Using last day vs no last day

[Guest JPIngold]

Ever since the advent of making every participant their own allocation class, I have normally eliminated the last day and service requirements to receive an allocation, figuring I just leave those terminees out of the allocation (assuming I satisfy 410(b) and 401(a)(4)). About the only downside that I have found is that I can not exclude a participant who terminates with less than 500 hours in the testing, when normally such employees are excludable.

On the flip side, I find that it gives me more flexibility in that: (1) I can use a young terminee with low pay and give them an allocation to help testing without the need to adopt an 11(g) amendment; and (2) that young terminee may not even be vested, and I can't do that with an 11(g) amendment.

However, I see some cross-tested documents out there that still have EOY and service requirements, and I am just wondering if I am missing something or others have reasons that outweigh my reasons.

Any thoughts would be much appreciated!!!

[ETA Consulting LLC]

You're not missing anything. I seldom seek logic in what someone else is doing. You know how the test is calculated. So, the ability to give an early terminee a relative small amount (while creating a higher percentage) should offset the advantage of excluding them from the test. In these type designs, the more flexibility the better; so why ever tie your hands with such an accrual requirement?

Good Luck!

[John Feldt ERPA CPC QPA]

Not really a big deal, but FYI:

If the plan is not safe harbor, when you do the coverage test you won't be able to exclude terminees with under 500 hours if they are allocated nothing, because that exclusion from the test only applies if the SOLE reason for them getting no allocation is because they did not satisfy the allocation conditions. Thus, with no allocation conditions in place, they stay in the count for the test. This would be rare plan to find out there, but just FYI.

With no allocation conditions, you have the most flexibility, adding conditions will just limit the plan's options.

[Guest JPIngold]

Not really a big deal, but FYI:

If the plan is not safe harbor, when you do the coverage test you won't be able to exclude terminees with under 500 hours if they are allocated nothing, because that exclusion from the test only applies if the SOLE reason for them getting no allocation is because they did not satisfy the allocation conditions. Thus, with no allocation conditions in place, they stay in the count for the test. This would be rare plan to find out there, but just FYI.

With no allocation conditions, you have the most flexibility, adding conditions will just limit the plan's options.

Agree --- that was the one downside I mentioned. The old 1.410(b)-6(f)(1) rule.

[K2retire]

Not really a big deal, but FYI:

If the plan is not safe harbor, when you do the coverage test you won't be able to exclude terminees with under 500 hours if they are allocated nothing, because that exclusion from the test only applies if the SOLE reason for them getting no allocation is because they did not satisfy the allocation conditions. Thus, with no allocation conditions in place, they stay in the count for the test. This would be rare plan to find out there, but just FYI.

With no allocation conditions, you have the most flexibility, adding conditions will just limit the plan's options.

But you could use a requirement that terminees have more than 500 HOS (instead of no allocation conditions) to get around this.

[John Feldt ERPA CPC QPA]

After thinking about this, maybe there's more. Let's look at two plans: Traditional Old Plan A and New Plan Design B

The provisions of Plan A require a last day and 1000 hours requirement and the allocation is pro-rata on compensation.

The provisions of Plan B have no allocation conditions but each participant is in their own allocation rate class.

Suppose each plan is otherwise the same and covers the same demographics in terms of wages and employee counts: they each have 1 HCE and 4 NHCEs during the year. Of the 4 NHCEs, 1 quits voluntarily with 1,000 hours in the year.

Both employers contribute the same amount, and Employer B just happens to pick the 3 active NHCEs for the allocation and gives them a uniform percent of pay allocation, the same percent as the HCEs. Both plans pass coverage: 75% is good to go.

So far, A and B look identical, but what about nondiscrimination?

Plan A allocates a uniform amount as described under 1.401(a)(4)-2(b)(2)(i) and uses the exceptions in 1.401(a)(4)-2(b)(4)(iii) to say the "zero" allocation given to the terminee can be disregarded for uniformity purposes.

Does Plan B get to apply the exceptions offered in (4)(iii)? According to the regulation, a plan does not fail to satisfy the uniform allocation formula "merely because the plan contains one or more of the provisions described in this paragraph (b)(4)"

The way that regulation is written, it seems to me that if the plan document does not contain a last day or 1000 hour provision, then I think the plan must satisfy 401(a)(4) in some other manner (including the terminees), such as testing the allocations or by testing the equivalent benefit accrual rates derived from those allocations.

Thus, I think New Plan Design B, although much more flexible, could be more dependent on the demographics than Plan A would be.

Agree? Disagree?

edit:typo

[Tom Poje]

As I understand it:

1 HCE, 3 NHCE, 1 of the NHCEs quits, more than 500 hours. That NHCE will not get a contribution due to last day rule.

Plan A - 2 groups, owner and all others

Plan B - each emp[loyee in its own group.

in either case, only 2 of the 3 NHCEs benefit, which is 66.7% which fails ratio % test.

In plan B, because each person is in its own group, the plan fails reasonable classification and can't rely on the avg ben test for coverage purposes.

Plan A could rely on this test and might pass the avg ben % test if there are deferrals (and match) on an allocation basis, or even pass testing on an accrual basis. Testing on an accrual basis for the avg ben % test does not mean plan A is cross tested or needs to provided a gateway. at this point in time it's coverage that is being looked at, not nondiscrimination.

[Guest JPIngold]

After thinking about this, maybe there's more. Let's look at two plans: Traditional Old Plan A and New Plan Design B

The provisions of Plan A require a last day and 1000 hours requirement and the allocation is pro-rata on compensation.

The provisions of Plan B have no allocation conditions but each participant is in their own allocation rate class.

Suppose each plan is otherwise the same and covers the same demographics in terms of wages and employee counts: they each have 1 HCE and 4 NHCEs during the year. Of the 4 NHCEs, 1 quits voluntarily with 1,000 hours in the year.

Both employers contribute the same amount, and Employer B just happens to pick the 3 active NHCEs for the allocation and gives them a uniform percent of pay allocation, the same percent as the HCEs. Both plans pass coverage: 75% is good to go.

So far, A and B look identical, but what about nondiscrimination?

Plan A allocates a uniform amount as described under 1.401(a)(4)-2(b)(2)(i) and uses the exceptions in 1.401(a)(4)-2(b)(4)(iii) to say the "zero" allocation given to the terminee can be disregarded for uniformity purposes.

Does Plan B get to apply the exceptions offered in (4)(iii)? According to the regulation, a plan does not fail to satisfy the uniform allocation formula "merely because the plan contains one or more of the provisions described in this paragraph (b)(4)"

The way that regulation is written, it seems to me that if the plan document does not contain a last day or 1000 hour provision, then I think the plan must satisfy 401(a)(4) in some other manner (including the terminees), such as testing the allocations or by testing the equivalent benefit accrual rates derived from those allocations.

Thus, I think New Plan Design B, although much more flexible, could be more dependent on the demographics than Plan A would be.

Agree? Disagree?

edit:typo

I agree that you are always more dependent on demographics when you use a non-safe harbor allocation method. I think you have to know this going in and be prepared to not have the disparity you strive for when designing the plan. However, I think that as long as the demographics, on average, support the design, I think it will be more likely to be beneficial to the targeted HCE's in most cases.

I agree with the fact that for 401(a)(4), you need to include that terminee and you won't automatically satisfy any safe harbor. But, this is where you still hope the general test will bring success.

[chc93]

As I understand it:

1 HCE, 3 NHCE, 1 of the NHCEs quits, more than 500 hours. That NHCE will not get a contribution due to last day rule.

Plan A - 2 groups, owner and all others

Plan B - each emp[loyee in its own group.

in either case, only 2 of the 3 NHCEs benefit, which is 66.7% which fails ratio % test.

In plan B, because each person is in its own group, the plan fails reasonable classification and can't rely on the avg ben test for coverage purposes.

Plan A could rely on this test and might pass the avg ben % test if there are deferrals (and match) on an allocation basis, or even pass testing on an accrual basis. Testing on an accrual basis for the avg ben % test does not mean plan A is cross tested or needs to provided a gateway. at this point in time it's coverage that is being looked at, not nondiscrimination.

Tom, can you please explain Plan B. The terminated NHCE is not getting a benefit because of the last day rule, as opposed to not getting a benefit because his "individual group" contribution is zero. So why would this preclude ABPT for coverage (failed reasonable classification) if the terminated NHCE is not getting a contribution for a reason that is not related to his "individual group". Seems like the last day rule can be used to allow ABPT for coverage for both Plans A and B. Thanks...

[Tom Poje]

I believe the IRS at different ASPPA Q and A's (and always have to be cautious about such statements because they don't necessarily reflect actual Treasury opinion) has stated that once you put everyone in their own group, then 1.410(b)-4(b) reasonable classification rules kick in "...an enumeration of employees by name or other specific criteria having substantially the same effect as an enumeration by name is not considered a reasonable classification."

so basically they are saying"While you can put everyone in their own group, this is not a reasonable classification"

[John Feldt ERPA CPC QPA]

Thus, for coverage, you'll need to give that person a dollar to avoid that issue. However, if you have everyone in their own class but still have a last day condition, you'll have to amend out that condition in order to get through the coverage test. After that you can deal with nondiscrimination testing.

edit: typo

[12AX7]

Am I understanding this correctly (or perhaps incorrectly) that I'm never able to use the Nondiscriminatory Classification Test if I have all participants in their own allocation group even if the plan has last day/1000 hours requirements for an allocation? Then rate group testing can only be satisfied by the Ratio Percentage Test?

[Tom Poje]

I think you are misunderstanding things.

there are 2 tests

Coverage (how many people get) and Nondiscrim (How much people get).

There is no problem using the classification test for rate group testing.

Its the coverage testing that is the issue. If you classify people by name (or something which is basically the same as doing things by name) then you can't use nondiscrimination classification testing to pass coverage.

That normally shoudn't be a problem, but I have a boss that sometimes likes to exclude people entirely from the plan to cut costs. (DB/DC Combos) this then becomes a problem if the DC plan has everyone in their own group, even though due to ages the plan passes nondiscrimination testing using rate group testing., because less than 70% actually received something (coverage issue)

[12AX7]

I am looking at this from a coverage perspective and this is perhaps where I'm having difficulty with this. Suppose there are two HCEs (each getting a different allocation) and the first HCE (group) passes rate group testing with the RPT. The second HCE rate group is subject to the ABT. All eligible employees are in their own allocation group. Each rate group must pass coverage, no?

[Tom Poje]

I guess I'm not quite sure what you are asking.

before getting to nondiscrim testing, the plan itself must pass coverage. you don't have different rate groups for this portion.

so if both HCEs benefit, at least 70% of the NHCEs must receive something as well, if the plan as a whole fails reasonable classification.

If you pass that, you can do whatever you want for rate group testing, at least as far as I know.

1.401(a)(4)-2©(3) says a plan passes reasonable classification if ratio % is greater than the midpoint. there is no mention of having to also avoid having people listed by name, etc.

[12AX7]

Ok, I get it now. It's the overall Coverage Test, rather than each Rate Group passing the General Test. It seems that's where the thread started, but took a turn into this direction. Thanks for your patience with me.