Tom Poje

nope. that appears to be about as much as you can do. assuming 11000 deferral, 8000 match, 6000 due to shnec, that only leaves 15000 left. obviously the only possible improvement would be to go cross tested instead of integrated, but remember that you can not impute disparity on the 3% safe harbor piece.

it sounds to me if you have the following: a group of employees. part of this group are PDFs, and a special class has been carved out for them. they can defer, but they will receive no match. For 410(B) the 401(k) portion is no problem. One who defers 0 is treated as benefiting. for 401(m) all pdfers who have worked a year/age 21 will be treated as includable and not benefiting. the pdfs will show on the ADP test, but are exclued from the ACP test.

I don't think anyone knows for sure if you can switch midyear or not if you are not using the model documnet. ceratinly you can't if you were. good grief, you only have 3 1/2 months left, I'm not sure how much of a difference it can make this late in the year. I would think the vesting only accompanies the SIMPLE match or nonelectives that have been made. in other words the vesting follows the source, just like a QNEC - but the person is still subject to vesting in other 'traditional' contributions.

from the ERISA Outline Book (2001 edition) page 11.321 5.c New Plan The same rules would apply to a new plan.... this refers back to the prior page Compensation taken into account. The plan may limit compensation to the period of eligibility....see IX.B of IRS Notice 98-52. This rule is not mandatory. the plan may instead determine compensation for the entire plan year...the document will control.

If someone is ineligible to receive a match, then they should not be included in the ACP test. you have people who have received a match who shouldn't have. now, the safe harbor method of correction is to amend the plan and bring these people in, and therefore include them on the tests. the non safe harbor method is forfeit the match (reduce future contributions) see Q and A 150, 151 under 'correcting plan defects' in that forum.

Big John: (or is it Little John, from Sherwood Forest) Or maybe I should say if it wasn't for the statement, maybe I would never figure it out anyway. I have never seen 2 401(k) plans with different plan years, so this would be something new to me. 410(B) says we can't aggregate the plans for coverage (except for the average benefits percentage test) Ah, actually I think I get it. The average benefits percentage test, is, an amounts test, so it is like its own ADP test - if all you would consider were deferrals. Now, when performing the average benefits percentage test under this condition, you include all contributions for plan years ending in the same calendar year. 1.410(B)-5(d)(3)(ii) and (d)(5)(ii) 1.401(m)-1(f)(1)(ii)(B) or 1.401(k)-1(g)(1)(ii)(B) say the same thing - if there are more than one plan you combine the numbers. the ERISA Outline book then describes the 401(k) test as follows: "This rule (1.401(k)-1(g)) applies even if the plans are not permissively aggregated for nondicrimination purposes." So it is obvious we must aggregate 401(k) even if we can't for 410(B). But actually you sort of aggregate for 410(B), but only for the avg ben % test. 1.401(k)-1(g)(3) simply restates what 410(B) said for the avg ben pct test - how you handle plans with different plan years

Danger Danger Danger Dr. Smith. Under 1.410(B)-7(d)(5) Two or more plans may not be aggregated and treated as a single plan under permissive aggregation unless they have the same plan year. Do not pass go. do not collect $200. Note: this rule does not apply to the average benefits percentage test.

yes, you can do that in the case of a new plan (or a new 401k feature added to an existing profit sharing plan) the notice requirement can be as late as the effective date of the 401(k), but why wait.

There are a number of testing implications. If each plan can pass coverage separately, then you run separate ADP tests. or, if necessary, you can combine plans in any way, shape or form, as long as you are consistent - whatever you combine for coverage, you combine for ADP testing. But remember, if you test each plan separately, all other employees (since they are treated as being employed by one employer) who meet the eligibility requirements of the plan being looked at are includable and not benefiting - since they received nothing under that plan. Watch out for eligibility! you may suddenly cause some ees who are ineligible to become includable and not benefiting - though you can usually use 'otherwise excludable' option and avoid this trap. you will need to test b/r/f (benefits/rights/features.) as well. for example, suppose you have 2 plans - one matches 100% up to 3% and the other matches 50% up to 6%. The plan may pass the ACP test, but now you have 2 groups of employees with different rates of match.

I believe your interpretation is correct. all the regs are saying is that you cant 'also' apply integration to the safe harbor contribution. there is no reason you can't make an additional contribution under whatever formula you desire. Many are going with a cross tested formula of some type. remember, you can use the safe harbor in testing as well, but you can not impute disparity on that portion of it.

based on your notes, no, that doesn't sound aggressive - simply using the otherwise excludable option that is available. I assume you are using prior year testing method as well. I suppose if the match has a last day provision, then you will make clear to the client the numbers could easily change if ees quit. With the elimination of the multiple use test for 2002 it makes the projection easier since now it is simply a '2 test' to worry about.

hmmmm. I hadn't thought about that. I could be wrong on my following comments ..... I hadn't read it that way. the way I read it is based on a year by year basis. In other words, if a plan chooses not to allocate a safe-harbor for a given year (e.g. instead of safe harbor match, they make a discretionary match), then they would have to provide a top-heavy minimum.

Lets be careful out there.... the new rule regarding top-heavy applies to plan years begining 1/1/2002 AND 'the term 'top-heavy plan' shall NOT include a plan which consists SOLELY of - (i)...401(k)(12).... [ADP safe harbor ], and (ii) ...401(m)(11)..... [ACP safe harbor] emphasis mine. in other words, it looks like if additional contributions are made you have lost the free pass on top heavy. You are correct. You can not convert an existing 401(k) until 2002.

there can be no hours/last day restriction on the safe harbor contribution. You can have a last day or hours requirement on any additional contribution. note, that is one of the big differences between a QNEC and the safe harbor. you can limit the safe harbor to only those who have completed a year of service and age 21 (ELIGIBILITY) if the plan provides for immediate eligibility to defer, but once an ee meets these requirements he/she must receive a safe harbor even if their hours drop below 1000.

I have not seen the second edition yet. (certainly the first book didn't) I sent a bunch of stuff to Amy in regards to this. one of my suggestions was for some examples in the appendix, which would have included disparity calc - I don't know it they made it in.

my understanding is that you wouldn't have separate listings, even though you had different accrual requirements. remember, schedule T is for coverage. as an example, suppose I had a mp plan and a ps plan, each with different formuals. I am allowed to aggregate the plans for coverage purposes. with a single plan you still have your nonelective contribution - actually you indicated you have two - one that is a 'special' nonelective, so to speak, but it is still a non elective - whether it is a QNEC or a SHNEC or whatever. A typical cross testing plan today would be a safe harbor 401k with a shnec and an additional ps contribution for the owners only. if you were to test coverage how you desire (one coverage test for the SHNEC and another for the PS) you would never pass the ps. so you end up passing coverage, but now you have to pass amounts testing.

a bit humbled at this end! The Panel Publisher Book came about in an interesting way. Every once and a while my boss 'rents' me out for a combination software/pension training. (Actually, they can't stand me here, so they get me out of the office when they can) Anyway, this one time it was at the company Amy Cavanaugh was working for, and I noticed a copy of the nondiscrim book on the shelf. Of course, I was fascinated, and 'stupid' enough to indicate so Amy asked me to review the first edition, etc., and so I ended up on the second edition. and yes, I am scheduled to give one of the talks at the ASPA conference at the end of October. The talk is intended to be more on the basic / intermediate level, so it doesn't get heavy into cross testing issues such as 'is this company a good choice for cross testing, and why/why not'. It is more on the level of coverage, participation, amounts testing, brf issues, etc. rather than a finely detailed cross testing presentation. thanks again for the friendly comments!

I don't even want to know who set up an obnoxious situation like that. you have some major grief in store if my mind is working properly, (some would argue that is an oxymoron, but what the heck) you probably would be better of setting up a plan for each different service requirement, and then combine for testing (all under the same employer) remember if one plan has immediate eligibility and another has 1 year wait, you end up with a group of ees who could end up as includable and not benefiting - even though they are not eligible for their own plan. (e.g. ees who have only worked 6 months) on the other hand, you can probably get around this by using the statutory exclusion option - but even that could mess up if you have an HCE. proceed with caution my friend! another possibility, there is a field in census DER - met eligibility requirements - I suppose you could plug a date there for all employees (different for each group of course) and then run eligibility. I have never tried that, but logically that should work - then you could get by with just one plan. then run contribution by division. as before, I think you will have to run statutory exclusions to make testing easier, and hopefully you will have no HCEs with less than 1 year svc/age 21.

perhaps my response was unclear. I don't disagregate the nonelectives. I was merely indicating that a nonelective is not simply the run of the mill profit sharing, but can include other items.

are you talking eligibility to enter the plan or eligibility to receive a contribution? my initial response is to simply put people into different divisions - and a big note in the folder to remember to enter divisions on all new people each year. then you can run the contribution by division, one transaction for each division.

I don't know if the term 'embarrassed' is a good one. I use a lot of those fields for other things out of 'desperation'. and so this was a good post on your part, because others might not have thought of using them. by the way, I am not quite sure how to pull the info if it was in beneficiary 2 (or more) so maybe my solution was only part correct!

wave the white flag and surrender. the original instructions that came with 5.0 carried the recomendation that you replace planee2.userdef... with the new user fields. however, it was quickly discovered this wouldn't work, and I think they even removed this statement from the instructions. I actually tried back then to pull fields, but the only success I had was if I pulled only field. If I tried to pull more than one it stopped working. I am guessing at this point, but maybe that is why you can pull the data into a subreport - but maybe if you only pull one field. which would mean a separate subreport for each user defined field. that is only a guess on my part. I too have always been told the same - 'it is complex' - and nothing else, so I have never bothered with it. what you might try, depending on how many 'extra' fields - and if you are not using them - is to use the beneficiary fields - e.g. name, street 1, street 2, city - or other census fields, such as internet, street 1, etc. I have used those fields with success. be careful if you are importing data into these fields - I know at one time some of these fields would only import 20 characters even though you can enter up to 30 characters.

While I can sympathise with your situation, it ' comes to this' because a retirement plan is just for that purpose - retirement. In addition, the govt has given a tax break as well, so they stick to their guns. You did not indicate if your plan allowed for hardship withdrawals. That is the only scenario in which the govt says you can override the situation. it is actually recognized hardships may occur. Are hardships available? not to take anything away from your circumstances which sounds rough, my experience has been that a hardship in some people's mind is that the neighbor has a new 'boat' or other toy, and the govt tries very hard to prevent people pulling retirement money for these things. and then, as it turns out, these are the same people who complain the most at retirement that the govt should turn around and give them an even bigger hand out. my prayers that your problem can be resolved soon!

close, but no cigar. I will use allocation method rather than accrual method as an explanation, but the concept is the same. Maximum disparity is lesser of 5.7 or the ps amount EE defers 5%, match 2%, ps 4%, shnec 3% avg ben % is 5 + 2 + 4 + 4(imputed disparity) + 3 = 18 for the nondiscrim test (401(a)(4)) you would have 4 + 4(imputed) + 3 = 11 in other words, you get to impute for the avg ben % test as well. or, if given a spreadsheet with values entered and cute little columns, you only impute on the purely nonelective column (this can include forfeitures) but not on deferral, match, QNEC, QMAC, SHNEC, ESOP. If it helps, you would not integrate any of those items, so you impute on them either. the only piece you could integrate is the purely nonelective, no strings attached piece.

Fred: I know you have done a bang up job on your spreadsheet, at least one or two versions i have seen and have played with. It has made some things easier along the way. To clear things up: you said (I've been told that if the Plan is a 401K Safe Harbor using the 3% non-elective, disparity hurts you so don't use it in that situation.) this is maybe, maybe not. example: safe harbor 401(k) using 3% SHNEC (Safe harbor nonelective) owner is only one who gets additional contribution. lets say 12% extra. If you impute you would have: rank and file 3% SHNEC - no impute allowed. owner additonal 12% - impute and his E Bar increases while no one leses does. in 2002, the minimum gateway would require the NHCEs receive 5%. Now you have rank and file 3% SHNEC - no impute 2% additional - impute owner - 3% shnec - no impute 12% additional - impute but imputing disparity will increase an E-Bar at most .65 (or up to .75 depending on the SSRA). in the case of an individual will large comp, this increase is even smaller. so, if there is an additional contribution to ALL parties involved, imputing disparity may help.