Tom Poje

(i) a cash or deferred arrangement which meets the requirements of section 401(k)(12), this refers to the ADPsafe harbor, which could be 3% SHNEC or BAsic Match or Enhanced Match (ii) matching contributions with respect to which the requirements of section 401(m)(11) are met." This refers to the ACP safe harbor only. the 3% SHNEC is not used to satisfy this test so the 3% is ok to prevent the plan from being considered top-heavy. Well, of course, if everybody received 3%, including termineed the plan would have satisfied top heavy anyway. (except that you could use comp from date of participation)

yes to a bunch and maybe to the last point Active ee, once having met eligiblity, is included in coverage. treated as includable and not benefiting. Possibly if the person is active, has a break in service, but continues working, the participation is suspended following the year of break in service, but that depends on the document. If the participant terminates with > 500 hours they are treated as includable and not benefiting if they had < 500 hours you could treat them as includable and benefiting - that is your option. if they are an hce it might help your test, but you have to treat everyone equally. If it is a controlled group, watch out. they are includable and not benefiting under the other company's plan. you can only use the 500 rule for 'participants', and that individual was never a participant in the other plan.

I handle too many. But then everybody does this time of year. I suspect I don't handle as many as some of you. I am suppose to help out on the Coverage and Nondiscrimination Answer Book as well. Plus I am suppose to talk at the fall ASPA this year (on 401k, I asked for entry/intermediate level) actually I wanted beginners cross tested, but you have to go with what you are assigned. I'd like to be able to study for C-3, but there isn't the time. I think I am lucky in the sense I have a 'feel' [logic?]for what is needed in plan processing, so maybe I don't do a lot of spinning wheels. I have a fairly good step by step processing as well. Plus, I can be running plans (e.g. printing) and hop out to Benefits Link at the same time. It helps me learn a lot reading the different questions and answers and doing some research if I am puzzled. Creating reports has become more of like solving a puzzle - I need a report for a reason, now go and solve. It is that challenge that keeps me going. The term 'saint' is an interesting one you use, because I do feel that sharing info is merely a part of my Christianity. It really does mean a lot to sacrifice some time and help others. But I am honest enough to admit that the extra work I bring on myself does have wear and tear at times. well, plus I am not or ever married (rats) or fool around, so maybe I have a little more time on my hands than most. ah heck, I ended up being a pension geek.

the ee does not benefit under the plan (read that 'the nonelective portion') so I would not provide a gateway minimum. Otherwise you would also have to provide the minimum for all terminees in a 401k as well.

oh, I know its possible, you would run it from Crystal, but at this time of year I can't fathom trying to write a report like that. too swamped. I have a plan list report, and I assume it would mean pulling ee name, acct bal and put a 'select' on it. But I also know that the report I have would have to have things moved around from one spot (read that: 'details' to headers or footers, along with what ever formats I used. The list report might actually exist on an earlier thread. I actually have done a few custom reports for people on an hourly basis, but it ain't happening right now.

actually, you might not have had a bad teacher. A few years ago (I don't recall which year) the instructions for the 5500 said you had to report the people the same year they quit. Maybe your teacher was using those!The IRS later issued an ooops, but obviously that doesn't change printed instructions. While perhaps it may require some time, it is worth the trouble coding paid outs the 'D' mgb- decades from now I hope to be retired! but it wouldn't be fair to let someone else down the road have to put up with call from the participant. I had to do it once on a takeover case. Not a lot of fun.

or for example: Plan A has 3 NHCEs and 1 HCE and plan b has 3 NHCEs plans are not aggregated so fails ratio % since NHCE % = 3/6 = 50% But you test avg ben test (treating the 3 NHCEs from B as big fat zeroes, and plan passes. Since plan passes 410(B) without aggregation, you do not have to provide minimum gateway to the folks in B

The ERISA Outline Book has a small discussion on this issue (2.49 in the 2001 edition) it basically refers to it as an administrative nightmare - especially if you had to wait to defer as well. at least you don't have that problem! I would assume you would have to provide the match, which also means you would have to add the employee to the ACP test and rerun. Wouldn't that be fun if it changed the test results ! Imagine if you refunded $ and now the test passes! or now the test fails and you have to refund $ a year after the fact. Another consideration: suppose the plan is also top heavy! then you have to provide that as well.

I agree with Mike, no prohibition against banding when cross testing. Relius (Quantech) will perform it, but as Mike pointed out there are limitations regarding favoring HCEs. Unless you are at 7.0 update or higher, then I would never use rate banding because the system sets the HCE at the highest level possible. that is a no-no. At 7.0 you are allowed to set your bands, but you better know what you are doing

prototypes have to be consistent (use the same testing method for ADP and ACP). If the match was 'introduced' in 2002, that is, wasn't ever possible before then you should be able to use the 3% prior year rule. If match was simply never implemented in prior years, then I can see why they wants to use prior ADP and current ACP since ACP would be 0 By the way, if you want to 'shift', you also have to be consistent

I haven't seen anything, but based on what the regs say I would say you have become top-heavy again. A similar scenario exists if you want to put into an IRA, if I recall correctly if you receive only forfeitures you still suffer the consequences. Heck, (in the future) use forfeitures to reduce plan expenses.

avg ben % test - you would add all the contributions and forfeitures and divide by comp if using allocation method. you would add all contr and forfs and convert to a benefit and divide by comp if using the accrual method. This is per individual, so each person has his or her E-Bar that is then averaged comp can be from date of participation, and the 5% gateway is satisfied if based on date of participation. if plan is top heavy, the participant would have to get at least 3% of total comp

I have grown to detest the term 'the devil's advocate'. afterall, I wouldn't think you would really want to be an advocate for the devil. that thought sends shivers up and down my spine! Maybe it is just me!!!!!! we'll have to come up with a better term! ....................................... I have a couple of thoughts 1.The ERISA Outline Book has a note which seems to read if your safe harbor notice is late, then the contribution is still safe harbor, but you also have to test. 2. suppose the plan wasn't safe harbor. Plan fails ADP test. As of this date you can not make a QNEC, except under the special rule of refunding the failure and also making a QNEC to match. so, in both instances, if you are late it seems to imply the worst of cases. Of course if the plan was a money purchase and you were late you would also pay a penalty, so I guess this late business is a good fund raiser for someone!

I think you still might have to test ADP/ACP 98-52 VII B a coda will not satisfy the adp safe harbor unless...contribution ...made...with the allocation and timing of section 1.401(k)-1(B)(4) This would be no different than with a QNEC. the govt generally won't let you make one if it is more than a year past due. The worst of all worlds, a 100% vested contribution, probably stuck using current year testing, etc. I think you are right, you can use SCP

based on what you say, I would agree, plan 1 will not pass coverage. FJR The issue boils down to whether you aggregate the plans. It is not simply a matter of 'everyone can defer' so I must pass coverage. If the plans are not aggregated, then not everyone can defer (e.g. the ees in plan 2 can't defer in plan 1)

Rosemary: I believe the 'z80' is a code that basically tells you that field has been put to sleep 'zzzzzzzzzzzzzzz' and no longer works. I heard tell that the user fields are now in a different table, and therefore you will have to edit the report to continue pulling them, but since I am not at 8.0 I don't know its name. Not sure what will happen if you try and run the report if you are at 7. I get an error message about invalid columns, but the report does eventually show on the screen. I believe Relius changed the tables because they didn't think I have enough work to do. : )

FJR: lets say I have 2 plans plan 1 has 7 NHCEs and 5HCEs and plan 2 has 3 NHCEs so for rati % testing you have plan 1 7/10 / 3/3 = 70% plan 2 3/10 / 0/3 = no hces so passes (If I understand you correctly)

Well bless your sweet little heart! Looks like you took my landscape one and moved the fields around. I later added 'forfeitures' so I ended up with a 3 by 3 effect. "Cosmetically pleasing" so to speak. My only suggestion would be to move the top-heavy info down a little bit so it lines up with the 'Dates' info. It will balance out the look a little better. Otherwise looks pretty good! (I am guessing this is a Relius 8 version, so some of the fields might not work at 7. This is based on the fact one of the fields says PLANSTAT.Z80USERALPHA1, which I believe means 'z80' no longer works at 8.0. Interesting, I heard they moved the fields to a differeny table!

IfI understand you correctly: You indicated that you need 2 of the ees to help pass testing. Therefore, that implies you are will not test otherwise excludables separately, so all ees and all contributions will be included in test. This will also require increasing the top heavy only ees to cover the gateway.

see IRS Notice 2000-3 Q 11 (which deals with adding a CODA to an existing plan) point #3.....the requirements of Notice 98-52 are otherwise satisfied for the entire period FROM THE EFFECTIVE DATE of the CODA to the end of the plan year. (Emphasis mine) so, hopefully the document says the effective date of the CODA is 7/1

Man, that sounds like the ultimate in fail safe language!

You still need to follow the terms of the document. If the document says only those with 1000 hours / employed on last day get the extra contribution, then those are the only ones who get the contribution. Chances are (e.g. if you are using a Corbel document, and I am sure there are others), there is nothing in the document that allows you to increase those ees who only received a safe harbor to the gateway minimum. to do that you would need a corrective amendment.

short on words, huh Mike? You should write the regs next time. Ha. I suppose if the safe harbor 401(k) plan covered at least 70% of the NHCEs you would not have to worry because you would pass testing without aggregating the plans. Of course if I was in the other plan and could take a job elsewhere I would probably would.

The answer given at the ASPA Fall conference was that if no other contributions were made (and that would include forfeitures) then the safe harbor plan is deemed to be 'not top heavy' for that particular year.

It doesn't really matter what order you do things, or what order your thinking process applies as long as the 1099-r reflects a 402g refund (since the W-2 will indicate the overage.)