrondo
(2005), using a sample consisting of Spanish companies, found that when monitoring by directors
increases, monitoring by shareholders decreases and vice versa. Coles et al. (2001) argue that com-
panies may substitute governance choices in selecting the monitoring mechanisms most suitable
for their organisational and environmental contexts. However, a major limitation of these studies
concerned the existence of endogeneity. Indeed, failing to fully control for all forms of endogeneity
can lead to spurious results (Schultz et al., 2010).
Companies’ performance depends on how effective is their mix of monitoring mechanisms to
control the agency problem (Agrawal and Knoeber, 1996; Rediker and Seth, 1995; Ward et al.,
2009). Even though the overall bundle is effective in aligning managers’ and shareholders’ inter-
ests, the impact of any one mechanism may not provide similar results due to the substitute and
complementary relationships (Ward et al., 2009). This issue can be explored by using SEM to

Azim 3
explore the multivariate relationships that exist between different monitoring mechanisms and
their performance consequences. This research aims to provide insights into how the bundle of
monitoring mechanisms works to reduce the agency problem.
SEM estimates a series of separate, but interdependent, multiple regression equations simulta-
neously by specifying the structural model. The same structural model can express relationships
among independent and dependent variables, even when dependent variables become independent
variables in other relationships. There are two major advantages of using SEM over other statistical
tools; firstly, it provides a straightforward method for dealing with multiple relationships simulta-
neously while providing statistical efficiency; and secondly, it has the ability to analyse the rela-
tionships comprehensively and provide a transition from exploratory to confirmatory analysis.
In research on corporate monitoring, there are some variables that cannot be observed directly.
These variables are termed latent variables in SEM. In this research examples of latent variables
are monitored as undertaken by the board of directors, shareholders and auditors. As these varia-
bles are not observed directly, they need to be related to the observed variables to determine their
value. In SEM, it is possible to include both observed and latent variables. Observed variables are
represented by data and are usually continuous. A latent variable is a hypothesised and unobserved
concept that can only be approximated by observable or measurable variables. Latent variables are
expressed in terms of observed variables. The latent variables in SEM are continuous regardless of
the properties of the observed variables.
Regulatory responses in Australia, such as the Australian Securities Exchange Corporate
Governance Council (ASX CGC) and CLERP 9 (Commonwealth of Australia, 2002), have not
addressed the substitute and complementary governance issues. While companies operate in an
increasingly globalised economy and very competitive environment, they are also subject to
regional and local societal, regulatory and business contexts, which may give rise to differences in
corporate governance structures and practices.
This paper is structured as follows: Section 2 develops a conceptual view of monitoring mecha-
nisms and performance using an agency perspective; Section 3 discusses the research propositions
and methodology; Section 4 describes sample selection and descriptive statistics for the research;
Section 5 provides the analysis; and Section 6 presents concluding comments and highlights pos-
sible avenues for future research.
2. Theoretical basis for model specification and causality

Theory is important for all multivariate analysis, but it is particularly important for SEM because
it is considered a confirmatory form of analysis; that is, it is useful for testing and potentially con-
firming theory. Theory is needed to specify relationships in structural models. The SEM process
begins by choosing the variables that will be measured. It concludes with assessing the overall
structural model fit. In the whole process theory plays a key role at each step. The goal of SEM is
to test the theory because without theory, a true SEM test cannot be conducted.
Drawing substantially on agency theory literature and previous empirical research results, this
research maps how primary monitoring mechanisms may interrelate and affect company perfor-
mance. A conceptual model has been constructed for this study in order to provide an integrated
understanding of how monitoring mechanisms affect company performance. The scope of moni-
toring research is depicted in Figure 1.
Agency risk will decrease if and when managers can be induced to act in the best interests of
shareholders (Jensen and Meckling, 1976). Denis (2001) finds that there are three devices that encour-
age management to act in this way: contractually bonding management, providing contractual

incentives and monitoring management activities. Monitoring, however, is a necessary condition for
using the first two devices. Firstly, contractually bonding management requires canvassing the pos-
sible eventualities and actions that a manager should take. This is not impossible, but requires a high
level of predictability and efficient monitoring. Secondly, incentive contracts require monitoring to
evaluate the agent’s performance and provide incentives accordingly (Denis, 2001). Thus, an effec-
tive monitoring mechanism is required for all situations to induce management to work in the inter-
ests of shareholders (Shleifer and Vishny, 1997).
There are different kinds of monitoring mechanisms (internal and external) available and they can be
classified into three broad groupings: market, internal monitoring and regulatory. Market mechanisms
may involve block shareholders, the capital market and the managerial labour market. In a single-
country study, the capital market and managerial labour market are common to all companies and there
is little scope to differentiate among these general market monitoring mechanisms (Agrawal and
Knoeber, 1996; Denis and McConnell, 2003). Generally, the reputation of individuals in the labour
market is a concern to management. In terms of the managerial labour market, if the performance of the
company, in which the managers are serving, is not satisfactory, it will be difficult for managers to sell
themselves in the competitive market. As an internal monitoring mechanism, this study concentrates on
insider shareholders and boards of directors. Another important internal monitoring mechanism is man-
agerial compensation. However, it is difficult to collect the breakdown of managerial compensation for
all the companies. Regarding regulatory mechanisms, Jensen (1993) acknowledges that the legal sys-
tem is characterised as ‘too blunt an instrument’ to deal with agency problems. Furthermore the legal
system is common to all companies in a single-country study. This study therefore focuses its attention
on the external auditors as a monitoring instrument operating in the form of a regulatory mechanism.
Shareholder Conflict of interest Management

Agency
Problem
Reduce Agency problem

Research Context
Perfect Contracting Monitoring Incentives to Management
Market Internal Control Regulatory

Mechanisms Mechanisms Mechanisms
Capital Managerial Managerial Regulatory Acts &

Market Labour Market compensation Agencies Regulations
Research Focus
Block Shareholders Board of Directors and Committees External Auditors
Company
Performance
Perfo
f rmance
Figure 1. Conceptual model of corporate governance mechanisms and their impact on company
performance.

Azim 5
As shown in Figure 1, the main focus of this paper is on the shareholders, boards of directors,
and auditors as monitoring mechanisms and the impact they have on company performance. The
following review of relevant studies provides the structure that informed this study.
2.1. Monitoring by shareholders

Monitoring by shareholders is classified into two groups: firstly, monitoring by the insiders (execu-
tive and non-executive directors); and secondly, by the major outside shareholders (block share-
holders). The role of director ownership as a monitoring mechanism has been the subject of much
empirical analysis (Hermalin and Weisbach, 1991; McConnell and Servaes, 1990; Morck et al.,
1988). When board members have considerable holdings in a company’s shares, their decisions
have an impact on their personal wealth. The degree of ownership concentration in a company
determines the distribution of power between its managers and shareholders (Welch, 2003). When
ownership is dispersed, shareholders’ control tends to be weak; when ownership is concentrated,
major shareholders play an important role in monitoring and reducing the scope for managerial
opportunism (Shleifer and Vishny, 1986).
2.2. Monitoring by the board of directors

The theoretical role of the board in monitoring and disciplining management is firmly grounded in
the agency framework developed by Fama and Jensen (1983). Empirical examination of board
characteristics and company performance focuses on the following criteria.
2.2.1. Board size. Researchers have emphasised the influence of board size on company perfor-
mance (Jensen, 1993). Larger boards are less effective monitors due to potential free riding, com-
munication breakdowns and inefficiencies (Boo and Sharma, 2008; Bushman et al., 2004).
Although the direction of influence is unclear (Kent and Stewart, 2008), most studies find a posi-
tive relationship between size of the board and both company performance (Chiang, 2005; Haniffa
and Hudaib, 2006) and board monitoring (Anderson et al., 2004; Williams et al., 2005). Kiel and
Nicholson (2003) suggest that there is an inverted U relationship between board size and company
performance in which adding directors can bring to the board a high level and mix of optimal skills
and experience. Beyond that point the difficult dynamics of a large board prevail over the skills and
expertise that additional directors might bring.
2.2.2. Independent directors. Boards of directors’ ability to act as effective monitoring mechanisms
rely on their independence from management (Beasley, 1996). Researchers have focused on the
proportion of executives to independent directors1 as an indicator of board independence (David-
son et al., 2005; Koh et al., 2007; Peasnell et al., 2005). Some previous studies have suggested that
independent directors are effective monitors because they do not have financial interests in the
company or psychological ties to management (Boo and Sharma, 2008). They are in a better posi-
tion to objectively challenge management (Abbott et al., 2004; Klein, 2002). Bedard and Johnstone
(2004) have also argued that higher independent director representation on the board provides
more vigilant oversight of the monitoring process.
2.2.3. Separate role of CEO and chairman. Conflict of interest may exist where the roles of CEO and
chairman (chair) of the board are held by the same person. When a single individual fulfils both
roles, managerial dominance is likely to be enhanced, since there is greater alignment of the board

with management rather than shareholders. Where the two roles are combined, the CEO will be
able to control the board, reduce the board’s independence from management and make decisions
in their own interest (Jensen, 1993). Corporate governance guidelines also assume a concentration
of power when the CEO is also the chairperson of the board (ASX CGC, 2007; Cadbury, 1992;
Standards Australia International, 2003). However, in Australia, unlike the US, not all executive
chairs are the CEO (e.g. Macquarie Bank). In this paper, duality measured as the combined role of
chair/CEO captures a subset of the ‘non-independent chair’.
2.2.4. Financial expertise of directors. According to the Blue Ribbon Committee Report (1999), a
well-balanced and effective board should have directors with an array of talent, experience and
expertise that influence different aspects of the company’s activities; such diverse contributions are
often made by different directors. DeZoort and Salterio (2001) and Cohen et al. (2002) conclude
that it is important for committee members to have accounting and financial expertise. Similarly,
Ramsay (2001) notes that financial literacy is ‘an important component of the general standards of
care, skill and diligence required of company directors’ (p. 155). The directors’ financial literacy
helps them to understand the implications of basic financial decisions. Financial literacy can be
acquired through both formal and self-guided education (ASX CGC, 2007; Cohen et al., 2002;
DeZoort and Salterio, 2001; Livingston, 2002). The general implication is that financial literacy
assists directors in monitoring management.
2.2.5. Board diligence. The number of board meetings per year influences the monitoring ability of
the board. Too many or too few meetings can be a threat to effective board monitoring. Too few
meetings may indicate the directors are not paying proper attention to the company; again, too
many may indicate that there is some difficulty in the firm (Kang et al., 2007; Vafeas, 1999).
Boards that meet frequently are more likely to perform their duties diligently and effectively,
thereby enhancing their level of oversight (Yatim et al., 2006). Boards of directors need to be active
in ensuring high-quality transparent reporting in annual reports (Kent and Stewart, 2008).
2.2.6. Committees of the board. Board committees improve the efficiency of board monitoring by
effecting closer scrutiny of management activities and decision-making. This is particularly true
when the board size is large (Menon and Williams, 1994). ASX CGC (2007) recommends estab-
lishing three different committees of the board: audit, nomination and remuneration committee.
Audit committee. A board audit committee focuses on issues relevant to the integrity of the com-
pany’s financial reporting (ASX CGC, 2007; Chen and Zhou, 2007; Davidson et al., 2005). Audit
committees have received considerable attention following both more distant and more recent
corporate scandals (Sarens et al., 2009). Prior research has found that an audit committee meeting
frequently can reduce the incidence of financial reporting problems (Farber, 2005; Hughes, 1999).
Similar to the full board, the effectiveness of audit committee monitoring depends on: (i) the num-
ber of meetings (Farber, 2005); (ii) the independence of the board (ASX CGC, 2007); and (iii)
the financial literacy of the directors (DeZoort and Salterio, 2001; Karamanou and Vafeas, 2005).
Although the ASX CGC recommends that all listed companies should have an audit committee,
listing rules mandate that only those companies in the Standard & Poor’s (S & P) All Ordinaries
Index (the top-500 companies by market capitalisation) must have an audit committee.
Remuneration committee. A board remuneration committee is an efficient mechanism for focus-

ing the company on appropriate remuneration policies for senior executives. The monitoring

Azim 7
ability of the remuneration committee is marked by: (i) the number of meetings; and (ii) the pro-
portion of independent directors on the committee (Bosch, 1995).
Nomination committee. The monitoring ability of the nomination committee depends on their
independence and the frequency of meetings (Leblanc, 2004). The monitoring ability of the nomi-
nation committee can be captured by: (i) the number of meetings; and (ii) the proportion of inde-
pendent directors on the committee.
2.3. Monitoring by external auditors

The auditor’s role includes suggesting improvements in internal controls (ASA 265) and ongoing
concern issues (see ASA 570, 705, 706 in CPA Australia Handbook, 2012), but nothing directly
related to company performance. Publicly traded companies in Australia are required to have
audits, as stipulated in the Corporations Act 2001. However, the quality of audits and subsequent
ability to reduce agency costs varies significantly (DeAngelo, 1981). DeAngelo (1981) defines
audit quality as the joint probability that an auditor will: (i) detect a material misstatement in the
financial report if one exists (auditor competence); and (ii) report the misstatement if it is detected
(auditor independence). Lack of auditor independence will reduce audit quality through a reluc-
tance of the auditor to report any misstatements detected. Therefore, with respect to auditor moni-
toring, the general finding is that high levels of competence and independence are necessary for a
high-quality audit.
Empirical studies by Kim et al. (2003), Krishnan (2003), DeAngelo (1981), Dye (1993) and
Craswell et al. (1995) provide strong support for the view that large audit firms provide higher
quality audits compared to small firms. Kim et al. (2003) and Krishnan (2003) argue that large
audit firms have stronger incentives to protect their reputations, because they lose clients if they
produce low-quality audits. Researchers, policy makers and practitioners have debated the provi-
sion by audit firms of audit and non-audit services, because this dual provision can have conflicting
impacts on the quality of an audit.
2.4. Performance measures

This research uses both accounting and accounting-market (hybrid) performance measures to
examine the effect of monitoring mechanisms on company performance. It is expected that the
different monitoring mechanisms will influence management to work for the company’s best inter-
ests and this will eventually induce them to act in the interests of the stakeholders. Most perfor-
mance measurement models, such as return on equity (ROE) and return on assets (ROA), are
widely regarded by users as most useful and the ultimate bottom line of business performance.
Although accounting information is useful and important in corporate governance studies, all
agency costs are not reflected in the accounting measures (Wiwattanakantang, 2001). Bacidore
et al. (1997) argued in favour of using hybrid-based performance measurement – that is, the price–
earnings ratio (PER), market-to-book value (MBV) and dividend yield (DY) – to measure improved
value returned to shareholders.
3. Propositions and methodology

The various indicators or proxy elements associated with the (i) board and its committees, (ii)
shareholders, and (iii) independent auditors in the monitoring process and their relationship to

company performance are presented as an empirical schema. The research questions tested in this
study are as follows:
i. Do the various monitoring mechanisms associated with the three categories of players have
a substitution effect?
ii. Do the various monitoring mechanisms associated with the three categories of players have
a complementary effect?
To answer questions (i) and (ii), SEM is used as it can deal with multicollinearity and it reveals
potentially complex inter-relationships among monitoring mechanisms. By using SEM it is possi-
ble to model important latent variables while taking into account any unreliability of indicators
(Wooldridge, 2003). Most of the corporate governance variables are latent, which means they are
otherwise difficult to measure. Again, in all multivariate analyses it is assumed that there is no error
in the variables. However, from practical and theoretical perspectives it is impossible to perfectly
measure a latent concept without considering error. SEM improves the statistical estimate by con-
sidering this type of error. SEM examines a series of dependent relationships simultaneously –
which is particularly useful when one dependent variable becomes an independent variable in
subsequent dependent relationships – this helps to avoid the problem of multicollinearity.
No single statistical test can estimate the goodness-of-fit of the SEM model. Researchers have
developed a number of goodness-of-fit measures to assess the results. It is necessary to have suf-
ficient data to facilitate the observation of important differences or relationships. SEM applications
typically use 200–500 cases and 10–18 observed variables (Hair et al., 2009). This research uses
approximately 500 companies for each period: 2004–2006 and 17 observed variables (monitoring
mechanisms).
Block shareholders, because of their greater bargaining power over the firm, are more likely to
influence management. Block shareholders exercise a ‘voice’ through direct negotiation with man-
agement and proxy contests (Chowdhury, 2004). In so doing, block shareholders determine the
distribution of power between managers and shareholders. Ownership concentration can function
as a monitor to reduce the scope of managerial opportunism (Shleifer and Vishny, 1986). Monitoring
by block shareholder is measured by: (1) the percentage of shares held by the top shareholder; (2)
the percentage of shares held by the next top-19 shareholders; and (3) the percentage of shares held
by the executive and non-executive directors.
Board size is measured by number of directors on the board, and board diligence is measured by
number of board meetings per year. Board independence is measured by proportion of independent
directors to total directors. Financial expertise signifies past employment experience in finance or
accounting, requisite professional certification in accounting or any other comparable experience
or background that results in the individual’s financial sophistication, including being or having
been a CEO or other senior officer with financial oversight responsibilities (according to Blue
Ribbon Committee, 1999). To represent the degree of separation between the roles of board chair
and corporate CEO, the analysis employed a dummy variable, taking a value of 0 if the roles of the
chair and CEO are exercised by the same individual and 1 if the role is exercised by different
individuals.
Indicators of audit committee effectiveness are independence and expertise of committee mem-
bers. Independence is the proportion of committee members who are described as independent.
Similar to the board, the financial expertise of audit committee members is measured as the propor-
tion of members with financial literacy or having been a CEO/chief financial officer (CFO) or
other senior officer with financial oversight responsibilities. Since another recommendation is to
have regular meetings, the number of audit committee meetings is also measured. Effectiveness

Azim 9
and independence of the nomination and remuneration committee are measured by the number of
directors assigned to those committees and independence is determined by the proportion of com-
mittee members who are described as independent.
Auditors do not directly monitor management; however, they provide an assurance service that
improves the quality of financial information. External audits are used to increase the reliability of
financial statements (Chow, 1982). The extent to which financial statements can reduce agency
costs depends on the quality of the audit; thus, the quality of audit acts as a control mechanism. To
characterise the audit quality, this study classifies audit firms into large and small. Large firms
comprise the Big Four. Here, a value of 1 is assigned when the company uses a large audit firm and
0 otherwise. Another factor that can influence the monitoring of the auditor is the joint provision
of audit and non-audit services. This research therefore quantifies the amount of non-audit fees as
a ratio of total audit fees to determine the level of independence.
4. Data and company characteristics

The initial sample consists of 1500 company-year observations from the top-500 ASX-listed compa-
nies from 2004 to 2006. This study chooses 2004–2006 because this was the most critical period
following major corporate collapses and subsequent implementation of the ASX good corporate gov-
ernance guidelines in Australia, but before the GFC. This made it possible to investigate the nature of
mandatory corporate monitoring responses to the rash of corporate collapses. Top-500 companies
were chosen as they covered most of the ASX CGC requirements and were more likely to have sound
monitoring processes. Companies are ranked according to market capitalisation for each year. To
ensure consistency in the database, companies with missing information are excluded. Ninety com-
panies from the total 1500 company-year sample were excluded due to insufficient information. The
remaining companies consist of 475 in 2004, 472 in 2005 and 463 in 2006 (Table 1).
Table 1. Total number of companies per year in this study
Top-500 companies in 2004

De-listed (year 2005) 24
Suspended (year 2005) 1
Missing data (current and/or lagged year) 0
Final group of companies in 2004 475
Total number of company-years in the sample 1410
To test robustness, this research also collected data from the top-200 companies for the 2009–
2010 financial year following the GFC. This research excludes the GFC period (2008–2009), as
shareholders, board of directors and auditors were well aware of the GFC and would be very

conservative in their monitoring requirements. Therefore, the results during the crisis period might
not be typical or representative.
The archival sources used to collect data were the Aspect Financial Analysis database; Connect
4; and Aspect Huntley Financial Analysis. The information provided in Aspect and Connect 4 were
used to manually collect data on corporate monitoring variables, that is, board of directors,
committees, external auditors and shareholder information. Aspect Huntley, another Australian
comprehensive source of data for listed companies, is used to collect data on company perfor-
mance. Information is cross-checked with the annual reports obtained from Aspect and Connect 4.
The ASX website was used to rank the companies according to market capitalisation, obtain indus-
try classifications and establish whether the company is listed, de-listed or has changed its name.
Performance data for these companies were collected for the years 2004–2008 (and for 2010 for
a robustness test), so that one- and two-year lagged analysis is possible. We have calculated indus-
try average performance data, which help to avoid bias in the performance data. A probable con-
cern when dealing with current-year performance is that it is difficult to understand the cause–effect
relationship – which leads to the problem of endogeneity. This problem can be avoided when SEM
is employed (Renders and Gaeremynck, 2006).
With reference to industry classification (based on four-digit Global Industry Classification
Standards (GICS) codes), most company-years in the sample operate in the financial sector
(21.63%), followed by the materials sector (18.93%). The remaining 60% are involved in energy,
industrials, consumer discretionary, consumer staples, healthcare, information technology, tele-
communication and utilities.
As shown in Table 2, on average the top-20 shareholders held 63% of shares. In our sample, in
1059 company-years (75%) the top-20 shareholders held more than 50% of the issued shares and
in 431 company-years 31% held more than 75% of the issued shares. On average, the directors of
the board held 16% of issued shares.
In the study, the average board size was 6 directors (maximum = 17, minimum = 3). Most (n =
1246, 88%) had a board size of 4–9 directors. The average number of board meetings was 10 per
annum (maximum = 372, minimum = 2). More than 82% of company-years (1,158) had a board
with a majority of independent members. Eighty percent (1143) had 1–4 independent directors on
the board. In 56 (4%) company-years, there were no financially literate members on the board and
in 229 (16%), all directors were financially literate. In the sample, there were 168 (12%) where the
roles of chair and CEO were occupied by one person. One hundred and fifty-three (7%) did not
convene any audit committee meetings, and 66% (929) held 2–4 audit committee meetings per
annum. In the sample, 79% (1117 company-years) held between 1 and 3 meetings per annum. Only
288 (20%) company-years had a nomination committee. Of these, 239 companies convened 1–4
nomination committee meetings per annum. Only in six committees were there no independent
directors.
From the sample, 1158 company-years (82%) were audited by a Big Four auditor. It is expected
that the auditing by a Big Four auditor will improve the audit quality. The average non-audit fees
earned by the external auditor are $AU8369. In the sample of companies studied, 36% (508) earned
higher non-audit fees compared to audit fees.
The Spearman Correlation Matrix (Table 3) shows that there are few pairs of independent vari-
ables with a high correlation (i.e. more than 0.6). For example, there is a correlation of 0.6886
between the top-19 shareholders and top one shareholder; correlation of 0.6164 between the pro-
portion of financially literate directors on the audit committee and the proportion of board finan-
cially literate directors; and correlation of 0.6218 between the proportion of independent directors
on the nomination committee and nomination committee meetings. However, as this research uses
SEM, it can competently manage the multicollinearity issue.

Table 2. Descriptive statistics regarding monitoring measures (sample size: 1410 company-years)
Monitoring variables Combined sample 2004 2005 2006
Minimum Maximum Mean S.D. Minimum Maximum Mean S.D. Minimum Maximum Mean S.D. Minimum Maximum Mean S.D.
Top shareholder (%) 0.96 97.01 23.64 17.79 0.96 93.99 23.7 18.28 1.00 97 23.7 18.09 1.42 88.77 23.40 17
Top-19 shareholders (%) 7.53 100 63.3 20.1 8 100 63.3 20.63 7.56 99.88 63.7 20.39 7.53 99.86 63.02 19.29
Boards of directors’ 0.00 96.69 15.93 21.03 0.00 92.41 16.13 21.21 0.00 96.69 16.75 21.69 0.00 91.87 14.90 20.13
shareholdings (%)
Size of the boards 3 17 6.31 2.11 3 17 6.24 2.153 3 17 6.31 2.132 3 15 6.38 2.07
(number)
Boards of directors’ 2 37 10.78 4.33 2 33 10.7 4.284 2 32 10.8 4.217 2 37 10.80 4.52
meetings (per year)
Proportion of independent 0 1 0.71 0.195 0 1 0.71 0.203 0 1 0.7 0.202 0 1 0.71 0.18
directors on the boards
Proportion of financial 0 1 0.2139 0.1529 0 1 0.2265 .1505 0 1 0.2350 0.1474 0 1 0.2136 0.2135
literate directors on the
boards
Dual role of chair and 0 1 0.12 0.320 0 1 0.14 0.344 0 1 0.14 0.343 0 1 0.08 0.27
CEO (0,1)
Number of audit 0 15 3.03 2.02 0 12 2.83 1.85 0 14 3.04 2.024 0 15 3.24 2.16
committee meetings (per
year, N = 1265)
members on audit
committee (N = 1265)
Proportion of financially 0 1 0.44 0.34 0 1 0.46 0.34 0 1 0.46 0.34 0 1 0.39 0.34
AC (N = 1265)
Number of remuneration 0 15 1.49 2.12 0 14 1.32 1.98 0 15 1.51 2.245 0 15 1.64 2.11
year, N = 815)

directors on RC (N = 612)
Number of nomination 0 17 0.55 1.57 0 17 0.45 1.626 0 13 0.56 1.674 0 13 0.66 1.41
year, N =288)
directors on NC (N = 157)
Big Four audit firms (0.1) 0 1 0.82 0.38 0 1 0.81 0.389 0 1 0.83 0.379 0 1 0.82 0.38
Proportion of non-audit 0 0.96 0.39 0.24 0 0.96 0.41 0.25 0 0.92 0.4 0.24 0 0.96 0.35 0.23
service fees (NAF/TAF)
AC: audit committee, RC: remuneration committee, NC: nomination committee, NAF: non-audit fees, TAF: total audit fees
Table 3. Correlation matrix (sample size: 1410 company-years)
TOP 1 T 19 ISH% BSIZ BDM PBI PBFL CECH ACM PAI PAFL RCM PRI NCM PNI BIG 4 PNAF
Top shareholder (TOP 1) 1.0000

Top-19 shareholders 0.6886** 1.0000
(TOP 19)
Board of directors 0.2212** 0.3084** 1.0000
shareholdings (DSH %)
Size of the boards (BSIZ) 0.0034 0.0061 −0.1668** 1.0000
Board of directors −0.0901** −0.1449** 0.0066 −0.0292 1.0000
meetings (BDM)
Proportion of −0.0460 −0.1164** −0.2003** 0.1771** 0.1194** 1.0000
independent directors on
the boards (PBFL)
Proportion of financially −0.0413 −0.0757** 0.0586* 0.0189 0.0921** −0.0333 1.0000
boards (PBFL)
Dual role of chair and 0.0928** 0.1160** 0.1380** −0.1226** −0.1154** −0.3741** 0.0368 1.0000
CEO (CHCE)
Number of audit −0.0386 −0.0833** −0.1291** 0.4025** 0.2669** 0.2494** 0.1363** −0.1363** 1.0000
committee meetings (ACM)
Proportion of independent 0.0029 −0.0166 −0.1273** 0.2570** 0.1631** 0.3739** 0.0705** −0.2365** 0.3893** 1.0000
members on audit
committee (PAI)
Proportion of financial −0.0244 0.0031 0.0872** 0.1490** 0.1852** 0.0639* 0.6164** −0.0800** 0.2999** 0.3277** 1.0000
AC (PAFL)
Number of remuneration −0.0416 −0.0432 −0.1228** 0.3056** 0.1576** 0.1571** 0.0362 −0.1384** 0.3594** 0.2217** 0.1295** 1.0000
committee meetings (RCM)
Proportion of −0.0182 0.0069 −0.0489* 0.1934** 0.1988** 0.2155** 0.0286 −0.1276** 0.2137** 0.2896** 0.2731** 0.4736** 1.0000
independent directors on

RC (PRI)
Number of nomination −0.0900** −0.0770** −0.1172** 0.2140** 0.1205** 0.1083** −0.0262 −0.0831** 0.2400** 0.1250** 0.0143 0.4773** 0.2594** 1.0000
committee meetings
(NCM)
Proportion of independent −0.0531** −0.0488** −0.1291** 0.2109** 0.2154** 0.2707** −0.1173 −0.2312** 0.2023** 0.2924** 0.0279 0.2541** 0.2645** 0.6218** 1.0000
directors on NC (PNI)
Big Four audit firms (BIG 4) 0.0122 0.0041 −0.2267** 0.2329** −0.0636* 0.0828** −0.0331 −0.0341 0.1316** 0.1019** 0.0018 0.1170** 0.1537** 0.0957** 0.1703** 1.0000
Proportion of non-audit −0.0448 −0.0188 −0.0708** 0.2166** 0.1048** 0.0943** 0.0199 −0.0880** 0.1073** 0.1042** 0.0507 0.1292** 0.1356** 0.0956** 0.1160** 0.2211** 1.0000
service fees (PNAF)
** Significant at the 5% level; *Significant at the 10% level.

DSH: director shareholdings, AC: audit committee, RC: remuneration committee, NC: nomination committee, ISH: insider shareholders, PBI: proportion of independent directors in the board
Azim 13
5. Analysis of the results

Initially models were run to test the effect of monitoring variables on the performance of the full
dataset (1410 company-years). The models were run for each year separately, which assisted in
comparing the changes, if any. As a test of robustness, the model was then run to test whether there
was any variation in the result when considering one-, two- and three-year lagged performance for
these companies throughout the sample period. The same dataset was run using the Ordinary Least
Square (OLS) method and the results were compared to SEM. Data for the year 2010 were also
collected to observe the effect after the GFC.
5.1. Fit of the structural models

The overall goodness-of-fit for structural equation models depends on many factors. No single
value for the fit indices separates good from poor models, and it is not practical to apply a single
set of cut-off rules to all measurement models and, for that matter, to all structural equation models
of any type (Hair et al., 2009). The application of multiple fit measures enables a consensus across
types of measures regarding acceptability of the proposed model to be attained. Various goodness-
of-fit measures assess the results from three perspectives: (i) absolute fit measure; (ii) incremental
fit measures; and (iii) parsimonious fit measure.
5.1.1. Absolute fit measures. Absolute fit measures are used to understand the degree to which the
overall model (structural and measurement models) predicts the observed covariance or correlation
matrix. For this reason, this study used two commonly used absolute fit measures: the chi-square
statistic (χ2) and root mean square error of approximation (RMSEA). The chi-square (χ2) statistic
determines the independence or relatedness of the variables and strength of their relationship with
the model. This research finds that χ2 ranges from 234.11 to 297.03 for 2004, 2005 and 2006.
These results are consistent for all of the models, which suggest monitoring variables are strongly
interrelated. Again, in this study RMSEA ranges from 0.044 to 0.054, results that are consistent
with Browne and Cudeck’s (1993) suggestion of the value being around 0.08 or less as an indica-
tion of model fit (Table 4).
5.1.2. Incremental fit measure. Incremental fit assesses how well a specified model fits relative to
some alternative baseline model. To measure incremental fit, the following indices are used:
adjusted goodness-of-fit index (AGFI), comparative fit index (CFI) and normal fit index (NFI).
AGFIs range from 0.902 to 0.963. For all models, a recommended acceptable level is a value
greater than or equal to 0.90 (Hu and Bentler, 1999). The CFI has been found to be more appropri-
ate in a model’s development. This research found the CFI value ranges from 0.942 to 0.963. This
is consistent with the findings of Hair et al. (2009), who discovered that values fall between 0 and
1, with higher values indicating superior goodness-of-fit. The NFI, which is a relative comparison
of the proposed model to the null model, indicates results ranging from 0.0897 to 0.937. Hair et al.
(2009) showed that the NFI ranges between 0 and 1 with a cut-off value of 0.90, where 0 indicates
no fit and 1 indicates perfect fit. Considering this standard, these results indicate that the models
are near to perfect fit.
5.1.3. Parsimonious fit measures. Parsimonious fit provides information about which model is best.
Conceptually, parsimonious fit indices are similar to the notion of an adjusted R2 in the sense that
they relate to model fitness (Hu and Bentler, 1999). One of the measures of parsimonious fit is
normed chi-square (χ2/df). The value of normed chi-square (χ2/df) ranges from 1.888 to 2.395. As

suggested by Carmines and Mclver (1981), normed chi-square ranges from 2 to 1 or 3 to 1, indicat-
ing an acceptable fit between the hypothetical model and sample data.
Table 4. Model fitness
Model χ2 (df) ∆χ2 (∆ df) RMSES AGFI CFI NFI

2004
Return on equity 267.367 2.156 .049 .911 .960 .929
Return on assets 274.206 2.211 .051 .910 .958 .927
Price–earnings ratio 267.438 2.157 .049 .912 .960 .929
Market-to-book value 268.517 2.165 .050 .911 .960 .928
Dividend yield 278.258 2.244 .051 .909 .957 .926
2005
Return on equity 291.620 2.352 .054 .903 .962 .936
Return on assets 283.807 2.289 .052 .906 .963 .937
Dividend yield 295.637 2.384 .054 .902 .961 .935
2006
Return on equity 239.456 1.931 .045 .923 .950 .904
Return on assets 257.017 2.073 .048 .918 .943 .897
Dividend yield 239.337 1.930 .045 .922 .950 .904
χ2 (df): chi-square, AGFI: adjusted goodness-of-fit index (acceptable limit ≥ .90), ∆χ2 (∆ df): normed chi-square (accept-
able limit 1–5; 1 = best fit, 5 = reasonable fit), CFI: comparative fit index (0 = no fit at all, 1 = perfect fit), RMSES = root
mean square (.05 or less indicate a close fit), NFI: normal fit index (0 = no fit at all, 1 = perfect fit).
Source: Hair et al. (2009).
5.2. Substitute or complementary effect of monitoring

There are three groups of monitoring variables in the model: shareholders, board of directors and
external auditors. For all models, the relationships are significant among shareholders, external
auditors and board-monitoring variables for all the identified paths. The research question is whether,
in the Australian context, a substitution or complementary effect exists among the monitoring vari-
ables. Prior studies point to the existence of such effects, although country-specific institutional
factors (such as regulation, labour force market, stakeholders’ expectations) that shape monitoring
variables and their effectiveness make the generalisability of findings from these prior studies open
to question.
Various monitoring mechanisms have strong links and it is important to consider these link-
ages between corporate governance variables and how their relationships are structured. It is
expected that, in practice, various monitoring mechanisms are set up to operate jointly in order to
resolve the shareholder–manager agency problem. For example, if it is accepted that the board of
directors, through the audit committee, monitors the role of external auditors, this means the
board influences the quality of the auditor function. Similarly, shareholder concentration appears
to influence auditors with respect to their impact on performance. When there are concentrated

Azim 15
shareholdings, shareholders will keep pressure on auditors by acting as a proper monitor of

reported performance.
5.2.1. Shareholder and board of directors monitoring. Of all the accounting and hybrid measurement
models, Table 5 shows that there is a significantly negative correlation concerning 2004 (p = .001),
2005 (p = .015) and 2006 (p = .000) models for shareholder monitoring and monitoring by the
board of directors. This indicates that shareholder monitoring can substitute for board monitoring.
Agency theory emphasises the board’s role in monitoring the activities of management and repre-
senting the interests of shareholders. In this respect it can be argued that greater shareholder moni-
toring is a substitute for weaker board monitoring or vice versa.
Table 5. Substitution effect between shareholders and board monitoring (sample size 1410 company-
years)
ROE ROA PER MBV DY

Shareholders and board −0.053 −0.057 −0.058 −0.051 −0.059
of directors (2004) (0.001) *** (0.001) *** (0.001) *** (0.001) *** (0.001) ***
of directors (2005) (0.014) ** (0.015) ** (0.015) ** (0.015) ** (0.015) **
of directors (2006) (0.000) *** (0.000) *** (0.000) *** (0.000) *** (0.000) ***
***Significant at the 1% level; ** Significant at the 5% level.
ROE: return on equity, ROA: return on assets, PER: price–earnings ratio, MBV: market-to-book ratio, DY: dividend yield
5.2.2. Board of directors and auditor. The next set of findings is a complementary relationship for all
models from 2004 to 2006 between board of directors monitoring and monitoring by auditors. This
complementary effect is indicated by the significantly positive correlations in Table 6. The audi-
tor’s function is confirmed as a value-adding activity for the board-monitoring process. An impor-
tant role of the board of directors is to oversee the financial information provided in the annual
report. To do this the board engages auditors to assure the quality of the published financial reports
prepared by management. Hence, a higher quality auditor complements a more independent and
active board.
The findings of this study are consistent with that of Beasley and Petroni (2001), who noted that
independent boards are more likely to hire a high-quality, industry specialist audit firm for quality
Table 6. Complementary effect between board and auditor monitoring (sample size 1410 company-
years)
ROE ROA PER MBV DY

Board of directors and 0.162 0.161 0.160 0.158 0.156
auditors (2004) (0.000) *** (0.000) *** (0.000) *** (0.000) *** (0.000) ***
auditors (2005) (0.000) *** (0.000) *** (0.000) *** (0.000) *** (0.000) ***
auditors (2006) (0.000) *** (0.000) *** (0.000) *** (0.000) *** (0.000) ***
***Significant at the 1% level.

monitoring. Until recently, boards of directors tended to select audit firms. This role has increas-
ingly been taken over by an audit committee that is a committee of the board (Gay and Simnett,
2006). Therefore, both board and auditor monitoring is essential in the monitoring process and they
cannot substitute for each other.
5.2.3. Shareholders and auditor monitoring. The external auditing function is mandatory for all listed
companies. Therefore, it is supplied for the advantage of all shareholders. However, the question
to be addressed relates to the quality of auditor monitoring of reported corporate financial informa-
tion and the extent to which the shareholders depend on the auditor’s role. If the shareholders have
sufficient privilege or influence in accessing the monitoring and control functions of the compa-
ny’s management, they can become less dependent on the auditor’s monitoring. This influence is
shown in Table 7, where a significantly inverse relationship between shareholder monitoring and
auditor monitoring is demonstrated.
This result indicates that shareholder monitoring can be a substitute for auditor monitoring.
Shareholder monitoring is measured by insider shareholder/managers and block shareholder own-
ership. If shareholdings by managers and block shareholders’ ownership levels are high, they tend
to become less dependent on auditor monitoring. Therefore, shareholder monitoring can be a sub-
stitute for auditor monitoring.
It is likely the complementary/substitution effects within one company will vary little over time.
Yet in Table 5 the years 2004 and 2006 appear very similar, but the coefficients in 2005 appear
much lower; this pattern does not appear to be repeated in Tables 6 and 7. This coefficient explains
that the substitution effect between shareholders and board monitoring was lower. The explanation
for this may lie in the CLERP 9 amendments to the Corporations Act, which took effect on 1 July
2004 (reflected in financial reports published in 2005) and introduced substantial changes to the
Australian corporate reporting and disclosure landscape. In 2005, instead of depending on share-
holder or board monitoring, the focus shifted towards reporting and disclosure.
Table 7. Substitution effect between shareholder and auditor monitoring (sample size 1410 company-
years)
ROE ROA PER MBV DY

Shareholder and −0.013 −0.011 −0.013 −0.013 −0.014
auditor monitoring (0.000) *** (0.000) *** (0.000) *** (0.000) *** (0.000) ***
(2004)
Shareholder and −0.016 −0.015 −0.016 −0.013 −0.013
(2005)
Shareholder and −0.021 −0.023 −0.025 −0.025 −0.025
(2006)
***Significant at the 1% level.
5.3. Robustness tests

This research finds a similar substitution and complementary relationship among corporate moni-
toring mechanisms when examining the relationship between board, shareholders and auditors
monitoring on one-, two- and three-year lagged performance. All the models indicate that there is

Azim 17
a substitution effect existing between shareholder monitoring with board monitoring and auditor
monitoring, and a complementary relationship between board and auditor monitoring. Substitution
and complementary relationships among monitoring mechanisms may explain why previous stud-
ies failed to find any consistent results when examining the effect of monitoring on company
performance.
Given that most readers would be more familiar with regressions models, analysis of OLS was
done using the same data set (2004–2006) and finding which statistical model (OLS or SEM) bet-
ter explains the relationship between corporate governance mechanisms and company perfor-
mance. The OLS model results in general show a poor fit and the coefficients on corporate
board-monitoring variables are in general not significant. Results for the current-year (Table 8)
model using 2005 data show the following: multiple R2 = 1.5–9.7%; adjusted R2 = 1.2–8.5%; and
ANOVA: F = 1.0640 (P = .000) to F = 8.151 (P = .000). Results for the lagged-year (Table 9)
model (monitoring data of 2005 and performance data of 2006) show the following: multiple
Table 8. Results from Ordinary Least Square models using current-year monitoring (2005) and
performance (2005)
ROE ROA PER M/BV DY

(Intercept) −.859 (−4.995***) −.481 (−9.368***) −20.648 (−.618) 8.353 (7.646***) .020 (.872)
Top 1 −.859 (−1.510) −.001 (−2.332) .110 (.379) −.012 (−1.283) .000 (.519)
Top 19 .003 (1.889) .002 (3.880***) .222 (.830) .015 (1.731*) .000 (−1.679*)
ISH .002 (1.876) .001 (2.895**) −.214 (−1.071) .005 (.839) .000 (.757)
BSIZ −.011 (−.893) −.003 (−.846) −.060 (−.024) .166 (2.047**) −.004 (−2.085**)
BDM −.006 (−1.310) −.001 (−.364) .813 (.870) −.023 (−.764) .000 (−.739)
PBI −.092 (−.814) −.040 (−1.172) 23.951 (1.087) .795 (1.102) −.016 (−1.055)
BFL .003 (.220) .001 (.185) −1.324 (−.442) .156 (1.596) .000 (.233)
CHCE .089 (1.368) .068 (3.487***) 8.468 (.672) −.215 (−.520) .003 (.318)
ACM −.003 (−.248) −.001 (−.392) −1.091 (−.462) .049 (.630) 3.360 (.020)
PAI .042 (.651) .011 (.554) −5.081 (−.409) .334 (.821) .011 (1.302)
AFL .001 (.022) .010 (1.331) 3.636 (.753) −.037 (−.234) −.002 (−.474)
RCM .002 (.148) .000 (−.108) 4.437 (1.817*) .075 (.933) −.001 (−.329)
PRI .045 (.812) −.016 (−.988) −8.600 (−.801) −.297 (−.846) −.002 (−.238)
NCM −.004 (−.250) .002 (.329) −2.313 (−.683) −.047 (−.427) −.001 (−.453)
PNI −.073 (−.991) −.031 (−1.422) −6.544 (−.457) .838 (1.787*) .001 (.078)
BIG 4 .021 (.390) −.002 (−.132) ` 8.279 (.779) −.358 (−1.030) .013 (1.769*)
PNAF .021 (.254) −.007 (−.306) −4.923 (−.310) 1.571 (3.027***) −.009 (−.845)
SIZE .147 (5.060***) .074 (8.504***) −.808 (−.143) −1.610 (−8.734***) .011 (2.744**)
F-stat 2.743** 8.272** .640*** 5.872*** 1.139***
Multiple R2 .034 .097 .019 .071 .015
Adjusted R2 .022 .085 .015 .059 .012
***Significant at the 1% level; ** Significant at the 5% level; * Significant at the 10% level.
ROE: return on equity, ROA: return on assets, PER: price–earnings ratio, MBV: market-to-book ratio, DY: dividend yield,
ISH: insider shareholders, BSIZ: number of directors on the board, BDM: number of board meetings, PBI: proportion of
independent directors on the board, BFL: number of financially literate directors on the board, CHCE: CEO and chair of
the board; ACM: number of audit committee meetings, PAI: proportion of independent directors on the audit committee,
AFL: number of financially literate directors on the audit committee, RCM: number of remuneration committee meet-
ings, PRI: proportion of independent directors on the remuneration committee, NCM: number of nomination committee
meetings, PNI: proportion of independent directors of the nomination committee, BIG 4: audited by one of the Big Four,
PNAF: proportion of audit/non-audit fees, SIZE: size of the firm based on the log of total assets

Table 9. Results from Ordinary Least Square models using current-year monitoring (2005) and lagged-
year performance (2006)
ROE ROA PER M/BV DY

(Intercept) −.588 (−1.980*) −.587 (−4.489***) (−44.876) (−2.182**) 6.303 (8.198***) −.010 (−.849)
Top 1 −.001 (−.547) 2.345 (.021) .149 (.831) −.003 (−.458) −7.957 (−.813)
Top 19 .002 (.945) .001 (.947) .241 (1.463) .000 (.017) −5.375 (−.597)
ISH .003 (1.456) .001 (1.194) −.035 (−.288) .002 (.418) .000 (1.969**)
BSIZ −.003 (−.152) .008 (.854) 2.439 (1.600) .160 (2.812**) .000 (−.582)
BDM −.001 (−.072) .000 (.128) −.021 (−.037) −.007 (−.310) .000 (−.541)
PBI .085 (.436) −.015 (−.177) 21.953 (1.617) −.133 (−.261) −.003 (−.396)
BFL −.025 (−.934) −.011 (−.899) .051 (.028) .057 (.827) .001 (.758)
CHCE .074 (.659) .067 (1.365) −2.784 (−.358) .053 (.182) −.004 (−.860)
ACM .051 (2.448**) .002 (.235) −1.406 (−.966) −.030 (−.550) .000 (−.521)
PAI −.073 (−.662) .007 (.144) 1.703 (.222) .515 (1.799*) .005 (1.311)
AFL −.028 (−.644) .027 (1.449) .507 (.170) −.062 (−.557) .002 (1.008)
RCM .021 (.964) .005 (.503) −1.361 (−.905) .014 (.255) .000 (−.388)
PRI −.039 (−.405) −.078 (−1.855*) 2.858 (.432) .091 (.366) −.008 (−2.172**)
NCM .027 (.888) .002 (.126) .889 (.426) −.017 (−.220) −.001 (−.603)
PNI −.149 (−1.169) .006 (.106) −3.002 (−.340) .415 (1.257) .007 (1.426)
BIG 4 −.031 (−.329) −.018 (−.432) −10.315 (−1.577) −.195 (−.797) .002 (.463)
PNAF .032 (.224) −.064 (−1.033) −.065 (−.007) .891 (2.440**) −.011 (−2.028**)
SIZE .076 (1.521) .082 (3.698***) 4.480 (1.291) −1.012 (−7.79***) .008 (4.156***)
F-stat 1.160*** 2.262** 1.263*** 4.525*** 2.713***
Multiple R2 .015 .028 .019 .058 .034
Adjusted R2 .012 .016 .013 .030 .021
***Significant at the 1% level; ** Significant at the 5% level; * Significant at the 10% level.
ROE: return on equity, ROA: return on assets, PER: price–earnings ratio, MBV: market-to-book ratio, DY: dividend yield,
ISH: insider shareholders, BSIZ: number of directors on the board, BDM: number of board meetings, PBI: proportion of
independent directors on the board, BFL: number of financially literate directors on the board, CHCE: CEO and chair of
the board; ACM: number of audit committee meetings, PAI: proportion of independent directors on the audit committee,
AFL: number of financially literate directors on the audit committee, RCM: number of remuneration committee meet-
ings, PRI: proportion of independent directors on the remuneration committee, NCM: number of nomination committee
meetings, PNI: proportion of independent directors of the nomination committee, BIG 4: audited by one of the Big Four,
PNAF: proportion of audit/non-audit fees, SIZE: size of the firm based on the log of total assets
R2 = 1.5–7.2%; adjusted R2 = 1.2–6.0%; and ANOVA: F = 1.160 (P = .000) to F = 8.787 (P =

.000). When we compare the above results with SEM results, clearly the current models and
lagged models for the period show that coefficients are neither consistent nor significant. This
may be due to the existence of a substitute and complementary relationship among monitoring
mechanisms; alternatively, it may be explained by the poorly specified models, omitted variables,
there being no relationship, etc.
Again, despite the use of multiple performance measures, this paper loads them onto (i) one
single performance measure and (ii) two separate (accounting and market) measures. The results
in any case are consistent with multiple performance measures.
Again, a robustness test was done using data (top-200 companies) for the 2009–2010 financial
year. In Table 10, results of SEM show that there is a substitute and complementary relationship
among monitoring mechanisms. However, the effects of the GFC changed the coefficients (lower
than 2004–2006) of board, shareholders and auditor monitoring.

Azim 19
Table 10. Substitution/complementary effect on monitoring mechanisms (sample size 200 company-
years)
ROE ROA PER MBV DY

Shareholders and −0.073 (0.014) ** −0.073 (0.015) ** −0.075 (0.018) ** −0.073 (0.021) ** −0.072 (0.019) **
board of directors
(2010)
Board of directors 0.121 (0.028) ** 0.123 (0.029) ** 0.122 (0.038) ** 0.121 (0.033) ** 0.123 (0.034) **
and auditors
(2010)
Shareholders and −0.013 (0.025) ** −0.011 (0.022) ** −0.015 (0.021) ** −0.013 (0.022) ** −0.014 (0.023) **
auditor monitoring
(2010)
***Significant at the 1% level; ** Significant at the 5% level.

ROE: return on equity, ROA: return on assets, PER: price–earnings ratio, MBV: market-to-book ratio, DY: dividend
yield
It seems that the governing boards of listed companies seemed unable to prevent the risky and
ill-fated decisions that jeopardised their companies, devastated their investors and helped precipi-
tate a financial meltdown that morphed into a global recession. This furthermore had a negative
impact on investors’ trust in board monitoring. Similarly, after the GFC, shareholders became more
careful about their financial interest and auditors were aware of the GFC and become more con-
servative in their requirements.
6. Conclusion
This study developed and tested a conceptual model to better understand the complex relationships
between various monitoring mechanisms and their impact on company performance. As evident in
this paper, SEM is superior in identifying the cause–effect relationships among variables and the
existence of reverse causality. Very few studies have employed SEM to examine the inter-relation-
ships among the corporate governance variables. The prospect of changes to monitoring mecha-
nisms in response to performance deficiencies gives rise to simultaneity issues. Effective
combinations of corporate governance mechanisms would mean that a company can perform better
and satisfy its shareholders’ interests.
Results show a greater consistency across all the models when examining the substitution or
complementary effects between the combinations of (a) shareholders and board of directors moni-
toring; (b) board of directors and auditor monitoring; and (c) shareholder and auditor monitoring.
Any individual monitoring mechanism will be complemented to some extent by an alternative one.
This finding is consistent with the conclusions of Ward et al. (2009), Boo and Sharma (2008),
Fernández and Arrondo (2005), and Coles et al. (2001). Specifically, if governance mechanisms
can complement or substitute for each other, then no clear relationship could be established between
monitoring mechanisms (when considered independently of each other) and company perfor-
mance. This fact can explain why previous research has found inconsistent results. By confirming
these relationships, this study adds value to the body of literature on monitoring and its influence
on company performance.
Both the literature review and empirical findings of this research point to the practical reality
of diversity in corporate governance structures, activities and practices across countries and

corporations. Differences in business circumstances, competitive conditions in Australia’s capital

market compared to those of the USA and UK and differences in Australian corporate board struc-
ture compared to Japan and Germany give rise to the importance of developing a modified corpo-
rate governance system. It needs to be a system that best suits the operations of corporations in
their local, national or regional context.
A number of limitations influenced the results of this study, and these need to be addressed in
order to improve the integrity of future research in this area. Firstly, this study did not classify
block shareholders into different categories (e.g. institutional shareholders, government owner-
ship, family ownership, corporate ownership). This issue is important in accommodating the
varying motivations of block shareholders who want to retain control over the companies in
which they have invested. Secondly, two accounting bases and three hybrid measurements were
used in this study. This strategy may impede some important performance features that could be
obtained through other tools, such as stock market return. Therefore, this study may not accu-
rately report companies’ intrinsic business performance. Accounting measures of performance are
subject to accounting policy choice, while hybrid measures of performance are affected by market
inefficiencies. Thirdly, this research tests the attendance of directors at company meetings as a
form of monitoring. However, it does not consider the degree of their involvement, and whether
their business experience may have an impact on the quality of monitoring. Fourthly, the notion
that monitoring effectiveness can be observed in performance is largely a long-term view, and
short-term effects are likely to be weak. However, changes in mechanisms and conditions pre-
clude testing lag effects for arbitrarily long periods. Fifthly and finally, this research only uses
information concerning the top-500 companies. Other listed companies may have very different
monitoring mechanisms and these may exhibit different kinds of impacts on performance.
Therefore, the results may have varied if the remainder of the listed companies or the bottom-500
listed companies were sampled.
For future research, the model in this study could be expanded to include more alternative moni-
toring mechanisms that influence a company’s performance. For example, equity-based incentives,
such as granting of rights, bonus shares and share options to managers, are widely used to align the
interests of the principals (owners) and the agent (employees). Similarly, there are different control
mechanisms, such as those derived from the market, where managerial talent has the expertise and
integrity to reduce agency-related problems. Performance-based cash bonuses represent another
mechanism for encouraging the agent to work so that the company’s performance targets are
achieved. Furthermore, the interlocking of directors and auditors can influence monitoring proce-
dures and effectiveness. While this study did not consider the impact this would have on company
performance, the way is open for such a theme to be pursued in future research.
Funding
This research received no specific grant from any funding agency in the public, commercial or not-for-profit
sectors.
Acknowledgements
The author appreciates helpful comments from the participants at the British Accounting Association
Conference, 2009, Dundee, Scotland; and the Accounting and Finance Association of Australia and New
Zealand Annual Conference, 2010, Christchurch, New Zealand. The author is indebted to Professor Lee
Parker (UniSA), Professor Dennis Taylor (RMIT) and Professor John Dalrymple (Swinburne) for their help-
ful comments. The author would also like to acknowledge two anonymous reviewers for their helpful
comments.

Azim 21
Notes
1. In Australia, directors are classified into three categories: executive (insider), independent and grey.
2. Annual reports show that Pinnacle VRB Ltd (PCE) had 33 board meetings in 2004; Anaconda Nickel
Ltd (ANL) had 32 board meetings in 2005 and ERG Group Ltd. (ERG) had 37 board meetings in 2006.
They had separate committee meetings for their audit committee and remuneration committee/nomina-
tion committee.
References
Abbott LJ, Parker S and Peters GF (2004) Audit committee characteristics and restatements. Auditing: A
Journal of Practice and Theory 23: 69–87.
Agrawal A and Knoeber CR (1996) Firm performance and mechanisms to control agency problems between
managers and shareholders. Journal of Financial and Quantitative Analysis 31: 377–397.
Anderson RC, Mansi SA and Reeb DM (2004) Board characteristics, accounting report integrity, and the cost
of debt. Journal of Accounting and Economics 37: 315–342.
Australian Securities Exchange Corporate Governance Council (2007) Corporate Governance Principles and
Recommendations. Australian Securities Exchange, Sydney, NSW.
Bacidore JM, Boquist JA, Milbourn TTY, et al. (1997) The search for the best financial performance measure.
Financial Analysis Journal 53: 11–20.
Beasley MS (1996) An empirical investigation of the relation between board of director composition and
financial statement fraud. The Accounting Review 71(4): 443–465.
Beasley MS and Petroni KR (2001) Board independence and audit-firm type. Auditing: A Journal of Practice
and Theory 20: 97–114.
Bedard JC and Johnstone KM (2004) Earnings manipulation risk, corporate governance risk, and auditors’
planning and pricing decisions. The Accounting Review 79: 277–304.
Blue Ribbon Committee (1999) Available at: http://www.nasdaq.com/about/Blue_Ribbon_Panel.pdf
(accessed 10 August 2010).
Bollen K (1990) Overall fit in covariance structure models: Two types of sample size effects. Psychological
Bulletin 107: 256–259.
Boo E and Sharma D (2008) Effect of regulatory oversight on the association between internal governance
characteristics and audit fees. Accounting and Finance 48: 51–71.
Bosch H (1995) Corporate Practice and Conduct. Business Council of Australia. Melbourne: Pittman
Publishing.
Browne MW and Cudeck R (1993) Alternative ways of assessing model fit, cited in Bollen KA and Long
JS (Eds.) Testing Structural Equation Models: 136–162. Beverly Hills, CA: SAGE.
Bushman R, Chen Q, Engel E, et al. (2004) Financial accounting information, organizational complexity and
corporate governance systems. Journal of Accounting and Economics 37: 167–201.
Cadbury A (1992) Report of the Committee on the Financial Aspects of Corporate Governance. London:
Professional Publishing Ltd.
Carmines E and Mclver JM (1981) Analysing models with unobserved variables: Analysis of covariance
structures, cited in Bohrnstedt G and Borgatta E (Eds.), Social Measurement: Current Issues: 65–115.
Beverly Hills, CA: SAGE..
Chen KY and Zhou J (2007) Audit committee, board characteristics, and audit switch decisions by Andersen’s
clients. Contemporary Accounting Research 24: 1085–1117.
Chiang H-T (2005) An empirical study of corporate governance and corporate performance. Journal of
American Academy of Business 6: 95–101.
Chow CW (1982) The demand for external auditing: Size debt and ownership influence. The Accounting
Review LVII: 272–291.
Chowdhury D (2004) Incentives, Control and Development: Governance in Private and Public Sector with
Special Reference to Bangladesh. Dhaka, Bangladesh: Dhaka Viswavidyalay Prakashana Samstha,
University of Dhaka.

Cohen J, Krishnamoorthy G and Wright A (2002) Corporate governance and the audit process. Contemporary
Accounting Research 19: 573–592.
Coles JW, McWilliams VB and Sen N (2001) An examination of the relationship of governance mechanisms
to perform. Journal of Management 27: 23–50.
Commonwealth of Australia (2002) Corporate Disclosure: Strengthening the Financial Reporting
Framework. Corporate Law Economic Reform Program (CLERP 9). Canberra, ACT, Australia: Australian
Government Publishing Service.
CPA Australia (2012) Auditing and Assurance Handbook 2012. Sydney, NSW, Australia: Pearson
Education.
Craswell AT, Francis JR and Taylor SL (1995) Auditor brand name reputations and industry specializations.
Journal of Accounting and Economics 20: 297–322.
Daily CM, Johnson JL and Dalton DR (1999) On the measurements of board composition: poor consistency
and a serious mismatch of theory and operationalization. Decision Sciences 30: 83–106.
Davidson R, Goodwin-Stewart J and Kent P (2005) Internal governance structures and earnings management.
Accounting and Finance 45: 241–267.
DeAngelo LE (1981) Auditor size and audit quality. Journal of Accounting and Economics 3: 183–199.
Denis DK (2001) Twenty-five years of corporate governance research … and counting. Review of Financial
Economics 10: 191–212.
Denis DK and McConnell JJ (2003) International corporate governance. Journal of Financial and Quantitative
Analysis 38: 1–36.
DeZoort FT and Salterio S (2001) The effects of corporate governance experience, financial reporting and
audit knowledge on audit committee members’ judgments. Auditing: A Journal of Practice and Theory
20: 31–47.
Dye R (1993) Auditing standards, legal liability, and auditor wealth. Journal of Political Economy 101:
887–914.
Fama EF and Jensen MC (1983) Separation of control and ownership. Journal of Law and Economics 26:
301–325.
Fan X, Thompson B and Wang L (1999) Effects of sample size, estimation methods and model specification
on structural equation modelling fit indexes. Structural Equation Modelling 6: 56–83.
Farber D (2005) Restoring trust after fraud: Does corporate governance matter? The Accounting Review 80:
539–561.
Fernández C and Arrondo R (2005) Alternative internal controls as substitutes of the board of directors.
Corporate Governance: An International Review 13: 856–866.
Gay G and Simnett R (2006) Auditing and Assurance Services in Australia. New York: McGraw Hill Irwin.
Hair JF, Black WC, Babin BJ, et al. (2009) Multivariate Data Analysis. Upper Saddle River, NJ: Pearson
Education International.
Haniffa R and Hudaib M (2006) Corporate governance structures and performance of Malaysian listed com-
panies. Journal of Business Finance and Accounting 33: 1034–1062.
Hermalin BE and Weisbach MS (1991) The effects of board composition and direct incentives on firm per-
formance. Financial Management 20: 101–112.
Hu L and Bentler PM (1999) Cutoff criteria for fit indexes in covariances structure analysis: Covariance cri-
teria versus new alternatives. Structural Equation Modeling 6: 1–55.
Hughes R (1999) The rise and rise of the audit committee. Accountancy 123: 59.
Jarratt D (2000) Outshopping behaviour: An explanation of behaviour by shopper segment using structural
equation modeling. The International Review of Retail, Distribution and Consumer Behaviour 10: 257–304.
Jensen MC (1993) The modern industrial revolution, exit and the failure of internal control systems. Journal
of Finance 48: 831–880.
Jensen MC and Meckling W (1976) Theory of the firm: Managerial behavior, agency costs and ownership
structure. Journal of Financial Economics 3: 305–360.
Kang H, Cheng M and Gray SJ (2007) Corporate governance and board composition: Diversity and independ-
ence of Australian boards. Corporate Governance: An International Review 15: 194–207.

Azim 23
Karamanou I and Vafeas N (2005) The association between corporate boards, audit committees, and manage-
ment earnings forecasts: An empirical analysis. Journal of Accounting Research 43: 453–486.
Kelloway E (1998) Using LISREL for SEM: A Researchers Guide. London: SAGE.
Kent P and Stewart J (2008) Corporate governance and disclosures on the transition to International Financial
Reporting Standards. Accounting and Finance 48: 649–671.
Kiel GC and Nicholson GJ (2003) Board composition and corporate performance: How the Australian expe-
rience informs contrasting theories of corporate governance. Corporate Governance: An International
Review 11: 189–205.
Kim J-B, Chung R and Firth M (2003) Auditor conservatism, asymmetric monitoring, and earnings manage-
ment. Contemporary Accounting Research 20: 323–359.
Klein A (2002) Audit committee, board of directors characteristics, and earning management. Journal of
Accounting and Economics 33: 375–400.
Koh PS, LaPlante SK and Tong YH (2007) Accountability and value enhancement roles of corporate govern-
ance. Accounting and Finance 47: 305–333.
Krishnan GV (2003) Does Big 6 auditor industry expertise constrain earnings management? Accounting
Horizons 17: 1–16.
Leblanc RW (2004) What’s wrong with corporate governance: A note. Corporate Governance: An
International Review 12: 436–441.
Lin YF (2005) Corporate governance, leadership structure and CEO compensation: Evidence from Taiwan.
Corporate Governance 13: 824–835.
Livingston P (2002) Test your financial literacy. Directors and Boards 26: 21.
McConnell J and Servaes H (1990) Additional evidence on equity ownership and corporate value. Journal of
Financial Economics 27: 595–612.
Marcoulides G and Hershberger S (1999) Multivariate statistical methods: A first course. Structural Equation
Modelling 6: 127–133.
Martens MP (2005) The use of structural equation modelling in counselling psychology research. The
Counseling Psychologist 33: 269–296.
Menon K and Williams JD (1994) The use of audit committees for monitoring. Journal of Accounting and
Public Policy 13: 121–139.
Morck R, Shleifer A and Vishny R (1988) Management ownership and market valuation: An empirical analy-
sis. Journal of Financial Economics 20: 293–316.
Peasnell KV, Pope PF and Young S (2005) Board monitoring and earning management: Do outside directors
influence abnormal accruals? Journal of Business Finance and Accounting 32: 1311–1346.
Ramsay I (2001) Independent of Australian company auditors: Review of current Australian requirements
and proposals for reform. Report, October. The University of Melbourne, VIC, Australia.
Raykov T and Marcoulides G (1999) On desirability of parsimony in structural equation model selection.
Structural Equation Modelling 6: 292–300.
Rediker KJ and Seth A (1995) Boards of directors and substitution effects of alternative governance mecha-
nisms. Strategic Management Journal 16: 85–99.
Renders A and Gaeremynck A (2006) Corporate governance and performance: controlling for sample
selection bias and endogeneity. Working Paper. Katholieke Universiteit, Leuven, Belgium.
Sahnoun MH and Zarai MA (2009) Auditor-auditee negotiation outcome: Effects of auditee business risk,
audit risk, and auditor business risk in Tunisian context. Corporate Governance: An International Review
17: 559–572.
Sarens G, Beelde ID and Everaert P (2009) Internal audit: A comfort provider to the audit committee. The
British Accounting Review 41: 90–106.
Schultz EL, Tan DT and Walsh KD (2010) Endogeneity and the corporate governance – performance relation.
Australian Journal of Management 35: 145–163.
Shleifer A and Vishny RW (1986) Large shareholders and corporate control. The Journal of Political
Economy 94: 461–488.
Shleifer A and Vishny RW (1997) A survey of corporate governance. Journal of Finance 52: 69–82.

Standards Australia International (2003) AS 8000–2003, Australian Standard – Good Governance Principles.
Sydney, NSW, Australia: Standards Australia International.
Tam OK and Tan MGS (2007) Ownership, governance and firm performance in Malaysia. Corporate
Governance 15: 208–222.
Vafeas N (1999) Board meeting frequency and firm performance. Journal of Financial Economics 53:
113–143.
Ward AJ, Brown JA and Rodriguez D (2009) Governance bundles, firm performance, and the substitutability
and complementarity of governance mechanisms. Corporate Governance: An International Review 17:
646–660.
Welch E (2003) The relationship between ownership structure and performance in listed Australian compa-
nies. Australian Journal of Management 28: 287–305.
Williams RJ, Fadil PA and Armstrong RW (2005) Top management team tenure and corporate illegal activ-
ity: The moderating influence of board size. Journal of Managerial Issues 17: 479–493.
Wiwattanakantang Y (2001) Controlling shareholders and corporate value: Evidence from Thailand. Pacific-
Basin Finance Journal 9: 323–362.
Wooldridge MJ (2003) Introductory Econometrics: A Modern Approach. Mason, OH: Thomson.
Yatim P, Kent P and Clarkson P (2006) Governance structures, ethnicity, and audit fees of Malaysian listed
firms. Managerial Auditing Journal 21: 757–782.
Zhang P (2010) Board information and strategic tasks performance. Corporate Governance: An International
Review 18: 473–487.
Date of acceptance of final transcript: 10 May 2012.

Accepted by Associate Editor, Peter Clarkson (Accounting).
Appendix variables for Ordinary Least Square and structural

equation modelling
Major shareholders
Top 1: Share hold by the top shareholder
Top 19: Share hold by the next top-19 shareholders
ISH: insider (directors and senior management) shareholders
Board of directors and committees

BSIZ: number of directors on the board
BDM: number of board meetings
PBI: proportion of independent directors on the board
BFL: number of financially literate directors on the board
CHCE: CEO and chair of the board
ACM: number of audit committee meetings
PAI: proportion of independent directors on the audit committee
AFL: number of financially literate directors on the audit committee
RCM: number of remuneration committee meetings
PRI: proportion of independent directors on the remuneration committee
NCM: number of nomination committee meetings
PNI: proportion of independent directors on the nomination committee

Azim 25
External auditors:
BIG 4: audited by the Big Four
PNAF: proportion of audit/non-audit fees
Control variable:
SIZE: size of the firm based on log of total assets
Performance measures
ROE: return on equity
ROA: return on asset
PER: price–earnings ratio
MBV: market-to-book value
DY: dividend yield

Corporate Governance Mechanisms and Thei

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

Corporate Governance Mechanisms and Thei

Încărcat de

Drepturi de autor:

Formate disponibile

Australian Journal of Management

Corporate governance mechanisms and their impact on company performance: A

The online version of this article can be found at:

Email Alerts: http://aum.sagepub.com/cgi/alerts

>> OnlineFirst Version of Record - Aug 2, 2012

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

Australian Journal of Management

equation model analysis

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

2. Theoretical basis for model specification and causality

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

Shareholder Conflict of interest Management

Reduce Agency problem

Perfect Contracting Monitoring Incentives to Management

Market Internal Control Regulatory

Capital Managerial Managerial Regulatory Acts &

Block Shareholders Board of Directors and Committees External Auditors

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

2.1. Monitoring by shareholders

2.2. Monitoring by the board of directors

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

Remuneration committee. A board remuneration committee is an efficient mechanism for focus-

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

2.3. Monitoring by external auditors

2.4. Performance measures

3. Propositions and methodology

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

4. Data and company characteristics

Table 1. Total number of companies per year in this study

Top-500 companies in 2004

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

Monitoring variables Combined sample 2004 2005 2006

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

Top shareholder (TOP 1) 1.0000

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

** Significant at the 5% level; *Significant at the 10% level.

5. Analysis of the results

5.1. Fit of the structural models

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

Table 4. Model fitness

Model χ2 (df) ∆χ2 (∆ df) RMSES AGFI CFI NFI

5.2. Substitute or complementary effect of monitoring

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

shareholdings, shareholders will keep pressure on auditors by acting as a proper monitor of

ROE ROA PER MBV DY

ROE ROA PER MBV DY

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

ROE ROA PER MBV DY

5.3. Robustness tests

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

ROE ROA PER M/BV DY

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

ROE ROA PER M/BV DY

R2 = 1.5–7.2%; adjusted R2 = 1.2–6.0%; and ANOVA: F = 1.160 (P = .000) to F = 8.787 (P =

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

ROE ROA PER MBV DY

***Significant at the 1% level; ** Significant at the 5% level.

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

corporations. Differences in business circumstances, competitive conditions in Australia’s capital

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

Downloaded from aum.sagepub.com at Swinburne Univ of Technology on August 21, 2012

*Significant at the 1% level; Significant at the 5% level.