Tom Poje

I think I would amend so keys are not eligible. based on your description it sounds lik they have to follow the terms of the document. plan is top heavy a key person deferred therefore, must provide a minimum to all ees. ugh. there is no provision I know of that says 'if there are only keys, then plan is not top heavy' well, I guess you could make the plan safe harbor and exclude hces from the safe harbor. I guess that would be deemed not top heavy.

even more important,or at least just as important, what is the definition of comp? is it whole year or date of participation?

It is my understanding that it works, quite possibly it is only on the most recent update. I get everything jumbled up in my head. if the plan has provided the gateway minimum (e.g. an additional 2% nonelective) than all NHCE with SSRA = 67 will have an E-BAR > .65 on the 2% piece anyway, so they wont matter- at least not until 2007, if I did my 'ciphering' correctly. That leaves you with ees who are age 50 or older that might not get the full disparity. hopefully you aren't relying on them to pass rate group testing since their EBARs will be very small. they might effect the avg ben % test, but again not much. Still, of course you should always be concerned, especially if you have provide a smaller additional nonelective contribution.

and predictions for next year: $210,000, this based on the CPI numbers

as long as you are providing a SNEC or a SHMAC, you can provide any additional nonelective contributions you desire.

the theory why e bars are the same can be explained mathematically. gotta love it! hopefully I have made it easy enough to follow in the enclosed

why is your goal to have everybody with the same E-BAR? There is no requirement for this. In fact, when you test, you could actually impute disparity ( a bizarre concept in an age weighted plan, but years ago Iread an article about actually setting up such a plan) you could use different testing assumption (e.g. different interest rate or mortality table) your first step is to follow the plan document. next step is to pass nondiscrim testing. you always pass if EBARs the same, there is a chance you fail if they are different.since it is the owner you are talking about, he is am hce and you indicated he had a lower EBAR. thus you will still pass.

on the report printout for whatever reason it will show all people within a range at the lowest point, rather than at the midpoint. rather confusing, not a great way of showing this. dang, cross testing is confusing enough without having a report adding to the confusion.

let midpoint = X and Hi point = Y then X + X(.05) = Y or 1.05(X) = Y so Relius simply sets the HCE at Y and figures backwards (Or at least that is what I understand) now, as for whether you can use rate banding, it does not matter if it is a small plan or not. Rate banding is allowed, but the Relius default method does not seem to be in sync with the regs that says you can rate band as long as it does not significantly favor the HCEs. setting the HCEs at the Hi Point would seem to be in conflict with this. if the HCE is at 8.0 and you have an NHCE at 8.2 and another at 7.6, that is probably a legitimate grouping with a midpoint of 8 since you are bringing an NHCE up and bringing an NHCE down. Relius default would be Hi point 8.0 then midpoint = 7.62 (8/1.05) which means lo point = 7.24 (5% this side of 7.62) depending on how close your EBARs are it might help to use 7.5 Post Interest, 1983 IAF as mortality. or possibly age definition nearest (or last) depending on whether the HCE was born in the first half of the year.

there isn't anything per se, as close as you will come is 1.401(a)(4)-4(e)(3)(iii)(F) right to each rate of match...also treating different rates as existing if they are based of definitions of comp or other requirements that are not substantially the same. Now, it is generally easier for someone who makes 100,000 to defer a higher percentage than someone who makes only 20,000. that is simply a fact of life. Can you 'mathematically' prove it? I'd argue if over a 5 year period that only a small percentage of NHCEs defer more than 3% that seems proof enough that effectively you are shutting out the lower paid people. But I think it is a gray area, and unless you have really ticked off the IRS agent it probably wouldn't be an issue. I don't recall anything in the proposed regs, but since they are cracking down on blatantly odd QNEC formulas I am not sure whether I would write a document in which the match % is greater the more you defer.

scenario 1: I believe you are correct, though the group of ees receiving 50% would include those receiving 100%, since that group received 'at least 50%'. as for scenario 2 - I would agree 'effectively' it might not be available to all, but that is awful hard to prove (though in reality we know that probably only HCEs can take advantage of it). I suspect unless the ACP test failed it wouldn't be an issue. On the other hand, there is precedent for the IRS cracking down on such formulas, in light of the safe harbor rules which say you can't have a formula that increases as rate of deferral increases. In fact, if I was to argue against such a match formula that is what I would point to, that for effective availability the IRS has said NO to such a formula.

you have said a mouthful, so lets get some things straight, and a bit of history as well. Back when I worked at Corbel they hopped into the office one day and said could I look at the QNEC. they wanted to add it to the system but were having problems. I actually figured the logic out over dinner, and the original reports on the pentabs system were labeled QNEC ADP and QNEC ACP. Actually, it was more complicated back then because you had the multiple use test to worry about. when the system converted from Pentabs to Quantech (now Relius) someone 'decided' that a QNEC ACP, since used in the ACP test, was the same thing as a QMAC. As I said I gave up years ago getting them to change the name to accurately describe what it is. The system simply doesn't calculate a QMAC at this time. Now, for example HCE = 6% and NHCE = 3% and there are 6 NHCEs. to pass the NHCEs must = 4% in total. that is 1% per NHCE. But lets say only 4 of 6 are eligible. Then that 1% must be increased by a factor of 6/4 or 1.5% to those eligible. the system then multiplies that figure by the comp of those 4 ees and says here is how much you need. All I have done for the QMAC is simply take the $ amount of QNEC and reconvert it back to a % and then readjust it to a new % of those who are deferring only (and also eligible for a match) lets say one of these people had a real low compensation. In that case the QMAC could come out less than the QNEC. If you are talking about people with a higher average of comp then it would increase the contribution. The second condition you noted is usually referred to as a bottom up QMAC. this is usally the least expensive, and will be eliminated as soon as the proposed regs are out in force. lets go back to the original example. I needed an average of 4% for 6 NHCE or a total of 24%. I already have a total of 18% (6 nhce* 3%) that means I need 6% more to pass. I could increase 1 person by 6% and pass. or if it is just those deferring and eligible for the match (lets pretend it is 3 people) then I would have to raise each of the 3 by 2% (3 * 2% = 6) hope that helps.

terminees are not normally eligible for a top heavy, so no, they should not get a top heavy, unless you have a strange document requiring it. everything else being equal, you are probably ok. If you have HCEs that are non key, then you would have to run a test proving that coverage would pass (just in case you have a flock of NHCEs that quit with greater than 500 hours.

first, I suppose, you need to be sure on the terms being used. what relius calls QMAC is really a QNEC for the ACP test. I gave up years ago trying to get them to make a cosmetic change. now, to solve for the QNEC: take the suggested QNEC amount / Total comp for those elig for QNEC this would be the % of QNEC now take this % * # elig for QNEC / # deferring and elig for Match this will give you the % match required to pass. you could run this as a discretionary match. you would have to code HCEs as not eligibile for match (override) If your desire is to pass ACP test, then simply use what relius calls suggested QMAC in the above formula. this should actually be easy to create / modify a crystal report for, but I have no time at the moment. ................... another method is to simply code your HCEs as not eligible for match override and run a discretionary match. check your results. Increase or decrease your discretionary match as needed. you should be able to zero in on an amount in 3 or 4 tries. no need to run eligibility - just the transactions.

in addition, a SHNEC is subject to the same withdrawal restrictions as a deferral. the age weighted allocation, though 100% vested would not be

for coverage you are at 100%. A safe harbor is still a nonelective contribution. By the way, the answer would be the same if one ee received top-heavy only andeveryone else received more. Now for the nondiscrim testing. each and every NHCE, no matter how much nonelective they received, must receive the gateway minimum if the plan is to be crosstested. Doesn't matter what eligibility conditions you have.Thus if the HCE receives more than 9% nonelective(total, including the safe harbor), you must give the terminated NHCE at least 1/3 the rate the HCE received up to 5%. You must also follow the terms of the document. If the document doesn't call for this additional contribution you must put in a corrective amendment for the year. If the plan is not cross tested, then you would not have to bump up the amount.

Something is amiss. the only terminees that can be excluded from coverage testing would be terminees with less than 500 hours. And only if they receive zippo. (I am not referring to the 401k portion of coverage, of course)It doesn't matter whether plan is safe harbor or not. If you have a terminee with more than 500 hours he is includable and either benefitting or not benefiting depending on your plan requirements. Since the SHNEC goes to all ees, your coverage for nonelective portion of the test is 100% vested. all you care about in coverage is if someone got something, not how much. Part 2: nondiscrimination. Basically same as above except now you are concerned with how much someone gets rather than simply did they get something. excluding terminees with less than 500 hours is entirely optional. if plan is safe harbor, then terminees who are excluded from the remaining profit sharing allocation MUST be bumped up to the gateway minimum. Make sure document contains such language. I believe Corbel has a snap on amendment to cover this, if that is the document you have.

basically a person within 14 or so years of retirement who is given an additional 2% ps contribution to satisfy the gateway minimum will have an EBAR of .65 on that 2% alone. therefore that person will be able to impute at the maximum anyway. usually to pass testing you are not depending on a 51 year old to help pass testing.

Again though, unless the proposed regs are modified the following is what the IRS desires. (desIReS) This is from the preamble The proposed regs would provide that a plan MUST provide for satisifaction of one of the specific nondiscrimination alternatives.....similarly, a plan that uses the safe harbor method must specify whether the safe harbor will be the nonelective safe harbor contribution or the matching safe harbor contribution and it is NOT permitted to provide that ADP testing will be used if the requirements for the safe harbor are not satisfied. The safe harbors are intended to provide employees with a minimum threshold in benefits in exchange for easier compliance for the plan sponsor. It would be inconsistent with this approach to providing benefits to allow an employer to deliver benefits to NHCEs and revert to testing. ......an employer could adopt a section 401(k) safe harbor plan which has a contingent non-elective contributions, provided the employer notifies employees of the contingent arrangement before the start of the year, AMENDS the plan to provide the nonelective contributions no less than 30 days before the end of the year. ..... Based on that, I don't think the IRS likes the 'flexibility' approach, rather they are looking for something more permament

you aren't the only one confused. It would be my understanding (at least based on the proposed regs) that if the language is in there, then the plan is safe harbor. of course the proposed regs aren't in effect yet and I imagine there are a lot of documents they say things are driven by the notice. again, this only my understanding of how things are suppose to be done. For example, plan gives notice 12/03 it may go safe harbor in 2004. It has until 12/04 to make that deceision, and if it does, the document is amended to contain safe harbor language. it also gives notice on 12/04 it might be safe harbor for 2005. on 12/05 plan would have to be amended to remove the language if it was decided not to go this route. Of course I am talking about the SHNEC not the SHMAC. yes, Corbel documents say its safe harbor, and then I think the SPD said 'if the company so chooses' again, my opinion (and quite possibly my opinion ony, but I think I actually read this stuff somewhere else) this is not what the IRS wants. it either is or isn't, which seems to make the notice an information reminder rather than something that drives the whole affair.

kangerooster: welcome aboard. I think you have it correct. the SNHECs count as a nonelective in determining the gateway minimum. I will go out on a limb and guess that some people may be interpreting the part of safe harbors that says 'may be used'... as meaning one doesn't have to. (e.g. a match 'may be used' to satsify top heavy, but you don't have to) Why you would want to ignore the SHNEC and give the NHCEs more is beyond me, especially in a cross tested plan is beyond me.

to pass nondiscrim a plan must pass either 1. ratio percentage test for each HCE or 2. average benefits test there are 2 parts to the average benefits test. 1. average benefits percentage test. 2. nondiscrimination classification test to pass nondiscrimination classification test, the ratio percentage must be greater than the midpoint percentage(based on the NHCE concentration %). so, yes it does matter what the ratio percenatge of each HCE is, because each of those ratio percentage must be greater than the midpoint %. If one of those failed(even with a mess of HCEs), then you fail the average benefits test. you indicated that test passed, so it sounds like there is a misunderstanding about what is involved in testing. especially since you said that deferral were 'brought in'. so lets go back to the average benefits test there are 2 parts to the average benefits test. 1. average benefits percentage test. (includes ALL contributions except catch ups and after tax) 2. nondiscrimination classification test (Only includes nonelectives)thsi is the rate group testing. each HCE's ratio % must be greater than the midpoint % somehow or other I am suppose to condense all this stuff into a 75 minute talk at ASPA in the fall. Ha. It would be easier to explain the theory of relativity in that time frame.

in order to even be able to cross test a plan must 'pass through the gateway' one way to pass through the gateway is provide a 'nonelective contribution' of 5% (or less if the HCE has less than 15% nonelective) Nonelectives = just what it says. nonelectives. Is a QNEC a nonelective? yes. is a SHNEC a nonelective? yes is a QMAC a nonelective? no, its a match is a SHMAC a nonelective? no, it also is a match.

I would agree - you have to follow the terms of the document. It would be hard to imagine a scenario in which something other than that would come into play anyway. consider a plan with an HCE who received a 3% match and an HCE who received a 3% match. are you considering skipping step 1 since both received a top heavy minimum? In that case the HCE would receive 3% integrated. at that point you have a nonelective contribution 0% base + 3% excess. hardly seems legal. Or are you trying to argue since both received a minimum you will allocate 1.5% base plus 1.5% excess. I guess you could write a document this way (not what yours says), but even this would 'fall apart' if you had an NHCE who received 0 match, because you would still have to provide the 3% to start with.

well,1.401(k)-1(a)(3)(iv) talks about 0ne time elections. They are NOT treated as a cash or deferred election. irrevocable, and only if first becoming eligible under ANY plan of the employer. (Therefore, if there was a prior existing plan, it is too late)(see example 3 that follows in the regs) ee can not participate in any plan that might be established in the future as well. If it is an actual one time election, then the election is not considered to be a decision to defer 0, but rather, the ee has elected out of the plan and therefore is not eligible. [safe harbors only need go to eligible ees.] I would assume based on the regs to read that if someone elected out of the plan (but would not be considered a one-time election) then such an election would be considered a cash or deferred election. Or at least that is how I would understand it. ........ Perhaps another way of looking at it, if you excluded all hourly people from the plan, would you provide a safe-harbor to them? They are not eligible, so I wouldn't think so. You might not pass coverage, but...