INTRODUCTION TO RISK MANAGEMENT

1
INTRODUCTIONTORISK MANAGEMENT
2
WHAT IS RISK?
3
DEFINITIONS I

• Arabic - Fortuitous and favorable.
• Greek - Fortuitous and neither favorable nor
  unfavorable.
• Latin (risicum) - the challenge that a barrier
  reef presents to a sailor.
• French (risque) - mainly negative connotation,
  but sometimes positive.
• Oxford Dictionary - ... the chance of hazard,
  bad consequences, loss, etc....

4
DEFINITIONS II

• Economic risk - the chance of loss due to .
• Business risk - the chance of loss
  associated with
• Market risk - the chance that a portfolio
  of investments can lose
  money because..
• Inflation risk - the danger that a general
  increase in prices ...
• Interest-rate risk - market risk due to interest
  rate fluctuations
• Credit risk - the chance that of a
  default on a loan ...
• Liquidity risk - the difficulty in selling
  a fixed asset ...
• Derivative risk - the chance of financial loss
  due to increased volatility .
• Cultural risk - the chance of loss because
  of product market ..

CHANCE Random Occurrence
BAD CONSEQUENCE Sense of Loss
Hirschey Pappas, Fundamentals of Managerial
Economics Dryden Press, 1998
5
A LITTLE BIT OF PROBABILITY
6
PROBABILITY

• Its a number its JUST A NUMBER!
• Its a number between 0 and 1 ( 0 P 1)
• It quantifies the likelihood of an event
• Its a function of experience, judgment,
  subjective assessment, available data
• Its uses all information you think is relevant
  to the determination of the likelihood of
  occurrence of an event

7
Probability Rules

• Probability 0 if never/impossible
• Probability 1 if always/certain
• If we are collectively exhaustive and
• mutually exclusive then
• the probabilities over the outocmes
• SUM to 1.

8
Probability Rules

• If mutually exclusive then
• P(A or B) P(A) P(B)
• If independent
• P(A and B) P(A) x P(B)

9
Probability from Data

• Given data we can always derive approximate
  probabilities using relative frequency.
• Relative frequency can be used as an estimate of
  the probability of the observed value
• Taken all together, these can represent the
  underlying PROBABILITY DISTRUBUTION FUNCTION

10
Earthquakes
11
(No Transcript)
12
Frequency Table
How Big?
How Many?
13
Relative Frequency Table
How Big?
How Many?
14
Relative Frequency Histogram
0.50
0.474
0.45
0.40
0.35
0.30
0.240
0.25
0.20
0.158
0.15
0.10
0.065
0.05
0.038
0.020
0.004
0.001
0.00
1.5 - 2.5)
2.5 - 3.5)
3.5 - 4.5)
4.5 - 5.5)
5.5 - 6.5)
6.5 - 7.5)
7.5 - 8.5)
8.5 - 9.5)
15
Prob. of an event proportion of observations
that correspond to the event
percent of observations that correspond
to the event
area of histogram that corresponds to the
event
16
(No Transcript)
17
AN INVESTMENT DECISION

• Planning for retirement
• Two options for investment
• Each has a track record, the historical
  rates-of-return over a specified time period
• Each can be used to compute various statistics
  e.g., average rate-of-return, etc.

18
AN INVESTMENT DECISION
Expected Value
Std. Dev.
Variance
A1
5.00
1.25
1.5625
A2
5.70
2.75
7.5625
19
(No Transcript)
20
Whats the likelihood of ?
r lt 0
I dont want a rate of return lt 0!
I want a rate of return gt 0!
Whats the likelihood of ?
r lt 0
21
THE FREQUENCY HISTOGRAM( The KEY to it ALL )

• The relative frequency histogram over the
  outcomes contains all relevant information.
• This information allows us to quantify risk.
• This is provides our most powerful tool for risk
  management.

22
A QUANTITATIVE DEFINITION OF RISK
23
DEFINITIONS III

• What can go wrong?
• How likely is it to go wrong?
• If it goes wrong, what are the consequences?

FUTURE SCENARIOS
PROBABILITIES
PREFERENCES
Adapted from S. Kaplan and B. John Garrick, On
the Quantitative Definition of Risk, Risk
Management, Vol.1, no.1, 1981
24
RISK is a combination of the answers to the three
questions..
FUTURE (bad)
POBABILITY of (bad) F
HOW YOU FEEL about the results due to the (bad)
F
25
WHAT CAN GO WRONG?(Specifying the Future
Scenarios)
1. Housing Market Collapse, Financial Market
Fails,.
2. Major Damaging Earthquake, Tsunami, .
3. Terrorist Attack with WMD, .
4. Nuclear deterrence fails!
26
HOW LIKELY IS IT TO GO WRONG?(Specifying the
Probabilities)
1. Probability of financial market failure,
likelihood of house prices declining, .
2. Probability of damaging earthquake,
probability of submarine fault, ..
3. Likelihood of terroroist attack with dirty
bomb, with biological agent, with chemical
agent, ..
4. Probability of nuclear war
27
WHAT ARE THE OUTCOMES?
1. Real Rate of Return 0,
2. Economic loss 100M 200M 300M .
3. Fatalities 10000 20000 1000000 .
4. Too horrible to contemplate!
28
WHAT ARE THE CONSEQUENCES?
1. How do you feel about real rate of return
0, ?
2. How do you feel about economic loss 100M,
... ?
3. How do you feel about fatalities 100000 ?
4. Too horrible to contemplate!
29
(No Transcript)
30
(No Transcript)
31
HOW YOU FEEL about the results
This is plain talk for what the analyst calls
your PREFERENCES
How YOU FEEL ? What YOU PREFER
Your PREFERENCES (over the outcomes) matter
32
Y Outcome g(F)
Consequences preferences over the outcomes
33
WHAT ARE THE CONSEQUENCES?(Subjective Valuation
of the Outcomes)
v(Y)
()
Outcome
BAD
GOOD
WORSE
BETTER
WOW
(-)
Consequences
34
WHAT ARE THE CONSEQUENCES?(Subjective Valuation
of the Outcomes)
Reference point identifies the distinction
between gains and losses on the outcome axis
v(Y)
Y
Losses (Prefer to Avoid)
Gains (Prefer to Experience)
35
CONSEQUENCES SPECIAL CASE
v(Y)
Y
PREFER TO EXPERIENCE (All Equally)
PREFER TO AVOID ( All Equally)
Consequences
36
CONSEQUENCES SPECIAL CASE
Fatalities 1000
Y
Economic Loss M
Consequences
PREFER TO EXPERIENCE (All Equally)
PREFER TO AVOID ( All Equally)
37
QUANTITATIVE DEFINITION
What can go wrong? ? range of
outcomes
How likely is it? ?
probabilities over the
range of outcomes
What are the consequences? ? the least
preferable
outcomes
38
QUANTITATIVE DEFINITION

• the probability of least preferred outcome.
• P least preferred outcome
• Pbenefit is too little - in terms of gain
• Pcost is too much - in terms of loss

PB lt Bmin
PC gt Cmax
39
Who uses this stuff?.......
OVERALL C-RATING System for Readiness C-1 MAE
gt 89 Pnot capable 0.11 C-2
MAE 80-89 0.11 Pnot capable 0.20 C-3
MAE 70-79 0.21 Pnot capable 0.30 C-4
MAE 50-69 0.31 Pnot capable 0.50 C-5
MAE lt 50 0.50 Pnot capable
Senator Strom Thurmond, Chairman of Senate Armed
Services Committee, terminology used in arguments
before the committee, Feb. 1997
40
A QUANTITATIVE APPROACH TO RISK MANAGEMENT
41
RISK MANAGEMENT PROCESS
Identify Risks
Assess Risks
What can go wrong?
What is the likelihood? What are the
consequences?
Prevent Mitigate Negotiate
Implement Monitor
Management Action
42
The History of Risk Management

• 1950 B.C. Code of Hamurabi formalization of
  bottomry contracts containing a risk premium for
  chance of loss of ships and cargo.
• 750 B.C. Greece the use of bottomry
  contracts.
• 1285 A.D. King Edward - forbids use of soft
  coal in kilns to manage air pollution in
  London.
• 1583 A.D. 1st life insurance policy issued in
  England.
• 19th and 20th century water and garbage
  sanitation, building codes, fire codes, boiler
  inspections, railroads, steamboats, autos.
• 1959 A.D. H. Markowitz, stock portfolio
  diversification.

43
Tools of Risk Management

• Prevention.
• Mitigation.
• Hedging.
• Diversification.

44
Determining the Outcome Distribution

• theoretical derivation
• direct assessment
• simulation

Generating your own data
45
EARTHQUAKES
46
P Total Cost x
1.00
0.652
0.11
0.012
0.002
0.00
47
A1 Do Nothing
A2 New Building Codes
A3 Retro-Fit New Codes
48
P Total Cost x
The RISK CURVE for A1
49
A1 (red)
A2 (blue)
A3 (green)
50
(No Transcript)
51
ACCEPTABLE RISK
The perennial question free people ask with
regard to defense is How much is enough? To
this there can be no precise answer. A countrys
security is a function of the DEGREE OF RISK
A COUNTRY IS WILLING TO ACCEPT.
Hitch McKean, The Economics of Defense in the
Nuclear Age, Atheneum, 1986
52
ACCEPTABLE RISK
Risk
Too Risky
Acceptable Risk
Proposed Budget
Required Budget
Cost
53
Who uses this stuff?.......
Ultimately, policy makers must decide how much
the United States is willing to pay to lower the
risks associated with de-ploying forces abroad.
But some might argue that defense planners
occasionally focus on absolute requirements the
minimum number of forces that they believe will
meet DoDs military needs without fully
weighing the relative risks and costs of
alternative levels.
Moving U.S. Forces Options for Strategic
Mobility Congressional Budget Office, Feb. 1997
54
Who uses this stuff?.......
Our armed forces remain capable, within an
acceptable level of risk, of meeting the demands
of our strategy.
Maj. Gen. John J. Maher, Vice Director for
Operations, Joint Staff testimony before House
National Security readiness subcommittee, Feb.
1997
55
Who uses this stuff?.......
Computer security is basically risk management.
. Managers have to decide what they are trying
to protect and how much they are willing to
spend, both in cost and convenience, to defend
it.
Stephen H. Wildstrom, review of the book Secrets
and Lies by Bruce Schneier, Businessweek Sept.
2000
56
Who uses this stuff?.......
we continue to believe the federal government
can benefit from risk management.
. An effective risk management approach
includes a threat assessment, a vulnerability
assessment and a criticality assessment ...
Raymond J. Decker, Director, Defense Capabilities
and Management, GAO, Testimony before the Senate
Committee on Governmental Affairs Oct. 2001
57
RISK MANAGEMENT
Risk
old
Cost
58
RISK MANAGEMENT
Risk
Proposed Budget
Cost
59
RISK MANAGEMENT
Risk
Acceptable Risk
Cost
60
APPLICATION I ENTERPRISE BUSINESS RISK
61
P .02
Error
P(uncorrected) 0.1
Reconciliation Check
P(corrected) 0.9
P .18
P(error) 0.2
Correct
Data Entry
P(correct) 0.8
P .80
Correct
SECNAV M-5200.35 March 2007
62
DIR.
Define Needs, prepare Purchase Requisition Form
Forward to ASA
End User
INDIR.
Forward to SPFA
NO
YES
ASA reviews PR, confirms funds, obtains approval
Admin. Support
YES
SPFA assigns PR number and form to PA
Sponsored Program
SPFA Reviews PR
NO
Purchase Agent
63
Clarify requirements with end user
End User
ASA /OA will Assign req., number and task
Purchaser
Admin. Support
Sponsored Program
NO
Screen request for mandatory sources of supply,
prohibited or special items, and authority to buy
YES
Purchaser reviews for completeness of
documentation
Purchase Agent
64
Receive ordered items and sign acknowledging
STOP
End User
Admin. Support
Sponsored Program
YES
Place order with source, direct delivery point,
and provide estimated delivery date
NO
Receive order (if delivery point)
Reconcile with vendor
Purchase Agent
65
P .3122
NO ERROR
0.9
ERROR
P .1
0.1
0.9
0.1
P .0729 .0729
ERROR
0.9
0.9
0.1
P .081 .081
Reimburse Or Direct Funds
ERROR
0.9
0.1
P .09 .09
ERROR
P .6878
P .1
0.1
ERROR
66
P .34865
NO ERROR
0.9
ERROR
P .1
0.1
0.9
0.1
P .0729.07695
ERROR
0.9
0.9
0.1
P .081.0855
Reimburse Or Direct Funds
ERROR
0.95
0.1
P .09.095
ERROR
P .65135
0.05
P .05
ERROR
67
P .3861
NO ERROR
0.9
ERROR
P .05
0.05
0.9
0.1
P .07695.07695
ERROR
0.9
0.95
0.1
P .0855.0855
Reimburse Or Direct Funds
ERROR
0.95
0.1
P .095.095
ERROR
P .6139
0.05
P .05
ERROR
68
APPLICATION II COST RISK ASSESSMENT
69
ESTIMATING SHIPBOARD HELICOPTER OM COSTS
Life-cycle cost estimates for the helicopter are
needed. The cost analysis staff is organized
into four groups, one each for the four main
components of the life-cycle cost (1) RD (2)
Procurement (3) Operations and Maintenance and
(4) Salvage/Residual. As leader of the OM cost
estimating group you have decided to use a factor
cost estimates since

• Relevant OM cost data produce reliable CERs for
  the three components of the OM cost POL, Parts,
  and Other as functions of the procurement
  cost.
• The helicopter is an off the shelf item so
  procurement cost is known 3.7 million.

70
POL 112.84 30.16 ACQ error
Parts -84.96 57.21 ACQ error
Other 45.21 10.12 ACQ error
71
PROBABILISTIC COST ESTIMATING
y a b x e
72
PROBABILISTIC COST ESTIMATING
y a b x e
73
PROBABILISTIC COST ESTIMATING
?
y a b x e
74
APPLICATION III PROJECT MANAGEMENT
75
FACILITIES FOR THE FLIGHT SIMULATOR
DEPARTMENT(ACTIVITIES, TIME ESTIMATES, AND
DEPENDENCIES)
76
APPLICATION V HOMELAND SECURITY
77
Homeland Security
Threat Assessment evaluate the likelihood of a
terrorist attack.
Vulnerability Assessment systematically
identify weaknesses in physical structures,
personnel protection, or other areas that may be
exploited.
Criticality Assessment systematically identify
and evaluate important assets in terms of
loss-of-life, damage to economy, damage to
infrastructure, propaganda value, etc.
A Risk Management Approach Can Guide
Preparedness Efforts GAO-02-208T
78
Homeland Security
Reduce the risk and mitigate the consequences of
an attack
PA
PD
79
Homeland Security
Threat Assessment
P(A)
Vulnerability Assessment
P(SA)
Criticality Assessment
D(A,S) ? V(D)
80
Homeland Security
Threat (Measure) probability a specific target
is attacked in a specific way in a specific time
interval.
Vulnerability (Measure) probability that damage
(deaths, property, etc.) occurs, given specific
attack type, at a specific time, at a given
target.
Consequences (Measure) the expected magnitude
of damage (deaths, injuries, property damage,
etc.), given a specific type of attack, at a
specific time, that results in damage to a
specific target.
Estimating Terrorism Risk by H. H. Willis
RAND-MG-388, 2005
81
Homeland Security
Threat Measure
P(A)
Vulnerability Measure
P(SA)
Consequence Measure
E D(A,S)
82
SUMMARY

• RISK is a factor in every decision with
  significant uncertainty
• RISK is a function of the answers to 3 questions
• What can go wrong?
• How likely is it to go wrong?
• If it does go wrong, what are the consequences?

83
SUMMARY

• RISK is quantified using PROBABILITY.
• Use it to express the riskiness an alternative.
• Use it to find the least risky alternative.
• THINK about the RISK vs COST tradeoff curve.

84
SUMMARY

• MANAGING RISK requires the information provided
  by the tradeoff curve!
• THINK about where you want to be on the curve.
• THINK about changing the tradeoff curve!
• USE THE MODEL to help find how to change things!