Simple General Testing Question

[AndyH]

Having a brain cramp, need help on how to handle past service grants in a new plan for testing purposes.

Looking at a proposal. New plan, two owners with 4 years of past service each, none for employees.

Simplified example of design:

Owners accrue 2% of pay x YOS = 8% of pay at end of Plan Year 1.

Employees accrue $100 per month per YOS = $100 at end of Plan Year 1. Assume this works out to 2.1% of pay.

1. It seems to me that this must be general tested because the formulas are not uniform, right?

2. If so, how is the past service handled for testing purposes in year 1?. Do I have a NAR of 2% for the owners, or a NAR of 8%?

[Andy the Actuary]

Are the facts that the NHCEs have no prior service (not that they have prior service but are not getting credited)? If the case, do you need any testing initially? Presumably, NHCEs (aka non-key employees) will get the 2% top-heavy minimum accrual. 4 years of past service is within the safe-harbor for grants of past service under the 401(a)(4) regs.

[AndyH]

NHCEs have no past service.

If this were a safe harbor design I think it would be fine.

Forget top heavy - the actual benefits are above 2% - I tried to simplify the actual benefit levels.

Because there are two formulas that apply to HCEs vs. NHCEs I think it needs to be general tested due to lack of uniformity. So, does general testing get messed up by the past service grant numerically? I think not, but I'm not 1000% sure, so I thought I'd bounce it off the esteemed panel. It comes down to what is the amount of testing service and what benefit is considered earned in the testing year, which happens to be the only plan year. Using the annual method, testing service must be one. But I got 5 years of accrual, so is accrued to date with 5 years of testing service OK?

[Andy the Actuary]

NHCEs have no past service.

If this were a safe harbor design I think it would be fine.

Forget top heavy - the actual benefits are above 2% - I tried to simplify the actual benefit levels.

Because there are two formulas that apply to HCEs vs. NHCEs I think it needs to be general tested due to lack of uniformity. So, does general testing get messed up by the past service grant numerically? I think not, but I'm not 1000% sure, so I thought I'd bounce it off the esteemed panel. It comes down to what is the amount of testing service and what benefit is considered earned in the testing year, which happens to be the only plan year. Using the annual method, testing service must be one. But I got 5 years of accrual, so is accrued to date with 5 years of testing service OK?

Feh, we're splitting hairs. Make the NHCE's benefits the greater of 2% of pay or $100/mo/yos and you should be fine without testing.

[Mike Preston]

I don't have time to look it up today, but accrued-to-date can base testing years on the same basis that is used to accrue benefits. Hence, if your design was (A) divided by (B) where (A) is a function of service, but (B) is a function of participation, you are stuck with one year as testing service. But if (B) is a function of YOS, your testing service should be YOS.

Let me know if you need me to paw through the 401(a)(4) reg for confirmation of my recollection.

[AndyH]

No, I think you are right with respect to the accrued to date method. The testing period would need to be the current year and all prior years, I think, which does have other ramifications such as the required use of average (rather than plan year) comp if I remember correctly. But I'd have to double check that before being certain.

Thank you.

[Mike Preston]

ATA's design, of course, obviates the need for general testing completely, if the "greater of" concept fits your situation.

[Bill3]

Whether the general test is used or a safe harbor formula is adopted, the requirements of 1.401(a)(4)-5(a)(3) need to be met. A red flag is that none of the NHCEs have any past service -- why is that?

[AndyH]

I don't have any more information other than the census presented, but your point is understood and is a valid area of inquiry.