Discretionary match option

[Molly the cat]

Client would like to have one match formula for executives & office staff and a second match formula for warehouse staff. We are told that the match for the warehouse staff will be a better formula than the formula for executives & office staff (which is where we are told HCEs are). We don't have a breakdown by group yet.

If we do this in the FT William document, we were told to use the discretionary match of D.6.a "A discretionary amount and percentage of Matched Employee Contributions". The Note in document states "The discretionary formula in D.6.a. must meet the nondiscrimination requirements regarding benefits, rights or features described in Treas. Reg. section 1.401(a)(4)-4.

While I understand that if you have different match formulas for different groups of people, you must do BRF testing, I'm not convinced that it is o.k. to use a discretionary match option in the document. Using discretionary in the document should mean having one discretionary formula for everyone in the plan, shouldn't it?

And, I understand that while the client is telling us that the match formula for the warehouse staff is better than the one for executives & office staff, we wouldn't know that until we actually saw what formulas they are thinking of using. And who knows what they would think up for the match formulas for future years.

Anyone else run into a formula like this or have any idea where/if it will fit into the FT William document?

I'm not looking forward to this at all as the BRF will be something we have to calculate manually as the testing software doesn't do that test for any match formulas.

Any thoughts? (maybe I should say useful thoughts )

[Tom Poje]

hopefully the famous last words "the HCEs are in the group that gets less" are true.

for example, lets say the executive group gets 50% match, and the warehouse group gets 75%.

then for BRF you have 2 groups -

1. how many get at least 50% which would be everyone (unless there is a last day rule or hours or something) so that passes

2. how many get at least 75% which, if there are no HCE would always pass (since no HCEs get the better of two formulas)

done plenty of BRF testing, but always with fixed match formulas, but I don't see why a discretionary formula should be any different