October 2012 q4

[walking_alone]

I don't quite understand the solution to part iii for this question. The question states that the policy year is from 1 November to 31 October - I had assumed that this meant that coverage was always this period, ie all policies have the same start date.
So why do we then gross the sum insured up to a full year for the current year? I understand that this assumes that policies are written all year round, but then surely we need to earn the sum insured as well?

Any ideas? Thanks!

 

[Katherine Young]

So why do we then gross the sum insured up to a full year for the current year?

The data is as-at August 2012. So the latest year only consists of 9 months of data. You need to gross it up to the full year so that it is consistent with all the other years.

I understand that this assumes that policies are written all year round

Policies are not written all year round. We're told they're all written on 1st Nov. Your idea of "earning the sum insured" isn't quite right. We've go 9 months of SI earned already. We gross it up to 12 months.

 

[walking_alone]

Policies are not written all year round. We're told they're all written on 1st Nov.

It doesn't quite say this - it says "all policy years offer coverage from 1 Nov.." rather than "all policies". To me these don't feel equivalent!

"Claim numbers and sums insured are reported as at 1 August 2012."

But if all policies are written on 1 Nov then all of the SI is written by August 2012 (ignoring any adjustments). The question doesn't say anything about the SI given being earned to this point, it says reported, hence shouldn't need grossing up. Grossing up implies that we haven't written all of our policies, ie policies are written all year round. Even if the SI was given as at November 2nd, surely we would still have a full year's SI to report?

 

[Katherine Young]

Quote:
it says "all policy years offer coverage from 1 Nov.." rather than "all policies". To me these don't feel equivalent!

The examiners intended you to read this as "all policies incept from 1st Nov".

Try re-working your solution based on this assumption and I'm sure you'll find it better.

Then, if you want to include the latest year in your calculations, you need to ensure it is on a consistent basis with the rest of the data; ie you need to gross it up to a full year's-worth of data.

 

[r_v.s]

In the comments in the examiner's report it says
"Weaker candidates used a method that contradicted information given in the question; for example, applying the development factors as if based on months since the midpoint of the policy year."

Would you please explain what this contradiction was about and it is different from the M1 in the solution?

 

[Darren Michaels]

Hi rvs

I suspect candidates used a variety of alternative methods. As we don't have access to student's scripts, we cant really say exactly what they were and what the contradictions would have been.

However, the examiners' preferred solution will be the one shown in the Examiners' report.

The Examiner's said that some students applied the development factors as if they were based on months since the midpoint of the policy year. This clearly contradicts what it says in the question - that the development factors are based on the months since policy inception - and so it will have affected their projected ultimate number of claims for 2007-2011.

 

[a.virdee]

Hi,

I am confused why the months of development have been adjusted to 9 months. Since the policy starts on 1 November of each policy year, shouldn't the months of development start at 10 months?

Also, a question on the sum insured adjustment - why have the sum insured been inflated using the mid policy year date? I thought we assume that on claims occur on the average policy date. I didn't know that it also applied to exposures too.

Thank you!

 

[Katherine Young]

shouldn't the months of development start at 10 months?

All policies incept on 1st November, and the "as at" date is 1st August. That's 9 months.

I didn't know that it also applied to exposures too.

Certainly it does. The past data applies to a time when yachts are cheaper, so the sums insured need to be brought on level. See chapter 12 page 10.

 

[Geraldine]

The question doesn't say anything about the SI given being earned to this point, it says reported, hence shouldn't need grossing up.

Sum insured isn't something you "earn". In this question, sum insured was the exposure measure. So it was saying that, as at August, they'd had X amount of exposure. So now you have to calculate the exposure to risk for an entire year (corresponding to which, is the full development of associated claims)