Permissively aggregating 403(b) Plan with a 401(a) Plan for top heavy testing of the 401(a) Plan?

Another TPA has asserted that this is possible. I think this is dead wrong, and I can find no support for such a position. However, I'm always willing to question myself. Does anyone believe this is possible?

Belgarath, I'll be interested in seeing whether anyone is aware of specific guidance on point. I could not find any. I can see the argument, in that you can aggregate 401(a) and 403(b) for 410(b) and 401(a)(4), and you can read 416 and the regs as saying you can aggregate for top-heavy any plan that you can aggregate for 410(b) and 401(a)(4). On the other hand, I don't see in the statute or 416 regs where "plan" is defined, so one could also argue that the only "plan" being referred to in 416 and the regs is a plan that is subject to 416, which of course 403(b)'s are not. On the policy side, it is arguably unfair to not let you aggregate just because the plan that you are aggregating with could not itself be top-heavy, since the point of the top-heavy rules is to make sure that non-control folks get significant benefits and the 403(b) balances are benefits. But as you know the IRS generally prefer technicalities over policy, and the courts will use policy at best as a tie-breaker, and arguably this doesn't rise to the level of a tie on the technicalities.

416 specifically mentions plans under 401(a) and states that SEPs are DC plans for purpose of 416. I don't see any mention of 403(b).

38 minutes ago, Lou S. said:

416 specifically mentions plans under 401(a) and states that SEPs are DC plans for purpose of 416. I don't see any mention of 403(b).

But Lou S., you didn't find a definition of "plan," right, just a lot of references to lower-case p "plans" like you're supposed to know what that means? One could well interpret that as being a reference to the "plans" that are described in the first part of the regs as being subject to 416, but I didn't find where the regs actually spell that out definitively.

I believe that the question comes down to if there are any key employees in the 403(b) plan and if the plans must be aggregated for purposes of 401(a)(4) or 410(b).  Then one must look to required aggregation groups and permissive aggregation groups.  Code Section 416(g)(1).  But it also just refers to Defined Contribution Plan (and a 403b plan is a DC Plan).  

I also could not find any specific reference to aggregation with a 403(b) although SEPs are specifically singled out.