Audit Committee Effectiveness and Earnings Conservatism ... Web viewa sample of 100 Australian...
Transcript of Audit Committee Effectiveness and Earnings Conservatism ... Web viewa sample of 100 Australian...
Board of Director Effectiveness and Earnings Conservatism: Preliminary Australian Analysis
Abstract:The study’s overarching objective is to examine influence of board of director effectiveness on the extent of earnings conservatism amongst Australian listed firms. The study collected data from a sample of 100 Australian publicly listed firms stratified-randomly selected based on market capitalization to reduce size-biases. Data is collected for the 2008 financial year. Empirical results indicate more effective board of directors a more likely to incorporate negative news into earnings significantly faster (i.e., more timely) than less effective board of directors. This implies firms with more effective board of directors are more likely to report more reliable accounting information than firms with less effective board of directors. Additional analysis also shows that board of directors that are more independent, have independent members who are financially qualified and meet more frequently were more likely to be associated with disclosure of negative news in a more timely manner. However, results show board of directors composed of at least one independent director with prior committee experience were just as likely to report negative news in a similar time frame as board of directors without such experienced members.
1. Introduction and background
Conservatism is one of the principal foundations of financial accounting with a longstanding
and substantial influence.1 Empirical research indicates that conservatism has increased in the past
three decades (Balkrishna, Coulton, and Taylor, 2007; Givoly and Hayn, 2002; Givoly and Hayn,
2000; Huijgen and Lubberink, 2005; Krishnan and Visvanathan, 2007a). Despite the longevity and
reputed importance of conservatism, capital market regulators, accounting standard-setters and
scholars frequently criticize conservatism.2 The lengthy survival and resilience to criticism,
nonetheless, strongly suggests conservatism has significant benefits overlooked by critics.
Whilst there is a need to determine and understand the benefits of conservatism, there is also a
significant need to identify factors influencing conservatism. Identifying determinants of conservatism
is essential because though capital market regulators and accounting standard-setters may not directly
seek to eliminate conservatism, actions affecting the determinants of conservatism could have serious
indirect repercussions on financial reporting quality. The need to understand determinants of
conservatism has been amplified in the past decade following major corporate financial accounting
scandals. It has long been acknowledged corporate management have strong incentives to overstate
earnings (Fama, 1980; Jensen and Meckling, 1976). However, escalating pressures during the past
decade to meet investor and market expectations for higher revenues and earnings is thought to be
increasingly enticing corporate management to adopt more aggressive financial accounting practices.
Consequently, this undermines conservatism and the quality of earnings (LaFond and Watts, 2008).
A majority of prior research examining determinants of conservatism has concentrated on the
influence of institutional factors (Ball, Kothari, and Robin, 2000; Grambovas, Giner, and
Christodoulou, 2006; Lara, Osma, and Mora, 2005). Mounting attention to corporate governance
1 Conservatism is an open-ended phrase that is frequently used to refer to various concepts such as earnings conservatism and balance sheet conservatism. The author acknowledges the broad use of this phrase but within the specific context of this study, unless stated, conservatism is used purely from an earnings conservatism point of view. Other forms of conservatism, if used, will be prefixed by a leading word to describe the other form of conservatism.2 This is because the persistent asymmetric treatment of gains and losses is argued to lead to understatements in the current period that prompt the overstatement of earnings in future periods via the understatement of future expenditure.
1 | P a g e
issues in the past decade has spurned a new stream of investigation examining the impact of corporate
governance mechanisms on conservatism. To date, much attention in this new stream has concentrated
on ownership structure, the auditing process, and audit firms and fees (Ahmed and Duellman, 2007;
Beekes, Pope, and Young, 2004; Hamilton, Ruddock, Stokes, and Taylor, 2005; Lafond and
Roychowdhury, 2008; Ruddock, Taylor, and Taylor, 2006). Other key corporate governance
mechanisms, pivotal to the financial accounting process, however, may likely be of significance but as
yet have received little, if any, attention. The board of directors is one such feature.
Board of directors have been placed at the centre stage of the global debate on corporate
governance reforms since the early 1990s (Blue Ribbon Committee, 1999; Corporate Governance
Committee, 2001; New York Stock Exchange and National Association of Securities Dealers, 1999).
Whilst boards of directors are acknowledged as the ultimate corporate governance mechanism for
monitoring the financial reporting process, responsibility for the general and day-to-day oversight is
increasingly delegated to the board of directors. The introduction of new corporate governance
regulations such as the ASX Corporate Governance Council Recommendations (ASX 2003) and the
Corporations Law Economic Reform Programme 9 (CLERP 9) have served to increase the
responsibilities and influence of board of directors in the financial reporting process.
Given the unprecedented and ever increasing interest in earnings by investors, combined with
escalating demands for higher quality earnings, there is an urgent need to identify influential factors
and the resulting impact on earnings. As discussed above, both the conservatism principle and the
board of directors are alleged to have a significant bearing on earnings (Beasley and Salterio, 2001;
Klein, 2002a; Watts, 2003). Currently the influence of a board of directors and its effectiveness on
conservatism (and ultimately earnings quality) remains an open empirical question that has not been
formally investigated. A pivotal aim of this study is to address this void in the extant literature
surrounding conservatism and the board of directors.
Studying board of director effectiveness/earnings conservatism linkage is important and
timely. Presently, international capital markets (including Australia) are highly volatile, raising
questions about earnings quality. Understanding the board of director effectiveness/earnings
conservatism link, and the impact on earnings quality, assists regulators derive strategies to restore
credibility in earnings. More generically, corporate governance is continuously under reform. Studying
the earnings conservatism/board of director effectiveness linkage, therefore, is of value as results will
determine whether there is a likelihood any changes to regulations governing board of directors will
impact on conservatism and earnings quality. Considering the earnings conservatism/board of director
effectiveness linkage in the context of Australia is also of interest. For example, corporate governance
has been a hot-bed of debate with the introduction of regulations such as ASX 2003 and CLERP 9. The
effectiveness and impact of these corporate governance reforms is still a matter of debate. Findings
from this study will aid in shedding new light and understanding on the debate. Also, the study focuses
on Australia because of the (a) paucity of existing research examining the relationship between board
2 | P a g e
of directors and financial reporting quality in Australia, and (b) mixed empirical results to date using
data from this nation (Francis and Stokes, 1986; Goodwin, 2003; Stewart and Munro, 2007).
The primary objective of this study is to examine the association between board of director
effectiveness and earnings conservatism of Australian publicly listed firms. There is currently a lack of
consensus in the extant literature of a precise definition or factors determining board of director
effectiveness. Corporate governance advocates generally argue structural and operational composition
factors influence a board of director’s effectiveness. For example, in respect to structural composition,
Klein (2002a) argues more independent board of directors are effective in constraining managerial
opportunism, thereby, leading to greater quality of reported earnings. In the case of operational
composition, Abbott and Parker (2000a) and Abbott et al., (2003) state more diligent board of
directors (i.e., meet more frequently) are better able to address key financial reporting issues such as
earnings management and financial misstatement. In line with prior literature (Beasley and Salterio,
2001; Van der Zahn and Tower, 2004) this study focuses on four prime components underlying board
of director effectiveness: (a) sub-committee independence; (b) financial expertise; (c) experience; and
(d) diligence.
Overall, this study will benefit 1) firms in determining the optimal composition of board of
directors, 2) policymakers and investors by identifying the impact of board of director features; and 3)
regulators in determining the true role board of directors play in improving the level of earnings
quality reported by firms and therefore, the integrity of the financial reporting process. Finally,
findings from this study can be the basis of a more structured and systematic approach to board of
director formation and the ability of the board of director to uphold key regulatory reforms.
2. Literature Review and Hypotheses Development
Scholars have a lengthy history of supporting the importance of board of directors to the
financial accounting process including influencing earnings quality (e.g., Beasley and Salterio, 2001;
Pincus, Rusbarsky, and Wong, 1989). It is argued the central role of the board of director is to reduce
the magnitude of positive or negative abnormal accruals, thereby, enhancing earnings quality. The
board of director’s ability to accomplish any required and/or perceived roles and responsibilities in
monitoring the financial reporting process will depend primarily upon the sub-committee’s
effectiveness. Despite extensive debate there is little empirical research surrounding the issue of a
board of director’s effectiveness. Aside from studies focusing on earnings management, very few
empirical studies have sought to determine the influence of board of director effectiveness on other
aspects of financial accounting such as conservatism.
The conservatism concept is an effective governance mechanism. Watts (2003), for example,
suggests that the use of conservative accounting figures in contractual arrangements amongst various
parties associated with the firm reduces information asymmetry and moral hazard problems derived
from agency conflicts. A growing number of researchers (Ahmed and Duellman, 2007; Beekes et al.,
3 | P a g e
2004; Lara et al., 2005) argue other corporate governance mechanisms may provide an additional layer
of conservatism important in determining earnings quality. Actions by regulators affecting corporate
governance mechanisms, therefore, could inadvertently affect conservatism and earnings quality.
2.1 Theoretical Perspective
Since Jensen and Meckling’s (1976) seminal work developing a theory of the firm based on
conflicts between contracting parties, a vast body of literature has developed explaining the nature of
the conflicts and means by which conflicts may be resolved. This theoretical perspective, commonly
known as agency theory, is strongly linked to earnings conservatism and board of directors’ role and
responsibilities. Watts (1992), for example, suggests conservatism likely evolved from accounting’s
contracting role. Watts (1992) argues accounting conservatism helps avoid inappropriate distributions
to claim holders. Basu (1997) suggests the conservatism principle evolved with audited financial
statements as a means of management bonding that prevents management exploiting information
asymmetry (a premise of Fama's 1980, agency theory). Ahmed et al., (2002) support the conservatism
and contractual arrangements link (and, in turn, agency theory). Specifically, Ahmed et al.,(2002)
argue conservatism mitigates conflicts of interest over dividend policy between shareholders and
bondholders. Finally, Watts (2003) expresses the view earnings conservatism is a component of
efficient contracting that restricts management’s opportunistic behaviour in contracting purposes. In
the case of board of directors, agency theorists strongly argue the board of directors is a pivotal
governance mechanism to mitigate conflicts between contractual parties. Carcello and Neal (2000), for
example, suggest a board of directors plays an important monitoring role. Beasley (1996), for
example, argues a board of directors is an effective monitoring mechanism in assuring the quality of
financial reporting and corporate accountability. This view is supported by others such as Carcello and
Neal (2000), Klein (2002a) and Van der Zahn and Tower (2004). Given the close association of
conservatism and board of directorss to contractual arrangements and agency conflicts, agency theory
provides the study’s underlying theoretical perspective.
2.2 Earnings Conservatism
Various renowned scholars (Ball and Shivakumar, 2005; Francis, LaFond, Olsson, and
Schipper, 2005; Watts, 2003) consider earnings conservatism a central indicator of earnings quality or
a desirable property of accounting earnings. Therefore, studies have examined the different aspects of
conservatism including existence, time series pattern, and determinants of conservatism (Basu, 1995;
1997; Givoly and Hayn, 2000; Lubberink and Huijgen, 2001; Givoly and Hayn, 2002; Penman and
Zhang, 2002; Lobo and Zhou, 2006; Ruddock et al., 2006; Givoly et al., 2007; Krishnan, 2007;
Krishnan and Visvanathan, 2007a; Roychowdhury and Watts, 2007).
The existence of earnings conservatism is strongly supported by the seminal study by Basu
(1997). After the groundbreaking findings of Basu (1997), Pae, Thornton and Welker (2005) found
4 | P a g e
that earnings were conservative in all periods examined. Similarly, Balkrishna, Coulton and Taylor
(2007b) sought to provide evidence of earnings conservatism in the financial reporting practices of
Australian firms. Apart from providing evidence about the existence of earnings conservatism, some
studies (Basu, 1997; Givoly and Hayn, 2000; Givoly and Hayn, 2002; Huijgen and Lubberink, 2005;
Lobo and Zhou, 2006; Balkrishna, Coulton and Taylor, 2007a; Krishnan, 2007) sought to examine the
time-series pattern of earnings conservatism and concluded that earnings conservatism has increased
over time (Basu, 1997; Givoly and Hayn, 2000; Givoly and Hayn, 2002; Huijgen and Lubberink,
2005; Lobo and Zhou, 2006; Balkrishna et al., 2007a; Krishnan, 2007).
Prior research on earnings conservatism has also identified a number of determinants which
influence earnings conservatism (Ball et al., 2000a; Gigler and Hemmer, 2001; Easton and Pae, 2004;
Gassen et al., 2006a; O'Connell, 2006; Qiang, 2007a; Lafond and Roychowdhury, 2008). The
determinants can be classified into three broad categories: (1) firm characteristics (2) institutional
factors and (3) governance structures.
In relation to firm characteristics, Easton and Pae (2004) found evidence that change in cash
investment provided significant incremental explanatory power for returns over earnings and earnings
changes. Similarly, Pae et al. (2005) showed that earnings conservatism was substantially greater in
portfolios of firms with lower price-to-book ratios than in portfolios of firms with higher price-to-book
ratios. Kwon, Yin and Han (2006), further indicated that distribution of earnings and discretionary
accruals between the two groups of firms were consistent with a higher level of accounting
conservatism evident in high-tech firms vis-à-vis low-tech firms. In addition, Pae (2007) found that
conditional accounting conservatism was attributable mainly to unexpected accruals suggesting that
managers exercise discretion over accruals to expedite the recognition of bad news rather than good
news.
Prior research has concluded that institutional factors such as accounting polices and
regulation, legal regimes (i.e. code law versus common law), political influence and cross-listing of
firms in different jurisdictions have a significant impact on the level of earnings conservatism reported
by firms (Pope and Walker, 1999; Huijgen and Lubberink, 2005; Lara et al., 2005; Brown Jr. et al.,
2006; Bushman and Piotroski, 2006). Ball, Kothari and Robin’s (2000b) results indicated that
common-law accounting income did exhibit greater timeliness than code-law accounting income. Pope
and Walker (1999), for example, concluded that the US accounting regime was significantly more
conservative than the UK regime, when comparing earnings before extraordinary items. Similarly,
Huijgen and Lubberink (2005) found that earnings of UK cross-listed firms (in the US) were
significantly more conservative than earnings of UK firms without a US listing. In addition, Lara et al.
(2005) reported that in certain institutional contexts (that is, weaker investor protection and less
dispersed ownership structures), earnings management drove the measures of conservatism. Finally,
Bushman and Piotroski (2006) found that investor protections embodied in corporate law and the
5 | P a g e
efficiency and impartiality of the judicial system played a significant role in creating incentives for
timely loss recognition.
Recent studies highlight the link between earnings conservatism and corporate governance
structures. Beekes, Pope and Young (2004) found that firms with a higher proportion of outside
members are more likely to recognize bad news in earnings. Similarly, Ahmed and Duellman (2007)
discovered a negative relationship between the percentage of the inside directors on the board and
earnings conservatism. They (Ahmed and Duellman, 2007) also found a positive relationship between
the ownership percentage of outside directors and earnings conservatism. Similarly, Lafond and
Roychowdhury (2008) concluded that firms with lower managerial ownership report more
conservative earnings. Ruddock et al., (2006), meanwhile, stated that the provision of non-audit
services is positively associated with a reduction in the extent to which earnings reflect bad news on a
more timely basis than good news (i.e., earnings conservatism). Finally, Hamilton, Ruddock, Stokes
and Taylor (2005) found a significant increase in the asymmetrically timeliness of economic losses
when a firm changes an audit partner.
Of various corporate governance mechanisms discussed in the popular press and scholarly
research during the past decade, board of directors have been a central focus in debates about how to
assure and enhance the quality of financial reporting and corporate accounting. Despite the
acknowledged role of the board of director, studies have yet to consider the relationship to earnings
conservatism. Research does, however, provide evidence of the linkage between board of directors and
earnings quality. Various incentives exist to suggest why a board of directors is motivated to ensure
the credibility of the financial reporting process and that quality of earnings is preserved. For example,
independent directors on the board of directors have a strong incentive to ensure the sub-committee’s
roles and responsibilities are fulfilled so as to ensure their (i.e., the independent directors) reputational
capital and opportunities for appointment to other boards. As high quality earnings will likely be
perceived as a key benchmark of a board of director’s success and reputation (and that of its
members), this provides strong motivation for the sub-committee to undertake actions that enhance
earnings quality. As noted earlier, greater earnings conservatism is thought to enhance earnings
quality. Consequently, it follows that a board of directors will actively engage in conservative earnings
practices with the aim of improving earnings quality.
Despite incentives to engage in greater earnings conservatism, the mere presence of the board
of directors does not automatically mean such practices will be undertaken. Rather, as highlighted in
the extant literature, the effectiveness of the board of directors will determine whether the sub-
committee actively seeks to conserve earnings or not. A more effective board of directors will be
better able to mitigate opportunities for corporate management to engage in opportunistic behaviour
that can affect earnings quality. Furthermore, a more effective board of directors will have greater
ability to override aggressive financial accounting policy choices initiated by corporate management
that could promote less conservative earnings results. Finally, if a board of directors is more effective
6 | P a g e
in its arbitration role, this will aid in the development of more systematic compromises between
corporate management and external parties (such as the external auditor). Improved resolution of
conflicts, therefore, will likely enhance the acceptance of conservative earnings practices. Based on
the above discussion, therefore, the general hypothesis tested is postulated as:
GH: Australian publicly listed firms with more effective board of directors are more likely to have higher levels of earnings conservatism.
2.3 Components of Board of Director Effectiveness and Influence on Conservatism
There are four main categories of board of director effectiveness determinants are identified
as: (1) arrangement (i.e., sub-committee independence, size and duality); (2) resources (i.e., financial
expertise, committee experience); (3) authority (i.e., power enshrined in the sub-committee); and (4)
diligence (i.e., active meeting of sub-committee).3 The first two categories relate to structural
composition whilst the latter two to operational composition features. Whilst postulated above that
overall board of director effectiveness is positively associated with earnings conservatism, different
components underlying a board of director’s effectiveness could have differing degrees of influence.
This study is extended to examine this proposition with, hypotheses developed below in respect to four
key components (i.e., independence, financial expertise, prior experience and diligence) frequently
cited as central determinants of board of director effectiveness.
2.3.1 Board of Director Independence
Corporate governance advocates, regulators and scholars frequently argue a board of directors
with a higher proportion of outside directors is less likely to be compromised in undertaking the sub-
committees roles and responsibilities. Furthermore, a more independent board of directors is likely to
be better able to constrain opportunistic behavior of corporate management. Empirical findings have
generally supported the perception an independent board of directors is more effective in constraining
corporate management and improving earnings quality. Van der Zahn and Tower (2004), for example,
observed a significant positive relationship between earnings management and board of directors with
less than a majority of outside directors. Meanwhile, others have determined board of directors
comprising solely of non-related directors had a positive association with the quality of the firm’s
financial reports. McMullen and Raghunandan (1996) also found that firms with reporting problems
were less likely to have board of directors composed solely of outside directors. Similarly, Abbott and
Parker (2000b) concluded that firms with more independent directors were less likely to be sanctioned
by the SEC for fraudulent or misleading financial reporting compared to firms whose board of
directors did not comprise of independent directors. Additionally, it is reported fraudulent firms had
less independent board of directors than no-fraud industry benchmarks. Overall, prior theoretical and 3 There is currently no consensus on a definition of audit committee effectiveness. Rather, a number of determinants have been detailed in the extant literature that researchers propose affect the effectiveness of the audit committee. DeZoort et al.,(2002), for example, suggest that audit committee effectiveness occurs when audit committees comprise qualified members who have the authority and resources to protect stakeholder interests by ensuring reliable financial reporting, strong internal controls and comprehensive risk management practices through diligent oversight efforts.
7 | P a g e
empirical research provides a sound foundation to imply that independent board of director members
are more likely to (perhaps) require the recognition of bad news sooner and delay the reporting of
good news (i.e., actions that support earnings conservatism) in order to protect their reputational
capital, reduce contracting and mitigate litigation costs. Thus, more independent board of directors are
likely to support more actively actions promoting earnings conservatism. To test this assertion, the
following hypothesis is proposed:
H1: Australian publicly listed firms with a majority of independent directors on the board of director are more likely to have higher levels of earnings conservatism.
2.3.2 Board of Director Financial Expertise
As new corporate governance regulations continue to expand a board of directors’ role and
responsibilities, there is growing pressure on members to develop greater financial expertise to counter
the escalating complexity and sophistication of the financial reporting. Theoretical and applied views
have been forwarded in the extant literature suggesting greater financial expertise amongst board of
director members enhances effectiveness. McDaniel, Marint and Maines (2002) argue the presence of
a financial expert improves the quality of the firm’s financial statements. DeZoort, Hermanson and
Houston (2003) state financial experts provide greater resolve to support the external auditor during
auditor-management disagreements. Abbott, Parker and Peters (2004), meanwhile, argue greater
financial expertise is better in preventing occurrences of financial misstatements. Finally, Defond et
al., (2005) suggest greater financial expertise enhances the firm’s overall internal control. Empirical
research has generally supported the notion the board of directors’ effectiveness is enhanced with the
presence of a financial expert (or experts) as a member (e.g., Van der Zahn and Tower, 2004).
Therefore, the following hypothesis is proposed to test this assertion:
H2: Australian publicly listed firms with board of directors comprising members with financial expertise are more likely to have higher levels of earnings conservatism.
2.3.3 Board of Director Experience
Corporate governance experience of board of director members is cited by regulators,
corporate governance advocates and scholars as a major factor determining the board of directors’
effectiveness (Blue Ribbon Committee, 1999). Empirical evidence has supported these views. Dezoort
(1998), for example, observed experienced directors made more consistent judgements, had better
self-insight and higher technical content levels. He (Dezoort, 1998) concluded this signified
improvements in effectiveness. Research has found board of director members with greater corporate
governance experience were more likely to support the auditor in an auditor-management conflict, and
more likely to address and detect material misstatement. Again, experience is seen to have a positive
influence on board of director effectiveness. Finally, Carcello and Neal (2003) found directors who sat
on other board of directors (i.e. governance expertise) were more effective in protecting the auditor
from dismissal following the issuance of a going concern report. Based on theoretical arguments and
8 | P a g e
empirical findings, it is expected board of directors with more experienced members will be more
effective in ensuring firms adopt appropriate conservative earnings practices relative to firms with less
experienced board of director members. The following proposed hypothesis tests this assertion:
H3: Australian publicly listed firms with board of directors comprising members with prior governance experience are more likely to have higher levels of earnings conservatism.
2.3.4 Board of Director Diligence
It is strongly argued in the extant literature (Abbott et al., 2004; Bedard et al., 2004; Menon
and Williams, 1994; Xie, Davidson, and DaDalt, 2003) that an effective board of directors meets
regularly; thus, is more capable of ensuring the financial reporting process is functioning properly. A
more active board of directors is thought to be able to detect and prevent opportunistic behaviour by
management and to ensure the integrity of reported earnings. Past research indicates the board of
directors of fraud firms met less often than board of directors in firms not experiencing fraud.
Similarly, firms with board of directors meeting at least biannually were less likely to be sanctioned by
the SEC for financial reporting problems. Others such as McMullen and Raghunandan (1996), Abbott
et al., (2004) and Vafeas (2005) find firms with reporting problems had less frequent meetings. Xie et
al., (2003), meanwhile, find the number of board of director meetings is negatively associated with
discretionary current accruals implying diligence is an important factor in constraining the
management’s propensity to manage earnings. Finally, Krishnan and Visvanathan (2007b) show that
board of directors meeting more regularly were more likely to detect internal control weaknesses.
Prior literature considering board of director diligence literature clearly demonstrates the
importance of having sufficient board of director meetings per year and its impact on the financial
reporting process. Based on the prior empirical findings, therefore, it is likely more diligence board of
directors will be better able to ensure the adoption of conservative earnings practices to ensure greater
earnings quality. To formally test this assertion, the following hypothesis is forwarded:
H4: Australian publicly listed firms with more diligent board of directors (i.e., meet more frequently) are more likely to have higher levels of earnings conservatism.
3. Sample and Method
3.1 Selection and documentation
Data for this study is obtained from a number of sources. Data for the earnings conservatism
measure is obtained from two different databases. Accounting data is collected from FinAnalysis
whilst stock return data is obtained from DataStream. Data for the main independent variable, board
of director effectiveness (i.e. AC) comprising four primary sub-components (e.g., board of director
independence, financial expertise, experience and diligence), is hand collected the 2008 financial year
annual report of firms included in the sample.4 The annual reports were obtained from the Annual 4 In financial (or fiscal) year in Australia runs from July 1 to June 30 of the following calendar year. For ease of administration this study sought to align the financial year as close as possible with the calendar year. Consequently, a firm’s annual report is defined to be from the 2008 financial year if the firm’s end of financial year is between April 1, 2008 and March 31, 2009. Of the final useable sample on nine [9] firms had financial year ending on dates other than June 30, 2008. Eight [8] firms had December 31, 2008 year ends and one [1] firm had an August 31, 2008 year end.
9 | P a g e
Reports Collection (Connect 4 Pty Ltd). Where annual reports were not available from the Annual
Reports Collection (Connect 4 Pty Ltd) database the documentation was from DatAnalysis or direct
from the firm.
Analysis involves an examination of a sample of 100 stratified-randomly firms listed on the
Australian Stock Exchange (ASX) during the 2008 calendar year.5 The initial sample comprises all
publicly listed firms listed on the ASX as at January 1, 2008. Consistent with prior research, financial,
banking, insurance and utilities are excluded from the sample (Ball et al., 2000a; Givoly and Hayn,
2000; Goodwin, 2003; Ruddock et al., 2006; Givoly et al., 2007). Firms not continuously listed on the
ASX during the observation period (specifically IPO firms and firms de-listed for a period of time and
re-listed) are also excluded in order to avoid undue influences of unexpected share price changes.
Consistent with Clifford and Evans (1997), foreign firms domiciled outside Australia are also
excluded since their (foreign firms domiciled outside Australia) financial statements are not always
prepared in accordance with the normal disclosure requirements for other firms listed on the ASX.
Firms that have an indication of missing data during the observation period are also excluded (Klein,
2002b) leaving a sample pool of 1,628 firms.
From the resulting sample pool of 1,628 firms, 100 are randomly selected using a stratified-
random approach which involves stratifying each year into quartiles by market capitalization and
randomly selecting a sample of 25 firms within each quartile (Balvers et al., 1990). 6 However, from
the final sample of 100 firms, 6 firms are further excluded given incomplete information leaving a
final usable sample of 94 firms. Table 1 Panel B presents the industry breakdown of the sample firms.
Industrials and materials collectively represent the highest proportion of final sample firms by
industry. This representation is proportionally representative of the market as a whole. Thus, each
industry contains enough observations to control for the industry effects in the regression.
[Insert Table 1 About Here]
3.2 Measurement of Earnings Conservatism
Prior research suggests earnings conservatism can be conceptualized in several different ways.
Three dominant views detailed in the literature perceive earnings conservatism as being a function of
either: (a) timeliness, (b) persistence or (c) differences in current period accruals and cash flows (Basu,
1997; Ruddock et al., 2006; Watts, 2003). A comprehensive analysis of earnings conservatism and the
three major perceptions of this concept such be undertaken to develop a more definitive picture. Such
5 Results reported in this study are preliminary results for a larger longitudinal study. The longitudinal study covers a five calendar year period (January 1 2004 and December 31 2008). The five-year period is selected to minimize any significant extraneous influences on findings as a result of fallout from the ‘Dot.Com Bubble’, the introduction of new International Financial Reporting Standards ( IFRS) and excessive volatility due to the 2009 Global Financial Crisis. The period for the broader longitudinal study is also selected as it transcends the introduction of key corporate governance reforms in Australia (i.e., CLERP 9 and ASX 2003). Findings from this study may indicate whether recommendations related to audit committees in CLERP 9 and ASX 2003 impact the audit committee effectiveness/earnings conservatism linkage. Due to time constraints the current paper focuses on only a single year. In future versions the full study analysis will be reported.6 Since one of the major drivers of firm performance is the need to maximize shareholder value, this measure is best reflected by the firm’s market capitalization. For this study firms will be the same for each of the calendar years examined. This raises a possible independence issue. However, it is not considered detrimental to the study because: (a) it only applies to the longitudinal OLS regression models to be used and (b) almost all of the past literature (in both accounting and finance fields) using firm-year observations for multivariate testing accept the fact that independence of samples may be of concern but that there is no other parsimonious way to undertake such length-of-time analysis where the changes in selected firm’s results are of interest to the researcher/s.
10 | P a g e
analysis is undertaken in additional work not reported in this current paper due to time constraints.
Hence, for purposes of this initial exploratory study analysis focuses solely on the timeliness of
conservatism. To measure the timeliness of earnings conservatism the methodology developed by
Basu (1997) is used. Equations 1details the basic model underlying Basu’s (1997) measure of the
timeliness of earnings conservatism:
Xjt = α0 + α1DRjt + β0Rjt + β1DRjt*Rjt + εjt [1]Where:
Xjt = Operating profit after tax of firm j deflated by market value of equity (MVE t-1 - Market value of equity of firm j at beginning of the fiscal year t) of firm j at beginning of the fiscal year t;Rjt = Buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j;DRjt = Indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0);DRjt*Rjt = Two-way interaction between indicator variable where firm j is scored one (1) if Rjt
is negative, otherwise firm j is scored zero (0) and the buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j;αk, βk = Coefficients; andεjt = Error term.
Basu’s (1997) timeliness model of earnings conservatism (i.e., Equation 1) presumes losses are
recognized more quickly than gains so share prices reflect bad news in contemporaneous market losses
earlier than good news via market gains..
3.3 Measurement of Board of Director Effectiveness
Presently there is no consensus on a precise measure for board of director effectiveness.
Following prior work (Beasley and Salterio, 2001; Klein, 2002a, 2002b; Van der Zahn and Tower,
2004), this study develops a composite score for board of director effectiveness based on the sub-
committees independence, expertise, experience and diligence. Specifically, for firm j in time period t,
a composite score for board of director effectiveness (henceforth denoted BoDjt) is based on a score of
one (1) being awarded for each of the following individual characteristics being met:
The board of director of firm j in time t consists of a majority of independent directors; At least one independent board of director member of firm j in time period t is a qualified
person with accounting expertise possessing necessary educational qualifications (i.e., degree in accounting) and professional credentials (i.e., member professional accounting body) and/or work experience as a chief executive officer in a non-for-profit organization7;
At least one independent member of the board of director of firm j in time period t has prior experience on the board of director of another firm; and
The board of director of firm j met 4 or more times during the time period t.
7 In order to effective manage the finances of a non-for-profit organization a chief executive officer will need to be financially literate and component with a major understanding of financial accounting and auditing issues. Publicly listed firms may desire individuals with a non-for-profit background to fill the role of an independent director so as to enhance to concept of ‘independence’.
11 | P a g e
The range of scores for BoDjt, therefore, is zero (0) to four (4). For additional analysis, BoDjt is
decomposed into four measures representing each individual component. Specifically, the measure
for: (a) board of director independence is denoted as Ind_BoDjt; (b) board of director expertise is
denoted as Expt_BoDjt; (c) board of director experience is denoted as Exp_BoDjt; and (d) board of
director diligence as Dil_BoDjt. Each individual board of director effectiveness component metric is
scored as per the respective criteria outlined in scoring BoDjt. Data to calculate respective board of
director effectiveness measures is obtained from annual reports of firms selected in each firm year.
The calculation can be represented mathematically in Equation 2 as:
BoDjt. = ∑ (Ind_BoDjt + Expt_BoDjt + Exp_BoDjt + Dil_BoDjt) [2]Where:
BoDjt = Composite audit committee effectiveness score of firm j in fiscal year t;Ind_BoDjt = Indicator variable that takes the value of 1 if the majority of the board of directors of firm j in time period t are independent directors, and 0 otherwise;Expt_BoDjt = Indicator variable that takes the value of 1 if at least one independent director of the board of directors of firm j in time period t has necessary expertise (based on educational, professional affiliations and/or non-for-profit role) to be financially qualified, and 0 otherwise;Exp_BoDjt = Indicator variable that takes the value of 1 if at least one independent director of the board of directors of firm j in time period t has prior board of director experience, and 0 otherwise;Dig_BoDjt = Indicator variable that takes the value of 1 if the board of directors of firm j in time period t met four [4] or more times during time period t, and 0 otherwise;
3.4 Statistical Analysis
To formally test the general hypothesis, Equations 1 is extended to incorporate intercept and
slope coefficients for the interactive effects of a board of directors’ effectiveness. The test model is
defined by Equations 3:
Xjt = α0 + α1DRjt + α2BoDjt + α3DRjt*BoDjt + β0Rjt + β1Rjt*DRjt + β2Rjt*BoDjt + β3Rjt*DRjt*BoDjt
+ εjt [3]Where:
Xjt = Operating profit after tax of firm j deflated by market value of equity (MVE t-1 - Market value of equity of firm j at beginning of the fiscal year t) of firm j at beginning of the fiscal year t;DRjt = Indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0);BoDjt = Composite board of director effectiveness score of firm j in fiscal year t;DRjt*BoDjt = Two-way interaction between indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0), and the composite board of director effectiveness score of firm j in fiscal year t;Rjt = Buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j;Rjt*DRjt = Two-way interaction between the buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-
1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j and indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0);Rjt*BoDjt = Two-way interaction between the buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and
12 | P a g e
Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j and the composite board of director effectiveness score of firm j in fiscal year t;Rjt*DRjt*BoDjt = Three-way interaction between the buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j, indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0), and the composite board of director effectiveness score of firm j in fiscal year t; αk, βk = Coefficients; andεjt = Error term.
In determining the influence of board of director effectiveness (and individual components) on
earnings conservatism, the sign and significance of the coefficients on β2 and β3 are considered. If
higher board of director effectiveness is associated with firms reporting conservative earnings on a
timely basis, then the β2 and β3 coefficients in tests of Equation 3 should be positive and significant.
Positive and significant β2 and β3 coefficients would indicate greater asymmetric timeliness in the
recognition of good and bad news.
4. Main Findings
4.1 Descriptive statistics
Table 2 contains the descriptive statistics. For the pooled-sample Table 2 results indicate
average operating income (OPATjt) declined 62.277% during the 2008 financial year. Average and
median for 2007 and 2008 financial year operating income values (i.e., OPATjt-1 and OPATjt) for firms
in the 1st Quartile and 2nd Quartile (3rd Quartile and 4th Quartile) are positive (negative). Average
operating income for the 1st Quartile and 4th Quartile decreased year-on-year (i.e., 2007 financial year
compared to 2008 financial year) but interesting increased for firms in the 2nd Quartile and 3rd Quartile.
Declines in operating income can likely be attributed to the tough economic conditions that developed
during the 2008 calendar year that culminated in the Global Financial Crisis.8
Consistent with Lara et al. (2009), Table 2 shows that median value of deflated earnings (Xjt)
for the pooled-sample is greater than the mean (i.e., 0.003 compared to -0.032). This indicates
earnings for the pooled-sample is negatively skewed. However, when the pooled-sample is partitioned
by quartiles, earnings are negatively skewed (median values exceed mean values) for the 1 st Quartile
and 4th Quartile (i.e., largest and smallest firms by market capitalization). Earnings are positively
skewed (mean values exceed the median values) for 2nd Quartile and 3rd Quartile; this result which is
consistent with Ruddock et al. (2006).
8 The roots of the Global Financial Crisis and the associated global recession (though unlike the majority of developed and emerging economies worldwide Australia never entered recession) can be traced to October 2007 though the most severe ramifications (particularly in respect to stock and commodity markets) did not take hold till the 3 rd and 4th quarters of 2008. Economic growth (Gross Domestic Product) was steady during the 3rd and 4th quarters of 2007 (i.e., 4.2%) before decline in subsequent quarters in 2008 (i.e., 3.40% 1st Quarter, 3.00% 2nd
Quarter, 2.40% 3rd Quarter and 0.70% 4th Quarter). The decline in operating profits likely reflects the slowing growth of the global and domestic markets during 2008. The operating year for the majority of the final useable sample firms is from July 1, 2007 to June 30, 2008; hence, 3rd Quarter 2007 to 2nd Quarter 2008. Thus, whilst operating income decline on average, the annual results for the majority of firms were derived prior to the true impact of the Global Financial Crisis taking hold.
13 | P a g e
In line with the decline in average operating income, the average market capitalization of the
pooled-sample decline year-on-year (i.e., MarCapjt-1 and MarCapjt) by 17.112%. Declines in market
capitalization are across each quartile sub-sample. The decline is highest (least) amongst firms in the
1st Quartile (2nd Quartile). Relative to the market, firms of the final useable sample underperformed as
indicated by the negative sign on the average RR-All Ordjt value for the pooled-sample.9 Whilst firms
in the 2nd, 3rd and 4th Quartiles by market capitalization underperformed the general market, firms of
the 1st Quartile outperformed the market on average. Overall, 57 firm (or 60.638%) of the final useable
sample suffered negative market-adjusted 2008 financial year returns. Of firms in the 1 st Quartile only
11 (or 45.833%) had negative market-adjusted 2008 financial year returns whilst 18 (or 75.000%) of
those in the 4th Quartile had negative market-adjusted 2008 financial year returns. RR-All Ordjt and
DRjt results by quartile suggest smaller firms on the ASX were likely to have suffered greater stock
declines for the 2008 financial year than larger counterparts.
[Insert Table 2 About Here]
The average size of the board of directors of the pooled-sample (i.e., 5.470) is relatively (or
perhaps slightly lower) than reported in other countries such as the United States, United Kingdom and
Singapore. Consistent with prior research in Australia and abroad, the size of the board of directors
tends on average to be smaller when the size of the firm declines. Independent directors accounted on
average for 47.763% of the members of the board of directors of firms in the pooled-sample. Firms in
the 1st Quartile (4th Quartile) had the highest (lowest) average percentage of independent director
representation on the board of directors. In total 50 (or 53.191%) firms in the pooled-sample had board
of directors comprising a majority of independent directors (see Ind_BoDjt). Nearly all 1st Quartile
firms had board of directors with a majority of independent directors whilst only eight [8] firms in the
4th Quartile had a majority.
As argued above an independent director is deemed to have financial accounting expertise if
possessing necessary accounting education qualifications and professional affiliations, or is (has) been
employed as a chief executive officer of a non-for-profit organization. Table 2 results show
approximately a quarter (i.e., 24.439%) of the independent directors on the board of directors of
pooled-sample firms had accounting expertise based on accounting qualifications and professional
affiliations (see Expt_Acc_BoDjt). As expected firms in the 1st and 2nd Quartiles had the highest
percentage of a board of directors comprising independent directors with relevant financial accounting
qualifications and professional affiliations. Overall, 69 (or 73.404%) firms in the pooled-sample had
at least one independent director on the board of directors with financial expertise based on accounting
education and/or professional affiliation credentials (see Expt_Acc_BoDjt). In contrast, 32 (or
34.043%) of the pooled-sample had at least one independent member on the board of directors who 9 There is considerable debate in the literature on the appropriate market index to be used in calculating market-adjusted returns. Values reported in Table 2 are based on use of the All Ordinaries Index for purposes of market-adjustment. The All Ordinaries Index is a relatively broad industry based market measure but is generally comprised of larger firms. Returns using alternative market indices to compensate for size and industry biases were also calculated. Sensitivity tests (as reported in Table 3) were performed using the alternative return measures.
14 | P a g e
had prior or existing chief executive officer experience with a for-profit organization (see
Expt_CEONP_BoDjt). Firms in the 1st and 2nd Quartiles had the highest percentage of firms having at
least one independent director on the board of directors who had prior or existing chief executive
officer experience with a non-for-profit organization. In total 75 (or 79.787%) of the pooled-sample
are found to have at least one independent director on the board of directors who is deemed to be
financially qualified (see Expt_BoDjt).
Table 2 results indicate that on average 41.943% of the members of the board of directors of
pooled-sample firms are independent directors with prior board of director experience (see
%AC_Exp_BoDjt). As per other board of director characteristics and effectiveness traits, board of
directors of firms in the 4th Quartile had the lowest average of board of directors composed of
independent directors with prior board of director experience (i.e., 50.000%) with firms in the 3 rd
Quartile having the highest average (i.e., 100.000%). Overall, 62 (or 65.957%) of the pooled-sample
had at least one independent on the board of director with prior board of director experience (see
Exp_BoDjt).
The pooled-sample’s average number of board of director meetings (i.e., 10.710) is
comparable to that reported for US firms (e.g., Abbott et al., 2004) and those firms in neighbouring
Asian nations (e.g., Van der Zahn and Tower, 2004). Table 2 results indicate a distinctive division in
the number of board of director meetings held by larger and smaller firms. Specifically, 1 st and 2nd
Quartile firms met (on average) three times more during the year than counterparts in the 3 rd and 4th
Quartiles (i.e., 11.380 and 12.170 meetings compared to 9.320 and 9.880 meetings). Overall, 54 (or
57.447%) of the firms in the pooled-sample met ten [10] or more times during the year.
Finally, the pooled-sample’s average (median) board of director effectiveness score (i.e.,
BoDjt) is 2.330 (3.000). If using two [2] as an arbitrary benchmark for defining average effectiveness
then the majority of firms in the pooled-sample had above average effective board of directors.
However, as potentially expected, board of directors of larger firms scored higher on average for their
effectiveness. This result may be explained by larger firms having more resources available to ensure
board of directors have members that better contribute to the committee’s effective. Alternatively, in
line with agency theory, larger firms may have more at stake (e.g., reputation) than smaller firms.
Consequently, larger firms are more politically visible than smaller counterparts, and will have greater
incentive to ensure the effectiveness of the board of directors.
4.2 Regression results
Table 3 presents the initial results testing the association between earnings conservatism and
board of director effectiveness. Results in Table 3 Column I report the basic association between
earnings and annual returns without consideration for other compounding influences. The coefficient
on Rjt is positive and highly significant (p<0.01). This result is inconsistent with other related
Australian earnings conservatism research (e.g., Taylor and Taylor, 2003; Hamilton et al., 2005) but in
line with recent European studies (e.g., Lara et al., 2007; Kankaanpaa, 2008). The inconsistency with
15 | P a g e
prior Australian earnings conservatism could be the result of temporal differences in datasets or the
relatively low sample size in this study that may not provide an (as yet) comprehensive picture cross-
section of Australian firms. When additional explanatory factors are introduced into the model (see
Table 3 Column II and Column III), the coefficient on Rjt continues to be positive. The coefficient,
however, is only moderately significant (p<0.10) in Basu’s (1997) basic base model specification (see
Table 3 Column II) when the indicator variable DRjt and two-way interaction term Rjt*DRjt are
included.
[Insert Table 3 About Here]
In the test of Basu’s (1997) basic base model (see Table 3 Column II) and the study’s full
model (see Table 3 Column III) the coefficient on DRjt is positive and significant (p<0.10).
Meanwhile, the coefficient on the two-way interaction term Rjt*DRjt is negative and statistically
significant in tests of Basu’s (1997) basic base model (see Table 3 Column II) and the study’s full
model (see Table 3 Column III). Findings related to the two-way interaction term Rjt*DRjt is consistent
with prior research suggesting that existence of earnings conservatism across all firms in the sample.
The result also indicates that negative news is conveyed more swiftly to the market than good news
(Lara et al., 2007).
In Table 3 Column III Basu’s (1997) basic base model of the timeliness of earnings
conservatism is extended to include the board of director effectiveness measure and relevant
interaction terms. The coefficient on BoDjt is positive and significant (p<0.01) suggesting firms with
more effective board of directors were associated with better earnings. Meanwhile, the coefficient on
the two-way DRjt*BoDjt variable is negative and significant (p<0.05). One of the two variables of
prime interest to this study is defined by the two-way Rjt*BoDjt variable. Table 3 Column III results
indicate the coefficient on the Rjt*BoDjt variable is negative and significant (p<0.01). The significant
negative result implies that firms with a more effective board of directors appears to recognize
negative news faster than counterparts with less effective board of directors. With respect to the main
variable of interest – defined by the three-way Rjt* DRjt*BoDjt interaction variable – the coefficient is
positive and significant (p<0.01). The result related to the Rjt* DRjt*BoDjt interaction variable
reinforces the proposition firms with more effective board of directors report more conservative
earnings figures.
To further explore the association between board of director effectiveness and earnings
conservatism the main regression model (as defined by Equation 3) is performed again but with the
individual components of board of director effectiveness respective replacing the composite score.
Results from the additional tests are reported in Table 4.
[Insert Table 4 About Here]
Consistent with Table 3 results the coefficient on Rjt is positive and significant (p<0.01 or
p<0.10) in three of the four regressions reported in Table 4. Interestingly, the coefficient on Rjt is
positive but insignificant when testing for the association between the prior board of director
16 | P a g e
experience of independent directors on the committee and earnings conservatism (see Table 4 Column
III). Similar to the Rjt results, the coefficients on the DRjt and Rjt*DRjt variables are positive across the
four regressions reported in Table 4. However, the coefficients are only significant in regressions
testing the specific influence of the board of directors’ financial expertise of independent committee
members (see Table 4 Column II).
With respect to individual components of board of director effectiveness, the coefficients on
Ind_BoDjt, Expt_BoDjt and Exp_BoDjt are positive (Table 4 Column I, II and III). The coefficient on
Expt_BoDjt is the only one significant (Table 4 Column II; p<0.05). The coefficient on Dil_BoDjt,
meanwhile, is negative but insignificant from zero (see Table 4 Column IV). The findings suggest
those boards composed of independent directors with financial qualifications and prior experience is
associated with higher earnings. The number of meetings held by the board of directors, however, is
not associated with earnings.
Meanwhile, the coefficients on the two-way Rjt*Ind_BoDjt, Rjt*Expt_BoDjt and Rjt*Dil_BoDjt
interaction variables (see Table 4 Column I, II and IV) are negative but only significant for the first
two variables (p<0.01respectively). The coefficient on the two-way Rjt*Exp_BoDjt interaction variable
is positive but insignificant from zero (see Table 4 Column III). The findings suggest board of
directors that are more independent and had financial qualified independent directors as members were
likely to report negative news quicker than counterparts with less independent, expert and diligent
board of directors. Meanwhile, board of directors with more experienced independent members and
met more frequently are likely to recognize negative news just as readily as board of directors without
independent members with prior experience and that met less frequently.
Finally, in respect to the key three-way interaction variables (i.e., Rjt* DRjt*Ind_BoDjt, Rjt*
DRjt*Expt_BoDjt, Rjt* DRjt*Exp_BoDjt and Rjt* DRjt*Dil_BoDjt) results tabulated in Table 4 are similar
to the associated two-way interaction variables. Specifically, the coefficients of each three-way
interaction variable are positive. Meanwhile, the coefficients on Rjt* DRjt*Ind_BoDjt and Rjt*
DRjt*Expt_BoDjt are significant (see Table 4 Column I and II; p<0.01 respectively). The three-way
interaction results imply an earnings are more conservative amongst firms with board of directors that
are of greater independence and composed of independent directors with financial expertise.
4.3 Robustness Tests
Prior research highlights a number of contentious issue regarding tests of accounting conservatism that
could influence empirical findings. For example, there is lengthy debate in the extant literature about
difficulties in measuring market-adjusted returns. Many studies commonly use a single market-index.
However, researchers (e.g., Ahmad-Zaluki et al., 2007) argue listed firms that constitute the whole
market represent a variety of characteristics that cannot be easily captured by a narrower index. For
instance, many major market indices (e.g., Dow Jones Index, FTSE 100, DAX Index, SMI Index)
often only include well known large entities. Consequently, this may introduce size biases in the
17 | P a g e
calculation of market-adjusted returns. Following Ahmad-Zaluki et al., (2007), market-adjusted
returns are recalculated using three alternative indices in an effort to better control for biases such as
size and industry. First, market-adjusted returns are calculated using the ASX 200 Index. This is a
narrow index than the All Ordinaries Index. Second, to better control for size market-adjusted returns
for firms included in the: (a) 1st Quartile are based on the All Ordinaries Index; (b) 2nd Quartile are
based on the ASX 200 Index; (c) 3rd Quartile are based on the Mid-Cap 50 Index; and (d) 4th Quartile
based on the Small Ordinaries Index. Finally, with consideration for industry differences market-
adjusted returns calculated using an industry index relevant to the major industry sector of each firm.
Tests performed using alternative market-adjusted return calculations are reported in Table 5.
[Insert Table 5 About Here]
Table 5 results related to variables of primary interest to this study (i.e., Rjt*BoDjt and Rjt*
DRjt*BoDjt) are consistent with Table 3 results. Specifically, the coefficients on Rjt*BoDjt are negative
and significant (p<0.01) in each of the three regressions shown in Table 5. Meanwhile, the coefficients
Rjt* DRjt*BoDjt are positive and significant (p<0.01). Table 5 results imply that regardless of the index
used to measure the market-adjusted returns, more effective board of directors appear to recognize
negative news more swiftly than less effective board of directors, and also report more conservative
earnings numbers.
Whilst Table 5 results related to the main variables of interest are consistent with Table 3
results, it is important to note differences do arise in respect to other variables. For example, the
coefficient on DRjt is positive across the three regressions reported in Table 5. However, when market-
adjusted returns are controlled for size or industry biases the coefficients are not significant.
5. Conclusions
This study analyses the association between earnings conservatism and board of director
effectiveness. Accounting conservatism is reputed to produce earnings that reflect negative news faster
than good news. Earnings conservatism has been conceptualized in different manners such as
timeliness and persistence. This study focuses on the timeliness of earnings news using Basu’s (1997)
model. Board of director effectiveness is considered to be the composite outcome of interrelated
characteristics. Specifically, board of director effectiveness in this study is constructed based on the
concepts of independence, financial expertise, prior experience and diligence.
The study collected data from a sample of 100 Australian publicly listed firms stratified-
randomly selected based on market capitalization to reduce size-biases. Data is collected for the 2008
financial year. Consistent with expectations (see GH), empirical results indicate more effective board
of directors a more likely to incorporate negative news into earnings significantly faster (i.e., more
timely) than less effective board of directors. This implies firms with more effective board of directors
are more likely to report more reliable accounting information than firms with less effective board of
directors. Additional analysis also shows that board of directors that are more independent, have
18 | P a g e
independent members who are financially qualified and meet more frequently were more likely to be
associated with disclosure of negative news in a more timely manner. This finding is consistent with
expectations presented in H1, H2 and H4. However, results show board of directors composed of at
least one independent director with prior committee experience were just as likely to report negative
news in a similar time frame as board of directors without such experienced members. This finding is
contrary to expectations (see H3).
Our findings are consistent with evidence reported by others such as Ahmed and Duellmann
(2007), Beekes et al., (2004), Lara et al., (2007, 2009) and Lobo and Zhou (2006) investigating the
link between corporate governance and earnings conservatism. Findings of this study are of merit as it
focuses on a key component of corporate governance that has come under intense scrutiny during the
past decade. Results from this study aid in indicating the importance of corporate governance – and
specifically board of directors – to earnings conservatism and the quality of reported accounting
information. The results provide confidence in the robustness of the respective links by showing the
links present in a jurisdiction where the demand for less conservative accounting numbers may be less
due to reduced litigation risk.
While this study provides interesting results, it is acknowledged that findings are preliminary.
This study is part of a larger and broader analysis during which a more comprehensive data set
spanning a longer time period will be collected. Future research using the additional data will add in
supporting or refuting of evidence presented in this paper. Despite any limitations this paper does
provide grounds for additional research.
19 | P a g e
ReferencesAbbott, L. J., and Parker, S. (2000a). Audit committee characteristics and auditor selection. Auditing: A Journal
of Practice and Theory, 19(2), 47-66.Abbott, L. J., and Parker, S. (2000b). Auditor selection and audit committee characteristics. Auditing: A Journal
of Practice & Theory, 19(Fall), 47-66.Abbott, L. J., Parker, S., and Peters, G. (2004). Audit committee characteristics and restatements. Auditing: A
Journal of Practice and Theory, 23(1), 69-88.Abbott, L. J., Parker, S., Peters, G., and Raghunandan, K. (2003). The effect of audit committee characteristics
and audit fees. Auditing: A Journal of Practice and Theory, 22(2), 17-32.Ahmed, A. S., Billings, B. K., Morton, and Stanford-Harris, M. (2002). The role of accounting conservatism in
mitigating bondholder-shareholder conflict. The Accounting Review, 77(4), 867-890.Ahmed, A. S., and Duellman, S. (2007). Accounting conservatism and board of director characteristics: An
empirical analysis. Journal of Accounting and Economics, 43, 411-437.Balkrishna, H., Coulton, J. J., and Taylor, S. L. (2007). Accounting losses and earnings conservatism: Evidence
from Australian Generally Accepted Accounting Principles. Accounting & Finance, 47, 381-400.Ball, R., Kothari, S. P., and Robin, A. (2000). The effect of institutional factors on properties of accounting
earnings. Journal of Accounting and Economics, 29(February), 1-51.Ball, R., and Shivakumar, L. (2005). Earnings quality in UK private firms: comparative loss recognition
timeliness. Journal of Accounting and Economics, 39, 83-128.Ball, R., and Shivakumar, L. (2006). The role of accruals in asymmetrically timely gains and loss recognition.
Journal of Accounting Research, 44, 207-242.Balvers, R. J., Cosimano, T. F., and McDonald, B. (1990). Predicting stock returns in an efficient market. The
Journal of Finance, 45(4), 1109-1128.Basu, S. (1997). The conservatism principle and the asymmetric timeliness of earnings. Journal of Accounting
and Economics, 24(1), 3-37.Beasley, M. S. (1996). An empirical analysis of the relation between the board of director composition and
financial statement fraud. The Accounting Review, 71, 443-465.Beasley, M. S., Carcello, J. V., Hermanson, D. R., and Lapides, P. D. (2000). Fraudulent financial reporting:
Consideration of industry traits and corporate governance mechanisms. Accounting Horizons, 14(4), 441-454.
Beasley, M. S., and Salterio, S. (2001). The relationship between board characteristics and voluntary improvements in audit committee composition and experience. Contemporary Accounting Research, 18(4), 539-570.
Bedard, J., Chtourou, S. M., and Courteau, L. (2004). The effect of audit committee expertise, independence, and activity on aggressive earnings management. Auditing: A Journal of Practice & Theory, 23(2), 13-35.
Beekes, W., Pope, P., and Young, S. (2004). The link between earnings timeliness, earnings conservatism and board composition: Evidence from the UK. Corporate Governance, 12(1), 47-59.
Blue Ribbon Committee. (1999). The Report and Recommendations of the Blue Ribbon Committee on Improving the Effectiveness of Corporate Audit Committees. New York, New York, United States of America: New York Stock Exchange.
Carcello, J. V., and Neal, T. (2003). Audit committee characteristics and auditor dismissals following new going-concern reports. The Accounting Review, 78(January), 95-117.
Carcello, J. V., and Neal, T. L. (2000). Audit committee composition and auditor reporting. The Accounting Review, 75(4), 453-467.
Corporate Governance Committee. (2001). Report of the Committee and Code of Corporate Governance. Singapore: Corporate Governance Committee.
Dechow, P. M., Kothari, S. P., and Watts, R. L. (1998). The relation between earnings and cash flows. Journal of Accounting & Economics, 25(2), 133-169.
Dechow, P. M., Sloan, R. G., and Soliman, M. T. (2004). Implied equity duration: A new measure of equity risk. Review of Accounting Studies, 9(2-3), 197-212.
DeFond, M., Hann, R. N., and Hu, X. (2005). Does the market value financial expertise on audit committees of board of directors? Journal of Accounting Research, 43(2), 153-193.
Dezoort, F. T. (1998). An analysis of experience effects on audit committee members' oversight judgments. Accounting, Organizations and Society, 23(1), 1-21.
DeZoort, F. T., Hermanson, D. R., Archambeault, D. S., and Reed, S. A. (2002). Audit committee effectiveness: A synthesis of the empirical audit committee literature. Journal of Accounting Literature, 21, 38-75.
Dezoort, F. T., Hermanson, D. R., and Houston, R. W. (2003). Audit committee support for auditors: The effects of materiality justification and accounting precision. Journal of Accounting and Public Policy.
20 | P a g e
DeZoort, F. T., and Salterio, S. (2001). The effects of corporate governance experience and financial reporting and audit knowledge on audit committee members' judgments. Auditing: A Journal of Practice and Theory, 20(2), 31-47.
Fama, E. (1980). Agency problems and the theory of the firm. Journal of Political Economy, 288-307.Farber, D. B. (2005). Restoring trust after fraud: Does corporate governance matter? The Accounting Review,
80(2), 539-561.Francis, J., and Stokes, D. (1986). Audit prices, product differentiation and scale economies: Further evidence
from the Australian audit market. Journal of Accounting Research, 24(4), 383-393.Francis, J. R., LaFond, R., Olsson, P., and Schipper, K. (2005). The market pricing of accruals quality. Journal
of Accounting and Economics, 39, 295-327.Gamble, R. (2008). Crunch time for treasury financing. Treasury & Risk, May, 34.Gassen, J., Fulbier, R. U., and Sellhom, T. (2006). International differences in conditional conservatism - The
role of unconditional conservatism and income smoothing. European Accounting Review, 15(4), 527-564.
Givoly, D., and Hayn, C. (2002). Rising conservatism: Implications for financial analysts. Financial Analysts Journal, 58(1), 56-74.
Givoly, D., and Hayn, C. K. (2000). The changing time series properties of earnings, cash flows and accruals: Has financial reporting become more conservative? Journal of Accounting and Economics, 29(3), 287-320.
Givoly, D., Hayn, C. K., and Natarajan, A. (2007). Measuring reporting conservatism. The Accounting Review, 82(1), 65-106.
Goodwin, J. (2003). The relationship between the audit committee and the internal audit function: Evidence from Australia and New Zealand. International Journal of Auditing.
Grambovas, C. A., Giner, B., and Christodoulou, D. (2006). Earnings conservatism: Panel data evidence from the European Union and the United States. Abacus, 42(3/4), 354-378.
Gregoriou, A., and Skerratt, L. (2007). Does the Basu model really measure the conservatism of earnings? Working paper. Brunel Business School, UK.
Hamilton, J., Ruddock, C., Stokes, D., and Taylor, S. (2005). Audit partner rotation, earnings quality and earnings conservatism. Working Paper, University of Technology, Sydney.
Healy, P. M., and Wahlen, J. M. (1999). A review of the earnings management literature and its implications for standard setting. Accounting Horizons, 13(4), 365-384.
Huijgen, C., and Lubberink, M. (2005). Earnings conservatism, litigation and contracting: The case of cross-listed firms. Journal of Business Finance & Accounting, 32(7/8), 1275-1309.
Jensen, M. C., and Meckling, W. H. (1976). Theory of the firm: Managerial behaviour, agency costs and ownership structure. Journal of Financial Economics, 3(3), 305-360.
Klein, A. (2002a). Audit committee, board of director characteristics, and earnings management. Journal of Accounting and Economics, 33, 375-400.
Klein, A. (2002b). Economic determinants of audit committee independence. The Accounting Review, 77(2), 435-452.
Krishnan, G. V. (2007). Did earnings conservatism increase for Andersen clients? Journal of Accounting, Auditing & Finance, 22(2), 141-163.
Krishnan, G. V., and Visvanathan, G. (2007a). Does the SOX definition of an accounting expert matter? The association between audit committee directors' accounting expertise and accounting conservatism. Working Paper, George Mason University.
Krishnan, G. V., and Visvanathan, G. (2007b). Reporting internal control deficiencies in the post-Sarbanes-Oxley era: The role of auditors and corporate governance. International Journal of Auditing, 11, 73-90.
LaFond, R., and Roychowdhury, S. (2008). Managerial ownership and accounting conservatism. Journal of Accounting Research, 46(1), 101-135.
LaFond, R., and Watts, R. L. (2008). The information role of conservatism. The Accounting Review, 83(2), 447-478.
Lara, J. M. G., Osma, B. G., and Mora, A. (2005). The effect of earnings management on the asymmetric timeliness of earnings. Journal of Business Finance & Accounting, 32(3/4), 691-726.
Lee, P., Stokes, D., Taylor, S., and Walter, T. (2003). The association between audit quality, accounting disclosures and firm-specific risk: Evidence from initial public offerings. Journal of Accounting and Public Policy, 22(5), 377-400.
McDaniel, L., Marint, R., and Maines, L. A. (2002). Evaluating financial report quality: The effects of financial expertise vs. financial literacy. The Accounting Review.
McMullen, D. A., and Raghunandan, K. (1996). Enhancing audit committee effectiveness. Journal of Accountancy, 182(August), 79-81.
21 | P a g e
Menon, K., and Williams, J. D. (1994). The use of audit committees for monitoring. Journal of Accounting and Public Policy, 13(2), 121-139.
New York Stock Exchange and National Association of Securities Dealers. (1999). Report and Recommendations of the Blue Ribbon Committee on Improving the Effectiveness of Corporate Audit Committees. New York, New York, U.S.A.: NYSE and NASD.
Pincus, K., Rusbarsky, M., and Wong, J. (1989). Voluntary formation of audit committees among NASDAQ firms. Journal of Accounting and Public Policy, 8(4), 239-265.
Roychowdhury, S., and Watts, R. L. (2007). Asymmetric timeliness of earnings, market-to-book and conservatism in financial reporting. Journal of Accounting & Economics, 44, 2-31.
Ruddock, C., Taylor, S. J., and Taylor, S. L. (2006). Nonaudit services and earnings conservatism: Is auditor independence impaired? Contemporary Accounting Research, 23(3), 701-746.
Stewart, J., and Munro, L. (2007). The impact of audit committee existence and audit committee meeting frequency on the external audit: Perceptions of Australian auditors. International Journal of Auditing, 11, 51-69.
Vafeas, N. (2005). Audit committees, boards and the quality of reported earnings. Contemporary Accounting Research, 22(4), 1093-1122.
Van der Zahn, J.-L. W. M., and Tower, G. (2004). Audit committee features and earnings management: Further evidence from Singapore. International Journal of Business Governance and Ethics, 1(2/3), 233-258.
Watts, R. (1992). Accounting choice theory and market-based research in accounting. The British Accounting Review, 24(3), 235-249.
Watts, R. (2003). Conservatism in accounting - Part 1: Explanations and implications. Accounting Horizons, 17(September), 207-221.
Wright, D. W. (1996). Evidence of the relation between corporate governance characteristics and the quality of financial reporting. Working Paper, University of Michigan, Michigan, United States of America.
Xie, B., Davidson, W. N., and DaDalt, P. J. (2003). Earnings management and corporate governance: the role of the board and audit committee. Journal of Corporate Finance, 9, 295-316.
22 | P a g e
Table 1: Sample selection and industry breakdown
Panel A: Sample selectionNumber of firms listed on the ASX as at January 1, 2008 2121Exclusions:
Financial Institutions 31Insurance 22Utilities 9IPO Firms 81Foreign incorporated firms 236Firms not continuously listed 43Missing data 71
Sample pool for random selection 1628Number randomly selected 100Excluded due to missing data 6
Final useable sample 94Panel B: Sample firm breakdown by industryASX Industry No. of Firms % of Sample
Consumer Discretionary 13 13.829Consumer Staples 4 4.256Energy 6 6.384Health Care 13 13.829Industrials 26 27.659Information Technology 7 7.447Materials 23 24.469Telecommunication Services 2 2.127
Total 94 100
23 | P a g e
Table 2: Descriptive statistics
Variable Statistic Pooled-Sample
1st Quartile 2nd Quartile 3rd Quartile 4th Quartile
(n=94) (n=24) (n=24) (n=22) (n=24)
OPATjt ($’000) Mean(Median)
19,873(211)
74,775(68,074)
6,682(3,025)
-1,676(-1,646)
-2,084(-1,123)
OPATjt-1 ($’000) Mean(Median)
52,682(344)
208,289(92,069)
909(2,106)
-2,321(-1,386)
-732(-859)
XjtMean
(Median)-0.032(0.003)
0.012(0.053)
0.035(0.023)
-0.016(-0.035)
-0.160(-0.120)
MarCapjt ($’000) Mean(Median)
1,044,853(47,557)
3,891,918(1,435,846)
160,746(111,269)
32,814(25,435)
9,597(6,977)
MarCapjt-1 ($’000) Mean(Median)
1,260,557(65,716)
4,706,131(2,389,310)
183,902(152,931)
39,443(35,973)
10,991(10,641)
RjtMean
(Median)-0.073
(-0.107)0.070
(0.012)-0.064
(-0.086)-0.136
(-0.107)-0.169
(-0.216)
DRjtCount
(%Sample)57
(60.638)11
(45.833)14
(58.333)14
(63.636)18
(75.000)
Size_BoDjt Mean
(Median)5.470
(5.000)7.960
(7.500)5.710
(5.000)4.140
(4.000)3.960
(3.500)
%Ind_BoDjt Mean
(Median)47.763
(50.000)65.483
(71.429)46.690
(50.000)46.364
(36.667)32.401
(33.333)
Ind_BoDjtCount
(%Sample)50
(53.191)18
(75.000)14
(58.333)10
(45.455)8
(33.333)
%Expt_Acc_BoDjt Mean
(Median)24.439
(22.500)26.271
(26.136)25.372
(20.000)23.258
(22.500)22.758
(20.000)
Expt_Acc_BoDjtCount
(%Sample)69
(73.404)22
(91.667)19
(79.167)15
(68.182)13
(54.167)
%Expt_CEONP_BoDjt Mean
(Median)8.049
(0.000)8.089
(0.000)9.716
(0.000)9.242
(0.000)5.248
(0.000)
Expt_CEONP_BoDjtCount
(%Sample)32
(34.043)10
(41.667)9
(37.500)8
(36.363)5
(20.833)
Expt_BoDjtCount
(%Sample)75
(79.787)22
(91.667)20
(83.333)18
(75.000)15
(62.500)
%AC_Exp_BoDjt Mean
(Median)41.943
(40.000)53.465
(56.349)41.071
(37.500)50.000
(50.000)23.909
(16.667)
Exp_BoDjtCount
(%Sample)77
(81.915)23
(95.833)20
(83.333)22
(100.0)12
(50.000)
#Act_Meetg_BoDjtMean
(Median)10.710
(10)11.380
(10.000)12.170
(11.500)9.320
(9.500)9.880
(8.500)
Dil_BoDjtCount
(%Sample)54
(57.447)14
(58.333)18
(75.000)11
(50.000)11
(45.833)
BoDjtMean
(Median)2.330
(3.000)3.708
(4.000)2.667
(3.000)1.818
(2.000)1.083
(1.000)Legend:OPATjt ($’000) = Operating profit after tax and interest of firm j for time period t expressed in AUD$’000; OPATjt-1 ($’000) = Operating profit after tax and interest of firm j for time period t-1 expressed in AUD$’000; Xjt = Operating profit after tax of firm j deflated by market value of equity (MVE t-1 - Market value of equity of firm j at beginning of the fiscal year t) of firm j at beginning of the fiscal year t; MarCapjt ($’000) = Market capitalization of firm j for time period t expressed in AUD$’000;MarCapjt-1 ($’000) = Market capitalization of firm j for time period t-1 expressed in AUD$’000; Rjt = Buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1
is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the All Ordinaries Index between dates relevant to the buy-and-hold return period of firm j; DRjt = Indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0); Size_BoDjt = Number of individuals on the audit committee of firm j at the end of time period t; %Ind_BoDjt = Percentage of members of the audit committee of firm j at the end of time period t designated as independent directors; Ind_All_BoDjt = Indicator variable that takes the value of 1 if all members of the audit committee of firm j in time period t are independent directors, and 0 otherwise; Ind_BoDjt = Indicator variable that takes the value of 1 if the majority of the audit committee of firm j in time period t are independent directors, and 0 otherwise; %Expt_Acc_BoDjt = Percentage of the independent directors of the audit committee of firm j at the end of time period t having necessary accounting educational and/or professional affiliations to be defined as a financial expert; Expt_Acc_BoDjt = Indicator
24 | P a g e
variable where firm j is scored 1 if at least one independent director of the audit committee of firm j at the end of time period t has necessary accounting educational and/or professional affiliations to be defined as a financial expert; %Expt_NPCEO_BoDjt = Percentage of the independent directors of the audit committee of firm j at the end of time period t having non-for-profit chief executive officer experience to be defined as a financial expert; Expt_NPCEO_BoDjt = Indicator variable where firm j is scored 1 if at least one independent director of the audit committee of firm j at the end of time period t has necessary non-for-profit chief executive officer experience to be defined as a financial expert; Expt_BoDjt = Indicator variable that takes the value of 1 if at least one independent director of the audit committee of firm j in time period t has necessary expertise (based on educational, professional affiliations and/or non-for-profit role) to be financially qualified, and 0 otherwise; %AC_Exp_BoDjt = Percentage of the independent directors of the audit committee of firm j at the end of time period t having prior audit committee experience with another listed firm; Exp_BoDjt = Indicator variable that takes the value of 1 if at least one independent director of the audit committee of firm j in time period t has prior audit committee experience, and 0 otherwise; Dig_BoDjt = Indicator variable that takes the value of 1 if the audit committee of firm j in time period t met ten [10] or more times during time period t, and 0 otherwise; and BoDjt = Composite audit committee effectiveness score of firm j in fiscal year t.
25 | P a g e
Table 3: Multiple regression analysis of audit committee effectiveness and earnings conservatism
Column I: Univariate Model Column II: Base Model Column III: Full ModelVariables Beta t-Stat P-value Beta t-Stat P-value Beta t-Stat P-valueIntercept -0.013 -0.463 0.645 0.030 0.561 0.576 -0.441 -2.182 0.032**DRjt 0.432 2.362 0.027** 0.427 1.767 0.081***BoDjt 0.149 2.372 0.020**DRjt*BoDjt -0.145 -1.829 0.071***Rjt 0.262 4.102 0.000* 0.183 1.848 0.068 1.858 4.808 0.000*Rjt*DRjt 1.158 2.756 0.007** -1.481 -2.804 0.006**Rjt*BoDjt -0.604 -4.536 0.000*Rjt*DRjt*BoDjt 0.589 3.116 0.002**R2 0.155 0.165 0.580Adjusted-R2 0.145 0.137 0.336N 94 94 94F-Statistic 16.828* 7.575* 6.230*
Legend: Column I Equation: Xjt = α0 + β0Rjt + εjt
Column II Equation: Xjt = α0 + α1DRjt + β0Rjt + β1Rjt*DRjt + εjt
Column III Equation: Xjt = α0 + α1DRjt + α2BoDjt + α3DRjt*BoDjt + β0Rjt + β1Rjt*DRjt + β2Rjt*BoDjt + β3Rjt*DRjt*BoDjt + εjt
Where: Xjt = Operating profit after tax of firm j deflated by market value of equity (MVE t-1 - Market value of equity of firm j at beginning of the fiscal year t) of firm j at beginning of the fiscal year t; DRjt = Indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0); BoDjt = Composite audit committee effectiveness score of firm j in fiscal year t; DRjt*BoDjt = Two-way interaction between indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0), and the composite audit committee effectiveness score of firm j in fiscal year t; Rjt = Buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j; Rjt*DRjt = Two-way interaction between the buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j and indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0); Rjt*BoDjt = Two-way interaction between the buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j and the composite audit committee effectiveness score of firm j in fiscal year t; Rjt*DRjt*BoDjt = Three-way interaction between the buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1
where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j, indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0), and the composite audit committee effectiveness score of firm j in fiscal year t; αk, βk = Coefficients; εjt = Error term; and *, **, *** = significant at the 0.001, 0.05 and 0.10 confidence levels.
26 | P a g e
Table 4: Multiple regression analysis of audit committee effectiveness components and earnings conservatism
Column I: BoD Independence Column II: BoD Expertise Column III: BoD Experience Column IV: BoD DiligenceVariables Beta t-Stat P-value Beta t-Stat P-value Beta t-Stat P-value Beta t-Stat P-valueIntercept -0.062 -0.643 0.522 -0.259 -1.819 0.072*** 0.014 0.035 0.972 0.051 0.579 0.564DRjt 0.018 0.141 0.888 0.348 1.876 0.064*** 0.035 0.087 0.931 0.012 0.090 0.928Ind_BoDjt 0.089 0.777 0.439Expt_BoDjt 0.306 2.020 0.046**Exp_BoDjt 0.035 0.090 0.928Dil_BoDjt -0.007 -0.056 0.955DRjt*Ind_BoDjt 0.103 0.577 0.566DRjt*Expt_BoDjt -0.427 -2.080 0.041**DRjt*Exp_BoDjt -0.014 -0.034 0.973DRjt*Dil_BoDjt -0.060 -0.333 0.740Rjt 0.619 3.314 0.001** 1.311 5.191 0.000* -0.268 -0.265 0.792 0.205 1.796 0.076***Rjt*DRjt 0.361 1.259 0.211 0.687 1.560 0.122 0.707 0.658 0.512 0.104 0.351 0.726Rjt*Ind_BoDjt -0.611 -2.813 0.006**Rjt*Expt_BoDjt -1.305 -4.840 0.000*Rjt*Exp_BoDjt 0.461 0.453 0.652Rjt*Dil_BoDjt -0.176 -0.686 0.495Rjt*DRjt*Ind_BoDjt 0.879 2.030 0.045**Rjt*DRjt*Expt_BoDjt 0.941 1.930 0.057***Rjt*DRjt*Exp_BoDjt 0.609 0.554 0.581Rjt*DRjt*Dil_BoDjt 0.261 0.574 0.567R2 0.271 0.368 0.191 0.188Adjusted-R2 0.212 0.317 0.126 0.122N 94 94 94 94F-Statistic 4.576* 7.158* 2.908* 2.849*
Legend: Column I Equation: Xjt = α0 + α1DRjt + α2Ind_BoDjt + α3DRjt*Ind_BoDjt + β0Rjt + β1Rjt*DRjt + β2Rjt*Ind_BoDjt + β3Rjt*DRjt*Ind_BoDjt + εjt
Column II Equation: Xjt = α0 + α1DRjt + α2Expt_BoDjt + α3DRjt*Expt_BoDjt + β0Rjt + β1Rjt*DRjt + β2Rjt*Expt_BoDjt + β3Rjt*DRjt*Expt_BoDjt + εjt
27 | P a g e
Column III Equation: Xjt = α0 + α1DRjt + α2Exp_BoDjt + α3DRjt*Exp_BoDjt + β0Rjt + β1Rjt*DRjt + β2Rjt*Exp_BoDjt + β3Rjt*DRjt*Exp_BoDjt + εjt
Column IV Equation: Xjt = α0 + α1DRjt + α2Dil_BoDjt + α3DRjt*Dil_BoDjt + β0Rjt + β1Rjt*DRjt + β2Rjt*Dil_BoDjt + β3Rjt*DRjt*Dil_BoDjt + εjt
Where: Xjt = Operating profit after tax of firm j deflated by market value of equity (MVE t-1 - Market value of equity of firm j at beginning of the fiscal year t) of firm j at beginning of the fiscal year t; DRjt = Indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0); Ind_BoDjt = Indicator variable that takes the value of 1 if the majority of the audit committee of firm j in time period t are independent directors, and 0 otherwise; Expt_BoDjt = Indicator variable that takes the value of 1 if at least one independent director of the audit committee of firm j in time period t has necessary expertise (based on educational, professional affiliations and/or non-for-profit role) to be financially qualified, and 0 otherwise; Exp_BoDjt = Indicator variable that takes the value of 1 if at least one independent director of the audit committee of firm j in time period t has prior audit committee experience, and 0 otherwise; Dig_BoDjt = Indicator variable that takes the value of 1 if the audit committee of firm j in time period t met ten [10] or more times during time period t, and 0 otherwise; DRjt*X_BoDjt - Two-way interaction between indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0), and each applicable individual component of audit committee effectiveness where X_BoDjt is: (a) Ind_BoDjt in Column I; (b) Expt_BoDjt in Column I; (c) Exp_BoDjt in Column III; and (d) Dil_BoDjt in Column IV; Rjt = Buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j; Rjt*DRjt = Two-way interaction between the buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j and indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0); Rjt* X_BoDjt - Two-way interaction between the buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j and each applicable individual component of audit committee effectiveness where X_BoDjt is: (a) Ind_BoDjt in Column I; (b) Expt_BoDjt in Column I; (c) Exp_BoDjt in Column III; and (d) Dil_BoDjt in Column IV; Rjt*DRjt*X_BoDjt = Three-way interaction between the buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j, indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0), and each applicable individual component of audit committee effectiveness where X_BoDjt is: (a) Ind_BoDjt in Column I; (b) Expt_BoDjt in Column I; (c) Exp_BoDjt in Column III; and (d) Dil_BoDjt in Column IV; αk, βk = Coefficients; εjt = Error term; and *, **, *** = significant at the 0.001, 0.05 and 0.10 confidence levels.
28 | P a g e
Table 5: Multiple regression analysis alternative market index
Column I: ASX 200 Index† Column II: Size adjusted indexΨ Column III: Industry specific index¥
Variables Beta t-Stat P-value Beta t-Stat P-value Beta t-Stat P-valueIntercept -0.444 -2.200 0.030** -0.298 -1.768 0.081*** -0.388 -2.191 0.031**DRjt 0.425 1.769 0.080*** 0.247 1.102 0.273 0.329 1.464 0.147BoDjt 0.151 2.395 0.019** 0.103 1.932 0.057*** 0.148 2.607 0.011**DRjt*BoDjt -0.144 -1.824 0.072*** -0.096 -1.274 0.206 -0.139 -1.881 0.063***Rjt 1.864 4.840 0.000* 1.441 4.362 0.000* 1.596 4.935 0.000*Rjt*DRjt 1.496 2.842 0.006** 1.124 2.081 0.040** 1.439 3.348 0.001**Rjt*BoDjt -0.606 -4.567 0.000* -0.465 -3.986 0.000* -0.523 -4.549 0.000*Rjt*DRjt*BoDjt 0.597 3.171 0.002** 0.442 2.269 0.026** 0.549 3.462 0.001**R2 0.583 0.534 0.597Adjusted-R2 0.339 0.286 0.357N 94 94 94F-Statistic 6.310* 4.913* 6.818*
Legend: † - Market-adjusted returns are calculated using the ASX 200 Index;Ψ – Market-adjusted returns for firms included in the: (a) 1 st Quartile are based on the All Ordinaries Index; (b) 2nd Quartile are based on the ASX 200 Index; (c) 3rd Quartile are based on the Mid-Cap 50 Index; and (d) 4th Quartile based on the Small Ordinaries Index;¥ - Market-adjusted returns are based on the industry index relevant to the major industry sector each firm is classified under;Column I, II and III Equation:Xjt = α0 + α1DRjt + α2BoDjt + α3DRjt*BoDjt + β0Rjt + β1Rjt*DRjt + β2Rjt*BoDjt + β3Rjt*DRjt*BoDjt + εjt
Where: Xjt = Operating profit after tax of firm j deflated by market value of equity (MVE t-1 - Market value of equity of firm j at beginning of the fiscal year t) of firm j at beginning of the fiscal year t; DRjt = Indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0); BoDjt = Composite audit committee effectiveness score of firm j in fiscal year t; DRjt*BoDjt = Two-way interaction between indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0), and the composite audit committee effectiveness score of firm j in fiscal year t; Rjt = Buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j; Rjt*DRjt = Two-way interaction between the buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j and indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0); Rjt*BoDjt = Two-way interaction between the buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1 where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j and the composite audit committee effectiveness score of firm j in fiscal year t; Rjt*DRjt*BoDjt = Three-way interaction between the buy-and-hold return over the fiscal year t of firm j (i.e., (Pt – Pt-1)/Pt-1
where Pt is the price of shares for firm j at the end of the fiscal year t and Pt-1 is the price of shares for firm j at the start of the fiscal year t) adjusted for the change in the market index between dates relevant to the buy-and-hold return period of firm j, indicator variable where firm j is scored one (1) if Rjt is negative, otherwise firm j is scored zero (0), and the composite audit committee effectiveness score of firm j in fiscal year t; αk, βk = Coefficients; εjt = Error term; and *, **, *** = significant at the 0.001, 0.05 and 0.10 confidence levels.
29 | P a g e