18.1 Introduction
VaR by itself it too simplistic for insurance companies
Advantages attributed to Economic Capital
Unifying measure for all risks in an organization
More meaningful to management than RBC or Capital Adequacy Ratios (e.g. Prem/Surplus, Res/Surplus)
Forces firm to quantify risks it faces and combine them into a probability distn
Provides a framework for setting acceptable risk levels for the company and its business units
Remark.
Many other risk measures have the advantages above (not just “economic capital” under VaR)
Insurers typically use multiple risk measure to see if a consistent picture emerges
VaR is typically calculated from the distribution of all risks and then allocate down to inidividual units
This provides consistent measurement of risk across units
Target Probability Level
Current modeling approaches are not able to accurately estimate losses deep in the tail like 1-in-3000 (99.97%) event
Bond ratings are discusses at this level but the ratings are not tied to the probabilities
Ratings are defined by factors (e.g. interest coverage)
Probabilities are published retrospectively after observing rated bonds for many organizations for many years
Target probability level is artificial (no theoretical support)
- Sometimes it is just backed out based on the modeled distribution
Suggestion: Should just express actual capital under various risk measures
The company sometimes compare held capital to the loss distn stating that it’s holding capital to cover a 1-in-x year event. But not the other way around like setting the capital based on it
Modeling difficulty of the tail can be circumvented by focusing on events that can lead to impairment of the company, not just insolvency