Data and Reserving

Question 1

Data and Reserving

Answer 1

• Uses of data
• Problems with industry wide data
• Factors influencing data quality/quantity
• Reasons for estimating reserves
• Factors determining the choice of reserving method
• Main methods for outwards reinsurance reserving
• Uses of stochastic reserving
• Key assumptions for the Mack Model
• Key assumptions for bootstrapping the ODP
• Objectives in communicating uncertainty
• Factors to consider when asked to adjust reserves
• Additional data issues associated with inwards reinsurance
• Considerations when assessing the suitability of outwards reinsurance reserving methods
• Requirements of a good model

Question 2

Uses of Data

SIR CP MAPEMAP

Answer 2

SIR CP MAPEMAP

• Statutory returns (IFRS)
• Investment strategy
• Risk management
• Capital Modelling (SAM)
• Performance analysis
• Management information and financial control
• Accounting
• Provisioning (Reserving)
• Experience analysis and statistics
• Marketing
• Administration
• Premium Rating and Product costing (Pricing)

Question 3

Problems with industry wide data

ONE CLINIQUE

Answer 3

ONE CLINIQUE

• Out of date - takes time to collect, collate and distribute
• Nature of data stored by different companies will not always be the same
• Economic and geographical sectors - operate in different socio-economic and geographical sections of the market
• Coding used for risk factors may vary from company to company
• Less detailed and less flexible
• Identical - policies sold by different companies are not identical
• Not all companies will contribute (not reflective of industry as a whole)
• Industry - not reflective of industry as a whole
• Quality of data depends on participants
• Underwriting, claims handling, reserving and risk management practices differ
• Errors

Short form:
- Out of date
- Nature
- Economic and geographical sectors

• Coding of risk factors
• Less detailed and less flexible
• Identical
• Not all companies contribute
• Industry
• Quality
• Underwriting, claims handling, reserving and risk management
• Errors

Question 4

Factors influencing data quality/quantity

SEMI COM

Answer 4

SEMI COM

• Size and age of the company
• Existence of legacy systems
• Management and staff responsible for collecting and maintaining date
• Integrity of data systems
• Class of business
• Organization - nature of the organization
• Method of distribution

Short form:
- Size
- Existence of legacy
- Management
- Integrity

• Class of business
• Organization
• Method of distribution

Question 5

Reasons for estimating reserves

PS I REST Mi CAT

Answer 5

PS I REST Mi CAT

(Determine insurers reserves for):
- Published accounts
- Supervisory authority

• Internal management accounts, business plans and budgets
• Reasonableness - check reasonableness of reserves to form an independent opinion
• Estimate ultimate claim costs for premium rating
• Sale/purchase - value of an insurer for sale/purchase
• Transfer a book of business
• Management - provide management with information on business performance by class of business
• Commutation - negotiate a commutation for buyer/seller with regards to run off reinsurance
• Adequacy - test adequacy of case estimates
• Tax liabilities - determine tax liabilities of insurer

Short form:
- Published
- Supervisory authority

• Internal management accounts
• Reasonableness
• Estimate ultimate cost of claims for premium rating
• Sale/purchase
• Transfer
• Commutation
• Adequacy
• Tax liabilities

Question 6

Factors determining choice of reserving method

TECH KAP

Answer 6

TECH KAP

• Type of claims that have been incurred/may be expected to occur
• Extent and quality of data available
• Class of business (Claims development [length of tail] and exposure period)
• Historical trends and patterns
• Key factors that determine claims development (claim delays) and payments
• Age of business/ cohort of claims and historic claims development information
• Purpose of reserving exercise

Short form:
- Type of claims
- Extent and quality of data
- Class of business - claims development and exposure period
- Historical trends and patterns

• Key factors
• Age
• Purpose

Question 7

Main methods for outwards reinsurance reserving

BITCHD

Answer 7

BITCHD

• Broad brush approach - adjust gross data using a broad-brush approach
• Individual losses - develop all individual losses then apply reinsurance to each one
• Triangulation techniques - perform standard triangulation techniques directly on reinsurance data alone
• Case-by-case approach on only the largest losses
• Huse data gross and net of reinsurance, then find the difference
• Distribution - derive a reserve distribution net of reinsurance

Short form:
- Broad brush
- Individual losses
- Triangulation techniques
- Case-by-case
- Huse gross and net and take the diff
- Distribution reserve net of reinsurance

Question 8

Uses of stochastic reserving

CAMP AD

Answer 8

CAMP AD

• Compare different estimates and datasets
• Allocate capital for reserve risk
• Monitor performance to see if claim movements are material
• Provide information to investors (assess relative attractiveness)
• Assess reserve adequacy in absolute and relative terms
• Discussions - facilitate discussions with regulators (run off risk)

Short form:
- Compare
- Allocate
- Monitor
- Provide

• Adequacy
• Discussions

Question 9

Key assumptions of the Mack Model

VIR

Answer 9

VIR

• Variance of cumulative claims to development time t is proportional to cumulative claims amount to time t-1
• Independent - Future development of a cohort is independent of historical factors
• Run-off pattern is the same for each origin period

Short form:
- Variance
- Independent
- Run of pattern same

Question 10

Key assumptions for bootstrapping the ODP

RIPP

Answer 10

RIPP

• Run-off pattern is the same for each origin period
• Incremental claims are statistically independent
• Proportional - variance of incremental claim amounts is proportional to the mean
• Positive - incremental claims are positive for all development periods

Short form:
- Run of pattern same
- Independent incremental
- Proportional incremental variance to mean
- Positive

Question 10

Objectives in communicating uncertainty

CEASES

Answer 10

CEASES

• Consistent - be consistent with vocabulary used by other professionals and explain terms
• Emphasize bigger and unusual issues
• Avoid misunderstandings
• Stakeholder - ensure stakeholders understand the level of uncertainty
• Explain what has been allowed for in the best estimate and what has not
• Scope and purpose - commenting in the context of the scope and purpose

Short form:
- Consistent
- Emphasize
- Avoid
- Stakeholder
- Explain
- Scope

Question 11

Factors to consider when adjusting reserves

Ping Pong People & Radio Raps on FM

Answer 11

Ping Pong People & Radio Raps on FM

• Purpose of the reserving exercise
• Peer review - seek peer review
• Professional guidance/legislation and actions taken by competitors
• Reasons suggested by management
• Range - current estimate relative to range of reasonable estimates
• Financial strength - current financial strength of insurer and potential cushions
• Materiality - overall materiality of adjustment

Question 12

Additional data issues associated with inwards reinsurance

BIG DUHS

Answer 12

BIG DUHS

• Benchmarks less relevant
• IT constraints
• Group data - more opportunity to group data differently
• Delays - longer claim reporting delays
• Upwards - greater tendency for claims to develop upwards (non-proportional reinsurance)
• Heterogenous - exposure can be very heterogenous
• Sparse - data can be sparse

Question 13

Considerations when assessing the suitability of outwards reinsurance reserving methods

Answer 13

• Simplicity vs accuracy
• Consistency of gross and net estimates
• Whether it can be used to assess volatility of net outcomes / reinsurance recoveries
• Compliance with regulation
• How method copes with
• Different types of reinsurance
• Sparse data
• Changes in reinsurance programmes/panel over time
• CAT and large claims
• Aggregate features such as profit commissions, and loss sensitive contracts such as stop loss cover
• Interactions between covers
• Whether the method can be used to investigate
• Capital requirements and ERM
• Credit risk

Question 14

Requirements of a good model

VARIABLE CRRISPS

Answer 14

VARIABLE CRRISPS

• Valid
• Adequately documented
• Rigorous
• Inputs to parameter values are appropriate
• Arbitrage free
• Behavior reasonable
• Length appropriate
• Easy to explain and understand workings
• Clearly displayed, verifiable and communicable results
• Reflects business accurately - model points
• Refinement and development is possible
• Independent verification of results
• Sensible joint behavior of variables
• Parameters allow for material features of the business
• Simple - not overly complex