Tom Poje

at the 2000 ASPPA Conference the question was: 26. A Company with an October 31 plan and fiscal year-end elects S status and switches its fiscal year and plan year to a calendar year basis. This results in a 2-month fiscal year and plan year. The owner of the Company (who is one of 2 HCEs) receives no compensation during this 2-month year (assume for example that he is paid on the first day of each calendar quarter). Is the owner an “eligible employee” such that he’s a zero in ADP testing? It has been argued that he is an eligible employee because he worked and was only limited from making deferrals by §415 (and the last sentence of 1.401(k)-1(g)(4)(i) says that an employee doesn’t fail to be treated as an eligible employee if he may reason no additional annual additions due to 415). Conversely it has been argued that such employee is not an eligible employee because he could not defer since he had no compensation. This will be discussed from the podium. (Supposedly the answer was it would be reasonable to treat the person as ineligible. at the 2008 ASPPA Conference Q and A 1 A 401(k) plan disregards commissions as compensation for plan purposes. One employee has only commissions, therefore could not defer. Assuming the compensation definition satisfies 414(s) testing, is this person included in the ADP test as a 0%, or would they be excluded from the test completely since they have no compensation for plan purposes? If the person is excludable, does the answer change if, for purposes of testing, the plan included all compensation (including the commissions) for ADP testing purposes? IRS Response Refer to Treas. Reg. §1.401(k)-2(a)(3)(i), last sentence, which requires that you include all eligible employees in the testing. What is an eligible employee? Per Treas. Reg. §1.401(k)-6, it is someone who is directly or indirectly eligible to make a deferral for the year. Is this guy eligible? Yes, you must include him as 0% ASPPA Counter response: Request was made for further discussion on the issue. Some at ASPPA feel an ee with no comp should not be in the test at all, rather than as a 0. .............. lets say the hardship rules didn't change, and for the sake of the argument the person could not defer for a full plan year. The regs are clear you would include that person even though they could not defer. same if the person hits the 415 limit (don't even want to know how that is possible) so saying a person doesn't have an effective way to defer could be tredding on thin ice. I guess one question would be Does it make a difference? as an interesting side note: suppose the plan has immediate eligibility to defer. Person is hired in late December, so he enters the plan, lets say 12/28/2011. now, do you include him in the test at 0 in 2011? (Since I view things as 'paychecks with a plan year, I would not)

buried in 1.401(k)-3(f)(1) ...a plan that provides for the use of the current year testing method may be amended [to a safe harbor] granted, this is in regards to a safe harbor 'maybe' situation, but it is sort of in the regs.

I understand his point. my point is the documents I looked at would seem to suggest it might depend on the document. my other point was even the FT William document could be viewed as saying its the form of benefit not the amount that is restricted. regardless, I think all would agree the IRS wouldn't push the issue of the amount, though a particularly disgruntled agent might argeu about no withholding on amounts about the minimum, if there was no withholding.

but I could read that to say "just because I hit RMD I can't create an installment benefit if the plan doesn't call for installment benefits" the document examples I provided (I believe one was Accudraft and the other Corbel -both are worded very similar) definitely say The Participant's entire interest will be distributed, or begin to be distributed, to the Participant no later than the Participant's Required Beginning Date. so I'm not creating anything. If I don't distribute the entire balance, I have to begin to distribute something and the rules only require a minimum to be distributed.

I've looked at a couple of different documents and they are basically worded as follows: (1) Amount of Required Distribution for Each Distribution Calendar Year. During the Participant's lifetime, the minimum amount that will be distributed each Distribution Calendar Year is the lesser of (A) the quotient obtained by dividing the Participant's Account Balance by the distribution period in the Uniform Lifetime Table set forth in Regulation §1.401(a)(9)-9, using the Participant's age as of the Participant's birthday in the Distribution Calendar Year; or (B) if the Participant's sole Designated Beneficiary for the Distribution Calendar Year is the Participant's Spouse, then the quotient obtained by dividing the Participant's Account Balance by the number in the Joint and Last Survivor Table set forth in Regulation §1.401(a)(9)-9, using the Participant's and Spouse's attained ages as of the Participant's and Spouse's birthdays in the Distribution Calendar Year. it simply says the minimum amount that will be distributed each year. I see nothing in that paragraph that says more than the minimum can not be distributed. the definition of required begining date is as follows: Required Beginning Date. The Participant's entire interest will be distributed, or begin to be distributed, to the Participant no later than the Participant's Required Beginning Date. that says you could distribute the entire amount once someone hits the magic date, but at the very least you have to satrt distributions and they have to be a minimum. I don't see either of them saying "You can only take the minimum and nothing more", but then maybe that is just me.

if it truly is after tax, this means the people can either (example only) after tax $5,000 earn interest and pay taxes on that interest when the money is taken out or forget the plan, put the same amount into a ROTH and pay no taxes on the interest. hmmmmmmmmm.

wow. how much dust is on that plan. haven't heard of one of those in years. was it found at the antique shop? it's not a money purchase. and there is certainly nothing wrong with having a plan with after tax contributions, though why bother in this day and age. and nothing wrong with matching after tax contributions. as I recall, everything tested using the ACP test. that's about all I remeber about those plans.

musings of someone who could be way off. I thought the idea was that when you hit the magic date you had to take a 'minimum' amount, that there was nothing to prevent you from taking more, but maybe that is document driven. 1.401(a)(9)-5 Q and A 2 says What if the amount distributed exceeds the minimum required, will credit be given in subsequent years for the excess distribution? A. If, for any distribution calendar year, the amount distributed exceeds the required minimum, no credit will be given in sibsequent calendar years for such excess distribution. there is no mention made of penalties for excess distributions under the minimum distribution rules that I can find. So that Q and A seems to me to say you could take more, but you don't get brownie points for future years and reduce or eliminate next years min distribution. The ERSIA Outline Book speaks about tax consequences - that anything above the min distrib would be eligible for rollover so should have had 20% withholding, but that is a different issue.

well, I suppose if the plan was top heavy,and since the document has to have top heavy language you could be forced in certain cases to provide a nonelective even if the document didn't permit nonelectives, but that's an exception to the rule.

I'd agree with that. In fact, that is the basis of EPCRS, that procedures are in place to prevent the things from happening over and over. And in you self correct, there is no 100% guarantee your method will be accepted. I think other factors come into play. for instance I would imagine it is less likely a problem if all employees receive the same % of contribution, or if the total contribution was something like a nice round 200,000. on the other hand if it is a cross tested plan, and its an HCE that is involved it might not look as good. still, at the end of the day, the person ends up at 49,000 for the year. if he receives 5000 additional profit sharing then he receives a distribution of 5000 and pays taxes on it. If you limit the profit so there is no excess, he is still at 49,000, and the company puts in 5000 less, so the company pays the taxes (I guess) I should have pointed out in example 22 there is a secong person with excess. And instead of distributiong deferrals, the nonelective (because the person was 0% vested) was forfeited. so same situation for 2 people, one was handled one way and the other was handled another way - both methods in EPCRS. oh well, someday soon hopefully they will release a new EPCRS (they have promised with language about correcting missed safe harbor notices, etc.) and we will see if the 415 limit example is still there.

you've lost me a little. 'top paid' usually refers to top paid group election, which applies to HCEs not to key employees.

that's the problem, I have no idea what the document language is. again, the old regs said you could write the document to say "the nonelective will be capped to prevent a 415 limit, w/o causing discrimination issues" on the other hand, if the document simply states no one will receive above the 415 limit, I think its a different issue. The EPCRS example 22 is 7500 nonelective 10000 deferral 500 after tax total contrib 18000 because of comp, 415 c limit is 15,000. so excess is 3000. the example does not say The nonelective was improper, it violated the terms of the document, you need to reallocate the excess to other NHCEs, rather the example says 500 after tax plus 2500 deferral (all adjusted for earnings) are distributed. I vaguely (but only vaguely) recall at one of the ASPPA Conferences the IRS saying if you alloacte a 5% profit sharing to employees and limit someone to the 415 limit without language indicating that is how the alloaction will be performed when someone hits the limit then you violate the terms of the document in regards to the formula.

but see examples 22 and 23 of Appendix B. no mention is made that "since deferrals were deposited before the nonelective they can't come out first" Appendix B refers back to Appendix A.08 last sentence reads For limitationm years beginning on or after 1/1/2009, the failure to limit annual additions allocated to participants in a dc plan as required in section 415 is corrected in accordance with section 6.06(2) and (3) [which would be the retun of deferrals first] so unless the plan is specific and says the 'nonelective' will be capped... this was possible under the old regs 1.415-1(d)(2)

check to make sure document doesn't contain fail safe provisions that tells you who exactly to bring into the plan without an -11g. it is an option, but some people simply check items on the checklist and up with this without realizing it.

1.401(a)(4)-9©(3)(ii) states that the min aggregate gateway cannot be satisfied on the basis of component plans. my understanding of that is before you even consider cross testing (e.g. in this case splitting the plan into component plans) the plan as a whole has to satisfy the gateway requirements.)

the only proration mentioned in the regs in regards to integration is if you have a short plan year (1.401.(l)-5) which you do not have. sounds like the person gets 5% of 200,000 + 5% (200,000 - 106,800) in addition, see 1.401(a)(17)-1(b)(3)(iii)(B) no proration [of comp] for participation for less than a full plan year......a plan is not treated as using comp for less than 12 months merely because the plan formula is based on comp for that portion which the employee is a participant.

you actually have 2 tests: 1. The ACP test - everyone, 1 test 2. BRF - each group/division must be tested. the typical example is match based on svc e.g. 25% for less than 5 years 5 - 10 years 50%, 11-15 years - 75% more than 15 years 100% match. if you have 6 groups/formulas then you have 6 BRF tests so how many people get at least a 25% match - everyone, so thats 100% formula 1 passes now, looking at group how many people get at least 50%? that means anyone with less than 5 years is included and not benefitting. etc etc.

that sounds correct, it's a matter of 1. did you work 1000 hours in a 12 month period? 2. It doesn't matter if you work 1000 in the first 3 months, you can't enter until the 12 month period is completed. the 1000 hours in a 12 month period is required by law. so some documents are written 300 hours in a quarter...but in no case will someone who works 1000 hours be excluded. so someone could work 250 hours each quarter and therefore fail the 300 hours in a quarter, but since they worked 1000 hours in a 12 month period they still enter at that point in time.

the Accudraft language for that situation is as follows: (e) Reemployment of an Employee Before a Break In Service and After Eligibility Requirements Are Satisfied. For any Plan Year in which the eligibility requirements under Section 2.1 are based on Years of Service, if an Employee Terminates Employment prior to the Employee's Entry Date in Section 2.1, the Employee had satisfied the eligibility requirements in Section 2.1 as of the Employee's Termination of Employment, and the Employee is subsequently reemployed by the Employer before incurring a Break in Service, then (1) the Employee will become a Participant as of the later of (A) the date that the Employee would enter the Plan had he or she not Terminated Employment with the Employer, or (B) the Employee's Reemployment Commencement Date, (2) the Employee's pre-termination Year(s) of Service (and Hours of Service during any computation period) will be counted for all purposes, and (3) the Vesting Computation Period and/or benefit accrual computation period, as applicable, will remain unchanged. I don't think it really varies from document to document, just how one document describes the situation might be clearer than another.

for coverage you would have NHCE benefitting 35 /120 = 29.167% HCE benefitting = 15 /30 = 50% ratio % 29.167 / 50% = 58.33% well short of the required 70% for the ratio % test.

only 'eligible' people show up in the ADP or ACP tests. so they wouldn't show on the tests, but for coverage they are includable and not benefiting (once they have met plan's eligibility) for profit sharing they show as 0 for nondiscrim and also as includable and not benefiting for coverage. the rule that applies to terminees with less than 500 only applies to participants. since they are not participants, even if they quit it apears they get counted as includable and not benefiting.

match could be dependent on terms of the document...e.g. per payroll basis with no true up at the end of the year could produce different results if one caps out at max deferral early in the process.

the actual preamble reads as follwos (if you are keeping score) “As noted above, the final regulations provide that a plan cannot take into account compensation in excess of the section 401(a)(17) limit. In addition, the final regulations provide that elective deferrals can only be made from compensation as defined in section 415©(3). However, in applying these two rules, a plan is not required to determine a participant’s compensation on the basis of the earliest payments of compensation during a year.”

technically if you have 3 groups you could have more than 3 allocation pecentages if there are top heavy/gateway minimum issues, but as a general rule that is a true statement.

I'm going to guess the accounts are set up for 'share' accounting. to me that would then make sense to reverse any transactions. change the account investment to a cash account investment. then processing/compute beginning balances to set the units = beg bal $ of course, if you have multiple accounts/investments that would seem to be a lot of work. I guess you could also set up 1 cash account (per source) and transfer all $ to that account at any point in time, and then run averything to that investment from this point. That would seem easier. just grasping without knowing more details of how things are set up and what has been done.