Chapter Outline

1
Chapter Outline

• 12.1 Risk Identification and Evaluation
• Identifying Exposures
• Property Loss Exposures
• Liability Losses
• Losses to Human Capital
• Losses from External Economic Forces
• Evaluating the Frequency And Severity of Loss
• Frequency
• Severity
• Expected Loss and Standard Deviation

2
Chapter Outline

• 12.2 Retention and Insurance Revisited
• Benefits of Increased Retention
• Savings on Premium Loadings
• Reducing Exposure to Insurance Market
  Volatility
• Reducing Moral Hazard
• Avoiding High Premiums Caused by Asymmetric
  Information
• Avoiding Implicit Taxes due to Insurance Price
  Regulation
• Maintaining Use of Funds
• Costs of Increased Retention
• Closely-held vs. Publicly-Traded Firms with
  Widely Held Stock
• Firm Size and Correlation Among Losses
• Correlation of Losses with Other Cash Flows
  and with
• Investment Opportunities
• Financial Leverage
• A Basic Guideline for Optimal Retention

3
Chapter Outline

• 12.3 Benefits and Costs of Loss Control
• Basic Cost-Benefit Tradeoff
• Examples of Identifying Benefits and Costs
• Installation of Automatic Sprinkler System
• Installation of Safety Guards
• Child-Resistant Packaging of Non-Prescription
  Drugs
• Qualitative vs. Quantitative Decision-Making
• 12.4 Statistical Analysis in Risk Management
• Approximating Loss Distributions with the
  Normal Distribution
• Illustration
• Problems and Limitation
• Computer Simulation of Loss Distributions
• Illustration
• Comparison of Results to Assuming the Normal
  Distribution
• Limitations of Computer Simulation

4
Chapter Outline

• 12.5 Use of Discounted Cash Flow Analysis
• The Net Present Value Criterion
• Example Forming a Captive Insurer
• The Appropriate Cost of Capital
• 12.6 Summary

5
Identifying Exposures

• Types of Exposures
• Property
• Liability
• Human resource
• External economic factors (e.g., price changes)
• Methods of identification
• Lists
• Understanding business

6
Assessing Loss Exposures

• Ideally, a risk manager would have information
  about the probability distribution of losses
• Then assess how different risk management
  approaches would change the distribution
• Summary measures of probability distributions
• frequency
• severity
• expected loss
• standard deviation

7
Calculating the Frequency and Severity of Loss

• Example
• 10,000 employees in each of the past five years
• 1,500 injuries over the five-year period
• 3 million in total injury costs
• Frequency of injury per year 1.500 / 50,000
  0.03
• Average severity of injury 3 m/ 1,500
  2,000
• Annual expected loss per employee 0.03 x 2,000
  60
• Ideally, also calculate the standard deviation of
  loss

8
Benefits of Increased Retention

• Savings on insurance premium loadings
• Administrative costs
• Reducing moral hazard
• Avoid being pooled with higher risk policyholders
• Avoid implicit taxes from insurance regulation
• Reducing exposure to insurance market volatility
• Allows firm to maintain use of funds
• Questionable

9
Costs of Increased Retention

• Increased probability of financial distress
• Bankruptcy is costly
• Possibility of distress affects contractual terms
  with other claimants
• Increased probability of raising external capital
• Forego tax benefits
• Forego efficiencies in bundling services

10
Factors Affecting Costs of Increased Retention

• Ownership structure (closely-held vs. widely-held
  firms)
• Firm size
• Correlation among losses
• Correlation of losses with other cash flows
• Correlation of losses with investment
  opportunities
• Financial leverage

11
Basic Guideline for Optimal Retention

• Retain reasonably predictable losses and insure
  potentially large disruptive losses
• Not always right (BP case)
• But often is

12
British Petroleum Case

• British Petroleum
• Perspective
• risk management strategy
• 1990
• Basic Businesses
• Exploration
• Oil Refining, Distribution, and Retailing
• Chemicals (small)
• Nutrition (small)

13
British Petroleum Case

• Financial Data
• Capital Structure
• Equity 35 billion
• Debt 15 billion
• After-tax profit
• average 1.9 billion
• standard deviation 1.1 billion
• Assets
• Diversified 13,000 service stations in 50
  countries
• Undiversified oil production

14
British Petroleum Case

• Loss Exposures (in million)
• 
  Expected
• Range Number Average
  Annual Standard
• million per year Severity
  Loss Deviation
• 0 - 10 1845 0.03
  52 12
• (vehicle accidents, injuries, small fires,
  equipment failures)
• 10 - 500 1.7 40.0
  70 98
• (refinery fires, explosions, minor oil spills)
• 500 0.03 1000
  35 233
• (major oil spills, tort claims from release of
  chemicals, major loss of life, defective fuel
  causing airplane disaster)

15
British Petroleum Case

• Previous Strategy
• Range Approach
• 0 - 10 centralized insurance
• purchases and
• self insurance
• 10 - 500 externally insured
• 500 self-insured

16
British Petroleum Case

• Conclusions for First Range of Exposures
• Decentralize insurance decisions
• More insurance (Why?)
• local insurers are more efficient in
• loss control
• claims processing
• insurance markets are competitive
• insurer insolvency not a concern

17
British Petroleum Case

• Conclusions for Second Range of Exposures
• Self-Insure (Why?)
• impact of losses on equity and income is small
• little competiiton in insurance market
• insurer insolvency a concern
• contract enforcement is costly
• BP has comparative advantage in loss control

18
British Petroleum Case

• Conclusions for Third Range of Exposures
• Continue to Self-Insure
• insurance is not availability (not credible)

19
Making Loss Control Decisions

• Ideally, calculate the present value of the
  benefits and costs of loss control
• Primary benefit reduction in expected loss
• Primary costs cost of loss control device
• Other harder to quantify effects of loss control
• Lost productivity
• Improved contractual terms with employees, etc.
• Quantitative versus qualitative decision making

20
Statistical Analysis in Risk Management

• Two main approaches
• Approximate losses using normal distribution
• Computer simulation of loss distributions
• Maximum probable loss
• if 5 million is the maximum probable loss at the
  95 percent level, then the firms losses will be
  less than 5 million with probability 0.95.
• Same concept as Value at risk

21
When to Use the Normal Distribution

• Most loss distributions are not normal
• From the central limit theorem, using the normal
  distribution will nevertheless be appropriate
  when
• Number of exposures is large
• Losses across exposures are independent
• Example where it might be appropriate
• worker injury losses for firms with a large
  number of employees
• automobile accident losses for firms with large
  fleets of cars

22
Using the Normal Distribution

• Important property
• If Losses are normally distributed with
• mean m
• standard deviation s
• Then
• Probability (Loss lt m 1.645 s) 0.95
• Probability (Loss lt m 2.33 s) 0.99

23
Using the Normal Distribution - An Example

• Worker compensation losses for Stallone Steel
• sample mean loss per worker 300
• sample standard deviation per worker 20,000
• number of workers 10,000
• Assume total losses are normally distributed with
• mean 3 million
• standard deviation 100 x 20,000 2million
• Then maximum probable loss at the 95 percent
  level is
• 3 million 1.645 (2 million) 6.3 million

24
A Limitation of the Normal Distribution

• Applies only to aggregate losses, not individual
  losses
• Thus, it cannot be used to analyze decisions
  about per occurrence deductibles and limits

25
Monte Carlo Simulation

• Overcomes some of the shortcomings of the normal
  distribution approach
• Overview
• Make assumptions about distributions for
  frequency and severity of individual losses
• Randomly draw from each distribution and
  calculate the firms total losses under
  alternative risk management strategies
• Redo step two many times to obtain a distribution
  for total losses under each of the alternative
  strategies
• Compare strategies (distributions)

26
Simulation Example - Assumptions

• Claim frequency follows a Poisson distribution
• Important property Poisson distribution gives
  the probability of 0 claims, 1 claim, 2 claims,
  etc.
• Expected value of distribution depends on other
  uncertain events
• Expected value equals
• 20 with probability 1/3
• 30 with probability 1/3
• 40 with probability 1/3

27
Simulation Example - Assumptions

• Claim severity follows a Lognormal distribution
  with
• expected value 100,000
• standard deviation 300,00
• note skewness

28
Simulation Example - Assumptions
29
Simulation Example - Alternative Strategies

• Policy 1 2 3
• Per Occurrence Deductible 500,000 1,000,000 no
  ne
• Per Occurrence Policy Limit 5,000,000 5,000,000
  none
• Aggregate Deductible none none 6,000,000
• Aggregate Policy Limit none none 10,000,000
• Premium 780,000 415,000 165,000

30
Simulation Example - Results
31
Simulation Example - Results

• Statistic Policy 1 Policy 2 Policy
  3 No insurance
• Mean value of retained losses 2,414 2,716 2,92
  5 3,042
• Standard deviation of retained losses 1,065 1,293
  1,494 1,839
• Maximum probable retained loss at 95
  level 4,254 5,003 6,000 6,462
• Maximum value of retained losses 11,325 12,125 7,
  899 18,898
• Probability that losses exceed policy
  limits 1.1 0.7 0.1 n.a.
• Probability that retained losses ? 6
  million 99.7 98.7 99.9 92.7
• Premium 780 415 165 0
• Mean total cost 3,194 3,131 3,090 3,042
• Maximum probable total cost at 95
  level 5,034 5,418 6,165 6,462

32
Discounted Cash Flow (DCF) Analysis

• When risk management decisions affect cash flows
  over multiple periods, the effect on value should
  be calculated using DCF analysis
• Calculate the Net Present Value (NPV) of the
  alternative decisions
• NPV
• where
• NCFt net cash flow in year t
• r opportunity cost of capital (reflects the
  risk of the cash flows)