Offset DB plans

[Blinky the 3-eyed Fish]

This might have been asked before, but I can't find it so...

In attempting to pass 401(a)(26), what would preclude a DB offset plan being set up where the owners' benefits are not subject to being offset, but the ancillary employees' benefits are? Note that all of the ancillaries' DB amounts would be completely offset by their DC nonelective balances.

[AndyH]

I'm glad you raised this because I've been meaning to look it up. It appears that this would qualify as a concurrent offset arrangement and it appears that 1.401(a)(26)-5 requires that such an arrangement provide benefits to employees on a "reasonably uniform basis". I think thats the problem with doing as you propose. Please let me know if you agree after reading that.

[Dougsbpc]

What is the definition of uniform and reasonable?

Although not certain, we contend that a safe harbor plan would be considered as providing benefits on a uniform and reasonable basis. Owner employees need not benefit for a plan to be a safe harbor plan. Therefore, as long as all participants other than the owners benefit on a uniform basis, you may satisfy 1.401(a)(26)-5.

[AndyH]

Dougsbpc, how would it to be uniform and reasonable if the owner is the only one who does not get offset? I don't think Blinky is saying he is out of the DC, just that the DC does not reduce the DB for him.

[Blinky the 3-eyed Fish]

Andy, I have read and re-read this reg many times. 1.401(a)(26)-5(a)(iii)(A)(2) is the specific passage you reference and it states:

"The employees who benefit under the formula being tested also benefit under the other plan on a reasonable and uniform basis; and"

I read that the other plan in this case is the DC plan and the people being tested are those DB participants subject to the offset. It seems to me that the owners are not subject to the offset and do not fit into this criteria. They can benefit 401(a)(26) on the DB alone.

I know it's debatable whether the DC plan has to be a safe harbor or not, but I am not trying to address that question.

Yes Andy, that is what I am saying.

[AndyH]

Blinky, for the benefit of the class, not me of course, are the owners in the DC plan or not? I'm still not sure of the specifics of your question.

Are they (1) not in the DC or (2) in the DC but not subject to the offset in the DB in your question?

It seems to me that (1) is ok but (2) is not. I take it you don't agree about (2).

And to Doug, I don't follow your comments about safe harbor. Because two plans are safe harbors certainly doesn't mean that they are uniform. If you have a safe harbor DB plan covering everybody and a safe harbor DC plan covering one NHCE and the DC plan offsets the DB, is that uniform?

[Blinky the 3-eyed Fish]

It's (2). They are in both plans, but the benefit for them in the DB is not offset by the DC, but is for everyone else.

Is your first post why you don't agree that (2) is okay? If so, perhaps you could rebut my comments as to the meaning of the cite.

BTW, this is purely a hypothetical situation. I have no plans designed such as this nor have I seen any ever designed this way, but it sure would help if it was possible.

[AndyH]

My read is that nobody in an offset plan that does not accrue a net benefit can be counted as benefitting unless the requirements for either concurrent or sequential offset are met. I think that is a precondition according to the following:

"An employee is treated as accruing a benefit under a plan that

includes an offset or reduction of benefits that satisfies either

paragraph (a)(2)(ii) or (a)(2)(iii) of this section if either the

employee accrues a benefit under the plan for the year, or the employee

would have accrued a benefit if the offset or reduction portion of the

benefit formula were disregarded.."

And the requirements of a concurrent offset include the following:

(iii) Concurrent benefit offset arrangements--(A) General rule. An

offset or reduction of benefits under a defined benefit plan satisfies

the requirements of this paragraph (a)(2)(iii) if the benefit formula

provides a benefit that is offset or reduced by contributions or

benefits under another plan that is maintained by the same employer and

the following additional requirements are met:

(1) The contributions or benefits under a plan that are used to

offset or reduce the benefits under the positive portion of the fomu1a

being tested accrued under such other plan;

(2) The employees who benefit under the formula being tested also

benefit under the other plan on a reasonable and uniform basis; and

I think your example violates (2) which is a precondition to having anybody with a $0 net benefit being counted as benefitting.

I agree that the people with a net benefit are benefitting and the people with DC accounts benefit due to aggregation but I'm focusing on 401(a)(26), not 410(b). How would you propose to pass 401(a)(26) since we cannot aggregate for that purpose? If you fail the above you've only got the owner benefitting under 401(a)(26), right?

And, yes, I understand it's hypothetical. I've seen a few offset candidates recently but don't have any actual offset plans now so I've been meaning to brush up.

[Blinky the 3-eyed Fish]

Andy, I agree that this language is designed to allow for a person with a net benefit of $0 to be considered benefiting for 401(a)(26) purposes. You say my arrangement fails (2), but why? Here is how I read (2) to mean.

(2) The employees who benefit under the formula being tested (the formula being tested is the DB formula here; is it not?) also benefit under the other plan on a reasonable and uniform basis (the other plan is the DC plan, which the participants in the DB plan do benefit on a reasonable and uniform basis) ; and

Where have I gone wrong in your opinion Andy? I don't see how my arrangement violates this cite.

[AndyH]

Well, having read some more stuff on this, in particular an outline done by Norm Levinrad #15 at the ASPA National conference in 2001, I think you are right.

I think you've got a non-safe harbor floor offset but since the offset is uniform you are ok under 401(a)(26). Sorry for the confusion. Like I said, I was a little rusty on this and this was very useful to me.

[Mike Preston]

Can you share snippets of the outline that are pertinent to the determination? Also, can you recap the plan design that you have come to the conclusion satisfies a26?

[AndyH]

Mike, send me an email with your email address and I'll send it to you. I know you've had a few of them. I just tried sending it to you through Benefitslink email and I didn't seem to be able to attach it.

The outline doesn't say you can do what Blinky wants. It outlines the requirements that an offset must meet, and Blinky's design passes all of them. And at the end of the outline there's a note about 401(a)(26), and it is consistent with Blinky's interpretation.

[Mike Preston]

The outline only has one reference to a26 that I could find:

"There could be a concern that the results after the offset do not satisfy 401(a)(26) which require that the benefits are offset by contributions which are reasonable and uniform. One could avoid this problem by a variation to the design to have all participants covered in both plans, with limits to the benefits and contributions."

With that said, can you recap the plan design that you have come to the conclusion satisfies a26?

[Blinky the 3-eyed Fish]

Mike I assume you are talking to Andy. Anyway, I would be interested in your opinion how my design as it pertains to 401(a)(26).

[Mike Preston]

Actually, I'm trying to figure out what the design *IS*! Not sure I will comment (sometimes my hands are tied because of current engagements), but I would really like to know what the design is that I thought you and Andy had settled on.

[Blinky the 3-eyed Fish]

The hypothetical design is as such (I added some specifics to assist in your overall fun):

1. DB plan with a 10% career average pay formula for the owner and .5% to everyone else. The .5% is offset by the AE of the PS balance, while the 10% is not offset at all.

2. DC plan where everyone gets a pro rata allocation (like I said, I am not trying to go into whether a cross-tested allocation is appropriate here).

Both plans are new and cover the same employees.

[Mike Preston]

And I presume your theory is that if the owner does not receive a contribution in the DC plan then you must consider those as benefitting only those that have a positive accrual after consideration of the offset so that you would only have the the owner benefitting (which of course would therefore fail a26).

In either event I find it strangely discomfiting that by giving the owner less in the DC plan the result would be that the DB plan fails 401(a)(26).

[AndyH]

Sorry, I've been away a bit doing the tax thing. Kind of hard to put that off on 4/15.

I was questioning what had to be reasonable and uniform (as Doug also did in his initial comments), whether it was the DB plan, the DC plan, or the relationship between the two. Clearly it is not the DB plan or the combined plans because a safe harbor floor/offset can have one plan be a general tested plan provided that the other is a safe harbor and the same employees benefit in each plan.

My conclusion is that the uniformity requirement applies only to the PS plan and that Blinky's example satisfies that. And therefore it satisfies the condition for the concurrent offset program outlined in the regulation that I cited.

Levinrad's outline did contain only one paragraph on 401(a)(26) but I thought that was telling in and of itself. And I believe that the one paragraph supported the interpretation that only the PS needs to be uniform and reasonable.

Mike, now you can do a more complete Floor/Offset outline for us and we'll shut up.

[Mike Preston]

In this thread, I'm just trying to gather information.

I am not aware of any case where the IRS has stomped on somebody for a 401(a)(26) failure in the situation you describe. To go even further, I'm not aware of any case where the IRS has stomped on somebody for a 401(a)(26) failure when the situation is similar to what you have described, where the only difference being that, while the DC plan is a safe-harbor as far as 401(a)(4) is concerend should it be viewed by itself, it provides a lower benefit (or none at all) to selected HCE's.

I've heard that there are some that take the position that a safe-harbor DC plan is all that is necessary to satisfy the 401(a)(26) rules.

I think it would be good if the IRS would clarify this at some point. Further, if they take the position that a safe-harbor DC plan is not, by definition, a plan that satisfies the rules under 401(a)(26) as an offset plan such that the determination under 401(a)(26) can be made before application of the offset, I would hope that they would provide some sort of rule that says: "those that have done this up till now are ok, but they can't do this for plan years beginning after xx/xx/xx". Where xx/xx/xx is some date after the guidance is issued.

[Blinky the 3-eyed Fish]

Mike, just to be clear from my perspective is that I am not proposing giving less to the owner in the DC plan. Besides in my hypothetical the DC contribution to the owner has no bearing on his DB amount.

[Mike Preston]

I think I am now clear on the design you are describing.

It seems to me that if the design "works" as far as 401(a)(26) is concerned, that it should also work if the sole difference is that the HCE/Owner receives **LESS** than he otherwise would.

Maybe this is why we haven't heard of the IRS attacking any offset plans for failure under 401(a)(26)?

[Guest Dave Peckham]

Mike, it is not clear from your comments that you appreciate what Blinky is trying to do: have a DB offset plan where the 10% career average pay formula for the owner is NOT offset by his or her (say, 7.5%) allocation in the DC plan, whereas everyone else's .5% career average pay formula in the DB plan IS offset by his or her (say, 7.5%) allocation in the DC plan.

Doug and I have never seen this plan design, nor has Blinky, nor has Larry Deutsch brought it up in his nondiscrimination seminars (although I am going to fax him this discussion to see if he will comment).

We are all aware of Norm Levinrad's outline showing the HCE getting a 0% allocation in the DC plan. That's not what this thread is about.

If it turns out that Blinky's is a valid plan design, then of course we will have succeeded in allowing the HCE benefits to be HIGHER than otherwise allowable.

Dave Peckham

[AndyH]

It still must pass the general test, so I'm not sure how we are giving HCEs anything higher than otherwise allowable. And you'll probably need to aggregate under the general test, so you've got the db/dc gateways to consider.

[Guest jgordon]

I have designed offsets such as the one blinky describes. I have favorable letters on the design. The only attack I received from the IRS under 401(a)(26) was whether the NHCEs in the DB plan received a significant minimum benefit. The Plans were amended to provide that the benefit which the NHCEs would be either the greater of a .5% of compensation annuity or the accrued benefits in the PS plan. That was accepted.

There was no attack on the reasonable & uniform (because the benefits in DC Plan - the offset- are reasonable & uniform).

We do need to aggergate under the general test and therefore do have to deal with the DB/DC gateways (usually 7.5% or broadly available plans exception to 7.5% gateway).