Tom Poje

it is on the excess amount unadjusted for gains/losses. see example under Reg 54.4979-1©(4)(iii)

you are allowed to choose whatever method you want. Alphabetical, based on date of hire, whatever as long as you arrive at the HCE with 0 deferrals you will arrive at what I suspect most people would do.

you might want to check http://www.aspa.org/edu/study-guide-references.htm these are corrections for all the study guides. page 40 for c-2 is amongst the corrections. (I have submitted what I consider a few other 'errors' for the study guides, haven't checked to see whether they have been posted)

I would think what you would like to do would not be possible. what you want to do is change the formula, and my understanding is that the notice provided at the begining of the year indicates a formula that will be provided for the entire year. notice 2000- 3 does indicate the formula could be reduced or eliminated, provided 15 days notice is given and that the ADP and ACP test will be performed using the current year testing. I suppose in a round about way you could 'stop' the match, and then provide an additional match after plan year end and perform testing, but that sounds like a lot of extra work, especially if the plan will be safe harbor next year. There is nothing in the regs or notices that indicate a formula could be otherwise changed, and i have never seen an example as such from anyone out there in the pension world arguing this can be done. I don't think the argument that changing the formula to include a true up could result in a greater match is valid. (While true there are other factors involved) Using the same argument an individual could argue if he had known he would have ended up with a greater match he would have deferred less at the begining of the year and moe at the end of the year. the whole idea of the notice is so individuals can make an informed decision at the start, and now the desire is to change the rules in the middle of the year. In addition, how would you handle the issue of gains? Suppose the stock market takes off. the match amount someone gets now would have the advantage of the earnings, whereas a match at the end of the year wouldn't.

It is the law, found in Code section 401(k)(2) and (10) Basically, deferrals are not permitted to be 'withdrawn' unless one has a distributable event. Being active is not considered a distributable event. Some plans will allow an option for a hardship withdrawal of deferrals, those are supposed to be limited to some type of financial hardship.( I know down here in Florida some people think just because the neighbor has gotten a new boat or pickup truck that an immediate financial need exists, but that is not the case) It is possible that a plan could have an option for in-service distributions of match and profit sharing contributions after meeting certain conditions, but it is certainly not required a plan have this option for in service withdrawal. with the penalties involved it usually isn't worth it. Remember, they are referred to as retirement plans, not credit union plans in which an individual has easy acess to the money.

my guess is there is no problem with that. conceivable once the ee has otherwise met the eligibility requirements you could fail coverage if the plan had only 2 or 3 NHCEs.

lets suppose I have a document that allows both nonelectives and QNEC(To all nonkeys) deferrals: HCE(key) 10% HCE(nonkey) 0% NHCE 0% I fail ADP test, and make a 3% QNEC. I choose to 'only use' the NHCE QNEC in the ADP test. I now pass ADP test. I test a(4) with all nonelectives. That would pass since the NHCE avg > HCE avg. I now test a(4) without QNECs used in the ADP test. I now have an HCE at 3% and no NHCEs so I fail. as far as I can tell all of the above follows the regs. in fact, I think that is why the regs are written the way they are, so as to prevent someone from taking advantage of this situation. Now lets modify the above HCE(key) 8% HCE(nonkey)0% NHCE 0% I now only need a 2% QNEC to pass testing. However, this would be insufficient if the plan is top heavy. I could provide a 1% nonelective, but then the HCE(key) would receive, and he is trying to put the absolute minimu in the plan. Therefore I still make a 3% QNEC but only use 2% of the NHCE in the ADP test. Now when I test a(4) using alll nonelectives not used in the ADP test I have: HCE(key) 0% HCE(nonkey) 3% avg 1.5% NHCE 1% 1 / 1.5 = 66.67% almost enough to pass the 70% test. but this was testing on an allocation basis. I might very well pass if I test on an accrual basis.

actually, that one has puzzled me. I am not really sure. Do I have to use a QNEC in either the ADP test or the ACP test, or can I simply designate some of the elective contributions as QNECs and not use them in either test? Why would I want to would be beyond me, except in this case where your document might force you to provide a QNEC to HCEs. hmmm. 1.401(k)-1(g)(13)(iii) [definitions] says ...qualified nonelective contributions must satisfy © [100% vested] and (d) [distribution rules] without regard to whether the contributions are actually taken into account as elective contributions under the ADP test. the definition under the (m) regs simply references you back to the (k) regs. I could easily read that to say I can provide a QNEC and not use it in the test. But I have been wrong before.

1.401(k)-1(b)(5)(i) basically says run the a(4) using all nonelective contributions plus those qualified nonelectives used in the ADP test 1.401(k)-1(b)(5)(ii) basically says run the a(4) excluding those QNECs treated as elective contributions. (similar argument applies if QNECs used as a match) and you need to pass both. so, if there are no other nonelectives AND the QNEC provided to the HCEs was not used in the ADP test, you would fail a(4) testing because under (ii) the only people with a contribution would be HCEs

if it is not an age weighted plan, but simply a cross tested plan I dont think you need an override since the contribution (I am assuming) is simply by class or group. the guy is age 66, he receives the contribution, I would use testing age 66. If plan is age weighted, I think you have to see what the document says. I know I have one that uses the same factor at age 65, but that has the language to do that.

I'm never 100% sure of anything. 1.401(a)(4)-12 definitions (testing age) says: 1. if plan provides same uniform NRA for all ees, the ees testing age is the ees NRA under the plan. 4. if an ee is beyond the testing age otherwise determined under 1, the ees testing age is the ees current age. so, I have always used #4. ERISA outline Book says 'normally regs require using current age if participant is beyond NRA. ' however, the book goes on to quote 1.401(a)(4)-8(b)(1)(ii) and says there appears to be support for using the factor at NRA. 9.30, 2003 edition. How is that for adding confusion.

I think the only thing that changes is that the APR used for that individual will be different than everyone else, but other than that not much changes.

I assume you mean the one- to -one correction method under SCP. Appendix B 2.01(b)(iv)(B)(1) states that the contribution is allocated either as the same percentage of comp or the same dollar amount to each employee. this would seem to rule out a bottom up QNEC

I would think that you could add one for the future (just like you could modify the document and switch from one type of safe harbor contribution to another) as to whether you could add a match during the current year....well, I am not sure - I would tend to agree I don't think it would be possible. How would you handle someone who deferred maximum at the start of the year and wouldn't get a match if match was from date of amendment. If match was made retroactive, you have a similar problem - someone who didn't defer at the start of the year will get shortchanged. now, it would certainly be possible to add the match, take the safe harbor on the ADP test and run an ACP test, and if plan is top heavy provide a top heavy.

maybe - it depends. you indicated plan has basic match so ADP safe harbor is satisfied. now, all matching contributions must be tested under the ACP test. The basic match on its own satisfies the ACP safe harbor. A discretionary match, if limited to 4% of compensation and up to 6% deferred. you indicated both these conditions are satisfied. so where does the problem come in? If you have an eligibility condition on the discretionary match. example - last day rule, 100% discretionary match provided HCE receives basic match plus discretionary or a total of 100% of first 3% + 50% of next 2% plus 100% of 4% NHCE quits and only receives the basic match. When you combine test under ACP the HCE has received a higher rate and therefore the safe harbor is blown out of the water. (Note: this is different than if the 3% SHNEC was provided and the only match was discretionary. currently it appears you could have eligibility restrictions. however, under the proposed regs an hours requirement or last day rule will fail safe harbor. Under the proposed regs an ee who receives 0 match due to an eligibility requirement is still considered 'eligible', and therefore an hce will have received a gr4ater rate of match.)

well, in the eyes of the IRS...hmmm. EyeAreUs... no after-tax basis is created. you simply end up paying taxes twice, no matter how unfair that is. That being the case, perhaps there is no rollover issue - after missing the APril 15 deadline the excess deferrals would appear to be treated like any other defrral. (I have never seen the issue of rollover addressed before)

maybe my brain is on hold from the weekend, but this would be my logic. you indicated the plan makes a 3% SHNEC so ADP test is satisfied. I will assume the match is discretionary, since it is 50% of the first 6% deferred the maximum match would be 3% of comp. This should satisfy the ACP safe harbor and no testing is required. since it sounds like only safe harbor contributions were made, I believe you are top-heavy free.

well 1.402(g)-1(e)(8)(iii) states that excess deferrals may only be distributed when permitted under 401(k)(2)(B) (That would be to have a distributable event ) As mentioned earlier, since this is not a plan problem (the excess deferral was due to aggregation of unrelated employers) there simply is no correction I know of - in fact why bother anyway. 1. ee had excess deferrals and pays taxes in year deferrals were made. 2. since excess deferrals not distributed timely, ee will pay taxes a second time. 3. neither plan is not in danger of disqualification. that is the only reason for using self correction. 4. therefore, whether the ee gets the distribution this year or in a later year is a moot point. 5. after April 15, there must be a 'legitimate' reason for the distribution - termination, age 59 1/2, death, disability. 6. in this case there is probably no distributable event, so therefore the $ sits until such event occurs at least that is the way I understand it. 7. employee has learned a costly lesson - or employee goes to H & R Blockhead (or similar tax preparer) and complains that the tax preparer should have noticed this problem and make them pay the penalty.

in your example anyone who made over 85,000 in 2001 would be an hce in 2002. (of course if top paid group election was made this might exclude some ees who made over 85,000. you never rely on current year data

line 6 should be 7f plus anyone who entered the first day of the plan year. but remember, unless this number is greater than 120 you can keep filing the I rather than the H

lets use your example I have 10 ees , 3 are owners 20% says I need 2 ees due to comp ee 1 200000 ee 2 200000 ee 3 100000 owner ee 4 100000 owner ee 5 100000 owner ee 6 - ee 10 I will have 5 HCEs in the plan. I brought in the top 2 by comp, and then added back any owners. If I start with the owners, I might say, I have 3 owners. Since my 20% count was only 2, I don't need to bring anyone else in. that would be incorrect. That is why I always recomend listing the ees by comp and bring in the number in the count. then add back any additional owners.

mwyatt I am confused by your statement Corrective method under plan document is to add in folks, but how do you do this when increasing those who already received a match isn't going to get me to pass 410(b)? It is one or the other. In the case of a plan failing coverage, you will have to add people, quite possibly through a corrective amendment under 1.401(a)(4)-11(g)(3)(vii) Special rule for section 401(k) plans and section 401(m) plans buried in the paragraph (it is the last sentence) provide a "....QNEC to NHCEs who were not eligible....equal to the product of the [nhce you are adding] compensation and the ACP percentage for the group of NHCEs who were eligible you have 9 1/2 months after plan year end to correct problem.

to make it simple: first list your people by comp. the top 20% with comp > than the limit are HCEs. now, in addition to those people, add anyone else who is a 5% owner

that can be done. if plan only has language for deferral and match you might have problems if the plan is top heavy. my guess is that you might even be able to accomplish all that in one plan - certainly the profit sharing plan could put people into classes. I haven't seen that done for purposes of match, but I imagine you could have an individually designed plan accomplish that.

testing comp is 414s comp, so that tosses it back to the document definition. e.g. Corbel's document now simply says 'any definition that satisfies 414s...' Years ago the definiton was more specific (e.g. from date of entry, etc) Since there are plenty of other documents out there, I wouldn't want to pass judgement on what yours may say.