Tom Poje

did a little more research, at least at the 2009 ASPPA Q and A 48-53 (a whole bunch of questions on this one!) the IRS voiced an opinion such comp is eligible 48. An employee terminates in December 2009. A final payment of salary due for services is made in January 2010. The plan does not use the “first few weeks” rule in the IRC §415 regulations to treat the January payment as made in 2009. The plan year is the calendar year. The plan includes a section 401(k) arrangement that defines compensation eligible for deferral to be section 415 compensation. Is the individual included in the 2010 ADP test, even though he terminated employment in the 2009 plan year? Yes. Since he could defer out of the compensation paid in January 2010, he is an eligible employee under the 401(k) arrangement for the 2010 plan year. The 401(k) regulations do not treat active and former employees differently 50. Suppose instead that the plan is a safe harbor plan that provides the safe harbor nonelective contribution. Is this individual entitled to that contribution? Yes. Since, as discussed in Q-1, the individual is treated as an eligible employee for 2010, he is entitled to a safe harbor contribution. However, if this individual is an HCE, and the plan does not provide the safe harbor contribution to HCEs, then no safe harbor contribution is made on his behalf. The same answer would apply to a safeharbor match if the individual made elective deferrals out of the 2010 compensation. ..................... Robert Kaplan gave an ASPPA presentation the following year and included this stuff in his talk http://www.asppa-net.org/Portals/2/Defining Compensation in Qualified Plans.pdf

Freddie the Unlucky is hired 1/3/2017. age 30 or 40, depends on the age of the lady he is talking to at the bar. he works 1000 hours in 2017. the regulations say 6 months after meeting 1 year of service or age 21.but his next entry date is 1/1/2019

if not familiar with it, on Relius you would have plan specs set to the comp definition then run eligibility. it does no good to change specs after the fact. as an alternative you could plug the desired comp in the computed fields if for some reason you desire not to rerun eligibility. just because match comp does not include bonuses you could test on total comp, in fact would have to if you failed the comp test.

without guidance I think you do a god faith effort / interpretation of how to do things. perhaps some point to consider - how many people involved in 2017 and 2018, and how many left it at 3% or switched? I find it hard to believe there are are a lot involved, but then I have no idea what size of group you are talking about.

without other guidance, I'd lean toward saying it is probably ok. obviously if the plan wasn't a QACA to begin with I could set it up and only those without an election in place would start at 6%, everyone stays where they are at. I'm not sure this is any different. certainly those less than 3% have already indicated what they want, so they wouldn't change. and it is only those who are between 3 and 6% that would be, as you say a 'second' QACA. but if someone is at 3% they could have switched and deferred more if they wanted to originally. Anyone who was there before 2017 would be at 6% now (assuming 1% increase each year) someone hired in 2017 would have been 3% in 2017, then 4% in 2018 and now 5% in 2019 (instead of 6%) so really the only people hired in 2017 and 2018 would be 'different', and they could have increased their deferral % to 6% if they wanted to anyway.

lets explain the reason first. generally, the people with the most years of service would be HCEs, so in addition to the ACP test you run a 2nd test to verify you are not 'favoring the HCEs. so let's say I have 10 NHCEs, 5 with less than 5 years, 5 with 5 or more years I have 1 HCE with at least 5 years. using the example I described above, how many folks could get at least 500? (It doesn't matter if you deferred or not) everyone can get at least 500. so 10/10 NHCEs and 1/1 HCEs so 100%. now, how many an get 1000. 5 of 10 NHCEs, 1 of 1 HCEs, so ration % of 50%. that fails ratio % test because it is not 70%, but if you can pass avg ben pct test AND safe harbor % based on the NHCE concentration % of 10/11 then you pass the BRF (benefits Rights Features)

I've never heard of such language before, but that doesn't mean it isn't possible. so lets say I make 2 groups. those with less than 5 years and those with 5 or more years. It sound like anyone with less than 5 years who deferred gets a set amount of match...e.g. $500 and then anyone who deferred with 5 or more years gets a set amount...e.g. $1000 in addition to the ACP test you also have to run a BRF test. but maybe someone else has run such a plan before and has a better idea than I do. of course you could also say 'no match this year"

this might be a little easier to understand 1.415(c)-1(b)(2)(ii)(D) Excess Deferrals that are distributed in accordance with section 1.402(g)-1(e)(2) or (3) do not give rise to annual additions so if corrected timely ignore.

in other words, 410b (in this case, for purposes of the nonelective) only asks did you receive a nonelective (not "how much"?) since plan is safe harbor, then the answer should yes. as noted it is a few people who received a safe harbor and not the extra profit sharing. though the safe harbor is a QNEC, the rules are slightly different, you do not have to perform 410b test, one with QNEC and one without QNEC. 1.401(k)-3(h)(2) if plan is cross-tested then those folks might have to be bumped up to the gateway minimum, which is a requirement, so it is possible no corrective amendment is due, simply follow the rules. if it is from 2016 then self correct under PCRS

depending on your software's capability, you could use total comp for deferral and comp less bonus for match (assuming that comp definition passes 414s).

maybe he means definition of comp excludes something and they fail the comp test???

And I still don't see how you can simply ignore the compensation in either year. the person receives a W-2 in 2018. the IRS treats the person as having 'worked' in 2018, though you say only 'paid'. assuming the document wasn't checked, comp earned in 2017 but not paid is counted in 2018. you are playing games saying the person 'worked' in 2017 but 'paid' in 2018 and therefore I can ignore the comp in either year, hence an exception to rule of no allocation conditions on a safe harbor. I bet if the person worked 994 hours in 2017 and worked Dec 27 and Dec 28 a total of 16 hours, you also exclude the hours and don't credit him with a year of service in 2017. I guess this is like basketball. over and back. yes we treat the comp as being 2018, but we get to ignore the hours in either year and the compensation for allocation purposes because it is actually 2017. since Bird and earlier post by Larry both concur on this I will refer to it as the "Larry Bird basketball exception rule" and we will apparently disagree on the issue.

so in other words, even tough the document says include comp in 2018, there is an exception to the rule that you can't have any allocation conditions on a safe harbor. that you actually have to perform services in 2018 (even though for personal taxes and everything else you are treated as having worked in 2018)

see the note on the blurb from the FT William document. (I assume other documents have similar logic somewhere) a. In-service withdrawals are allowed on attainment of age 59-1/2 (Section 8.02): i. [ ] None ii. [ X ] All Accounts iii. [ ] Selected Accounts b. If Selected Accounts is selected, specified age withdrawals may be made from the following Accounts: i. [ ] Elective Deferral Account ii. [ ] Matching Account iii. [ ] Profit Sharing Contribution Account iv. [ ] Qualified Non-Elective Contribution Account v. [ ] Voluntary Contribution Account vi. [ ] Rollover Contribution Account vii. [ ] Transfer Account viii. [ ] Other: c. If a Participant may receive a withdrawal upon the attainment of a specified age from his Elective Deferral Account, permit such withdrawals from the Participant's Roth Elective Deferral Account subject to the same terms and conditions as apply to the Participant's Elective Deferral Account: i. [ ] Yes ii. [ ] Yes - only if the withdrawal from the Roth Elective Deferral Account qualifies as a "qualified distribution" within the meaning of Code section 402A(d)(2) iii. [ ] No NOTE: If G.5a is less than age 59-1/2, Elective Deferrals, Qualified Non-Elective Contributions, Qualified Matching Contributions and the portion of any Account that has been used to satisfy the safe harbor requirements of Code sections 401(k)(12) or 401(k)(13) and/or 401(m)(11) or 401(m)(12) shall not be eligible for withdrawal until the Participant attains age 59-1/2; but only to the extent withdrawals are permitted from such Accounts pursuant to G.5a and G.5b.

agree, unless they were to change the wording that the 'add back' applies only to terminees (I could possibly see the logic of that) who quit the last week it would be insane, especially with a plan of any size. however, now you have the problem you would need to request comp for the last week for anyone falling into that category, because usually you don't receive that until the following year. so, at least for me, it falls, for all practicality, to the logic of a term date attached to the W-2 that falls within the plan year. so someone who quits Dec 27,2017 and worked 10 hours, but the paycheck shows in 2018 worked those 10 hours in '2018'. thus it makes sense to include them in ADP testing, etc. in 2018, though they didn't 'work' in 2018. otherwise you are back to saying you don't have to give a contribution on the last paycheck earned, yes you may have had comp, but no hours, sorry, how dare you claim you should get something.

at first I was going to say yes, but now I'm not so sure. that says if I determined an ADP failure and then decide to treat those ADP failures as after tax, then I run my ACP test, which would make sense, because you can't run an ACP test, then suddenly add after tax after the fact.

let's take it one step further. let's say the person was eligible in the prior year. you are saying if you quit the last week of December, and the paycheck shows up the following year, you never have to calculate a contribution because such people work no hours in the new year. we need a term for that. the last week termination you get screwed rule.

the ft William document has language such as (and this is not post severance comp, that is a different check box) 15. Post Year End Compensation [ ] Determine Compensation using Post Year End Compensation NOTE: If selected, amounts earned during the current year and paid during the first few weeks of the next year will be included in current year Compensation. so if you don't check the box(and I know few that don't check the box, you are already indicating to include comp even though 'no hours worked'. unless you figure if I am counting the comp I am also counting the hours as well because I am counting the W-2s that fall within the plan year.

the document should contain language such as (b) Refund of Excess Elective Deferrals. In the event that Elective Deferrals under this Plan when added to a Participant's other elective deferrals under any other plan or arrangement (whether or not maintained by the Employer) exceed the limit described in the preceding Subsection, the Plan Administrator shall distribute, by April 15 of the following calendar year, the excess amount of Elective Deferrals plus income thereon. If it is an HCE the excess is included in ADP testing, if NHCE it is not included. if there is a related match on the excess it would be forfeited. gains adjustment, etc are treated the same as you would for an ADP failures. from the Coverage/nondiscrimination answer book 12:14 Excess deferrals should be distributed with earnings to the participant no later than the April 15 following the calendar year in which they were made. The excess deferral is taxed in the year it was deferred, and the earnings are taxed in the year of distribution. Therefore, two 1099-R forms are needed: one reports the excess deferral with a reporting code “P” indicating the distribution is taxable in the prior year; the other reports the earnings (losses), with a reporting code “8” indicating the distribution is taxable in the current year. ............. The IRS looks at the W-2 so it knows the excess deferral exists so, famous last words, when looking at the W-2 it will adjust things even if the individual doesn't indicate the excess. What is 'confusing' to the individual is the 1099 won't come until the following year even though they need to indicate the amount on this years form, hence the code P

ignoring the safe harbor issue for the moment. suppose the person regularly deferred and had deferrals in 2017. so now there is a final paycheck which of course shows in 2018. most documents would indicate to include the deferral in testing in 2018 (unless they have checked the 'pain in the rear' first few weeks rule, which would seem to imply you treat everybody the same way. so now you are including this paycheck in 2018. I would generally include it as comp earned in 2018, certainly for tax purposes the IRS treats it as such as well. so for me, if the doc says use W-2 income, I would lean toward providing the 3% on the small amount. but then I look at it as comp earned within the plan year, rather than actual date of termination. otherwise it seems rather inconsistent to include the person in ADP in 2018 even though they didn't work, but for all other purposes you don't count them. (but then I'm a bit off my rocker anyway)

Based on the info I gathered I came to the following conclusion. There is no requirement that the ADP test be performed first, followed by forfeiture of related matching contributions if there were excess contributions . Both Treasury Regulations Sections 1.401(m)-2(b)(3)(v)(B) and 1.401(a)(4)-4(e)(3)(iii)(G) indicate that the rate of match is determined after any corrections are made. IRS officials also agreed with this conclusion (at the conference they indicated they thought most plans utilize the second option - run ACP first. [Q and A #18, 2004 ASPPA Annual Conference] Thus, aside from any document restrictions, it is recommended to perform the ACP test first, make any corrections necessary, and then perform the ADP test. In this way the HCE might not forfeit any matching contributions.

that depends. maybe . check the document. for instance 8. Top-Heavy Allocations Top-Heavy allocations are made to a. [ X ] This Plan. Participants who share in Top-Heavy minimum allocations: i. [ ] Non-Key only. Any Participant who is employed by the Employer on the last day of the Plan Year and is not a Key Employee ii. [ X ] All Participants. Any Participant who is employed by the Employer on the last day of the Plan Year you indicated the top-heavy was a 1% "allocation". but if all employees receive the allocation, then it sound like the key person receives as well (his deferral doesn't sound like an allocation). but since he deferred 1% he is now at 2%. oops I just increased my top heavy. but I haven't seen a write up on how this is supposed to be interpreted, so what do I know.

the issue is because the IRS said vesting is 'protected' so some think it is 'permanent'. so they think someone who has 2 years of service if 100% vested is always 100%. but the IRS, even in it's example only says the vesting to those benefits accrued to data. so even if the person was 0% vested with 2 years of service their accrued benefit to date is under the old schedule (and the new if portions are better)

I'd think if only 1 payment was missed (in May), then it couldn't have been much beyond the 5 year limit on the loan if that was the case, and the total loan balance wouldn't amount to much. in other words, it doesn't sound like the 5 years were up.