Task 2

[ILAmember]

Task 2 questions and comments here

[Joseph Martin]

Hey guys. Not sure who all is in this group and still working on this, but I can get us started with a question:

Possible errors in Excel file? 

1. The formula in cell I5 was subtracting E5 from E4, but I think it makes more since to do E5 - E3 to compare to the baseline case similar to K5. 
2. The approved pol count in the excel file was 725 and 800 for Triage and Simple respectively - yet in the prompt, it is 700 and 850. Which makes more sense to use?

Cheapest Underwriting Type

1. I am getting the Traditional Medical as the cheapest, but how the prompt is worded, it makes it seem like one of the other two should be cheaper. Anyone else getting the same thing?

[DragonV]

1. Yes, you are right about E5-E3.

2. In my handout, it is 725 and 800. I would suggest you use either you feel comfortable with and then explain there is a discrepancy.

3. I think from a total expense perspective, Traditional Medical is the cheapest. However, the way I see the cheapest is to consider both Underwriting and Death Benefit at a Per policy level. Although simple is more expensive, but per policy expense it not the highest. Assuming they all charge the same price, then Traditional does not have the highest profit.

Just my own opinion, hope it help.

[Joseph Martin]

This helps a lot. Putting PV of DB on a "per policy approved" basis gives you a different perspective than just looking at total expense saving. Thanks.

[Anne-Laure Theall]

You see traditional as the cheapest? Mine for total and for per unit expenses has simple < triage < traditional. I definitely think they are supposed to be cheaper and less invasive. That’s the whole point.

Did anyone else normalize the death benefits by assuming same number of policies for each? I felt like triage and simple were being unfairly treated here because of higher number of policies approved. If we have different number of policies then I feel like we need to see premium as well.

[Anne-Laure Theall]

Oh wait, are you looking at pre-tax change in expenses when saying traditional is cheapest?

Since I normalized for number of polices I have triage<traditional< simple implying the increase in DB for simple is not offset by the decrease in expenses. Sorry everyone I am quite tired haha but trying really hard to get these handed in my Dec 20 for the fac deadline!

I still feel like that’s the best way to compare expense savings and risk... really curious what others think. Thanks!

[DragonV]

I didn’t say that I think traditional is the cheapest. Sorry if there is any confusion. Another thing here is, I suggest to not only look at expenses/profits, but also think about the impact of each underwriting type on policyholders...They probably have a preference to certain types of u/w than other types...

[Hailey]

Do you only see triage being the cheapest when you normalized? Triage is the method I want to recommend but having trouble justifying it since I am seeing it as the most expensive. I wasn't planning on normalizing since in reality, they will approve more policies under the less strict methods and therefore have more exposure to mortality risk which I think is relevant when making a final recommendation. can you explain why you think it makes sense to normalize? thanks! 

[Anne-Laure Theall]

Yes. My thought process is that we shouldn’t penalize an underwriting method for approving more policies and ultimately making more sales. At least I don’t think so. I understand that the additional policies that they’re approving are the unhealthy ones but we would be collecting additional premium too.

So I was really trying to get at whether the increase in benefits paid was offset by decrease in expenses and hence not impacting or favorably impacting profit. Not sure I got it right.

[Anne-Laure Theall]

Another approach perhaps, looking at total underwriting cost per policy sold vs PV of Benefits per policy sold. This has simple come out on top and as far as implementation and third party desires and healthy population, seems there’s a lot to be able to justify.

I worry that I’m going above what soa wants here and over complicating. But I am having a lot of trouble with their set up. I don’t like it

[EZ]

For the part on the UW expense basis used by actuaries and underwriters, is it referring to per policy vs per per application or is it something else?

[Joseph Martin]

I think what it's trying to get at is UWs would care more about per application expenses whereas actuaries care more about per policy expense. Possibly? That's my gut feeling at least. Could be wrong.

[Vandan Patel]

I see discussions on adjusting either total underwriting expenses or PV of benefits so that they are either both based on application count, or both based on approved policies. 

To me, it seems that the total underwriting expense should be based on # of applications as an expense is still incurred for underwriting applications that aren't approved. Also, PV of future benefits would be per approved policies as those would be the policies in-force in future durations. 

This is just my opinion though. 

[Shannon]

Is the 'different basis' for actuaries vs. underwriting leaders explained/defined anywhere, or are we supposed to just assume this is the difference between expenses on a per-policy vs. per-application basis? And, if we are supposed to assume, which basis goes with which group? 

[Nick Odegaard]

I think it’s related to per application and per policy basis. Per application would be for UW and per policy would be for actuaries. The data that actuaries get are only on paid policies so they wouldn’t know application count unlike UW

[Shannon]

Ok, that makes sense. Thanks.

[Nabeel Khairani]

Anyone here, who can help me understand why actuaries and underwriting leaders view underwriting expenses on a different basis.