adp testing - controlled group

[Chippy]

I have two companies that are a controlled group. Plan A has deferrals and profit sharing, Plan B has deferrals and match. The combined plans pass coverage testing using the ratio percentage test. They also pass the Average Benefit Test. Is there another test I should be doing?

When doing the ADP Test separately, Plan A fails and Plan B passes. IF I combine them, Plan A fails less, but a participant of B will now receive a refund. Is this acceptable to combine for adp testing even if they are not required to be tested together? I think it is, but is it fair to the employee of Plan B that will now receive a refund?

[Tom Poje]

1.410(b)-7(d)

If an employer treats two or more separate plans as a single plan under this paragraph, the plans MUST be treated as a single plan for all purposes under sections 401(a)(4) and 410(b)

............

in other words, if you combined them coverage you have no choice but to combine them for the ADP test.

if you don't combine them for coverage then you can't combine them for ADP testing.

and yes, it could happen that someone who might normally not get a refund if tested separately gets a refund if you have to combine.

[Chippy]

Maybe combine for coverage isn't the correct term. By combining , I mean there are 22 comibined HCEs and 70 combined NHCE. In Plan B 7 out of 22 hce receive the match, and 25 out of 70 NHCE receive the match. to pass at 112.23%.

[Tom Poje]

ok, that makes more sense.

more accurately then,

you have 2 plans as part of a controlled group.

so 2 choices

1. if they are not permissively aggregated they both pass testing on their own and separate tests are run for ADP testing

2. plans are permissively aggregated, therefore ADP tests are aggregated as well.

when aggregated a person from plan B who did not have a refund when tested separately now has a refund. yes, while that seems unfair that can happen, but it could well be the overall result is the total refund is a lot less than when each plan is tested separately.