Controlled group--different SH's? and/or match?

[BG5150]

Controlled group.  3 companies A B & C.

A does not pass coverage on its own, B & C do.  

I can test A & B together, b/c together they pass coverage excluding C.  C passes excluding A & B.

Can A & B have different SH formulas?  Ie, one a 3% and the other SH Match?

If not, could I have A with SHM plus a discretionary match (that satisfies ACP SH) and B with just the SHM?

[John Feldt ERPA CPC QPA]

If they are aggregated for coverage, they must be aggregated for nondiscrimination.

You either cannot aggregate or you at least lose the safe harbor status if you do aggregate the plans. In either scenario, it is possible for an HCE in one plan to get a larger employer benefit than a NHCE in the other plan if they both defer the same percent of pay. My understanding is that at least blows the safe harbor.

Try running the average benefits test for coverage, there are only about a zillion scenarios you could try.

Otherwise, if they are separate lines of business,  and all have the same plan year end, and have enough employees, perhaps put two employers in QSLOB #1 and one employer in QSLOB #2. Too late to do that for calendar year 2018 though.

Perhaps amend under -11(g) to add participants and their corresponding QNECs/QMACs. Unless you’re trying to fix calendar year 2018, as it’s too late for that again. 

[Luke Bailey]

John Feldt, are you sure that, assuming A and B adopt separate plans, they could not have different SH provisions?

[John Feldt ERPA CPC QPA]

If they pass coverage without aggregating the two plans, yes, no problem.

Otherwise, when you aggregate for coverage, the nondiscrimination testing is done as if it was one single plan - can you do what you're describing in one single plan?

Meaning, could a single plan say that only some employees get safe harbor type #1 and another employee group gets a safe harbor type #2 plus a discretionary match?

[Purplemandinga]

I thought the requirement was that no HCE could receive a greater allocation than than of an NHCE in a safe harbor arrangement so if you aggregated this would occur.

[Purplemandinga]

It applies to matching allocations, I suppose this wouldn't apply if one plan was a NE and one plan was a match.

(4) Limitation on HCE matching contributions. The safe harbor matching contribution requirement of this paragraph (c) is not satisfied if the ratio of matching contributions made on account of an HCE's elective contributions under the cash or deferred arrangement for a plan year to those elective contributions is greater than the ratio of matching contributions to elective contributions that would apply with respect to any eligible NHCE with elective contributions at the same percentage of safe harbor compensation.

[Purplemandinga]

But if one plan was NE and one plan was Match perhaps it wouldn't pass coverage?!? So many questions!

[Luke Bailey]

1 hour ago, John Feldt ERPA CPC QPA said:

If they pass coverage without aggregating the two plans, yes, no problem.

Otherwise, when you aggregate for coverage, the nondiscrimination testing is done as if it was one single plan - can you do what you're describing in one single plan?

Meaning, could a single plan say that only some employees get safe harbor type #1 and another employee group gets a safe harbor type #2 plus a discretionary match?

OK, this stuff is complicated, to be sure. I don't find any equivalent in 401(k)(12) or 1.401(k)-3 to the indented "flush language" at the end of 401(k)(3)(A), which clearly does require aggregation for ADP if you aggregated for 410(b).

[John Feldt ERPA CPC QPA]

Maybe someone could volunteer to write up a plan where the safe harbor match is 221% of deferrals up to 6% of pay for only Group 1 (just happens to be the doctors) and the other group (the rest of the employees) only have a 3% safe harbor non-elective? Submit the document for an IRS determination letter and be sure to point out in the cover letter the dissimilar but aggregated safe harbor formulas.

I was about to say "I can't imagine getting a favorable determination letter on that!", but then I actually imagined it before I could type that all out.

[Mike Preston]

9 hours ago, John Feldt ERPA CPC QPA said:

I was about to say "I can't imagine getting a favorable determination letter on that!", but then I actually imagined it before I could type that all out.

You have an over active imagination. 

[Kevin C]

The definition of "plan" in 1.401(k)-6 leads you on a trail to the definition of "plan" in 1.410(b)-7 and 1.401(k)-1(b)(4)(iii)(A) specifically says two permissively aggregated plans are treated as a single plan under section 401(k).  The 1.401(k)-6 definition of "eligible employee" is an employee eligible to defer under the plan. An "eligible NHCE" is defined as an eligible employee who is not an HCE. The requirements for the non-elective SH in 1.401(k)-3(b) require that each "eligible NHCE" receive the non-elective SH.  Likewise, 1.401(k)-3(c) has a requirement that each eligible NHCE receive the SH Match.  When you try to permissively aggregate a 3% SH plan and a SH Match plan, the aggregated plan is not SH because neither SH formula is provided to each eligible NHCE.

1.401(k)-1(b)(4)(iii)(B) says you can't aggregate plans with different testing methods.  It specifically mentions that you can't aggregate a prior year testing plan with a current year testing one and that you can't aggregate a SH plan with a non-SH plan.  If you read it as allowing plans with different SH formulas to be aggregated, the aggregated plan would not be safe harbor and would be subject to ADP and ACP testing.

   

[John Feldt ERPA CPC QPA]

Give that man a trophy!

[Luke Bailey]

3 hours ago, Kevin C said:

401(k)-1(b)(4)(iii)(A) specifically says two permissively aggregated plans are treated as a single plan under section 401(k)

OK. I was looking for that in 1.401(k)-3, but 401(k)-1(b)(4)(iii)(A) clearly says that it applies for 1.401(k)-1 through -6, so that ties it up. Thanks.

[KEC79]

If there are two plans with the same safe harbor match formula in a controlled group, would they satisfy nondiscrimination testing (regardless of whether they have to be aggregated for 410b testing)?