Reserving-Related Working Parties to Present Results at CLRS

Unlike past years where the CAS Committee on Reserves (CASCOR) focused on eliciting reserve-related call papers dealing with the complex issues that challenge the property/casualty insurance industry, this year CASCOR focused on producing research, analysis, and papers that would assist the actuary and the student in dealing with specific areas not widely addressed in the past. Thus, two working parties were formed in late 2004, one to focus on Tail Factors, and the other on Bornhuetter Ferguson-Initial Expected Losses. Both have been making much progress over the last several months.

The Tail Factor Working Party (TFWP), co-chaired by Steve Herman and Mark Shapland, has been "working its tails off" evaluating numerous tail factor methods. Starting from day one, the focus was to identify all such methods contained within CAS literature and then identify methods currently being used, but not contained within the literature. The laborious process yielded much fruit as more than twenty methods were identified. The next step involved obtaining data from numerous sources so that sufficiently detailed data could be used to test the various methods in order to judge which were feasible to use under what circumstances. As this is written, the TFWP is completing its work and will be presenting its findings at the CLRS in Boston.

The first task completed by the Bornhuetter-Ferguson Initial Expected Losses Working Party (BFIELWP) was a survey of the member's workplaces (and the workplaces of other actuaries) regarding methods currently in use. The next task was identifying the currently available literature on this topic (literally dozens of papers) and summarizing the methods used therein to establish initial expected losses. Many of the methods that have been documented will be familiar to experienced actuaries, although everyone seems to have their own twist. Also, many methods work fine under several circumstances, but fail in others. The BFIELWP will compare and discuss the strengths and weaknesses of each method to address this issue. The BFIELWP will present its findings at the CLRS in September. The paper is slated for completion during fall 2005. Jeff Carlson and Chris Olson are co-chairing the working party.

Both papers will be available online when completed.