How I'm running nondiscrimination testing for doctor's plan; please check me

[pbarrett]

Just to make sure I've got the testing down right...., I have a client (doctor) who makes well over $200,000. He has establilshed a new comp plan with a last day rule. There are 2 groups: 1.) the dr 2.) all others.

The dr. is set to get $40,000. The employees we're giving 1/3 of his % which is 6.7%. Actually, we giving group #2 7%.

My 410(b) coverage (70%) passes.

My 410(b) average benefit fails

The minimum allocation gateway and safe harbor minimum both pass.

Based upon the above testing, the allocation is ok. I don't need to worry about the average benefit test failing because the gateway passed. Right?

I generally have the new comp plans pass both gateway tests (minimum and safe harbor). Am I being too conservative?

Pat

[Blinky the 3-eyed Fish]

Based upon the above testing, the allocation is ok.  I don't need to worry about the average benefit test failing because the gateway passed.  Right?

Wrong. The gateway is merely that. It is what gives you the right to cross-test the plan. So, first you provide the gateway minimum, then you run the cross-testing. You haven't provided enough information to say whether that passes or not.

I am not sure what you are referring to by "safe harbor minimum" and "both gateway tests".

You mention that your average benefits test for 410(b) fails. Why are you running this when you pass coverage on the ratio percentage test?

I have the feeling a few pieces of the puzzle are missing for you. Perhaps you could post the ages of the participants (if not too many) and someone can walk you through the test.

[pbarrett]

The Dr. is 62. His ebar is 3.13%. The 3 NHCE's are receiving 7% of pay -- ebar 1.02, 5.37, 3.48 for an average of 3.29%.

The plan is top heavy. There is no one eligible with less than 1000 hours so I think I am ok here because everyone is getting 7%.

Our software system (Quantech) prints out both the safe harbor gateway and the minimum (minimum is 1/3). The safe harbor gives everyone 5%.

Another question too if you all can figure out my first question is: Does any employer really state how much they want to put in before the year begins? My clients say give me the most and give as little as possible to the others.

Oh, I run all the tests. I don't know why I run the average benefits tests other than to see how it looks.

Any tips would be appreciated.

Thanks!

[Blinky the 3-eyed Fish]

Okay, good info.

First, we look at the gateway requirements (the lesser of 1/3 of the highest HCE and 5%). You are providing 7%, so we move through the gate and on to cross-testing.

Next, based on the nonhighly concentration percentage of 75%we need to have the lowest rate group (which is the only rate group in this case) be greater than 33.75%. So, at least 2 of the NHCE's need to have a higher rate group than the HCE.

In this case the your only rate group looks like: (2/3) / (1/1) = 66.7%, so you pass, but must move to the average benefits test since your percentage is under 70%.

Average Benefits Test:

Avg NHCE's: (1.02 + 5.37 + 3.48)/3 = 3.29%

Avg HCE's: 3.13%

Avg Benefit % = 3.29/3.13 = 105.1%

So, you pass that test as well and you PASS the cross-testing. In fact, you could set the allocation to the NHCE's lower, so that the NHCE with the 3.48% EBAR is right at 3.13%. To not do that will not be achieving your client's objective.

Going back to the question of your original post now that I understand your terminology, yes, I think you are being too conservative in giving the GREATER of the 1/3 or 5% versus the required LESSER. No reason to do it if the testing would pass otherwise, unless the client were aware of what you were doing and why.

[Tom Poje]

The critical point is that the NHCE concentration percentage is 75%, which produced the magic midpoint of 33.75%, as the 'fish' indicated.

With 3 NHCEs, that means you must have 2 NHCEs with E-Bars > or = to the HCE. you indicated that the HCE is at 3.13.

One NHCE is at 5.37, no problem

The next NHCE is at 3.48.

that does not give you a lot of room to spare

Ignoring rate banding, imputing disparity you could have gotten by with

(3.13 / 3.48) * 7%allocation = 6.3% and still pass the rate group portion of testing, so you really couldn'y do that much better than what you did.

as 3.13 / 3.48 = 90% then the average ben % test would have passed as well.

I imagine the extra .7% is not enough of a difference....oh wait, this is a doctor, giving anything way is too much.

[pbarrett]

Thanks Blinky & Tom. The info really clarified it for me.

[AndyH]

Wow, Blinky, I just used my new Dave-Baker-given power to zap my incoherent message from yesterday along with your (correct) comment that my message made no sense! What power! Tom, how do you control yourself? Next thing there'll be a movie and action figures.

[Blinky the 3-eyed Fish]

I wish I could zap my incoherent and incorrect messages.

[KJohnson]

Nice explanation by Tom and Blinky as long as the IRS does not determine that you have set up a non-qualified CODA for the Doc....

http://www.benefitslink.com/boards/index.p...=23&t=19772&hl=

[Blinky the 3-eyed Fish]

Not a chance. This is a very conservative design with only 2 rate groups. If this was deemed a nonqualified CODA, then every cross-tested plan sans a select few would be.

[Tom Poje]

Andy -

give a man a fish, feed him today

give a man a hook and feed him forever.

Ok, hopefully it won't be a 3 eyed fish!

No, really, sometimes answers, though correct, don't explain the reason why, so the person will be again asking similar questions. I always hope I can provide a 'little extra' so one can see what's going on.

(Now if I can just get cracking on the talk I am suppose to do for ASPA in the fall. sorry, it is to be of 401k, not cross testing)

[KJohnson]

Blinky,

I agree with you. However the comment by Shultz and Holland at the Mid Atlantic Employee Benefits Conference raised the issue of whether such a two rate group scenario with the sole owner/director as one of the two groups would raise a CODA issue. Tripodi raises this issue on page 1.52 of his 2003 Outline but believes that you should be able to take comfort in the determ letter.

[Blinky the 3-eyed Fish]

That is a concern only for performing the average benefit test for coverage, not nondiscrimination testing.

It looks as if the post I was reponding to by gkaley was deleted by said person. It's okay to be wrong g.

[Guest crosseyedtester]

Looking for clarification on the nonhighly concentration percentage part of the test so I can make sure I'm applying what I read in this thread correctly.

I am using Relius Administration on a plan which has 401(k), Safe Harbor 3%, and discretionary Employer contribution.

There are 1 HCE and 5 NHCE's.

Looking at the Average Benefit Percentage Test for ALL contributions, the test passes.

Looking at the Ratio Percentage Test for ALL contributions, the test also passes.

When running just the discretionary contribution, there are only 3 NHCE > HCE for a 60% ratio which does not pass.

However, the R.A. test results show the "Modified Facts & Circumstances Portion of Average Benefits Test".

NHCE Concentration = 5/6 = 83% for a midpoint of 27.75% (I still have to research that calculation.)

Since 60% > 27.75%, does that mean then that the overall test passes? Can both the ratio of just the discretionary contribution and of all contributions just be above 27.75% to pass? Does it also mean there may be room to spare (as in take away from NHCE and give to HCE)?

Thank you

[AndyH]

What are you testing, and why? Nondiscrimination or coverage? If coverage, don't you pass the ratio/percentage test? Does the discretionary contribution need to be general tested, or is it a safe harbor (i.e. same pct of pay)?

I'd be happy to comment if you would clarify what you are trying to do.

If you are general testing an employer contribution, then you have left out the Average Benefits Percentage Test, where the average % for NHCEs needs to equal or exceed 70% of the average % for HCEs. Maybe that answers your question? Passing the midpoint only lets you progress to the ABPT.

[Guest crosseyedtester]

I am attempting to allocate the discretionary contribution using a cross-tested formula and determine the maximum amount that may be allocated to the HCE. It is not a safe harbor contribution.

To do so, I thought all tests need to pass, so I'm checking both coverage and non-discrimination. I think it's very possible I have some of my test names mixed up, as well as the order of conducting the tests. Because the Safe Harbor is 5%, the gateway passes.

Ratio Percentage Test - 410(b) - tests coverage on the sum of all contributions. In this case, 4 of the 5 NHCE have an ebar higher than that of the HCE. Passes.

Average Benefit Percentage Test - 410(b) - tests the rate of contribution of all contributions. In this case, the average of the NHCE far exceeds the average of the HCE. Passes.

Non-discrimination Test - 401(a)(4) - similar to the Ratio Percentage Test above except it is only testing the employer contribution (safe harbor plus discretionary). This is where my confusion is centered.

In my example, 3 out of 5 of the NHCE's have a great ebar than the HCE. This is 60%. That is why I'm looking at the midpoint. From what you said, it seems that since 60% > the midpoint of 27.75%, the testing can be done on all contributions, as indicated above for the Ratio Percentage Test, and it passes, even though the individual test of Employer contributions is less than 70%.

Am I on the right track?

Thank you very much.

[Tom Poje]

it sounds correct, but lets make sure on the facts since not everything is 100% clear.

for example, you indicated 4 of 5 NHCEs had E-Bars greater than the HCE, but you did not indicate what test you were talking about.

for example, suppose an NHCE defers $10,000 and the HCE defers nothing. Further suppose the NHCE receives no profit sharing but the HCE received a profit sharing contribution. In that scenario, it is entirely possible the NHCE has a greater E-Bar on the AVg Ben % test than the HCE but that doesn't count.

Remember, when testing ratio % you could have 3 tests (401(k), 401(m) and nonelective)

Coverage - 410(b)

Ratio % - it doesn't matter how much someone gets, it only matters did someone get something - anything.

therefore, the fact that 4 NHCEs have a greater E-Bar than the HCE doesn't make a bit of difference.

you said plan was safe harbor. that means all NHCEs received a nonelective contribution. that sounds like 100% for coverage.

You want to cross test.. each NHCE received at least 5% so you are permitted to cross test.

You indicated 4 of 5 NHCEs had E-Bars > the HCE. If you are only talking about E-Bars involving non elective contributions, then you have a ratio percentage of 80% and don't need to proceed further. Unusual, rarely see a cross tested plan that passes this test when looking at the rate groups.

[Guest crosseyedtester]

More details, based on further review since my last post and my current understanding of the testing....

There are three types of contributions: 401(k), 3% Safe Harbor (sorry I said 5% earlier), Discretionary Employer which is allocated using cross tested methodology, looking to allocate maximum to HCE based on specific deposit amount.

The highest percentage allocated to the 1 HCE which I believe passes all the testing is 6.75%. The percentage to the NHCE is 2.8%. Gateway passes.

Since there is a Safe Harbor, no testing needed on the 401(k).

410(b) - ER contribution passes Ratio Test at 100%, so no further testing needed.

401(a)(4) - need to test the individual benefit, in this case the sum of the S/H and ER discretionary. For 410(b), that means looking at the rate of the HCE contribution, since there is one HCE, just one Rate Group. At 6.75%, the HCE ebar is such that only 2 NHCE are greater for a Ratio Percentage of 40% which fails. There is still the option to pass though if passing two more tests.

1. The Average Benefit Percentage Test on these two contributions shows the NHCE average greater than

the HCE.

2. The Modified Facts & Circumstances Test shows that 40% is greater than the Safe Harbor % (5/6 = 83% => 32.75%)

Therefore, I believe all tests pass. If I raise the ER discretionary to 7% to the HCE, the Ratio Percentage of the Rate Group falls to 20% which fails. Am I on the right track?

Thank you.

[Blinky the 3-eyed Fish]

Close, but not quite for 401(a)(4). To pass the rate group portion you only have to be above the midpoint of the safe harbor and non-safe harbor percentage (27.75% in your case).

Also, I think you should post the ages, allocation percentages (split by safe harbor and regular non-elective amounts) and testing assumptions for each of the participants. It would be unusual for the results to be so poor with the HCE getting 6.75% and the NHCE's getting 5.8%. There could be better results with testing on a contributions basis an imputing permitted disparity. What is the comp and nonelective allocation to the only HCE?

[Guest crosseyedtester]

Clarification on one issue, the 6.75% is just the discretionary portion. So the total HCE percentage is 9.75%. HCE is second from last on list.

As such here are the results, with two options, listing Age, Safe Harbor %, Allocation % and ebar w/HCE @ 6.75%, Allocation % and ebar w/HCE @7.00%:

52 3.00% 2.82% 1.89% 2.55% 1.80%

49 3.00% 2.82% 2.41% 2.55% 2.30%

47 3.00% 2.82% 2.84% 2.55% 2.71%

45 3.00% 2.82% 3.34% 2.55% 3.19%

32 3.00% 2.82% 9.66% 2.55% 9.21%

52 3.00% 6.75% 3.17% 7.00% 3.25%

[Blinky the 3-eyed Fish]

Looks right to me. You could do an -11(g) amendment to increase just the 45 year-old, but probably worth saving that (can't be a pattern of those) for another time.

[Tom Poje]

ok, you have 5 NHCEs, 1 hce

nhce concentration % = 5/6 = 83% (always round down)

this translates to a midpoint of 27.75%

This means you need 2 NHCEs in the rate group.

(2/5 = 40%)

The 2 youngest NHCEs are age 32 and age 45, HCE is age 52.

Therefore you have an age difference of 7 years between the 45 and the 52 year old.

maximum interest rate is 8.5%

1.085 ^ 7 = 1.77

so if the NHCEs get 5.8% then the HCE can get 10.26% (5.8% * 1.77)

That is without imputing disparity.

This of course only passes the rate group test, avg ben % test would still need to be passed. But it does provide the maximum the HCE can receive.

[Guest crosseyedtester]

That's an interesting test which I'm not famliar with. What is it called? I don't see it in the output from Relius Administration.

Thanks.

[Tom Poje]

wich test are you referring to? Relius does them all.

the NHCE concentration % test is hopelessly buried on the rate group report

[Guest crosseyedtester]

Your calculation of 10.26% (5.8% * 1.77) - I'd like to see that in the Relius output for the NHCE Concentration Test. I see the 83%, the Midpoint of 27.75%, the Rate Group Ratio of 40%, but not a comparison of 10.26% with something else. I thought it was good enough that the 40% was greater than the 27.5%?

Thanks.

[Mike Preston]

Blinky, pattern of -11g amendments are only restricted for b,r&f's, not for amounts testing. Somewhere in the vast archive of BenefitsLink this has been pointed out before.

[Tom Poje]

There is no calculation of the 10.26%. that is simply a mathematical conclusion based on all the facts.

you had 5 NHCEs. if only one was in the rate group you would be at 20%, and 2 NHCEs would be 40%. so you are correct, all you need is 40%.

So the question is, how do I get 2 NHCEs in my rate group so I pass testing?

A E-Bar is a conversion of the contribution to a benefit - therefore it is tied to the interest rate, and the age difference.

I used a simple case ignoring imputing disparity, plus only 1 HCE.

Suppose you had more than 1 HCE.

You would need a report (and certainly programable since one can discuss it)

that says

HCE 1 needs x NHCEs

HCE 2 needs y NHCEs

etc.

Then the report needs to indicate the age of the 'oldest' NHCE in each group compared to the age of the HCE looked at.

[Blinky the 3-eyed Fish]

Mike, you are correct. I had wondered where you were. AndyH thought you were dead, but I just figured you were wounded in a remote area and left for dead.

[Mike Preston]

You are both correct.

Hope to see you guys at ASPA.

[AndyH]

Well, we can't see Blinky unless he divulges his true identity,

or unless he actually looks like that!

[Blinky the 3-eyed Fish]

I won't be there . Not enough lakes.

[Guest crosseyedtester]

Tom -

I think I see what you are doing.

Once you determine the number of NHCE's needed (in this case 2 since 40% > 27.5%), you need to determine the maximum contribution percent to the HCE (along with the minimum to the NHCE). This is determined by taking the youngest NHCE's and putting them in a group, then the oldest age of those is used to calculate the required minimum contribution percentage to the HCE (in this case 10.26%). This would help determine the passing allocation percentages which allows a maximum to the HCE. As it turns out, I've only worked on formulas which involve using the highest passing .25% to the HCE, so never had to determine the actual highest passing amount in a such a manner.

Thank you.

[Tom Poje]

correct.

Now, look what happens to your plan if the youngest NHCE quits. not pretty, is it.