Tom Poje

no one knows for sure (unless it is buried in the proposed regs somewhere.) Sal argues for as late as 12 months after the end of the plan year because that is the time limit for correcting an ADP failure. but then he adds that IRS has indicated that future regs may require 1st day of plan year for which changes takes effect. so what's a body to do? I don't know - are we still in the restatement period, if so, then that is the easy way to go. ......... I still find this one real interesting. the schedule T you will check 'the employer employs only HCEs' which is Sal's argument 'ah, then you pass'

Blinky: I think the problem is that if you don't amend to current year testing, you end up with the situation of never using the NHCE results. (e.g. if the prior year results were real low for the NHCEs then you have circumvented the sytem using this logic) the reg cited is 1.401(k)-1(b)(2)(i)(B) "an arrangement does not fail to satisfy...merely because all of the eligible employees under an arangement for a year are highly compensated" I understand Sal's argument. It just seems strange, probably more so as a result of the fact this reg was written before the ability to use prior year testing. personally, to be safe, I think I would amend to current year.

A plan with only HCEs is deemed to pass. Interesting question. According to the big E book(11.269, 2003 edition), the determination of whether all eligible ees are HCEs is based on the current year, since this relates to coverage testing. Prior year testing is simply a means of calculating the ADP of NHCs to determine if plan passes for the current year. That would seem to say in 2004, no nhces, so you get free ride even though there was a prior year nhce amount! or at least if I understand the argument correctly implications are that if no nhces in prior year, but there are in the current year, you don't get a free ride but are stuck with current year testing! hardly a goofy question following all that logic.

correct. current year never matters. always use lookback year and the value for the calendar year that begins in that year

prior 12 months began in 2001, so the the threshhold was 85,000. It could have been possible to use a calendar year election, which would have used comp from 1/1/2002 - 12/31/2002 and then the ee would have had to have 90,000. But that depends on what the document says.

or, I suppose, put another, does it pass the smell test. lets see, partners (HCEs) are able to receive an additional contribution on a payroll basis, thereby getting earnings throughout the year. the NHCEs only receive a one time contribution at the end of the year. Smells to me like you might have a BRF failure.

if, on the other hand ee is age 50 or older and plan allows catch up contributions then at least part of the deferrals would have to be treated as catch up, and these amounts are not counted when determining the top heavy %.

it is unclear from your note what the document says. If the document has 26 classes (1 each for each HCE and the last class being all others) and each class definition says the allocation is comp to comp,and a contribution is made to each class, then you need do nothing more. one or more classes could get a zero allocation.

If the formula for match catch ups is the same as for the regular match I don't think you won't have to worry about the 70%. In other words, a single formula doesn't have to worry about BRF however, the possible issue is how the document is worded as to the cap on deferrals. the example used by Corbel was:(and this is a brief version) Plan limits deferrals (regular and catch-up) to 60% of pay. Jim is catch-up eligible and makes $20,000. Thus Jim's deferrals cannot exceed $12,000. Since the plan has imposed a limit that a catch-up eligible participant can't exceed, the plan violates the universability requirement.

Earl- Maybe I will backtrack a little on my earlier comment, depending on what the document says. If the formula is simply 100% of all deferrals, and a cap on deferrals at $1000 I think you are ok, even under the universality rule. If there was a different rate for catch ups then you would have a nondiscrim issue.

the reason: if you say maybe to the 3%, it shouldn't effect an individuals choice to defer but the match will have a bearing if someone chooses to defer, therefore you are not allowed to say 'maybe'

that would seem to violate 1.414(v)-1(d)(4) of the regs [its 1(d)(3) of the original proposed regs] ...a catch up eligible participant is not treated as having a right to a different rate of match of allocation of matching contributions merely because an OTHERWISE NONDISCRIMINATORY SCHEDULE of matching rates is applied to elective deferrals that include catch-up contributions.

based on the IRS opinion, yes. although if no ps is made and there are forfeitures, then you are still top heavy

my guess is that you could probably use one definition for ADP and another for ACP as you indicated, I don't think you would be able to 'shift' deferrals under those conditions because of consistency. I seem to recall you could test ADP using otherwise excludables and ACP using everyone, and fun things like that as well, but don't quote me on that one either!

I would hold that anyone who made 85,000 in 2001 is an HCE, regardless. suppose there is no controlled group. a company starts a new plan. you look at last year's comp to determine who is an hce. the fact a controlled group already exists shouldn't make a difference. all employees are treated as working for the same employer. the only difference might be if this is a situation in which a controlled group was just created (e.g. asset or stock aquisition) I am guessing that if there was an asset aquisition you ignore prior comp because because the prior 'employer' has ceased to exist, but that is a big guess.

Yes, if ps made. however IRS was clear if no ps made you are ok. at 2002 ASPA conference, Q 21 the following was asked: Plan has the following contribution types 1. 401k 2. 401m 3 discretionary profit sharing. plan is top heavy. no profit sharing will be made in the future. can match safe harbor satisfy top heavy? Answer - Yes, however if no profit sharing contribution is made such that the only contributions made are deferrals and safe harbor match, then the plan is not top heavy, and no top heavy is required.

ERISA Outline book specifically states (2003 edition, 11.294) that a tiered match similar to the one listed must be tested for effective availability. as for Joe's enhanced formula: I don't think that is possible for an enhanced match. The rate of match cannot increase as the rate of deferral increases. Nice try, though.

I believe the answer is yes. of course, as you noted, the document rules

the safe harbor top heavy freebie rule implies no allocations other than deferral and safe harbor. once you give someone a forfeiture you have given them an allocation other than permitted.

I think to be on the 'safe' side, have any forfeitures from the profit sharing reduce plan expenses. otherwise I think you have the top heavy issue.

the preamble to the final catch-up regs says "For example, in a plan year ending on June 30,2005, elective deferrals in excess of the...ADP limit for the plan year ending June 30,2005, would be treated as catch-up contributions as of the last day of the plan year, up to the catch-up limit for 2005." Based on that, In your example you had 3000+ in excess contributions. Sinc plan year ends 2003 only those deferrals up to the limit of $2000 can be treated as catch up.

see if this works. this was originally written for cross tested, so ees are grouped by numeric user field #1 in census. ever since 8 or 8.2 I have found I have to type in a 0 on the 'common' people, even though 0 is in the field already

it is the year in which the lookback year begins. In your case the lookback year is 9/1/01-8/30/02 since this begins in 01 use 85,000.

close. its a cap at 4% of comp and 6% deferred. so a formula of 66.67% of the first 6% deferred, would work for a safe harbor match. proposed regs, if I understand them correctly would require the discretionary to go to all employees (no hours/last day requirement permitted)

you are correct in your reasoning. the match will require ACP testing, and it will also change the average benefits % test, it will not count toward the gateway minimum, nor used in rate group testing.