Attributes and Structure of An Effective Board of Directors: A Theoretical Investigation

ABACUS, 2018 doi: 10.1111/abac.
12132
ADI MASLI, MATTHEW G. SHERWOOD, AND

RAJENDRA P. SRIVASTAVA
Attributes and Structure of an Effective

Board of Directors: A Theoretical
Investigation
Despite the board of directors’ importance, there is a lack of well-

developed theory regarding the interrelationship of the attributes of an
effective board. Based on prior board literature, we develop and analyze,
via Bayesian logic, a theoretical framework of key board attributes
(Independence, Competency, Diligence, and Behavioural Attitudes), plus
general board safeguards. We base our framework on the non-linear
‘and’ relationship, among board attributes, instead of the linear ‘or’
relationship, as currently considered in academic research and assumed
by board independence standard setters like the Securities Exchange
Commission. Using inferential statistics we demonstrate that a board
requires the presence of each attribute, plus safeguards to be effective,
and no level of one attribute can substitute for the non-existence of
another.
Key words: Attributes of board of directors; Bayesian Model; Board of
directors; Corporate governance.
Corporate boards worldwide have come under increasing scrutiny due to growing
social protest and concerns about the performance and governance of
corporations (Ingley and Van der Walt, 2001; Hendry and Kiel, 2004; Kumar and
Zattoni, 2014). In their editorial piece, Kumar and Zattoni (2014) emphasize that
the role and effectiveness of the board of directors has been, and will continue to
be, at the centre of corporate governance research. In this study, we develop and
analyze, through Bayesian logic, a theoretical framework comprised of four key
attributes of the board of directors: independence from top management,
competency in performing board duties, activeness/diligence in carrying out board
ADI MASLI (adi@ku.edu) is an Associate Professor and E&Y CARAT Fellow at the School of
Business, University of Kansas. MATTHEW G. SHERWOOD (msherwood@isenberg.umass.edu) is an
Assistant Professor at the Isenberg School of Management, University of Massachusetts—Amherst.
RAJENDRA P. SRIVASTAVA (rsrivastava@ku.edu) is the Ernst & Young Professor of Accounting and
Information Systems, School of Business, University of Kansas. The authors would like to thank Mike
Ettredge, Theresa Libby, Ted Mock, John Roberts, workshop participants at Deakin University,
University of Queensland, University of Sydney, and Wichita State University, and attendees of the
2013 American Accounting Association Strategic and Emerging Technologies Section meeting, the
2013 International Conference on Corporate Governance, co-sponsored by the University of Essex and
Sri Venkateswara University, and the 2014 EY Center for Auditing Research and Technology
Research Conference for their helpful comments on a previous version of the paper. Professors
Srivastava and Masli acknowledge research support from the EY CARAT at The University of Kansas.
1
© 2018 Accounting Foundation, The University of Sydney
ABACUS
responsibilities, and having the appropriate behavioural disposition towards fellow

board members. We assess the impact of the association between these attributes
on board effectiveness using the non-linear ‘and’ relationship, instead of the linear
‘or’ relationship common in academic research. Within our framework, we take
into account the effect of the construct of public and private safeguards in place to
limit conduct that violates the board’s fiduciary duties. We also demonstrate its
value in enhancing the board of directors’ effectiveness. While the attributes in
our framework are not new, this is the first study, to our knowledge, to utilize
applied decision making and inferential statistics to examine and define the
relationship between them. We further examine how the attributes’
interconnection affects the overall effectiveness of the board, and how external
observers can utilize these inferences when assessing the strength of a board.
On 8 July 2002, the US Senate’s Permanent Subcommittee on Investigations
issued a report entitled ‘The Role of the Board of Directors in Enron’s Collapse’.
The report highlights a number of significant fiduciary failures by Enron’s board,
including a lack of independence between board members and management,
inactivity in the performance of key fiduciary duties, and a lack of strong intra-
board behaviours, such as questioning one another and providing constructive
criticism of suggestions made by fellow board members.1 The report identifies
over a dozen red flags that should have caused Enron’s board to ask hard
questions, examine policies, and consider changing course. However, the board
simply ignored the red flags. We quote from page 55 of the report’s conclusion:
In too many instances, by going along with questionable practices and relying on
management and auditor representations, the Enron Board failed to provide the
prudent oversight and checks and balances that its fiduciary obligations required and
a company like Enron needed. By failing to provide sufficient oversight and restraint
to stop management excess, the Enron Board contributed to the company’s collapse
and bears a share of the responsibility for it.
Less than a year later WorldCom Inc., which like Enron went bankrupt following
a significant financial statement fraud, issued a report regarding its board of
directors (WorldCom, Incorporated, 2003). The report identifies numerous
characteristic similarities between the WorldCom and Enron boards. While
ultimately concluding the board was unaware of the fraud, the report notes
significant deficiencies in its role of direction and cultural supervision of the
company. The reports conclude that while appearing to satisfy the checklist
mentality, the board failed to have the necessary energy, leadership, or courage.
These board failures either contributed to (i.e., Enron’s board), or failed to
impede (i.e., WorldCom’s board), two of the largest frauds and subsequent
bankruptcies in the history of American industry and played a prominent role in the
passage of the Sarbanes-Oxley Act. While the above discussion presents only two
1
In this case, inactivity does not relate to the number of meetings per se, rather it implies an
unwillingness to take proactive action on behalf of the stakeholders. In effect the board members
were willing to turn a blind eye to the firm’s true state of affairs.
2
ATTRIBUTES AND STRUCTURE OF AN EFFECTIVE BOARD OF DIRECTORS
examples of the potentially severe consequences of poor board behaviour, there are
numerous others.2 It is clear a connection exists between the characteristics of the
board of directors and the potential for extremely negative outcomes.
It is widely recognized that corporate governance arrangements, in which the
board of directors plays a primary role, substantially influence corporate decisions
and shareholder wealth (e.g., Core et al., 1999; Cyert et al., 2002; Klein, 2002; Deutsch
and Ross, 2003; Larcker et al., 2007; Bebchuk et al., 2009; Fisman et al., 2013).3
Hendry and Kiel (2004) suggest that boards play an important role in corporate
strategy, noting that the old perspective in which boards were seen as ‘rubber stamps’
has been overtaken by an ‘active’ perspective in which boards are viewed more as
independent thinkers shaping the strategic direction of their organizations.
Additionally, Ingley and Van der Walt (2001) suggest that boards have been under
pressure to develop a broader mindset and new skills to deal with the uncertainty of
higher-level issues such as direction giving and strategy implementation.
Prior research indicates that when properly designed and implemented by an
effective board of directors, corporate governance can be an assurance mechanism
that restores stakeholders’ trust in corporations (Hilb, 2012). However, prior recent
literature also calls into question the degree to which researchers are developing
and testing theory regarding the role of corporate governance and the board of
directors. In particular, Brown et al. (2011) voice this concern in their review of
corporate governance indices, such as the board of directors. A similar sentiment is
expressed by Larcker et al. (2007), who also raise concerns regarding the lack of
well-developed theory surrounding the complex and multi-dimensional nature of
various aspects of corporate governance. Further, Van den Berge and Levrau
(2004) observe that ‘traditional’ academic research has focused on only a limited
number of quantifiable board characteristics, while practitioners attach greater
importance to ‘soft’ elements, which are nearly absent in the academic literature.
Van den Berge and Levrau (2004) consequently highlight the need for a better
understanding of all elements that determine board effectiveness. Leblanc and
Schwartz (2007) also lament the fact that while ‘board process’ has been identified
as a critical factor for corporate governance research, gaining access to corporate
boardrooms is extremely difficult, if not virtually impossible, for most researchers.
While prior literature lacks theory on the collective intra-workings of the board
of directors, it does provide insight into the attributes that influence board
performance. These attributes include independence from management,
possessing the competency to perform key board duties, and maintaining an
2
Other financial frauds include, but are not limited to Olympus, Parmalat, Satyam, and Madoff,
which have resulted in hundreds of billions of dollars in losses to the global economy (Brickey, 2003;
Anand, 2009). In addition, we have recently witnessed a global financial crisis, which resulted in
trillions of dollars in losses to society and threatened the very existence of the financial markets,
financial institutions, and, in some instances, national governments (Claessens et al., 2010;
Helleiner, 2011).
3
Corporate governance involves a set of relationships between a company’s management, its board,
its shareholders, and other stakeholders (OECD, Principles of Corporate Governance, revised
May 2004).
3
ABACUS
appropriate level of activity (or diligence) in carrying out its duties (e.g., NACD,
1996; Carcello et al., 2002, Deutsch and Ross, 2003; Cohen et al., 2012; Cicero
et al., 2013). Academics (e.g., Finkelstein and Mooney, 2003; Huse, 2005; Roberts
et al., 2005) also suggest that characteristics pertaining to the inner-workings of the
board, as well as directors’ behavioural disposition and professional relationships
with one another influence board effectiveness.4
Following prior studies (e.g., Stewart and Kinney, 2013; Srivastava et al., 2012;
Srivastava et al., 2009), we use Bayesian logic to assess our theoretical model of
board effectiveness. The theoretical framework is comprised of the four primary
attributes noted in the preceding paragraph. Prior research examines these board
attributes primarily from an ex post point of view (e.g., Cicero et al., 2013;
Boone et al., 2007; Knyazeva et al., 2013). Moreover, the effects of these
attributes on corporate performance have been empirically investigated only by
considering one or two attributes at a time and thus it is difficult to draw
conclusions about whether the presence or absence of the other attributes at the
same time strengthens or weakens the effectiveness of the board. Prior research
also lacks an ex ante theoretical discussion regarding how the interrelationships
among these attributes influence each other and the overall effectiveness of the
board.
In addition, some rating agencies such as Corporate Library (see Van den Berghe
and Levrau, 2004 for details) assume a linear relationship between board attributes
when assessing board effectiveness. Each board attribute is sub-divided into a
number of sub-attributes, and each sub-attribute is scored based on if it is present.
The overall attribute score is then obtained by summing the sub-attribute scores per
attribute. The linear (i.e., ‘or’) relationship implies that one can substitute one
attribute for another. However, when one tries to interpret the linear relationship,
the attribution substation approach appears illogical. For example, consider the
following situation where all board members are independent, regularly meet with
the CEO and CFO to discuss the company’s strategies, and have the appropriate
behavioural characteristics, but are incompetent in their ability to carry out their
board duties. A linear relationship could identify this as an effective board if the
attribute level among the three present attributes is great enough to substitute for
the missing fourth attribute. In fact, however, the board is unable to be truly
effective, as it lacks the inherent ability it needs. As a result, we only consider the
‘and’ relationship among the key board attributes in our analytical model.
Through the use of our Bayesian model, we demonstrate that the board is most
effective if—and only if—the directors are independent, competent, active, and
display appropriate behavioural attributes. While this result is far from surprising,
we also extend the use of inferential statistics to identify the probable impact on
board effectiveness when any combination of these attributes is diminished or
absent from the board. Thus making this the first paper, to our knowledge, to
4
Roberts et al. (2005) observe that research on corporate governance and boards of directors lacks
understanding of the behavioural processes within boards and has not fully considered such factors
in evaluating board effectiveness.
4
provide evidence on the implied inferential influence of the presence or absence

of the four board attributes on board effectiveness.
As mentioned earlier, in addition to the four primary board attributes, we also
introduce into our framework the construct of safeguards (controls) within the
board, as an important component of the board’s effectiveness. These safeguards
are represented by mechanisms (both internal and external to the firm) that can
regulate or influence a board members actions. These include a director’s level of
attentiveness to how the board’s actions can threaten his or her reputation,
heighten his or her potential of litigation, or subject him or her to regulatory
enforcement. The effectiveness of safeguards varies between boards, as not all
board members view potential risks in a similar manner. Additionally, the weight
each board member puts on the safeguards is likely to vary among the four
attributes. To address these concerns in our model, safeguards function as a direct
control to board effectiveness, and not as an individual board attribute.
Ultimately, boards are most effective when a high level of safeguards accompany a
strong presence of board independence, competence, activity, and appropriate
behavioural attributes. It is important to note that some of the current literature
identifies safeguards, that is, control mechanisms, for controlling board
effectiveness as ‘accountability’ and considers them to be an attribute of individual
board members. We argue that safeguards are not an attribute of individual board
members per se, rather they are a control mechanism influencing each key
board attribute. Although it is an empirical question, one can argue that if a board
member knows he or she will be held accountable, he or she will ensure that he or
she possess, and can demonstrate, each attribute required to fulfill his or her
board responsibilities prior to accepting a board position.
This paper contributes in a number of ways to the extant literature on the
attributes that comprise an effective board of directors. First, using prior
literature, we develop a theoretical framework of core attributes required in order
for a board of directors to be effective. We then apply inferential statistics,
through a Bayesian methodology, to examine how the presence or absence of
these attributes (with varying levels of certainty) influence board effectiveness.
This answers the calls for research to advance formal theoretical work on
corporate governance and the board of directors (Larcker et al., 2007; Brown
et al., 2011). At the same time, the use of applied logic allows us to take an ex ante
point of view in answering questions that prior studies fail to address. These
questions include, ‘How does the presence or absence of a core attribute, with
certainty, influence board effectiveness?’; ‘Can perceived strength in one attribute
compensate for perceived weakness in another?’; ‘To what degree would
enforcing boards to absolute adherence of these attributes provide additional
benefit to the outside stakeholders the board represents?’; and ‘To what degree do
external safeguards influence the collective board’s actions, when properly
implemented?’. We present these questions not as formal research questions, but
as examples of the type of questions one might consider addressing via Bayesian
logic and address them via multiple case scenarios below. Further, this is the first
paper, to our knowledge, to provide theoretical support to the notion that it is
5
ABACUS
essential for researchers, when assessing the effectiveness of the board’s function,
to implement the ‘and’ relationship, not the ‘or’ relationship among the board
attributes of interest.
While the primary support for our choice of attributes comes from the finance
and accounting literature, due to the diverse nature of the subject, we intentionally
exclude any discipline-specific focus from the paper. The mathematical methods
we employ and the conclusions we draw are equally applicable to all fields of
study. Further given the influence of board effectiveness on corporate outcomes is
of importance to all stakeholders, our study should be relevant beyond the
academic realm. Regulators, auditors, executives, investors, and even current
board members should benefit from both the framework we present and the
conclusions from our Bayesian analysis. Using the interrelationships of four core
attributes, we provide a unique manner in which all stakeholders can evaluate
potential board effectiveness.
ATTRIBUTES OF AN EFFECTIVE BOARD OF DIRECTORS
The principles of agency theory imply that agents (i.e., management) may be
reluctant to impose strict and binding mechanisms of corporate governance that
limit the agents’ ability to act in their best interest (Ross, 1973; Eisenhardt, 1989).
As a result, there arises a need for stakeholders to have an entity on the inside to
play a role in governing and monitoring the firm’s actions on the stakeholder’s
behalf, that is, the board of directors (Baysinger and Butler, 1985; Hillman and
Dalziel, 2003; Fisman et al., 2014; Kuhner and Pelger, 2015). Prior research
establishes that boards face a dual set of responsibilities that can, and do, compete
with one another to serve as the board’s primary area of focus. In particular,
companies elect a board to provide operating guidance to the firm’s management
team and/or serve as a monitoring mechanism over firm management (e.g., Cyert
et al., 2002; Boone et al., 2007; Dontoh et al., 2013; Baldenius et al., 2014; Fisman
et al., 2014; Kim et al., 2014).
The board of director attributes we describe and employ throughout this paper
are relevant to boards focusing on either monitoring management or providing
guidance to management. Boards do not treat these two objectives as mutually
exclusive (Kim et al., 2014), nor is there a ‘one-size fits all’ model applicable to all
boards (Wintoki, 2007; Coles et al., 2008). As a result, there is clearly overlap in the
characteristics of importance to both the monitoring and the guidance functions.
Thus, the same basic characteristics are of significant importance regardless of the
board’s primary focus (Coles et al., 2008; Cicero et al., 2013; Kim et al., 2014).
Prior research (e.g., Cotter and Silvester, 2003; Gompers et al., 2003; Deutsch,
2005; Brown et al., 2011; Roberts, 2012) suggests that efficacy of the board of
directors can be influenced by certain characteristics, including independence (I) of
the board members from management, competence (C) or expertise of the board
members (e.g., knowledge about business processes within the firm, particular fields
6
(accounting, law, etc.), industry expertise, etc.), active (A) board members (i.e., the
members frequently meet with the CEO and CFO to monitor their activities and
provide advice, are diligent in attending board meetings and being active in those
meetings), and appropriate behavioural attributes (B) by directors to act cohesively
and constructively. We discuss each of these four attributes in turn.
Independence of Board Members

Prior studies suggest that more independent boards act in the best interest of
shareholders and have stronger incentives to prevent and detect opportunistic
behaviour by management (Fama and Jensen, 1983; Beasley, 1996; Cotter et al.,
1997). Gordon (2007) reports that between 1950 and 2005 the composition of
large public company boards shifted from approximately 20% to 75%
independent. He suggests that this dramatic increase is due to the increasing
importance of shareholder value as the primary corporate objective and the
greater information content in relation stock prices. He further notes that when
the company commits to a shareholder maximizing strategy, independent directors
are more valuable than insiders.
To enhance the corporate governance of publicly traded companies, the
Sarbanes-Oxley Act (SOX) of 2002 and major US stock exchanges have promoted
and imposed requirements for board independence (Public Law 107-204;
Chhaochharia and Grinstein, 2007). For example, Section 301 of SOX requires that
each member of the audit committee be independent. The rule changes adopted by
the NYSE and NASDAQ require firms to have a majority of independent board
members and entirely independent compensation and nominating/governance
committees (Engel et al., 2005; Linck et al., 2008). Linck et al. (2008) find that boards
of directors during the post-SOX period are more independent and more likely to
separate the two posts of CEO and chairman of the board.
Boone et al. (2007) examine changes in board structure as firms age from IPO to
10 years of age. They find that as firms age and grow the board takes on more of a
monitoring role as evidenced by firms adding independent board members, as well
as firms increasing board size. However, they also find that boards of growing
firms add independent board members to tap into outside expertise as the firm
moves into new regions or diversified service offerings. Bertoni et al. (2014) also
examine the role of board independence in the valuation of IPO firms. They find
that in young companies, the value-creation role of the board dominates, and the
impact of board independence on firm valuation decreases with firm age; in
mature companies, the value-protection role of the board dominates, and the
impact of board independence on firm valuation increases with firm age.
A number of other studies show the value of having independent directors. For
example, Beasley (1996) finds that the inclusion of larger proportions of outside
members on the board of directors reduces the likelihood of financial statement
fraud.5 Klein (2002) shows a negative relation between board independence and
5
Examining a sample of US firms from 1978 to 2001, Uzun et al. (2004) find that firms with more
independent boards and board sub-committees are associated with lower levels of corporate fraud.
7
ABACUS
abnormal accruals. Further, she finds significant increases in abnormal accruals

accompany reductions in board or audit committee independence. Using a sample
of firms in the United Kingdom, Peasnell et al. (2005) find the likelihood of
managers making income-increasing abnormal accruals to avoid reporting losses
and earnings reductions is negatively related to the proportion of outsiders on the
board. Anderson et al. (2004) find that board and audit committee independence
are associated with lower cost of debt financing, indicating that creditors place
importance on director independence in monitoring financial accounting reports.
Using a sample of UK firms, Osma (2008) posits and finds that more independent
boards are more efficient at detecting and constraining manipulation of research
and development (R&D) expenditure.
More recently, Knyazeva et al. (2013) show that proximity to larger pools of
local director talent leads to more independent boards. Using the local pool of
potential directors as an instrument for determining board independence, they
find that board independence has a positive effect on firm value, operating
performance, fraction of CEO incentive-based pay, and CEO turnover.
Finally, a number of studies examine ties between directors and the
management team. Chen (2011) finds that the effects of top management team
tenure and international experience on internationalization are stronger as more
independent directors are included on the board, suggesting that independent
directors (as monitors and resource providers) offer better counsel to executives
and enhance their strategic decision capabilities. Cohen et al. (2012) investigate
firms that appoint independent directors who are overly sympathetic to
management while still technically independent according to regulatory
definitions. They find that such firms are more likely to have management on the
board nominating committee, appear to be poorly governed, and more likely to
increase both earnings management and CEO compensation following these
board appointments. Nguyen (2012) further shows that when the CEO and a
number of directors belong to the same social networks, the CEO is less likely
to be dismissed for poor performance. Such findings suggest that close social ties
between board members and CEOs impact the workings of the board of
directors.
Activity of Board Members

The second attribute that we examine is the activity level of the board of directors.
As noted by Carcello et al. (2002, p. 371), the activity or diligence of the board
consists of ‘factors such as the number of board meetings and the behaviour of
individual board members surrounding such meetings (e.g., preparation for
meetings, attentiveness and participation during meetings, and post-meeting
follow-up)’. Regulators and academics largely support the proposition that high
activity level and diligence in completing board duties increase board effectiveness
(Lipton and Lorsch, 1992; Vafeas, 1999; Carcello et al., 2002). The Blue Ribbon
Committee, for example, advocates that the audit committee of the board can
8
provide better assurance over the quality of the financial statements by having at
least four meetings a year (Anderson et al., 2004).
The vast majority of academic empirical work measures board activity and
diligence by the number of board meetings, finding associations between such
measure and economic outcomes. For example, Vafeas (1999) finds that share
price declines are followed by higher meeting frequencies and years with an
abnormally high meeting frequency are followed by improvements in operating
performance. MacAvoy and Millstein (1999) find boards that frequently meet help
companies produce higher rates of return for their shareholders, as measured by
earnings in excess of the cost of capital and net of the industry average.
Furthermore, Anderson et al. (2004) find evidence suggesting that audit committee
activity is quite important to creditors. Specifically, their findings reveal that firms
with more audit committee meetings are associated with a significantly lower cost
of debt financing.
A stream of research also associates board meeting frequency with better
monitoring effectiveness. For example, Carcello et al. (2002) find that diligent
boards promote shareholder interests by purchasing higher-quality audit
services differentially. In addition, Xie et al. (2003) show that board and audit
committee meeting frequencies are associated with reduced levels of earnings
management.
Competence of Board Members

The third board effectiveness attribute is competence or the ability of the board.
Even if the mix of independent-to-non-independent board members is ideal and
these board members are actively engaged in performing their duties, if the board
does not comprise individuals properly qualified to provide guidance to, and
monitoring of, management, the board will ultimately be ineffective. According to
Ingley and Van der Walt (2001), capabilities such as strategic vision and
leadership are key requirements for directors, but in practice there is a variation
among individual directors in terms of their competence in this regard.
Deutsch and Ross (2003) examine reputation of directors as a signalling
mechanism for young firms. Using analytical models, they demonstrate that in the
face of a market failure in which stakeholders refuse to align themselves with new
firms, high-quality new ventures may attain credibility by appointing reputable
directors to their boards. Some have argued that directors who hold multiple
directorships have made a significant investment in developing their reputation
capital as competent directors (e.g., Fama and Jensen, 1983; Carcello et al., 2002;
Abbott et al., 2003). Ferris et al. (2003) note that firm performance, during a
director’s tenure, has a positive effect on the number of board seats a director
subsequently obtains, thus suggesting that reputation matters in the market for
directors. Additionally, some studies show benefits to multiple directorships. For
example, Cotter et al. (1997) show that boards with directors holding multiple
directorships can obtain larger premiums in tender offers for their shareholders.
Carcello et al. (2002) find that boards in which multiple directorships are common
9
ABACUS
are more likely to protect shareholder interests and purchase differentially higher-
quality audit services.6
Regulatory provisions and prior academic work also associate competence with
expertise or knowledge held by directors. For example, the SEC acknowledges the
importance of financial expertise on the audit committee and requires companies
to disclose whether they have at least one ‘audit committee financial expert’
serving on their audit committee. In this case, such an expert has characteristics
that are particularly relevant to the functions of the audit committee, such as a
thorough understanding of the audit committee’s oversight role, expertise in
accounting matters as well as understanding of financial statements, and the ability
to ask the right questions to determine whether the company’s financial
statements are complete and accurate (SEC, 2003).7
Studies have shown value from having directors with specific expertise. Kroll et al.
(2008) show that boards comprised of directors that are vigilant as well as having
appropriate knowledge gained through experience are better monitors and more
useful advisors to top managers. DeFond et al. (2005) find a positive market reaction
to the appointment of accounting financial experts assigned to audit committees,
suggesting that the market supports accounting-based financial skills to improve the
audit committee’s ability to ensure high-quality financial reporting. Other works also
show the positive effects of having financial and legal experts on the audit committee
on internal controls and financial reporting quality (e.g., Carcello et al., 2002;
Krishnan, 2005; Krishnan and Visvanathan, 2008; Krishnan et al., 2011).
More recently, Kim et al. (2014) suggest that a director’s tenure with the firm is
positively associated with the director’s competency to advise and monitor firm
performance. More specifically, they contend that ‘longer director tenures reflect
more board and committee meetings attended, likely increased committee
assignments, greater experience with the firm’s strategies and policies, and greater
within-firm deal-level experience’ (Kim et al., 2014, p. 111). Consistent with their
hypotheses, they find outside director tenure to be associated with better firm
acquisition/investment policy advising performance, lower CEO rent extraction,
and higher operating performance measures.
Behavioural Attributes of Board Members

The final attribute necessary for board effectiveness is the behavioural attributes
and relationships among directors. The board’s behavioural attributes can be
6
Ferris et al. (2003) test the hypothesis that directors who serve on multiple boards become so busy
that they cannot monitor management adequately (i.e., the ‘busyness’ hypothesis). Their results
suggest that market participants do not view the appointment of a multiple director as a negative
event for the firm; multiple directors actually sit on more committees and attend more meetings; no
statistically significant evidence exists that firms with multiple directors are more vulnerable to
securities fraud litigation. They conclude that the evidence does not support the proposition that
multiple board membership harms firm value.
7
Linck et al. (2008) find that, in the post-SOX period, board members are more likely to be lawyers,
consultants, financial experts, or retired executives.
10
broadly defined as the ability of the board to act as a cohesive unit, allow for
constructive suggestions among its members, work well together as a team, and
foster a positive intra-working relationship among team members (Forbes and
Milliken, 1999; Finkelstein and Mooney, 2003; Roberts, 2012). Using in-depth
interviews with company directors, Roberts et al. (2005) examine the behaviours
and relationships of non-executive directors to evaluate board effectiveness. The
authors suggest that while board structure, composition, and independence each
play an important role, it is ultimately the actions and conduct of individual board
members that determine board effectiveness. Specifically, they suggest that a
variety of behaviours—challenging, questioning, probing, discussing, testing,
informing, debating, and exploring—are crucial for directors to be effective.
Forbes and Milliken (1999) integrate research on workgroup effectiveness with
research on the board of directors. They identify group cohesiveness (agreement
on the common goals of the board) along with the degree to which task-oriented
disagreement (constructive criticism of another’s idea) is permissible among board
members, as necessary elements in assessing board effectiveness. Furthermore,
Finkelstein and Mooney (2003) identify five interrelated board behavioural
processes that can make boards more effective. These are engage in constructive
conflict, avoid destructive conflict, work together as a team, know the appropriate
level of strategic involvement, and address decisions comprehensively.
Safeguards within the Board of Directors

Directors are attentive to protecting their reputational capital (Fama and Jensen,
1983; Gilson, 1990). Indeed, prior academic work highlights that directors who are
associated with corporate catastrophes and fail to exercise reasonable care in
fulfilling their responsibilities are often subject to labour market penalties (Fama,
1980; Gilson, 1990; Carcello et al., 2002). Srinivasan (2005), for example, finds that
director (especially audit committee) turnover is higher for firms that restate their
earnings (i.e., overstated their earnings) compared to non-restating firms.8 Fich
and Shivdasani (2007) investigate the reputational impact of financial fraud for
outside directors, finding that directors of sued fraud-committing firms experience
a significant decline in board seats held by other companies.
Directors are also attentive to avoiding or minimizing legal liability (Gilson, 1990;
Kesner and Johnson, 1990; Carcello et al., 2002). Recent lawsuits against directors
of companies such as Target, Groupon, and News Corp. highlight examples of
current cases where directors are targets of shareholder lawsuits. As suggested by
Brochet and Srinivasan (2013), directors named as defendants in securities lawsuits
face the possibility of financial and reputational harm, lost time, and emotional
distress. Brochet and Srinivasan (2013) also show that directors named in lawsuits
receive more negative recommendations from the proxy advisory firm and attain
significantly more negative shareholder votes than directors in a benchmark sample.
8
Srinivasan (2005) also shows that directors of firms with overstated earnings are no longer present in
25% of their positions on other boards.
11
ABACUS
Furthermore, directors are often wary of conducting improprieties that could result
in predicaments with regulatory authorities (i.e., SEC, IRS, etc.). Section 305 of the
Sarbanes-Oxley Act, for example, states that the SEC may issue an order to
prohibit any person from acting as a director of an issuer if the SEC has found that
such person’s conduct ‘demonstrates unfitness’ to serve as a director of any such
issuer (Marden and Edwards, 2005).
We contend that safeguards greatly influence board effectiveness. Our concept of
safeguards pertains to the strength of mechanisms that regulate and oversee the
board’s conduct. The mechanisms include numerous factors, such as: the company’s
own disciplining procedures for board misconduct; extensiveness of corporate
bylaws and code of ethics; monitoring over the board by shareholders, institutional
investors, and analysts; penalties in the labour market for directors; professionalism
guidelines, board evaluations; and industry-specific and/or market-wide regulations.
Safeguards can also reflect individual director’s (and the overall board’s) proclivity
to guard against actions and decisions that can jeopardize reputational capital,
heighten litigation risk, and violate regulatory provisions or rules of conduct. We
treat this construct of safeguards as a direct control to board effectiveness because
the level of safeguards has a direct impact on each of the other four attributes
(i.e., independence, competence, activity, and behaviour attributes).
We maintain that the level of safeguards can vary from one board (or individual
director) to another. While all directors should ideally be concerned about their
reputation capital, litigation risk, and compliance with regulatory provisions, it is
likely some directors are less alarmed by such issues than others. Based on a
sample of securities class-action lawsuits from 1996 to 2010, Brochet and
Srinivasan (2013) show that only a minority of independent directors has been
named as defendants. Black et al. (2006) argue that litigation risk for directors
may be exaggerated and report that outside directors of US public companies who
fail to meet their fiduciary duties almost never face personal liability losses
because of such factors as directors’ and officers’ (D&O) liability insurance.
BAYESIAN MODEL FOR THE EFFECTIVENESS OF THE BOARD OF

DIRECTORS
In this section, we develop a theoretical model for the effectiveness of a board of

directors using the Bayesian framework in terms of the basic attributes of board
effectiveness. We use the Mautz and Sharaf (1961) philosophical approach to
determine the basic attributes of the board and their relationship to board
effectiveness. As established by several researchers (e.g., Brown et al., 2011;
Gompers et al., 2003; and Roberts, 2012), and as discussed in the previous sections,
the board of directors’ effectiveness is determined through the presence of key
attributes, such as independence (I) of board members from the management,
competence (C) of board members in carrying out their duties, board members
being active (A) in the fulfilling of their assigned board tasks and possessing
12
appropriate behavioural attributes (B) that allow them to act cohesively yet
constructively with other board members. As discussed in the previous section, we
consider an additional construct ‘safeguards’ for the effectiveness of the board. We
argue that the construct ‘safeguards’ is not a basic attribute of the board members;
rather it is a control mechanism governing the basic attributes of the board
members. Some examples of ‘safeguards’, such as legal liability against board
members and reputation loss of board members of companies that are involved in
fraudulent behavior, were discussed in the previous section.
The current literature on the characteristics of an effective board simply
identifies the above attributes (independence, I; competence, C; active members,
A; and behavioural attributes, B) without providing any insights into their
interrelationships. In addition, as mentioned earlier, the significance of these
attributes for board effectiveness has been established through empirical research
by considering one or two attributes at a time and thus it is difficult to draw
conclusions about whether presence or absence of the other attributes at the same
time would strengthen or weaken the board’s effectiveness. The question is how
should these factors be related to board effectiveness?
As argued in the introduction, we use the ‘and’ relationship among these attributes
that influence board effectiveness. In other words, the board will be most effective
only when all four attributes, I, C, A, and B, are present. That is, the board is most
effective if and only if the board members are independent (I), competent (C), active
(A), and the members possess appropriate behavioural characteristics (B). We
analyze various scenarios in this section after developing the probabilistic model. It is
important to point out that we employ a probability model because of the
uncertainties involved in assessing whether certain attributes are present or absent.
The ‘and’ relationship can be expressed in terms of the following set notations:
eb = i \ c \ a \ b,
where ‘eb’ represents the state that the board is effective, ‘i’ represents the state
that BOARD members are independent, ‘c’ represents the state that board
members are competent, ‘a’ represents the state that board members are active,
and ‘b’ represents the state that board members possess appropriate behavioural
characteristics. In terms of the negations of the above states, one can write the
above relationship as an ‘or’ relationship:
eb = ð i \ c \ a \ bÞ [ ði\ c \ a \ bÞ [ ði \ c\ a \ bÞ [ ði \ c \ a\ bÞ
[ ð i\ c \ a \ bÞ [ ð i \ c\ a \ bÞ [ ð i \ c \ a\ bÞ [ ði\ c\ a \ bÞ
[ ði\ c \ a\ bÞ [ ði \ c\ a\ bÞ [ ð i\ c\ a \ bÞ [ ð i\ c \ a\ bÞ
[ ð i \ c\ a\ bÞ [ ði\ c\ a\ bÞ [ ð i\ c\ a\ bÞ:
The above equation implies that the board will be ineffective (~eb) if any of the
attributes are absent. In the present discussion, we have assumed all the attributes
to be binary variables, which is rarely the case. Thus, we want to develop a
13
ABACUS
formula for the posterior probability that the board is effective given that we have
information regarding the degree to which each of the four attributes, I, C, A,
and B, are present.
Similar arguments are common in audit and information systems literature. For
example, the audit literature argues that an account is fairly stated if and only if all
its management assertions such as existence, completeness, valuation, rights and
obligation, and so on, are true (see AICPA’s SAS 106, 2006). In developing a
fraud risk assessment formula, Srivastava et al. (2007) have used the ‘and’
relationship among the fraud triangle factors, incentives to benefit from fraud,
attitude to commit fraud, and opportunity to commit fraud, with the commitment
of fraud by management. All three fraud factors have to be present for
management to commit fraud. Desai et al. (2010) have used similar arguments in
determining the strength of the internal audit function. They argue that the
internal audit function will be strong if and only if the internal auditors are
competent, they are objective, and their work is of high quality.
Figure 1 depicts an evidential diagram (Srivastava, 2011) for the problem at
hand. The main assertion that the board is effective is related to four attributes
(i.e., constructs) through the ‘and’ relationship. In general, they are known as
variables. These variables are assumed to be binary and are represented by
rounded boxes. The rectangular boxes in Figure 1 represent items of evidence
pertaining to the variables to which they are connected. We have derived a
general formula, equation (A9), in Appendix A, for the posterior probability of
the main variable being true as a function of the strength of evidence pertaining to
the main assertion and each sub-assertion expressed in terms of the likelihood
FIGURE 1
EVIDENTIAL DIAGRAM FOR BOARD OF DIRECTOR’S EFFECTIVENESS WITH FOUR

ATTRIBUTES: INDEPENDENCE (I), COMPETENCE (C), ACTIVE (A), AND APPROPRIATE
BEHAVIOURAL CHARACTERISTICS (B)
BOD is Indepedent (I) E I : Evidence pertaining to I

{i, ~i}
E1EB: Direct Evidence pertaining to

Safeguards related to BOD Members being effective
Board of Directors (BOD) is Effective (EB) BOD is Competent (C) EC : Evidence pertaining to C
AND
{eb, ~eb} {c, ~c}
E2 EB: Direct Evidence pertaining to BOD BOD is Active (A) EA : Evidence pertaining to A
{a, ~a}
BOD Members possess

appropriate Behavioral EB : Evidence pertaining to B
Characteristics (B)
{b, ~b}
The rectangular boxes with rounded corners represent the attributes (variables), and the rectangular
boxes represent items of evidence pertaining to the variables they are connected to.
14
ratio, λ, and the corresponding prior odds, O. In the case of the board, we have
five variables, the main assertion, EB, representing the effectiveness of the board
with two values {eb, ~eb}; ‘eb’ representing ‘effective board’ and ‘~eb’
representing ‘ineffective board’, and four sub-assertions with two values,
representing the presence and absence of the corresponding variable,
independence (I), competence (C), active (A), and appropriate behaviour (B).
The corresponding items of evidence are represented by E1EB, E2EB, EI, EC, EA,
and EB, and the related likelihood ratios, representing the strength of evidence,
are given by λ1EB = P(E1EB|eb)/P(E1EB|~eb), λ2EB = P(E2EB|eb)/P(E2EB|~eb),
λI = P(EI|i)/P(EI|~i), λC = P(EC|c)/P(EC|~c), λA = P(EA|a)/P(EA|~a), and λB = P
(EB|b)/P(EB|~b). The corresponding prior odds, O’s, and the related β’s are given
by: OEB = P(eb)/P(~eb), OI = P(i)/P(~i), OC = P(c)/P(~c), OA = P(a)/P(~a), and
OB = P(b)/P(~b), and βEB = λ1EBλ2EBOEB, βI = λIOI, βC = λCOC, βA = λAOA, and
βB = λBOB. Substituting these values in equation (A9) with n = 4, we obtain the
following expression for the posterior probability that the board is effective given
all the information about all the variables:
βI βC βA βB βEB
PðebjEI EC EA EB E1EB E2EB Þ = P P P
1+ i βi + i, j, i6¼j βi βj + i, j, k, i6¼j6¼k βi βj βk + βI βC βA βB β EB
ð1Þ
where i, j, k 2{I, C, A, B}.

Equation (1) is a general expression for the posterior probability for the
strength of the board being effective given that we have evidence related to all
the variables (attributes) and the two items of evidence, E1EB and E2EB at the
main assertion level. In our model, we consider the evidence E1EB to represent
the presence or absence of the factor ‘safeguards’, which serves as a control and
the evidence E2EB represents the other potential evidence that pertains to the
overall board effectiveness but does not pertain to ‘safeguard’.9 The presence or
absence of safeguards impacts all the attributes. For example, if there is a
strong legal liability punishable by jail terms for not serving on the board
effectively or being negligent, the members will not serve on the board if they
think they lack the competence or cannot meet the challenges of the
responsibilities for fear of being prosecuted for being negligent. Under such a
situation, even if they decide to serve, they will make sure that they maintain
independence, and acquire the required competence and possess other attributes
to be effective.
Next, we use equation (1) to analyze the impact of various attributes being
present or absent on the posterior probability of board effectiveness. Also, we use
equation (1) to analyze the impact of ‘safeguards’ on board effectiveness. We want
9
This item of evidence is added to make the model more general. If such an item of evidence is not
obtained or considered then one can just replace the corresponding likelihood ratio, λ2EB, in
equation (1) by 1.0 (i.e., λ2EB = 1).
15
ABACUS
to emphasize here that the measure of presence or absence of attributes in our

approach is modelled through the likelihood ratios. An advantage of using a
likelihood ratio that it is a continuous variable and hence is a better measure of
the presence or absence of an attribute than the measures, 1 or 0, an indicator
approach often employed in academic research.
APPLICATION OF BAYESIAN MODEL UNDER VARYING LEVELS

ATTRIBUTE EXISTENCE
Case 1: Impact of Presence or Absence of Attributes with Certainty on Board

Effectiveness
In the case we assume that we have information that I, C, A, and B are present
with certainty, that is, board members are independent, they are competent, they
are active, and they possess appropriate behavioural characteristics to be effective.
This situation is the ideal situation and implies that λI = λC = λA = λB! ∞.
For this case, equation (1) yields the posterior probability, P(eb|
EIECEAEBE1EBE2EB) = 1, irrespective of the value of λ1EB (i.e., irrespective of
the presence of ‘safeguards’) and λ2EB, and for any value of the prior odds for all
the variables. This is a logical result. If we know for sure that board members are
independent of the management, are competent, are active in overseeing the
management, and possess appropriate behavioural characteristics then the board
by definition will be effective. However, when any one or more of the four of the
factors is absent, the board will, by definition, be ineffective. Equation (1) yields
this result. The condition that any one factor is absent is obtained by setting the
corresponding likelihood ratio to zero. For example, if board members are not
independent then λI = 0, that is, βI = 0. This condition yields P(eb|
EIECEAEBE1EBE2EB) = 0, and P(~eb| EIECEAEBE1EBE2EB) = 1, irrespective of
the prior odds and whether the other three factors are present or not present.
Similarly, one can show that P(eb|EIECEAEBE1EBE2EB) = 0, and P(~eb|
EIECEAEBE1EBE2EB) = 1, when any two or more of the factors are absent. The
above results are again logical as one would expect under the ‘and’ relationship.
Next, we discuss situations where the items of evidence are uncertain about the
presence or absence of the variables. Such cases are more interesting because, in
real world situations, it is hard to determine if board members are 100%
independent (I), 100% competent (C), 100% active (A), or 100% possess the
right behaviour (B). It would be an empirical question as to how to measure these
constructs on a scale from 0 to 1 whether they are present or absent.
Case 2: Impact of Presence or Absence of Attributes with Uncertainties on Board

Effectiveness
In this case, we demonstrate the effect of the presence or absence of the attributes
on board effectiveness, especially under the condition when we do not have full
16
FIGURE 2
EFFECTIVENESS OF THE BOARD AS A FUNCTION OF THE STRENGTH OF VARIOUS

ATTRIBUTES OF THE BOARD WITH NO CONSIDERATION OF SAFEGUARDS
1.0
0.9
Posterior Probability of BOD being Effective
1
0.8
0.7
0.6
0.5 2
0.4
0.3 3
0.2
0.1
4
0.0
0 2 4 6 8 10 12 14 16 18 20
Strength of Attributes in terms of Likelihood ratio, λ
The strength of an attribute is represented by the likelihood ratio, λ, corresponding to the attribute
where λ =1 means no knowledge of its presence and λ = 19 means 0.95 probability that the attribute is
present (Probability of attribute being present based on the evidence = λ/(1 + λ)).
1. All the four attributes (Independence, Competence, Active, and Behavioural characteristics) of the
board are present.
2. Any three of the four attributes (Independence, Competence, Active, and Behavioural characteristics)
of the board are present.
3. Any two of the four attributes (Independence, Competence, Active, and Behavioural characteristics)
of the board are present.
4. Only one of the four attributes (Independence, Competence, Active, and Behavioural characteristics)
of the board is present.
knowledge about their presence or absence, that is, there is uncertainty about
their presence or absence. Figure 2 is a graph of the posterior probability that the
board is effective as a function of the strength of evidence showing the presence
or absence of various attributes. In this figure, we do not consider the presence or
absence of safeguards (i.e., we do not have any information whether safeguards
are present or not, thus, the corresponding likelihood ratios, λ1EB = 1 and
λ2EB = 1). As we can see from the top, solid line (Line 1) in Figure 2, the posterior
probability of the board being effective is the highest when all four attributes are
17
ABACUS
FIGURE 3
COMPARISON OF EFFECTIVENESS OF THE BOARD AS A FUNCTION OF THE

STRENGTH OF VARIOUS ATTRIBUTES OF THE BOARD WITH NO CONSIDERATION OF
SAFEGUARDS
1.0
0.9
0.8 1
0.7
0.6
0.5
0.4
2
0.3
0.2 3
0.1
0.0
0 2 4 6 8 10 12 14 16 18 20
Strength of Attributes in terms of Likelihood Ratio (λ)
The strength of an attribute is represented by the likelihood ratio, λ, corresponding to the attribute
where λ = 1 means no knowledge of its presence, and λ = 19 means 0.95 probability that the attribute is
present (Probability of attribute being present based on the evidence = λ/(1 + λ)).
1. All the four attributes (Independence, Competence, Active, and Behavioural characteristics) of the
board are present.
2. Any three of the four attributes (Independence, Competence, Active, and Behavioural characteristics)
of the board are present with the fourth one not present with probability 0.66.
3. Any two of the four attributes (Independence, Competence, Active, and Behavioural characteristics)
of the board are present with the other two not present with probability 0.66.
present, and increases the strength of attributes (i.e., as the support for the
presence of all the attributes increases). Since the strength of evidence is
represented by the likelihood ratio, a value of λ = 9 means the corresponding
attribute is present with probability 0.90 (Posterior Probability = λ/(1 + λ) with the
prior odds = 1, that is, with no prior knowledge of the presence or absence of the
variable. The second line from the top (Line 2) in Figure 2 represents the
posterior probability of the board being effective in the presence of only three of
the four attributes. This posterior probability is much smaller than the posterior
18
FIGURE 4
EFFECTIVENESS OF THE BOARD AS A FUNCTION OF THE STRENGTH OF

SAFEGUARDS WITH FIXED (GIVEN) LEVEL OF STRENGTH FOR VARIOUS ATTRIBUTES
1.0
0.9
1
0.8
0.7
0.6
2
0.5
0.4
3
0.3
4
0.2
0.1
0.0
0 2 4 6 8 10 12 14 16 18 20
Strength of Safeguard in terms of Likelihood ratio, λ
The strength of Safeguards is represented by the likelihood ratio, λ, based on the evidence pertaining
to the presence of Safeguards, where λ = 1 means no knowledge of its presence and λ = 19 means 0.95
probability that effective Safeguards are present.
1. All four attributes are present with probability 0.66 each.
2. Two attributes are present with probability 0.66 and two are absent with probability 0.66.
3. One attribute is present with probability 0.66 and three are absent with probability 0.66.
4. All four attributes are absent with probability 0.66.
probability of the board being effective when all four attributes were present. The
posterior probability further decreases as additional attributes are eliminated from
consideration as demonstrated from Lines 3 and 4 in Figure 2. Thus, board
effectiveness is the highest when all four attributes are present and increases
further as the strength of these attributes increase.
Figure 3 compares the posterior probability of the board being effective when
all four attributes are present with the posterior probability of board effectiveness
when some attributes are known to be absent. From Figure 3, we see that the
posterior probability of board effectiveness increases, in general, as the likelihood
of the presence of the attributes increases and is highest when all four attributes
are present (see Line 1 in Figure 3). In principle, the posterior probability of
board effectiveness will approach unity when all four attributes are present for
19
ABACUS
FIGURE 5
EFFECTIVENESS OF THE BOARD AS A FUNCTION OF THE STRENGTH OF ATTRIBUTES

WITH ONE OR TWO ATTRIBUTES BEING ABSENT FOR DIFFERENT LEVELS OF THE
STRENGTH OF SAFEGUARDS
1.0
0.9
1
0.8
0.7
0.6 2
0.5
0.4
3
0.3
0.2 4
0.1
0.0
0 2 4 6 8 10 12 14 16 18 20
1. Strong Safeguards (λ1EB = 9, Probability of its presence = 0.9) with three attributes being present
with increasing likelihoods, and one attribute is absent with probability 0.66.
2. Strong Safeguards (λ1EB = 9, Probability of its presence = 0.9) with two attributes being present with
increasing likelihoods, and two attributes being absent with probability 0.66.
3. No Safeguards (λ1EB = 1, Probability of its presence = 0.5) with three attributes being present with
increasing likelihoods, and one attribute being absent with probability 0.66.
4. No Safeguards (λ1EB = 1, Probability of its presence = 0.5) with two attributes being present with
increasing likelihoods, and two attributes being absent with probability 0.66.
sure (i.e., when the likelihood ratios, λs, for all the attributes approach infinity).
Notice the significant decrease in the posterior probability when we have the
knowledge that one of the attributes is not present (say with probability 0.66 in
our case,that is, the corresponding likelihood ratio that the attribute is present is
0.5) compared to when we have knowledge that all four attributes are present
(compare Line 2 with Line 1 in Figure 3). The posterior probability of the board
being effective further decreases when we have evidence in support of any two of
the attributes and the other two are known to be absent, say, with probability 0.66
(Compare Line 2 with Line 3 in Figure 3). In our illustration, Line 5 represents
the worst scenario case, where two of the factors are known to be absent with
20
FIGURE 6
EFFECTIVENESS OF THE BOARD AS A FUNCTION OF THE STRENGTH OF ATTRIBUTES

WITH DIFFERENT LEVELS OF THE STRENGTH OF SAFEGUARDS
1.0
0.9
1
0.8
0.7
0.6
2
0.5
0.4
0.3 3
0.2
0.1 4
0.0
2 5 10 15 20

1. Moderately strong Safeguards (λ 1EB = 4, Probability of its presence = 0.8) with all the four
Attributes being present with increasing likelihoods.
2. Strong Safeguards (λ1EB = 9, Probability of its presence =0.9) with only three Attributes being
present with increasing likelihoods, and one known to be absent with probability 0.66.
3. No Safeguards (λ 1EB = 1, Probability of its presence = 0.5) with all the four Attributes being present
with increasing likelihoods.
4. No Safeguards (λ 1EB = 1, Probability of its presence = 0.5) with only three Attributes being present
with increasing likelihoods and one known to be absent with probability 0.66.
probability 0.66 (the corresponding likelihood ratio being 0.5). This is a logical
result because when you know that two of the attributes are absent, board
effectiveness will be lower than for the situation where only one attribute is
known to be absent, which will be lower than for the situation when all the
attributes are present (compare Lines 1, 2, and 3 in Figure 3).
Case 3: Impacts of Safeguards on the Board Effectiveness
Here we analyze the impact of safeguards on the posterior probability of board
effectiveness. As discussed earlier, safeguard factors impact board effectiveness as a
whole, and their role is more like a control mechanism. Figure 4 illustrates the
variation of the posterior probability of board effectiveness as a function of the
21
ABACUS
strength of safeguards. As we see in Figure 4, in general, the posterior probability of

board effectiveness increases as the strength of safeguards increases. This result
makes logical sense; the stronger the safeguards the higher the board effectiveness.
Next, we examine the impact of the presence or absence of the attributes on
board effectiveness as a function of the strength of safeguards. From Figure 4, we
observe that the posterior probability of board effectiveness is the highest (Line 1)
when we have the knowledge that all four attributes are present with certain
probability (in Figure 4 it is 0.66 with the corresponding likelihood ratio equal to
2.0). The second line from the top (Line 2) represents the posterior probability of
board effectiveness when two of the four attributes are present with probability
0.66, and the other two are absent with probability 0.66. Line 3 represents the
posterior probability when one of the attributes is present with probability 0.66,
and the other three are absent with probability 0.66. Finally, Line 4 in Figure 4
represents the posterior probability of board effectiveness when all four attributes
are known to be absent with probability 0.66. It is clear that as the strength of
safeguards increases the posterior probability of board effectiveness increases and
also it increases the presence of more attributes (Line 1 is higher than Line
2, which is higher than Line 3, which is higher than Line 4). These results suggest
that the board is most effective when all the attributes are present, and the board
members are accountable through the safeguards for their acts. It is important to
note that even if few attributes are absent as in the case of Lines 2, 3, and 4, as the
strength of safeguard effectiveness increases the posterior probability of board
effectiveness increases and in the limit it approaches unity for infinitely strong
safeguards irrespective of whether some of the attributes are absent. This result
demonstrates the value of safeguards. Safeguards work as a control and can be
introduced through some kind of regulations to make board members accountable
for their acts. The stronger the safeguards and higher the probability of the
presence of all four attributes, the higher the board effectiveness.
Figure 5 depicts the variation of board effectiveness as a function of the strength
of the attributes being present when there is a strong safeguard and when there is
no safeguard. Lines 1 and 2 represents board effectiveness under strong safeguard
(the corresponding likelihood ratio, λ = 9, i.e., an effective safeguard is present
with probability 0.9) as a function of the strength of the attributes being present
with one attribute and two attributes being absent, respectively, with probability
0.66. Lines 3 and 4 are similar to Lines 1 and 2, except when there is no safeguard.
As one can see by comparing the lines in Figure 5 (Compare Line 1 with Line
3, and Line 2 with Line 4), for a set of attributes being present and the others
being absent, board effectiveness is significantly higher under the presence of an
effective safeguard than when there is no safeguard. Again, this result shows the
importance of safeguards.
Case 4: Board Effectiveness—Additional Analysis

Figure 6 demonstrates the value of the presence of all four attributes along with
the presence of safeguards. Lines 1 and 3, respectively represent the board
22
effectiveness when all four attributes are present with moderately strong
safeguards (the corresponding likelihood ratio λ1EB = 4) and with no consideration
of safeguards. Lines 2 and 4 respectively represent the board effectiveness when
only three attributes are present and one attribute is absent with probability 0.66
with strong safeguards (the corresponding likelihood ratio λ1EB = 9) and with no
consideration of safeguards. In Figure 6 we see that Line 1 is higher than Line
2. This suggests that board effectiveness is higher for the case where all the four
attributes are present with moderately strong safeguards than for the case when
only three attributes are present with one being absent (with probability 0.66) but
with strong safeguards.
It is important to note that all four attributes must be present for a high level of
board effectiveness (compare Lines 1 and 3 with Lines 2 and 4). Moreover, board
effectiveness would further increase if safeguards were present (compare Lines
1 and 2 with Lines 3 and 4). Board effectiveness would not be as high even with
strong safeguards if one or more attributes are absent. However, in the limit, if we
consider infinitely strong safeguards then board effectiveness for the case where
one or more attributes is absent can be higher than board effectiveness where we
have all four attributes present but moderately strong safeguards. It is not easy to
think of examples of infinitely strong safeguards in the real world. One could
consider ‘execution’ as being such a safeguard to control board members’
behaviours. However, this ideal situation will never be possible in reality.10 Thus,
for realistic situations with moderately strong safeguards board effectiveness is
higher for the case where all four attributes are present than for the case where
only three or fewer attributes are present, and the others are absent even with
strong safeguards. This shows the importance of all four attributes and the
safeguards.
MEASUREMENT ISSUES ASSOCIATED WITH BOARD OF DIRECTOR

ATTRIBUTES
In this section, we provide a list of potential measures associated with the board of
director attributes modelled in the study (see Table 1). While the list is not
necessarily an exhaustive inventory of possible measures, it is intended to provide
guidance for future researchers investigating board of director effectiveness.
Board of Director Independence

The first critical element in our theoretical framework indicates that the board of
directors ought to be independent from management. Researchers can measure
the proportion of the board of directors that is not a member of the management
10
China executed its former top food-and-drug regulator, Zheng Xiaoyu, who was taking bribes from
pharmaceutical companies (see http://www.denverpost.com/headlines/ci_6389912).
23
ABACUS
TABLE 1
MEASURES FOR ATTRIBUTES OF AN EFFECTIVE BOARD OF DIRECTORS
Board Attributes No. Measures
Independence 1 Percentage of directors not a member of management team

2 Separation of the CEO position and chairman of the board of
directors
3 Percentage of director that have not had any business, financial or
other connections with the company
4 Percentage of directors without ‘social ties’ with members of
management team
5 Percentage of directors not affiliated with members of management
team (via family relationship, business dealings, etc.)
6 Complete independence within critical board committees such as
audit committee
7 Level of involvement of CEO and other key executives in
nominating/selecting individuals to serve on the board
Competence 1 Directors’ knowledge about the company and its business objectives
2 Directors’ knowledge about the industry and competitive landscape
3 Tenure as director of the company
4 Tenure being member of any board of directors
5 Number of directorships held in other corporations (i.e., multiple
directorships)
6 Education level and background of directors
7 Directors possessing specific expertise and knowledge
8 Director’s participation in training programs and professional
development conference
Active 1 Number of meetings held by the board of directors

2 Number of meetings held by the committees within the board of
directors
3 Attendance of directors in overall board meetings
4 Attendance of directors in committee meetings
5 Number of hours directors devote to board related work (outside of
meetings)
6 Participation and attentiveness of directors while in meetings
Behavioural Attributes 1 Development of strong relationship and familiarity among directors

2 Social interactions among directors
3 Development of expectations and performance standards for directors
4 Positive and conducive working environment within the boardroom
team. Often, the CEO is listed as a member of the board of directors, but lower
level executives (such as chief financial officer, chief operating officer, etc.) could
also be listed as directors. In addition, researchers can specifically focus on the
chairman of the board of directors position to investigate whether the chairman
position is held by the current CEO.
Next, researchers can investigate the proportion of the board of directors that
does not have (and has not had) any business, financial, or other dealings with the
company. For example, a director that is a former employee or business supplier
of the company is deemed to have had connections with the company. Next,
24
researchers can examine the social ties between the board of directors and the
CEO (and other executives). As Krishnan et al. (2011) illustrate, the efficacy of a
formal independent director could be undermined by social ties with the CEO.
Krishnan et al. (2011, p. 537) define social ties as ‘non-familial, informal ties
between individuals arising from current or prior employment, education, and
other activities (such as golf clubs or charity organizations)’. While a director may
not be employed as a current executive, he or she may be affiliated with
management through family or other relationships. For example, a director that is
a cousin of the CEO is not independent from management.
While the board of directors in not required to be comprised entirely of
independent directors, there are certain committees within the board that are
expected to be completely filled by directors that are independent. For example,
the audit committee plays a crucial role in contracting with external auditors and
monitoring of the financial reporting process. The audit committee is expected to
be completely filled with independent directors. Finally, researchers can assess the
level of involvement that the CEO and other executives have in nominating and
selecting potential members of the board of directors. If a formally independent
director was handpicked by the CEO, then his or her independence may be
compromised.
Board of Director Competence

The second element in our theoretical framework indicates that the board of
directors ought to be competent. To measure board of director competence,
research can investigate the level of directors’ knowledge about the company as
well as about the industry and the competitive landscape. Archival studies have
empirically measured director knowledge by examining the tenure of the director.
Directors that have served more years with the company are expected to have
greater competence because they have more experience dealing with corporate
strategies, policies, and resources. Researchers should also focus on the number of
years a director has served as a board member of any company (not necessarily
the current company). Having longer tenure as a director signals more experience
and knowledge gained as a director of a business organization.
Beyond measuring tenure of the director, we recommend future researchers
devise other creative means of capturing director knowledge. For example, field
and survey-based research could formulate survey questions whereby the
chairman of the board would be asked to rate the knowledge level of the directors
pertaining to specific areas of the business, such as operational, financial,
compliance, corporate strategy, supply chain, and so on.
To measure director competence, researchers could also compute the number of
other directorships that a director holds outside of the company. Prior research
suggests that directors who hold multiple directorships have developed greater
reputational capital and have demonstrated their proficiency as a director. Next,
director competency could be measured by the individual’s education level and
background. In terms of education, researchers could assess the director’s academic
25
ABACUS
degree(s) (i.e., bachelor, masters, doctoral), academic discipline (i.e., business,

engineering, natural science, etc.), quality of the academic institution (i.e., ivy league,
university rankings, etc.), and professional certifications (i.e., certified public
accountant, chartered financial analyst, etc.). To gauge the business background of
the director, researchers could examine the extensiveness of past work experiences
and job functions. For example, when did the director first hold a top executive
position with a company? Has the director held a high-level position in a publicly
traded company? Is the director a current or past CEO?
Prior research has also focused on particular competencies of the director that are
relevant for his or her work on the committee. For example, the audit committee
ought to have directors that have greater financial accounting and legal expertise; the
science and technology committee ought to have directors with more technical
technology or engineering knowledge. Finally, researchers could investigate a
director’s participation level in professional conferences. There are programs that
offer training and professional development for corporate directors. By participating
in such programs, a director learns about best practice. He or she can also access
resources and training materials that can help him or her become a more effective
director.
Board of Director Activity

The next element in our theoretical framework indicates that the board of directors
ought to be active. Based on prior research, the predominant measure for board of
directors activeness is the number of meetings held by the board of directors (and
its various committees). The more meetings the board of directors holds, the more
active the board is in discussing company matters and performing its duties. For
public companies, the number of meetings held by the overall board and its
committees are typically disclosed in public financial disclosures. In addition to the
number of meetings held, it is important to assess the attendance rate of the
directors. Even if the board of directors holds many meetings to discuss issues, if
only half of the members regularly attend then the activeness of the board is sub-
optimal. When evaluating attendance records, researchers ought to measure the
attendance rate of directors for both overall board and specific committee meetings.
While the number of meetings and attendance rate are important to assess,
researchers also ought to examine the number of hours that directors devote to
board-related work outside of meetings. In this regard, researchers can assess the
amount of time a director spends on studying corporate materials, planning for
board meetings, communicating with management and other directors about
corporate matters, reviewing corporate documents and disclosures, and so on. This
information is unlikely to be available through public disclosures or archival sources
but can be explored through field studies and survey-based research methods.
Another dimension of director activeness is his or her participation and
attentiveness during board activities and meetings. A director may physically
attend all board and committee meetings but not actively participate in the
deliberations, in which case that director would not be considered to be highly
26
active. To gauge a director’s participation level, researchers can assess, for

example, the amount of time he or she speaks during the meeting, the number of
questions he or she asks, and the number of issues he or she puts forward to other
directors. Again, such information is unavailable through public diclosures or
archival sources, but can be investigated through field and survey-based research
techniques.
Board of Director Behavioural Attributes

The fourth element in our theoretical framework indicates that the board of
directors ought to have the appropriate behavioural disposition. The first
characteristic we list is that the board ought to develop a strong relationship and
familiarity among the directors. While there are many potential ways to assess this
characteristic, we offer a couple of suggestions. First, a researcher could assess the
number of common directorships held among the directors. For example, director
A and director B serve on the board of directors of three different companies. Since
these two directors serve together on multiple boards, they are likely to have
developed a strong familiarity and close relationship with each other. Second,
researchers could examine the ‘combined’ tenure of the directors. The longer that
the board members have served together, the better they know each other.
Researchers may also consider more direct measures and survey directors about
how they perceive their relationship with the other directors. For example, directors
can rate how much they enjoy working with other members of the board, how close
they perceive the relationship with other directors, and to what extent they would
consider other board members as ‘friends’.
Researchers could also examine the level of social interactions among directors
beyond the boardroom. Directors that have more frequent social interactions are
likely to develop stronger bonds and closer communication ties. Researchers
could consider, for example, measuring the dispersion of age among the
directors. Directors around the same age are more likely to have common social
ties and life experiences. Another measure to consider is the average distance
between each director’s primary place of employment and the primary place of
employment of the other directors. The closer the proximity of directors the
greater likelihood for social interactions. Finally, researchers can investigate the
level of social ties between the directors. As an example, multiple directors on
the same board may belong to the same golf country club or non-profit
organization. Stronger social ties among directors may facilitate stronger
collegiality. It is important to note that these are not ‘perfect’ measures and may
only capture certain aspects of positive social interactions. For example, directors
that belong to the same social club may not necessarily always agree on
corporate issues. However, belonging to the same social club may afford these
directors with more opportunities to interact with one another and develop
closer relationships.
To ensure appropriate behaviour by directors it is often necessary to develop
expectations for professional conduct. Researchers can investigate the extent and
rigour of the company’s code of conduct for its directors. The code of conduct can
27
ABACUS
be informative as it outlines the expectations and acceptable practices of board

members. In addition, researchers can investigate the method and extent to which
the company evaluates the performance of its directors. For example, is the
evaluation process formal and comprehensive, formal but limited in scope, or not
formal? Having robust performance evaluation procedures in place can help
ensure appropriate director conduct.
Finally, researchers can assess the working environment within the board. They
can investigate certain behavioural cues, including: (1) willingness of directors to
accept constructive suggestions from others; (2) ability of directors to work
together as a team rather than working individually; (3) aptitude of directors to
resolve disagreements productively. Again, such data are unavailable through
archival research methods. However, field-based researchers may get appropriate
access to the inner-workings of the boardroom and can document observations
pertaining to these behavioural cues. Researchers can also develop survey
instruments that investigate the chairman of the board in order to rate the board’s
working dynamics. Alternatively, researchers can directly survey directors to rate
the quality of interactions between board members.
CONCLUSION
Corporate governance provides the structure through which the objectives of the
company are set, and the means of attaining those objectives and monitoring
performance are determined (OECD, 2004). An effective corporate governance
framework necessitates strategic guidance of business activities and monitoring of
management actions, among others (OECD 2004; Hilb 2002). In which case, the
board constitutes an essential component of the corporate governance system. The
board is a trust-engendering mechanism intended to ensure that managers are
made accountable for their actions and that decisions are made to the benefit of
shareholders’ interests (Baysinger and Hoskisson, 1990; Hilb, 2012).
While a vast number of studies examine the association between characteristics of
the board and corporate outcomes, there is scant formal theoretical work underlying
the inter-relationships between attributes of an effective board (Larcker et al., 2007;
Brown et al., 2011). Given the impetus to develop a theoretical framework on board
effectiveness, we present an analytical model demonstrating that for a board to be
effective, the following four criteria must be met: (1) independence from
management; (2) competent in performing its functions and responsibilities;
(3) active in board duties; and (4) have appropriate behavioural attributes.
We identify these four attributes from professional guidance and prior academic
research that examine the efficacy of corporate boards. Our theoretical model
defines an ‘and’ relationship between a board being effective and the four
attributes of independence, competency, activity, and behavioural attributes.
Using inferential statistics, we demonstrate that one cannot replace an attribute by
28
strengthening another. Thus, we conclude that a board of directors is most

effective if and only if the board exhibits high levels of independence,
competency, activity, and behavioural traits. In addition, the results of our
Bayesian model demonstrate that even when all four attributes are present the
presence of ‘safeguards’ further increases the effectiveness of the board. Thus, as
an implication of this study, we propose that some manner of ‘safeguard’
mechanism should be in place to control the behaviour of board members to make
the board more effective. Further, we encourage future research on the board of
directors to include the safeguard construct.
Of particular interest to regulators and other governance bodies may be the
identification of the importance of the ‘and’ relationship in assessing board
effectiveness. Prior attempts to address board effectiveness, such as the SEC’s 2003
adoption of board compensation requirements as proposed by the NYSE and Nasdaq
for listed firms, take an ‘or’ approach, focusing on an individual attribute (in this case
board independence), while failing to address the interrelationship with other
attributes.
The analytical model we present in this paper is a strong first step in understanding
the core requirements necessary for a board to be effective; however, we recognize
that it is only a beginning. As with any analytic model, there are questions to which
answers can be formed but which cannot be directly answered. To help address these
shortcomings we encourage researchers to take this framework and use it as a
starting point for future research. In particular we encourage archival based
researchers to develop and validate proxies for each of the model’s attributes and to
use those proxies further to explain the role the board of directors plays in firm
guidance and governance. The attributes of board competency and the board’s
behavioural attributes are of particular interest as they have, to a certain extent, been
ignored by prior archival research. Upon developing suitable proxies, researchers can
consider examining changes to the role and structure of the board following an
economic shock, such as the recent financial crisis, or changes to board culture
following a negative accounting outcome, such as a restatement of financial reports.
Additionally, there is significant room within the behavioural research space to
develop and test research questions regarding the degree of confidence one needs
when assessing the board’s attributes. In particular, it would be very insightful to
know if each attribute is viewed as carrying the same level of importance and, if not,
to identify the pecking order of importance among attributes. Finally, given the broad
influence of the board of directors, we encourage our fellow scholars in disciplines
such as management, finance, and psychology to consider our theoretical framework,
and to apply its characteristics and attributes to situations from their field of study.
REFERENCES
Abbott, L. J., S. Parker, G. F. Peters, and K. Raghunandan (2003), ‘The Association Between Audit
Committee Characteristics and Audit Fees’, Auditing: A Journal of Practice & Theory, Vol. 22,
No. 2, pp. 17–32.
29
ABACUS
American Institute of Certified Public Accountants (AICPA) (2006), Statement on Auditing Standards
No. 106: Audit Evidence, AICPA, Durham.
Anand, S. (2009), ‘Satyam Fraud Shakes Investor Confidence’, Finance Asia, pp. 18–19.
Anderson, R. C., S. A. Mansi, and D. M. Reeb (2004), ‘Board Characteristics, Accounting Report
Integrity, and the Cost of Debt’, Journal of Accounting and Economics, Vol. 37, No. 3,
pp. 315–42.
Baldenius, T., N. Melumad, and X. Meng (2014), ‘Board Composition and CEO Power’, Journal of
Financial Economics, Vol. 112, No. 1, pp. 53–68.
Baysinger, B. and R. E Hoskisson (1990), ‘The Composition of Boards of Directors and Strategic
Control: Effects on Corporate Strategy’, Academy of Management Review, Vol. 15, No. 1,
pp. 72–87.
Baysinger, B. D. and H. N. Butler (1985), ‘Corporate Governance and the Board of Directors:
Performance Effects of Changes in Board Composition’, Journal of Economics and
Organizations, Vol. 1, pp. 101.
Beasley, M. S. (1996), ‘An Empirical Analysis of the Relation Between the Board of Director
Composition and Financial Statement Fraud’, Accounting Review, Vol. 71, pp. 443–65.
Bebchuk, L., A. Cohen, and A. Ferrell (2009), ‘What Matters in Corporate Governance?’, Review of
Financial Studies, Vol. 22, No. 2, pp. 783–827.
Bertoni, F., M. Meoli, and S. Vismara (2014), ‘Board Independence, Ownership Structure and the
Valuation of IPOs in Continental Europe’, Corporate Governance: An International Review,
Vol. 22, No. 2, pp. 116–31.
Black, B., B. Cheffins, and M. Klausner (2006), ‘Outside Director Liability’, Stanford Law Review,
Vol. 58, pp. 1055–160.
Boone, A. L., L. Casares Field, J. M. Karpoff, and C. G. Raheja (2007), ‘The Determinants of
Corporate Board Size and Composition: An Empirical Analysis’, Journal of Financial
Economics, Vol. 85, No. 1, pp. 66–101.
Brickey, K. F. (2003), ‘From Enron to Worldcom and Beyond: Life and Crime After Sarbanes-Oxley’,
Washington University Law Quarterly, Vol. 81, pp. 357.
Brochet, F. and S. Srinivasan (2013), ‘Accountability of Independent Directors: Evidence from Firms
Subject to Securities Litigation’, Journal of Financial Economics, Vol. 111, No. 2, pp. 430–49.
Brown, P., W. Beekes, and P. Verhoeven (2011), ‘Corporate Governance, Accounting and Finance:
A Review’, Accounting and Finance, Vol. 51, pp. 96–172.
Carcello, J. V., D. R. Hermanson, T. L. Neal, and R. A. Riley (2002), ‘Board Characteristics and Audit
Fees’, Contemporary Accounting Research, Vol. 19, No. 3, pp. 365–84.
Chen, H. L. (2011), ‘Does Board Independence Influence the Top Management Team? Evidence from
Strategic Decisions Toward Internationalization’, Corporate Governance: An International
Review, Vol. 19, No. 4, pp. 334–50.
Chhaochharia, V. and Y. Grinstein (2007), ‘Corporate Governance and Firm Value: The Impact of the
2002 Governance Rules’, Journal of Finance, Vol. 62, pp. 789–825.
Cicero, D., M. B. Wintoki, and T. Yang (2013), ‘How Do Public Companies Adjust Their Board
Structures?’, Journal of Corporate Finance, Vol. 23, pp. 108–27.
Claessens, S., M. L. Laeven, D. Igan, and M. G. Dell’Ariccia (2010), Lessons and Policy Implications
from the Global Financial Crisis, International Monetary Fund (Epub), No. 10–44.
Cohen, L., A. Frazzini, and C. J. Malloy (2012), ‘Hiring Cheerleaders: Board Appointments of
“Independent’ Directors”, Management Science, Vol. 58, No. 6, pp. 1039–58.
Coles, J. L., N. D. Daniel, and L. Naveen (2008), ‘Boards: Does One Size Fit All?’, Journal of Financial
Economics, Vol. 87, No. 2, pp. 329–56.
Core, J. E., R. W. Holthausen, and D. F. Larcker (1999), ‘Corporate Governance, Chief Executive
Officer Compensation, and Firm Performance’, Journal of Financial Economics, Vol. 51, No. 3,
pp. 371–406.
Cotter, J. and M. Silvester (2003), ‘Board and Monitoring Committee Independence’, Abacus, Vol. 39,
No. 2, pp. 211–32.
30
Cotter, J. R., A. Shivdasani, and M. Zenner (1997), ‘Do Independent Directors Enhance Target
Shareholder Wealth During Tender Offers?’, Journal of Financial Economics, Vol. 43,
pp. 195–218.
Cyert, R. M., S. H. Kang, and P. Kumar (2002), ‘Corporate Governance, Takeovers, and
Top-Management Compensation: Theory and Evidence’, Management Science, Vol. 48, No. 4,
pp. 453–69.
Davidoff, S. (2011), ‘Ex-Directors of Failed Firms Have Little to Fear’, New York Times, retrieved
from http://Dealbook.Nytimes.Com/2011/08/02/Ex-Directors-Of-Failed-Firms-Have-Little-To-Fear/.
Defond, M. L., R. N. Hann, and X. Hu (2005), ‘Does the Market Value Financial Expertise on Audit
Committees of Boards of Directors?’, Journal of Accounting Research, Vol. 43, No. 2, pp. 153–93.
Desai, V., R. Roberts, and R P. Srivastava (2010), ‘A Conceptual Model for External Auditor
Evaluation of the Internal Audit Function Using Belief Functions’, Contemporary Accounting
Research, Vol. 27, No. 2, pp. 537–75.
Deutsch, Y. (2005), ‘The Impact of Board Composition on Firms’ Critical Decisions: A Meta-Analytic
Review’, Journal of Management, Vol. 31, No. 3, pp. 424–44.
Deutsch, Y. and T. W. Ross (2003), ‘You Are Known by the Directors You Keep: Reputable Directors
as a Signaling Mechanism for Young Firms’, Management Science, Vol. 49, No. 8, pp. 1003–17.
Dontoh, A., J. Ronen, and B. Sarath (2013), ‘Financial Statements Insurance’, Abacus, Vol. 49, No. 3,
pp. 269–307.
Eisenhardt, K. M. (1989), ‘Agency Theory: An Assessment and Review’, Academy of Management
Review, Vol. 14, No. 1, pp. 57–74.
Engel, E., R. M. Hayes, and X. Wang (2005), ‘The Sarbanes-Oxley Act and Firms’ Going Private
Decisions’, Journal of Accounting and Economics, Vol. 44, pp. 116–45.
Fama, E. F. (1980), ‘Agency Problems and the Theory of the Firm’, The Journal of Political Economy,
pp. 288–307.
Fama, E. F. and M. C. Jensen (1983), ‘Separation of Ownership and Control’, Journal of Law &
Economics, Vol. 26, pp. 301–25.
Ferris, S. P., M. Jagannathan, and A. C. Pritchard (2003), ‘Too Busy to Mind the Business? Monitoring by
Directors with Multiple Board Appointments’, The Journal of Finance, Vol. 58, No. 3, pp. 1087–112.
Fich, E. M. and A. Shivdasani (2007), ‘Financial Fraud, Director Reputation, and Shareholder Wealth’,
Journal of Financial Economics, Vol. 86, No. 2, pp. 306–36.
Finkelstein, S. and A. Mooney (2003), ‘Not the Usual Suspects: How to Use Board Process to Make
Boards Better’, Academy of Management Executive, Vol. 17, pp. 101–13.
Fisman, R. J., R. Khurana, M. Rhodes-Kropf, and S. Yim (2013), ‘Governance and CEO Turnover: Do
Something or Do the Right Thing?’, Management Science, Vol. 60, No. 2, pp. 319–37.
Forbes, D. P. and F. J. Milliken (1999), ‘Cognition and Corporate Governance: Understanding Boards
of Directors as Strategic Decision-making Groups’, Academy of Management Review, Vol. 24,
No. 3, pp. 489–505.
Gilson, S. C. (1990), ‘Bankruptcy, Boards, Banks, and Blockholders: Evidence on Changes in
Corporate Ownership and Control When Firms Default’, Journal of Financial Economics, Vol.
27, No. 2, pp. 355–87.
Gompers, P., J. Ishii, and A. Metrick (2003), ‘Corporate Governance and Equity Prices’, The Quarterly
Journal of Economics, Vol. 118, No. 1, pp. 107–56.
Gordon, J. N. (2007), ‘The Rise of Independent Directors in the United States, Vol. 1950–2005 of
Shareholder Value and Stock Market Prices’, Stanford Law Review, Vol. 59, pp. 1465–568.
Helleiner, E. (2011), ‘Understanding the 2007–2008 Global Financial Crisis: Lessons for Scholars of
International Political Economy’, Annual Review of Political Science, Vol. 14, pp. 67–87.
Hendry, K. and G. C. Kiel (2004), ‘The Role of the Board in Firm Strategy: Integrating Agency and
Organisational Control Perspectives’, Corporate Governance: An International Review, Vol. 12,
No. 4, pp. 500–20.
Hilb, M. (2012), New Corporate Governance: Successful Board Management Tools, Springer,
New York.
31
ABACUS
Hillman, A. J. and T. Dalziel (2003), ‘Boards of Directors and Firm Performance: Integrating Agency
and Resource Dependence Perspectives’, Academy of Management Review, Vol. 28, No. 3,
pp. 383–96.
Huse, M. (2005), ‘Accountability and Creating Accountability: A Framework for Exploring
Behavioural Perspectives of Corporate Governance’, British Journal of Management, Vol. 16,
S65–S79.
Ingley, C. B. and N. T. Van Der Walt (2001), ‘The Strategic Board: The Changing Role of Directors in
Developing and Maintaining Corporate Capability’, Corporate Governance: An International
Review, Vol. 9, No. 3, pp. 174–85.
Jensen, M. C. and W. H. Meckling (1976), ‘Theory of the Firm: Managerial Behavior, Agency Costs
and Ownership Structure’, Journal of Financial Economics, Vol. 3, No. 4, pp. 305–60.
Kesner, I. F. and R. B. Johnson (1990), ‘An Investigation of the Relationship Between Board
Composition and Stockholder Suits’, Strategic Management Journal, Vol. 11, No. 4, pp. 327–36.
Kim, K., E. Mauldin, and S. Patro (2014), "Outside Directors and Board Advising and Monitoring
Performance”, Journal of Accounting and Economics, Vol. 57, No. 2, pp. 110-31.
Klein, A. (2002), ‘Audit Committee, Board of Director Characteristics, and Earnings Management’,
Journal of Accounting and Economics, Vol. 33, No. 3, pp. 375–400.
Knyazeva, A., D. Knyazeva, and R. W. Masulis (2013), ‘The Supply of Corporate Directors and Board
Independence’, Review of Financial Studies, Vol. 26, No. 6, pp. 1561–605.
Krishnan, G. V. and G. Visvanathan (2008), ‘Does the SOX Definition of an Accounting Expert
Matter? The Association between Audit Committee Directors’ Accounting Expertise and
Accounting Conservatism’, Contemporary Accounting Research, Vol. 25, No. 3, pp. 827–58.
Krishnan, J. (2005), ‘Audit Committee Quality and Internal Control: An Empirical Analysis’,
The Accounting Review, Vol. 80, No. 2, pp. 649–75.
Krishnan, J., Y. Wen, and W. Zhao (2011), ‘Legal Expertise on Corporate Audit Committees and
Financial Reporting Quality’, The Accounting Review, Vol. 86, No. 6, pp. 2099–130.
Kroll, M., B. A. Walters, and P. Wright (2008), ‘Board Vigilance, Director Experience, and Corporate
Outcomes’, Strategic Management Journal, Vol. 29, No. 4, pp. 363–82.
Kuhner, C. and C. Pelger (2015), ‘On the Relationship of Stewardship and Valuation—An Analytical
Viewpoint’, Abacus, Vol. 51, No. 3, pp. 379–411.
Kumar, P. and A. Zattoni (2014), ‘Corporate Governance, Board of Directors, and the Firm: A
Maturing Field’, Corporate Governance: An International Review, Vol. 22, No. 5, pp. 365–66.
Larcker, D. F., S. A. Richardson, and I. Tuna (2007), ‘Corporate Governance, Accounting Outcomes, and
Organizational Performance’, The Accounting Review, Vol. 82, No. 4, pp. 963–1008.
Leblanc, R. and M. S. Schwartz (2007), ‘The Black Box of Board Process: Gaining Access to a Difficult
Subject’, Corporate Governance: An International Review, Vol. 15, No. 5, pp. 843–51.
Linck, J. S., J. M. Netter, and T. Yang (2008), ‘The Determinants of Board Structure’, Journal of
Financial Economics, Vol. 87, pp. 308–28.
Lipton, M. and J. Lorsch (1992), ‘A Modest Proposal for Improved Corporate Governance’, Business
Lawyer, Vol. 48, No. 1, pp. 59–77.
Macavoy, P. W. and I. M. Millstein (1999), ‘The Active Board of Directors and its Effect on the
Performance of the Large Publicly Traded Corporation’, Journal of Applied Corporate Finance,
Vol. 11, No. 4, pp. 8–20.
Marden, R. and R. Edwards (2005), ‘The Sarbanes-Oxley “Ax”, The CPA Journal, Vol. 75, No. 4,
pp. 6–9.
Mautz, R. and H. Sharaf (1961), The Philosophy of Auditing, American Accounting Association,
Sarasota, Florida.
National Association of Corporate Directors (NACD) (1996), Report of the NACD Blue Ribbon
Commission on Director Professionalism, NACD, Washington, DC.
Nguyen, B. D. (2012), ‘Does the Rolodex Matter? Corporate Elite’s Small World and the Effectiveness
of Boards of Directors’, Management Science, Vol. 58, No. 2, pp. 236–52.
32
Organization for Economic Co-Operation and Development (OECD) (2004), OECD Principles of
Corporate Governance, OECD Publications Service, Paris.
Osma, B. G. (2008), ‘Board Independence and Real Earnings Management: The Case of R&D
Expenditure’, Corporate Governance: An International Review, Vol. 16, No. 2, pp. 116–31.
Peasnell, K. V., P. F. Pope, and S. Young (2005), ‘Board Monitoring and Earnings Management: Do
Outside Directors Influence Abnormal Accruals?’ Journal of Business Finance & Accounting,
Vol. 32, Nos 7–8, pp. 1311–46.
Public Oversight Board (POB) (1993), Issues Confronting the Accounting Profession, POB, Stamford CT.
Roberts, J. (2012), ‘Between the Letter and the Spirit: Defensive and Extensive Modes of Compliance
with the UK Code of Corporate Governance’, in The Sage Handbook of Corporate Governance,
Sage, Thousand Oaks, pp. 196–216
Roberts, J., T. Mcnulty, and P. Stiles (2005), ‘Beyond Agency Conceptions of the Work of the Non-
Executive Director: Creating Accountability in the Boardroom’, British Journal of Management,
Vol. 16, S5–S26.
Ross, S. A. (1973), ‘The Economic Theory of Agency: The Principal’s Problem’, The American
Economic Review, Vol. 63, No. 2, pp. 134–39.
Securities and Exchange Commission (SEC) (2003), Final Rule: Disclosure Required by Sections
406 and 407 of the Sarbanes-Oxley Act of 2002, SEC, Washington, DC.
Shafer, G. (1976), A Mathematical Theory of Evidence, Princeton University Press, Princeton NJ.
Shenoy, P. and G. Shafer (1990), ‘Axioms for Probability and Belief-Function Computation’, in
R. Shachter, T. Levitt, J. Lemmer, and L. Kanal (eds), Uncertainty in Artificial Intelligence,
North Holland, Amsterdam, pp.169–98.
Srinivasan, S. (2005), ‘Consequences of Financial Reporting Failure for Outside Directors: Evidence
from Accounting Restatements and Audit Committee Members’, Journal of Accounting
Research, Vol. 43, No. 2, pp. 291–334.
Srivastava, R. P. (2011), ‘An Introduction to Evidential Reasoning for Decision Making Under
Uncertainty: Bayesian and Belief Functions Perspectives’, International Journal of Accounting
Information Systems, Vol. 12, pp. 126–35.
Srivastava, R. P., T. Mock, and J. Turner (2007), ‘Analytical Formulas for Risk Assessment for a Class
of Problems Where Risk Depends on Three Interrelated Variables’, International Journal of
Approximate Reasoning, Vol. 45, pp. 123–51.
(2009), ‘Bayesian Fraud Risk Formula for Financial Statement Audits’, Abacus, Vol. 45, No. 1,
pp. 66–87.
Srivastava, R. P., T. J. Mock, K. Pincus, and A. Wright (2012), ‘Causal Inference in Auditing: A
Framework’, Auditing: A Journal of Practice and Theory, Vol. 31, No. 3, pp. 177-201.
Stewart, T. R. and W. R. Kinney Jr (2013), ‘Group Audits, Group-Level Controls, and Component
Materiality: How Much Auditing is Enough?’, The Accounting Review, Vol. 88, No. 2, pp. 707–37.
Tomberlin, M. (2003), ‘Five Board Members at Healthsouth will Resign’, The Birmingham News,
retrieved from http://www.Al.Com/Specialreport/Birminghamnews/ Index.Ssf?Healthsouth/
Healthsouth154.Html.
United States Senate (2002), The Role of the Board of Directors in Enron’s Collapse, Report Prepared
by the Permanent Subcommittee on Investigations of the Senate of the United States, US
Senate, Washington DC.
Uzun, H., S. H. Szewczyk, and R. Varma (2004), ‘Board Composition and Corporate Fraud’, Financial
Analysts Journal, Vol. 60, pp. 33–43.
Vafeas, N. (1999), ‘Board Meeting Frequency and Firm Performance’, Journal of Financial Economics,
Vol. 53, pp. 113–42.
Van Den Berghe, L. A. A. and A. Levrau (2004), ‘Evaluating Boards of Directors: What Constitutes a
Good Corporate Board?’, Corporate Governance, Vol. 12, No. 4, pp. 461–78.
Wintoki, M. B. (2007), ‘Corporate Boards and Regulation: The Effect of the Sarbanes–Oxley Act and
the Exchange Listing Requirements on Firm Value’, Journal of Corporate Finance, Vol. 13, Nos
2–3, pp. 229–50.
33
ABACUS
Worldcom Incorporated (2003), Report of Investigation by the Special Investigative Committee of the
Board of Directors of Worldcom, Inc, retrieved from https://www.sec.gov/Archives/edgar/
data/723527/000093176303001862/dex991.htm
Xie, B., W. N. Davidson III, and P. J. Dadalt (2003), ‘Earnings Management and Corporate
Governance: The Role of the Board and the Audit Committee’, Journal of Corporate Finance,
Vol. 9, No. 3, pp. 295–316.
APPENDIX A
Posterior Probability of Main Assertion Given the Knowledge About
‘N’ Sub-Assertions in an ‘And’ Tree
We use the induction approach to develop the formula for posterior probability of
the main assertion, M, which consists of n-sub-assertions, A1... An, that are related
to the main assertion through the ‘and’ relationship, as depicted in Figure 1A. The
rounded boxes represent variables (assertion and sub-assertions), and the
rectangular boxes represent items of evidence pertaining to the variables to which
they are connected. The ‘and’ relationship implies that the main assertion is met,
that is, it is true if and only if all the sub-assertions are true. We assume these
variables to be binary in nature. An upper case letter is used to represent the
name of the variable while the lower case letter is used for its values. For
example, M represents the main assertion with ‘m’ representing the value that it is
true, that is, the assertion is met, and ‘~m’ represents the negation of the variable
M that it is not true. Figure 1A is a tree type evidential diagram, where each item
of evidence pertains to only one variable. If an item of evidence pertains to more
than one variable, then the evidential diagram becomes a network. Deriving an
analytical formula for a network evidential diagram becomes intractable.
FIGURE 1A
GENERIC EVIDENTIAL DIAGRAM FOR N SUB-ASSERTIONS
Sub-assertion A1 E1 : Evidence pertaining to

{a1, ~a1} A1
Sub-assertion A2 E2 : Evidence pertaining to

{a2, ~a2} A2
Main Assertion (M) AND

{m, ~m}
Sub-assertion Ai Ei : Evidence pertaining
{ai, ~ai} to Ai
EM : Evidence pertaining to M
Sub-assertion An En : Evidence pertaining to

{an, ~an} An
34
We want to determine the posterior probability that the main assertion is true
given that we have observed all the evidence related to all the variables: EM, E1,
… En (see Figure 1A). In other words, we want to answer the question: what is
the likelihood that the main assertion is true given that we have collected and
evaluated all the evidence related to all the variables as given in Figure 1A? We
use Shenoy and Shafer’s (1990, see also Srivastava, 2011) technique to derive the
general formula. Srivastava et al. (2012) have used the Shenoy and Shafer
approach to derive a general formula for the likelihood of a particular cause to be
the reason for an effect under the situation where multiple causes are the reason
for the effect. In some sense, our problem is similar to theirs, except we want to
determine the likelihood of the presence of the effect (the main assertion) based
on the presence of all the causes (all the sub-assertions).
First, we develop the analytical formula for the posterior probability that the
main assertion, M, is true for the simple case with two sub-assertions, A1 and A2,
and then by induction we derive the formula for n sub-assertions. In general, we
also assume that we have partial knowledge about the presence of the main
assertion, M, through evidence EM.
Two Sub-assertions Case

Consider a situation where two sub-assertions, A1 and A2, are related to the main
assertion M through an ‘and’ relationship. This relationship implies that the main
assertion M is true if and only if A1 is true and A2 is true. Under the set notations, one
can write this relationship as: m = a1\a2, where a1 represents the state that sub-assertion
A1 is true, a2 represents the state that sub-assertion A2 is true, and m represents the
state of the main assertion M that it is true. The ‘and’ relationship between M, and A1,
and A2 can be expressed in terms of ‘or’ relationship for the negation of the main
assertion M as represented by the following equation in terms of set notations:
m = ða1 \ a2 Þ [ ð a1 \ a2 Þ [ ð a1 \ a2 Þ,
which means that the main assertion M is not true when any one of the sub-
assertions is not true or when both of them are not true.
To derive the formula for the posterior probability that the main assertion is true
given that we have observed and evaluated all the evidence, P(m|EME1E2), we use
the Shenoy and Shafer (1990) approach of combining probability information. Under
this approach, we first identify all the probability information relevant in the problem.
Probability information on a variable (variables being M, A1, and A2) is expressed in
terms of what Bayesian literature refers to as probability potentials (Shenoy and
Shafer, 1990). The probability potentials at a variable essentially are probabilities or
conditional probabilities associated with the variable but are not necessarily
normalized, that is, they do not necessarily add to one. For example, the conditional
probabilities associated with variable M due to the evidence EM could be expressed
as potentials at M with two values, one for ‘m’ that it is present and the other for
‘~m’ that it is not present and are represented as P(EM|m)P(m) and P(EM|~m)P(~m).
In general, we use the symbol φ(.) to express the potential for the argument given in
35
ABACUS
the parenthesis. For example, the potentials on the state space {m, ~m} of the binary
variable ‘M’ based on the conditional probabilities can be expressed as: φ(m) = P(EM|
m)P(m), and φ(~m) = P(EM|~m)P(~m). Basically, under the Shenoy and Shafer
(1990) approach, all the potentials are vacuously extended11 to the joint space of all
the variables, point-wise multiplied,12 and then marginalized13 to variables of interest
to determine the posterior probability of that particular variable. The following
discussion provides the details of combining all the potentials and finally determining
the overall potentials at M, the variable of interest.
Step 1: Identify all the probability potentials for two sub-assertions case in
Figure 1A
φ ð mÞ PðEM jmÞPðmÞ
Probability potentials at ‘M ‘due to evidence EM: =
φð mÞ PðEM j mÞPð mÞ

φða1 Þ
Probability potentials at variable A1 due to evidence E1:
φð a1 Þ

PðE1 ja1 ÞPða1 Þ
=
PðE1 j a1 ÞPð a1 Þ

φða2 Þ
Probability potentials at variable A2 due to evidence E2:
φð a2 Þ

PðE2 ja2 ÞPða2 Þ
=
PðE2 j a2 ÞPð a2 Þ
Probability potentials related to ‘and’ relationship:
2 3 2 3 2 3
φða1 a2 mÞ Pða1 a2 mÞ 1:0
6 φða1 a2 mÞ 7 6 Pða1 a2 mÞ 7 6 1:0 7
6 7 6 7 6 7
4 φð a1 a2 mÞ 5 = 4 Pð a1 a2 mÞ 5 = 4 1:0 5
φð a1 a2 mÞ Pð a1 a2 mÞ 1:0
Step 2: Combination of Potentials

We combine the potentials using the Shenoy and Shafer (1990) approach. Under
their approach one needs to perform point-wise multiplication of all the
potentials by first vacuously extending the potentials at M, A1, and A2 to the
11
Vacuous extension can be easily explained in terms of an example. Suppose we have the following
probability information about, say a binary variable X: P(x) and P(~x). We can vacuously extend
the above probability information to the joint space {xy, x~y, ~xy, ~x~y} as follows: P({xy,
x~y}) = P(x), and P({~xy, ~x~y}) = P(~x). There is no information about Y in the above probability
information. That is the reason it is called vacuous extension.
12
Point-wise multiplication means that one element of a potential is multiplied to the same element
of another potential yielding the resultant potential.
13
Marginalization is the opposite of vacuous extension. Suppose we want to marginalize the probability
information, P({xy, x~y}), and P({~xy, ~x~y}) to variable X. This is achieved by summing the
probability information over the variable Y that is not needed. Thus, we obtain the following: P({xy,
x~y}) marginalized to X yields P(x), and P({~xy, ~x~y}) marginalized to X yields P(~x).
36
joint space {a1a1m, a1~a2~m, ~a1a1~m, ~a1~a1~m}, which happens to be the state
space of the ‘and’ relationship. As mentioned earlier, under point-wise
multiplication, each element of a potential is multiplied with the same element of
another potential. Given below is an example of the vacuous extension of the
potential at M.
2 3 2 3
φða1 a2 mÞ PðEM jmÞPðmÞ
6 φða1 a2 mÞ 7 6 PðEM j mÞPð mÞ 7
6 7 6 7 ðA1Þ
4 φð a1 a2 mÞ 5 = 4 PðEM j mÞPð mÞ 5
φð a1 a2 mÞ PðEM j mÞPð mÞ
Next, we point-wise multiply the four sets of potentials (three sets from the
three variables A1, A2, and M, and one set from the ‘and’ relationship). In other
words, we point-wise multiply the potentials defined earlier at M, A1, A2, and the
potential for the ‘and’ relationship after vacuously extending the potentials onto
the joint space of the three variables. This multiplication yields the following
potentials:
2 3 2 3
Φða1 a2 mÞ PðE1 ja1 ÞPða1 ÞPðE2 j a2 ÞPða2 ÞPðEM j mÞPðmÞ
6 Φða1 a2 mÞ 7 6 PðE1 ja1 ÞPða1 ÞPðE2 j a2 ÞPð a2 ÞPðEM j mÞPð mÞ 7
6 7 6 7
4 Φð a1 a2 mÞ 5 = 4 PðE1 j a1 ÞPð a1 ÞPðE2 ja2 ÞPða2 ÞPðEM j mÞPð mÞ 5
Φð a1 a2 mÞ PðE1 j a1 ÞPð a1 ÞPðE2 j a2 ÞPð a2 ÞPðEM j mÞPð mÞ
ðA2Þ
To determine the combined potentials at variable M, we marginalize the above

potential in (A2) onto the state space of M. The marginalization process yields the
following potentials at M.

Φ ð mÞ X1
= , ðA3Þ
Φ ð m Þ X2 + X3 + X4
where X1, X2, X3, and X4 are defined below:
X1 = PðE1 j a1 ÞPða1 ÞPðE2 ja2 ÞPða2 ÞPðEM jmÞPðmÞ,

X2 = PðE1 j a1 ÞPða1 ÞPðE2 j a2 ÞPð a2 ÞPðEM j mÞPð mÞ,
ðA4Þ
X3 = PðE1 j a1 ÞPð a1 ÞPðE2 j a2 ÞPða2 ÞPðEM j mÞPð mÞ,
X4 = PðE1 j a1 ÞPð a1 ÞPðE2 j a2 ÞPð a2 ÞPðEM j mÞPð mÞ
Equation (A3) provides the overall potentials at variable M after combining all
the items of evidence including the priors. The potentials in (A3) when
normalized yield the posterior probability that M is true given all the evidence
about M, A1, and A2. To simplify the final result, we divide both the numerator
and denominator of the normalized potentials by P(E1|~a1)P(~a1)P(E2|~a2)P(~a2)
37
ABACUS
P(EF|~m)P(~m) and define the following likelihood ratios, λ’s, and prior
odds, O’s:
λ1 = PðE1 ja1 Þ=PðE1 j a1 Þ;λ2 = PðE2 ja2 Þ=PðE2 j a2 Þ;λM = PðEM jmÞ=PðEM j mÞ,
O1 = Pða1 Þ=Pð a1 Þ,O2 = Pða2 Þ=Pð a2 Þ, Om = PðmÞ=Pð mÞ,
ðA5Þ
We obtain the following posterior probability of m:
λ 1 O1 λ 2 O2 λ M OM
PðmjE1 E2 EM Þ = ðA6Þ
1 + λ 1 O 1 + λ 2 O2 + λ 1 O 1 λ 2 O 2 λ M O M
We can rewrite the above posterior probability by defining βi = λiOi as the

product of the likelihood ratio and the prior odds as:
β1 β2 βM
PðmjE1 E2 EM Þ = ðA7Þ
1 + β 1 + β 2 + β1 β 2 β M
Equation (A7) represents the desired posterior probability that the main
assertion is true for the two sub-assertions case in terms of the likelihood ratios,
λ’s, and prior odds O’s, as given by β’s, β = λO, given that we have observed the
following evidence: E1, E2, and EM. Equation (A7) is a general expression. Next,
we extend the above formula for the situation where there are ‘n’ sub-assertions.
Extension to ‘n’ Sub-Assertions

As we can see from (A7), the two-assertions case, there is a definite pattern in
the formula. For example, the denominator consists of four terms, one plus
three additional terms, two for the two individual sub-assertions and one joint
term consisting of the product of the likelihood ratios and prior odds for all the
three variables, A1, A2, and M. For a three sub-assertion system one would
have the following expression for the posterior probability that the main
assertion is true:
β1 β 2 β 3 βM
PðmjE1 E2 E3 EM Þ = ðA8Þ
1 + β 1 + β 2 + β 3 + β 1 β2 + β1 β 3 + + β 2 β 3 + β1 β 2 β 3 βM
Thus, by induction we can generalize the result for the case where we have n
sub-assertions as:
Q
n
βi βM
i=1
PðmjE1 E2 …En EM Þ = ðA9Þ
P
n P
n P
n Q
n
1+ βi + βi βj + βi βj βk + … + βi βM
i=1 i, j, i6¼j i, j, k, i6¼j6¼k i=1
38
where i, j, and k, vary from 1 to n. The symbol β is defined to be the product of

the likelihood ratio, λ, and the corresponding prior odds, O, that is, βi = λiOi.
In general, the likelihood ratio, λ = P(E|a)/P(E|~a), represents the strength of
evidence E pertaining to A, where ‘a’ represents the state that Assertion A is true,
and ~a represents the state that A is not true. For a positive piece of evidence the
likelihood ratio takes the following values: ∞>λ > 1, for a neutral piece of
evidence λ = 1, and for a negative piece of evidence the value of the likelihood
ratio lies between 1 and zero, that is, 1 > λ > 0. The extreme values of the
likelihood ratio represent certainties. For example, λ!∞ implies that the assertion
is true for such a piece of evidence, that is, P(a|E) = 1. Similarly, λ = 0 implies that
the assertion is not true with certainty, that is, P(~a|E) = 1.
39

Attributes and Structure of An Effective Board of Directors: A Theoretical Investigation

Uploaded by

Copyright:

Available Formats

Attributes and Structure of An Effective Board of Directors: A Theoretical Investigation

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Attributes and Structure of An Effective Board of Directors: A Theoretical Investigation

Uploaded by

Copyright:

Available Formats

ABACUS, 2018 doi: 10.1111/abac.

ADI MASLI, MATTHEW G. SHERWOOD, AND

Attributes and Structure of an Effective

Despite the board of directors’ importance, there is a lack of well-

responsibilities, and having the appropriate behavioural disposition towards fellow

provide evidence on the implied inferential inﬂuence of the presence or absence

ATTRIBUTES OF AN EFFECTIVE BOARD OF DIRECTORS

Independence of Board Members

abnormal accruals. Further, she ﬁnds signiﬁcant increases in abnormal accruals

Activity of Board Members

Competence of Board Members

Behavioural Attributes of Board Members

Safeguards within the Board of Directors

BAYESIAN MODEL FOR THE EFFECTIVENESS OF THE BOARD OF

In this section, we develop a theoretical model for the effectiveness of a board of

EVIDENTIAL DIAGRAM FOR BOARD OF DIRECTOR’S EFFECTIVENESS WITH FOUR

BOD is Indepedent (I) E I : Evidence pertaining to I

E1EB: Direct Evidence pertaining to

BOD Members possess

where i, j, k 2{I, C, A, B}.

to emphasize here that the measure of presence or absence of attributes in our

APPLICATION OF BAYESIAN MODEL UNDER VARYING LEVELS

Case 1: Impact of Presence or Absence of Attributes with Certainty on Board

Case 2: Impact of Presence or Absence of Attributes with Uncertainties on Board

EFFECTIVENESS OF THE BOARD AS A FUNCTION OF THE STRENGTH OF VARIOUS

COMPARISON OF EFFECTIVENESS OF THE BOARD AS A FUNCTION OF THE

EFFECTIVENESS OF THE BOARD AS A FUNCTION OF THE STRENGTH OF

EFFECTIVENESS OF THE BOARD AS A FUNCTION OF THE STRENGTH OF ATTRIBUTES

EFFECTIVENESS OF THE BOARD AS A FUNCTION OF THE STRENGTH OF ATTRIBUTES

Strength of Attributes in terms of Likelihood ratio, λ

strength of safeguards. As we see in Figure 4, in general, the posterior probability of

Case 4: Board Effectiveness—Additional Analysis

MEASUREMENT ISSUES ASSOCIATED WITH BOARD OF DIRECTOR

Board of Director Independence

MEASURES FOR ATTRIBUTES OF AN EFFECTIVE BOARD OF DIRECTORS

Board Attributes No. Measures

Independence 1 Percentage of directors not a member of management team

Active 1 Number of meetings held by the board of directors

Behavioural Attributes 1 Development of strong relationship and familiarity among directors

Board of Director Competence

degree(s) (i.e., bachelor, masters, doctoral), academic discipline (i.e., business,

Board of Director Activity

active. To gauge a director’s participation level, researchers can assess, for

Board of Director Behavioural Attributes

be informative as it outlines the expectations and acceptable practices of board

strengthening another. Thus, we conclude that a board of directors is most

GENERIC EVIDENTIAL DIAGRAM FOR N SUB-ASSERTIONS

Sub-assertion A1 E1 : Evidence pertaining to

Sub-assertion A2 E2 : Evidence pertaining to

Main Assertion (M) AND

Sub-assertion An En : Evidence pertaining to

Two Sub-assertions Case

Step 2: Combination of Potentials

To determine the combined potentials at variable M, we marginalize the above

where X1, X2, X3, and X4 are deﬁned below:

X1 = PðE1 j a1 ÞPða1 ÞPðE2 ja2 ÞPða2 ÞPðEM jmÞPðmÞ,

We obtain the following posterior probability of m:

We can rewrite the above posterior probability by deﬁning βi = λiOi as the

Extension to ‘n’ Sub-Assertions