Tom Poje

unclear what you are really asking. (since you indicated hours, I assume this is not elapsed time. there are 2 issues - servide for eligibility, and service for vesting. in 04 person did not work 1000 hours so 0 yrs for vesting. in 05 worked 1000 hours so at the end of 05 1 yr of service. now, eligibility. 7/28/04 - 7/28/05 person worked 1000 hours, so if plan has 1 yr svc requirement then that has been satisfied, person enters on next entry date. if plan has 2 yrs svc requirement, then it depends on what the document says. most documents switch to plan year in the person's second year, in which case the ee would complete a 2nd yr by 12/31/05. yes, you do end up counting hours twice for eligibility in this particular case.

I have no clue how this report works. I didn't write it. I have no idea how the fields are even set into this report - I looked at it simply to update the report to the current level, and was totaly baffled by its format. quite the idiot I am on this one. haven't used this report in years. it will print 30 people per page (3 x 10) have no clue what labels that will be for. It won't select out people (except I see it is coded not tp print 'paid out' folks (unless I guess if you are at Relius 12, if I remember they said they are getting rid of that status to make life difficult on us folks ...er...well maybe that wasn't the reason but I understand they are getting rid of that. anyway, from report writer you can always say don't print ineligibles. Can't do anything about those without acct balances - though I guess you could put people without balances into a division and then supress that in the report.

this could be a big 'it depends'. its not really a BRF issue, but could be a coverage issue. years ago at one of the conferences, someone asked this question about the same thing in regards to a Standardized Plan, and the IRS response was something to the effect that a Standardized document can't fail coverage, so you must be ok. I do not know if the same rule wiuld hold true if the plan is not on a standardized document. and you did not indicate if this was a totaly new company, or whether the HCEs have been around awhile. if the HCEs have been there over a year, then you could always test using the otherwise excludable option and you would pass.

a little bit confused on what is being asked. at 12/31/06 you determined plan is top heavy for 2007. now I thought the issue becomes if a key ee defers in 2007 (or receives a non elective) then the top heavy must be given. deferrals do not count to determine if the top heavy has been satisfied. (e.g. a non key who defers in 2007 has not received a top heavy)

hmmmm. probably poor choice of heroes, but if the report will work for you, fine. one thing the vesting won't catch (unless you modify it) is ees who hit normal retirement (or early retirement) with less than the 'required' years of service. Because of the way I have accounts set up on my system I don't have that problem because I can pull vesting. However, the idea of modifying things slightly and making the report available to others I didn't take NRD (and ERD) into consideration.

This is an attempt (of course, use at your own risk - but someone has to try to make life a little easier) at generating the required participant quarterly notice. In order for this report to work, the user fields in Plan Specs would have to be coded as follows: Date field 26 = date vesting was last updated alpha field 29 = hours needed to accrue a year of service (e.g. 1000) alpha field 31 = vesting yr 1 (e.g. 0%) alpha field 32 = vesting yr 2 alpha field 33 = vesting yr 3 alpha field 34 = vesting yr 4 alpha field 35 = vesting yr 5 alpha field 36 = vesting yr 6 must have something in these fields, e.g. put 100% fields 4 5 and 6 if fully vested after 4 years. alpha field 37 = no if plan is not subject to permitted disparity. This report will look at ees years of service and compare to the alpha field to print vesting %. of course some modifications would always be necessary. this particular statement says you are always 100% vested in your deferral account and any rollover. If you have QNECs, safe harbor, etc you would have to add that as well.

I believe it is a BRF issue. while true, ACP test may pass and while true 'currently' all employees could get the max match, it becomes an 'effective' availability issue as well - this is where the BRF comes into play. In fact, I see in the ERISA Outline Book 11.364 (2006 edition) at 3.b.1 this eventually refernecs the BRF rules, and with effective availability there is no 'real' mathematical test - it is simply a facts and circumstances test - so how do you prove it one way or another? I suppose you end up using whatever mathematical tests are available - such as what % of NHCEs take advantage of the highest rate , how 'easily' the ACP test is passed, etc. the usual facts and circumstances see 1.410(b)-c(3)

well, lets see if I have the picture affiliated service group so all employees must be included in the denominator. you have 9 total NHCE and 12 total HCE aggregation of plans is permissive which you dont want to do. so plan A nhce 8/9 = 88.9% 11/12 hce 91.6% so coverage not a problem. plan B nonelective only nhce 1 / 9 = 11.11% (assuming a corrective contribution) hce 1/12 = 8.33% so coverage is not a problem either. gateway appears to be covered in either plan by itself. so if the 5% is sufficient to get the NHCE in the rate group of plan B that would work (if the NHCE is 0% vested then vesting must also be applied to the corrective amendment.)

the general advice from the IRS in the past is that severance pay is not to be deferred on. therefore the $ should have been returned (along with gains losses) you are correct when you say it is a gray area - some of the severance could have been due unused vacation pay or something similar. since if you were active you could have deferred on vacation pay, then it would seem there is some types of severance pay it would be possible to defer on, but at this time there are no clear guidelines.

the way I understand catch-ups is that you don't 'make' them. If however, you exceed a limit, then that amount could be treated as a catch-up. so, I guess (again, if I understand how these things work) you could avoid the top-heavy issue by writing a document that says 'deferral limit on key employees is 0%'. but you can't simply treat deferrals by key ees as a catch up without exceeding some limit - and top-heavy isn't one of those limits.

yes, in that case the HCE is out of luck unless they were also catch-up eligible well, no wait a second. since we are talking 2007, the plan could be amended to current year testing.

are you saying: 2006 NHCE who are statutory includable = none. zip. zero. non existant 2006 NHCE who are otherwise excludbale = doesn't matter, they exist in 2007 I compare the HCE with the NHCEs If I use the otherwise excludable option, then the 'plan' being tested consists of only the HCE, and by regulation, such a plan automatically passes or at least that is how I understand the rules

if I understand your question properly the following could occur. plan has usual 1 year wait, entry dates of 1/1 and 7/1 Fred is hired 2/2/05 works 1000 hours so enters plan on 7/1/06. however in 06 he only worked 888 hours. it takes 1000 hours to receive a contribution, so he could be a participant and not have an account balance.

without further guidance, I'd agree with your conclusion.

so are you saying that you have a 401(k) plan (deferrals only, no match, so no ACP) and an ESOP (which has no deferrals, but has the match, hence the ACP). hence the question regarding shifting? hadn't thought about something like that. I have thought about the situation in which an NHCE defers, but quits so gets no match, so doesn't show up on the ACP test. I don't think you could shift that person's deferrals, but I'm not even 100% sure on that.

at last years Western Benefits Conference the IRS official 'informally' indicated it would probably be possible to add a Roth feature as long as a supplemental safe harbor notice was provided. It was also indicated future guidance would be provided, but I don't recall ever seeing any. At the same time the officials expressed concern if you were to add other features - such as a match, etc. so there ceratinly is no blanket approval for adding anything. In connection with the Roth, they understand the importance of it because it starts the 5 year clock retroactive to the first day of the plan year (just the clock, not a retroactive ability to make Roth contributions), so it does make a difference.

1. if you combine plans for coverage you must combine for nondiscrimination 2. see #1 now a 401k plan is made up of 3 components. each has its own coverage (and nondiscrim test) 401k 401m nonelective in plan 1 there are no HCEs. since all nhces can defer (unless you have a real screwball document) then if you combine with plan that will have 3 HCEs and 3 nhces able to defer you have to pass coverage. everyone benefits. but the trade off is that you have to include everyone in the ADP test.

Tests: since you can't have allocation conditions then coverage should be a moot point. there is no ADP test maybe an ACP test depending on exactly what you are providing no top heavy unless you provide additional contributions. must provide notice at least 30 days before plan year begins every year. filings: same as any other plan

I'm not quite sure if you are really shifting from one plan to another. you are aggregating plans for coverage, so you must aggregate them for ADP testing. now, you are allowed (provided the ADP test passes before and after) to use the 'extra' deferrals in the ACP test. nothing has physically happened. you are simply 'moving' numbers on paper.

a. true. if plan is top heavy, anyone who is a participant and is employed on the last day must receive the top heavy. since the top heavy is a nonelective contribution (the person did not 'elect' to receive it) then the rules clearly state that anyone receiving any type of nonelective must receive the gateway as well. an exception applies if you were to test otherwise excludables separateluy. b. true, you can't have hours or last day provision for a SHNEC or a SHMAC. once a participant, you get. again, you could have the plan exclude otherwise excludables from the safe harbor. or the plan could have immediate eligibility for deferrals and a 1 year wait for all other contributions, but since it is top heavy, then you have to give that anyway, and so you are stuck with the gateway - but then again you could test otherwise excludables separately. c.its 1/3 the allocation rate of HCEs or 5% of 415 comp, which can be from date of participation. just in case you have a formula based on a starnge definition of comp. plus if you use comp from mid year entry, you have to make sure the ee received top heavy based on full year comp. a match is a match, and not a nonelective. a SHNEC is a type of nonelective, hence counts towards the gateway

yes, matches that are forfeited are not taken into account for the ACP test. would agree with you there. but that is all it says. the issue is are those matches forfeitures because they are related to excess contributions or are they excess aggregate contributions to start with and therefore don't pertain this question. so it boils down to which comes first the chicken or the egg, or rather, are you required to run the ADP test before performing the ACP test? I'd hold it is more or less at the same time, not one before the other. so for example I run my tests and ADP has an excess contribution of $1000 and ACP has an excess aggregate contribution of $500. the match was $ for $, so the related match is $1000. but wait, this was determined before any corrections were made. but that is not what the regs said. the regs said I determine if I have a rate violation AFTER corrections are made. or at least that is how I see it.

"1.401(m)-2(a)(5)(i) which tells us not to include matching contributions that do not meet 401(a)(4) in the ADP Test." my copy of the regs says 1.401(m)-2(a)(5)(i) which tells us not to include matching contributions that do not staisfy the requirements of paragraph (a)(4)(iii) of this section in the ACP Test. and that section says match is counted only if (A) allocated to the employee's account under the terms of the plan my comment "Duh, if you didn't folow the terms of the document, then why include it" (B) made on the basis of deferrals or employee (after tax) contributions my comment "Hence the term 'match' © actually paid to the trust no later than the end of the 12 month period following the year that contains that date. my comment, in other words if match was required, and you made it really late - over a year after the plan year end, then you must test it, not as a match, but as a nonelective under 401(a)(4) - which is what 1.401(m)-2(a)(5)(i) says.

If you have a failed ADP test and distribute deferrals, then you might have to forfeit match because te benefits rights and features rule clearly state you can't do that. 1.401(a)(4)-4(e)(3)(iii)(G). but those same rules also state that to determine if you have a violation, that the rate of match is determined AFTER corrections made under 1.401(m)-2(b)(1)(i) [which are excess aggregate contributions] we read further that (in fact the very first sentence of 1.401(m)-2(b)(3)(v)(B) Excess aggregate contributions are NOT considered when determining rate of match an individual received. so, based on that, I am indeed one of the folks that believes you can run the ACP test, correct a failed ACP test (as well as the ADP test) and then determine if you have to forfeit match due to a bad rate of match. in fact, I'd say the regs say you have to do that way. hope that helps.

at meetings the IRS personal has consistently used 'additional hardships' as an example of a discretionary feature which would have to be in place before plan year end. though since under EPCRS you can amend to add a hardship after the fact, it seems rather a moot point, or at least arguably so.

again, there is no formal guidance in this area. best I can come up with is the following: though there is a note in the ERISA Outline Book that a merged plan is a 'sucessor' plan (11.407 2006 Edition) and initial plan year of safe harbor plan can't be a successor plan. Now, it is true that it is not a new plan, well, except for some of the folks sort of. 11.538 says if two (safe harbor) plans merge with different plan years then one would have to be treated as having its plan year end as of date of merger. this creates a short plan year for testing purposes, and that kills the safe harbor. 11.402 says if two 401(k) plans (non safe harbor) merged with same plan year then test plans as single plan for the entire plan year. but if you do that, you are now combining a safe harbor with a non safe harbor and that kills the safe harbor.