MWeddell

WMiller, The explanation of the true-up match doesn't change my conclusion. On the second part of your post, I would tell the client to contact legal counsel. The plan document and the notification to employees issued in late 2013 both say (1) the plan is 401(k) / 401(m) safe harbor and (2) all eligible employees must be employed on the last day of the plan year to get the match. Both of those statements cannot be true simultaneously, yet they are in your plan document. You already are violating your plan document, which is a potential disqualification concern. Amending the plan document now to eliminate the EOY condition for NHCEs is kind of like a "no harm, therefore no foul?" defense, but it doesn't really work legally. The technically correct answer is to make that change and do a VCP filing, but that's imposing a fair amount of extra costs on the plan sponsor.

That's right. You cannot impose a last day rule on a non-safe harbor match even though the 3% nonelective is the safe harbor contribution. The rule is in Treas. Reg. Section 1.401(m)-3(d)(3) and it applies to 401(m) safe harbor plans regardless of which dollars fulfill the safe harbor requirements.

Suppose that the employer wants to comply with the 401(k) and the 401(m) safe harbor rules. You will need to make all eligible NHCEs eligible for the true-up match without regard to whether they were employed on the last day of the plan year. You may require employment on the last day of the plan year as a condition for HCEs to receive the true-up match (although one suspects that the employer will abandon the idea completely once you inform it of the restrictions). If you want the explanation, let me continue. The match must be evaluated as a full plan year match because the true-up portion does not meet the timing requirements for safe harbor matching contributions made during the year. Evaluated based on the full year, you cannot have any NHCE have available to him or her a lower available match rate than what an HCE has available. The only way to do that is to have all NHCEs receive the true-up match.

The contingent benefit rule does not apply to plans sponsored by a "church" or a "qualified church-controlled organization." I don't mean to quibble, but those definitions are narrower than the ERISA Seciton 3(33) definition of "church plan," so you can't gloss over the details.

The contingent benefit rule does apply to 403(b) plans. See Treas. Reg. 1.403(b)-5(b)(2)(last sentence).

A participant does not have a right to access his or her account without spousal consent if the plan document provides for spousal consent rules. The plan document is what establishes rights and rules. Trying to put this into other words ... Code Section 401(a) generally is structured to provide that to be a qualified plan, one must meet the following conditions. It doesn't say that a plan can only provide for the following rules. The background rule is that the employer can put into plan document any provision that is not forbidden by Code Section 401(a) and ERISA.

Yes to karl and yes to BG5150. One can elect to receive the after-tax portion of the distribution and to rollover the pre-tax portion. Roth 401(k) (unlike Roth IRA) partial withdrawals are taken pro rata from earnings and contributions. However, that doesn't prevent one from rolling over just the earnings portion.

I don't see any problem with permissively aggregating the plans for testing purposes based on what you said.

If the lump sum payment is the primary form of benefit payment and the annuity forms are optional, then the J&S rules only affect a profit-sharing plan once a participant affects an annuity form of payment, which occurs after loans and in-service lump sum distributions are taken. So yes, you can eliminate the spousal consent for the loans and hardships. However, this is a matter of interpretation, not something that is crystal clear from the regulations, so I would tell the client to have legal counsel review the issue before an plan amendment is made.

The same question also is being discussed at http://benefitslink.com/boards/index.php?/topic/55420-loans-hardship-only/

Yes, that is permitted. As you point out, all eligible employees are covered by the plan no later than the first day of the plan year of six months after they have attained age 21 with 1 year of service.

I am guessing that all three plans are sponsored by the same controlled group or affiliated service group, so that it is one "employer." If Plan B was a safe harbor plan, then there is no ADP testing and (assuming no traditional after-tax contributions) no ACP testing. As long as the HCEs in the Plan B weren't allowed to defer to the other plans during 1/1/13 - 3/31/13, there's a exception to that odd HCE aggregation rule that will protect Plan B's safe harbor status. If Plan C's definition of plan year indicates that it began 4/1/2013, then the Plan B HCE contributions won't be included in the Plan C ADP / ACP tests. If the plan year for Plan C instead began 1/1/2013 but just the cash or deferred arrangement started up April 1, then you probably have a problem and need to aggregate those Plan B HCE contributions.

Must follow the allocation conditions in the written plan document. If the written plan document allows individual employees to waive their contributions, then you have to interpret the waiver. I doubt that the waiver clearly includes that the employee waived the right to receive contributions to the qualified retirement plan.

On the original post, I don't think you go back and rerun year 1's ADP test to generate prior year data for year 2. This was emphasized in notices or revenue procedures shortly after Pension Protection Act of 1996 was passed, introducing the prior year testing method for NHCEs, although none of the fact patterns matches yours exactly.

That is still generally the case. See Treas. Reg. 1.401(a)(4)-9(b)(3) for the special BRF testing rules when you are aggregating DC and DB plans.

It is easier for the 401(k) third party administrator to not have to deal with loan refinancings, but they are still permitted. Here's an article posted by a law firm just last month: https://www.wnj.com/Publications/The-Benefits-and-Pitfalls-of-Refinancing-401(k)-Pl

There might be BRFs in Plan A that are not present in Plan B, which would create a problem. That sounds like it may be less expensive to correct than a coverage test failure though.

Well, there are times (for example, one plan covers two locations and there are different match formulas for each location), where the plan may pass ACP testing and flunk BRF testing on the availability of match rates. My interpretation does not complete nullify the need to perform BRF testing on the availability of match rates. I am suggesting that it would strike me as unusual for (1) there to be one match formula currently available to all eligible employees, (2) for the plan to pass the ACP test (with or without the need for corrective action), but (3) the better match rate was effectively available to a group of employees that substantially favored HCEs.

Belgarath does point to an open issue, that cannot be completely dismissed. That said, I think it would be an hard argument for the IRS to make stick that the match rate was effectively available to a group of employees that substantially favored HCEs based on all the relevant facts and circumstances if the actual matching dollars allocated (either with or without corrective distributions) satisfied the ACP test.

The ERISA Outline Book covers the topic. No, the March 15 quasi-deadline is not the type of deadline that is automatically extended to the next Monday.

Thanks. That was helpful. Seems to me you get to the same result more directly. Dr. X owns 100% of both companies. Because he directly owns 50% of the dermadology practice, then the other 50% ownership by Dr. X's wife is attributed to him. Dr. X's direct ownership in the derm practice means that the exception from the spousal attribution rules in Code Section 1563(e)(5) is inapplicable. [Note, I don't usually work with small businesses, so I'm not an expert on family attribution rules.]

Why would having a child create a controlled group?

Depends on the type of plan and what is the plan document. For 401(k) elective deferrals, Treas. Reg. 1.401(k)-1(d)(1)(i) lists an employee's disability as a permissible distributable event. This is a separate distributable event from "severance from employment," so there must be some circumstances when one has a disability but not a severance from employment or else the wording is redundant.

I would say that the transition period ended February 2014. I would characterize last year's actions as "no significant change in the plan or in the coverage of the plan other than the acquisition" (which may include a stock acquisition). There was a significant change in the employer but not in the plan or the plan's coverage.

See Treas. Reg. 1.410(b)-4(b). I think it depends on how many employees are in each allocation group. If the groups tend to include multiple employees in each group, then the "reasonable classifications generally include specified job categories" language indicates one can run an average benefit test. If the groups tend to each have one person in them, then the key provision is "An enumeration of employees by name or other specific criteria having substantially the same effect as an enumeration by name is not considered a reasonable classification." Sounds to me like your client is in the second situation and that the plan has failed to satisfy coverage testing.