Tom Poje

I had a typo, one of many. my humble apologies.

we know if you have a controlled group, you have to include all employees for coverage purposes. if you aggregate for coverage, then you have to aggregate for nondiscrim (ADP testing) as well. but you can't do that unless both plans use the same testing method - but there is no requirement that all members be safe harbor. Otherwise the regulation (1.401(k)-1(b)(4)(iii)(B)) that says you may not aggregate a plan using ADP safe harbor provisions with another plan that is using the ADP test makes no sense. (This particular section of the regs will refer you to 1.410(b)-7(d) which are the rules for permissive aggregation (refers you to the 'employer' choosing how to treat 2 or more plans, including QSLOBs, etc) Under what other circumstances would you have the situation arise where you would have a non safe harbor 401(k) and a safe harbor 401(k)? you can't (Tom corrected his typo) aggregate a plan with a non-related company. I suppose you could have an odd company that sponsors both a 401(k) plan and a safe harbor 401(k) plan... possibly what you might have read or are thinking of, is that even if both plans are safe harbor, they have to have the same formula as well.

its cleverly hidden in 1.401(k)-1(b)(4)(iii)(B) the very last sentence.

only if my sources take good notes. They have a ball and chain around my ankles, plus a tracking collar, so there is no way I can attend the conference.

my line of reasoning would work something like this: plan year 1/1/2008 - 12/31/2008 first paycheck runs from 12/27/07 - 1/2/08. ee defers $30 and works 40 hours. I doubt very few people say, oh he worked 5 days in 2007, so I will count $30/5 for ADP credited in 2007 and $30/2 for ADP credited in 2008. and therefore I count 40/5 for hours in 2007 and 40/2 for hours in 2008. Rather I look at the year as a whole and use the whole payroll running from 12/27/07 - 12/25/08 and therefore the hours credited during that payroll period. The document generally says use W-2 comp and that would be what was on the W-2 statement, so why would hours be any different. but then, thats my opinion only.

the topic of pro-rating comp, etc has been submitted as a Q and A for the upcoming ASPPA Western Benefits Conference (end of this month). hopefully the IRS will provide a deeper explanation then.

I suppose it depends on how the document is worded (but then many documents do not describe things accurately enough) for example, the new Corbel documents have an option "Administrative delay ("the first few weeks") rule. 415 comp for a limiation year shall not include, unless otherwise elected in section XXX of this amendment, amounts earned but not paid during the limitation year solely because of the timing of pay periods and pay dates.........if elected, ...shall include ...provided the amounts are paid in the first few weeks... suppose the ee deferred. He would have a W-2 indicating during in 2008 even though he terminated late 2007, simply because of the timing. My own personal opinion is that for a given calendar plan year, you are not using calendar year comp, but rather payroll comp that falls withing that plan year. so (unless the document specifically says include the comp in the prior year) I treat the person who terminated 12/27/07 as having terminated '1/3/08' [or whatever the last payroll period was] lets take it one step further. suppose that person had deferred 15,500 in 2007 before the last paycheck. do you now say he has excess deferrals??? but then that is only an opinion.

certainly if you have a discretionary match you most likely would want to use current year testing for the ACP testing to avoid the problem in years in which there is no match. If I remember correctly if you use prior year testing for the ADP test and current year testing for ACP then you could not use the option of 'shifting' deferrals to the ACP test. I prefer to have as many testing options available so that would be a drawback in my opinion.

in addition,according to the regs, a SIMPLE plan must be the only plan in a calendar year in which participants accrue benefits.

my notes say 'while yes, it could not be done in the dreaded standardized document' in addition, under the old rules, it rarely helped because it blew the ACP test out of the water. and Sieve, lest I forget, Go Red Wings!

no you can not have an allocation condition (other than excluding otherwise excludeble ees) this was made clear when the final 401k regs were released a few years ago. it was never intended that you could jave conditions for the the discretionary, hence the clarification when the final regs were released you can have vesting schedules apply

then if I understand you correctly, if Mr Big made 300,000 in the first 6 months (30 days after amendment), he would only be entitled to 1/2 of 245,000 as far as safe harbor goes (or $3675). but if the company had been kicking in 3% on a payroll basis, then you would have to forfeit the additional amount he had received - though if the plan is top heavy and keys were not excluded, then he would still be ok except his 'extra $' become unsafe harbor money. (unless of course he quits before the last day of the plan year in which case he wouldn't get the top heavy) (though I really do appreciate the fact someone called her, and hopefully as a result they will clarify the preamble when they release the final regs - since apparently a number of people have called on the issue) (interesting, 100 downloads on this one so far. you think its a hot topic?)

I'd say they need to address how the proration is to be applied. note, it says the rule applies to any safe harbor, including matching contributions! never heard that before.

hot off the presses. possible ability to eliminate 3% safe harbor during the year proposed amendment. I haven't even looked over it yet it is so hot and out of the oven

I agree (somewhat), what I wrote didn't quite come off what I intended. my apologies! my point initially was intended to simply say that you have to follow the terms of the document. however, you can't simply say "The regulations do not say that the maximum disparity has to be 5.7%, it says it cannot exceed 5.7%." becasue paragraph 'd' says if you have an integration level less than the taxable wage base then the maximum 5.7% must be reduced" (so, for instance, you couldn't say I can always use 5% becasue that is less than the maximum 5.7%) what I should have said, was that the regs tell you what the max % is if you use more than 20% of the taxable wage base but less than 100% than the max% is either 4.3% or 5.4%. The 5.7% can be the max if you use less than 20% (and therefore using less than 5.7% would be ok as well). What I had intended to say (hmmm, maybe I shouldn't try to write a response while someone else is in the office asking misc questions, sometimes you hit 'post' before finishing your answer...) anyway, what I had initially intended to say was that even if you tested using the allocation method you would pass. the point being, why do the regs describe 'safe harbor formulas' for profit sharing plans? if you allocate comp to comp you have a safe harbor formula and don't have to test. Wow. and if you tested on an allocation basis, everyone would have the same %, so its no big deal. If you use 100% of the taxable wage at 5.7%, and tested on an allocation basis imputing disparity everyone would still have the same % (assuming the contribution is large enough) - thus the fact the regs tell you that you have a formula that doesn't have to be tested is a foregone conclusion based on the formulas involved. (I'm old fashioned enough to believe people should try running the numbers once) at 100% of the TWB at 5.7% just to prove that it works out that way

you can not ignore the terms of the document (if the terms are hard coded exactly as you have described) all that means is you have a formula that does not meet 'safe harbor', so should be tested for nondiscrim. If you test on an allocation basis and impute disparity I would find it hard to believe it wouldn't pass testing.

you have a document you follow the terms of the document if it says HCEs are eligible to receive, what reason is there for not providing the match? none. they must receive and then you fail and correct any such failures.

technically if anyone in a plan could defer 5% or 10% or whatever to get the extra match, then option 'currently' availability to all. but a plan must both current availability and effective availability. This exciting and breathtaking nondiscrimination feature is described in greater detail under 1.401(a)(4)-4. but why wait for the movie when you can read up for it on your own.... how do you prove effectively availability when its a facts and circumstance test? I'd think you would say an extra match above 10% is too high, but at point do you say the % is okay? if all the NHCEs are minimum wage, then that would be another factor to consider, etc. as for fixing a problem, 1.401(a)(4)-11(g) describe corrective amendments

this concept applies if you are using a 3% safe harbor, or for that matter some people received a 3% top heavy. Suppose the group that received only 3% consists of less than 30% of the NHCEs. assume they received 0%. if you were test on an allocation basis (not accrual basis) then at least 70% have received comparable to the HCEs. even if the plan was integrated, because if you impute disparity the E-BARs on an allocation basis will be equal. (this assumes the plan is not a class allocation) by the way, 1.401(a)(4)-2(b)(4)(vi)(D)(3) even talk about treating top-heavy only people as includable and not benefiting and if you pass 410(b) you are ok.

based on your comments of having semi- annual entry dates, (assuming its a calendar year plan) most administrators would request on people who enter the plan 7/1 both comp for the whole year and from 7/1 testing comp can be from 7/1, if plan is top heavy you need to use full year comp. to determine max deduction you use full comp as well if an ee enters mid year and you only asked for comp from 7/1 but the person made over 105,000 you would never know if that person would be an HCE next year

there is no requirement that testing use the same assumption every year -you can use otherwise excludable one year and not the following if you so desire. if you are using prior year testing, then whatever method you use for the prior year NHCE you use for the current year HCE. but that has no effect on what you choose for the 'current year' NHCEs. hence next yeari if you use OE for the prior year' NHCEs the HCEs would use OE as well. also remember, that there are actually 2 options available under OE - one is simply that you can never have an HCE who is otherwise ecludable.

I was asked to look at a strange animal. a target benefit with mandatory contributions???????? the regs have instructions on safe harbor targets which say that you can't fund the benefit with ee money. does that mean you can do this , making it non safe harbor plan, and if so, how the heck would you even test?

yes, because the rules (putting it in rather simplistic terms) count anyone who received 'at least at this level...'

if I understand your description, you would have for BRF testing: the third group is all people since all receive the base match of 150% of 1 and 50% of the next 4%. the second group are those who get at least 2.5% additional (so this includes those at the 7% additional level as well since they received at least 2.5% extra ) from your description, it sounds like people are in the different groups, so whether they take advantage of being in the group is not so important?

I would guess that most testing is done on a year to date basis. (Annual testing as oppossed to accrued to date) in other words, only contributions (and forfs) are tested. so ending balances never come into play for testing purposes.