Tom Poje

While we use Relius for plan admin, I wrote a report that pulls that data and easily imports into FT William, but since it is an excel spreadsheet it wouldn't matter what software you have as long as you can get it into excel.

one possible blurb besides the regs is found at http://www.irs.gov/irm/part4/irm_04-072-002.html in particular 4.72.2.5.2 this is pretty much a repeat of the regs but adds a little, such as what happens if the employee status changes. according to the ERISA Outline Book [under the definitions] (and I am sure other sources) "If two or more organizations are treated as an affiliated service group, the organizations are treated as a single employer" since you have 1 employer, and one time election applies to any plan of the employer including plans not yet established, well....

As I recall, when the safe harbor rules first came out, people thought you could have allocation requirements on discretionary match, but the IRS came out and said Absolutely NOT, or something like that. the only thing special about the discretionary match was the ability to apply vesting. so I guess I don't understand the example above. if the plan has allocation requirements for the match, then it would have to be tested under ACP. My brain gears are really rusty, but as I recall you could include the safe harbor match in ACP testing as well. Lets suppose you had an enhanced match of 100 up to 8%. that satisfies ADP, but we know that it will fail ACP. does that mean the ADP fails as well? no, it just means you have to test ACP. I see the ERISA Outline Book says pre 2006 could have had allocation conditions (because the regs weren't clear, I guess)

agree that for HCEs you have to count all comp and contributions. but the original comments said "I ran coverage for both and they pass when combined" I took that to mean the plans were combined, and I don't see how that is possible with different testing methods. combining all contributions for the HCEs should not change the coverage testing, that should only affect the ADP test.

good grief. you can't aggregate for testing because they have different testing options, one is prior the other is current. but you have to perform coverage and nondiscrim in the same manner - either aggregate or not. see 1.401(a)(4)-1©(4) "whatever testing method used under 1.410(b)(8)...must also be used for testing under 401(a)(4)" that is of course, one of the the whole concepts behind the transition rule, you have time to modify one or both plans

FT William sent out the following blurb the other day - at least it gives a clear explanation of what is going on: (Gotta love the IRS concept "submit your form 8955 later in August...") Dear 5500 Customer: We recently contacted the IRS due to several customers receiving Penalty Notices for filing the Form 5500/8955-SSAs after the original due date, even though a Form 5558 was filed timely. An IRS Representative responded with the following information: At times when a filer submits a Form 5558 extension request and soon thereafter submits their Form 5500 or Form 8955-SSA filing, all forms will post to their account in the same cycle. Should this occur, if a return happens to post before the extension request, the return will erroneously be marked as late and the filer will receive a penalty notice. This is a programming issue that the IRS is working to resolve. A filer experiencing this issue will have to either call the IRS Help Desk or respond to the penalty letter sending the supporting documentation. The penalty will be abated and their file will indicate that this was an IRS error. Filers are advised that to avoid this problem in the future, either submit the Form 5558 extension request earlier in July, and/or submit the Form 5500/8955-SSA later in August to give the Form 5558 extension request a chance to be posted to their account prior to the filing. As the IRS Representative noted, they are working on resolving the issue, but it will take some time. In the meantime, please contact support@ftwilliam.com or the IRS with any further questions you may have in regards to this issue.

well, (ii) says you have to be a participant to be able to be excludable under these rules. If one has met the eligibility requirement but is not a participant you are out of luck based on those rules item (iii) says the plan has to have a minimum service or last day rule which you don't have so I don't see how the rules would kick in. as you indicated, generally such a person would be benefiting anyway and by item (i) you couldn't exclude them anyway even if you wanted to.

My notes say the deferral limit for 2013 is 17,500 and catch up is 5,500 for a total of 23,000. am I missing something? remember, in regards to excess deferrals, they are calendar year. you send in your tax forms to the IRS. they look at all your W-2s. they add up all the deferrals and they know whether you are over or not, regardless whether you report it. therefore the deadline is April 15 (no extensions) this is why you could end up getting taxed twice if excess is not removed timely. once for the plan year when it was due, and then a second time when the distribution actually takes place if it is late.

doesn't matter whether it is DB or DC see 1.410(b)-6(f) certain terminating employees (I) employee does NOT benefit during the plan year (ii) employee is eligible to participate (Thus in a controlled group scenario you can't exclude those folks who aren't participants in the plan) (iii)plan has a minimum period of service requirement or last day rule (iv) participant fails to accrue because of (iii) (v) employee terminates < 501 hours (vi)basically says you could ignore all of the above and include all employees (this would be done if you HCEs in this group) ............................. in addition, your statement "treating anyone entitled to the accrual as benefiting" might not pass the smell test. years ago the IRS issued a memo indicating a participant should receive a minimal accrual of .5% to be considered having a meaningful benefit in addition, since you indicated this is part of a large controlled group, hopefully the DB plan has at least 50 people benefiting otherwise the minimum participation test fails unless it meets one of the exceptions.. shultz db memo.pdf

well, one example of such a plan that exists (though possibly not intended to be a cross tested plan) is a top heavy plan. such a plan has a 1000 hours requirement, so non key active employees working less than 1000 hours would receive only a top-heavy. thus such a plan has 2 'allocation' groups' the regs actually have this as an example under 1.401(a)(4)-2(b)(4)(vi)((D)(2) phew! under multiple formulas such a formula would be considered 'safe harbor' providing you pass 410(b). such a plan would probably pass using 'allocation rates' in testing rather than converting to accrual rates. in reality, this is somewhat a moot point. it is not much more different than the plans satisfying 'broadly available allocation rates' under the gateway minimum rules! of course, if you don't pass 410(b) then you have to provide the gateway, so you probably haven't accomplished much.

Paper is Not Dead (I guess I should have seen the punch line coming...) http://www.youtube.com/embed/V gOZDWQj3Q?rel=0

well, there is only 1 more month before the figures will be released, but based on the figures currently available (and of course assuming my spread sheet still works) next year we will have 260,000 as the comp limit 52,000 as the 415 limit 170,000 key ee 210,000 DB limit everything else "same as same as". I can't imagine things changing so dramatically to change these results. catch up is getting close to incrementing (up to 5934, but regulations permit increases in units of $500) if I understand things correctly, (at least my mom might be curious) soc sec increases would be based on the following CPI-W values for July Aug Sept Of course we won't know Sept 2013 until next month, but again, I can't see things changing that much - I will use June July Aug for 2013) 2012 values 225.568 + 227.056 + 228.184 = 680.808 2013 values 230.002 + 230.084 + 230.359 = 690.445 690.445 - 680.808 = 9.637 9.637 / 680.808 * 100 = 1.4% increase

while you may change a vesting schedule, anyone with 3 or more years of service has to be given the choice of the old vesting schedule or new vesting schedule. IRC 411(a)(10)(B). now, lets suppose for ADP/ACP testing you choose the default (prior year testing). would you argue you can use 3% first year look back or not? I'd think the answer would be yes, and conclude (possibly incorrectly) that the old document made no sense in saying match was 100% vested when in fact it didn't even exist.

that's what the regs say, by last day of taxable year. kind of hard to know since you don't know what comp you will have. but I guess that is the comment in the preamble, you need to actually have the comp when all is said and dine. as I recall from an ASPPA conference, the IRS indicated if you deferred during the year and ended up with no comp, the deferral comes back due to 415 violation. well, of course how does anyone know when the partner has actually made his election. kind of hard to prove, but I guess if you ever were audited... but all that aside, I suppose you could write an election form to say something like 3% if my earned income is $x and 5% if my earned income is $y and have it signed by 12/31 (or whatever)

see also the 401k regs page 12 of the pdf (from the preamble) - can make deferrals even before you know how much comp you will (with a caveat of course) p 63 the election itself to defer (I assume and how much) is supposed to be in place before the end of your tax year. 401k regs.pdf

if a ps is made, then top heavy free is lost there is nothing to stop you from 1st providing 3% to those who didn't defer to cover top heavy, then allocate the rest to everyone else. Or another way to look at it, you allocate 1.4% to everyone, but some people didn't receive enough for top heavy. so you 'skim' enough from everyone to provide top heavy to the people that need it. hopefully there isn't an HCE non key in the group.

if everyone deferred at least 3%, and therefore received at least 3% match, then that would be acceptable (assuming match can be used to satisfy top heavy) well now, if you are using Relius it depends on how things are checked. 1. does the system even know the plan is top heavy for the year 2. was Top heavy checked on the transaction 3. what is checked for 'match excluded from top heavy' 4. what is checked for top heavy first or top heavy skim (I tired of people forgetting to run the top heavy report, so my 'plan spec' report prints a great big message "Top Heavy not processed " if you forget. even I forget from time to time, so it comes in handy)

I almost asked what software you are using earlier today. Haven't heard from any of our clients. we use FT William. had one form accepted 8/8, bot nothing filed on 8/9.

the rule of thumb is any NHCE who is considered benefiting under the plan must receive the gateway. (Aside from the items AndyH lists above) you are most likely aggregating the plans, so for coverage, do you consider the person to be benefiting for nonelective purposes? In addition, most such plans are top heavy anyway, and the terminated person who received the cash balance probably had 1000 hours. therefore he has to receive a top heavy in some way shape or form.

what is being referred to under BRF (1.401(a)(4)-4) is 'current' availability and 'effective' availability. Currently - can everyone defer enough to get the match? Well, yes, of course, there is nothing in the plan itself that says "group B can only defer <4%" Effective - probably best summed up as "are people actually able to afford to defer 4%?" e.g. what if all the NHCEs are making minimum wage? generally there is no mathematical test, it is facts and circumstances situation. a rate of match based on svc can be proved mathematically because it is not dependent on how much one defers, but rather simply service and you can split the HCEs and NHCEs based on that - that would fall under 'current' availability. a rate of match based on how much one is deferring (especially if the rate is higher as deferrals increase) gets a lot tougher to prove - just because HCEs don't take advantage it doesn't mean a plan passes. That is why, for instance, the safe harbor 401(k) plan rules have the rule the rate of match can't increase as deferrals increase, you can't receive a discretionary match greater than 4%, etc.... someone once asked at an ASPPA conference could you start an automatically enrollment at 6% and the response was such a plan would probably fail effective availability.

somewhat along similar lines: EPCRS, Appendix B, section 2.06 ( it's example 24) has an example of failure to limit comp in 2006 they gave 8% profit sharing but used comp of 250,000 instead of limiting it to 220,000 the correction method is (Tom's wording) Hey idiot. just take the stupid extra $2,400 (adjusted for earnings) out of the participant's account and put it in an unallocated account to be used to reduce future contributions.

well, I certainly don't think the 'intent' of safe harbor plans would be to have this situation arise. Arguably, such a situation could lend itself to the possibility an HCE would receive a match at a higher rate than an NHCE, which of course is forbidden. For sure it would happen if you based the rate of the definition of comp on which one could defer. oh crap. I just realized. this is post number 5330 for me. does that mean I have to fill out a form? Hopefully at least a few of my posts have been informative.

see PDF page 132 (actual page 129) of the LRM (Listing of Required Modification) this was released 10/2011 so probably the most up date. About the closest I can find on the issue. lrm.pdf

ERISAtoolkit: I wasn't even thinking about that possibility. that sounds so artificial. I can't let an HCE have his own outside directed investments within the plan unless it is available to everyone, but can get around that rule with in-service withdrawal. Masteff: I looked it up from am ASPPA talk I gave a few years ago on distributions. Makes me feel good, because those were the same cites I had listed, so at least someone else has the same info I had! Completely forgot about that talk.

plus the Code 414(q) simply says and HCE is ANY employee who 1(a). was a 5% owner (b) in the preceding year had comp >80,000 (as indexed) there is no exception for someone who had comp while member of an excluded group