London Market & Lloyd's

Exposures

The requirement to deeply understand the exposures in the book means that it is common to strip particular features out of large, heterogeneous books of business.  These may be lineslips or binding agents or they may simply be problem contracts or catastrophic losses. 

These features may already have separate market studies performed on them, if they are known market-wide issues.  Alternatively, it is common to perform an exposure analysis on problem contracts.  This involves investigating the potential coverage available to the insured and estimating how much of this is likely to be used.  Finally, it may be appropriate to simply project the experience of the stripped-out feature individually.
Development Factor Models

For the remaining large, relatively homogeneous book of business, it is common to use Development Factor Models as a first approach to establishing reserves.  Development Factor Models assume that historic development of claims is a good guide to future development.  For example, if the cumulative level of claims has historically increased by 50% in the second year after underwriting, we may assume that future claims will also increase by 50% in the second year after underwriting.

Development Factor Models also allow for the projection of claims beyond what has been experienced so far, which is known as a tail factor.  Tail factors are projected by fitting curves to observed experience and extrapolating from that curve.

Bringing in Other Information

Development Factor Models often give only a partial way of tackling reserving.  For example, they may give misleading results for years with little development to date. 

We may have independent views on the probable ultimate claim level.  For example, an underwriter will probably have written the book with a particular target loss ratio in mind.  Alternatively, we may perform an exposure analysis on the book.

There are various techniques for combining this independent information.  The most common is called the Bornhuetter-Ferguson method.  This combines the results of the Development Factor Model and the independent assessment based on the credibility of the claims development to date.

Development Factor Models also rely on historic development being applicable to future claims, which may not be appropriate.  It is important to understand how development patterns may have changed and to build that into the final model.

Benchmarking

Another important technique used to establish reserves for London Market business concerns the use of benchmarks.  This is discussed elsewhere on the website.

Characteristics of London Market Business