cohen et al - 2004 - the corporate governance mosaic and financial reporting quality

THE CORPORATE GOVERNANCE MOSAIC AND FINANCIAL
REPORTING QUALITY







Jeffrey Cohen
Associate Professor, Boston College


Ganesh Krishnamoorthy
Associate Professor, Northeastern University


Arnie Wright
Professor, Boston College




Published in

Journal of Accounting Literature (2004, pp. 87-152)

1

THE CORPORATE GOVERNANCE MOSAIC
AND FINANCIAL REPORTING QUALITY

INTRODUCTION
One of the most important functions that corporate governance can play is in ensuring the
quality of the financial reporting process. Levitt (1999 2) stated in a speech to directors, “the link
between a company’s directors and its financial reporting system has never been more crucial.”
Further, the Blue Ribbon Commission (1999) called for auditors to discuss with the audit
committee the quality and not just the acceptability of the financial reporting alternatives.
Corporate governance has received increasing emphasis both in practice and in academic
research (e.g., Blue Ribbon Committee Report 1999; Ramsay Report 2001; Sarbanes-Oxley 2002;
Bebchuk and Cohen 2004). This emphasis is due in part, to the prevalence of highly publicized and
egregious financial reporting frauds such as Enron, WorldCom, Aldelphia, and Parmalat, an
unprecedented number of earnings restatements (Loomis 1999; Wu 2002; Palmrose and Scholz
2002; Larcker et al. 2004) and claims of blatant earnings manipulation by corporate management
(Krugman 2002). Further, academic research has found an association between weaknesses in
governance and poor financial reporting quality, earnings manipulation, financial statement fraud,
and weaker internal controls (e.g., Dechow et al. 1996; Beasley 1996; McMullen 1996; Beasley et
al. 1999; Beasley et al. 2000; Carcello and Neal 2000; Krishnan 2001; Klein 2002b). Given these
developments, there has been an emphasis on the need to improve corporate governance over the
financial reporting process (e.g., Levitt 1998, 1999, 2000), such as enacting reforms to improve the
effectiveness of the audit committee (Blue Ribbon Committee 1999; Sarbanes-Oxley Act 2002)
and to make the board of directors and management more accountable for ensuring the integrity of
the financial reports (SEC 2002, The Business Roundtable 2002) as well as a rapid expansion of
research on corporate governance.

2

The purpose of this paper is to review research on corporate governance and its impact on
financial reporting quality. This review will serve three purposes: (1) to suggest a corporate
governance “mosaic”(i.e., the interactions among the actors and institutions that affect corporate
governance) that encompasses a broader view of governance than has been considered in prior
accounting research; (2) to provide an overview of the principal findings of prior research; and (3)
to identify important gaps in the research that represent promising avenues for future study.
Accordingly, the remainder of the paper is divided into the following three sections. The next
section provides a general framework for understanding the corporate governance mosaic and its
impact on financial reporting quality. This section is followed by a discussion of prior research,
dealing respectively with the role of the following actors in the corporate governance mosaic: (1)
the board of directors and the audit committee; (2) the external auditor; and (3) the internal
auditors. The final section provides a summary of areas for future research.
THE CORPORATE GOVERNANCE MOSAIC
Figure 1 provides an overview of the corporate governance mosaic and its impact on
financial reporting quality. Prior accounting research and the accounting profession have focused
primarily on the board of directors and the audit committee. For instance, the Public Oversight
Board (POB 1993) defined corporate governance as “those oversight activities undertaken by the
board of directors and audit committee to ensure the integrity of the financial reporting process.”
However, a narrow view of corporate governance restricting it to only monitoring activities may
potentially undervalue the role that corporate governance can play.
Further, in a recent meta analysis of corporate governance research, Larcker et al. (2004, 1)
conclude that “the typical structural indicators used in academic research and institutional rating
services have very limited ability to explain managerial behavior and organizational performance.”
Thus, as depicted in Figure 1, a more comprehensive framework should consider all major
stakeholders in the governance mosaic, including those inside and outside the firm. For instance,
the external auditor plays a significant role in monitoring financial reporting quality and hence can

3

be viewed as an important participant in the governance process. We do not suggest that extant
research has not looked at the role of the auditor but rather that the role of the auditor in the
governance process is very complex as the auditor interacts with other stakeholders in the
governance mosaic such as the audit committee and the management. In turn, the interplay among
the stakeholders is affected by outside forces such as by regulators and stock exchanges as well as
pressure to meet financial analysts. Further, the corporate governance mosaic suggests we need to
look beyond much of the focus of current research in corporate governance that has concentrated
on documenting associations and not causal relationships (Larcker et al. 2004) and to complement
the current research by also investigating the substance of the interactions in the corporate
governance arena. For example, although the emphasis in corporate governance research has been
on looking at issues of independence, Cohen et al. (2002) document that unless management allows
itself to be monitored the substance of governance activities will be subverted.
Figure 1 also indicates interrelationships between the various actors and mechanisms
within the corporate governance mosaic. For example, the interactions among the audit committee,
the external auditor, the internal auditor, the board, and the management are crucial to effective
governance and to achieving high quality financial reporting (Sarbanes-Oxley Act 2002). An
interview study with experienced auditors (Cohen et al. 2002) revealed that management has a
significant influence over these parties. Some of the auditors in that study argue that if management
does not want to be “governed”, they can’t be (Cohen et al. 2002 582). Further, management may
place passive, compliant members on the board who may satisfy regulatory requirements but are
reluctant to challenge management. For example, QWEST had no outside directors with experience
in the company’s core business. They also had a compensation committee that consistently
awarded excessive bonuses to management in spite of the firm’s relatively subpar performance
(Business Week 2002).
Other actors and mechanisms depicted in Figure 1, largely external to the corporation, also
influence its effective governance in significant ways and are integral to safeguarding the interest

4

of a company’s stakeholders. Examples of such actors include, but are not limited to, regulators,
legislators, financial analysts, stock exchanges, courts and the legal system, and the stockholders.
These external players often shape and influence the interactions among the actors who are more
directly involved in the governance of the corporation. For instance, the Sarbanes-Oxley Act
(2002) has significantly impacted all direct players in the corporate governance mosaic not only in
terms of their role and function in the governance process but also in terms of how the players
interact with one another. Under Sarbanes-Oxley, the audit committee now has the responsibility to
hire and fire the auditor and to approve the non-audit services that the auditing firm can perform
(Sarbanes-Oxley Act 2002). Further, management must state that it has the responsibility for
maintaining the internal control system and for evaluating its effectiveness (Geiger and Taylor
2003).
1
In summary, Figure 1 depicts the actors in the governance process, highlights their
potential interactions, and suggests that the governance process impacts the quality of financial
reporting (e.g., transparency, objectivity) and, in the extreme, earnings manipulation and outright
fraud.
Quality of Financial Reporting
Although one should expect that “better” corporate governance leads to improved financial
reporting, there is a lack of consensus as to what constitutes “financial reporting quality.” For
example, although, the BRC (1999) and Sarbanes-Oxley (2002) require auditors to discuss the
quality of the financial reporting methods and not just their acceptability, the notion of financial
reporting quality remains a vague concept. As Jonas and Blanchet (2000, 353) state, “in light of
these new requirement, auditors, audit committee members, and management are now struggling to
define “quality of financial reporting.”
Rather than define “quality of financial reporting,” prior literature has focused on factors
such as earnings management, financial restatements, and fraud that clearly inhibit the attainment

1
Although we don’t discuss the courts in this paper, we do include the courts in the governance mosaic and

recognize that they play a critical role in the governance process. For instance, the courts can define and can

5

of high quality financial reports and have used the presence of these factors as evidence of a
breakdown in the financial reporting process. Specifically, prior literature has examined the role of
the various players in the governance mosaic (e.g., board, audit committees, external auditor,
internal auditors) and the extent to which these players have either individually or collectively
influenced the attainment of financial reports that are free from material misstatements and
misrepresentations. The principal players identified in prior literature include the board of
directors, the audit committee, the external auditor, and the internal auditors.
Accordingly, in the rest of this paper, we discuss the role of the following players, the
interactions among them, and their collective influence in helping attain high quality financial
reporting.: (1) the board of directors and the audit committee (2) the external auditor; and (3) the
internal auditors.
2
A summary of prior research is contained in Table 1.
BOARD OF DIRECTORS AND AUDIT COMMITTEE

Various attributes of the board and audit committee may influence their effectiveness as
corporate governance mechanisms.
3
For example, the BRC’s (1999) recommendations looked at
strengthening both the independence and expertise of audit committees. In this section, we examine
the research of various characteristics of the board and audit committee including issues of (1)
composition, (2) independence, (3) knowledge and expertise, (4) effectiveness, (5) power, (6)
duties and responsibilities and (7) the association between board characteristics and earnings
manipulation and fraud.
Composition

expand the duties of directors and officers to the corporation such as the duty of care and loyalty.
2
Management potentially has a significant impact on the effectiveness of the governance process and is
listed in the corporate governance mosaic. However, since there has been almost no research that explicitly
and directly examines the role of management in the governance process, we discuss management primarily
in relation to opportunities for future research.
3
DeZoort et al. (2003) provides a framework for the evaluation of audit committee effectiveness and
synthesizes the existing literature into four components: audit committee composition, authority, resources,
and diligence. However, their study does not directly address board characteristics, nor does their framework
provide a basis to synthesize and interpret prior literature in terms of their impact on the financial reporting
process.

6

The only study to directly examine the composition of AC members in terms of board
experience and independence was conducted by Vafeas (2001). He found that members appointed
to the AC have significantly less board tenure with the firm, serve on fewer other committees, and
are less likely to serve on the important compensation committee. Surprisingly, they hold the same
level of equity interest in the firm and are as likely to be a grey
4
director as other members of the
board. The overall results led Vafeas to conclude that AC appointees are less seasoned board
members, who are not chosen because of their greater experience or independence but consistent
with a “next in line” strategy. He notes that future research is needed to examine whether the
characteristics studied are linked to improved AC performance or financial reporting quality and
that the composition of the AC in terms of other characteristics such as financial literacy need
consideration.
To answer the question of what audit committees are actually doing, Carcello et al. (2002)
examined recent disclosures of audit committee charters and reports included in proxy statements.

The major finding of this study is that there is a gap between what audit committees say they are
doing and what is mandated by their charter. Although this gap may be due to several reasons
including liability concerns, it raises the general issue of transparency with respect to activities of
the audit committee despite the changes made in disclosure requirements based on BRC
recommendations. The study also found that while there is generally a high level of compliance
across firms with respect to exchange mandated disclosures, voluntary disclosures of AC activities
were more prevalent in larger companies, depository institutions, NYSE firms, and firms with
independent audit committees. An important limitation of the study is that it did not explore the
specific reasons for the gap between what is stated in the AC reports and the mandate in the AC


4
Independent directors are defined as non-employees with no tie to the firm or its management except in
their role as a director. Grey directors are defined as non-employees who may have past or present
relationship with the firm or its management such as relatives of management or consultants and suppliers.



7

charter. Future studies should address this issue by using other research methods such as interviews
or internal documentation that will complement the archival data.
Collectively, these two studies suggest that audit committees prior to Sarbanes-Oxley may
have underutilized its potential as the committees had less experienced members on it and there are
questions about whether committees fully fulfilled its mandate. Of course, since the data in these
studies were collected prior to Sarbanes-Oxley, it is imperative to determine if these shortcomings
still hold true in a world where there is more scrutiny placed on AC activities.
Independence
The recent reforms in Sarbanes-Oxley (2002) enacted to strengthen the audit committee
(AC) has implicitly assumed that independence will improve the effectiveness of the audit

committee. An early study that investigated this issue was conducted by Vicknair, Hickman and
Carnes (1993) who examine the level of “grey” directors who are members of the audit committees
of a sample of NYSE firms. They find that about a third of the members of the audit committees
sampled were grey directors and, thus, of questionable independence. Since this data was gathered
about 15 years ago, it is unlikely that the findings are reflective of the current situation given the
movement to improve the independence of audit committees (e.g., Blue Ribbon Committee 1999).
Wolnizer (1995) uses an a priori argument approach to evaluate whether independent ACs
can significantly improve financial reporting quality. He argues that this is unlikely, because
current accounting practices allow wide discretion by management in the choice of accounting
methods and estimates. With limited exceptions (e.g., count of cash or inventory) accounting
values are not subject to validation through impartial evidence (e.g., market based data). Thus,
auditors and the AC can only evaluate or review management’s potentially biased choices. The
most significant contribution of this work is to highlight the link between the nature (limitations) of
the accounting system and those who seek to monitor the system.
Another study relating to the importance of AC independence was conducted by DeZoort
and Salterio (2001). They examined the judgments of audit committee (AC) members who were

8

asked to determine their level of support for the auditor vis-à-vis management in a case situation
involving a dispute over proper revenue recognition. The primary issue addressed was how AC
independence and knowledge affect audit committee members’ propensity to support the auditor’s
position. The findings indicate that independent, more knowledgeable AC members were more
likely to support the auditor in a dispute with management.
A question arises as to what causes the demand for AC independence. To explore this
issue, Klein (2002a) examined whether AC independence was affected by a number of board
factors and by substitute monitoring mechanisms. Her results indicate a positive association
between AC independence and board size and AC independence and the proportion of outside
board members. Klein also found negative associations between AC independence and a firm’s
growth opportunities, AC independence and the existence of a large blockholder on the audit

committee and finally AC independence and firm size. There was no effect for creditors, CEO on
compensation committee, and outside directors' shareholdings. The study suggests that prior to the
regulations enacted recently after the BRC, the ability of an AC to be independent was affected by
the larger outside board and a firm's financial health. This study highlights the importance of the
board in its power over assignments and authority afforded to the AC.
Collectively, these studies, especially Klein (2002a), suggest that the independence of audit
committees may be affected by the independence of the board in general. Although there is nothing
in Sarbanes-Oxley (2002) that mandates the selection of powerful AC members who are
independent in fact as well as in form, there is at least the potential that stronger boards in general
will seek out AC members who are willing to confront management to a greater degree than
previously was documented prior to the enactment of the BRC (1999) reforms.
Knowledge and Expertise
With the requirement in Sarbanes-Oxley (2002) that all AC members have financial
literacy and that at least one member be a financial expert, an understanding of the link that
knowledge and expertise has on audit committee effectiveness is quite important. DeZoort (1997)

9

investigated the views of AC members regarding the formal responsibilities of the AC, other duties
performed, and the importance of potential duties from those compiled by Wolnizer (1995). The
duties from Wolnizer fall into the general categories of financial reporting (including controls);
auditing; and other corporate governance (e.g., facilitate communications between the board and
the external auditors).The results suggest that AC members were not fully aware of their formal
responsibilities when comparing their responses to those reported in the company’s proxy
statement. Noteworthy, the majority felt that all AC members should have sufficient knowledge in
accounting, auditing, and legal issues and that they perceived they did not have enough knowledge
in many of these areas. DeZoort calls for further research examining the divergence of publicly
disclosed responsibilities of the AC and those identified by participants. He also emphasizes the
need for further work regarding the types and composition of expertise needed by AC members.
DeZoort (1998) evaluated whether AC members with experience in auditing and internal

controls would make different internal control evaluations than members without this experience.
DeZoort found that as hypothesized, the AC members with experience were more likely than AC
members without such experience, to make control evaluations more in line with external auditors.
The AC members with greater experience also were more consistent and demonstrated a higher
degree of consensus. These results suggest that ACs that have members with appropriate domain
related experience may at least have a better understanding of the auditor’s side in disputes with
management and potentially may even lend support to the auditor in their dispute.
Beasley and Salterio (2001) posit that the board has a significant influence on the quality of
the AC in terms of independence and knowledge, since it is the board who selects AC members.
They argue that strong, independent boards, as evidenced by the proportion of outside members, an
independent chair who is not the CEO of the company, and larger size, will be more likely to
appoint a higher quality AC. Their findings supported expectations, although a weaker association
was found for AC knowledge. This study is noteworthy in that it is the first one to explicitly
consider the AC as part of the corporate governance rubric, highlighting that there are many other

10

interrelated mechanisms such as the board, the external auditor, and holders of large blocks of
stock. Beasley and Salterio note that future research is needed to examine the causal links (beyond
the association found in their study) between AC quality and other governance mechanisms and to
examine whether and how AC characteristics impact monitoring effectiveness.
The issue of how the expertise and financial literacy of AC members potentially affect the
quality of the financial reporting process was examined by McDaniel et al. (2002). The BRC had
recommended that ACs be comprised of individuals possessing financial literacy with at least one
member being a financial expert. In their study, McDaniel et al. compared how financial experts
may differ from financial literates in the evaluation of the quality of financial reporting items and
whether the salience and the recurring nature of the items would affect the groups' evaluations.
They found that in contrast to literates, the experts' assessment of quality was related to elements of
the SFAC framework (relevance and reliability) and they identified reporting concerns that were
recurring in nature and would receive little business press. In contrast, literates raised concerns

about high salience items that were nonrecurring in nature and those that received attention from
the press. Thus, the paper suggests that each group may bring different perspectives to AC
meetings and thereby improve financial reporting quality.
Finally, DeZoort and Salterio (2001), examine the effect of AC member accounting and
auditing knowledge on the propensity to support the auditor’s position in a dispute with
management over an ambiguous accounting issue. They find that greater auditing knowledge is
positively related to support for the auditor, while, unexpectedly, no relationship is found for
accounting knowledge. The latter finding may have been due to the non-technical, generic nature
of the accounting issue at hand.
Collectively, these studies suggest that getting more knowledgeable audit committee
members may lead to greater cooperation between auditors and the audit committee members of
their clients. However, consistent with the findings of DeZoort and Salterio (2001) this knowledge
potentially may be more important if it relates to knowledge of the complex nature of the

11

accounting problem on hand for a specific industry. Thus, an issue for future research might be to
extend this line of research by examining how the support for auditors from AC members may vary
as a function of the AC knowledge of complex industry specific accounting issues and of auditing.
Effectiveness
Although all factors discussed in the previous subsections can potentially influence AC
effectiveness, studies that have directly examined this issue are discussed in this section. Spangler
and Braiotta (1990) focus on the impact of the AC chair possessing transactional and
transformational leadership attributes. In essence, the transactional leadership attribute pertains to
the ability of the AC chair to help provide opportunities and rewards and thus motivate
management to act in the best interest in the shareholder while the transformational leadership
attribute refers to the ability of the AC chair to provide a vision for management to follow. They
found a positive association between AC effectiveness and transformational leadership and some
transactional leadership characteristics (contingent rewards and active management by exception).
Spangler and Braiotta suggest that further research should examine company specific situational

variables that may impact the importance of AC chair leadership characteristics in committee
effectiveness.
Kalbers and Fogarty (1993) investigate the relationship between various dimensions of
power and the effectiveness of ACs in discharging three oversight roles (financial reporting,
oversight of external auditors, and oversight of internal controls). Their results indicate that
organization types of power are mediated by personal power factors. Specifically, effective ACs
require a strong organizational charter or mandate, institutional support (information support from
management and auditors and a supportive environment by top management), and diligence.
Organization power, however, is not useful unless members exercise strong will and determination.
Knowledge (expert power) was important only for the financial reporting oversight function,
implying ACs can rely on the support of other parties such as the external and internal auditors to
effectively discharge their other functions. Future research could further explore how the complex

12

nature of power dimensions such as the personal relationship between management and the audit
committee members could potentially impact AC effectiveness.
To explore the impact of symbolic activities (e.g. frequency and quality of the audit
committee meetings, the act of asking questions at the audit committee meeting) on the perceived
effectiveness of ACs and to determine what impacts the social construction of the effectiveness of
the AC, Gendron and Bédard (2004) interviewed governance actors both internal to firm and the
external auditors from two corporations. Social construction is the manner in which AC members
achieve legitimacy in the eyes of other attendees at audit committee meetings. They found that the
social construction was affected by the ability of the AC to ask questions, have private meetings
with the external auditors and through the ceremonial and substantive components of the meetings.
They conclude that AC meetings are more than mere symbolism and that the performance of
members plays a significant role in the social construction of their effectiveness. Their research
demonstrates the value of qualitative research methodology to uncover the underpinnings of the
workings of audit committees.
AC’s may be formed primarily for cosmetic reasons to make it appear to outside

stakeholders that the company desires monitoring of financial reporting and controls. Menon and
Williams (1994) examine this issue by looking at the relationship between reliance on the AC and
factors suggested in the literature that potentially drive the need for an effective AC: management
stock ownership; leverage; company size; type of auditor (Big 8 vs. Non-Big 8); board
composition; and board size. AC reliance is measured by activity (number of meetings) and
independence (presence of a member of management). The sample included OTC companies not
required to form an AC. The results indicated that AC activity and independence was positively
associated, as expected, with the proportion of outside directors on the board and that AC activity
was also greater for larger companies.
In an extension of their earlier study (Kalbers and Fogarty 1993), Fogarty and Kalbers
(1998) investigate whether AC effectiveness is more closely aligned with agency or institutional

13

theory. Agency theory would predict that factors that create a need for closer monitoring of
management (e.g., higher leverage) produce the need for effective ACs. In contrast, institutional
theory posits that many organizational structures such as the AC are merely symbolic to conform to
social expectations. To test this issue, they reanalyze the data obtained in their earlier study along
with considering an additional measure of AC effectiveness (number of AC meetings).
5
Although
they find support for the importance of some agency variables (e.g., company size), the results
generally do not show a strong link between AC effectiveness and agency theory factors. There is
also a weak correlation between effectiveness and some measures of organizational bases of power
for the AC (sanctionary power relating to the scope of the AC charter). Fogarty and Kalbers call for
additional consideration of the joint effects of agency and institutional factors in research
explaining AC effectiveness.
Collier and Gregory (1999) replicate and extend the study by Menon and Williams (1994)
by looking at AC activity for large companies and by examining another measure of AC
effectiveness—duration of meetings. The results failed to support the findings of Menon and

Williams regarding the impact of agency variables on the number of AC meetings, most likely
because of limited variance in this measure (usually two meetings per year). However, the type of
auditor (Big 6) was found to be associated with increased duration of AC meetings while leverage
had a marginally significant relation with the duration of AC meetings. Importantly, a dominant
CEO and AC insiders were found to lead to a significantly lower duration of meetings. These latter
results suggest the importance of having an independent chair of the board and independent AC
members to foster an active AC. Future research may work on developing a more testable construct
of audit effectiveness.
Haka and Chalos (1990) investigated the perception of an agency conflict among chief
operating officers, internal auditors, external auditors, and audit committee chairs. Examining

5
It should also be noted that Kalbers and Fogarty define AC independence more strictly than many
prior studies, excluding members of management and “grey” directors.

14

factors that constitute complete financial statement disclosure and factors that should influence
accounting procedure choice, they found that audit committee members wanted greater disclosure
than other groups. Further, audit committee chairs and external auditors disagreed on the influence
of a number of factors (e.g., cost of audit, government intervention) that are important in the
financial reporting process. External auditors and management tended to agree on both what should
constitute complete financial statement disclosure and what should influence accounting procedure
choice.
Krishnamoorthy et al. (2002a) surveyed both audit partners and managers to understand
their conceptualization of financial reporting quality and to evaluate factors that influence the
effectiveness of audit committees. They found that the expertise and willingness of AC members to
confront management are strong influences on the effectiveness of audit committees. They also
found that management is perceived to play a significant role in influencing the extent and the
quality of communication between the external auditors and the audit committee and that the AC

should play a greater role than they currently do in ensuring the quality of the financial reporting
process. Also, consistent with the BRC, clarity and consistency of financial disclosures and degree
of aggressiveness in accounting principles and estimates were cited as the most important
determinants of financial reporting quality. Krishnamoorthy et al. (2002a) did not examine whether
there is consensus between auditors and audit committees about what constitutes quality financial
reporting. However, they suggest that if audit committee members are in agreement with auditors,
then audit committee members will be more likely to support the auditors in disagreements with
management.
Collectively, these studies suggest that it is important to examine what actually transpires
in audit committee meetings and interactions to determine if its reported activities are effective. For
example, both Fogarty and Kalbers (1998) and Gendron and Bedard (2004) demonstrate the need
to look at the substance of the AC activities to help determine if the AC is effective. The mere



15

recording of activities may be symbolic or a means of complying with regulation. This suggests
that future research may explore whether a principles-based approach may be better than a rules-
based approach in achieving the desired goals of audit committees to effectively serve as an
oversight to management in the financial reporting process.
Power
The only studies to explicitly consider the importance of AC power were conducted by
Kalbers and Fogarty (1993) and Fogarty and Kalbers (1998). In the former study, they examine the
relationship between various dimensions of power and AC effectiveness. Power could take the
form of support from the organization in terms of the AC receiving appropriate and timely
information from management or power for an AC could be manifested in the expertise of its
members. The findings indicate that effective ACs require a broad mandate, good institutional
support, and a willingness to undertake its responsibilities. Organization power was only valuable
when the AC exercised diligence. Interestingly, expert power (knowledge) was important only for

the financial reporting oversight function, suggesting the AC could draw on other sources of
expertise in addressing other functions such as overseeing the functioning of controls. Also, as
reviewed earlier, the primary focus of Fogarty and Kalbers (1998) was to examine agency and
institutional theory factors impacting AC effectiveness. They report a weak correlation between
AC effectiveness and some measures of organizational bases of power for the AC (sanctionary
power relating to the scope of the AC charter).
The research by Kalbers and Fogarty (1993) and Fogarty and Kalbers (1998) suggest that
it is difficult to measure the construct of power of the audit committee. It appears that research is
needed that uses an in depth interview approach similar to the methodology used by Gendron and
Bedard (2004) to isolate the characteristics of an audit committee that determine what creates the
power for an AC to be effective in its monitoring role over management. This in depth interview
approach can also isolate when an AC’s lacks real power and exists as a mere symbolic mechanism
in place to comply with regulation.

16

Duties & Responsibilities
Two studies that focus on the duties and responsibilities of the AC are by Wolnizer (1995)
and DeZoort (1997). In a comprehensive review of professional standards and the literature
Wolnizer identifies 17 functions for the AC that fall into three areas: financial reporting (including
controls); auditing; and other corporate governance (e.g., facilitate communications between the
board and the external auditors).
DeZoort provides empirical data to validate this list of functions by asking a sample of AC
members whether they perform these functions. He also compares the duties they note as assigned
to those indicated in the company’s proxy statement. Respondents indicate that the most important
duty for the AC is to evaluate controls. Other important functions were to review financial
statements, review the effectiveness of the internal and external audits, review the management
letter of the external auditor, and evaluate auditor independence. There was a weak association
between duties listed in the proxy statement and those that AC members indicated were assigned to
the committee.

Collectively, these studies suggest that the public disclosures of AC activities do not seem
to map well with those actually exercised. Since the duties and responsibilities of AC members are
evolving, future research can investigate the level of convergence between the activities outlined in
an audit committee charter and the actual duties that AC members perceive they perform as well as
consider factors that may account for greater or lesser convergence.
Earnings Manipulation and Fraud
The research to date in this area has used an archival approach to assess the link between
corporate governance characteristics and instances of earnings manipulation and fraud as
documented in Accounting & Auditing Enforcement Releases (AAERs) issued by the SEC. The
studies discussed below have generally been consistent in their findings that there appears to be a
link between board and/or audit committee characteristics and the incidence of earnings
manipulation and fraud.

17

McMullen (1996), using data from when audit committees were voluntary, examined the
association between the presence of an audit committee and various indicators of possible lapses in
the financial reporting process or the occurrence of illegal acts. Using archival data, she found that
the existence of an audit committee was associated with a lower incidence of shareholder litigation
of alleged fraud, correction of quarterly earnings, SEC enforcement actions, illegal acts, and
auditor turnover due to a reporting dispute with management. A future study could examine the
reasons for this finding. For example, does the ability and the willingness of audit committee
members to ask tough questions of management result in a lower incidence of these lapses than
companies that have audit committees that essentially provide a “rubber stamp” for management?
Dechow et al. (1996) compared firms receiving AAERs with a non-AAER sample
matched
6
on size and industry to determine motives and consequences of earnings manipulation.
They also investigated whether characteristics of the corporate governance structure are associated
with behavior consistent with earnings manipulation. They found that firms subject to enforcement

action have weaker governance structures (e.g. less independent boards and ACs) and are more
likely to engage in earnings manipulation. The existence of an audit committee as well as general
board characteristics (e. g. composition, power of CEO, stock ownership) were associated with
differentiating AAER from non-AAER firms. They found no effect for the existence of a bonus
plan but they did find support for motivation to lower the cost of external financing. Thus, this
study lends support for investigating the importance of governance structures in improving the
quality of the financial reporting process.
Beasley (1996) examined the relationship between board of directors’ composition and the
likelihood of financial statement fraud. In addition, he examined if the presence of an audit
committee reduces the likelihood of fraud. Using a matched sample of fraud and non-fraud firms,

6
Although a discussion of the methodological limitations of using a matched sample in research studies is
beyond the scope of this paper, it is important to recognize these limitations when interpreting the results
from these studies. Readers interested in a thorough discussion of the limitations of matching in empirical
research should refer to Kerlinger and Lee (2000, 489) or a similar treatise on research methods.

18

he reports that boards of no-fraud firms are more likely to have a larger proportion of outside (non-
employee) directors than fraud firms. The results were robust to a finer definition of outside
directors, separating outside directors into grey and independent directors. Surprisingly, the results
indicate that the presence of an audit committee was not associated with a reduced likelihood of
financial statement fraud. Finally, supplemental analysis suggests that the likelihood of fraud
decreases as (1) the level of stock ownership by outside directors increases; and (2) the number of
years of board service for outside directors increases. Further, the likelihood of fraud increases as
the number of directorships held by outside directors in other firms increases. Collectively, these
results highlight the importance of examining corporate board-level factors when evaluating the
impact of corporate governance on financial reporting quality.
7

Using three industries (high technology, health care and financial services), Beasley et al.
(2000) compared fraud companies (subject to SEC enforcement) with benchmarks of industry
corporate governance practices. They found that the only Board of Director measure that differed
was the proportion of outside directors, with fraud companies having a lower proportion of its
directors who are outsiders. Examining the AC variables, they found differences in all three
industries for the likelihood of independent directors with less fraud occurring when there were
more independent AC members. Further, the AC members in the high tech and health care fraud
companies met less frequently than was typical for their respective industries. Finally, they found
that the internal audit function was less likely to be evident in fraud companies. This study suggests
the need to investigate or at a minimum control unique industry effects of specific corporate
governance characteristics.
Abbott et al. (2000) posit that audit committee independence and increased activity will
result in a lower incidence of SEC enforcement actions. Using a matched control sample, they
found support for their expectations. Their results were qualitatively the same if they restricted the

7
Beasley’s (1996) results described are for univariate analysis. The results might be viewed with some
caution as the results for multivariate analysis are not as strong as those found with the univariate analysis.

19

study to the firms specifically sanctioned for fraud. Since their study is a test of association, a
future study could examine the causal link between audit committees characteristics and fraud.
Abbott et al. (2001) compared fraud/financial misstatement firms with a control group.
They found a negative relationship for both independence and activity of AC. No effect was found
for AC financial statement expertise or AC size and, in contrast to Beasley (1996), no effect was
found for any of the board characteristics. Their results suggest that some of the BRC's
recommendations (e.g., independence) may be effective in mitigating fraud or earnings
management while others (e.g., expertise) may either have no effect or may need time to have an

influence on the workings of the AC. It must be noted that since the data was from a pre BRC
period, the reforms enacted post BRC may be effective in improving the quality of the financial
reporting process.
When discussing earnings management , it is also important to explore the effect of
corporate governance mechanisms on earnings management behavior that does not strictly violate
GAAP rules but may nonetheless be potentially violating the spirit of GAAP (Krishnamoorthy et
al. 2002a).An archival study in this area was conducted by Klein (2002b) who examined whether
there is a link between the independence of the board and/or audit committee and the magnitude of
abnormal accruals. She found that firms with a majority-independent board and/or audit committee
have a lower level of discretionary accruals. However, there was no additional curtailment of
earnings management when the audit committee consisted wholly of independent directors. This
result calls into question whether the regulation requiring complete independence of the audit
committee is overly stringent as Klein found having a majority of independent directors would
suffice She also found that when firms switched from a majority-independent to minority
independent board/audit committee they experienced a higher amount of abnormal accruals.
Klein’s study demonstrates the importance of looking at quality of earnings, and not just the
incidence of fraud or earnings manipulation. Since Klein’s paper documents an association and not
causality, future research could disentangle whether the management is influencing the selection of

20

compliant boards in order to engage in aggressive earnings management or conversely do more
independent boards curtail management from overly aggressive earnings management behavior.
Although not directly dealing with earnings manipulation, another relevant study was
conducted by Wild (1996) who investigated the association between formation of an audit
committee and the quality of accounting earnings. Using an information economics framework,
Wild predicts that earnings reports disclosed after the formation of an audit committee will elicit a
greater stock price reaction than those disclosed before, after controlling for other relevant factors.
Using abnormal stock returns, results were consistent with expectations and showed that the stock
market reaction for earnings reports was 20 percent greater for earnings announcements after audit

committee formation than before. These results suggest that audit committees provide a useful
oversight mechanism for the financial reporting process and that this increased oversight results in
improved earnings quality. This study also suggests the value of using a longitudinal approach to
investigate the effect of corporate governance mechanisms on the quality of financial reporting
over time.

Collectively, these studies suggest that there is an association between the independence
of the audit committee and the incidence of fraud or AAERs as well as activities associated with
earnings management behavior. However, there is a need to document why this association exists.
Is it that more independent audit committees are more diligent in preventing management from
engaging in egregious behavior or is that the through the influence of management of better
companies who identify audit committee members that will ask the type of questions that will
mitigate the firm committing errors and irregularities?
Synthesis of Research on Board of Directors and Audit Committee
To synthesize, prior research suggests that audit committees are comprised of relatively
junior members of the board, whose effectiveness is largely dependent not only on their knowledge
and expertise in financial reporting, but also the extent to which the board supports and empowers

21

the AC. Further, AC independence, a key factor in the AC’s ability to confront management and
work effectively with auditors, is largely dependent on the attitude and independence of the board.
Prior studies also indicate that the role of the AC is crucial, especially in resolving auditor-
management disagreements on significant financial reporting issues, an aspect critically important
in supporting and enhancing auditor independence. Finally, there appears to be an association
between characteristics of the board and the audit committee and the incidences of earnings
manipulation and fraud. The results imply that for the AC to play an important role in the financial
reporting process, the AC must be vested by the greater board with real power and sufficient
expertise to serve as an effective monitor over management’s actions. The AC should be viewed as
an ally to auditors in being steadfast in the goal of having a high quality financial reporting process

and in the prevention of fraud.
INSERT TABLE 1 HERE

EXTERNAL AUDITOR
The external auditor plays a crucial role in helping to promote financial reporting quality.
As a result of Sarbanes-Oxley (2002), auditors must be prepared to discuss with the audit
committee the quality and not just the mere acceptability of financial statement issues. Through
curtailing excessive earnings management techniques such as unexpected discretionary accruals,
auditors can serve as an effective monitor on overly aggressive management. Prior research
examining the relationship between various actors in the corporate governance mosaic and the
external auditor can be summarized into the following themes: (a) auditor selection and client
acceptance; (b) audit quality and audit fees; and (c) audit opinion and the audit process. These
themes are discussed below.


22

Auditor Selection and Client Acceptance
With increased emphasis on the role of corporate boards and especially audit committees in
auditor selection and retention decisions, research in this area is of critical importance to both
auditors and other corporate stakeholders. Abbot and Parker (2000) argue that independent and
active audit committees will demand higher quality audits and are hence more likely to engage an
industry-specialist auditor. This reasoning is in line with the earlier cited research on the
relationship between governance factors and quality of earnings (e. g. Klein 2002a and 2002b) that
suggests that stronger boards may demand a higher quality audit to curtail management’s ability to
engage in rampant earnings management behavior. Abbot and Parker defined audit committees as
independent if they were comprised entirely of independent directors and as active if they met at
least twice a year. The study found support for the hypothesis. However, audit committee
independence or activity alone yielded insignificant results.
Beasley and Petroni (2001) examined the association between outside directors on the

board and the choice of external auditors for property-liability insurance companies. They argued
that boards with a higher percentage of outside directors will seek higher quality auditors in order
to provide more effective monitoring of corporate management. Using a sample of property-
liability insurers, Beasley and Petroni (2001) found that the likelihood that a specialist Big-6
auditor is selected increases with the percentage of outside directors on the board, but the outside
board membership percentage has no impact on the choice between a nonspecialist Big-6 and non
Big-6 auditor. Thus, board composition is associated with choice of auditors based on industry
specialization, rather than the blanket brand name (Big 6 vs. non-Big 6) differentiation found in
other settings. Together, these two studies suggest that a more independent and active governance
structure is associated with the use of higher quality auditors, assuming that industry specialization
is an indicator of quality.
Shareholders are taking a more active role in decisions relating to the hiring and retention
of external auditors (Clapman 2000). Two studies have examined shareholder ratification of

23

auditors selected by management. Raghunandan (2003) investigates whether shareholder
ratification of a management-selected auditor is affected by the relative magnitude of nonaudit fees
paid to the auditor. Based on the analysis of shareholder voting in a sample of Fortune-1000
companies, Raghunandan hypothesizes that the magnitude of nonaudit fees (relative to audit fees)
will negatively impact the percentage of shareholders voting to ratify the appointment of the
external auditor. Although the findings were consistent with expectations, the results of the study
should be interpreted with caution given that the shareholder ratification rates were very high
(about 97%) even in companies where a significantly large percentage of the total fees paid to
auditors consisted of nonaudit fees. Given the above finding, two questions that arise is what are
the characteristics of companies that actually have shareholders debate as well as ratify the choice
of auditors and secondly how how the composition of the audit committee influences shareholders’
decision to ratify an auditor appointment when nonaudit fees are high.
Raghunandan and Rama (2003) examined this issue and found that shareholders are less
likely to vote against auditor ratification even when the nonaudit fee is relatively high if the audit

committee consists of solely independent directors. However, audit committee expertise (i.e., the
presence of at least one member with accounting or financial expertise) was not associated with
shareholder votes on auditor ratification.
From an external auditor’s perspective, corporate governance and audit committee factors
could significantly influence the client acceptance decision. These factorswill influence the quality
of the financial reporting process, because in light of the heightened scrutiny in the post-Enron era,
auditing firms may be reluctant to associate themselves with clients where management is likely to
engage in egregious financial reporting behavior. Very little professional guidance or prior research
exists on how auditors consider governance factors when accepting new clients. Cohen and Hanno
(2000) was the first study on this issue. Specifically, they relied upon the monitoring framework of
the COSO Report (1992) and POB (1993) to develop an oversight perspective when describing
corporate governance factors. Manipulating the strength of the management control philosophy and

24

the strength of the corporate governance structure in an experimental setting, they asked audit
partners and managers to make a client acceptance recommendation. They found that both the
management control philosophy and the strength of the corporate governance structure
significantly affected client acceptance recommendations, control risk assessments and the extent
of substantive testing. However, they manipulate corporate governance in an aggregate manner,
employ a governance orientation based strictly on the COSO (1992) framework, and do not test the
effect of individual elements such as the independence of the audit committee. Their study suggests
the need to more fully capture the effect of individual components of corporate governance on
various stages of the audit process.
Collectively, the studies on auditor selection suggest that stronger governance structures
are associated with the selection of higher quality auditors. Further, the studies suggest that
although shareholder ratification is influenced by the provision of nonaudit services, stronger audit
committees may be viewed by stockholders as effective monitors to ensure that audit quality is not
compromised due to the provision of significant nonaudit services. Further, because of the paucity
of existing studies, more research needs to be conducted examining how the judgments of auditors

are affected by governance factors during the client acceptance process. For example, will auditors
view a potential client that has a board with a strong and effective monitoring focus as a client that
has a lower r likelihood of revealing overly aggressive reporting behavior by management and thus
be a client that is more desirable to actively seek? Finally, with the requirements of the Sarbanes-
Oxley Act that the audit committee select the external auditor, research is needed on factors that
impact their decision.
Audit Quality and Audit Fees
Since audit committees can play a key role in auditor selection, audit committee members’
assessment of audit quality is an important issue to consider. Knapp (1991) examined the impact of
auditor firm size (Big Eight versus local CPA firm), length of auditor tenure, and general audit
strategy (structured versus unstructured) on audit committee members’ assessment of audit quality.