414s Testing

[justatester]

We have a client that fails 414s testing for the compensation definition they use for the profit sharing allocation. For the 2011 plan year, the client opted to run the general test to prove the compensation(based on 415 comp) was non-discriminatory.

Since the ps compensation excludes large amounts of compensation, we are anticipating that the plan will fail again. (It was not remotely close in 2011).

The question is: Do they need to run the general test again-since it was being completed to prove the compensation was not discrimintory as opposed to the formula? Or would you be able to rely on the 3 year rule? I am saying you need to run a 2012 General Test to prove the 2012 compensation is not discriminatory. How can you rely on 2011 test results to prove the 2012 compensation is non-discriminatory--especially since the 414s test has an annual requirement?

Any thoughts or regulation references would be greatly appreciated!

[ETA Consulting LLC]

I think you may be using "compensation" too many times.

Let's suppose:

1) Your providing a uniform allocation to all employees (prorata)

2) However, the definition of Compensation used for providing the allocation is not safe-harbor (excluded bonus & overtime)

3) When you run the Compensation Ratio Test to show the compensation is non-discrimintory, it fails.

From here, you have the option of testing the actual "Contributions" provide to each employee against 415 Compensation (or any defintion of Compensation that actually satisfies 414(s).

Since you're having to provide additional testing (on a prorata formula) due to the fact Compensation is failing 414(s), you may consider revising the Compensation used to make the contributions in future years.

Is this consistent with your situation or question?

[justatester]

The client has chosen to the option of testing the actual "contributions". I assume I would need do the General Test baseed off the 2012 compensation and contibutions.

The client is asking whether or not they can rely on the 2011 General Test results to satisfy the 2012 414s requirement.

[ESOP Guy]

I THINK the answer is yes. I had one client in this situation in all my years and their attorney signed off on the idea of doing the test once every 3 years as long as the situation from year to year was rather consisitant. That is the rules in the regs they are clear if something material changes to the company/census during the normally non-test years you would test again. You can do both a 414(s) and a general test once evey three years is my understanding. It has been a few years since I have done that with any of my employeers For one thing my current firm has great testing software that makes doing an annual 414(s) or general test so easy we just assume do it annually.

Having said that I would advise on something as agressive as this to get the ERISA attorney who is going to have to defend the plan before the IRS or DOL to agree to this. An ERISA attoney gets paid more per hour then a TPA firm for a reason and one of those reasons is they get to make these kinds of risk calls. Higher reward = higher risk that is how the market works. (I will stop my editoral now.)

Edits:

Fixed a few typos to make what I was saying more clear

[justatester]

I was under the impression that 414s is required annually. Based on this, if the test fails, would I run a 2012 General Test (based on 2012 comp/contrib) OR can I use the 2011 General Test?

[ESOP Guy]

Justatester:

See sections 1 and 5 of this rev proc 93-42

I am having a hard time finding a link to a copy of the rev proc but if you read it I think it is relevant.

I am happy to be told I am wrong on this issue but I think I am right.

[justatester]

Ok..I finally found Rev Proc 93-42. And I see where you are saying about the 3 year testing cycle....I guess I am confused.

In Sal's book, if you look and 410(b) & 401(a)(4) sections, you can clearly see a reference to 93-42 and the 3 year testing cycle. However, the sections that talk about 414(s) test talk about annual requirements and there is no reference to the 3 year cycle. If fact, there are several references to each year and annual requirements. .

Was there another rev proc that changed 93-42?

[rcline46]

In my opinion, any plan that passes on a general test depends on demographics, and that changes every year. And getting there due to a 414(s) failure would push me to test every year. Not testing is a false savings. (editorial comment coming-) I would think it a sign of a professional to do the testing to prove the plan was in compliance rather than just assuming it was in compliance.

[Mike Preston]

The purpose of 414(s) compensation is to use it in a non-discrimination test. If you do a non-discrimination test in Year X and that satisfies 93-42 for years X+1 and X+2 then there is no need to analyze anything with respect to 414(s) in years X+1 or X+2. What am I missing?

[ESOP Guy]

Ok..I finally found Rev Proc 93-42. And I see where you are saying about the 3 year testing cycle....I guess I am confused.

In Sal's book, if you look and 410(b) & 401(a)(4) sections, you can clearly see a reference to 93-42 and the 3 year testing cycle. However, the sections that talk about 414(s) test talk about annual requirements and there is no reference to the 3 year cycle. If fact, there are several references to each year and annual requirements. .

Was there another rev proc that changed 93-42?

I don't believe 93-42 has been superceded.

[ESOP Guy]

In my opinion, any plan that passes on a general test depends on demographics, and that changes every year. And getting there due to a 414(s) failure would push me to test every year. Not testing is a false savings. (editorial comment coming-) I would think it a sign of a professional to do the testing to prove the plan was in compliance rather than just assuming it was in compliance.

I don't think one is assuming the plan is in compliance. The rules are clear that unless a major change has happened to the plan then 93-42 says once every three years meets the rules.

I am always conservative on making sure a major change hasn't happened. Or maybe a better way of thinking it is- if I have any reason to believe a major change happened such I can not rely on 93-42 I recommend a test. Then the client decided if they want to pay us for the test or risk it.

[Tom Poje]

e.g. if the plan has 5 HCEs and gains or loses one, I would consider that a significant change

[Mike Preston]

I would let an ERISA attorney adjudicate the wording of 93-52.

[LarryDavid]

As kind of an aside to this topic, is there anywhere in the Regs where it formally states that the method used here to prove compensation is nondiscriminatory is allowed (i.e., that instead of doing compensation percentage test under 414(s) you can instead just run the General Test using a 415-definition of compensation)? I've always assumed this was true so I'm not questioning it per se, however I now have a colleague that is asking me to confirm that we can do this by citing something formal in the Code or the Regs, and I am having trouble doing so.

Anyone know where it says we can do this?

[Tom Poje]

well, Code Section 414(s)(3) says you can use alternate definitions as provided by the Secretary

1.414(s)-1(d) discusses other definition including satisfying nondiscrim test of comp (e.g. comp test)

1.414(s)-1©(2) discusses comp within the definition of 415©(3)