Reserving E_01 Practice Problem #2 (2017.Fall 25)
Hi,
It looks like we can develop either the net reported or gross reported claims, per the examiner report. If we develop gross reported claims and ceded reported claims separately, can we develop reinsurance using a ratio approach? That seems like the most intuitive way to recognize the relationship between the two to me.
Additional question on this specific problem: Do we always assume that net paid losses will not go down when they're above the stop-loss limit? If 2015 net paid claims are $8,400 and the stop-loss limit is $8,000, shouldn't we ultimately get $400 back, which would partially offset net unpaid claims in AY 2016, if they're asking for all AYs combined?
Thank you so much!
Comments
When you say a ratio approach, do you mean ceded losses divided by gross losses? If so, I'm not sure that would be a great approach. You could try it and see how that compares to the given answers.
In any case, I don't think the ceded losses have enough credibility to do anything other than subtract them from the gross losses to get net losses. I see that the second sample answer actually developed the triangle of ceded losses but it really doesn't make sense to do that because you can't even calculate age-to-age factors for all cells. If I were the grader, I'm not sure I would have given sample answer 2 as much credit as sample answer 1 even though the answers were close to each other.
Regarding your other question: I suppose the net paid loss should not have gone about 8,000 but I don't think you had to worry about that detail in this problem because the same thing happened in the actual exam problem. They had a net paid loss of 5,102 in AY 2013 which had stop-loss limit of 5,000. That was just the way the net paid data was given.
I did mean ceded losses divided by gross losses and developing those ratios to the ultimate value and multiplying that ratio by the gross losses. And I thought the same thing, that there was no way there was enough credibility in the ceded losses to develop them, but they provided a triangle so it seemed like the required thing to do. It seems odd to me that we would develop net reported losses instead of gross reported losses when we don't know how the reinsurance treaty is being applied, but it seems like one of those "that's how they want it" circumstances. Thanks again!
True, in practice you would want to know more about the reinsurance treaty before rushing off into a calculation...