Most of the plans that I work on have a normal retirement age (NRA) of 65. When I process their cross testing, the EBARs are calculated by calculating years to retirement on the basis of 65 as the testing age. I recently got a plan that contains an NRA of 59 1/2. My questions/comments are as follows:

1) My assumption is that I should be using age 59 1/2 as my testing age. Is this correct?
2) If, so, I would assume that I base the calculating years to retirement on the year the participant turns age 59 1/2. Correct?
3) Or do I simply drop the 6 months requirement from the calculation of the exponent when calculating years to retirement? (Essentially makign the testing age 59?)
4) Am I permitted to use age 65 as my testing age?
5) Am I permitted to use a different age entirely as long as I do not exceed age 65? (I do not know why I would actually do this, but I am curious.)
6) What if the NRA in the doc was age 65 and 5 YOS? Is the testing age different for anyone who would not have 5 YOS by age 65? (66,67, etc...)
7) If the answer to #5 is that the ages would be different, would I not have a problem with some sort of uniform retirement age?
8) Is there anything else I should be aware of regarding NRA?

Thanks in advance for the replies.

My two cents:

1. From what you have said, yes.
2-3 I think any reasonable approach would be accepted as long as it does not skew results towards HCEs. I know we've used age 60 and had it approved.
4-5. I don't think so. You have a uniform retirement age from what you've said. That makes it testing age.
6. I'll let Tom answer that. I asked him that years ago and I didn't understand the answer. My view is that that reasonable people can disagree about that. I would default to using actual NRA (66+) for such a person and not worry about uniformity.
8. Let the document drafter have it.

6. If the age is uniform (and 65 and 5 YOS satisfies that definition in the regulation) did you really get an answer other than yes? Or was it that if the definition is NOT uniform then you use 65 even for those who would have a retirement age of 66 under a 65 and 5 YOS definition because, well, there is something that causes the definition not to be uniform (like the lower of two ages, one of which is determined with respect to 65 and 5 YOS and the other is determined based on something else)?
7. Not a problem.
8. If you want to use 65, then do something to your definition that pushes it out of being uniform.

Mike, would you try again on 6? I don't understand what your opinion is. I agree with your comments, but I still don't know what your answer is.

I agree that 65+5 is not a non-uniform testing age, although I'm not sure what age it actually is.

Is it a variation of 65 in which case there is one uniform testing age with different practical results (e.g. 65-70)? Or is it one of two (65 being the other) different uniform retirement ages, in which case the rules are convoluted as follows:

1.401(a)(4)-12 in such case says that Testing Age "is the employee's latest normal retirement age under any uniform normal retirement age under the plan, regardless of whether that particular uniform nra actually applies to the employee under the plan."



So, I still don't know what "the" answer is. What is your opinion? Thanks.

All:

Thank you for your replies. I appreciate your responses and I have a follow-up question. (It is a Relius question and does belong in the Relius board, but I don't know how to move it and I thought the history above would make it easier to discuss.)

Utilizing the info above, I am trying to run my cross tested allocation. When I run the ABT, it shows the projection to age 65. It is NOT projecting to age 59 1/2. Can anyone tell me how to instruct the system to change the projection to the NRA?

Any help would be greatly appreciated.

Mr. Poje - Is this something you might have in your vast knowledge of Relius?

My knowledge is better on "Don't Rely-on-Us"

anyway, the system should be using whatever you have coded for NRA in Plan Specs.
if you are getting everyone to show at age 65 it sounds like a coding problem. (I even tried changing the retirement date on a dummy plan I have and age 60 shows up as testing age (or it could be 59 depepnding on if you code age nearest or age last)
The only other ting I can think of is that you have the definition for retirement date as something other than
'date of event' or '1st of the month following' (this would apply to any definition - not just 59 and a half)
I could see this creating nonuniform ages, and therefore the system using age 65.

e.g. if it is coded plan val date nearest, some people would be age 59 and others 60. this is nonuniform and so I will go out on a limb and guess the system would use age 65.

Not sure how to respond. I think I made my position clear.
Why isn't it 65 for anybody who's 65/5 would be 65 and 66 for anybody who's 65/5 would be 66, etc.?


That section is trying to say that if you have a series of uniform NRA's and the employee is eligible for an earlier one, you still use the later one for testing. It does not mean that if one employee's 65/5 is 66 and another employee's 65/5 is 67 that you use 67 for both.

Clearer?

Yes, and I agree.

But I think the reg could be clearer and that such discussion is helpful as a result. I also don't think that a 65+5 testing age makes any sense for DC plan cross testing, but it is what it is.*

Thanks.



* better stated as "It says what you say it says"

Tom,

Thank you very much!!! You were right on the money. It had to do with the Date or Retirement. It was coded 1st of month during. It made for a non-unioform NRA and that was the problem. I chaged it to 1st of month following and it worked.

Thanks again!!!

And you see how easy it is to create a non-uniform NRA so that you can use 65 as your testing age.