I should have known better than to delve into this on a Friday afternoon. I started off thinking I knew what I was doing, and have managed to confuse myself. This isn't a real plan or document, just hypothetical.

Suppose you have a 401(k) that provides a dollar for dollar match, up to 3%, and also provides for an employer discretionary contribution on a cross tested basis.

Let's further assume that for a given year, the employer makes no discretionary contribution but the plan is top heavy. Some participants have deferred, some have not. The employer wants to use the match as counting toward the TH. Clearly this is allowable. Question is, is this satisfying the "multiple allocation" exception for TH in 1.401(a)(4)-2(b), and thus does not require general testing for the additional employer contribution to bring everyone up to 3% who doesn't already get 3% from the match?

Part two - assuming for the moment that the above is ok and general testing of the additional required top heavy isn't necessary, now the employer wants to contribute an additional discretionary contribution to be allocated on a cross tested basis. Can this be allocated on a cross tested basis separately, or must the rest of the employer constribution for TH be brought in? All subject to gateway, of course...

By the time I come back on Tuesday, maybe I'll no longer be confused. In advance, wishing you all a good long weekend.

Part One: I don't think you have to general test the topping-off TH min contribs as you describe them in Part One if the plan has the language described in Treas Reg 1.401(a)(4)-2(b)(4)(vi)(D)(3): "In the case of a plan that provides the greater of the allocations under two or more formulas, one of which is a top-heavy formula, the top-heavy formula does not fail to be available on the same terms to all employees merely because it is available solely to all non-key employees on the same terms as all the other formulas under the plan."

Part Two: I think that the topping-off TH min contrib should be taken into account as you compute your cross-testing of the discretionary contribution. Here, the formula other than the TH min contrib one does not meet a safe habor. You're testing. Also, I think you can only apply the topping-off TH min contribs after all other contributions have been taken into account.