I am working with an employer that is using a new prototype document that specifies the 'Participant Group Allocation Method' under LRM 94 language. The adoption agreement specifies that the employer will specify the groupings for the plan year by the time it makes the contribution.

After running several different grouping scenarios, it looks like cross-testing is not as good for the plan year ended April 30 as would a simple profit sharing allocation of the same percentage of earnings for everyone. Due to some personnel changes in this small company (10 employees, counting the two owners), the owners are almost all younger than the 7 other employees.

May the employer specify one grouping of all 10 employees under the LRM 94 language, allocate the entire contribution within that one group in respect to earnings, and not be discriminatory because all eligible employees are receiving the same percentage of earnings? Or must we use cross-testing to show discrimination, in which case the single grouping idea would cause the plan to be discriminatory?

Sure. Employer specifies same contribution for all groups. It passes as a uniform allocation. Unless the document has some weird provision that states contributions must pass the general test on a benefits basis, but can't imagine it has such language.

Since you have a non-safe harbor allocation formula, you must general test, but you don't have to cross test (i.e. on a benefits basis in a DC plan). An allocation that is effectively pro-rata as you suggest will pass the general test on a contributions basis. (If the owners make more than the other employees you might be able to give them a little bit more and impute permitted disparity to pass the test.)

Mucho gracias, Senors Norman and Bird.

Bird, to do integration in that context, would you have to put all of the employees that earn under your integration level in one group and allocate to them at least the base rate, and then each of those that earn more than the integration level into a separate group of his or her own and then make the contributions? It seems to me that since each making over the integration level would have a different excess compensation amount and within each grouping the contribution must be allocated in proportion to actual income, this might be necessary to achieve the same outcome as having a stated integration formula.

There might be the possibility of bunching a few of those excess earners into one or a few groups so long as what each HCE receives does not exceed the delta permitted under integration, vis-a-vis the base amount that the NHCEs are receiving.

Revised to correct a typo

J Simmons, I agree with everything you said. I don't know how the groups are set up and was just commenting that the allocation didn't necessarily have to be pro-rata across the board. It would likely not be the same as an allocation formula that used PD.