BRFs

[Guest qualified plan]

Question:

Company has two plans: each passes 410(B) test on its own.

Must the company perform BRF testing between the plans (assuming the plans offer different BRFs)?

Thanks in advance!

[jaemmons]

Without having any other information, I would say yes. Are they "like" plans (i.e.- DC, DB)? All like plans of the employer MUST be aggregated together in determining satisfaction of 401(a)(4), taking into account all nonexcludable employees under 410(a). Passing coverage on participation is only one test. Rates of accrual, vesting schedules, timing of benefits, availability of loans, etc. all must be looked at to see that they are not favoring HCE's for the entire company.

[Guest qualified plan]

Thanks for the response!

I am surprised by your response. It was my understanding that if a plan passes 410(B) testing on a disaggregated basis it need not perform 401(a)(4) (and by extension BRF) on an aggregated basis.

Could you please provide a citation to support your response?

Thanks again!

[Tom Poje]

I think one piece of important information is missing -

do the plans cover different employees, e.g. different divisions, etc.

if they cover different employees, then that is no different than having one plan and doing cross testing using class allocation, whcih of course requires 401(a)(4) testing.

if the plans cover the same employees, then if each passes on their own (assuming they are safe harbor formulas) then I believe the regs use the 'commutative' process...if A passes and B passes, then A + B passes.

well, I don't remember, I think that was grade school match. or is it a different term then 'commutative'

[Guest Harry O]

I'm not sure exactly what your question is asking . . . If you did not aggregate the plans to pass section 410(B) (that is, each plan passed 410(B) on its own), you do not have to aggregate for purposes of testing BRFs. Of course, when testing current availability of BRFs, you need to take into account all employees of the employer -- even employees participating in the other plan that passed 410(B) on its own.

[jaemmons]

What if there were different levels of matching contributions in both plans (assuming that they are 401(k))? Even though for 410(B) purposes you may pass on an unsafe harbor basis taking into account any participant who receives an allocation, the level of benefit must be tested under 1.401(a)(4) excluding all statutory exclusion under 410(a)/(B)(3)(A)&©. This discrimination testing is not satisfied under 401(m) because we are testing the formulas themselves and not the actual allocations received.

Even the availability of loans, hardships, timing of benefit payment, forms of benefit payment, and NRA, must be looked at. Depending on what each plan offers, you need to make sure that at least 70% of the NHCE workforce is able to receive this right or feature. I know it seems cumbersome, but EVERY BRF offerred under "like" plans (ie.-dc or db) must be looked at for 401(a)(4) testing purposes taking into account ALL employees of the employer, even if they are not covered under the plan.

Disaggregating the plans and testing only those employees covered for coverage and BRF's is not an option in a single employer plan. All employees must be taken into account.

[Guest qualified plan]

Thanks all--

To clarify, in my case, each plan is a 401(k)/PS plan and each plan covers a different company in the controlled group (controlled group consists of 2 companies). Each plan passes 410(B) testing on its own. In comparing each plan, each plan has a different level of Matching Contributions, a different level of Profit Sharing Contributions and different optional forms of benefit. All employees participating in a particiular plan are entitled to the same Matching Contributions, Profit Sharing Contributions and optional forms of benefit.

With respect to the fact that there are different levels of Profit Sharing Contributions between the Plans, no testing should be required since 401(a)(4) applies on a disaggregated basis. Agree? If you disagree, please provide a cite

Similarly, with respect to the different levels of Matching Contributions between the Plans no testing should be required. Agree? If dsagree, please provide a cite.

With respect to the different optional forms of benefit, it sounds like you are all saying that testing is required. Could someone please provide a cite to an IRC Section which would support your proposition?

Thanks again!

[AndyH]

It sounds to me like you are correct, and that Harry O's comments seem to support that as well, provided that each plan provides uniform BRF's within each plan.

[Tom Poje]

I would agree, providing,

1. when the statement that was made

'each plan passes on its own for 410(B)' means that all members of the other plan were treated as includable and not benefitting, (including terminees < 500 hours, if there were any)

2. matching formula is not based on svc..e.g. 50% for less than 5 years and 100% gretaer than 4 years.

note, see ERISA Outline Book (9.109, 2001 edition for BRF info)

If you think about it:

if plans are diasagregated and pass, it would be a moot point to test BRF.

for example, for 410(B)

Plan A passes ratio percentage at 90%

plan B passes ratio percentage at 85%

now, if I was to test my match formula for BRF on each plan, I hope I end up with the same percentages.

[Guest Harry O]

You can BRF problems even if the plans separately pass 410(B). This can happen if you have a "grandfathered" BRF in one of the plans. The plan as a whole could pass 410(B) but the limited group to whom the BRF is offered may not pass.

[Guest carsca]

What if a controlled group sponsors plans A, B and C, and plans A and B are aggregated for coverage testing -- does BRF testing involve plans A&B only, or A, B & C?

[Belgarath]

If the plans aren't aggregated for 410(b) testing, then I don't believe you must aggregate the BRF testing. You would just test each separately. And I think the 1.401(a)(4)-4 regs support this, in that they refer you back to 410(b) - so if you're passing 410(b) on a disaggregated basis, you test for BRF coverage on a disaggregated basis as well.