410b issue in a controlled group?

[AlbanyConsultant]

Took over a plan this year, and I'm drawing a blank on where the problem is... or is there one?

G and H are a controlled group, and each has their own 401(k) plan.

G: 118 participants that can't be segregated via statutory exclusion, of which 96 are union (included in the plan), has 3 HCEs (all non-union), and also an exclusion for "per diem" employees (14 of which are union (included in the 96), and 3 are non-union)

H: 25 participants, no union, no HCEs, also has a match provision

Clearly, there's no issue with the coverage for the match.

For the deferrals, after pulling out the union employees in G's plan I'm coming up with benefitting 16 NHCEs (118 total - 96 union - 3 HCE - 3 "per diem") out of 44 NHCEs (the 16 benefitting + 25 from H + 3 "per diem") = 36%. And of course all three HCEs benefit.

From here, I'm not sure what to do next. It seems I've found things that point in several different directions, so I'm just spinning my wheels. Any thoughts? Thanks.

[Jim Chad]

I am having trouble with the 36%. I can't follow where that came from.

If we test for coverage on acombined plan basis and exclude all union from numerator and denominator, what percentage do you get?

[12AX7]

The OP appears to be getting the 36% by testing Company G separately. The plans need to be aggregated to pass coverage.

[Mike Preston]

Yes, but it matters greatly what the "plans" are that are being tested. 401(k)? I don't see an issue. Match? Already determined not to be an issue. The only potential issue would be a PS contribution made solely to G. Is that what is going on here?

[12AX7]

What about BRFs? Where applicable, wouldn't there be some concerns?

[AlbanyConsultant]

It's the second plan that is messing with my head...

I think the 36% is what I get when I do the combined test without the union for G's plan:

NHCE (excluding union)

• 19 G NHCEs in total
• 3 G NHCEs that don't benefit (per plan exclusion, the "per diem" ee's), included in the 19
• 25 H NHCEs (none of whom benefit under G's plan)

So that's 16/44 = 36% for the NHCEs, compared to 3/3 = 100% for the HCEs, if I'm testing from G's point of view.

Or can I disaggregate* somehow? Because G would pass on it's own ([16/19] / [3/3] = 84%) and H passes on it's own (no HCEs). There is no profit sharing (or reallocation of forfeitures), and the plans have identical entry (or so I'm told... I'm still getting data from H because we don't do that plan). The only difference is that H has a match, which I think is OK for testing purposes.

* for both coverage and ADP

Mike Preston, you said that you don't see there being an issue on the deferral side... is that because they both are getting a deferral opportunity?

[Mike Preston]

Yes, as long as the BRF's don't create an issue, as already mentioned. I don't see any reason why you would want to test the 401(k) feature on the basis of G alone. Aggregating makes it pass, doesn't it?

[AlbanyConsultant]

I should be getting H's info any day, but I'm led to believe that their deferrals are pitiful; I'm concerned that it could weigh down the NHCE ADP of G, so I'd prefer to test separately. G is likely going to fail on it's own, true, but it will be worse with ~20 more NHCES (from H) who don't defer.

Not to be dense, but the way this works is how? Because there are two plans with equivalend BRF, I can disaggregate them for coverage testing, and therefore can disaggregate them for ADP testing? Whereas if I aggregated them for coverage, I'd have to aggregate them for ADP?

[Mike Preston]

Yup.