ch15 Doubts
- Fall 2016 (Q18)_part (b): If BF unpaid estimates are consistently higher than Paid CL unpaid est. as we go down the AYs, how does this imply decreased settlement rate as one possibility mentioned in the sample answer? Not able to connect..
Thanks
Comments
I think the sample answer 2 is better and more straightforward because the implied loss ratios can be calculated. You can then directly observe that the loss ratios are decreasing. Note that you're also told that the paid BF method uses the same ECR for each year, which is then apparently false. That means paid development should work better because it doesn't use the incorrect ECR of 56%.
Let's get back to your question about sample answer 1 however: I'm not sure they can definitely state that the settlement rate decreased, but here's what I think their reasoning is. Their observation is that the paid BF method is always higher than the paid development method. In other words, the paid development is always low. Well, what can cause the paid development to be low? One explanation is that the settlement rate is decreasing. They then conclude that the paid development method is more accurate because it's more responsive to a settlement rate decrease.
I'm not so sure this is correct however. If the settlement rate is decreasing, the paid development would tend to under-estimate and in the absence of any other changes, the ECR method (and also the BF method) should be more accurate because they are not affected as much by such a change.
So sample answer 1 is not really correct in my opinion.