Late Retirement

[nancy]

Has anyone ever seen any document language which provided for the sum of continued accruals plus actuarial increases vs the greater of the two? Would this be treated as a separate benefit structure and require (a)(4) testing?

[flosfur]

No, I have not seen it - not yet. And I don't this requires general testing.

Is there any reason you think it may require general testing.? Do you think this is a subsidized late retirement benefit in any way?

I don't think it is subsidized. Let's consider the formula. I assume the formula you have is:

Actuarial Equiv (AE) of NRB plus additional accruals.

Although it appears generous, it is still less than the theoretical actuarial equivalent of accruing benefits (if one agrees with the philosophy that benefit accruals should continue until late retirement date Vs. the old old practice of stopping accruals at NRA).

At any time after NRA, the mathematical true equivalent is:

AE of "prior year" accrued plus current year additional accrual.

So at age 66, the accrued would be:

AE of accrued at 65 plus additional accrual after age 65.

At 67, the accrued would be:

AE of accrued at 66 plus additional accrual after age 66.

and so on.

So the formula you have does not represent a subsizing late retirement benefit.

[nancy]

It's intended to do what a dc plan would do - provide an accrual (contribution) and give earnings on the prior account balance (actuarial eqivalence of prior years accrued benefit.