Aggregation of Plan for Top Heavy

[Lori Foresz]

Plans must be aggregated for top heavy testing if at least one key EE participates in both plans. But what does participate really mean? Does a KEY EE have to BENEFIT under the plan during the year or just have a balance in the plan. Sal Tripodi's book seems to say a KEY EE only need have a balance in the plan for the plan to be aggregated with another plan covering a KEY EE. So, a frozen plan in which a key EE has a balance would still be aggregated with an active plan that covers other KEY EEs and non-KEY. I guess this makes sense- I just want to be sure.

thanks!

[Mike Preston]

You can take what Sal says to the bank. [Or, as may be the case this year, to the government which has taken over said bank.]

[Guest Sieve]

As an aside, Mike, I assume you mean that Lori can take Tripodi to the bank (or government!!) on this issue, rather than that you can always take Tripodi to the bank. I think Sal is great, and covers many issues not addressed elsewhere, and gives citations to support his conclusions--but I, at least, do not agree with him all the time.

[RobN]

What if the Employer sponsors 2 401(k) Plans, with no Key Employees in one of the Plans. They do not have to be aggregated for Top Heavy, so the "Non-Key Employee Plan" does not have to provide Top Heavy minimum contributions.

Can you aggregate the Plans for ADP testing?

[Guest Sieve]

Lack of keys does not get you out from under the potential of required TH aggregation: if plan #2 (no keys) is permissibly aggregated with plan #1 in order for plan #1 to pass minimum coverage (IRC Section 410(b)), then the plans must be aggregated for top heavy purposes (even if one plan has no key employees) and a top heavy minimum may then be required.

Aside from that, however, I think you can aggregate plans #1 & #2 for ADP purposes without being required to aggregate for TH purposes (because required TH aggregation does not encompass plans which are permissively aggregated for ADP testing pusposes--only 410(b) & 401(a)(4)).

[Blinky the 3-eyed Fish]

Sieve, can you enlighten me on your last paragraph? I thought that aggregating for ADP purposes would be aggregation under 401(a)(4) and correspondingly under 410(b), which would cause both plans to be in a required aggregation group.

[Guest Sieve]

Blinky --

Here's my thinking . . .

Aggregation of plans for TH purposes is required (i) for each plan in which there is a key employee, and (ii) for each plan "which enables any plan [with a key employee] to meet the requirements of section 401(a)(4) or 410." (IRC Section 416(g)(2)(A)(i).)

In this situation, one plan does not have any keys, so (i) is not met. Now, let's assume that each plan separately passes 410(b), and therefore they are not being combined in order to enable the plan with keys to pass 410(b). (Note that it is not the combining of plans under 410(b) that is critical, but whether they are permissively aggregated specifically in order to cause a plan with keys to pass 410(b).)

So, the remaining question, then, is whether combining the plans for ADP testing purposes, specifically to enable one of the plans to pass ADP, is, in effect, nothing more than permissive aggregation in order to enable a plan with keys to pass 401(a)(4).

Well, of course, if the plan with keys is not the one that is failing ADP (i.e., the keys are not HCEs), then the permissive aggregation of the plans is not accomplished to enable the plan with keys to pass anything--the permissive aggregation is used for the plan without keys to pass something. Therefore, there would be no required TH aggregation (under (ii), above). (By the way, this is consistent with Tripodi (2008), Chapter I, Part B.1.a.2.)

What if, however, the permissive aggregation is in order to enable the plan with keys to pass an otherwise failed ADP?

An ADP test certainly is a non-disrimination test, but it is not a 401(a)(4) test. The regs do not say that a 401(k) plan must pass 401(a)(4) and that passing ADP relieves the 401(k) plan from passing 401(a)(4), or that a plan can pass 401(a)(4) by passing ADP. Rather, because a 401(k) plan must pass a different non-discrimination test, "A section 401(k) plan is deemed to satisfy [iRC Section 401(a)(4)] . . ." (Treas. Reg. Section 401(a)(4)-1(b)(2)(ii)(B)). So, the permissive aggregation for ADP purposes is not undertaken to cause another plan (with keys) to pass 401(a)(4).

Therefore, in my mind, permissively aggregating plans for ADP purposes does not cause them to be considered combined for 401(a)(4) purposes, so there is no required aggregation for TH purposes.

Have you seen something--or do you believe--to the contrary?

[Blinky the 3-eyed Fish]

Sieve, I finally had a chance to look at this. I agree with your analysis.

[jpod]

I like Sieve's analysis too, but do we know if the IRS likes it? If there is any doubt, query whether it is worth it to aggregate for passage of ADP on that basis while at the same time risk blowing the non-keys plan's shelter from the top-heavy minimum.

[Guest Sieve]

It is clear that if the plan failing ADP is NOT the one with keys (i.e., the plan with keys passes ADP on its own), then combining the plans for ADP will not require TH aggregation since aggregation only results when plans are combined to permit a plan with keys to pass 410(b) or 401(a)(4). (IRC Section 416(g)(2)(A)(i)(II).) That, I assume, we can agree on.

[jpod]

Yes, I agree that that part of the analysis is perfectly clear on the face of the statute.

As to the other part of the analysis, ????? Consider this scenario. Elective deferral-only 401k plan covering only associates in a law firm. The associates are all or virtually all NHCEs (because of the top 20% election). The associates' ADP is higher than the ADP of NHCEs in the other plan covering partners and non-attorney employees, and if aggregated both plans would pass ADP. If, however, IRS does not agree with your analysis and treats the two plans as a req. agg. group, you have to give 3% of comp. to all the associates.

Notwithstandnig your excellent analysis, I think I'd want to see something by IRS blessing it before I take the risk, because the liability for that 3% of comp. for, let's say, 80 associates making an average of $150,000, would be $360,000. Compare that to making some relatively small distributiions to some partners to bring down the HCE ADP in the partner plan. You are weighing the annoyance of causing some partners to lose a small portion of their tax deductions for the year against the risk of a $360,000 unbudgeted expense.