Doc Ument

If there are approved plans still using a "1-year" break in service, so be it, and don't be fooled by provisions that are discussing anything other than the rule of parity. As far as I know, you do not get to restart anyone as a new employee unless all prior service is disregarded, and the only time you get to disregard all prior service is because (1) the plan still has the rule of parity (of which one condition is a 5-year break in service, see IRC 410(a)(5)(D)(i)(I)), and (2) the participant meets all the conditions stated in the rule of parity (see the rest of IRC 410(a)(5)(D). Sometimes a plan uses both the "1-year" and the "5-year" provisions in the same section, which is not very useful is it? Such plans are probably trying to coordinate the provisions of 410(a)(5)(D) at the same time as 410(a)(5)(C). However, paragraph (C) of the reg is only a "temporary" hold-out, not a permanent loss of prior service credit, as under paragraph (D) of the reg. So, even if you get past the hold-out period of one year (assuming the plan uses that provision), then you qualify to progress on to evaluate the rule of parity with its 5-year BiS provision, and only then do you get to reset past service to zero (assuming that in addition to the 5-year BiS, you meet the other conditions in the plan, which should reflect the rest of 410(a)(5)(D)).

Some providers have been known to get this "allocation condition" option approved (and preapproved) by the IRS by providing benefit accruals throughout the year via a principal credit periods that are shorter than the plan year and for which there are no hours requirement, which means I guess that no accrual needs to be credited for the principal credit period that ends on the last day of the plan year if the participant is no longer employed. I am not going to take sides on the merits or lack thereof, or whether or not such a feature is useful. (I am just the messenger.)

The condition that is most frequently overlooked when applying the mid-year reduction/cessation regulation is the following requirement (i.e., not only must there be an amendment to cease (or reduce) the ADP SH contribution after giving 30-day advance notice, but the plan must also be amended as follows): 1.401(k)-3(g)(1)(i)(E): The plan is amended to provide that the ADP test will be satisfied for the entire plan year in which the reduction or suspension occurs using the current year testing method described in § 1.401(k)-2(a)(2)(ii) It is not sufficient to just amend to end the ADP SH contribution because to be qualified, a 401(k) plan must affirmatively state the specific method by which the ADP requirement is going to be met (see 1.401(k)-1(e)(7)). In this case, because one ADP-test method is being removed via a midyear amendment (i.e., the safe harbor method that had been in effect for that plan year), another method must be specified by the document to use in its place, and the safe harbor regulation requires that the plan be amended to use only the method specified above. (In some cases, though, a plan might already provide that upon any midyear amendment to cease or reduce ADP SH contributions, the plan will automatically use the current year ADP testing method described in § 1.401(k)-2(a)(2)(ii) for the entire plan year, in which case the drafter might reasonably choose to rely solely upon such pre-existing language.)

Yes and no. Can you amend the plan to prospectively remove the true-up? Yes, even though it is a prospective reduction. Can you keep the ADP safe harbor for the plan year if you make the amendment? No, because the employer did not keep the ADP SH promise it made for the current plan year. An ADP SH matching contribution can be the subject of a midyear amendment to reduce (or eliminate) the ADP SH matching contribution if the requirements of 1.401(k)-3(e) and (g) are met, e.g., 30-day advance notice of the amendment, the amendment also reinstates the ADP test (on a current-year testing basis) for the entire plan year, the distribution of a 30-day advance notice of the revised supplemental notice to participants, etc. In other words, the plan cannot operationally "revert" back to ADP-tested status, it must affirm that the ADP test will be used for the plan year. Some plans have a built-in provision of this nature (stating that the ADP test applies for the entire plan year using current-year ADP testing whenever an amendment is adopted to reduce an ADP SH contribution), and for such a plan, the amendment would not need to say repeat all that again. But rather than look for that language, I think it best to simply include the language in the amendment anyway because there is no harm done if you are only repeating something already stated in the plan, and it makes it more obvious to all parties that ADP SH status is lost. If more pain is desired, read on. To protect the plan from a likely violation of the anti-cutback rule, you would also need to protect the true-up feature for the portion of the plan year during which the plan year was used as the matching computation period. In other words, a true-up would be owed for the portion of the year starting on the first day of the PY as measured through the later of the effective date or the adoption date of the amendment to shorten the computation period. For that first portion of the PY, the rate of deferral would be the deferrals (taken into account by the formula) for the portion of the PY during which the "entire plan year" was the computation period, divided by the compensation (taken into account by the formula) for the portion of the PY that the "entire plan year" was the computation period. Only at that point can the match become determined using the payroll period computation period.This paragraph would also apply even if the fixed match formula was not an ADP SH match. But for the ADP SH plan, the effective date of the amendment changing the matching computation period to a shorter period must be at least 30 days into the future, so you must essentially add still another month to the portion of the PY during which the true-up calculation must be protected. I suspect that the employer will reconsider the risks and benefits associated with making such an amendment.

I agree with my immediate predecessor responder. I respectfully dissent a whole lot from at least one of the prior commentators. First, the QJSA/QPSA provisions must be used by every pension plan, i.e., both DB and money purchase DC plans. Second, the QJSA/QPSA provisions must be used for any other plan (i.e., profit sharing plans) unless the drafter takes advantage of the exception granted to such plans by drafting the plan to contain (or the AA option election that activates) the three conditions stated in the regulation for the profit sharing plan to be exempt from the QJSA/QPSA rules. The first condition is that 100% of the participant's account balance is the spousal death benefit. Both of my enumerated statements are right out of the first sentences of Regulation 1.401(a)-20 Q&A#3. You should try to find the "no-QJSA/QPSA" provisions in the plan. If you have a DC basic plan document with a QJSA/QPSA Section, and you do not see that non-QJSA/QPSA language in that Section, then there will be another Section with those provisions, and in that case, there should also be an AA option referring to the latter Section, and which would need to be selected if there was to be no QJSA/QPSA). For the basic plan document, a good search term is "412" because the plan most likely mimics the regulation, i.e., "for plans not subject to 412,..." The regulation is copied below. That means that for profit sharing rule that contains the three provisions in the cited regulation for the account balance to be free of the QJSA/QPSA rules, the new spouse is the beneficiary unless that spouses waives his or her right to 100% of the participant's account balance. I agree with the statement that 100% of all death benefits must go to the beneficiary, but the term "beneficiary" will be defined by the plan to first confer spousal benefits and then refer the employer to examine the designation form if the spouse has so consented to non-spouse beneficiaries. It sounds like the participant made the assumption that the old designation would survive. Not so. Just like participants should re-do designation forms immediately upon divorce (so as not to inadvertently leave their ex as still being the beneficiary they specifically designated), they should re-do them immediately upon marriage, at least if there are to be non-spouse beneficiaries. In the absence of a QDRO saying otherwise, the old designation form must yield to the language in the PS/K plan that likely states something along the regulatory lines of: (1) The plan provides that the participant's nonforfeitable accrued benefit is payable in full, upon the participant's death, to the participant's surviving spouse (unless the participant elects, with spousal consent that satisfies the requirements of section 417(a)(2), that such benefit be provided instead to a designated beneficiary); (2) The participant does not elect the payment of benefits in the form of a life annuity; and (3) With respect to the participant, the plan is not a transferee or an offset plan. (See Q&A 5 of this section.)

I think the IRS position would be that there was no plan until, at the earliest, 1/1/17. I say that because I don't see the distinction between a discretionary amendment and the discretionary act of adopting a plan. In both cases, I would conservatively argue, the latest possible date of adoption by the EMPLOYER is the last day of the plan year of implementation (i.e., using the discretionary amendment rule stated in a series of IRS Revenue Procedures). There might even be a better authority or doctrine that states that plans must be adopted by the end of the first plan year, but offhand, I cannot think of it. (I am almost certain there is another, better source (citation) for plan implementation.) In any event, I think a poll of practitioners would show 99% agreeing (for one reason or another) that the employer's deadline for signature is 12/31/16 for a calendar-year plan being implemented in 2016. In fact, you can Google it yourself, because at the end of every calendar year nearly every accounting firm handling personal accounts publishes a newsletter saying something along the lines of "unlike an employer tax-qualified plan, which needs to be adopted by 12/31/xx, you still have time left after the end of the year, until the tax filing deadline of 4/15/xx+1, to adopt your own IRA or a SEP retirement plan." On other facts, I might suggest that you have the situation where the single trustee signature on 12/19/16 is also the signature of the employer, i.e., one individual can act in both capacities. I think you would then have good reason to assert the plan was (in fact) established in 2016. I can only assume, though, that this is not so, since you now add that "the employer" signed on a date other than the date that "the trustee" signed, and so I can only presume that the employer is not also the trustee. Note that if this is a 401(k) plan, then in accordance with the 401(k) regulations no deferrals were permitted to be deducted from pay until the employer signs the plan document (and the document should so state), which means no deferrals could be taken out of pay until 1/5/17 on your facts, even if somehow you can argue successfully that the plan started in 2016 for purposes related to deferrals. You cannot easily correct this "error" because the date of signature is not an error. It is a legal representation by an individual attesting to an event that occurred at a specific time. The only error here is that the employer didn't know to sign (or knew but failed to sign) the document by the end of 2016. That is not a plan error. It is not a scrivener's error, which is when a plan document contains unintended language. You can deduce intent from the fact that the employer operated a plan in 2016 by all the rules but for the date of execution. Maybe that will work, and maybe it won't. (You can tell the IRS that you intended to file your tax return by the tax filing deadline, but it won't do you much good when you get an unsympathetic bill from the IRS.) The only "correction" that I can find is to re-do the administration retroactively as if 2017 was the year of implementation, and use VCP to obtain a resolution stating that all necessary retroactive corrections were correctly corrected. You are best advised to find legal counsel who has assisted similarly situated employers in the past, i.e., perhaps making the argument that the operation of the plan is sufficient evidence of the intent to operate a plan (to which I say "good luck" with that, but what do I know?). I can't help but point out that after you convince the IRS of the fact that the plan really started in 2016, you may find yourself needing to also convince the DOL of that same thing, and of the two, I would prefer to tangle with the IRS on tax matters than with the DOL on labor matters. That isn't to say that the DOL will not be sympathetic (depending upon the facts), I merely point out that the IRS is not the only sheriff in this town. Think of the plan document as being like a will. The will was drawn up correctly, but perhaps it was executed too late to enforce the intent of the deceased as to a particular beneficiary or the disposition of a particular asset. There is nothing to fix in the will itself, and, in any event, it is too late to execute it at a different time. Maybe all the stakeholders will agree to acknowledge what can clearly be shown by other evidence as the signer's intent when drafting the will, but maybe not.

I can imagine a public or private auditor or other investigator reviewing two plans that each say "this Plan shall make a 3% top heavy contribution for all non-key employees for any top-heavy plan year" and requiring the employer to fund that 3% contribution for employees who are in both plans (notwithstanding that there is a regulation saying it is unnecessary). And since in my example the plan actually stated "3%" without saying more, then I would even say that the full 3% would be required even if no key employee got more than 2%. Why? Because the plan says only "3%" and it fails to set forth the exception for when the non-key with the highest allocation rate gets less than 3% (in which case you need only provide non-keys with whatever that lower rate is). I say all this because the regulation doesn't say that an employer can disregard a plan's TH language, it only gives permission to draft a plan with a non-duplication provision in it with respect to employees who are in both plans (i.e., eligible for a contribution in both plans). See Questions M-1 (each plan responsible for its own participants) and M-8 (...unless an employee is in both plans) in Regulation 1.416-1. Typically, the IRS allows employers to say (or the employer to elect) that a particular plan is "primary," such as a money purchase plan or new comparability plan.

With regard to "shifting" and what to do when the 1,000th hour is met, the rules for eligibility and vesting can have dissimilar results. The first eligibility computation period (ECP) must be the 12-month period beginning on the date of first hire for all plans. So a plan that wants to use the plan year for eligibility must "shift" to the PY after that first ECP to that first overlapping PY. That is not necessary for vesting, i.e., the first vesting computation period (VCP) does not need to be the first 12 months after the date of hire (although it can be), so there does not need to be any shifting or any overlap (a plan can use the PY for vesting without regard to when during the PY the employee is hired). Also, even when plan is silent, many practitioners do not credit a YoS for eligibility until the last day of the ECP. One reason is that the employee can't enter the plan if they leave during the first ECP and never come back, even if they have the 1,000 hours. In contrast, for vesting, once the 1,000th hour is credited, there is no point in not vesting at that point, i.e., there is no reason to wait until the end of the VCP as the vesting credit will need to be credited no later than the end of the VCP anyway (you don't need to be employed to get the vesting credit). If you were to pay out immediately at the 1,000th hour without giving credit for that YoS, then you will end up needing to make another distribution to supplemental the first distribution to reflect the higher vesting percentage at the end of the VCP of severance. Everything above is subject to specific plan language to the contrary. For example, a document might have more specific provisions, such as plan entry upon when the 1,000th HoS is credited during an ECP, or perhaps even a "parallel" track for shifting the VCP so that each VCP coincides with each ECP. The text of the first paragraph of this response is intended only to illustrate how and why the YoS requirement will often play out differently for eligibility than for vesting for many plans that, despite the seemingly odd results, are fully compliant with the Code.

There is nothing to preclude a document from stating that the employer that "starts" the plan can choose whether other companies may adopt the plan, and to impose conditions on such adopting employers joining or staying on the plan. For example, a plan that does not contain any provisions for unrelated employers might require that any adopting employer continue to be a member of a controlled group or affiliated service group. That is why the attorneys who structured this deal should have read the document before proceeding with the transaction. That is why no such divestiture should occur without all the plan's advisors providing advice before the transaction takes place. I am not an expert on such matters, but it sounds to me like this group is still a controlled group (since 5 or fewer persons together own presumably equal shares that combine to a percentage over 50%). Even if I am wrong, it could be an affiliated service group. So, even if the plan is designed only to be a "single" employer plan, you might still have a single employer plan (i.e., a plan that is adopted only by related employers within the meaning of IRC 414(b)(c) and (m)). So you would need to look at the document to see if there are any other less common restrictions on the degree of "permission" that exists for adopting employers to remain an adopting employer. Probably not. Or, if the brothers have non-uniform ownership among the three companies, and it is not a controlled group for that reason (see IRC 1563(a)(2)), and is not an affiliated service group, then yes, you should make sure the plan contains provisions for unrelated employers, i.e., a multiple employer plan (MEP). If not, that could be a reason to expel an unrelated employer, though I could argue it means only that the plan should have been amended to accommodate the unrelated employer. In any event, the deal that was struck should have specifically addressed this situation and not left it up to the advisors to try to figure out once the deal was done. Did the deal contain a contractual commitment that the 49% company give up the plan? If so, then the brothers have a contractual remedy outside the plan to force the company to un-adopt the plan. The successor-plan rule is generally meant to mean the restriction of the distribution of deferral-related accounts upon a plan termination. So far, I haven't hear anything about the plan terminating. Nor have you said that any employees are terminating, so I don't even see any distributable event (an employer un-adopting a plan is not, in itself, a distributable event, except I suppose it could be for some types of funds). So I don't think anyone reading this has enough facts to make an educated guess as to what options the employers have. Accountant: "So, I assume nothing major changed with your tax situation during 2018 or I would have heard from you." Client: "No, not really. Not that I can recall. Unless you mean that I got divorced, sold the house, bought a condo rental unit that I will live in until my new business (I told you about that, right?) takes off, and then I will rent it out. Why do you ask? Isn't that why we have these annual interviews on April 14th?"

To add to my earlier response, and in response to Cardscrazy, if the document provides language similar to the following, then an employer does not have the discretion to wait for an additional period of time beyond the latest time prescribed by regulation (citation upon request): (a) General rule. An Eligible Employee who has satisfied the conditions of eligibility pursuant to Section 3.1 shall become a Participant effective as of the date elected in the Adoption Agreement if employed on that date. Regardless of any election in the Adoption Agreement to the contrary, an Eligible Employee who has satisfied the maximum age (21) and service requirements (one (1) Year (or Period) of Service (or, with respect to contributions other than Elective Deferrals, more than one (1) year if full and immediate vesting)) and who is otherwise entitled to participate, will become a Participant no later than the earlier of (1) six (6) months after such requirements are satisfied, or (2) the first day of the first Plan Year after such requirements are satisfied, unless the Employee separates from service before such participation date. (c) Recognition of predecessor service. Unless specifically provided otherwise in the Adoption Agreement, an Eligible Employee who satisfies the Plan's eligibility requirement conditions by reason of recognition of service with a predecessor employer will become a Participant as of the day the Plan credits service with a predecessor employer or, if later, the date the Employee would have otherwise entered the Plan had the service with the predecessor employer been service with the Employer.

Many preapproved plans provide for discretionary grants of predecessor service, so it seems to me you can do it, but I doubt that the plan document provides for, or can be construed as allowing, a second entry date waiting period. If you would like a more precise answer, let me know. I would bring anyone in as you would do in a comparable rehire situation. However, if it is not already too late to do so, the drafter can certainly impose a waiting requirement in an indirect manner. For example, a drafter could also put one or more acquired employees into a non-service-based excluded classification until an arbitrary later date, at which point the employee(s) will pop-in on that date (under your facts). Alternatively, the drafter could make the grant of such predecessor service effective as of a prospective date, e.g., the drafter could say something along the lines of "Effective July 1, 2019, service with Pamela's Pension Professionals (PPP) shall be granted to those employees of PPP who became employees of ABC Company (ABC) solely by reason of the acquisition of PPP by ABC on or any time prior to that date." In that case, of course, any such acquired employee would need to still be an employee of ABC on 7/1/19 to enter on 7/1/19. (Not every employer necessarily wants to limit service (as I have suggested) to only those who "come over" in a particular acquisition - but it's worth the drafter asking the employer if such language is desired before any broader discretionary grant of predecessor service becomes effective. I have seen "unfortunate" fact patterns emerge from an unnecessarily broad grant, which typically occurs when a drafter just fills in the blank with the name of the "predecessor" companies. The BPD might say that any service (whenever rendered) with the named employers will be counted.) Nothing in this response is responsive to the issue of potential discrimination.

At the risk of stating the obvious, the reason why the plan lumps fringe benefits in with the other things you are referring to by saying "...among other things" is because of Regulation 1.414(s)-1(c)(3). So if you are tempted to separate out only the fringe benefits, or only the GTL, I see a potential problem with your having a definition of compensation that automatically complies with the 414(s) safe harbor (which may or may not be important to you). If you stay with excluding all those other things, then I agree with Luke that "fringe benefits" not defined for this purpose, and also "welfare benefits" is not defined in the 414(s) regulation, which is an even bigger stumbling block for many employers. It is my understanding that the only amount included on the W-2 is the cost of the GTL in excess of a face amount of $50,000, so that is, I think, the amount that would be excluded by the "fringe among other things" option, and that cost would be the amount (or typically most of the amount) of the difference between the W-2 and the 3401(a) definitions.

I've been told that I'm not good at math and should leave the math to others (which is often the case), but I am good at run-on sentences, so assuming the entire amount being described is pure severance pay, such as that which is paid from a severance plan, then it seems to me that 415 compensation is zero, and that the 415 annual addition limit would therefore be zero, and if the plan document includes in compensation (for deferral purposes) only such elements of post-severance pay (if any) that are permitted under the final 415 regulations (which is likely), then plan compensation for purposes of making deferrals, it seems to me, should also be zero, and deferrals should therefore be limited to zero as a matter of plan administration even without regard to 415 or 402(g) (though I have made assumptions about the plan's language in that regard). With that assumption in mind, if forced to choose between 415 and 402(g) (as opposed to choosing self-correction for an administrative error, if there is one), I'm with Belgarath.

To amplify duckthing's response, suppose there was a last-day requirement but that there is also an exception of that requirement in the event of death, disability or retirement, and X dies after the start of the plan year but before an amendment to exclude HCEs is both adopted and made effective. In that case, it is too late for the amendment because X, whether or not an HCE, has already become entitled to an allocation, thus affecting everyone else's eventual share of the pie. {I presume this is a pro-rata allocation method.} The anti-cutback rule applies equally to HCEs and NHCEs, so it is not an argument that you are "discriminating" only against HCEs. If the effect of the amendment is to give at least one participant more as a relative percentage of the overall contribution, then it follows that some other participant(s) will get less as a relative percentage of the overall contribution, and that is what is prohibited under IRC 411. (In the case of a fixed contribution formula, the analysis is much simpler, as each participant gets whatever the formula says, i.e., the amount to be allocated to each participant is not dependent upon how many other participants are sharing.) That is why employers might choose not to have any exception to the last-day requirement, and even when that is the case, any such amendment must be adopted before the last day of the plan year, as adopting it on the last day of the plan year (for such a plan) is a day too late.

To amplify CuseFan's response, I usually express this to clients in such a way that they cannot read it too fast, by saying something like "the deadline for such a plan to change its formula is the day before the last day of the plan year, e.g., December 30th for a calendar year plan."

That IRS webpage is NOT saying that a participant cannot receive a hardship distribution for the purchase of a residence, but is saying only that the plan must specify which events qualify and which do not, and gives an example of a plan that chooses to permits only two "safe harbor" hardship events described in the regulation (medical and funeral expenses) but prohibits another one of those regulatory "safe harbor" hardship events, namely, the purchase of a principal residence. Quote: "Thus, for example, a plan may provide that a distribution can be made only for medical or funeral expenses, but not for the purchase of a principal residence or for payment of tuition and education expenses. " In fact, that IRS webpage is defective in that it does not acknowledge that under (current) regulations, an employer may choose to bypass the "safe harbor" hardship events altogether, and use instead a "facts and circumstances" method for determining if a hardship has occurred. Of course, the regulation is changing, presumably in 2019, to take into account recent statutory changes that will render the distinction between the safe harbor method and the facts-and-circumstances method obsolete starting no later than January 1, 2020. The fact that a plan may currently choose (as noted on the IRS webpage) to not allow a hardship for the purchase of a principle residence does not mean that a plan is prohibited from doing so, especially since Regulation 1.401(k)-1(c)(3) lists the purchase of a principle residence of the employee as a "safe harbor" event for granting a hardship distribution request. It is the second item on that list of safe harbor hardship events, following the event for medical expenses. The regulatory safe harbor event for the purchase of a principal residence even goes so far as to state "excluding mortgage payments." That means that although a plan MAY provide hardship provisions for the purchase of a principal residence, a plan relying on the safe harbor event may NOT provide a hardship distribution for mortgage payments, i.e., the funds must be applied directly to the purchase of the residence. That means the administrator has no reason to contact any lender when granting a hardship for the purchase of a residence because the only factual issues to be resolved are (1) Is a residence being purchased, and (2) if so, how will the residence be used? A hardship event may be declared even if there is no loan (e.g., mortgage) and therefore no lender. Most employers adopting the "safe harbor" events tend to select all of those events, which would therefore include the purchase of a principal residence. Some preapproved plans bunch them all together, which would necessarily include the hardship event of the purchase of a principle residence. The IRS page is giving an example where an employer has selected some, but not all, of the available safe harbor hardship events, and should state that this is applicable for only plans using the safe harbor method for determining hardship events. At one time, the regulation was more liberal, as I tend to recall it once allowed even the purchase of a vacation home, but was revised quite some time ago to limit it to only the principal residence of the employee.

If it is a preapproved plan you are looking at, then unfortunately the IRS can say to the document provider "Put this in the document if you want to get an approval letter." They can do so as a matter of policy. Their demand need not be one of the listed "do's" and "don'ts" in the reapproved program Rev. Proc. If you are using a preapproved DB plan with the .5% fail-safe (which to my knowledge will always be the case), you'd better be prepared for a spirited discussion with the IRS (on audit) as to why you didn't follow the plan language. Here's the thing: The IRS will entertain reasonable modifications to preapproved language on a Form 5307, so you might wish to give that a shot. That's their position, they say to providers to just tell the advisor to "send that situation in on a Form 5307." If the IRS consistently issues favorable determination letters on a certain design, then they will be more likely to want to approve that design in the next preapproved plan cycle because they have another competing policy, namely, to reduce the number of determination letter applications. And then, if approved, send that approved language that you devised to your document provider so it can be (maybe) included in that provider's next draft specimen plan. That's the best way to get the ball rolling. (Recall when "new comparability" was too radical for preapproved DC plans.)

Does the document itself contain a complete definition of compensation to use when 3401(a) is selected on the AA? In particular, does the document include the following underlined language for the definition of 3401(a) compensation? (The following language is derived from IRS model language for the definition of compensation when the 3401(a) definition is selected.) Compensation means "...wages within the meaning of section 3401(a). However, any rules that limit the remuneration included in wages based on the nature or location of the employment or the services performed are disregarded for this purpose." The underlined language is required in the case of determining 415 compensation, so I suspect that it is also used by the document for contribution purposes as well. If it does contain the underlined language, then why is the participant claiming to be exempt from withholding? If it is because of the nature or location of the employment or service performed (such as, for example, the exception for agricultural labor in section 3401(a)(2)) , then you would disregard the exemption from withholding and treat the entire reported amount as 3401(a) compensation, i.e., even if no federal tax withholding is required for that particular wage earner. In other words, it would be plan compensation. If the document doesn't define 3401(a) compensation (for contribution purposes) as described above, then you may need to follow up with the document provider as to whether the amount in Box 14 constitutes the definition of 3401(a) compensation under that particular document for purposes of contributions (as opposed to 415 purposes). For 415 purposes, the document should also state that you need to include amounts that would be included in 3401(a) wages but for an election under IRC section 125(a), 132(f)(4), 402(e)(3), 402(h)(1)(B), 402(k), or 457(b)), and many plan documents will therefore automatically include such amounts within the meaning of 3401(a) compensation for contribution purposes (subject to an AA option to exclude such amounts). Thus, the entire definition of 3401(a) compensation in the document needs to be examined to determine what amount to use for a particular participant for a plan that "uses 3401(a) compensation."

A drafter can specify different eligibility requirements for profit sharing contributions than for deferrals. That is because there is no regulation stating anyone eligible to defer must receive the profit sharing contribution. Contrast that to the existence of the requirement in 1.401(k)-3 that all NHCE "eligible employees" must receive the ADP SH contribution, and the definition of "eligible employee" in 1.401(k)-6 that states that an "eligible employee" is someone eligible to defer. Is the drafter of that preapproved document on notice that there is a GENERAL requirement that eligibility for the safe harbor must be the same as for deferrals UNLESS the rules set forth in the BPD regarding 410(b) disaggregation are followed (with the consequent likely loss of the safe harbor exemption)? If so, I have no objection. Otherwise, I would have made it much harder for the drafter to mistakenly conclude that having different eligibility for safe harbor contributions than for deferrals is no different than having different eligibility conditions for profit sharing contributions than for deferrals. If the employer doesn't know that they have essentially elected to use the disaggregation rules when they elect different eligibility conditions for the safe harbor contributions, they may end up with a serious potential qualification defect because of a regulation that is unique to safe harbor contributions. I will revise my previous post to say: In the absence of 410(b) disaggregation of otherwise excludable employees, eligibility for safe harbor contributions must be the same as for deferrals, at least for NHCEs.

No, you cannot impose a separate age and service requirement for ADP safe harbor contributions (different than for deferrals), you can only use the 410(b) disaggregation rules, which is not the same as "eligibility" (which is why most preapproved plans will refer you to an exclusion (i.e., disaggregation) of otherwise excludable employees, rather than offering separate age and service requirements). The 410(b) component plan that is comprised of those under age 21 and 1 year of service MUST be ADP-tested, and for that reason, no such plan will qualify for the top heavy exemption, since the plan as a whole does not consist entirely of an ADP safe harbor (even if there are no HCEs in the ADP-tested group). If there is an HCE in the ADP-tested disaggregated group (such as an owner who is under age 21&1), and the ADP test for that disaggregated group fails, then you will need to make corrective contributions or corrective distributions for that disaggregated group, even though the employer calls it a "safe harbor" plan. And, if you choose to make safe harbor contributions for those with six months of service, for example, then you will have a group of participants who are receiving ADP safe harbor contributions but who are nonetheless in the ADP-tested 410(b) component plan. In summary, disaggregating otherwise excludable employees is not the same thing as having an "eligibility" requirement of six months, or one year of service, for ADP SH contributions. All of the above would also apply even if only the age requirement was different for ADP SH than for deferrals. See: Revenue Ruling 2004-13 and https://benefitslink.com/boards/index.php?/topic/63746-shnec-top-heavy-and-otherwise-excludable-ees/&tab=comments#comment-290008

The disaggregation of otherwise-excludable employees is not a "restructuring" provision, it is a "disaggregation" provision. Restructuring is prohibited by Regulation 1.401(k)-1(b)(4)(iv)((B). Disaggregation is permitted by 1.401(k)-1(b)(4)(iv)(A). Regulation 1.401(k)-3 requires that the safe harbor contribution be made for all NHCE "eligible employees," which is defined by 1.401(k)-6 as being all (NHCE) participants "eligible to defer." If you disaggregate otherwise excludable employees, then the plan does not consist solely of deferrals and ADP and ACP safe harbor contributions, i.e., the plan will also consist of ADP-tested contributions (for the disaggregated otherwise excludable employees), and the exemption described by IRC 416 fails. In other words, you have NHCE participants who are eligible to defer who do not fall within the ADP safe harbor, thus causing the top heavy exemptions to fail. That also means you could end up with an owner in the ADP-tested group (the owner's spouse is hired midyear), and need to make corrective contributions or corrective distributions to one or more HCEs for a failed ADP test for your so-called "safe harbor" plan. Code Section 416 does not recognize 410(b) components plans. Only 401(k) does. 1.401(k)-3(h)(3) Early Participation Rules. "Section 401(k)(3)(F) and §1.401(k)-2(a)(1)(iii)(A), which provide an alternative nondiscrimination rule for certain plans that provide for early participation, do not apply for purposes of section 401(k)(12), section 401(k)(13), and this section. Thus, a plan is not treated as satisfying this section with respect to the eligible employees who have not completed the minimum age and service requirements of section 410(a)(1)(A) unless the plan satisfies the requirements of this section with respect to such eligible employees. However, a plan is permitted to apply the rules of section 410(b)(4)(B) to treat the plan as two separate plans for purposes of section 410(b) and apply the safe harbor requirements of this section to one plan and apply the requirements of Section 1.401(k)-2 to the other plan. See Section 1.401(k)-1(b)(4)(vi), Example 2."

Speaking for myself, I forgot to assume it was after March 15.

Belgrath, Although I am not a testing expert, as a participant, I would expect to receive the earnings (or losses) over the same period as every other participant who is receiving the same contribution under the same formula and who is making the same investment, i.e., this does not strike me as something that "cannot be expected to be of meaningful value." In addition, the right to a particular investment is a BRF, and while this question does not deal with specific investments, why should John get to invest in ABC Mutual Fund a month sooner that Jane, especially if there is a dramatic upswing or downswing in the market during that period? If one service provider is slower than the other, then the employer, it seems to me, should slow down the fast provider and make the deposit at the same time for everyone. Suppose Joe gets to get his deferrals deposited by ABC payroll company one month sooner than Judy gets her deferrals deposited by XYZ payroll company, even though they work for Wally's Widgets? Might the DOL conclude that the employer should be using only a single transmittal company, i.e., the faster transmitter, so as to separate deferrals from employer assets as soon as administratively feasible? Would not the same reasoning apply here? I acknowledge that deferrals are not the same as a match, but as a fiduciary, I'd be looking to move to uniform deposit timing just as fast as I could, solely out of an abundance of caution and best practice. (But, I suspect I might be preaching to the choir.) In addition, assuming the BRF does apply, are those sub-populations stable from year to year? And if not, what's the correction if the test unexpectedly fails? Whatever the fix is, I bet those calculations will be more expensive, when added to the "damages," than making uniformly timed deposits for all participants, at least if this is a big company with lots of participants making lots of different investments. You solicited thoughts, not an opinion, and such are my thoughts. In fact, I have no opinion, because I am in territory I normally decline to occupy, at least in this forum.

I am not familiar with that document, but now that you have provided additional context, I agree that an employer with such a document could rely on such a provision to avoid funding the match for HCEs. (I also agree with dmwe that rather than engaging in analysis paralysis, just fund the contribution and make corrections!)

I will revise my earlier post to agree that if the plan document can be reasonably interpreted as not conferring a protected benefit on HCEs under a uniform allocation formula, as when there is something in the document (AA) at least suggesting that the amount of discretionary matching contribution to be contributed for one or more HCEs can be a separate discretionary amount than the discretionary amount to be allocated to NHCEs, then I acquiesce to that reasonable interpretation. My intention had been only to suggest that such liberty does not automatically flow solely from the presence of the word "discretionary" immediately prior to the word "matching," and that further inquiry is necessary to determine if that discretion extends to things like the allocation formula and the allocation conditions. If no discretion is to be found in those latter contexts, there exists the possibility that HCEs have a protected benefit. I have heard of employers taking liberties that contradict plan language on the theory that a "discretionary match" means I can do whatever I want because I am the boss and it says "discretionary match." Under that theory, I could even exclude NHCEs, perhaps just those that live in a certain zip code. I speculate that the Ft. Wm. document excerpt that refers only to "excess contributions" in the second enumerated item might deliberately exclude any mention of "excess aggregate contributions." I would be reluctant to rely on that language in the context of ACP, but only because I'd rather be safe than sorry. In contrast, I would be delighted if a preapproved AA were to have an option providing that HCEs were not required to receive any "discretionary match" (even when the NHCEs receive that match), but that the employer could nevertheless choose to make a separate discretionary match for any HCE in an amount that would not exceed any limit on that HCE imposed by any applicable nondiscrimination test. I know that many plans have specific language saying that you can avoid a 415 failure by NOT contributing or allocating the full amount required by a contribution or allocation formula (but, if you do actually do allocate an excess annual addition, then you must correct via EPCRS). Some plans have interesting language with regard to ADP SH contributions for HCEs, i.e., HCEs are technically excluded from the arrangement, but the employer has the discretion to operationally add them back (not to exceed the amount that would be contributed if they were a NHCE). It appears to me that such language is structured so as to avoid any anti-cutback issue. [I don't know why my font size changes midstream when I cut and paste from my clipboard (and please don't clutter this thread with a response on that issue). I'll figure it out at another time. I've been trying to make all of this response have the same bigger font size as it appears in the first two paragraphs.]