Chapter 9 - Risk & Uncertainty

Question 1

Uncertainty in insurance policies

Answer 1

insurance transfer risk/uncertainty from PH to insurers

uncertainty is the inability to predict the future with confidence. Uncertainty exists in both the amount and timing of cashflows. need to include a loadin to cover this

risk is = Possibility of variation in financial results (gain or loss).

Question 2

process uncertainity and effect on claims

Answer 2

Process error refers to the inherent randomness underlying a book of business.

An insurer’s specific process uncertainties stem from:

EXTERNAL
-general claims uncertainty relating to the specific business written (external). Always uncertainty around claim frequency (propensity to claim Increase) claim severity and claim payment patterns (delay between when date of claim event, date of rpeorting claim and final settlement of claim)
-changing development patterns
-demand surge eg increase in builders after a flood
-climate change
-government legislation

INTERNAL
-internal influences, eg changes in reserving philosophy or mix of business. (internal)
-planned or unplanned changes in mix eg change in UW process or pricing
-new markets
-new distribution channels
-new claims handing process
-increase use of profit share arrangements

Question 3

What is data error?

Answer 3

Covers uncertainty arising from the way past claims are reported and stored

Question 4

What is model error

Answer 4

Covers uncertainty in the selection of the model
And how it is used

Question 5

process uncertainty - effect on other areas

Answer 5

aggregators

off-shoring

effect of economic on investments and expenses

competition

insurance cycle

expenses uncertainty

Question 6

parameter uncertainty

Answer 6

refers to the uncertainty arising from estiamtion of parameters used in a model.

uncertainty from data used that may be poor quality, internally inconsistent, incomplete and non-existent. format data

-changes in case estimated reserving philopshy

-large, cat and latent claims

-claims inflation not as expected

-new distribition channels

Question 7

model uncertainity

Answer 7

model error arises form the choice of or specification of the model. (less easy to detect so uncertainty than paramter error)
-programming error
-incorrect distibuitional assumptions in modelling claim value uncertqainty
-others like not using enough simulations in a stochastic model
-incorrect correlation assumptions