Tom Poje

under the old regs it was clear you were able to shift different amounts (or %) for different people. there was no reuirement that you treat everyone equally. the new regs 'imply' that is still the case. I would agree it doesn't seem to make sense to use deferrals from those that are not in the ACP test. however, there is certainly nothing to stop you from shifting a larger amount from the other NHCEs to the ACP test. I suppose if the numbers were just right you could still fail, but without seeing an actual case I wouldn't know. good question and observation.

plans are hard enough to run without you throwing in different what ifs... all Notice 2000-3 says is that you don't have to provide the SHNEC or SHMAC to those who have less than 1 yr svc or age 21. so, logic would be as follows: plan is testing otherwise excludable separately that sounds easy enough. the allocation for the group in the otherwise excludables is: no ps safe harbor for those with at least 6 months. since the regs now refer to the SHNEC sort of as a QNEC, I think what you have is a plan with the following: 3% QNEC to all employees with 6 months of service, and a profit sharing to those with 1 year or more. my understanding of how it works: you have a 3% QNEC with no hours/last day requirement, therefore it satisfies safe harbor conditions and therefore you get the free ride if you want. well, assuming proper notice was given as well. you could not get the free ride on the otherwise excludable group since not all received the 'QNEC' but then again I could be wrong on all that, but why you would even set up a plan like that is beyond me. I am really guessing on how things would work under those eligibility conditions.

as long as the eligibility is not longer than 1 year/ age 21, it can be whatever you want. by the way, the same would hold true for the safe harbor match. The regs are real clear that you do not have to provide safe harbor contributions to 'otherwise excludables' what you can not do is have a 2 year wait as is possible for profit sharing contributions.

you are correct. go to the head of the class.

since there is a different rate of match you have a BRF (benefit, rights ,features) which would be similar to 410(b). though for example, if one formula was 75% match and the other was 100% match, the group receiving 75% would say all get (since those at 100% receive at least 75%) the group at 100% would end up being like the 410(b) hope that is not to brief an explanation if my feeble brain is working, I believe you get a free ride on avg ben % test, so you realy only need to pass safe harbor % rather than 70%.

simply that the example used in the Q and A dealt with a specific situation. a db/dc combo that was otherwise tested separately. (hence the IRS response ....in certain situations... so does that mean in other situations, lets say you only have a dc plan. are you allowed to do the avg ben % test on an accrual basis without the gateway (besides Mike, I really need help with a db min distrib question I posted in the distribution section. since the Wolverines have apparently played themselves out of the big dance, Ifigured you would have more time for pension stuff.)

well actually you are probably close to answering what I want. but I guess it would help to see numbers. plan froze 1/1/05 so there sre no more increases. to keep numbers simple. ee has benefit of 1000 at 12/31/04 he turns age 70 1/2 by 12/31/05 so RMD is 4/1/06 how much should this be? is it the actuarial increase at 4/1/06 because that is how much the $1000 at 12/31/04 is worth? logically that would make sense to me as a monthly benefit? but if taken as an annual benefit then do you adjust that figure to account for the fact he is actually taking the monthly figure as an annual lump sum earlier?

participant must start receiving 4/1 is that the actuarial equivalnce of whatever the benefit was as of that date and if one chooses to make annual payments, is it simply 12 * the monthly amount, in which case the particpant would seem to come out ahead?

I thought it said the same thing. to me it sounds like it says ''in certain situations such as aggregating db/dc plans only for the avg ben % test dont worry about the cross testing rule pertaining to gateway because you are going to test the plans separately for all other purposes (e.g. not cross test) now, by implication does that mean you have to consider the gateway for other purposes.

the preamble to the final regs says in regards to prefunding 'only if contributions are made after the performance of services which relate to the compensation.... amounts contributed in anticiaption of future performance of services are not treated as elective contributions... it does not sound as if the situation you describe falls into that category, so it sounds like you are ok

I did find the Q and A I was looking for at http://www.abanet.org/jceb/2002/qa02irs.pdf 6. §401(a)(4) – New Comparability Regulations Do the new comparability regulations apply for purposes of the average benefits percentage test? Assume an employer sponsors both a defined benefit and defined contribution plan, each of which is tested for coverage and nondiscrimination separately, but uses the average benefits test for demonstrating compliance with the coverage requirements for one or both plans. Do the new comparability regulations affect the employer’s ability to do the average benefits percentage test on a benefits basis? Proposed response: The new comparability regulations do apply for purposes of the average benefits percentage test, but won’t restrict the employer’s ability to do the test on a benefits basis in the situation described. Treas. Reg. 1.410(b)-5(d)(5) indicates that the employee benefit percentage for the average benefits percentage test is “the rate that would be determined for that employee for purposes of applying the general test for nondiscrimination in Treas. Reg. 1.401(a)(4)-2, 1.401(a)(4)-3, 1.401(a)(4)-8 or 1.401(a)(4)-9, if all the plans in the testing group were aggregated for purposes of §410(b).” Since the new comparability regulation provisions in Treas. Reg. 1.401(a)(4)-8 and 1.401(a)(4)-9 affect the rate that would be determined for the general nondiscrimination test, those provisions apply for purposes of the average benefits percentage test. However, in the case of an employer sponsoring a defined benefit plan and a defined contribution plan that are tested separately under the coverage and nondiscrimination regulations, except that benefits under both plans are considered for purposes of the average benefits percentage test, the defined contribution plan will satisfy Treas. Reg. 1.401(a)(4)-9(b)(2)(v)©, the “broadly available separate plan” threshold. The average benefits percentage test in this situation can be determined on a benefits basis. IRS response: The IRS agrees with the proposed response.

I had to stop and look up the regs just so I had the cites correct. I think what Mike is referring to would be the following: plan is going to test on an allocations basis. however it runs the average benefits % test on an accrual basis. The ERISA Outline Book says that 1.410(b)-5(d)(5)(i) which explains the manner in which benefit percentages are calculated for the ABR test does not cross reference the gateway requirement under 1.401(a)(4)-8(b)(1) I am not 100% sure on that. the regs (first sentence of 1.410(b)-5(d)(5)(i)) says ...the employee benefit percentage for a testing period is the rate that would be determined for that employee for purposes of applying the general test for nondiscrim in...1.401(a)(4)-8 so, I am simply unclear how to read that (though maybe it was answered in a Q and A somewhere). certainly the gateway rules are found in 1.401(a)(4)-8(b)(1), but the rate determination is found in 1.401(a)(4)-8(b)(2)

I should have added the following to the initial response, but I got sidetracked or overfocused on the comment about "I do not pass the 1/3 gateway allocation mark since I have NHCE's getting 0.00% and HCEs getting 1.00%' this plan would also meet the 'broadly available test' since the ratio % is > than the safe harbor %. I haven't seen that before in an actual real life situation. therefore, no minimum allocation gateway would be needed. (I have created such an animal as an example, but gee whiz, you threw me for a loop with an actual case. must be slipping in my old age.

Mike: you would agree I can set a plan up with classes and name those classes however I want. so, net effect is that you are saying this could be deemed 'unreasonable' (because you set up a 10 yr wait for ps) and therefore I can not use the avg ben test to pass coverage because I fail one of the requirements to pass nondiscrim classification. hmmmm. I'm not 100% sure on that, but you never know with the IRS.

I will give you extra bonus points for supplying sufficient information, but the logic or premise or whatever falls apart - so at that point you join the ranks of so many. lets look at the facts: 50 NHCE benefitting / 195 total NHCE = 25.6% 30 HCEs benefitting / 55 total HCEs = 54.5% ratio % = 46.9% conclusion: fail 410(b) response : no, that only fails the ratio % test. you might still pass avg ben test. avg ben test: NHCE concentration % 195 / 250 = 78% safe harbor % is then 36.5% therefore, since the ratio % 46.9 > 36.5% plan pass nondiscrim classification so now you need to find out if plan can pass avg ben % test. you could test on an allocations basis and impute disparity. this could pass and therefore you wouldn't even get to the point of having to worry about the gateway. assume that fails, so you want to cross test the avg ben % test. who has to receive the gateway? only those who recevived an nonelective contribution. so your conclusion you fail 1/3 test is incorrect since some ees are at 0. those ees dont come into play. all others received the same %, so you pass the gateway and can test on a cross tested basis.

be careful when you say 'further testing' are you talking about ADP test or 414(s) testing? As Locust noted you could use total comp in your ADP test, and if you pass you are done (assuming you also don't have a profit sharing contribution based on comp less deferral) excluding bonuses is considered 'reasonable' so then normally, step 1 would be to run a 414(s) comp test. if you pass, then you could use comp less bonuses in your testing

you say you 'do not have a cross tested plan'. I would say there is no such thing as a 'cross tested plan. There are plans with an allocation which, because the allocation is considered to be non safe harbor, they are cross tested to prove nondiscrimination. so lets look at your facts. 3 hces received x% 6 Nhces received x% 5 NHCEs received less than x%. since only 6/11 NHCE received the same % as the HCEs, you do not have a safe harbor formula. If it helps, you have a plan with two groups or classes. one group received x% and the other received less than x%. therefore, you have to prove the plan is not discriminatory. you might pass testing on an allocation basis if you impute disparity, especially if your plan is not integrated. if not, then you either have to cross test and provide the gateway to the 5 NHCEs (and still pass testing) or put in a corrective amendment and increase the contribution to one or more NHCEs to pass testing on an allocation basis.

there is no requirement that one exercise the option of testing otherwise excludables separately. and in fact, you dont have to be consistent from year to year, it is simply an option (I guess somewhat like having an option to which mortality table you use in testing, what interest rate , etc.) so, if you have some young people in the otherwise excludable group you might do better to combine all ees and simply provide a gateway rather than an allocation. (Make sure your document allows the gateway) remember also, you are only talking about the a(4) test. your avg ben % test would include all ees if you choose not to use the otherwise excludable option. however, you could still run the ADP test using otherwise excludables. again, while the otherwise excludable is an option, you still have to be consistent when performing the ADP or ACP test if using prior year testing.

generally yes. however, since a match can be used to satisfy top heavy, it is possible that some people might satisfy top-heavy without receiving a non elective contribution. I have seen some documents refer to match as a non elective contribution, but I think that is a confusing use of the term.

plan is top heavy, so all active participants must receive at least that, whether they have worked a year or not. Thus all those people have received a nonelective contribution. now, plan is broken up into 2 'plans' - statutory includables and otherwise excludables. the otherwise excludable group consists of 0 HCEs. that 'plan' is tetsed on an allocation basis (well, since there are no HCEs it wouldn't really have to be tested, but for the sake of the argument, you could have tested it and of course it would pass.) If I cross test, I have to provide the gateway. but I am not crosstesting, so no gateway is needed. The statutory includables would have to provide the gateway if cross tested, of course.

aw shucks, but maybe I was only dazzled by that purr...er fur coat you have

in all seriousness, there is not much you can do unless your 401k plan allowed loans to terminees and allowed one to pay back outside of payroll. from the sounds of your situation I doubt that is an option. e.g. you dont get a free ride, you have to come up with the $ to pay back the loan even if the company approved it. I'd go out on a limb and say with most companies that would not be a loan provision to pay back outside of their own payroll. sorry, and good luck finding something.

1.401(m)-2(a)(2) ...ACP for a group of eligible employees...

on the other hand, I have heard it said that a loan could be denied to a bad credit risk (e.g. the plan could refuse the loan if there is a reasonable chance it will not get paid back) a person seeking a hardship (depending on the reason) could be considered a bad risk for paying back a loan.

when you create a new plan year, the system will grab the highest paid people and do everything for you. (hopefully you have updated your table if there has been a change in the comp limit) if you look in census/compliance under highly compensated - prior year you will see a bunch of different possibilities. now, areas where you may have to 'switch' something - rounding - any reasonable may be used, so if 20% was 6.6, there would be nothing to stop you from always rounding down. not sure what Relius does. this might make a difference in testing. ties - if two ees have the same comp, you gotta pick someone (preferrably the guy deferring 0?)