Goldfarb - risk sources

Question 1

5 main categories of risk to which insurer is exposed

Answer 1

Market

Credit

Insurance UW

Operational

Strategic

Question 2

Market Risk

Answer 2

risk to firm’s current investments from changes in market variables (equity indices, interest rates, foreign exchange, etc)

Question 3

Credit Risk

Answer 3

-risk of loss due to credit events

(counterparty default, changes in counterparty rating etc)

• market securities/derivatives/swap positions
• insured’s contingent premiums & deductibles receivable
• reinsurance recoveries

Question 4

market securities/derivatives/swap positions credit risk

Answer 4

firm’s investments in these securities may lose value from credit events

Question 5

insured’s contingent premiums & deductibles receivable credit risk

Answer 5

insurer is exposed to risk that insured may not pay balance that it owes

Question 6

reinsurance recoveries: most difficult to estimate due to 3 unique aspects:

Answer 6

1. definition of default: this should be modified to reflect the fact that if reinsurer’s credit rating gets downgraded below an investment grade level, it could enter a death spiral where its policyholders try to end their contracts resulting in severe liquidity problems; as result, reinsurer may settle claims for less than 100% of full amount
2. substantial contingent exposure: reinsurance recoverable at a given point in time may increase in future due to adverse loss development, causing credit risk exposure to increase as well
3. reinsurance credit risk is highly correlated with underlying insurance risk

Question 7

Insurance Underwriting Risk

3 categories

Answer 7

1. loss reserves on prior policy years: potential adverse development
2. UW risk for current policy year
3. property catastrophe risk

Question 8

operational & strategic risk & issue

Answer 8

Operational risks: failure of people/systems/processes

Strategic risks: related to competitors

-hard to quantify these

Question 9

UW risk

Answer 9

-risk that losses and expenses from the NB including renewals written and/or earned during risk horizon period exceeds the premiums collected

Question 10

reinsurance recoverables credit risk also includes

Answer 10

partial default (settling at less than full recovery due to disputes with reinsurer)

Question 11

methods to quantify UW eisk

Answer 11

-several methods can be used to quantify this risk

Loss ratio distribution model

Frequency and severity model

Inference from reserve risk models

Question 12

Property-Catastrophe Risk - why historical experience is not best indication of future losses

Answer 12

Events are rare

Exposures change over time

Severities change over time due to changes in building materials and designs

-catastrophe models are therefore used

Question 13

catastrophe models often have several modules

Answer 13

1. stochastic/hazard module: generate the events that can occur including location, intensity, etc
2. damage (vulnerability) module: derives the damage that would arise from an event based on exposure info
3. financial analysis module: applies the insurance/reinsurance terms to losses to determine the financial impact to the insurer

Question 14

once insurer has derived the individual distributions of risks

Answer 14

it will then need to aggregate them

-in order to do this, dependency between risks need to be considered

Question 15

dependency is different to correlation

Answer 15

correlation looks at linear relationship

Question 16

3 methods to quantify the dependency:

Answer 16

1. empirical analysis of historical data
2. subjective estimates
3. explicit factor models

Question 17

empirical analysis of historical data disadvantage

Answer 17

usually there is insufficient data to calculate the historical dependency, there is little insight as to how the dependencies will change during tail events

Question 18

subjective estimates adv & disadv

Answer 18

advantage: can account for tail events, reflects the user’s intuition
disadvantage: as number of risk categories increases, number of dependency parameters that need to be estimated increases exponentially

Question 19

explicit factor models

Answer 19

these link the variability of risks to common factors

Question 20

once dependency has been estimated, insurer can aggregate the distributions using one of several available techniques

Answer 20

1. closed form solutions
2. approximation methods
3. simulations methods
4. square root rule