Risk Management Framework

1. RISK MANAGEMENT
FRAMEWORK
MSC ACCOUNTANCY & FINANCE :
CORPORATE GOVERNANCE
& OPERATIONS RISK ANALYSIS AND CONTROL
Stephen Ong
BSc(Hons) Econs (LSE),
MBA International
Business(Bradford)
Visiting Fellow, Birmingham City University
Visiting Professor, Shenzhen University

2. • Discussion : Corporate
Cultures and Governance
1
•Risk Management
Framework
2
• Case Presentation: GM3
Today’s Overview

3. 1. Open Discussion
• Mayer, Colin (2002) “Corporate Cultures and
Governance: Ownership, Control and
Governance of European and US
Corporations”, TRANSATLANTIC
PERSPECTIVES ON US-EU ECONOMIC
RELATIONS:CONVERGENCE, COOPERATION
AND CONFLICT ,Conference paper, JFK School
of Government, Harvard University, April 11-
12

5. Corporate Governance Research in
Accounting & Auditing

6. Formal Risk Rating
Corporate Risk Rating must reflect a
transparent understanding of the risk appetite
of the Board
IMPACTONBUSINESS
Critical
4
4 8 12 16
Unacceptable level of risk exposure,
which requires extensive management
Major
3
3 6 9 12
Moderate
2
2 4 6 8
4 – 8: Risk management measures need
to be put in place and monitored
Minor
1
1 2 3 4
Almost
Never
1
Unlikely
2
Likely
3
Almost
Certain
4
1 – 3: Acceptable level of risk subject to
regular monitoring
LIKELIHOOD OF OCCURING

7. Where there is an action....
• Recognise real issues within
in organisation
• Focus appropriate
management attention
• Delivery of assurance
through review and closure
• Need to recognise
relationship to achievement
of corporate and
operational targets.
• Relevance to Governance
Statements in Annual
Reports
High
Medium
Low
Low Medium High
S
I
G
N
I
F
I
C
A
N
C
E
PROBABILITY
Requires
close
monitoring
Manage and
monitor
Significant
focus and
action
Accept but
monitor
Management
effort
worthwhile
Manage
and
monitor
Accept risks
Accept but
periodically
review
Accept but
monitor

8. Hierarchy of risk reporting
Board
Executive
Management
Operational
Management
• Continuous examination of
operational risks
• Consideration of material risks
within risk definitions
• Consideration at meetings
• Review of performance against business
expectations at EMT
• View to risk horizon through effective
scanning and intelligence
• Relationship with middle management
• Commitment to upwards reporting
• Review of business critical risks and
actions to manage
• Relationship with Executive
Management
• Focus on annual business performance
data link to Governance Statement
• Setting risk appetite

9. 2. RISK
MANAGEMENT
MODELS

10. COSO Risk Management Model (2004)
Committee of Sponsoring Organisations of the
Treadway Commission on Enterprise Risk
Management (ERM) characteristics :
1. Process
2. Operated at every level
3. Applied in strategy setting
4. Applied across the enterprise
5. Identifies key events and manage their risks
6. Geared to achievement of objectives

11. COSO FRAMEWORK (2004)
8 COMPONENTS
1. Internal Environment
2. Objective Setting
3. Event Identification
4. Risk Assessment
5. Risk Response
6. Control Activities
7. Information &
Communication
8. Monitoring

12. COSO FRAMEWORK (2004)
4 OBJECTIVES
1.Strategic
2.Operational
3.Reporting
4.Compliance

13. COSO FRAMEWORK (2004)
4 ENTERPRISE LEVELS
1.Subsidiary
2.Business Unit
3.Division
4.Entity-Level

14. Benefits of ERM (COSO, 2004)
• Alignment of risk appetite
and strategy
• Link growth, risk and return
• Choose best risk response
• Minimise surprises and
losses
• Identify and manage risks
across the organisation
• Provide responses to
multiple risks
• Seize opportunities
• Rationalise capital

15. CIMA Risk Management Cycle (2002)

16. ISO31000 Risk Management Framework
(IRM 2010)

17. ISO31000 Risk Management process

18. Risk Management policy (ISO31000)
• Risk management & internal control objectives (corporate
governance)
• Statement of the attitude of the organisation to risk (risk strategy)
• Description of the risk aware culture or control environment
• Level and nature of risk that is acceptable (risk appetite)
• Risk management organisation & arrangements (risk architecture)
• Details of procedures for risk recognition & ranking (risk
assessment)
• List of documentation for analysing & reporting risk (risk protocols)
• Risk mitigation requirements & control mechanisms (risk response)
• Allocation of risk management roles & responsibilities
• Risk management training topics & priorities
• Criteria for monitoring & benchmarking of risks
• Allocation of appropriate resources to risk management
• Risk activities and risk priorities for the coming year

19. Risk architecture at large PLC

20. WHO is Responsible for Risk Management?
(ISO31000)
• CEO/ Board
• Business unit manager
• Individual employees
• Risk manager
• Specialist risk management
functions
• Internal audit manager

21. ISO31000 Risk
Management
Responsibilities

22. ISO31000 Risk Management Techniques

23. FIRM Risk Scorecard

24. FIRM Risk Scorecard : Financial Risks
EXTERNALLY DRIVEN
• Accounting
Standards
• Interest Rates
• Foreign Exchange
• Funds & Credit
INTERNALLY DRIVEN
• Internal Control
• Fraud
• Historical Liabilities
• Investments
• CAPEX Decisions
• Liquidity & Cashflow

25. FIRM Risk Scorecard : lnfrastructure Risks
EXTERNALLY DRIVEN
• Communications
• Transport Links
• Supply Chain
• Terrorism
• Natural Disasters
• Pandemic
INTERNALLY DRIVEN
• Recruitment
• People Skills
• Health & Safety
• Premises
• IT Systems

26. FIRM Risk Scorecard : Marketplace Risks
EXTERNALLY DRIVEN
• Economic
Environment
• Technology
Developments
• Competition
• Customer Demand
• Regulatory
Requirements
INTERNALLY DRIVEN
• M&A Activity
• R&D Activity
• Intellectual Property
• Contracts

27. FIRM Risk Scorecard : Reputational Risks
EXTERNALLY DRIVEN
• Product Recall
• CSR
• Public Perception
• Regulator
Enforcement
• Competitor
Behaviour
INTERNALLY DRIVEN
• Brand Extensions
• Board Composition
• Control Environment

28. 3. RISK APPETITE,
CULTURE AND BEHAVIOUR

29. Risk Appetite, Culture and Behaviour
• Biases in human behaviour which limit
human rationality, reduce the quality of
corporate governance and cause bad
decision making.
• Review of biases including excessive
loyalty to the CEO by members of the
board of directors, excessive risk taking,
overconfidence and hubris by managers
as well as attitudes towards risk taking
according to gender and age.

30. Learning Outcomes
1. Perform a critique of the key assumption in
finance and economics that humans (always)
act rationally
2. Evaluate the potential impact of biases in
human behaviour on managerial decision
making
3. Explain the link between human psychology
and agency problems
4. Assess the usefulness and limitations of
possible solutions to behavioural biases.

31. Introduction
• Corporate governance theory, such as the principal–
agent model, predicts that conflicts of
interests result in value destruction or
expropriation.
• However, it nevertheless assumes that all economic
actors concerned behave rationally.
• More generally, economics and finance assume that
– individuals are value-maximising rational decision
makers, and
– they make optimal decisions in an uncertain world based
on all the information available at the time of the
decision.

32. Introduction (continued)
• The efficient market hypothesis (EMH),
which is central to finance and essential to
valuation models, states that security prices
always fully reflect all available information.
• Hence, market prices of securities are always
a true and fair reflection of their value.
• Hence, mispricings should NOT occur
– Incidences of short-term mispricing do not
necessarily call for the rejection of the EMH
– However, long-term, persistent mispricing does.

33. • The question arises as to whether there are
such incidences of persistent mispricings.
• The answer to this question seems
straightforward
– The internet stock market bubble of the late
1990s and the ensuing global stock market crash
of 2001/2 is just one fairly recent case where
stock markets got things wrong
– There are other more regularly occurring
incidences of persistent mispricings such as
those observed in initial public offerings (IPOs).
Introduction (continued)

34. • The existence of these persistent mispricings
has urged academics and practitioners to
seek possible explanations.
• One such explanation is human psychology.
• Contrary to what classical economics and
finance assume, humans may
not always be fully
rational.
• Their rationality may be bounded or limited.
Introduction (continued)

35. Bounded Rationality
• The concept of bounded rationality was
formalised by Herbert Simon in 1947.
• Humans are typically not able to analyse all the
information available to them due to
– time constraints, and
– limits to their cognitive resources.
• Limits to cognitiveresources include
limits to
– our intelligence,
– our memory, and
– our attention span.

36. Bounded Rationality (Continued)
• Hence, humans frequently do not make
optimal decisions.
• Our brains have been wired by evolution to
deal with a complex world and our limited
time and cognitive resources.
• Hence, our minds use rules of thumb
also called
– heuristics,
– algorithms, and
– mental modules.

37. • As this heuristic simplification is used by all humans,
this may cause biases in human behaviour, resulting in
systematic mistakes such as persistent mispricing in
capital markets.
• However, there are at least five other potential sources
of bias in human behaviour
1. emotions,
2. social interactions,
3. overconfidence,
4. risk taking according to gender
and age, and
5. reflexive loyalty.
Bounded Rationality (Continued)

38. Heuristic Simplification
• Given the limited time and cognitive
resources,
– we focus on subsets of information;
– we analyse problems in isolation;
– we focus on similarity; and
– we only slowly change our beliefs.

39. Focusing on Subsets of Information
• Some of the effects caused by focusing on subsets of
the information available include the halo effect.
• The halo effect consists of admiring an exceptional
characteristic of an individual and then widen this
positive assessment to all the other characteristics
of this individual.
• The halo effect may explain why in certain
corporations everybody is in awe of the CEO.
• At worst, this effect may prevent other
constituencies within the firm from scrutinising the
CEO’s decisions.

40. Analysing Problems in Isolation
• We tend to analyse decision problems in
isolation, ignoring the broader
context.
• We tend to compartmentalise decision
problems or frame them in a narrow way.
• Most problems can be easily and safely
compartmentalised.
• However, there is a danger of narrow framing
or packaging.
• The way the information is presented to the
decision maker may influence the actual
decision.

41. Analysing Problems in Isolation
(Continued)
• Examples of compartmentalisation or
narrow framing (packaging) include
–mental accounting,
–the disposition effect,
–loss aversion, and
–the house money effect.

42. • Mental accounting consists of most investors
keeping two accounts in their brains
– one for the gains from their investments, and
– one for the losses.
• It may explain the so called disposition effect
– Investors tend to sell winners too soon
and losers too late
– The reason why investors treat losses differently may
be due to the fact that losses are perceived to be
painful
– Postponing the sale of the losers avoids the painful
realisation of a loss.
Analysing Problems in Isolation (Continued)

43. • Loss aversion consists of the fact that
most investors tend to be very risk averse,
even when facing very small risks
–It may be explained by the fact that we tend
to turn a continuous return into a discrete
return
–We distinguish between two regions, i.e. the
region for losses and that for gains
–The two regions are clearly separated by the
break-even point.
Analysing Problems in Isolation (Continued)

44. • The house money effect is the greater
willingness of people to gamble with
money they have recently
won
– This behavioural pattern may be explained by the
fact that the recent positive sensation of a gain
may cancel out the possible future painful
sensation of a loss
– A manager may be more willing to take on
unnecessary risk after having had one or more
highly successful projects.
Analysing Problems in Isolation (Continued)

45. Focusing on Similarities
• We tend to give too much attention to
information that is perceived to be similar.
• Two examples of the focus on similarities are
– the representative heuristic, and
– gambler’s fallacy.

46. Focusing on Similarities (Continued)
• The representative heuristic consists of
determining the probability of an event
based on information judged to be
typical or similar to that event
– This is what statistical analysis is about
– Most people tend to draw
inferences from small
samples that are too strong
given the small sample size
– Inferences drawn from large samples are
too weak
– Hence, most people are subject to a
systematic bias which can be predicted by
the sample size.

47. • Gambler’s fallacy consists of mistakenly reading
patterns into random events that are clearly
independent of each other
– The classic example of gambler’s fallacy is that a lot of
lottery players, when choosing their numbers, tend to
avoid those numbers that have
come up in recent draws
– They turn a random event into a conditional event whose
outcome depends on the outcomes from previous events
– Hence, they underestimate the probability of some
numbers
– This phenomenon is also called base-rate underweighting.
Focusing on Similarities (Continued)

48. We Only Slowly Change Our Beliefs
• Under certain circumstances, we tend to be
conservative, i.e. we tend to change or update
our beliefs less frequently than would be
rational.
• This bias is exactly the opposite of the above
base-rate underweighting.
• Base-rate underweighting is about
overreacting to information whereas
conservatism is about underreacting.
• Base-rate underweighting is about the excessive
reliance on the strength of a signal.

49. We Only Slowly Change Our
Beliefs (Continued)
• Conservatism is about underreliance on the
weight of the information.
• Evidence suggests that we are more likely to
suffer from conservatism when
the information is cognitively
costly to analyse.

50. Emotions
• Emotions, such as anger,
resentment and love, are
another reason why humans
do not always act rationally.
• At worst, we may end up
suffering from a loss of self-
control.
• Evidence from experiments
suggests that
– people who are in a good mood
are less critical whereas
– those in a bad mood
spend more time
analysing the information
they are presented with.

51. Emotions (Continued)
• There is some evidence that stock market prices are
affected by seasonal affective disorder (SAD).
• SAD consists of suffering from clinical depression
during the winter months when daylight is limited.
• Some studies suggest that there is a correlation
between the amount of daylight
across the seasons and the stock
returns.
• Stock returns are higher during the winter months in
the Northern hemisphere, possibly compensating for
the increased risk aversion among investors with SAD.

52. Social Interactions
• An individual’s behaviour and decisions are
influenced by the individual’s interactions
with their social group.
• People tend to conform to the beliefs and
behaviours of others.
• This conformity effect also depends on an
individual’s
– culture,
– society, and
– history.

53. Social Interactions (Continued)
• An important means of social interaction is
conversation.
• However, conversation tends to be
a poor means of communication
because
1. The subject of most conversations tends be to
information already known to all participants
2. Conversations are subject to time constraints
3. The complexity of the information conveyed is
also subject to limits.

54. • Hence, information conveyed via conversations
tends to be sharpened and oversimplified.
• As a result, the listener in a conversation will
end shifting their beliefs to extremes.
• Other effects caused by social interactions
include
–the fundamental attribution error,
–the false consensus effect, and
–the curse of knowledge.
Social Interactions (Continued)

55. Social Interactions (Continued)
• The fundamental
attribution error
consists of
underestimating
external factors and
overestimating the
importance of a
person’s mind set
when explaining that
person’s actions
– For example, the non-
executives may think that
the reason the executives
have granted themselves
stock options is to improve
their incentives
– However, the true reason
is that the executives
believe that the firm’s
stock is undervalued and
that the options will be in
the money soon.

56. Social Interactions (Continued)
• The false consensus effect
consists of people having a
tendency to assume that
others share their
opinions more than they
actually do.
• The curse of knowledge
consists of assuming that
others who have less
knowledge have similar
beliefs than they actually
have.
• It is important for
boards of directors as
well as other corporate
committees to be aware
of these effects as they
are likely to bias their
assessment of the
executives’
performance and
intentions.

57. Self-deception
• The two main forms of self-deception are
– overconfidence, and
– confirmatory bias.
• Most of us suffer from overconfidence.
• Overconfidence is the tendency of
overestimating one’s knowledge and
abilities.
• Men tend to be more overconfident than women
– The difference in confidence increases for tasks that
are perceived to be more masculine.

58. Self-deception (Continued)
• Experts tend to be
overconfident when the task is
complex and the information is
highly opaque.
• For overconfidence to persist,
there needs to be biased self-
attribution.
• People tend to attribute
good outcomes to
their own abilities
and bad ones to
external influences.
• Whereas rational learning
would make individuals less
overconfident over time,
biased self-attribution
increases overconfidence over
TEMASEK
LOSSES

59. Self-deception (Continued)
• Richard Roll argues that
managerial overconfidence
or hubris explains why so
many mergers and
acquisitions fail to
realise the expected gains.
• While the majority of
mergers and acquisitions
are unsuccessful,
individual bidding
managers may believe that
their skills are superior to
those of other managers.

60. Self-deception (Continued)
• Confirmatory bias consists of
spending too little time on data
and other
information
that
contradicts
one’s beliefs
and discarding
it as bad luck or data.

61. Self Deception
• In 1984, The Economist
asked:
– 4 ex-finance ministers of OECD
countries,
– 4 chairmen of multinationals,
– 4 Oxford University economics
students and
– 4 garbage collectors
to predict several economic
factors for the OECD ten
years ahead.
• In 1994, the magazine
revisited the predictions and
checked their accuracy.
• On average, the forecasts were more
than 60 percent too high or too low.
• The average forecasted price of oil,
for example, was $40 compared with
an actual price of just $17.
• All the respondents said Singapore’s
GDP per capita would never overtake
Australia’s, but that had actually
happened in 1993.
• The most accurate
forecasters were the
London dustmen and the
chairmen of multinational companies
(a tie for first place); the finance
ministers came in last.
• But the performance of every group
was quite abysmal. The unpalatable
fact is that no one can predict the
long-term economic and market
environment with any real accuracy.

62. Risk Taking According to Gender and Age
• John Coates and Joe Herbert have found preliminary
evidence that human biology, in particular
testosterone, drives risk-taking behaviour.
• Testosterone in stock-exchange traders may have
two contrasting effects
– At first, testosterone has a positive effect on performance
as it makes traders more persistent, more willing to take
on risk and seek novelty
– However, if testosterone continues to increase and stays
at high levels, traders may seek unnecessary risk, i.e. risk
which does not increase expected performance.

63. Testosterone vs Income (Coates 2009)

64. Risk Taking According to Gender and
Age (Continued)
• Testosterone
has been shown
to decline
significantly
with age.
• This research suggests
that board diversity in
terms of gender and age
may be good in terms of
risk management.

65. How to Address Behavioural Issues?
• One way of increasing awareness of these
behavioural issues and reducing their impact is
to improve managers’ education.
• Another way is to ensure that boards of
directors are effective and that they rein in bad
decision making by managers, in particular the
CEO.
• However, there is little evidence that
boards of directors are effective
monitors.

66. How to Address Behavioural Issues?
(Continued)
• Why is this the case?
– One possible answer is that today’s board
composition is not exogenous and is likely to be the
result of past corporate governance issues
– For example, if a firm performed badly in the past
there may have been pressure from shareholders to
appoint more independent directors
– Hence, one firm with good current performance may
have a large proportion of independent directors on
its board due to historic reasons whereas another
firm with equally good current performance may
have a much lower proportion given that its past
performance has always been good.

67. • Randall Morck proposes another possible
reason for the observed lack of a relationship
between corporate performance and board
composition.
How to Address Behavioural
Issues? (Continued)

68. Reflexive Loyalty
• Morck argues that there are two principal–agent
problems and that both can be described by
referring to loyalty.
• The type I agency problem is the classic agency
problem between the managers and the
shareholders.
• It is caused by the managers acting in their own
interest rather than performing their duty.
• It consists of managers being disloyal to their
principal, i.e. the shareholders.

69. Reflexive Loyalty (Continued)
• The type II agency problem is caused by
excessive loyalty to one’s principal (the CEO or
large shareholder) rather than a lack of loyalty.
• Morck argues the recent corporate scandals of
Enron, Hollinger and Worldcom have been
caused by misplaced loyalty of
directors to powerful CEOs.
• This problem is caused by a behavioural bias
which consists of a reflex for loyalty to figures
of authorities.

70. The Milgram Experiment
• http://www.youtube.co
m/watch?v=BcvSNg0HZ
wk&feature=relmfu

71. • During the 1960s, Stanley Milgram, professor of
social psychology at Yale, conducted
experiments with human subjects.
• The subjects were recruited via newspaper ads
and paid for their participation.
• They were told that the experiments were about
the effects of punishment on learning and
memory.
• The subject would act as a “teacher” who would
ask questions to a “learner”.
Reflexive Loyalty (Continued)

72. • The subject would then administer an electric
shock of a certain severity to the learner for every
wrong answer.
• Unknown to the subject, the learner was a
professional actor and the machine administering
the electric shock a fake.
• The electric switches on the machine were
labelled from 15V to 450V and were also labelled
with the intensity of the pain (“slight” through
“very strong” through “danger severe” and
“XXX”).
Reflexive Loyalty (Continued)

73. • Each time the learner would give a wrong answer,
the severity of the electric shock would increase
and the actor feign increasing levels of pain.
• A psychologist would be present at the
experiments.
• More than 60% of the American subjects were
happy to go all the way to 450V (“XXX”).
• Milgram’s experiments were repeated across
other countries and cultures and the results were
similar.
• The results were also similar across genders.
Reflexive Loyalty (Continued)

75. • Milgram concluded that humans have a reflex
for loyalty
– Prehistoric men who were loyal to their tribal leader
were more likely to survive
– Through evolution this resulted in loyalty being hard-
wired into the human brain.
• Milgram followed up with his subjects and asked
them why they had behaved the way they had.
• Many were very upset about the experience, but
justified it with words such as “loyalty”
and “duty”.
Reflexive Loyalty (Continued)

76. • Morck argues that in the boardroom
directors have a tendency to be loyal to
their leader, the CEO, rather than
to ask probing questions and challenge
his/her decisions.
• This may explain why studies that look at the
effects of boards on corporate performance
have found very little evidence of such
effects.
Reflexive Loyalty (Continued)

77. • Milgram performed many variants of his
experiment.
• One consisted of having three “teachers”, two
dissenting teachers plus the actual subject.
• At 150V, the first teacher would walk out.
• The second teacher would walk out at 210V.
• The two “dissenting peers” caused a huge
increase in disobedience.
Reflexive Loyalty (Continued)

79. • Morck argues that boards of directors should
be designed in ways that promote
disloyalty to the CEO (or large
shareholder), i.e. in ways that prevent
type II agency problems.
• Both the US Sarbanes-Oxley Act and the UK
Higgs Report seem to have adopted that
approach by stressing the role of non-
executives on the board of directors as well as
the importance of their independence.
Reflexive Loyalty (Continued)

80. • One way to avoid type II agency problems would
then be to reduce the dominance of the CEO.
• Remember the narrow framing effect we
discussed earlier
– The CEO may influence the board’s decision by the
way he/she presents the information.
• The Higgs Report recommended that the CEO
does not chair board meetings
– However, this also has costs as it may give too much
power to less well informed outsider.
Reflexive Loyalty (Continued)

81. • Hence, there is a trade-off between
keeping the power of the CEO in check and
ensuring the effectiveness of executive
decision making.
Reflexive Loyalty (Continued)

82. How to Reduce Behavioural Biases
• These biases may be difficult to address in
practice.
• Reducing the dominance of the CEO is one way
forward.
• However, this may increase the power of less
well-informed actors.
• Board meetings should be conducted in such a
way that
– They are not dominated by the CEO
– The non-executives are able to scrutinise the policies
proposed by the executives.

83. Conclusions
• Bounded rationality.
• Heuristic simplification.
• Emotions.
• Social interactions.
• Overconfidence and hubris.
• Risk taking according to gender and age.
• Reflexive loyalty.
• Reducing behavioural biases.

85. Casestudy 2 : General Motors
1. Read and prepare the Casestudy on
General Motors (Monks & Minow
(2011)) for discussion next class.
Identify the corporate governance
issues faced.
2. In groups of four members you are
required to:
• Allocate responsibility to – a non executive
director, a CEO, A Health and Safety Manager
and a Risk Manager
• Analyse the scenario’s in the case study and
discuss which are the most critical risks that
the organisation faces.
• Plot the resulting risk analysis on an
appropriate risk map.
• Decide what is your group’s response.

86. Further Reading
• Solomon, Jill (2010) Corporate Governance
and Accountability 3rd Edition, Wiley, UK.
Ch.6
• Goergen, Marc (2012) International
Corporate Governance, Pearson. Ch.15
• Larker & Tayan (2011) Ch.6
• CIMA - Performance Strategy: Study Text
(2012) BPP Learning Media Ltd. Part B : 2

87. Additional Readings (1)
• COSO (2004) Enterprise Risk Management – Integrated
Framework, September 2004, Committee of Sponsoring
Organizations of the Treadway Commission
• IFAC (2006) Internal Controls : A review of current
developments, August 2006, New York : International
Federation of Accountants
• AIRMIC, Alarm & IRM (2010) A structured approach to
Enterprise Risk Management and the requirements of ISO
31000, Institute of Risk Management.
• CIMA (2008) Fraud Risk Management: A Guide to Good
Practice, Chartered Institute of Management Accountants.
• FRC (2004) The Turnbull guidance as an evaluation
framework for the purposes of Section 404(a) of the
Sarbanes-Oxley Act, 16 Dec 2004, FRC

88. NEXT Ideas for Discussion
• Morck, Randall and Yeung, Bernard (2003)
Agency problems in large Family Business
Groups, Entrepreneurship: Theory and
Practice, Summer 2003. Vol. 27, No. 4: pp.
367 – 382

89. QUESTIONS?