Patrik Flashcards
Reinsurance loss reserving problems (7)
- longer claim report lags
- persistent upward development of reserves
- heterogeneity - contracts, reporting patterns, etc.
- industry statistics are not very useful b/c of above
- reinsurer often lacking important info (summary only)
- often have data coding and IT systems problems
- size of adequate reserve to relative to surplus
Causes of persistent upward development of reinsurer loss reserves (3)
- economic and social inflation
- tendency to reserve at modal values
- tendency to under-reserve ALAE
Components of a reinsurer’s loss reserve (6)
- ceding co. case reserves
- reinsurer’s additional case reserves
- IBNER
- pure IBNR
- discount for future investment income
- risk load
Methods for reinsurance reserving for short-tailed exposure categories (2)
- set IBNR = % of latest yr EP
2. reserve up to a selected LR
Method for reinsurance reserving for medium-tailed exposure categories
CL - incurred preferable, but paid can be used if more stable
Methods for reinsurance reserving for long-tailed exposure categories (6)
- CL
- BF
- CC
- credibility IBNR estimate (CL and CC)
- other credibility procedures
- alternate estimation methods - ex: stochastic reserving or freq/sev models
Advantage of using CL method for reinsurer loss reserve estimates
(Patrik)
strongly correlates future development with overall lag pattern
Disadvantage of using CL method for reinsurer loss reserve estimates
(Patrik)
so heavily correlated that credibility is low for immature years
Advantage of using the BF method for reinsurer loss reserve estimates
(Patrik)
correlates future development with an exposure measure
Disadvantage of using the CC method for reinsurer loss reserve estimates
(Patrik)
highly dependent on rate-level adjusted premium
Credibility IBNR estimate
credibility IBNR = Z * CL IBNR + (1 - Z) * SB IBNR
where Z = cred. factor * report lag
» more weight to CL as AYs mature
Potential conclusions if actual claims > expected claims (3)
- purely random
- beginning IBNR was too low
- report lags were too short
Advantage of the CC method for loss reserve estimates
Patrik
more stable than CL results at early maturities