Jump to content

"Super-integrated" plans - looking for an example formula


Guest Elizabeth Gaskins

Recommended Posts

Guest Elizabeth Gaskins

i need information on super-integrated plans. maybe an example formula or something. i don't understand them or how the allocation works. from what i do understand of them i think they might work well on some of the plans i administer. either super-integrated or new-comp. but cross tested plans are my weak point. also, if anyone has info on seminars/web-casts/resources that are given on plan design please let me know. i can't find any. Thanks!

Link to comment
Share on other sites

ugh, there is no easy explanation to what you have asked. At the ASPA conferences there is usually a session on cross testing, and I am sure the same applies to other such groups.

My experience has been you have to have at least part of clue what is going on to make such sessions even worth while. (but please don't take that as a comment to discourage you if you have never looked at one before.

As a general rule of thumb, if you have aa average 10 year difference in age between the HCE and the rank and file, you have a good candidate.

Once you hit a difference of 15 or more, you have an ideal candidate. If all the employees are 18 years or more difference, an age weighted plan is the best way to go in a purely profit sharing plan.

But it all depends on how many HCEs you are talking about in a given plan. Depends on whether you are talking about a plan that also has a 401K feature, etc.

The ERISA Outline Book discusses a lot of the concepts and there are some examples as well. If you are doing anything in pensions that is a good resource to have anyway.

I've been asked to add some material to the Coverage and Nondiscrimination Answer Book, so I take question like you ask and try to work them into the book. There are already questions like

What are the steps in the cross-testing process, how is the ratio percentage test performed,etc.

Link to comment
Share on other sites

Super-Integrated Plans generally work as follows:

1) Everyone gets a base percent of compensation (usually 3%)

2) Funds then spill over to "fill" the next allocation (say 25% of Compensation in excess of $100,000 - hence "super" integration) with the intention of attaining most, if not all, of the 415 limit for your highly paids; then

3) If further funds are remaining, then allocated among all based on compensation (with exception of those already at 415 limit). Generally this amount raises allocations to level necessary to pass General Test on "cross-tested" basis.

This design was generally a reaction to the IRS's mid 90's position that the Class Allocation (where the Employer specifies various classes in document and then allocates contribution among classes) method was not "definitely determinable". Super Integration was developed to deal with this objection. I have one problem with the Super Integration approach, however. Look back at the method; say you calculate the optimal amount of contribution that gets your favored participants to 415 and allows you to pass the General Test. What happens when client may want to put in a lesser amount (or the amount exceeds the 15% IRC 404 deductible limit)? The above approach to allocation doesn't handle this real well, to put it mildly.

The IRS has since backed off of their objection and is allowing class allocation. The few clients that we have who took the Super Integrated approach we have amended to the Class approach. This is much cleaner and allows you to achieve goals of maximizing contribution while passing General Test without amendment to the plan each year.

Hope this is of help.

Link to comment
Share on other sites

Guest Elizabeth Gaskins

thank you mwyatt!!! that clarifies things so much for me. one question.... is class allocation your term for new comparability?

Link to comment
Share on other sites

You are correct (sorry for adding differing terminology). What I meant by "class allocation" was the typical New Comparability design (i.e., you specify different classes of employees, and then the employer provides written instructions to the trustee each year saying "The total contribution for the year will be $X, with $Y going to Class A and $Z going to Class B.")

I find this method to be the best for a number of reasons, as you basically determine the optimal contribution each year, and then get to specify the exact dollar amounts necessary to make it work without having to amend the benefit formula (with all of the problems that accompany multiple amendments). The only things that you may have to watch out for are census changes that necessitate adding or modifying your classes (young HCEs - the bane of any New Comp plan ;) ).

Link to comment
Share on other sites

other 'watch-outs'

you generally want to make sure there is a last day provision in your document so you can amend the current year if need be

if a class is defined as 'owners', make sure it is owners not by attribution. kids mess up a plan faster than anything.

make sure you bone up on testing options such as 'grouping accrual rates' and 'testing age = SSRA'. Yes, the regs permit these assumptions, and I have seen software use them, but use them incorrectly. (or maybe I should say, allow the user to simply click on options and expect the system to handle everything)

Link to comment
Share on other sites

Tom, the last day employment requirement is an interesting take. I prefer just the opposite-no last day. I just find it causes undesirable and unpredictable test results, because those people become 0%'s.

I'd like to hear you elaborate on why you prefer to have a last day requirement, i.e. your experience with the tradeoff.

Also, your comment about testing using SSRA. How can you do that if NRA is uniform? Even if NRA isn't uniform, I didn't think SSRA was generally accepted as ok for testing age. I thought it was a gray area that most shied away from.

Link to comment
Share on other sites

your comments on SSRA as testing are correct. the problem is that, for example, the software I use has a check box that allows it. It helps your results, so you check on it.

legally you cant do that since your testing age is plans normal retirement age, but that is why I pointed it out. you really have to know what you are doing before checking assumtions choices.

as to last day requirement, lets say I have a class plan and my class is all doctors. during the year and young doctor buys into the business. If I have a last day provision, I can amend my plan since no one has accrued yet. e.g. doctors with 5 or mor years of service and all other doctors.

if I dont have a last day provision, then chances are someone has already accrued a benefit and I cant amend. now I am stuck.

obviously there are other factors, etc, but what you are trying to avoid are surprise HCEs.

e.g. owners kids enter the plan and class only says owners. you still have time to amend to owners not by attribution.

I know, you are lucky to only deal with clients who provide all the info you need, on a timely basis, etc. I am not so fortunate.

or another example, lets say I have 5 doctors in a class. the plan has always passed. but in one year a bunch of older ees entered the plan and now the test will fail. if I run a preliminary test in November, I discover this, and can amend to change something. It might mean putting one of the doctors in a separate class, and giving him slightly less, but at least I can do it.

Link to comment
Share on other sites

Good discussion on "last day" usage. I do tend to fall w/ Andy H's observation of problem of generating 0% rates for terms. We've generally dealt with fairly small plans with 1 year eligibility, so that new hires during the year we're working on can be analyzed not too far into the next year so that if demographic changes look to be a problem in next year we do have time to make changes. This of course assumes that client gets you the data not too far after the end of year.

Another item for thought: when I was at the EA meeting last year, one presenter had the interesting idea of actually having all employees in the plan with a very minimum $ amount (say $100) guaranteed to all participants. In some ways this is kind of a sneaky "gift" as these part-timers aren't earning any vesting service, so unless they stick around to NRD they walk with nothing. But consider the accrual percentages that can be generated from a bunch of part-timers earning minimal dollars with a $100 contribution. In some circumstances (his example was the typical dentist office with rafts of PT employee) the extra minimal cost for these folks may generate much more favorable results. Can't say that this approach would be appropriate for all but made for an interesting discussion...

Link to comment
Share on other sites

Very interesting ideas. I wonder how one would put such a thing in a document. I frequently see flat dollar minimums in discussions, which in theory is great, but again how does that fit in a document?

I asked Joan Gucciardi that this October and she said it's most common application of a flat dollar minimum would be in a corrective amendment, but obviously that doesn't work for nonvested terminations.

Link to comment
Share on other sites

Guest pineapple

I like the idea of including PT employees to increase %, but how do you address new cross testing rules (1/3 or 5% gateway)? My understanding is that all participants, including those otherwise ineligible (i.e., part-timers), must meet 1/3 or 5% gateway.

Link to comment
Share on other sites

pineapple:

I think its 'all participants who receive a profit sharing contribution will have to be bumped up to receive the gateway.

If an individual is ineligible, he wouldn't be in the plan anyway. If you bring parttimers in, then they will get the minimum, but the $100 might be more than the minimum anyway.

It is possible in a class plan to exclude someone from the profit sharing, and he would not get the gateway unless the plan was top heavy, then since he benefits, he has to get the gateway.

or something like that, but its Friday, and its after 4:15, and I go by what the regs say...my 415 limit says I cant get anymore....

Link to comment
Share on other sites

WARNING: LONG POST.

The new cross-testing regs have an interesting provision. I seem to recall that this provision has been discussed somewhere on this board in the recent past. The preamble to the regs makes it clear that one can separately test under 410(B) and therefore end up separately testing under 401(a)(4) those that meet statutory eligibility and those that don't.

Consider the impact of this provision on the $100 minimum benefit provision to those who are not statutorily eligible. If the $100 minimums are being provided to those that are not statutorily eligible, they don't seem to require the 5% gateway in order to enable cross-testing.

My take on this is that application of this provision is going to "potentially" (see below) assist in passing the average benefits test.

However, it doesn't seem like the two groups can be tested together for 401(a)(4) if one goes down this path since they are being disaggregated for 410(B). Hence, while these people will "potentially" help in the average benefits test, they don't seem to help the rate group testing.

But there is a big caution that I have to throw in here. And that is why I used the word "potentially". There are some people who believe that if a plan tests 410(B) on the basis of disaggregating the statutorily eligibles, then the average beneifts test must be run separately as well! Let's not get into the argument as to whether they are right or wrong for the moment. Let's just consider the impact if that argument is true. In that case, the $100 minimums end up neither helping with the rate group testing nor with the average benefits test. Well, at least they don't HURT the testing!!!

Now, the IRS has stated, from the podium, and in writing, that there is only one average benefits test. See the first question and answer from the most recent ASPA annual conference.

1. Plan has immediate entry in 401(k) and 21/1 for profit sharing allocation. When running the average benefits percentage test for pupose (sic) of the general test how are employees with less than a year handled - i.e. are they in the ABPT?

A: Yes, they are in the ABPT.

Does it matter if we have bifurcated for ADP testing or not?

A. No, not for the ABPT.

Pretty clear that the IRS folks in Washington think that there is only one ABPT. So that seems to indicate that "potentially" is more like "most likely".

But I think that they are wrong. Getting into the nitty gritty one has to examine the definition of testing group for the ABPT. See 1.410(B)-7(e). If somebody else wants to read that section and let me know where I'm misinterpreting things, that would be appreciated. The key two sentences in that regulation are numbers 2 and 3.

Sentence 2 reads:

For this purpose, the plans in the testing group are the plan being tested and all other plans of the employer that could be permissively aggregated with that plan under paragaph (d) of this section.

Sentence 3 reads:

Whether two or more plans could be permissively aggregated under paragraph (d) of this section is determined:

(i) wihtout regard to the rule in paragraph (d)(4) of this section that portions of two or more plans benefiting employees of the same line of business may not be aggregated if any of the plans is tested under the special rule for employer wicde plans in Section 1.414®-1©(2)(ii).

(ii) without regard to paragraph (d)(5) (the same plan year requirement) of this section, and

(iii) by applying paragraph (d)(2) of this section without regard to paragraphs ©(1) through ©(2) of this section."

Whew! On the surface, Sentence 2 seems to agree with the IRS. Certainly it appears that the non-statutorily eligible group can be permissively aggregated under 410(B). However, let's look closer at the (iii) section of Sentence 3. What I _think_ it is saying is that you apply (d)(2) (the section that defines "Rules of disaggregaton") by looking back at the requirements for "mandatory disaggregation" defined in ©. However, I find it curious that ©(3) (the section that specifically deals with "Plans benefiting otherwise excludable employees.") is not included in (iii).

My conclusion has been that if an employer chooses to apply Section 410(B) separately with respect to a portion of a plan that benefits only employees who satisfy statutory age and service conditions, then the otherwise excludable employees who participate in that plan constitute a mandatory disaggregation population with respect to the plan being tested and therefore may not be included in the testing group of 1.410(B)-7(e). That is the group that is used for the ABPT.

Now, I have not gone into the detail that I could have, because the back references in (iii) require one to look at a spaghetti web of cross-references. But I don't _think_ there is anything in those sections that gets in the way of this determination.

OK, so what should be done if the IRS thinks one way and I (and the very small number of people I expect to be swayed by this argument) think otherwise? I think the answer is clear: submit to the IRS with a request to have them look at the 401(a)(4) testing and get whatever plan uses this approach an LOD!!!

With all that said, I'll repeat my request that if anybody can provide an analysis that indicates my logic is flawed, I would appreciate it.

mike

Link to comment
Share on other sites

I think the IRS either was wrong, misunderstood, or didn't understand the questions that were being asked in theQ & A you refer to. And Q & As don't necessarily carry weight, the IRS has changed its statement from one year to the next.

I think the regs clearly state their are two ABT in the example you state, and Sal Tripodi would agree also...see The ERISA Outline Book, 8.107 and 8.57 of the 2001 edition.

The proposed regs (34536 of the Federal Register, sec b) are clear that if 'a plan benefits employees who have not met the minimum age and service requiremnets of section 410(a)(1) the plan may be treated as two separate plans, one for those otherwise excludable employees and one for the other employees benefitting under the plan.' thus the minimum gateway need not be provided to that group of ees. However, once an ee has met the requirements (and benefits in some way for the profit sharing) they have to get the minimum. thus you could exclude an ee by class and avoid the minimum, but would still have to pass 410(B), etc...

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...