ST8 2010 Specimen Paper - Q7 and inflation

[test72738]

Could someone explain why the inflation factor for the year 2007 is 1.09^3? I would assume that it should be 1.07^0.5 (which gets you to the start of 2018) and then 1.09^2.5 which gets you to the midpoint of 2010. What's going on here?

Could someone also explain why the assumption that the claims inflation in 2010 is the same as in 2009? I don't see how we use that assumption in the calcs.

 

[Katherine Young]

Could someone explain why the inflation factor for the year 2007 is 1.09^3? I would assume that it should be 1.07^0.5 (which gets you to the start of 2018) and then 1.09^2.5 which gets you to the midpoint of 2010. What's going on here?

Ah yes, I see your logic. It sounds like you’ve used a valid approach, given the information.

The examiners used a simpler method. Their interpretation is that any claim event occurring in 2007 (for example) will be exactly 7% bigger than a corresponding claim event occurring in 2006.

Notice that the question says 'estimate', so in practice a wide range of approaches was allowable as long as you stated your (suitable) assumptions.

Could someone also explain why the assumption that the claims inflation in 2010 is the same as in 2009? I don't see how we use that assumption in the calcs.

Let’s look at the 2006 accident year as an example. We inflate that up to the 2007 accident year using the factor 1.07. Then we inflate it up to the 2009 accident year by using an extra factor of 1.09^2. Now we have to inflate it up to the 2010 accident year, using the appropriate inflation rate for that year… but hang on, the question doesn’t tell us what inflation rate to use for 2010, so we’ll just use the same as before and state the assumption that “inflation in 2010 is the same as in 2009”. So the total inflation factor for the 2006 accident year is 1.07*(1.09^2)*1.09.

 

[test72738]

Ah yes, I see your logic. It sounds like you’ve used a valid approach, given the information.

The examiners used a simpler method. Their interpretation is that any claim event occurring in 2007 (for example) will be exactly 7% bigger than a corresponding claim event occurring in 2006.

Notice that the question says 'estimate', so in practice a wide range of approaches was allowable as long as you stated your (suitable) assumptions.



Let’s look at the 2006 accident year as an example. We inflate that up to the 2007 accident year using the factor 1.07. Then we inflate it up to the 2009 accident year by using an extra factor of 1.09^2. Now we have to inflate it up to the 2010 accident year, using the appropriate inflation rate for that year… but hang on, the question doesn’t tell us what inflation rate to use for 2010, so we’ll just use the same as before and state the assumption that “inflation in 2010 is the same as in 2009”. So the total inflation factor for the 2006 accident year is 1.07*(1.09^2)*1.09.

Thanks. The second point makes sense.

Going back to their method, I've never seen this before in any other exam questions - I always thought we inflate from the middle of the claim period to the middle of the exposure period in question. Could you suggest any other questions where they've used this method or am I OK to stick with line of thinking for the exam?

 

[Katherine Young]

Both methods would be fine I'm sure, so if I were you I'd stick with whichever approach you're more comfortable with.

 

[test72738]

Thanks