Tom Poje

I think any answer in general always raises other question. In this particular situation, I just noticed the document mentioned was a prototype, and those scare me. It always seems like there are other rules attached to those, because they are guaranteed of passing averything. since there are no alloaction requirements, I'd think you use comp through the date of the amendment. I'd think you have to true up through that date through the comp at that date. While the comments said you didn't have to provide a notice (like you would for a safe harbor) you still have to eventually provide a SMM or a revised SPD. I'd provide some type of notice because the employees are going to find out eventually, and why casue any more grumbling then what has started. I'd admit I wasn't sure which way to lean on the answer to this one without looking for some other explanation. I thought the comments provided were well written. .......... Have a blessed Easter everyone!

Corbel has a comment regarding fixed match elimination dated 2/10/2009 and can be found here http://www.relius.net/News/TechnicalUpdates.aspx?ID=433 summed up: yes, but have to fund through date of amendment examples are provided, and additional comments

lets say I have 1 plan (profit sharing only) and 5 owners, each in his own group. The profit sharing contribution is being made, and some owners 'choose' no contribution. In the eyes of the IRS, you arguably (could?) have a CODA where none such animal exists. Even if it was a 401k then arguably you have people receiving more than the deferral limit. But for some reason, the IRS won't question this if each person has his own plan.

that reminds me of the following (I'll simplify it a bit though it misses all the magic involved) 2x - 6 = 5(x-3) 2(x-3) = 5(x-3) divide both sides by (x-3) 2 = 5(x-3) / (x-3) this leaves 2 = 5 ............ years ago I worked with an actuary and he suggested the following for an audit. put the value you want into memory. then, in front of the auditor start pucnhing different keys and the last step "recall memory" then, simply say, an alternative way would have been to do the following. punch a bunch of different keys, and for the last step, hit recall memory...you got it... same answer. in fact you could actually do it a third way....

copies of the draft and instructions can be found at http://www.form5500help.com/DRAFT8955_SSA_...instructNEW.pdf you have to love the fact you can now put a 'check mark' next to an individual if "the plan administrator is unable to determine from the records of the participant’s service as to whether or not the participant is vested in any deferred retirement benefit but there is a significant likelihood that the participant is vested in such a benefit. See 26 CFR 1.6057-1(b)(3)." no, really, thats what the instructions say.

The consumer price indexed was released today (CPI-U) and the value was 223.467 If this value stays the same (or increases) through Sept, then under the regulations next year the deferral limit will be 17,000 comp limt 250,000 and 415 limit 50,000 so, unless we bottom out like we did mid-year of 2008 all signs point to an increase in the limits next year.

well, since the regs state that an enumeration of employees by name or other specific criteria having the same effect as enumeration by name is not reasonable (1.410(b)-4(b)) I would agree with you - in the case of a controlled group in which you are not aggregating the plans it appears you run into problems under the rules.

very observant. I think that's just a starting point assumed by the softwares, for all practical purposes a minimum number and I have always increased by the number of employees entering on the first day. In rare cases I've seen the number drop beacuse the company didn't tell me about employees (Those with no balance) who actually quit the year before. I'm hesitant to change lower a number, but I've never had a problem increasing the number.

one such site (It says you can click here to generate a different card, so probably what you are looking for...) http://bullshitbingo.net/cards/bullshit/

well, since you asked for thoughts and not logic I'd lean toward the following... and using nondiscrim as an example before I can test nondiscrim I have to satisfy the gatewa so before I can ADP test I have to pass coverage I fail, but I can pass by 'making inelegible people eligible' by giving them a QNEC = to the avg of the NHCE. In effect I am adding them to my test, and my ADP % does not change because I have provided a QNEC = to that avg. But now I have tossed the people on the test with a QNEC, and once I do that, then I think you are back to the starting point of testing with and without the QNEC but remember, that is only my reasoning, which is certainly suspect at times.

you probably need to get a young goat or other such animal and if the you find favor with the nondiscrimination gods.... 1.401(a)(4)-9 two or more plans that are permissively aggregated under 1.410(b)-7(d)... but 1.410(b)-7(d)(5) says: Thou shalt not aggregate unless thou hast the same plan year (Old English Version of the Regs) but then for the avg ben pct test 1.410(b)-5(d)(5)(ii) plans with different plan years...normally thou shalt not aggregate such plans, but for this test (and this test only) thou shalt testeth the plans as follows... all kidding aside, yes, based on the above it is a little more difficult

to pass coverage a plan must pass eother the ratio % test or the average benefits test. you indicated plan 1 fails ratio % if the plans are not aggregated. For the average benefits test you have to pass both 1. average benefits % test 2. nondiscrimination classification test. the average benefits percentage test includes all contributions, all bodies, etc even if the plans are not aggregated for coverage. (see 1.410(b)-7(e) ...all plans of the employer that could be permissively aggregated... there is no gateway at this point because you are not testing for nondiscrimination.

oriecat: that's just a myth when you have both division and multiplication. for what is division but the inverse of multiplication? so you could have rewritten the equations as 48 * [1/2] * (9 + 3)

I asked the actuary. he simply said "What do you want the answer to be?"

my research says the following (which I made an attempt to put in the book I help edit, so I think I'm allowed to quote myself) The top-heavy regulations do not directly reference this issue. However, excess contributions, even though distributed, are still considered for purposes of sections 404 (deduction limits) and 415 (individual contribution limit), even though distributed from the plan. This regulation is silent in regards to section 416 (top-heavy). Treas. Reg. §1.401(k)-2(b)(2)(vii)(B). If the excess contributions were recharacterized as employee contributions - these are specifically mentioned as being continued to be treated as employer contributions for purposes of 404, 409 (ESOPs), 411 (vesting), 412 (minimum funding), 415, 416 (top heavy) and 417 (QJSA). [Treas. Reg. § 1.401(k)-2(b)(3)(iii)©]Therefore a top-heavy minimum allocation to non-key employees would be required, at least in that situation. At the 2004 Annual ASPPA Conference the IRS representatives indicated that a top-heavy minimum would be required even if it were returned as an excess contribution. (IRS Q and A #29) A reminder that any such comments in regards to the Q and As might not reflect an actual position of the IRS.

the preamble to the final 401k regs noted a sole proprietor / partner could defer...but..."self-employed individuals who take advantage of this opportunity to defer amounts during the year must make sure that the amount contributed during the year will not exceed the limits (such as the limits of section 515) that will apply to the individual, based on the individual's actual earned income for the relevant period." so I agree with 12AX7 in particular, see 1.414(v)-1©(1) ..catch ups...when added to all other deferrals...cannot exceed the participant's 415 compensation.

The ERISA Outline Book notes that there are no specific examples for cases with more than 2 loans involved. (thus, I don't know where you would find an example that says "Yes, do it this way") The book provides the 'literal' example along the lines as indicated above, but adds that without further guidance this is a 'reasonable' interpretation of the rules. as a result, I'm not sure I would say one opinion is correct or incorrect, one is certainly conservative and you know it doesn't violate the rules. my gut feeling is the IRS probably wouldn't push the other interpretation since that is a literal reading of what it says.

my understanding yes you could do that, since the right defer could be entirely eliminated. not sure you could word it "anyone over age 50" because that would be related to age (and age and service are no-no descriptions) the preamble to the original catch-up rules actually had an example of having a 0% cap on deferrals for HCEs, with the idea that they could only make catch ups. that example was removed from the final preamble, not necessarily because they decided you couldm't do that, but somethimes those thing get eliminated for space. It would seem kind of odd to say "If you have a cap at 0% you can't defer and therefore you can't have catch ups, but you could have a cap at 1 cent, and then since you could defer you can do the full catch-up, but what do I know about things like that.

I'd say the 1-1 is based on the total correction needed (Appendix B .01(b) does say excess contribution amount is either distributed or forfeited note, that example 7 under 1.401(m)-2(b)(5) would not even require forfeitures, even though the person is not 100% vested. this gives the HCE to 'vest' as long as they continue to work. you might have a strange looking participant statement for a few years (e.g. end balance = $600. vested % = 60%, but vested balance is not 360 but maybe 200. and then after a few years the person becomes 100% vested and it becomes a good deal for the HCE because they didn't forfeit anything. In exchange, you give up having some forfeitures which could reduce contributions. ...................... Example 7. (i) Employee H is an HCE in Employer X's profit sharing plan, which matches 100% of employee contributions up to 5% of compensation. The matching contribution is vested at the rate of 20% per year. In 2006, Employee H makes $5,000 in employee contributions and receives $5,000 of matching contributions. Employee H is 60% vested in the matching contributions at the end of the 2006 plan year. In February 2007, Employer X determines that Employee H has excess aggregate contributions of $1,000. The plan provides that only matching contributions will be distributed as excess aggregate contributions. (ii) Employer X has two options available in distributing Employee H's excess aggregate contributions. The first option is to distribute $600 of vested matching contributions and forfeit $400 of nonvested matching contributions. These amounts are in proportion to Employee H's vested and nonvested interests in all matching contributions. The second option is to distribute $1,000 of vested matching contributions, leaving the nonvested matching contributions in the plan. (iii) If the second option is chosen, the plan must also provide a separate vesting schedule for vesting these nonvested matching contributions. This is necessary because the nonvested matching contributions must vest as rapidly as they would have had no distribution been made. Thus, 50% must vest in each of the next 2 years.

correct, my bad typo

I believe the logic is a little flawed the top heavy test is run at the end of 2008 (12/31/2008) and more than 60% belongs to the key employees. this means the plan must provide top heavy in 2009 (unless there are no other contributions beside top heavy) [of course I meant safe harbor, my bad]but if other contributions are made then you lose your 'get out of top-heavy for free' card. the top heavy test is then run as of 12/31/2009 and the plan is again demeed to be top heavy, so top heavys must be provided in 2010 (unless there are no other contributions besides top heavy.)[of course I meant safe harbor, maybe my brain needs retirement] I don't think it ever a case of 'running the top heavy is meaningless'

too late. the person has already met the requirements to receive an allocation. you can't take that away from them. not a lot you can do. I've seen some suggest you could terminate the plan and put in a new plan, but good grief, I think I'd live with things for the current year and get it amended now for next year if it was the other way around in which you had a last day provision, then no one accrues until the last day, and you can amend things the way you desire.

the deferral limit of 16,500 is an individual limit, and a calendar year limit. Its not a plan limit. A plan is only required to limit someone to prevent them from deferring too much in a calendar year (and these totals show on the W-2)when you have an off calendar year plan then it certainly is possible to have more than 16,500 show up on the test, as far as I know.

but what is the intent of the reg (the English language is bad enough. for instance the regs says that when you have excess contributions that related match 'may' be forfeited. Does that mean I don't have to? no, it means that normally if someone was 100% vested in match you wouldn't forfeit, but in the case you are allowed to.) so in a safe harbor plan, you get a free ride on ACP testing if certain conditions are met. Its more likely that to have HCEs able to defer at a higher percent than NHCE are able to defer. so if I match nothing at 6%, but all of a sudden I match at 8% that definitely seems to favor the HCEs, given the fact you are getting a free ride on ACP testing. (Of course, since you are talking a match above 6% of comp perhaps the point is moot since you have already violated the safe harbor conditions)

if I understand the question at 4% everyone receives 100%. for the next 3% you receive nothing more, in effect 0%. then the match picks up again so someone who defers 8% or more. so I'd say have the following 1% 100% match 2% 100% match 3% 100% match 4% 100% match 5% 0% match 6% 0% match 7% 0% match 8% 50% match... at that point I'd think the rate is blown out of the water that doesn't seem any different than a formula like 1% 100% match 2% 100% match 3% 100% match 4% 100% match 5% 50% match 6% 50% match 7% 75% match 8% 75% match... (By the way Sieve, are you warming up for the hockey playoffs?)