Marshall Flashcards
Explain why both quantitative and qualitative analysis necessary to properly assess risk margins.
Quantitative analysis can only reflect uncertainty in historical experience and cannot adequately reflect all possible sources of future uncertainty.
Judgment is necessary to estimate future uncertainty.
Summarize the 9 steps of Risk Margin analysis framework
- Portfolio Preparation
Determine valuation portfolios, groups - Independent Risk Analysis
Quantitative analysis, benchmarking - Internal Systematic Risk
Balanced scorecard, map scores to CoVs - External Systemic Risk
Potential future external sources of risk - Correlation Effects
Correlations between classes and liability - Consolidation of Analysis
Consolidate COVs and correlations - Additional Analysis
Sensitivity testing, scenario testing, benchmarking, hindsight analysis - Documentation
Document analysis & judgment - Review
Review assumptions annually, full analysis every 3 years
Explain the framework for assessing risk margin
Start from Claims Portfolio.
Split in Valuation classes (ex: Home vs Auto)
Split valuation classes into homogenous claims groups (ex: Liability vs Property)
HCG include both systemic risk and independent risk.
Systemic risk includes external (labor costs, legislation risk) and internal (specification, data error) risks.
State 2 considerations for splitting claims portfolio into valuation classes
- Preferable to split same way used for developing central estimates.
This allows analyzed sources of uncertainty to be aligned with central estimate analysis. - If the valuation of the central estimate is at granular level, it may make sense to do the quantitative analysis on aggregated valuation classes (more credible) and allocate results down.
2 considerations when allocating valuation classes to claims groups
- If different groups of claims within a valuation class have materially different uncertainty, they should be treated separately in the risk margin analysis.
- balance benefit gained and practicality/cost
Identify the 2 different sources of uncertainty
- Systemic risk
- Independent risk
Define Systemic Risk
Risks that are potentially common across valuation classes or claim groups.
Define Independent Risk
Risks arising from randomness inherent in the insurance process.
Identify the 2 sources of Systemic Risk
- Internal Systemic Risk
- External Systemic Risk
Define Internal Systemic Risk
Risks internal to the liability valuation process.
Reflects the extent to which the actuarial valuation approach is an imperfect representation of a complex, real-life process.
Also referred as model specification risk.
Examples:
Model structure
Model parameterization
Data accuracy
Define External Systemic Risk
Risks external to the actuarial modelling process.
Even if the model represents current conditions well, future systemic trends may cause future experience to differ from current expectations.
Example:
Future trends in claim cost outcomes (change inflation) that may cause actual experience to differ from what is expected based on current environment and trends.
Name the 2 sources of Independent Risk
- Parameter risk
- Process risk
Define Parameter Risk
Randomness of the insurance process compromises ability to select appropriate parameters for valuation models.
Define Process Risk
Pure effect of randomness of the insurance process.
Identify when quantitative modelling is best able to assess risk and when it needs to be supplemented.
- Quantitative modeling is best for analyzing independent risk and past episodes of external systemic risk.
- Quantitative modeling must be supplemented with other qualitative or quantitative analysis to incorporate internal systemic risk and external systemic risk (future external systemic risk may differ from past episodes)
3 main sources of internal systemic risk
- Specification Error
- Parameter Selection Error
- Data Error
Define Specification Error
Error arising from an inability to build a model that fully represents the underlying insurance process.
Define Parameter Selection Error
Error that model cannot adequately measure all predictors of claim cost outcomes or trends in predictors. There may be more cost drivers than can be captured in valuation model.
Define Data Error
Error from poor data or unavailability of data required for a credible valuation model.
Name 3 categories of External Systemic Risk categories
- Economic and social risks
- Legislative political risks and claims inflation risks
- Claim management process change risk
- Expense risk
- Event risk
- Latent claim risk
- Recovery risk
Explain why quantitative methods might not be appropriate for assessing correlation effects
- Techniques tend to be complex and require substantial data (time/effort required may outweigh benefits)
- Correlations would be heavily influenced by past correlations
- Difficult to separate past correlations effects between independent risk and systemic risk or identify the effects of past systemic risks.
- Internal systemic risk cannot be modelled with standard correlation risk techniques.
- Results unlikely to be aligned with framework, which splits between independent internal systemic and external systemic risk.
Describe the correlation effects on independent risks
Assumed to be uncorrelated with any other source of uncertainty either within or between valuation classes.
Describe correlation effects on internal systemic risks
Assumed to be uncorrelated with independent risk and external systemic risk sources.
There is correlation between classes and between outstanding claim and premium liabilities due to:
1. Same actuary effect
2. Linkage between premium liability methodology and outcomes from outstanding claims valuation
Describe correlation effects on external systemic risks
Uncertainty from each risk category is assumed to be uncorrelated with independent risk, internal systemic risk and uncertainty from other external systemic risk categories.
There is correlation between classes and between outstanding claim and premium liabilities from similar risk categories (claims inflation risk across all long-tail portfolios)
Calculate Independent Risk CoV
CoVind ^2 = sumproduct(CoVi^2*weight^2)
Calculate Internal Systemic Risk CoV
CoVint ^2 = sum(rhowiwjCoViCoVj)
Calculate External Systemic Risk CoV
CoVext ^2 = sum(external risks CoVi^2)
Calculate Total consolidated CoV
CoVtot = sqrt(CoVind^2 + CoVint^2 + CoVext^2)
Calculate Risk Margin using Normal Distribution
Risk Margin % = CoVtot * z
Calculate Risk Margin using Lognormal Distribution
s^2 = ln(1+CoVtot^2)
Risk Margin % = exp(z*s - s^2 / 2) - 1
Describe the goal and steps of sensitivity testing
Goal is to gain insight about sensitivity of the final risk margin to key assumptions
Steps:
1. Independently, vary key assumptions (CoVs and correlations) and monitor the impact on the risk margin
2. Review key assumptions that have significant impact on the risk margin
Describe the goal of scenario testing and how it’s done.
Goal is to identify scenarios that would result in the central estimate increasing to the current central estimate plus the risk margin.
This is done by adjusting assumptions used for central estimate.
Describe the goal of internal benchmarking
Check for internal consistency and reasonableness by comparing CoVs:
1. Between similar valuation classes
2. Between outstanding claim and premium liability within classes
For independent risk, internal benchmarking should be done for which 2 main dimensions?
- Portfolio size: larger portfolios have lower CoVs due to lower volatility from random effects.
- Length of claim runoff: longer tailed lines have higher CoVs due to more time for random effects to have an impact.
Explain the 2 implications of internal benchmarking for independent risk.
- LT portfolios: PL CoV should be higher than OSC CoV since PL for a LT line is smaller than OSC for a LT line.
- ST portfolios: PL CoV should be lower than OSC CoV since PL for a ST is larger than OSC for a ST line.
Explain the 2 implications of internal benchmarking for internal systemic risk
- If template models are used for similar valuation classes, we would expect similar CoVs
- If similar valuation methodology is used on both short and long-tail classes, then we would expect a higher CoV for the long-tail class.
Explain the implication of internal benchmarking for external systemic risk
CoVs should be higher for long-tail portfolios, except for event risk for Property and liability risk for Home
Explain the goal of external benchmarking
Compare the reasonableness of CoVs and risk margins to those from external sources.
Should not be relied on instead of analysis, but may be useful when there is little information available, especially for independent risk.
Describe the goal of hindsight analysis
The goal is compare the past reserve estimates of liabilities against the latest view of the same liabilities to review the actual volatility in the past.
Name 3 considerations of hindsight analysis
- Valuation models may have changed from previous valuations
- May be different sources of external systemic risk in the future
- Less valuable for long-tail portfolios with serial correlations between consecutive valuations
Describe the 3 mechanical hindsight analysis steps
- Apply the chain ladder method to data as of the valuation date to calculate a current unpaid estimate.
- Iteratively, remove a diagonal of data and apply the same method to calculate unpaid estimates for prior valuation dates.
- Compare the past estimates to the current estimate for the same accident years. The relevant payments made between the past valuation dates and current valuation date should be added to the current unpaid estimate.
Describe the goal of independent risk assessment for outstanding claims liabilities.
Model is used to fit away past systemic episodes in order to analyze the residual volatility and get the CoVs for independent risk.
Methods include:
Mack method
Bootstrapping
Stochastic Chain Ladder
GLM techniques
Bayesian techniques
Describe the goal and the steps of the Balances Scorecard Approach
Assess the adequacy of modeling infrastructure and its ability to reflect and predict underlying insurance process.
Steps:
1. Conduct a qualitative assessment of the modeling infrastructure for each specification/parameter/data risk component (score 1-5)
2. Calculate a weighted-average score using weights for each risk indicator that reflect relative importance.
3. Calibrate wighted-average scores to a CoV scale for internal systemic risk.
Name 3 potential risk indicators for Specification Error
- Number of independent models used
- Range of results produced by the models
- Checks made on reasonableness of results
Name 3 potential risk indicators for Selection Error
- Best predictors have been identified
- Best predictors are stable over time (or change due to process changes)
- Value of the predictors used (close to best predictors)
Name 3 potential risk indicators for Data Error
- Good knowledge of past processes affecting predictors
- Extent, timeliness, consistency and reliability of information from business
- There are appropriate reconciliations and quality control for the data
Explain how to develop a scale to calibrate balanced scorecard scores to CoVs
Scale should represent uncertainty of internal systemic risk ranging from worst to best practice.
Relative difference in performance of modeling infrastructures over time can give insight into uncertainty of poor vs good modeling.
Min CoV for best practice is unlikely to be smaller than 5% (best models cannot completely model insurance process)
Might have CoV greater than 20% for single, aggregated model with limited data.
True or False?
Scale to calibrate balanced scorecard scores to CoVs is linear.
False
Marginal improvements to outcomes are much less from a fair to good model than from poor to fair models.
True or False?
CoVs for LT lines are higher than for ST lines.
True
LT lines are more difficult to model and have less stable predictors.
True or False?
We should use the same scale for OSC and Premium Liability.
True
Define Economic and Social Risk (External Systemic Risk)
Uncertainty associated with inflation, social trends, etc.
Includes:
1. Standard inflation
2. General economic conditions (unemployment, GDP growth, interest rates, asset returns)
3. Fuel prices
4. Changing driving patterns
5. Systemic shifts in claim frequency for ST lines
Define Legislative, political and claims inflation risk (External Systemic Risk)
Uncertainty associated with changes in political landscape, shifts/trends in the level of claim settlements, etc.
Includes:
1. Impact of recent legislation and court interpretations
2. Potential future legislation with retrospective impacts
3. Precedent setting in courts
4. Changes in medical technology costs
5. Changes in legal costs
6. Systemic shifts in large claim frequency/severity
Complete the sentence:
Legislative, political and claims inflation risk is more material for _______ and sub-risks are often _______.
More material for LT classes and sub-risks are often correlated.
True or False?
Claim Management Process Change Risk are more relevant for PL and for OSC.
False
They are more relevant for OSC than for PL.
Describe how we can quantify Claim Management Process Change Risk.
Discuss current and future process changes with claims management and how changes could impact reporting/payment patterns, closing and reopening rates and case reserves.
Could do sensitivity testing of key valuation assumptions to help assess CoV for this category.
Describe how we can quantify Expense Risk
Discuss with product/claims management to understand drivers of policy maintenance and claims handling expenses.
Can do sensitivity testing around key drivers.
Describe how we can assess Event risk for OSC
Focus on material outstanding events.
Discuss with claim management should help set expectations on final claim cost outcomes.
Range of development patterns on past events may influence the view on uncertainty.
Describe how we can assess Event risk for PL
Look at:
Past experience for event claims (could create a freq/sev model and adjust for future
Output from proprietary cat models
Output from models for perils not covered by cat models.
Important to allow for changes in portfolio size, geo spread, inflation, reinsurance arrangements, etc.
Define Event Risk.
Relates to single events that cause a large number of claims (ex: CATs)
Describe how we can quantify Latent Claim Risk
Negligible for most valuation classes, but material for some WC and Liab classes.
Low prob risk with very high potential severity, thus difficult to quantify.
If risk exposure is significant enough to be included in central estimate valuation or capital adequacy modeling, then it should be examined more closely for setting risk margins.
Discuss with underwriters to identify sources of latent claim risk.
Describe how we can quantify Recovery Risk
Likely to be insignificant except for Auto and TPD recoveries.
Non-reinsurance recovery rates can be analyzed to understand past systemic risks.
Discuss potential risks with Claims management.
Reinsurance recoverability: systemic risk of recovery could be driven by CATs or falling asset returns.
Discuss with reinsurance management to identify possible scenarios, likelihood and potential impact.
Describe the Bolt-On Approach to determine risk margins.
Separate analyses are used to determine the central estimate and the risk margin. Hence, the risk margin is bolted-on to the central estimate.
Define Claim management process change.
Uncertainty associated with changes in claim reporting, payment, estimation, etc.
Define Expense risk
Uncertainty associated with cost of managing the run-off of the insurance liabilities or the cost of maintaining the unexpired risk until the date of loss.
Define Latent Risk
Uncertainty associated with claims that arise from sources that are not currently covered (ex: asbestos)
Define Recovery Risk
Uncertainty associated with recoveries, either reinsurance or non-reinsurance.
Discuss an external system risk for which the assumption of constant correlation between lobs does not hold.
Suggest a method to incorporate variable correlation into the risk margin.
Event risk:
Different LOBs, such as Auto and Home, may be uncorrelated under normal circumstances, but significantly correlated in the tails for CATs that impact both lines.
This dependency can be modelled with a copula.
Discuss 2 possible reasons for unpaid claim reserves of property business being more uncertain than reserves for WC and medical malpractice.
- property business may have significant cats exposure. Recent cats would cause the unpaid claims reserve for property to be more uncertain than WC and medical malpractice.
- Insurer could have adjusted its reinsurance program to retain more property exposure or cede more casualty exposure. This would result in greater uncertainty for property compared to casualty lines.
Identify 3 sources of risk the company should consider in order to avoid underestimating the degree of uncertainty in its unpaid claim reserves for WC.
- Legislative changes to WC benefits
- Inflation on WC claims (LT line)
- Latent exposure risk, such as asbestos
Give an example of LOB for which legislative risk would be important and explain why.
Workers Compensation
Law changes can affect benefit levels. The potential for higher benefit levels would increase the exposure for an insurer and adds significant uncertainty (especially because WC is a long-tail lob)
Give an example of LOB for which event risk would be important and why.
Property
Event risk is important for Property because a single event can cause losses on a large number of policies at once. Event risk may also be correlated in the tails between lines of business (such as Comm Prop and Home)
Give an example of LOB for which latent risk would be important and explain why.
Workers Compensation
WC is a long-tail line of business. There is uncertainty from latent claims risk that the insurer will need to pay for claims from exposures that they did not know existed when policies were written. Latent risks take a long time to be discovered (e.g. asbestos).
Identify 3 questions and actuary could ask company managements to help determine how to segment claims portfolio if insurer writes Personal Auto and Home insurance in multiple geographical locations.
- Do we write in locations prone to catastrophes?
CAT and non-CAT losses develop differently and should be split out if there is significant CAT exposure. - What are the different limits written?
Different limited losses develop differently, so we might segment into different groups based on limit. - What coverages are written?
Different groups of coverages may have different development patterns and should be segmented.
Describe how the choice of valuation classes within a claims portfolio can affect internal systemic risk
The choice of classes must ensure that classes grouped together are homogeneous and in line with central estimate valuation. To reduce risk, they must have similar characteristics and must be sufficiently large to ensure reasonable credibility.
