Age weighted Profit Sharing allocation

[cpc0506]

I have a new plan with 4 employees, one of which is an HCE. I have allocated the Profit Sharing based upon the age-weighted formula as stated in the document.

I was under the impression that I needed to run the general test since this allocation method is not a safe harbor allocation formula. When I run the test, the plan fails 401(a)(4). Only 1 NHCE has a EBAR that is greater than the HCE's EBAR. It does not appear that the general test was run last year. How do I fix the allocation to get the plan to pass? What options do I have?

[Mike Preston]

It sounds like the assumptions that are being used in the document to determine the age-weighted allocations are different from the assumptions being used to run the general test.

If the age-weighted allocation uses 401(a)(4) compliant assumptions, the allocation is done correctly, and the general test is run using those same assumptions then there will be exactly one rate group and each HCE and NHCE will have the same EBAR and therefore all will be in that same rate group.

Before you jump to the conclusion that the fact that no general test was run last year leads to something that requires fixing, make sure you have something to fix!

[cpc0506]

I agree that the numbers should come out the same. My issue is that the HCE is over 65 and as a result, the testing APR is not the same for him as the other employees. If the same APR is used then, yes, everyone's EBAR is exactly the same.

I did decide to run the test using rate banding, and the plan then passes the rate group test. I have not used banding before. but will look into why it passes under this method of running the rate group test.

[Tom Poje]

years ago I came across a similar type situation. possibly this is what is going on in your case.

the example used was the following:

2 ees, one age 65 and one age 70.

both make the same compensation.

under age weighted the older person would normally receive a smaller contribution (because of the APR)

that's a bit discriminatory, so the doc was written to use the same APR when determining points. that works fine for getting the same contribution, but now testing the age 70 person has a larger e-bar, which would fail testing.

however, 1.401(a)(4)-8(b)(1)(ii) Allocations after testing age A plan does not fail to satisfy [testing] merely because allocations are made at the same rate for employees who are older than their testing age....as for employees who are at that age.

I'm sure most software won't catch something like that.

.................

rate banding - (in a rather simplified description- lets say the HCE has an e-bar of 2.50 and all NHCE have 2.45. under rate banding they would all be treated as having the same e-bar because 'they are close enough to each other'

[Mike Preston]

Rate banding doesn't really work. If you need it to pass, generally, then you aren't eligible to use it!

[BG5150]

Mike,

If and when CAN you use it? Are there any examples or explanations of this topic floating around?

[Tom Poje]

I don't think there are 'examples' floating around.

yes irate bandingt is in the regs, but the same reg also says you can only use if the HCEs do not fall significantly higher than the NHCEs within the range.

but how is that to be interpreted?

You establish a midpoint, for example 11.0. so everyone with in 5% is treated as having an E Bar of 11.

so everyone between 10.45 and 11.55 is treated the same. (5% of 11 is .55)

now if all the NHCEs are below 11 and all the HCEs are above 11 that probably is not so good, especially if the HCE is at 11.55. that sounds like you are significantly favoring the HCE.

If some of the NHCEs are above 11 it is probably less of a problem. e.g. you are bring some NHCEs down to a mid point rather than only HCEs.

even if all the HCEs are above the midpoint and all the NHCEs are below it is probably less of a problem if instead of a 5% range you only use 1%, but who is to say? 5% is simply the maximum range, so does that mean 1% would be considered 'close enough'? In the case of an age weighted plan like the example above I suspect it is less of a problem (I doubt the range needed is anywhere close to 5%), than a cross tested plan in which the HCE gets 20% and everyone else gets 5%.

..............

I suspect in the example above if disparity was imputed that would probably be enough to pass rate group testing. while I personally have never seen an integrated age weighted formula, I do recall reading an article about such a formula, the conclusion being it wasn't worth all the extra involved in such a formula, so unlike an ESOP or SHNEC which you can't impute disparity, my understanding is it should be possible even with an age weighted formula.

[Mike Preston]

What he said.