Top heavy in frozen DB plan

[Belgarath]

I'm not a DB person, so pardon my terminology if it isn't quite accurate.

With the EGTRRA change, you no longer count service after 2001 for additional TH accruals if the plan is frozen. However, it seems to me that there's no mention of freezing the benefit - if your average comp goes up, then you have the same TH percentage, but of a higher salary.

For example - prior to a plan freeze effective 1-1-2002, participant has accrued a TH benefit of 6% of a high-5 average salary of 20,000, or 1,200. Three years later, with the high-5 average being 30,000, is his TH benefit:

A. still 1,200, or

B. 1,800?

I'd vote for B, as I see nothing in the statute or regs which indicates otherwise.

Additional twist - suppose as of 1-1-02 the client adopted a PS plan that specified that the PS plan would provide TH benefits of 5% if required, but for 2004 there is no contribution to the PS plan. If the answer to the DB question is (B), then does this reqire a contribution to the PS plan? I guess that depends on whether the increase to 1,800 is considered "accruing" an additional TH benefit? Anybody wrestled with this question yet? My inclination is no contribution to the PS plan, as no HC is accruing a benefit in the DB plan.

[mwyatt]

Hey Belgarath:

As long as the plan continues to be top heavy, you include the compensation (you just aren't adding service) so I vote B. So in effect you have a "soft" freeze for non-keys, while a "hard" freeze for keys.

Went through this issue a little bit as far as TH minimums go under a DC plan operating with a frozen DB plan. Couple of points that came up in discussions and research with TAG were that your DC minimum operates at the 3%, not 5% level. Second, the 3% could drop if the keys' highest percentage was under 3%, so a $0 contemplated PS contribution wouldn't run you afoul of the TH minimum (although be careful if you have key 401(k) deferrals in the mix).

[Belgarath]

Thanks!

[Blinky the 3-eyed Fish]

I don't agree with the thought that the average compensation must continue to factor into the TH benefit for years beginning in 2002 when a plan is frozen. Here are 416©(1)©(iii) and 416©(1)(D):

(iii) Exception For Plan Under Which No Key Employee (or Former Key Employee) Benefits For Plan Year.--

For purposes of determining an employee's years of service with the employer, any service with the employer shall be disregarded to the extent that such service occurs during a plan year when the plan benefits (within the meaning of section 410(b)) no key employee or former key employee.

(D) Average Compensation For High 5 Years

For purposes of this paragraph--

(i) In General

A participant's testing period shall be the period of consecutive years (not exceeding 5) during which the participant had the greatest aggregate compensation from the employer.

(ii) Year Must Be Included In Year Of Service

The years taken into account under clause (i) shall be properly adjusted for years not included in a year of service.

It seems to me that because the year of service is not included for benefits purposes, that translates to not including it for average compensation purposes.

[Belgarath]

Hi Blinky - I understand your feeling - in fact, I started out with the same interpretation. But after reviewing it for a while, it seemed to me that 416©(i) is saying that except as provided in clause (ii) or (iii) of that paragraph ©, YOS are determined in accordance with (4), (5), and (6) of 411(a). And the purpose of (ii) and (iii) is to ignore those years when no HC benefits - which includes a frozen plan. But paragraph (D) of 416 seems to stand alone - that is, the modifications to "years" that are made in © do not modify 411 for purposes of paragraph (D). If (D) had specifically excluded years excluded under ©(iii) then I'd feel a lot happier. But since it doesn't, I reluctantly arrived at my current position.

Does anyone know if the IRS has addressed this from the podium at an ASPPA conference, or some other such venue? The consequences of interpreting this "incorrectly" (from the viewpoint of the IRS - I'm not sure that either interpretation is either completely correct or incorrect) - could be fairly significant with average comp rising over time.

[Blinky the 3-eyed Fish]

They did address it in LA last week, but I honestly don't recall the response offhand and my notes are at home. I guess the answer wasn't that compelling to me to sink in. SoCal was there so perhaps he can chime in.

I agree this is less than clear. As for your interpretation, take a look at this cite. If (D) stood alone, why would the regs exclude a YOS for average compensation when a person didn't meet the 411(a) requirements cited? My thought is that because the statement in 416©(1)(D)(ii) is not just excluding years defined in 416©(1)(D)(iii). Look at one of your documents and I bet it excludes years in which the person didn't earn meet the accrual requirements. IMHO, the same logic would apply to a frozen plan.

1.416-1 M A-2©

© In determining the average annual compensation for a period of consecutive years during which the employee had the largest aggregate compensation, years for which the employee did not earn a year of service under the rules of section 411(a) (4), (5), and (6) are to be disregarded. Thus, if an employee has received compensation from the employer during years one two, and three, and for each of these years the employee earned a year of service, then the employee's average annual compensation is determined by dividing the employee's aggregate compensation for these three years by three. If the employee fails to earn a year of service in the next year, but does earn a year of service in the fifth year, the employee's average annual compensation is calculated by dividing the employee's aggregate compensation for years one, two, three, and five by four. The compensation required to be taken into account is the compensation described in Question and Answer T-21. In addition, compensation received for years ending in plan years beginning before January 1, 1984, and compensation received for years beginning after the close of the last plan year in which the plan is top-heavy may be disregarded.

[Belgarath]

Blinky, thank you for the response. I sincerely hope the IRS (if SoCal has info on this?) agrees with your interpretation! I just get a different result from the same cites that you do. I read the reg merely to say that you don't count a YOS toward the % increase (so in my original example, you remain at 6% rather than moving to 8%) but that the averaging period only excludes years that do not count as a year under 411.

I like your interpretation better, and I look forward to hearing what the IRS had to say on this last week, and hopefully being all wet in my thought process on this garbage. Thanks again.

[could be me maybe not]

Blinky, go home and get those notes! Geez! I sure hope you are right.

[SoCalActuary]

Holland mentioned the issue of a "soft-freeze", where the non-keys could get an increase in their TH min with new salaries. If so, then you don't have a requirement to do a change in funding method.

But I do not recall any statement that you must do a soft-freeze post-EGTRRA. I stand with Blinky on his interpretation that you can ignore compensation in years where no top-heavy accrual occurs. Thus, a freeze becomes a hard freeze if properly documented.

[Belgarath]

Hoo Yah!

May the wind under your wings bear you to where the sun sails and the moon walks.

Thank you both.

[mwyatt]

Well, if you look at most documents (my experience primarily rests with Corbel), you will see that compensation is ignored (or can be ignored) prior to the establishment of the top heavy rules or after the last year in which the plan was top heavy. I too wish that a pre '02 frozen plan could ignore compensation after 2001, but that's not what I'm reading (especially fun in that I've had some plans that were previously not top heavy that popped back up due to the movement from 5 year back to 1 year back on terminations). Blink, I'd be happy to be wrong in this case...

[Guest dsyrett]

I recall hearing that Jim Holland said at a meeting within the last year or so that a "hard freeze" was possible (whereby TH comp would not increase).

[mwyatt]

Let's forget about our existing document language here (which to tell the truth is irrelevant to our discussion since noone has an EGTRRA document at this point). Either IRC 416 states that if the plan continues to be top heavy, you have to include compensation in the average for calculating the top heavy minimum, or it doesn't. We know that EGTRRA provides that you don't have to credit service after 2001 if benefits have been frozen; however, the existing language in IRC 416 and regs states that your averaging period has to include years while the plan is top heavy (the pre '84 stuff is somewhat irrelevant at this point).

I really don't see how Holland can say that you "could" have a hard freeze without some future guidance on how 416 will be interpreted.

[Blinky the 3-eyed Fish]

however, the existing language in IRC 416 and regs states that your averaging period has to include years while the plan is top heavy (the pre '84 stuff is somewhat irrelevant at this point).

That's just not the case though as I spelled out in the cites I quoted. You most definitely do not include years in the averaging period for which the person was not credited with a year of service. That is the mechanism for which you can logically argue that you don't include compensation in years for which the plan is frozen.

[mwyatt]

Actually Blink, that more takes into account the situation in an ongoing TH plan where the non-key is credited with less than 1,000 Hours of Service (so you don't beat down the average further).

[Blinky the 3-eyed Fish]

I realize that is the purpose of the reg, but consider that when that reg was written, EGTRRA didn't exist. It doesn't negate the argument I presented, nor does it negate the fact that Holland too has an opinion without needing to issue further guidance.

Speaking of purposes, what was the purpose of the EGTRRA change eliminating the TH minimum in a frozen DB plan? Was it to mirror DC plans in that a benefit need not be provided if no key benefits? Was it to keep DB plans from terminating, by offering a freeze option and no new accruals? I don't know the behind the scenes info (maybe MGB can chime in), but I have to think that it was one of the two or both. If either, then the intent has to be that compensation is frozen as well.

[mwyatt]

Het Blink:

I hope you understand that I'm rooting that you're right and I'm wrong about this! However, until we get some clarification with some post-EGTRRA updates to the 416 regs, who knows what the answer is.

I've seen too many plans over the years where not a very big benefit was offered that got thrown into horrific underfunding due to the TH minimum..

[Blinky the 3-eyed Fish]

Yes, I know we are on the same side. I guess it just boils down to me thinking there is a strong case for being able to not including the compensation post EGTRRA based on existing regs and you thinking differently. I guess time will tell, but in the meantime I feel most comfortable operating plans accordingly.

[mwyatt]

Hey Blink:

Hate to be a pest, but I haven't run through anything on RIA Checkpoint or TAG that supports blowing off post 2001 TH years for inclusion of compensation averaging accrual purposes...

[Guest merlin]

mwyatt,

I put the question to TAG after reading the earlier posts. They agree that if you don't count service you don't count the comp. No cites, tho'. But since their is supported by Blinky's cite I didn't press them on it.

[mwyatt]

Thanks Merlin and Blinky. If this is indeed the case, this is great news, as I have a previously frozen plan that actually wasn't top heavy but came back as TH when the five-year lookback went to one-year. Will be very glad to revise this val!

[Guest merlin]

Like Blinky I always thought that the EGTRRA change was meant to put frozen db plans on the same footing as terminated plans. Then your post, and "Oy Veh"! But Blinky swam to the rescue as usual. It pains me to think of eating so many of his relatives on Friday as a child. Mea culpa.