Actuarial Outpost
 
Go Back   Actuarial Outpost > Actuarial Discussion Forum > Life
FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions

2017 ACTUARIAL SALARY SURVEYS
Contact DW Simpson for a Personalized Salary Survey

Reply
 
Thread Tools Search this Thread Display Modes
  #1  
Old 01-20-2011, 12:05 PM
Jorge Julio Jorge Julio is offline
 
Join Date: Jan 2011
Posts: 17
Default Graded Benefit Life and IRC 7702

Iím curious about the compliance of a Graded Benefit Life products with Section 7702 of the IRC. A benefit grading feature Iíve seen, for annual pay policies, allows for the non-accident DB to be return of premium plus interest during the first couple policy years, then $1k/unit thereafter. The DB for accidental death, as defined in the contract, is $1k/unit at all times.

I believe Section 7702(e)(1) lists some computational rules, one of which is that the DB is assumed to not increase. If the ROP + interest DB from year-1 must be the assumed DB in all years when illustrating compliance with 7702, the product would clearly fail the test.

Section 7702(e)(2) allows some exceptions to the Ďno increaseí rule above, but, among other things, appears to limit allowed increases to 10% of the initial DB. Just going from ROP in year 1 to ROP in year 2 would violate this allowed 10% increase.

Is there something Iím missing that allows this type of product to pass Section 7702?
Reply With Quote
  #2  
Old 01-21-2011, 03:03 PM
Freebird's Avatar
Freebird Freebird is offline
Member
SOA AAA
 
Join Date: Dec 2001
Location: Topeka, Kansas
College: Graduated from Ohio State too many years ago to remember
Favorite beer: Newcastle
Posts: 822
Default

I don't know the 7702 rules of which you speak, but I do remember a product at a former company very similar to the one you mention. We never even considered 7702 as an issue, and the product was sold for years as a graded death benefit whole life. The product was graded that way because there was either no underwriting or very limited, and the grading was done to prevent anti-selection. IMHO (obviously not an SAO), I think it qualifies as life insurance under 7702.
Reply With Quote
  #3  
Old 01-21-2011, 07:21 PM
JimmyJoe's Avatar
JimmyJoe JimmyJoe is offline
Member
SOA AAA
 
Join Date: May 2007
Posts: 68
Default

If the guaranteed CV in any year of your graded plan is greater than the NSP then needed to fund the test plan to maturity, then the graded plan fails.

The test plan is what is defined in 7702.

Me thinks that's right
Reply With Quote
  #4  
Old 01-21-2011, 08:05 PM
JoJo JoJo is offline
Member
 
Join Date: Jan 2002
Posts: 1,704
Default

Yes, you should use the CVAT test. You'll find your cash values in the first 2 years are nearly nil so no problem there.

I think you're taking "the DB is assumed to not increase" too literally. This was put in the reg so companies didn't include future increases, such as guaranteed increase options, that may be opted out of, to increase the guideline premiums from inception. I would use the scheduled death benefits to calculate the NSP and ignore the ADB in the first 2 years.

On a side note: you'll probably will find most of the policies will be MECs.
Reply With Quote
  #5  
Old 01-24-2011, 08:55 AM
Jorge Julio Jorge Julio is offline
 
Join Date: Jan 2011
Posts: 17
Default

Thanks for the responses, but I'm not sure I follow them all.

Quote:
Originally Posted by JimmyJoe View Post
If the guaranteed CV in any year of your graded plan is greater than the NSP then needed to fund the test plan to maturity, then the graded plan fails.

The test plan is what is defined in 7702.

Me thinks that's right
Quote:
Originally Posted by JoJo View Post
Yes, you should use the CVAT test. You'll find your cash values in the first 2 years are nearly nil so no problem there.

I think you're taking "the DB is assumed to not increase" too literally. This was put in the reg so companies didn't include future increases, such as guaranteed increase options, that may be opted out of, to increase the guideline premiums from inception. I would use the scheduled death benefits to calculate the NSP and ignore the ADB in the first 2 years.

On a side note: you'll probably will find most of the policies will be MECs.
My understanding is that:
- For the CVAT, you compare the cash values of the actual product to the NSPs of the test plan mentioned above.
- Section 7702 defines the rules for the test plan, and one of those rules is that you assume the DB of the test plan is equal to the lowest DB of the actual plan.

The second point there is what I'm having difficulty with. If for the test plan I assume the actual plan's lowest DB (ROP+ in year 1) as the DB for all years, then I have a situation where the actual plan fails the CVAT - at some point the CV of the actual plan approachs $1k/unit where the NSP to cover a DB of YR-1 ROP+ is nowhere near $1k.

I'm guessing my understanding is off somewhere because this type of product does exist in the market. I figure that it exists at least partially due to the favorable tax treatment life insurace contracts get from passing 7702 tests.

Is there a chance that since the DB for accidental death is always $1k/unit the DB for the test plan is assumed to be $1k/unit at all times?
Reply With Quote
  #6  
Old 01-24-2011, 12:38 PM
JoJo JoJo is offline
Member
 
Join Date: Jan 2002
Posts: 1,704
Default

Do you have access to the book "Life Insurance & Modified Endowments"?
The answer is on the bottom of page 145.

It references actuarial guideline XXV and it's ability to meet CVAT and satisfy state NF.
Reply With Quote
  #7  
Old 01-24-2011, 01:17 PM
Jorge Julio Jorge Julio is offline
 
Join Date: Jan 2011
Posts: 17
Default

Quote:
Originally Posted by JoJo View Post
Do you have access to the book "Life Insurance & Modified Endowments"?
The answer is on the bottom of page 145.

It references actuarial guideline XXV and it's ability to meet CVAT and satisfy state NF.
I have read that part of that book (and AG XXV), but am not sure it applies to this case: that paragraph appears to apply to "plans where the death benefit is adjusted based on the Consumer Price Index," where the grading is not known ahead of time because the changes in the index are not known ahead of time, not a guaranteed grading of DB from ROP to $x/unit.
Reply With Quote
  #8  
Old 01-24-2011, 03:56 PM
JoJo JoJo is offline
Member
 
Join Date: Jan 2002
Posts: 1,704
Default

Quote:
Originally Posted by Jorge Julio View Post
I have read that part of that book (and AG XXV), but am not sure it applies to this case: that paragraph appears to apply to "plans where the death benefit is adjusted based on the Consumer Price Index," where the grading is not known ahead of time because the changes in the index are not known ahead of time, not a guaranteed grading of DB from ROP to $x/unit.
You're right. I hadn't looked at that reg and it doesn specify CPI only.

Here's another thought - CVAT says "contract may not at any time exceed the NSP which would have to be paid at such time to fun future benefits under the contract"

So, in year 1 & 2 you may be limited to using the premium refund amount in the NSP calc, but years 3+ you should be able to use the full face amount in your NSP calc. CVs in years 1 & 2 are very low and still should pass if you use the lower NSP values.
Reply With Quote
  #9  
Old 01-24-2011, 08:22 PM
CDesRochers CDesRochers is offline
Member
 
Join Date: Oct 2002
Posts: 236
Default

Under the CVAT, the comparison is the net single premium for the current death benefit, so for a graded benefit policy, the maximum allowable cash value is tested against the then current death benefit. However, except for some exceptions for preneed policies, the NSP is still based on a level death benefit, which ignores contractual increases.

In the ROP death benefit design, that would be used as as the assumed level benefit in each period.
Reply With Quote
  #10  
Old 01-26-2011, 07:36 PM
Jorge Julio Jorge Julio is offline
 
Join Date: Jan 2011
Posts: 17
Default

Quote:
Originally Posted by JoJo View Post
You're right. I hadn't looked at that reg and it doesn specify CPI only.

Here's another thought - CVAT says "contract may not at any time exceed the NSP which would have to be paid at such time to fun future benefits under the contract"

So, in year 1 & 2 you may be limited to using the premium refund amount in the NSP calc, but years 3+ you should be able to use the full face amount in your NSP calc. CVs in years 1 & 2 are very low and still should pass if you use the lower NSP values.
Quote:
Originally Posted by CDesRochers View Post
Under the CVAT, the comparison is the net single premium for the current death benefit, so for a graded benefit policy, the maximum allowable cash value is tested against the then current death benefit. However, except for some exceptions for preneed policies, the NSP is still based on a level death benefit, which ignores contractual increases.

In the ROP death benefit design, that would be used as as the assumed level benefit in each period.
Thanks a lot for the input. I think I've got a much better handle on this now.
Reply With Quote
Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


All times are GMT -4. The time now is 05:27 PM.


Powered by vBulletin®
Copyright ©2000 - 2018, Jelsoft Enterprises Ltd.
*PLEASE NOTE: Posts are not checked for accuracy, and do not
represent the views of the Actuarial Outpost or its sponsors.
Page generated in 0.26729 seconds with 9 queries