Tom Poje

Someone needs to tell the IRS that people should not read their example 7 too closely because it refers to the rate groups not satisfying things rather than the formula. Example 7 The facts are the same as in Example 6, except that the classification of employees who are entitled to benefit under the formula that applies to H1 includes N1 and N2, who are identified by name. Under paragraph ©(3)(ii) of this section, the rate group with respect to H1 does not satisfy the nondiscriminatory classification test under § 1.410(b)-4 because the classification of H1, N1 and N2 by name does not satisfy the reasonable classification requirement of § 1.410(b)-4(b). Therefore, the rate group with respect to H1 will satisfy paragraph ©(3) of this section only if the ratio percentage of the rate group is greater than or equal to 70 percent.

with a caveat at the 2010 ASPPA Conference Q and A 3, in regards to a terminated plan and top heavy the IRS person indicated the following: (1) Of course, if there is no employer contribution, there would not be an obligation to provide top heavy minimum contribution. But, if there were contributions to keys during the year, including elective deferrals, there is a top heavy minimum based on compensation and employment through 9/15/10. Plan must liquidate within a reasonable time under Rev. Rul. 89-87 or else 9/15 date may not be reasonable. There is effectively a short plan year for top heavy purposes. in other words, if the assets aren't paid out then you could lose the 'short plan year'

sort of: we gave you folks a chance to do something a little extra you have carried it to such an extreme that it is an abuse so time to reign it in somewhat.

The text itself (somewhat similar to Ferenzy article) was from FT Williams commentary, but it is still all based on what is in the Fed Register of the proposed regs sorry to be so confusing

on the other hand, the db/dc combo plans have resulted in some extremely large percents to HCE and at the most 7.5% to NHCEs, arguably an abuse of how the things should work. one of the problems is when the regs were written they fixed the DC interest between 7.5 and 8.5%, which at the time was probably understating things. so, can you imagine if instead they tied the testing % to the current interest rate?

they wouldn't say it is impermissible, just that you give up your rights to the avg ben test when performing nondiscrim testing.

I made it up and cleverly typed it up to look in the same format as the Treasury would, what else. but you caught me! see if the following link will work, though I simply copied what was there. (or also see Irene Ferenzy's article that has just been posted under 401(k) plans) https://www.federalregister.gov/articles/2016/01/29/2016-01675/nondiscrimination-relief-for-closed-defined-benefit-pension-plans-and-additional-changes-to-the

Thinking about this some more: if these regs go through this year and are applicable for plan years beginning 1/1/2017. (and assuming things work like they do, not until late in 2016. lets say your plan has no hours requirement or last day rule or whatever, but because you have a group by name you can no longer can use the avg ben test to pass testing. I guess you would need to amend by the end of 2016 to change something because once 2017 starts someone has 'accrued a benefit' under the old formula. one has until the end of April to make comments. and then we will see how fast things move from there. I certainly understand the IRS attempt to curtail some of the things I have seen - the formulas in which the owner gets 'a gazillion %' in a DB/DC combo plan just because of the mathematics of the way things work out and the NHCEs get the gateway only.

one of the possible changes: Under the current regulations a cross-tested plan can pass nondiscrimination testing using either the ratio percentage test or the average benefit test without requiring that each rate group be considered a “reasonable classification”. Under the proposed regulations, this will still apply to the ratio percentage test. However, in order to use the average benefit test, the rate groups will need to satisfy the reasonable classification test. Of greatest concern are plans where one or more of their rate groups are set by naming the individuals as traditionally this has not been considered a reasonable classification. If these proposed regulations become final, new comparability or cross tested DB/DC plans will need to review their plans to determine if (1) they can pass testing using the ratio percentage test or (2) their rate groups meet the requirements to be a reasonable classification so that the average benefit test can be used. Plans that cannot would need t o be amended to ensure that nondiscrimination testing could be passed. .............. so while the reasonable classification test used to only apply to coverage, it would now apply to nondiscrim testing as well if the proposed regs go through. I did submit a comment for clarification if 'one group per participant' is considered reasonable or interpreted as being 'by name' proposed reg.doc

let's suppose, instead of the NRA getting changed the plan had no hours requirement for a contribution. they amend the plan to 1000 hours. so three years later would you say the person is entitled to a contribution? I think any benefit accrued up to that point in time would be eligible for a distribution at age 59 1/2 (but that is protected benefit for distribution) but I don't think that rule applies to a right to a contribution. (unless you changed the nra in the year the person reached NRA) If the person was an hourly employee and you amend to exclude hourly employees would you say "but previously you said I would get a contribution in year of retirement, so I have to get it"

in 8 years the person never noticed or said anything? good grief, I can see some fun stuff on this one. "In addition to no deferrals being taken out, I came home from work and the wife and kids were gone. Come to think of it, I just noticed they haven't been around for 8 years either." since an SAR has to be given out yearly and the person should have received those, so he knows plan is on going and never said anything about deferrals not being taken out... since the plan would have been amended for EGTRRA within that time and the person should have received an SPD and never said anything. about deferrals not being taken out .....

somewhat similar to the IRS at the 2010 ASPPA Q and A at least based on the chicken scrawl note I wrote, since this was not in the handout. the handout merely said to be discussed from the podium. Q and A 27 (though they weren't numbered) The issue was raised about forfeitures arising from related match (those matches forfeited due to a return of deferrals for a failed ADP test). The response was that those are probably not “forfeitures” in the sense that is commonly used—resulting from unvested balances

the IRS frowns on forfeitures being used to fund safe harbors, because safe harbors are supposed to be 100% vested when made to the plan. the $ you are talking were not a result of someone forfeiting due to not being 100% vested. in fact, the $ you are talking about were actually 100% vested when made (even if done so in error) Maybe the 'line' is too fine, but I would hold their is a distinction in this case. so I could certainly be in error.

I'm more than curious if that is what the document really says. unless it was a QACA in which case you could actually generate forfeitures, I would have thought the incorrect match is tossed into suspense and used to reduce future safe harbor (it was after all 100% vested when made inadvertently to the plan.) I'd argue it is a correction under EPCRS, you are putting the plan in a position it should have been if the error hadn't occurred.

many years ago I worked out the following spreadsheet, only requires entering the new taxable Wage Base each year. hasn't failed me since 2005, well maybe once the rounding was off $1 for one year. so even if the IRS is slow in posting a new table... covered comp at 118500.xls

It accomplishes at least the one thing that wasn't in the regs, namely, the issuance of a new safe harbor notice. I've noted that before, how can you change something without being able to issue a new notice.

otherwise I would intentionally give all NHCEs a huge match they shouldn't get, use it in the test to help pass ACP then forfeit it after plan year end because they shouldn't get it in the first place. clever scheme!

one such write up is found here https://www.orrick.com/Events-and-Publications/Pages/New-401-k-Correction-Procedures.aspx of course, it wasn't really written to handle your particular case in which it was a one time 'extra large' deferral. still, the concept of the employee benefiting by getting the deferral put in by a QNEC and still keeping the comp is the type of windfall this was intended to stop. and yes, this says the match is still required

I see FIRE is now back up and running. this should be fun to see how many 'late' filing there may be as a result.

the cite is buried in the concept -namely, the person did not have a break in svc (e.g. work less than 500) common misconception that you have to terminate to have a break in svc. break in svc means you might have to reestablish svc to be eligible because you had a 'break'

by their own logic then, if the person in question worked 1000 hours from 5/2 - 12/31 then they should enter 1/1 and not 7/1 which would ignore a '1 yr svc computation period' in other words they want an entry upon completion of 1000 hours in any time period not exceeding 12 months. Certainly that can be done one option with the FT William document has: date on which 1000 hours of service is completed within the computation period

My understanding it is only optional for 2015, I hadn't heard they were going to revise things. The issue, I suppose, is the DOL has the info available for all to see. The IRS wants the additional compliance info. I can't see anything in those questions that make a difference, aside from the other optional question that asks who the preparer was. The Q and A the IRS released had the following note: The IRS has introduced new compliance questions on the 2015 Form 5500-series returns. These IRS compliance questions will help the IRS to effectively focus on specific factors and issues of federal tax law compliance. Although responses to these new questions are optional for the 2015 plan year, we strongly encourage you to answer them. These FAQs clarify some of these new questions I am assuming the software will roll most of the answers over from one year to the next, so answering them this years will speed things up next year. again, since I already have them ready on a report, I'm really not putting any extra time into it.

I am assuming the following ee did not work 1000 hours from 5/2/14 - 5/2/15 but did work 1000 hours in 2015 (but not by 5/2) otherwise the question makes no sense. I would say you are now in a new 12 month measuring period 1/1 - 12/31 and so the person doesn't enter until 1/1/2016 (assuming of course the document reads that way using 1000 hours in a 12 month period...) now if the clients wants to say they worked 1000 hours from 1/1/ -5/2 (must be in the pension industry running ADP tests! to accomplish that)

currently using Relius, every time we restated a document (or received one from the client) simply entered into some of the user fields the serial number and the date. most of the rest of the info (testing method for the plan, etc) can be pulled from the system, so I simply have a report generated that provides most of the answers (about the only one I can't pull is how much in service distribution), so, at least for me, it is not that big a deal. known for a year these questions would appear and set things up as the last year went along.

agreed, it is possible the new doc requires restablishment