Reported Claims
Reported Claims = Paid Claims + Case Outstanding
Unpaid Claims (2 formulas)
Unpaid Claims = Case Outstanding + IBNR (Broad Def.)
= Ultimate Claims - Paid Claims
IBNR (Broad Definition) (sum of 4 elements)
Ultimate Claims
= Paid Claims + Case Outstanding + IBNR
= Reported Claims + IBNR
3 Approaches to Select Tail Factors
Assumptions of development method (3)
Situations in which the development method works well (3)
For development method : Impact of speedup in settlement rate
Paid Claims : Overestimates development factors dans ultimate claims
Reported Claims : No effect
For development method : Impact of Increase in case reserve adequacy
Paid Claims : No effect
Reported Claims : Overestimates development factors and ultimate claims
For development method : Impact of increase in freq and/or sev
Claims increase, but development factors are unaffected
For development method : Impact of increase in exposures
Claims increase, but development factors are unaffected if the average accident date is held constant
For development method : Impact of
Underestimates development factors and ultimate claims
For development method : Impact of change in product mix
Development factors change to reflect the new mix
Expected Claims Method : Key Assumptions (2)
Uses of expected claims method
Advantage (1) and disadvantage (1) of expected claims method
Advantage : Not subject to data uncertainties, gives very stable ultimate claims estimate
Disadvantage : Not responsive to recent claims experience
For expected claims method : Impact of speedup in settlement rate
Paid : Overestimates ultimate claims, but not as much as the development method would
Reported : No impact
For expected claims method : Impact of increase in case reserve adequacy
Paid : No impact
Reported : Overestimates ultimate claims, but not as much as the development method would
For expected claims method : Impact of increase in freq/sev
The EC method accounts for changes unless they occurred within the most recent year
In that case, ultimate claims would be underestimated
For expected claims method : Impact of increase in exposures
No effect, as long as average accident date is held constant
For expected claims method : Impact of average accident date shifts forward
Underestimates ultimate claims, but not as much as development method
For expected claims method : Impact of change in product mix
Impacted if segments changing have different development patterns or different ECR
Main assumption of B-F method
Unreported/Unpaid Claims will develop based on expected claims
Situations in which B-F works well (3)