The current issue and full text archive of this journal is available at

www.emeraldinsight.com/1746-5664.htm

Agency theory
Agency theory and corporate and corporate
governance governance
A study of the effectiveness of board in their
monitoring of the CEO 7
Livia Bonazzi and Sardar M.N. Islam Received October 2005
Financial Modelling Program, Centre for Strategic Economic Studies, Revised December 2006
Accepted December 2006
Victoria University of Technology, Melbourne, Australia

Abstract
Purpose The effect of corporate governance on firm performance has long been of great interest to
financiers, economists, behavioural scientists, legal practitioners and business operators. Yet there is
no consensus over what constitutes an effective corporate governance mechanism that induces agents
or managers to consistently act in the interest of share value optimisation. The purpose of this study is
to develop a model to resolve an on-going issue in financial economics: how can CEOs be effectively
monitored by the board of directors?
Design/methodology/approach A survey of the literature on corporate governance and the
relationship between board composition and financial performance leads to the development of the
proposed model, which is based on a framework which takes into account the probability of success
representing a CEOs ability, and the active monitoring function (which is represented by the numbers
of control visits) carried-out by the directors.
Findings The design of the model is aimed at identifying an optimal level of monitoring, which will
maximise share value, to guide internal and independent directors.
Research limitations/implications The model has limitations: it does not address the input of
other directors and it focuses solely on the monitoring function, even though boards also play
important roles in providing information and advice to management.
Originality/value The finding of this study contributes to the Agency Theory debate, in essence
that the board monitoring of CEO will improve the performance of the CEO and avoid possible conflict
of interests.
Keywords Corporate governance, Mathematical modelling, Optimization techniques, Chief executives
Paper type Research paper

1. Introduction
Separation between ownership and control of corporations characterises the existence of a
firm. The design of mechanisms for effective corporate control to make managers act in the
best interest of shareholders has been a major concern in the area of corporate governance
and finance (Allen and Gale, 2001), and continuing research in agency theory attempts to
design an appropriate framework for such control. In a corporation, the shareholders are
the principals and the managers are the agents working on behalf of, and for the interests
of, the principals. In agency theory, a well-developed market for corporate controls is
assumed to be non-existant, thus leading to market failures, non-existence of markets, Journal of Modelling in Management
Vol. 2 No. 1, 2007
moral hazards, asymmetric information, incomplete contracts and adverse selection pp. 7-23
among others. Various governance mechanisms have been advocated which include q Emerald Group Publishing Limited
1746-5664
monitoring by financial institutions, prudent market competition, executive DOI 10.1108/17465660710733022
JM2 compensation, debt, developing an effective board of directors, markets for corporate
2,1 control, and concentrated holdings. Developing an effective board of directors remains an
important and feasible option for an optimal corporate governance mechanism.
The limitations of the current literature in this field are the following. Currently,
several formal models exist based on the effectiveness of the board of directors in
designing efficient control mechanisms for corporate governance. However, for a
8 multi-level decision system such as corporate governance, an improved model
including uncertainty in model parameters is necessary to resolve the uncertain nature
of the agency problem in corporate governance. In addition, the modelling studies of
monitoring CEOs by the board, using empirical data, to resolve the agency problem in
a more quantitative form are not well-known in the current literature.
The objectives of this paper are; to provide an improved model for corporate
governance based on an effective board of directors and to analyse the results of this
model for the effectiveness of boards and their monitoring of the CEO. This study is
also expected to enrich the current literature by providing an empirical assessment of
CEO performance and monitoring, in precise quantitative terms.
The paper is structured in the following way: Section 2 presents the principal theoretical
concepts covered by the selected literature on the agency theory in finance, and on the role
and composition of boards. Section 3 presents a model to prescribe optimal monitoring
levels, featuring a case study of a director that sits on different boards, and needs to decide
how to apportion her monitoring efforts across the different CEOs. A mathematical model
is developed to resolve the dilemmas presented in the case study. Section 4 covers a
discussion on the algorithm, data, solution and results, as well as their interpretations.
Section 5 discusses the implications and findings of the models being proposed. Section 6
discusses the limitations of the model. The conclusion is presented in Section 7.

2. Corporate governance: the issues and the role of the board

2.1 The issues
Agents or managers may not always act in the best interest of shareholders when the
control of a company is separate from its ownership. In June 1959, Simon Herbert
(quoted in Baysinger and Hoskisson, 1990) proclaimed that managers might be
satisfiers rather than maximisers, that is, they tend to play it safe and seek an
acceptable level of growth because they are more concerned with perpetuating their
own existence than with maximising the value of the firm to its shareholders. But
shareholders delegate decision-making authority to the agent (CEO) with the
expectation that the agent will act in their best interest.
A comprehensive theory of the firm under agency arrangements was developed by
Jensen and Meckling (1976), who show that the principals (the shareholders) can assure
themselves that the agent will make the optimal decisions only if appropriate incentives
are given and only if the agent is monitored. Incentives include such things as stock
options, bonuses and prerequisites which are directly related to how well the results of
managements decisions serve the interests of shareholders. Monitoring consists of
bonding the agent, systematic reviews of management prerequisites, financial audits, and
placing specific limits on management decisions. These involve costs, which are an
inevitable result of the separation of corporate ownership and control. Such costs are not
necessarily bad for shareholders, but the monitoring activity they cover needs to be
efficient.
 In contrast, Demesetz (1983) and Fama and Jensen (1983) suggest that the primary Agency theory
monitoring of managers comes not from the owners but from the managerial labour and corporate
market. It is argued that management control of a large corporation is completely
separate from its security ownership. Efficient capital markets provide signals about governance
the value of a companys securities and thus about the performance of its managers.
If the managerial labour market is competitive both within and outside the firm, it will
tend to discipline the manager. Therefore, the signals given by changes in the total 9
market value of the firms securities become very important.
Kaplan and Reishus (1990) find evidence consistent with this argument: directors of
poorly performing firms, who therefore may be perceived to have done a poor job
overseeing management, are less likely to become directors at other firms. On the other
hand, reputational concerns do not correct all agency problems and can, in fact, create
new ones.

2.2 The role of the board

Garrat (1997) defines the function of the board as a collective responsibility to:
.
determine the companys purpose and ethics;
.
decide the direction, that is, the strategy;
.
plan;
.
monitor and control managers and CEO; and
.
report and make recommendations to shareholders.

Individual directors have personal liability if the company can be shown to have been
trading wrongfully (trading whilst insolvent), continuing to trade when there was no
reasonable prospect of its being able to pay its debts, illegally, carrying out activities
contrary to laws and regulations, e.g. Emron, AWB.
Independence of thought is demanded of all directors when on a board and this
requires that they pursue discriminating questions until they get satisfactory
answers that they and other board members understand. Pursuing the companys
interests above all else should be their priority. Directors competence, independence of
imagination and thought plus the skill to run an effective enterprise, will determine an
organisations success. The model developed below relates specifically to monitoring
and control by the board as a function of attendance at board meetings.

2.3 Board composition: directors independence

Empirical evidence supports the position that outside directors have been more
effective in monitoring managers and protecting the interests of shareholders. Larger
numbers of outside directors have been associated with a negative relation between
CEO turnover and performance (Weisbach, 1988), a lower probability that the board
pays greenmail in a control contest (Kosnik, 1990) and a lower probability that the
board adopts a poison pill (Mallette and Fowler, 1992).
Byrd and Hickman (1992) argue that the greater the proportion of outside directors,
the better the stock markets reaction to their firms tender offers for other firms.
Weisbach (1988) finds that the sensitivity of CEO turnover to firm performance increases
with the number of outside directors on the board. He suggests that outside directors who
own a substantial number of shares and who hold more corporate directorships
JM2 (presumably measuring the value they place on their reputations) are better at negotiating
2,1 a favourable deal for shareholders who face a takeover bid. Lastly, Hallock (1997) finds
that firms whose boards are interlocked (contain a CEO on whose board the firms CEO
serves) tend to pay their CEOs more. He argues that interlocked directors are less
independent and, consequently, give the CEO a larger fraction of the rents than necessary.

10 2.4 Board composition and CEO performance

Hermalin and Weisbach (1998) ask how boards can be chosen through a process partially
controlled by the CEO, yet, in many instances, still be effective monitors of him. Weisbach
(1988) reports that boards with at least 60 percent independent directors are more likely
than other boards to fire a poorly performing CEO. This type of firing is likely to add value
because boards are generally slow to fire CEOs. The stock price reaction to such firings is
hard to interpret because the firing announcement conveys information to the market both
about the event and about how the firm performed under the fired CEO, but there is
evidence that investors believe that these firings increase firm value. He quotes evidence
(Daily and Dalton, 1995) that firm performance improves modestly on average after a
CEO is replaced. Weisbachs study suggests that independent directors behave differently
to inside directors with respect to retaining or firing a CEO, but it is not clear whether
independent directors make better or worse decisions, on average.
Hermalin and Weisbach (1998) assume that the board and the CEO negotiate over both
the CEOs wage and the identity of new directors. These negotiations could be either
explicit or implicit, that is the CEO could nominate new board members subject to a tacit
understanding about the set from which they may be chosen. Were the CEO to violate this
understanding, the board would refuse to approve his nominees. The CEOs bargaining
power in these negotiations comes from his perceived ability relative to a replacement.
These negotiations determine the boards level of independence. Director
independence is important because a directors willingness to monitor the CEO
increases with his or her independence. Monitoring provides information about the
CEO used by the board in deciding whether to retain or to replace him/her.
However, a study conducted by Carver (1997) revealed that the board should
evaluate the performance of the CEO based on:
(1) the final achievement which has been initially defined;
(2) the compliance of the CEO to a set of executive limitations which were set-out
by the board;
(3) the alignment with the point of CEOs accountability as being set by the board;
and
(4) the regular assessment whether the CEO achieves points (1), (2) and (3).

Evaluation is such a crucial issue to maximize CEOs performance with expected end
result to improve the future performance, not to fire the CEO.

3. A model to prescribe optimal monitoring levels

3.1 The need for a model to prescribe optimal monitoring levels
The debate over the alleged greater propensity of independent directors to perform the
monitoring function compared with internal directors, calls for an objective measure of
the monitoring requirements that transcends the individuals distaste or propensity for
monitoring. Different directors put different weights on profits, as directors concerns Agency theory
for building reputations as competent managers may vary. So, it can be said that and corporate
directors have a different utility for monitoring. Some inside directors careers are tied
to the CEOs, so they rarely find it in their interest to monitor him/her. Also, directors governance
who value the opportunity to serve on other boards could have an incentive to establish
reputations for not rocking the boat, i.e. for not intensely monitoring the CEO.
The requirement for monitoring changes is based on several factors: the estimated 11
ability of the management team and the CEO; their level of cohesiveness; the leadership
skills of the CEO; and the current economic circumstances and risks faced by
the company. The model, therefore, should be dynamic and able to accommodate
variations in the factors over a period of time. The model that determines an objective
level of monitoring based on the variables discussed would provide an unbiased
approach for all directors to adopt, and for a chairman to drive and implement, to
ensure a CEO has optimal chances to succeed.
The authors are not aware of any other attempt to develop such a model, prior to
this paper.

3.2 The case study

The case study developed here is based upon the following assumptions:
(1) that a CEO who performs poorly is more likely to be replaced than one who
performs well;
(2) that CEO turnover (firing) is more sensitive to performance when the board is
more active in its monitoring function;
(3) that the director has an incentive to optimize the financial performance of her
companies, but that there is a natural tendency to minimize the monitoring
costs, often at the expense of results; and
(4) that 1-3 are a function of director attendance at scheduled board meetings.

Directors of boards often sit on more than one board. Directors are also known to have
several commitments and often conflicting requirements. They have time constraints
and thus need to carefully manage their efforts for maximum results. The principal
function that tests the effectiveness of a board is that of monitoring and control of the
CEOs and their performance. The greater the level of monitoring, the greater the
probability of success or enhanced financial performance.
A director sitting on multiple boards (six) is trying to determine how to allocate her time
and monitoring activity across the six organisations, for the following financial period.
She knows that the probability of successful performance of each organisation partly
depends on the degree of monitoring she performs, but she can only afford to spend limited
time, which is represented as a maximum of a total of 20 visits to the six CEOs.
The Board of Directors has an estimate of the ability of each CEO based on their
past performance. If the CEO is new, the estimate of his ability is low. He requires more
monitoring than old CEOs, since less is known about his ability. More independent
boards have a greater tolerance for this added monitoring, so they can afford to be
tougher with an incumbent CEO whose performance is marginal.
When new information about a CEOs performance is observed as, either profits
or some signal, the directors update their beliefs about the CEOs ability.
JM2 Poor performance lowers the boards assessment of the CEOs ability. Similarly, if a
CEO keeps his job, then retaining him must be worth more to the directors than
2,1 replacing him. A firing reveals that a CEO who was previously seen as better than the
expected value of a replacement is now not seen that way.
The CEOs ability is expressed as an estimate of the probability of successful
performance, as a function of the degree of monitoring by the director, measured in
12 number of control visits. The probability function is:
X i 0:1
pi
X i ki
where X is the number of director visits, and k is a constant for each CEO that
determines the shape of his ability (probability of success) function.
The costs and benefits associated with the success or failure of the companys
financial performance under the leadership of a CEO are expressed in terms of share
value movements. If the company performs well and delivers increased profits, the
value of its shares increases. Conversely, if the firm performs badly and the market
obtains signals of poor financial performance, the effect is felt in the reduced share
price. The share price sensitivity of each firm is illustrated in the data table.
The directors remuneration is based on the share price performance, therefore she has
an interest in maximising share value. However, she has limited resources: as an example,
she can only afford 20 monitoring visits, and as the location of each CEO is different, each
destination takes a different time to reach, plus the time spent meeting with the CEO,
checking reports, etc. in hours. She has a total of 80 hours available for the visits.

3.3 The model

3.3.1 Decision variables. The director must make two separate decisions, which are
related:
(1) Determine whether to keep or fire a CEO. This is a binary decision where:
Yi (1 retain the CEO) or (0 fire the CEO).
(2) Determine how many monitoring visits to assign to each CEO: X 1,2,3, . . . 20.
3.3.2 Objective function. The director wants to maximise the expected share value of
her decisions, which is, for each respective CEO, the product of the probability of their
success by the increase in share value, minus the product of the probability of their
failure by the related drop in share value:
max:


A1 X 1 A2 X 2 A3 X 3 A4 X 4 A5 X 5 A6 X 6

X 1 p1 X 2 p2 X 3 p3 X 4 p4 X 5 p5 X 6 p6

 1
B1 X 1 B2 X 2 B3 X 3 B4 X 4 B5 X 5 B6 X 6
2
X 1 p1 X 2 p2 X 3 p3 X 4 p4 X 5 p5 X 6 p6
where A is share value increase if successful; and B is share value decrease if
unsuccessful.
3.3.3 Constraints. The constraints that apply to this model are the following.
The director has a maximum of 80 hours to dedicate towards the monitoring visits.
This is measured as follows:
 5X 1 Y 1 3X 2 Y 2 4X 3 Y 3 6X 4 Y 4 6X 5 Y 5 8X 6 Y 6 # 80 2 Agency theory
Next, the director must ensure she does not exceed her 20 visit quota. This can be done and corporate
by the following: governance
X 1 X 2 X 3 X 4 X 5 X 6 # 20 3
Also, she needs to ensure that the visits are not wasted on the CEOs that will be fired 13
(new CEOs replacing fired ones will have a new k value, which is to be entered into the
model, and a new solution run again). For simplicity purposes, when a CEO is to be
replaced, the model will recommend no visits. It follows that when a replacement is
found, and his/her k value established, the model will provide a revised response. The
linking constraints are the following:
X i 2 20Y i # 0 where i 1; 2; 3; 4; 5; 6 4
This ensures that an Xi variable can be greater than zero only of its respective Yi is one.
The expected share value (V) is the sum of the probability of each CEO succeeding
times the respective expected increased share value, minus the probability of failure
times the respective expected drop in share value.

4. Algorithm, data, solution, and results

4.1 Algorithm
The expected share value is:
X1
6
{Ifpi 0; then zero; otherwise; pi Ai 1 2 pi Bi } 5

The model for CEO monitoring optimisation is illustrated in the Table I. In essence, it
is:
maximise the expected share value:
X1
6
{Ifpi 0; then zero; otherwise; pi Ai 1 2 pi Bi } 6

subject to:
5X 1 Y 1 3X 2 Y 2 4X 3 Y 3 6X 4 Y 4 4X 5 Y 5 8X 6 Y 6 # 80
X 1 X 2 X 3 X 4 X 5 X 6 # 20
X i 2 20Y i # 0 where i 1; 2; 3; 4; 5; 6
X i $ 0 and integer
Y i binary

4.2 Data
The data used in the model (reported below) is illustrative only but based on a realistic
scenario. For this case study, the data were determined as follows:
.
number of boards/CEOs: 6;
.
total available number of monitoring visits: 20;
 2,1

14
JM2

problem
Table I.
Solution to the CEO
monitoring optimisation
Y X k P A B V
Fire CEO? No. of Prob. of Share value Hours
(1 keep, monitoring Linking Probability success increase if Share value decrease Expected share required per
CEO 0 fire) visits constraints parameter (percent) successful ($) if un-successful ($) movement ($) visit

1 1 8 212 4.2 66.4 8,000 2 2,000 4,639 5

2 0 0 0 18.0 0.6 2,000 2 4,400 N/A 3
3 1 2 218 3.6 37.5 1,000 2 1,500 2 563 4
4 1 5 215 1.2 82.3 7,800 2 6,000 5,352 6
5 0 0 0 5.0 2.0 2,000 2 7,000 N/A 6
6 0 0 0 6.8 1.5 7,600 2 1,500 N/A 8
Total used 15 Total 9,428 78
Maximum
Total available 20 hours available 80
 .
total available hours spent on visits: 80; Agency theory
.
time involved in monitoring each CEO (Table II); and corporate
.
ability factor, k Factor for each CEO (Table III); and governance
.
expected share movement due to success/failure of CEO (Table IV).

4.3 Solution and results

The Excel Function Solver provides a series of optimal solutions based on specific
15
variables given. Given the parameters in Section 4.2 data, the solution is represented in
Table I.
The model recommends to fire and replace three CEOs, and distribute the time
available across the remaining CEOs so that 78 hours are spent on a total of 15
monitoring visits. This behaviour will optimise the value of the shares and generate an
aggregate increase of $9,428 million.
The variables can easily be modified by the director to reflect current information
about market reactions and signals on the ability of each CEO. The model will continue
to provide an optimal solution upon each round of updated variables.

5. Implications and findings

Consistent with the widely held belief that entrenched CEOs or CEOs who have
cultivated personal loyalty (and whose ability is therefore more likely to be

CEO1 5h
CEO2 3h
CEO3 4h
CEO4 6h
CEO5 6h
CEO6 8h Table II.

CEO1 4.2
CEO2 18.0
CEO3 3.6
CEO4 1.2
CEO5 5.0
CEO6 6.8 Table III.

CEO (company) Increase if success ($m) Decrease if failed ($m)

Company 1 8,000 2,000

Company 2 2,000 4,400
Company 3 1,000 1,500
Company 4 7,800 6,000
Company 5 2,000 7,000
Company 6 7,600 1,500 Table IV.
JM2 overestimated), are less scrutinized and face lower standards, the research finding
2,1 reveals that, all other things being equal, the more the board values the incumbent
CEO, that is, the lower the k factor, the less intensely he will be monitored by the board.
As revealed by Carver (1997), this study also implies that the evaluation of the CEO
is to improve CEOs performance in the future, not to fire the CEO. However, once the
decision to fire a CEO needs to be made, such decision is also contingent upon the
16 potential risk and magnitude of loss or gain in share value, which is a firm specific
variable. If the magnitude of loss due to market signals of poor performance is very
large, compared to the magnitude of share value increase if the financial performance is
positive, even where the probability of failure is low, the expected benefit might still be
negative. This situation requires a very capable CEO whose k factor approaches zero.
The model addresses both sides of the equation by determining the optimal
intensity with which to monitor a CEO based upon the interrelation of the boards
perception of the ability of a CEO (measured as the k factor) and the expected market
reaction to the CEOs performance (expected share movement).
In the solution shown in the Appendix (Tables AI and AII and Figure A1), out of six
CEOs, only numbers 1, 3 and 4 shall be retained, as they are likely to provide the best
returns, with the right amount of monitoring. The director will need to replace the other
CEOs, estimate their respective k factor, and re-enter the new variables in the model to
achieve the new optimal distribution of monitoring resources.
Although this model has been developed in relation to boards of directors, its
application is fairly general. It can be extended to job-matching problems where some
workers are recognized to be more valuable than others.
It can also be used by a director to decide whether to keep a particular board
assignment, or to resign, based on the risk which is expressed as the number of
monitoring visits required to achieve a positive result.
The model is consistent with both realities observed by the literature: active
monitoring in some firms and CEO dominance in others. A company has a crisis
induced by poor profits and the board is forced to act. The new CEO has no bargaining
power, a high k factor, and, thus, has to contend with an active board. None of this
would happen if the previous managers performed better; they would maintain their
jobs and their control over the board.
The model is consistent with a number of empirical regularities: CEO turnover is
negatively related to performance and this relation is stronger when the board is more
independent, that is, monitors more actively. The probability that stronger
independent directors are added to the board increases following poor corporate
performance. And boards tend to become less scrutinising over the course of the CEOs
career.

6. Limitations of the model

The model being proposed in this study has a limitation where the focus is solely on
the monitoring of boards. The institutional literature (Mace, 1971; Vance, 1978)
emphasizes that boards also play important roles in providing information and advice
to management, and serving as a training ground for future CEOs. A richer model of
boards should take into account these roles as well.
More research is also required in devising objective measures to assess the ability of
potential directors to discern and address tough company situations, to know when
painful decisions are called for, to nurture their own independent opinions, and to Agency theory
decipher the conceptual errors of corporate officers. In fact, more work is required in and corporate
the area of pre-board selection tests, and an accreditation system that guarantees at
least some fundamental levels of skill, knowledge and experience. governance
The model does not address the input of other directors, and it assumes that each
director will act in a similar fashion, with respect to their own portfolios of board
positions. However, the issue of free-riding does arise and some directors might 17
perceive that the monitoring carried out by their fellow directors is adequate and
choose to decrease their share of monitoring. The literature reviewed highlights that
this behaviour is usually observed in non-independent directors, as internal directors
have an incentive to ingratiate themselves with the CEO, to ensure continuing support
as board members. This explains why some CEOs are able to avoid scrutiny.
Corporate policy should address this problem, and some commentators suggest
smaller boards (to reduce free-riding), for outside directors, more meetings, director
pay linked to stock performance, and appointment of a lead director (if not the
chairman) who is separate from the CEO. These policies would lead to better
monitoring of the CEO.
The model represents a practical, quick solution to a simple question: As a Director
on several boards, how do I apportion my time across the different boards?. However,
it does not fully address possible complications such as the extra time involved in
replacing non-performing CEOs, or lobbying other directors to make the right
decisions. It may, however, assist a director who is considering resigning from a board,
by providing improved objective insight into the financial outcomes following such
decision.

7. Conclusion
Reputed literature on the agency costs associated with the separation of ownership and
management in corporate governance has been reviewed in this study. A model has
been developed that determines the optimal level of monitoring of CEOs by directors,
which is considered a crucial function in determining the financial performance of the
firm. In the scenario illustrating the model, the optimal value of monitoring is
represented by 15 monitoring visits across three CEOs for a total of 78 hours and the
replacement of three CEOs. The expected share value movement is an increase of
$9,428 million.
The above-mentioned optimal level monitoring model also incorporates the
premise that it is important for the board to evaluate CEOs performance by
defining the final achievement as well as setting the executive limitations and
point of accountability of the CEO. Such a monitoring needs to be conducted on a
regular basis to maintain an optimal level of expected CEO performance.
Even though corporate law outlines that shareholders choose the board of directors,
in practice, shareholders almost always vote for the slate proposed by management.
Moreover, this slate is approved by, if not chosen by, the very CEO these directors are
supposed to monitor. The resulting governance system has been criticized as
ineffective in controlling management.
Consistent with the continuing research in agency theory, the finding of this
study contributes to resolve the agency problem, in essence that the board
JM2 monitoring of CEO will improve the performance of the CEO and avoid possible
2,1 conflict of interests.
In addition, an area of research that might draw light on the determinants of firm
performance is the role of social connections in determining who is chosen to fill board
seats, whether they are classified as independent or otherwise. A study on meritocracy
would inevitably need to measure the individual value of each director within the
18 function they are appointed to cover, the collective contribution of the board as a whole,
and the relationship between these measures and financial performance. Managing
corporate governance based on social business ethics is also essential (see arguments
in Islam, 2002). The next challenge is to structure a model which measures all
variables.

References
Allen, F. and Gale, D. (Eds) (2001), Comparing Financial Systems, MIT Press, Cambridge, MA.
Baysinger, B.D. and Hoskisson, R.E. (1990), The composition of boards of directors and
strategic control: effects on corporate strategy, Academy of Management Review, Vol. 15,
pp. 72-87.
Byrd, J.W. and Hickman, K.A. (1992), Do outside directors monitor managers? Evidence from
tender offer bids, Journal of Financial Economics, Vol. 32, pp. 195-222.
Carver, J. (1997), Board Assessment of the CEO, Jossey-Bass, New York, NY.
Daily, C.M. and Dalton, D.R. (1995), CEO and director turnover in failing firms: an illusion of
change?, Strategic Management Journal, Vol. 16, pp. 393-400.
Demesetz, H. (1983), The structure of ownership and the theory of the firm, Journal of Law &
Economics, Vol. 26, pp. 375-90.
Fama, E.F. and Jensen, M.C. (1983), Separation of ownership and control, Journal of Law &
Economics, Vol. 26, pp. 301-25.
Garrat, B. (1997), The Fish Rots from the Head, Harper Collins, London.
Hallock, K. (1997), Reciprocally interlocking boards of directors and executive compensation,
Journal of Financial and Quantitative Analysis, Vol. 32, pp. 331-44.
Hermalin, B.E. and Weisbach, M.S. (1998), Endogenously chosen boards of directors and their
monitoring of the CEO, American Economic Review, Vol. 88, pp. 96-118.
Islam, S.M.N. (2002), Modelling corporate governance and finance: a multi-level
optimisation approach and implications for governance mechanism design, paper
presented at the Victoria Graduate School of Business, 27 November, Victoria University,
Melbourne.
Jensen, M. and Meckling, W. (1976), Theory of the firm: managerial behavior, agency costs and
ownership structure, Journal of Financial Economics, Vol. 39, pp. 1021-39.
Kaplan, S.N. and Reishus, D. (1990), Outside directorships and corporate performance, Journal
of Financial Economics, Vol. 27, pp. 389-410.
Kosnik, R.D. (1990), Effects of board demography and directors incentives on corporate
greenmail decisions, Academy of Management Journal, Vol. 33, pp. 129-50.
Mace, M.L. (1971), Directors: Myth and Reality, Harvard Business School Press, Boston, MA.
Mallette, P. and Fowler, K.L. (1992), Effects of board composition and stock ownership on the
adoption of poison pills, Academy of Management Journal, Vol. 35, pp. 1010-35.
Vance, S.C. (1978), Corporate governance: assessing corporate performance by boardroom Agency theory
attributes, Journal of Business Research, Vol. 6, pp. 203-20.
and corporate
Weisbach, M.S. (1988), Outside directors and CEO turnover, Journal of Financial Economics,
Vol. 20, pp. 431-60. governance

Further reading 19
Agrawal, A. and Knoeber, C.R. (1996), Firm performance and mechanisms to control agency
problems between managers and shareholders, Journal of Financial and Quantitative
Analysis, Vol. 31, pp. 377-97.
Bavly, D. (1999), Corporate Governance and Accountability, Quorum Books, Westport, CT.
Bhagat, S. and Black, B. (1998), Do Independent Directors Matter? Draft Paper, Graduate School
of Business, Boulder, CO.
Black, S. and Moersch, M. (Eds) (1998), Competition and Convergence in Financial Markets:
The German and Anglo-American Models, North-Holland Elsevier, Amsterdam.
Blackburn, V. (1994), The effectiveness of corporate control in the US, Corporate Governance:
An International Review, Vol. 2, pp. 196-202.
Brickley, J.A., Coles, J.L. and Terry, R.L. (1994), The board of directors and the enactment of
poison pills, Journal of Financial Economics, Vol. 35, pp. 371-90.
Charkham, J. (1994), Keeping Good Company: A Study of Corporate Governance in Five
Countries, Clarendon Press, Oxford.
Chew, D. (Ed.) (1997), Studies in International Corporate Finance and Governance Systems,
Oxford University Press, New York, NY.
Colesa, J., McWilliams, V. and Senc, N. (2001), An examination of the relationship of governance
mechanisms to performance, Journal of Management, Vol. 27, p. 23.
Dalton, D.R., Daily, C.M., Ellstrand, A.E. and Johnson, J.L. (1998), Meta-analytic reviews of
board composition, leadership structure, and financial performance, Strategic
Management Journal, Vol. 19, pp. 269-90.
Davis, J.H., Schoorman, F.D. and Donaldson, L. (1997), Toward a stewardship theory of
management, Academy of Management Review, Vol. 22, pp. 20-47.
Demesetz, H. and Lehn, K. (1985), The structure of corporate ownership: causes and
consequences, Journal of Political Economy, Vol. 93, pp. 1155-77.
Ermann, M. and Lundman, R. (Eds) (1992), Corporate and Governmental Deviance, Oxford
University Press, New York, NY.
Forbes, D. (1999), Cognition and corporate governance: understanding boards of directors as
strategic decision-making groups, Academy of Management Review, Vol. 24, p. 489.
Hill, C.A. and Snell, S.A. (1989), Effects of ownership structure and control on corporate
productivity, Academy of Management Journal, Vol. 32, pp. 25-46.
Hopt, K.J., Kanda, H., Roe, M.J., Wymeersch, E. and Prigge, S. (Eds) (1998), Comparative
Corporate Governance: The State of the Art and Emerging Research, Clarendon Press,
Oxford.
Hubbard, R. (Ed.) (1990), Asymmetrical Information, Corporate Finance and Investment, Chicago
University Press, Chicago, IL.
Jensen, M.C. (1986), Agency costs of free cash flow, corporate finance and takeovers, American
Economic Review, Vol. 76, pp. 323-9.
JM2 Jensen, M.C. and Meckling, W.H. (1976), Theory of the firm: managerial behaviour, agency costs
and ownership structure, Journal of Financial Economics, Vol. 4, pp. 305-60.
2,1
Keong, L.C. (Ed.) (2002), Corporate Governance: An Asia-Pacific Critique, Sweet ands Maxwell
Asia, Hong Kong.
Kini, O., Kracaw, W. and Mian, S. (1995), Corporate takeovers, firm performance and board
composition, Journal of Corporate Finance, Vol. 1, pp. 383-412.
20
Lawrence, J. and Stapledon, G.P. (1997), Independent Directors in Australia: Empirical Evidence
and Policy Prescriptions, University of Melbourne Centre for Corporate Law and Securities
Regulation, Melbourne.
MacAvoy, P., Cantor, J., Dana, J. and Peck, S. (1983), ALI Proposals for Increased Control of the
Corporation by the Board of Directors: An Economic Analysis, Principles of Corporate
Governance and Structure, Business Roundtable, New York, NY.
Mathiesen, H. (2002), Managerial Ownership and Financial Performance, PhD thesis,
Copenhagen Business School, Department of International Economics and Management,
Copenhagen.
Mikkelson, W.H. and Partch, M.M. (1997), The decline of takeovers and disciplinary
management turnover, Journal of Financial Economics, Vol. 44, pp. 205-28.
Myers, S. (1977), Determinants of corporate borrowing, Journal of Financial Economics, Vol. 5,
pp. 147-75.
Ragsdale, C. (2001), Spreadsheet Modeling and Decision Analysis, Thomson Learning, Melbourne.
Ross, S. (1973), The economic theory of agency: the principals problem, American Economic
Review, Vol. 63, pp. 134-9.
Schellenger, M.H., Wood, D.D. and Tashakori, A. (1989), Board of director composition,
shareholder wealth and dividend policy, Journal of Management, Vol. 15, pp. 457-67.
Shleifer, A. and Visney, R. (1997), A survey of corporate governance, Journal of Finance, Vol. 52,
pp. 737-83.
Stiles, P. and Taylor, B. (1993), Maxwell the failure of corporate governance, Corporate
Governance An International Review, Vol. 1 No. 1, pp. 34-45.
Stoeberl, P. and Sherony, B. (1985), Board efficiency and effectiveness, in Matter, E. and Ball, M.
(Eds), Handbook for Corporate Directors, McGraw-Hill, New York, NY, pp. 12.1-12.10.
Vance, S.C. (1964), Boards of Directors: Structure and Performance, University of Oregon Press,
Eugene.
Williamson, O. (1975), Markets and Hierarchies: Analysis and Antitrust Implications, Free Press,
New York, NY.
Wilson, R. (1968), The theory of syndicates, Econometrica, Vol. 36, pp. 119-32.
 Appendix

Fire CEO? No. of Probability of Share value Share value Expected Hours
(1 keep, monitoring Linking Probability success increase if decrease if share required
CEO 0 fire) visits constraints parameter (percent) successful ($) unsuccessful ($) movement ($) per visit

1 1 8 212 4.2 66.4 8,000 2 2,000 4,639 5

2 0 0 0 18.0 0.6 2,000 2 4,400 N/A 3
3 1 2 218 3.6 37.5 1,000 2 1,500 2 563 4
4 1 5 215 1.2 82.3 7,800 2 6,000 5,352 6
5 0 0 0 5.0 2.0 2,000 2 7,000 N/A 6
6 0 0 0 6.8 1.5 7,600 2 1,500 N/A 8
Total used 15 Total 9,428.46 78
Maximum
Total hours
available 20 available 80
governance
and corporate

optimisation model
CEO monitoring
Agency theory

Table AI.
21
JM2
Monitoring 4.2 12.0 3.6 1.2 5.0 6.8
2,1 CEO 1 CEO 2 CEO 3 CEO 4 CEO 5 CEO 6
Visits (percent) (percent) (percent) (percent) (percent) (percent)

0 2.38 0.83 2.78 8.33 2.00 1.47

1 21.15 8.46 23.91 50.00 18.33 14.10
22 2 33.87 15.00 37.50 65.63 30.00 23.86
3 43.06 20.67 46.97 73.81 38.75 31.63
4 50.00 25.63 53.95 78.85 45.56 37.96
5 55.43 30.00 59.30 82.26 51.00 43.22
6 59.80 33.89 63.54 84.72 55.45 47.66
7 63.39 37.37 66.98 86.59 59.17 51.45
8 66.39 40.50 69.83 88.04 62.31 54.73
9 68.94 43.33 72.22 89.22 65.00 57.59
10 71.13 45.91 74.26 90.18 67.33 60.12
11 73.03 48.26 76.03 90.98 69.38 62.36
12 74.69 50.42 77.56 91.67 71.18 64.36
13 76.16 52.40 78.92 92.25 72.78 66.16
14 77.47 54.23 80.11 92.76 74.21 67.79
15 78.65 55.93 81.18 93.21 75.50 69.27
16 79.70 57.50 82.14 93.60 76.67 70.61
17 80.66 58.97 83.01 93.96 77.73 71.85
Table AII. 18 81.53 60.33 83.80 94.27 78.70 72.98
Probability parameter 19 82.33 61.61 84.51 94.55 79.58 74.03
(success factor) 20 83.06 62.81 85.17 94.81 80.40 75.00

100%
90%
80%
Probability of Success

70%
60%
50%
12.0 CEO 2
40%
1.2 CEO 4
30%
5.0 CEO 5
20%
10%
Figure A1. 0%
Selected probability 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
functions Monitoring Hours per month

About the authors

Livia Bonazzi graduated in Architecture (Honours), Arts, Planning and Design (1990) at University
of Melbourne and Master of Business Administration (1997) at Melbourne Business School. She is a
Management Consultant and Operations Director for boutique consulting firms with seven years
experience in corporate strategy and sustainable performance improvement for a diverse range of
industries and corporations. She held roles as Senior Executive in Local Government (Director City
Infrastructure), Operations Manager and Marketing Manager in the Manufacturing and Agency theory
Engineering Industry; previously practicing as an architect and Construction Project Manager.
Livia has extensive experience in strategic policy development and implementation across several and corporate
functional areas: strategic asset management; risk management; urban design; logistics; property governance
management; road safety and engineering; emergency management including disaster planning.
Livia Bonazzi is the corresponding author and can be contacted at: [email protected]
Sardar M.N. Islam is a Professor, and Director, Decision Sciences and Modelling Program,
Victoria University, Australia. He is also associated with the Financial Modelling Program and 23
the Law and Economics program. He has taught for more than 15 years at different universities
in various countries. He has interest in other academic fields in finance, business, and law where
he has expertise, interest, or experience in teaching, supervision, or research. He has published 16
books and monographs and more than 170 technical papers in the above areas. His research has
gained international reputation. E-mail: [email protected]

To purchase reprints of this article please e-mail: [email protected]

Or visit our web site for further details: www.emeraldinsight.com/reprints