That's a good observation. You're recalling, of course, that the standard range for the investment rate MfAD is 25-200bps. There is a permitted exception to the standard range, however, and this is described in Section 2260.04. of the CIA.CSOP paper. This is also discussed in Section 5 of the CIA.MfAD wiki article.
There are 2 reasons given to select an MfAD higher than the standard range, and this applies to all 3 categories of MfAD (claims development, investment rate, and reinsurance recovery):
If there is unusually high uncertainty that warrants a higher selection
If the resulting PfAD is unreasonably low because it's based on the best estimate and the best estimate is also unreasonably low.
Overall, I think 2015.Spring Q30a was not a well-constructed question. Unless part of the purpose of the question was to highlight a situation where MfADs can be outside the standard ranges, the question-creators should have made sure the answer fell inside the standard range. You are not the first person who has asked that question. When you're a Fellow and are on an exam committee, make sure you double-check the answers to the questions you create!
like reasons to select MdAD higher than standard range, are there reason to select "lower" than standard range? (Don't think these would be the same as selecting lowest range?)
Also for this Q, part b)-> can we say something like: asset risk is high, payment pattern is not predictable etc. [trying to see if i can use considerations under MfAD (inv) to answer part b & c]
MfAD(clms): There's a brief mention in the CSOP reading about selecting below 2.5% in the case of an insurer with aggregate stop loss coverage that is reserved at the stop loss limit. (I will add this to the BattleCards.)
MfAD(re): The lower bound here is already 0% so it wouldn't make sense to pick something negative.
MfAD(inv): you can select lower than 25bps when the discount rate itself is already lower than 25bps.
In general, there is always judgment involved, so there could be a very specific situation where the appointed actuary might be able to justify an out-of-range selection.
Part (b): That's a good idea to apply MfAD considerations to the specific situation.
If the asset risk is higher for strategy A then strategy B might be better. So I think that would be a valid answer for part (c) versus part (b)
I believe the payment pattern is equally predictable for both strategies. They both involve bonds, and a key feature of bonds is that the payment pattern is determined from the outset. Often bonds pay annual coupons so there is no variation.
I've noticed that in problems like this (where they give you 2 scenarios and ask you to pick which is better) the answer often jumps right out. I think your approach of relating the specific situation to general considerations is good, but try not to overthink it too much.
I have a follow up question for Part(b) of this problem. One of the reasons for choosing Strategy A is "Lower liabilities due to higher yields". I'm not sure I understand the connection between yield and level of liabilities. Is it not possible to have both high yield and high liability at the same time?
The key to understanding the connection between yield and level of liabilities is that an insurer holds liabilities on a discounted basis. If you start with a particular value for undiscounted liabilities and then apply the discounting process, a higher yield will result in lower discounted liabilities. You can see this from the basic formula for discounting:
And about your other question on reinvestment risk: Your reasoning is not correct. Reinvestment risk refers to something different. Here's the definition:
Reinvestment risk is the risk that an investor won't be able to reinvest at the same rate as the original investment.
So, reinvestment risk decreases as the asset duration increases. Strategy A has an asset duration of 4 years (better) whereas Strategy B only has an asset duration of 3 years (worse.) In other words, with strategy B, the insurer has to reinvest more often and that exposes them to more reinvestment risk.
Comments
That's a good observation. You're recalling, of course, that the standard range for the investment rate MfAD is 25-200bps. There is a permitted exception to the standard range, however, and this is described in Section 2260.04. of the CIA.CSOP paper. This is also discussed in Section 5 of the CIA.MfAD wiki article.
There are 2 reasons given to select an MfAD higher than the standard range, and this applies to all 3 categories of MfAD (claims development, investment rate, and reinsurance recovery):
Overall, I think 2015.Spring Q30a was not a well-constructed question. Unless part of the purpose of the question was to highlight a situation where MfADs can be outside the standard ranges, the question-creators should have made sure the answer fell inside the standard range. You are not the first person who has asked that question. When you're a Fellow and are on an exam committee, make sure you double-check the answers to the questions you create!
like reasons to select MdAD higher than standard range, are there reason to select "lower" than standard range? (Don't think these would be the same as selecting lowest range?)
Also for this Q, part b)-> can we say something like: asset risk is high, payment pattern is not predictable etc. [trying to see if i can use considerations under MfAD (inv) to answer part b & c]
Selecting lower than the standard range:
In general, there is always judgment involved, so there could be a very specific situation where the appointed actuary might be able to justify an out-of-range selection.
Part (b): That's a good idea to apply MfAD considerations to the specific situation.
I've noticed that in problems like this (where they give you 2 scenarios and ask you to pick which is better) the answer often jumps right out. I think your approach of relating the specific situation to general considerations is good, but try not to overthink it too much.
I have a follow up question for Part(b) of this problem. One of the reasons for choosing Strategy A is "Lower liabilities due to higher yields". I'm not sure I understand the connection between yield and level of liabilities. Is it not possible to have both high yield and high liability at the same time?
One more question, reinvestment risk is lower for Strategy A because the yields on the bonds is higher right?
The key to understanding the connection between yield and level of liabilities is that an insurer holds liabilities on a discounted basis. If you start with a particular value for undiscounted liabilities and then apply the discounting process, a higher yield will result in lower discounted liabilities. You can see this from the basic formula for discounting:
And about your other question on reinvestment risk: Your reasoning is not correct. Reinvestment risk refers to something different. Here's the definition:
So, reinvestment risk decreases as the asset duration increases. Strategy A has an asset duration of 4 years (better) whereas Strategy B only has an asset duration of 3 years (worse.) In other words, with strategy B, the insurer has to reinvest more often and that exposes them to more reinvestment risk.