Claim thresholds - how many years to consider? (Q9 Sept 06)

[AngelFace]

Dear all,

I'm asking in context of Experience Rating (Chapter 9). When you are calculating the claim threshold for probability whether a policyholder decides to make a claim or not, how many years do you generally consider? I realise this depends on the number of levels of discounts, so say three levels, e.g. 0%, 20% and 40%?

I'd have said the claim threshold needs to be two years (which most of the these type of questions seem to be using), as making a claim affects premium for the next two years, assuming no more claims are made in the future (the third year would be to the highest level - so as if no claims had been made and hence doesn't affect anything). However, Q9 from Sept 06 paper had consider just 1 year in the future - sure the question is a bit more complicated with having to consider two claims, but that doesn't mean you simplify the assumptions - if I was a policyholder I'd certainly still consider the difference it'd make to me on the whole rather than just in the next year. Or am I missing something here?

(The probability would have been quite different with the two assumptions and the examiner's solutions don't seem to have allowed for my assumption).

Any help much appreciated!

 

[Nadeem_Ladha]

From what I have seen - you will always be told in the exam what time horizon you should assume for the policyholders...

Dear all,

I'm asking in context of Experience Rating (Chapter 9). When you are calculating the claim threshold for probability whether a policyholder decides to make a claim or not, how many years do you generally consider? I realise this depends on the number of levels of discounts, so say three levels, e.g. 0%, 20% and 40%?

I'd have said the claim threshold needs to be two years (which most of the these type of questions seem to be using), as making a claim affects premium for the next two years, assuming no more claims are made in the future (the third year would be to the highest level - so as if no claims had been made and hence doesn't affect anything). However, Q9 from Sept 06 paper had consider just 1 year in the future - sure the question is a bit more complicated with having to consider two claims, but that doesn't mean you simplify the assumptions - if I was a policyholder I'd certainly still consider the difference it'd make to me on the whole rather than just in the next year. Or am I missing something here?

(The probability would have been quite different with the two assumptions and the examiner's solutions don't seem to have allowed for my assumption).

Any help much appreciated!

 

[Catrina]

I agree with the above, in all Questions I've practiced we've always been told how many years worth of premium increase the policy holder considers when deciding whether to submit his claim or not. I think where you're going wrong is by assuming dependence between the number of years taken into consideration and the number of discount categories. Not sure where you get the idea from? Good luck!!