Bri

On both a 5500 and an SF, the opening wording to question 4 or 10 (and their a through i or n subsections) says, "During the plan year:" So I don't think you're properly answering the question if you obtain it after the year-end but still suggest it was covered (unless retroactive). And as for the "part of the year" scenario - I think it's reasonable to say that "the plan year" does not specify the entire plan year. Meaning, if you got your coverage on December 31, then the answer of "yes" is a true answer because during the year it was indeed covered. Heck, I'd suggest you get to leave "yes" as your answer if your policy expired on December 30. (Recall on the SF, you're asked if the plan had loans, even if the they're all paid down to zero by 12/31, so you still admit there were loans, even if it wasn't all year long.) As for the dollar amount, I'd use the largest amount of policy in effect during the year. Or if the plan just uses an inflation guard, I'd use the 10% BOY amount.

I'd think there had better be a compelling alternative reason not to follow the plan's terms. I suppose "may" buys you some leeway, but as to how much.....and more importantly, why? They obviously signed a document presuming folks would come with rollovers ahead of their plan participation, so why don't they like the idea now?

I think what that's less-than-simply saying is, if the plan doesn't make its "computed more frequently than annually" match by the end of the next quarter (as we normally see required with pay period match computations), then by failing on that timing provision, the plan is essentially falling back into an annual computation and the true-up would be required. (Which makes sense because the plan would have bailed on its obligation as a pay-period calculator....and an annual calculation is what then leads to needing true-ups in the first place.)

A plan doesn't have to fully vest account balances upon Early retirement age (as opposed to NRA). Will there be a noticeable number of participants who might not be getting vesting credit each year? If you require 1000 hours a year for vesting credit, but everyone works that much anyway, then sure - anyone getting to the 5th anniversary of plan participation will have earned 3+ years of vesting credit. But if you have a good number of part-timers, you might have folks who don't typically vest in their match contributions - that would make the ERA definition more important, in terms of requiring years of participation to it. Typically in a DC plan I see the only "benefits" of naming an early retirement age are (a) possible waiver to allocation requirements for the year of termination, (b) possibly having in-service withdrawal ability tied to an ERA, or (c) getting less-than-65ers a faster track to 100% vesting. If terminees already can get distributions upon termination of employment, it's not as though the presence of an ERA speeds up their ability to receive benefits.

Technically speaking, can something be an "eligible rollover distribution" if it's not an eligible distribution in the first place? If the guy kept the cash, then that probably eliminates the potential to hold "your rollover is ineligible" over him, I suppose, in an effort to get the funds back.

So he turned 72 last year which means his RBD was 4/1/23, right? So his minimum at that point was still zero due to nonvestedness. Now we have an increase in the vested accrued benefit, which typically you have within 12 months to adjust to the payment stream, I think.

When's the RBD? Did he JUST turn 73 or did he JUST become vested in the accrued benefit? (Making sure the RBD hasn't already passed but with a $0 payment due earlier due to it being a year he was still nonvested)

Plan's been frozen so no new benefits after June 30. Plan's staff contribution credits tend to suck so bad that they always have to do an -11g amendment to pass 401(a)(26). (Annual $500 bucks each, allocated formally as $125 for each quarter they're active during, and then the -11g stuffs more into Q4 as necessary for folks.) Well now they're only going to be getting $250 for the most part. But is there any "accepted practice" for whether it's appropriate to measure the 0.5% benefit level relative only to the compensation through the freeze date? Otherwise everyone's going to be at HALF their typical accrual rate, if it's expected to be measured on full year's pay. --bri

Agree with Belgarath - use already-plan money to pay the premiums as an "investment transfer."

It's not just fixable as 415(c) excess?

Isn't the easiest technical fix just to allow Roth as a benefit for NHCEs only? Adjust the "universal availability" to make it easier for sponsors.

Hey, that top heavy paragraph (ii) hints that you can include (not electing to exclude) the LTPT balances in the top heavy ratio, but wouldn't have to make a THM or increase vesting if applicable for them. Seems best of both there - maybe the LTPT drop you from 61 to 59% and nobody gets a THM. But if they still leave you at 61 while including them, they don't get their 3%.

Will small employers' advisors drive their sponsor clients to these pooled arrangements? Wouldn't they prefer more of a finger in the pie, teaming with a local TPA/advisory firm to something more tailored?

I would think it's okay - I mean, you'd typically budget a normal cost for someone who then doesn't hit 1000 hours in the upcoming year, but it was the actuary's best guess as of the val date. You could consider a turnover decrement for people scheduled to hit 7/1/23 that "might end up turning out differently" as well, right? I realize this is just my gut but it "feels" acceptable.

Might they be able to get the -11g amendment executed prior to 3/15/24 rather than 10/15/24, and then they could do a 412(d)(2) election for it?

My "7-10 Split Plan" is just to roll it hard and hope for a bounce across the lane.

If the contribution is that onerous, though, they could always reduce their 2023 deferrals so that the business can afford it before 12/31/23.

Thanks, Lou - That was my thought as well, to have other assets outside that recordkeeper, so their version of the prepared 5500 would be wrong, but that's fine if the sponsor only actually files our replacement version instead. As it's not my plan, I was actually more curious about the legal argument, than how my colleague's gotta deal with it 🤪

So, here's 1.401(k)-1(d)(4) from the Cornell Law website folks The fact pattern a co-worker gave me was interesting DB/DC combo, plan needs to make DC allocations for 2022. Plan sponsor's assets were sold/employees terminated in 2022. DC plan has been already terminated, though, and everyone's paid out, maybe except for one straggler. There's concern their big former DC separate recordkeeper would balk about opening the plan back up after termination/payouts. Anyway, the "interesting part" concerns whether or not the sponsor can just start up a separate DC plan, retroactive to 2022, to receive these allocations for the testing. This regulation seems to preclude distribution upon the termination of a plan when the sponsor's going to set up another plan within 12 months. The thing is, in this case, all the employees' distributions are contingent upon their termination of employment with the Seller. And the owner has a distributable event upon age 59½. So none of these distributions seem to have the plan termination as the distributable event - does that mean a successor DC plan is going to be okay after all? --bri

I did just look to see what 8/30 would convert to as 6 months later and it came to 2/28. Starting with 10/31, though, it came up as 4/29. (This was straight plan entry-date calculations as opposed to benefit-eligibility calculation. Maybe Relius does those calculations identically, but I can't say for sure.) Interesting....

Yeah, I always chuckled at the basic concept of "the plan sponsor has the option to file the same version of the 5500 as last year" when they're in the 80-120 range. As in, why pay for the audit you don't have to get, now that you've come in under 100 for the first time in ages? Not like anyone jumps at the chance to do the audit at 101, they always wait until 121.... (don't they??)

Peter - I just took a quick peek at how Relius sets up distribution definitions, and it's looking for a minimum age of years/months. I don't have a dummy plan to test it on, but that sounds like they'll simply add the number of months to the birthdate day number.

Hopefully they aren't also staring at their 415 max lump sum on the other end of the calculations....

This will be easier once the calendar goes metric.

What about the exception for disability?