Correction to failed 414s

[pixmax]

What is the correction if a Plan is failing 414s and the Plan is a Safe Harbor Match? They are not allocating a PS contribution but they do exclude bonuses as their compensation definition for the Safe Harbor Match. They do not have a bonus deferral election.

[Guest Sieve]

Excluding bonuses from the definition of compensation for purposes of a SH match is not necessarily a "violation" of 414(s), since the regs under IRC Section 414(s) permit any definition of compensation to be used if it "does not by design favor highly compensated employees [HCEs], is reasonable . . . and satisfies the nondiscrimination requirement [i.e., cannot discriminate in favor of HCEs] . . ." (Treas. Reg. Section 1.414(s)-1(d)(1).)

A definition excluding bonuses is "reasonable" under the regulations (Treas. Reg. Section 1.414(s)-1(d)(2)(ii)). The issue, therefore, is whether a definition excluding bonuses is non-discriminatory. I can't say for sure--since a test must be performed, as described in Treas. Reg. Sectin 1.414(s)-1(d)(3))--but, in my exerience, bonuses tend to be a higher percentage of compensation for HCEs than for NHCEs, so a definition of compensation which excludes bonuses would generally discriminate against HCEs and therefore would be permissible.

[Kevin C]

Sieve,

How about if the HCE's are above the compensation limit without counting the bonuses? 1.414(s)-1(d)(3)(ii)(A) says total compensation used in the test may not exceed the annual compensation limit of section 401(a)(17).

pixmax,

If you are failing the 414(s) compensation test, you are right there is a problem. The safe harbor rules require that the compensation used to determine the SH contribution satisfy section 414(s). The SH rules also require the SH match provisions to be adopted before the first day of the plan year and remain in effect for the entire year.

In looking at Rev. Proc 2008-50, I doubt your problem would be considered an operational failure because you should be following the terms of the plan. Self correction is only available for operational failures. Even if it is an operational failure, I think the correction would need to be done through a plan amendment. Correction by amendment in SCP is limited to only certain areas listed in the Rev. Proc. I don't see any of them that looks like it would apply here. That leaves VCP.

I wonder how the IRS would react to a VCP filing with a proposed correction method of a retroactive amendment changing the compensation definition to something that satisfies 414(s)? I would expect them to insist that the correction does not reduce anyone's match.

[BG5150]

I don't think you'd have to do a VCP, especially if the entire contribution hasn't been funded yet (I think the SHMAC needs to be funded at least quarterly, and I would think you'd have time to do the 4th quarter).

But correcting for a match is an interesting problem. For a PS, you'd just come up with comp that would satisfy 414 and allocate that way. But you can't for a match, since the people didn't (obviously) defer on all of the comp that would satisfy 414.

I am interested in the resolution to this as well.

[Guest Sieve]

Kevin -- By limiting comp in the 414(s) test to the annual 401(a)(17) amount, you are correct that the very highest paid e'ees will be at 100%, even after excluding bonuses (since their comp will exceed the limit both with & without the bonus), and that might raise the avg of the HCEs as a group--which certainly reiterates the need to do the test with real numbers rather than speculate about its result.

I suspect that you'd correct a discrimination failure based on 414(s) as if it were a 401(a)(4) operational discrimination failure (based on the wording of Treas. Reg. Section 1.414(s)-1(d)(3)(iii)). That brings you into the correction procedures of 1.401(a)(4)-11(g), where you'd amend the plan to change the definition of compensation (as generally described, e.g., in 1.401(a)(4)-11(g)(3)(iii)) and the employer would make additional contributions accordingly. EPCRS would come into play only after the correction time period in the regs had expired, and then I think you'd first come under self-correction.

[Kevin C]

I don't think that whether it has been funded makes a difference. The problem is that the SH match as defined in the document does not comply with the SH rules because the compensation definition fails the 414(s) test.

You can use a different definition of compensation for deferrals than you do for the match.

[K2retire]

I don't think you'd have to do a VCP, especially if the entire contribution hasn't been funded yet (I think the SHMAC needs to be funded at least quarterly, and I would think you'd have time to do the 4th quarter).

I believe that the SH match only has to be funded quarterly if the document calls for matching per pay period without true up. If the document calls for matching on a plan year basis you have until the due date of the tax return.

But the deposit date is not the big issue.

[BG5150]

I believe that the SH match only has to be funded quarterly if the document calls for matching per pay period without true up. If the document calls for matching on a plan year basis you have until the due date of the tax return.

I was thinking about the pay-period calc when I was typing my resonse, but neglected to put it in there.

[Kevin C]

Sieve,

I don't follow how 1.414(s)-1(d)(3)(iii) {Employees taken into account} gets you to -11(g), but -11(g) certainly looks a lot better than VCP.

[Guest Sieve]

Kevin --

From the following language from Treas. Reg. Section 1.414(s)-1(d)(3)(iii), I deduce that a plan's definition of compensation is part and parcel of determining if a plan is discriminatory under IRC Section 401(a)(4): "For example, in determining whether a plan satisfies section 401(a)(4), an alternative definition [of compensation] must satisfy this paragraph (d)(3) . . ." and "If an employer is using the same alternative definition of compensation to determine whether more than one plan satisfies section 401(a)(4) . . .". (Treas. Reg. Section 1.414(s)-1(d)(3)(iii)(A)).

Therefore, if a discriminatory definition of 414(s) compensation causes a 401(a)(4) failure, then that definition--and thus the 401(a)(4) failure--can be corrected as per the -11(g) regs.

At least that's my analysis.