"Dual Eligibility" Issue

[Guest Jeff Underwood]

We have a client who wants to set up a plan (employer contributions only) effective as of the beginning of the year. We proposed standard eligibility requirements of 1 year of service and age 21 with an eligibility requirement waiver for anyone actively employed as of the effective date of the plan. The individual preparing the document says that we have a problem due to the "dual eligibility" provisions of 1.401(b)-6(b)(2). On pages 2.6 and 8.97 of the ERISA Outline Book, Sal Tripodi indicates that such waivers can potentially cause coverage problems, but does not provide a clear explanation as to how. I point out that such eligibility waivers are included in standardized prototype plans, which are supposedly designed to always pass coverage. Any ideas?

[Blinky the 3-eyed Fish]

I got a learnin' on this one a few months back, so I will try and explain. Basically, you need to run the 410(b) test based on the lowest eligibility requirements in effect for the plan year. In your case, there are no eligibility requirement, since you are making everyone employed on the adoption date eligible.

So your concern is those that terminated employment during the year. Let's do an example .

HCE's employed on adoption date and also benefiting: 2

NHCE's employed on adoption date and also benefiting: 10

Terminees during the year (all NHCE's): 20

Coverage test: 10/30 / 2/2 = 33.33% - could be a problem

Of course you could test using otherwise excludables and if no HCE's are in that group then you pass at 100%. So in reality, you really only have an issue if an HCE is being let in by the more generous eligibility provisions.

[Guest Jeff Underwood]

Here is my census for the year:

2 Owners employed at the beginning of the year;

1 NHCE employed at the beginning of the year; and

1 NHCE hired during the year.

All three individuals employed at the beginning of the year received contributions; the individual hired during the year did not.

If I understand your explanation, then my coverage ratio would be 50%. I don't believe that the "otherwise excludable" helps me because this is a new company. That means there is no prior service to count for the HCEs to move them out of the "otherwise excludable" group (right?).

[Blinky the 3-eyed Fish]

You say "received contributions" like it already happened, but your first post indicates the plan has not been set up yet. I will assume the latest post is a misuse of the proper tense and the plan has not indeed been adopted yet.

Your problem will then be solved, not by allowing for immediate eligibility this year, but by having say a 9 month eligibility requirement and entering the plan immediately. Then your one NHCE is ineligible for the year and your coverage ratio is 100% and all is happy and go-lucky.

[Guest Jeff Underwood]

It appears from your post that my testing assumptions (if not my tense) were correct - the "ineligible" NHCE would get pulled into coverage testing and I would not be able to use the "otherwise excludable" exception to my advantage.

If the eligibility requirement waiver can cause problems like this, why is it allowed as a standardized prototype option?

Also, if we set up eligibility provisions such as 9 months + immediate entry, how do we amend them to more conventional terms without having the same "dual eligibility" issues?

[Blinky the 3-eyed Fish]

It appears from your post that my testing assumptions (if not my tense) were correct - the "ineligible" NHCE would get pulled into coverage testing and I would not be able to use the "otherwise excludable" exception to my advantage.

It was correct. Otherwise excludable testing would not work because everyone would be in the OE category, which too has to be tested.

If the eligibility requirement waiver can cause problems like this, why is it allowed as a standardized prototype option?

I didn't know it was, but if so in your prototype, well I can't answer why. At least you recognized there was a problem.

Also, if we set up eligibility provisions such as 9 months + immediate entry, how do we amend them to more conventional terms without having the same "dual eligibility" issues?

Well each year is a separate testing issue. For the next year the only eligibility that would apply would be the 21/1. The 9 month eligibility would only be for the first year. No dual issues any longer.

[rcline46]

I think Mr. Tripodi is using a too conservative interpretation. The reg and the examples are all of concurrent eligibilty. That is two plans or two groups in one plan with different eligibility.

The situation described is consecutive eligibility. One on the first day of the year, and a second one from the second day forward. Anyone hired after the first day can in no circumstance (by changing groups or plans) acheive a different result. Therefore you do not have what is termed 'dual eligibility'. That is why it works in standardized prototypes.

In my opinion of course.

[Blinky the 3-eyed Fish]

To go into more detail a previous discussion is here:

http://benefitslink.com/boards/index.php?s...hl=feeling&st=0

I would say you are in the extreme minority in your opinion.

[rcline46]

Maybe so, maybe not too many have thought about it!

However, there is no refutation of my position (ok, maybe I am stubborn!) of concurrent vs consecutive. Has anyone tried to get a explanation from Mr. Tripodi?

Why has the IRS approved Standardized prototypes with this feature since under the design of the plan you cannot fail 401(b) by definition? But Mr. Tripodi's position is that you can fail.

THis alone would indicate the IRS agrees with me on this situation.

[AndyH]

I am unconvinced of this as well. Why does the plan have two sets of eligibility requirements for the year? I see it as one, 1 hour of service as of 1/1. Then you have a plan amendment effective in the second year.

Now I could see an argument for a 401(a)(4) violation on the grounds of discriminatory timing of an amendment, but I'm not getting the "most liberal eligibility" argument. That pertains to the ABPT when there are more than one plan. I don't see how it applies here.

Blinky, I'll accept a learnin, but it looks like a hard subject to teach.

p.s. Is the issue that there are two entry dates and the waiver only applied for the first? Then ya might have an issue. Absent that I don't get it.

[Tom Poje]

While the concept 'smells' of lending itself to being discriminatory, the following from Reish Luftman % Reicher's IRS technical tips (which is from an ASPA Q and A)

would support the idea that it is nondiscriminatory. I love the wording of the response "I do not think..." Does that mean "...but maybe we missed this one?"

Split Eligibility: Is it Discriminatory?

(Posted November 22, 2000)

Technical Tip 37: The following question and answer were from the IRS Q&A Session at the 1999 ASPPA Annual Conference:

Reg 1.401(a)(4)-5(a) says that the timing of a plan amendment or series of amendments must not be discriminatory, and that the establishment or termination of a plan is considered to be an amendment for this purpose.

If a new plan provides for immediate entry for those employed on a certain date, and a waiting period for those hired after that date, might this be considered a discriminatory amendment?

IRS Response: This provision is often seen in standardized prototypes which are supposed to be nondiscriminatory by design. I do not think present and future eligibility would have been approved in standardized prototype plans if this were not acceptable.

We do not see this eligibility provision as a discriminatory amendment under the -5 rules.

[Blinky the 3-eyed Fish]

It seems like we are talking about 2 different things here, 1) whether or not the change in amendment is discriminatory and 2) how the coverage testing should be done.

1) Addressed by Tom above. Fine and dandy.

2) Whether or not you classify it as consecutive or concurrent or dual eligibility or whatever, how do you do the coverage testing?

If you have a provision that allows for all those employed as of X date, then effectively you have no eligibility requirement for those people. The coverage testing then for that year should be based on that eligibility requirement (or lack thereof). If someone terminated during the year, then how can you treat them as an excludable employee for the coverage testing? Of course there is the otherwise excludable option, but if an HCE is in that group, you still need to test it. Now what troubles me is the fact that prototypes allows this to occur, so maybe I am missing something.

[AndyH]

Well, I see the eligibility requirement as 1 second of service as of 1/1 and you see it as none, so there is a difference in perspective there. As to terminees, if they worked one second as of 1/1 then they are not excludable; otherwise they are, or so it would seem under my argument.

[Blinky the 3-eyed Fish]

I am not sure I follow. All I am saying is that you are testing on the lowest age and service requirements in the plan year, whether you define that as none or 1 second shouldn't matter.

[AndyH]

If there are two participants, one hired December 31 and one hired January 2. Both terminate during the year and get no benefit. I say one is nonexcludable and non benefitting, the other is excluded and excludable. Aren't you saying that both are nonexcludable and nonbenefitting?

[Blinky the 3-eyed Fish]

I should let this die but...

First, Andy, yes I would say both are nonexcludable. There has been discussion of concurrent versus consecutive by rcline. Now apply that to a plan with no eligibility requirements for those employed as of the effective date and 21/1 for all else.

Rcline, you say

The reg and the examples are all of concurrent eligibilty. That is two plans or two groups in one plan with different eligibility.

. I would say that it too is concurrent eligibility in that we are testing a plan year and in that plan year there are 2 sets of eligibility requirements that are applicable to 2 different groups, those employed on the effective date and those not. Again, because we are testing a plan year and because 2 separate eligibility requirements apply to that plan year, my thought is that it treated the same as the examples in the regs.

[rcline46]

Ahhh Blinky, guess we had better ring up Sal? We disagree, but that's ok. Until a field agent decides to make an issue of it and we get a definitive ruling we can do our own thing.

Observers can make up their own minds as to which interpretation to take. And of course if you depend soley on your software to do the testing.......

YOU might be a redn.... er ..... surprised!