John Feldt ERPA CPC QPA

If they find a qualification defect, they can only go baack 3 years for purposes of determining the maximum penalty and sanction - calculated based on the numbers from those last 3 years. But to determine if the plan has an error, there is no limitation. They could ask for proof that the plan was originally qualified when it was first established, or ask for a copy of the plan's original opinion letter or determination letter.

You ask after they accept the job.

Suppose you restructure a safe harbor profit sharing plan for testing purposes. Also suppose the plan excludes HCEs from receiving safe harbor contributions (to minimize the required contribution in a bad year). The owner and most other HCEs are in component plan A: the benefits-tested group with enough of the youngest NHCEs to pass the ratio percent test. The nonelective portion is tested for nondiscrimination by cross-testing and by imputing disparity on the regular PS nonelective allocations (can't impute with the SH amounts). The owner's son however, is in the component plan B: the contributions-tested group with all the rest of the NHCEs where the intnet is that they all receive a total nonelective allocation of 5% of pay (5% PS for the son and 3% SH plus 2% PS for the others). The owner's son's wages are not over the taxable wage base. If component plan A imputes disparity, then imputing disparity in component plan B results in a 10% allocation rate for the son and 7% for the NHCEs (fail). Assuming there is no other integrated plan sponsored by the employer and assuming the son is not a self-employed individual, when imputing disparity, must the entire plan impute disparity, or could just one component plan impute disparity and not the other? How do you read 1.401(a)(4)-7(d)(2) for this?

If the accrued-to-date testing option is used, the "account balance" is tested: 1.401(a)(4)-8(b)(2). This worked very for this one plan since the plan was not cross-tested for the last 15 years (until last year). The target HCE is only 4 years older than the plan's one NHCE. This year, the spouse shows up on the census with low compendation and they want to know if the spouse should defer. I read 1.401(a)(4)-8(b)(2) to be the entire balance, not just the employer-provided nonelective. A high spouse deferral would not test well. Is it possible that the entire balance is used only for the average benefits testing and that just the employer-provided nonelective balance is used for the rate group testing if tested using the accrued-to-date method? I don't see it that way, but wanted to throw up the question. We'll test another way, suggest low spouse deferrals, or imply that higher spouse wages might help out. Any other ideas? Edit: After reviewing, I am instead convinced that the deferral and match balances are not part of the rate group testing when testing the nonelective. My misunderstanding there.

We keep a copy of our marriage certificate, birth certificates, and our kids birth certificates in a safe deposit box. I would hope that the local court records where you were married would easily be able to provide a copy of your certificate. For example: http://www.co.dallas.ia.us/index.aspx?page=729

If no "certificate" of marriage was available, we had fairly short list of acceptable alternative documentation, such as some type of court record, copy of court proceedings, or some other legal documentation that sufficiently proved marital status. When in doubt, we would have the employer make the final decision of whether or not thr proof was sufficient. I do not recall seeing an alternative actually in use for proof of marriage, normally that was for proof of date of birth.

When providing service to some larger plan soem years ago (10,000+ participant plans), we required proof of date of birth, proof of marriage, proof of date death, but the level of service being provided would be considered as outsourcing and is generally not the type of service that a typical service provider offers (a TPA). I think that with the small plans (under 100 lives), the employer typically has the burden of obtaining proof of date of birth, marriage and death.

To avoid estate taxes, which I believe was the case that got this to the supreme court, could a mother/daughter be considered "married"?

There have been cases where a "marriage" was disputed after a separation of the husband-wife when no actual divorce occurred, and then long after the separation (years, months, days, or hours after, who knows) one of the two gets married again while still legally married to the first spouse. The dispute here being over whether or not the two were still considered as married and/or whether or not the second "marriage" has any beneficiary rights as a "spouse" to and of the benefits in the plan. Of course this comes up when a lot of moola is at stake after the death of person who was double-married, and both "spouses" make a claim for the benefits. "Double-married?" . . . hmmm maybe that term will also be allowed to be considered as "married" now that the one man and one woman terminology has gone away?

But i think the total employee count can matter here. As my memory serves, I think the rule is not to cover "only", but to cover "primarily" - isn't it? So I think you can include some employees who are neither a management employee nor a highly compensated employee, as long as the plan "primarily" covers a select group of management or highly compensated employees. So, what is "primarily" then? Courts may differ here. Is it all management and HCEs plus a few other employees? Maybe, but as soon as it looks like the covered group is consisting of more non-HCE non-management employees, then the "primarily" requirement is certainly not met.

Can participants make pre-tax deferrals if the plan is not tax-qualified?

So, to get that exemption then, it must be for late deferral deposits, the excise tax must be contributed to the plan, the 5330 must go to the DOL with the VFCP application, and the excise tax must be under $100.

Same here. These code responses are very insignificant for 5500 filing purposes. However, unlike the voluntary 401(k) questionnaire where plan sponsors could simply mark any answer they wanted (just to finish the request without getting audited), the Form 5500 has this statement: Under penalties of perjury and other penalties set forth in the instructions, I declare that I have examined this return/report, including accompanying schedules, statements and attachments, as well as the electronic version of this return/report if it is being filed electronically, and to the best of my knowledge and belief, it is true, correct and complete. If, upon examination/audit, the government agent begins to think that the Form 5500 questions were just filled in with something, anything, just to get the filing done, then you've likely got more trouble, not just a short closing comment made by the auditor to end the exam. So, it is best to get the best/correct answer in all questions at all times.

You may want to confirm just how discretionary that match is (as defined under the terms of the document). I have seen a document where the time-period was allowed to be a discretionary annual determination made by the employer. edit: wording

I view a "plan characteristic" as "plan" + "feature". The "plan" is the written instrument that the sponsor of the plan adopts. The "features" of the plan include anything that would be available under the terms of the plan. Thus, if the document allows a discretionary match and a discretionary PS, but has never used those features, I would still enter codes 2E and 2K. Similarly, if the document allows only deferrals (does not allow a discretionary PS or match). I would not use either codes 2E or 2K. edit: typo

The participant elected to have their DB benefit paid as an annuity. The plan document has a formula that specifically defines the benefit amounts payable under that form, so once the benefit election is complete, the plan pays that benefit as elected from the plan assets. Alternatively, the plan fiduciaries could use plan assets to purchase that annuity from an insurance company, thereby removing that liability from the plan. It may very well cost much, much more to do that than the amount that was being funded into the plan (unless the actuary knew what the participant was going to elect). That will depend on the annuity quotes you get back, the type of annuity selected, and the actuarial equivalence definition used in the plan to determine the amounts payable under that elected form of payment.

Right, so be sure to only use the pre-approved checkboxes if using a prototype and only use pre-approved text (and/or checkboxes) in a vol sub. and stay within the parameters outlined for any "other" open text fields.

I agree with Sully regarding Section 6.05(1) of EPCRS Rev Proc 2013-12: "a determination letter application is not required ... if the correction by plan amendment is achieved through ... the adoption of a prototype of volume submitter plan with an opinion or advisory letter..." For SCP purposes, 6.05(2)(b) starts "Except as provided in 6.05(1) ..." Does that mean that the adoption of a plan amendment to fix the issue as described under SCP is therefore not subject to D letter submission if you were able amend by using a pre-approved document? Going outside a prototype or a vol sub you must submit. Here's an small excerpt from that SunGard Relius technical update. Full text is here: http://www.relius.net/News/TechnicalUpdates.aspx?ID=964 Must a plan that adopts retroactive corrective amendment under SCP file for a determination letter for the plan? EPCRS gives a general rule requiring a determination letter submission: Except as provided in §6.05(1), in the case of any correction of an operational failure through plan amendment under SCP that is permitted under §4.05(2) …, a plan sponsor must submit a determination letter application for the plan, including the corrective plan amendment, during the plan’s next on-cycle year, or if earlier, in connection with the plan’s termination. §6.05(1) has exceptions to this general rules: Notwithstanding any other part of this §6.05, a determination letter application is not required and may not be submitted with the VCP submission if: the correction by plan amendment is achieved through the adoption of an amendment that is designated as a model amendment by the Service or the adoption of a prototype or volume submitter plan with an opinion or advisory letter as provided in Rev. Proc. 2012-6, 2012-1 I.R.B. 197, on which the Plan Sponsor has reliance (or is treated as having reliance …), or the failure corrected is a demographic failure. Therefore, if a plan sponsor incorporates its retroactive corrective amendment within its approved prototype or volume submitter document, the plan sponsor not only does not need to submit a determination letter application with respect to the corrective amendment, the IRS prohibits such an application. However, if the plan is on an individually designed plan document, the plan sponsor will need to include the corrective amendment as part of its next determination letter application.

Is the match a safe harbor match?

That's good that you limited the amount to $51,000. I have seen more than one owner-only prospect bring an out-of-date MP plan to us (not restated for EGTRRA, GUST, etc.) where a buddy of theirs had been only applying the compensation limit and having them contribute 25% of the 401(a)(17) limit into the plan each year. In one case the owner was about to put in $62,500 for 2012 but they had just been selected for an IRS audit. Since they had not done anything to their plan document since GUST or maybe before that, they got scared and started calling around. Since we would not help them backdate any old documents (isn't that something like tax fraud?), they found an "ERISA Counsel" somewhere who "found" up-to-date documents and, by golly, the execution dates on the document were no problem. We assume they still had to deal with and somehow use up the accumulated excess unallocated amounts (because of the excess contributions).

Assuming the plans are required to be aggregated for coverage and for 401(a)(4), assuming a gateway minimum is the best option for passing 401(a)(4), assuming the TH minimum is provided in the DC plan, and assuming the highest HCE aggregate rate exceeds 35% of compensation: 1) Yes, by what you've described, it seems that a total 7.5% allocation is required for this participant. As to mattmc82's comment, by virtue of the plans being top heavy, wouldn't the TH minimum apply even to a separate 401(k) plan with just 1 YOS, which then triggers the TH minimum and the gateway contribution? What did I miss? 2) If the amendment was not clear that the change in vesting only applies to participants with one hour of service on/after the effective date of the amendment, then that terminee could be 100% vested. However, be sure to go back into the plan document itself to the vesting section and/or to the amendments section to see if it already has language saying something like "in the event this section or the vesting section is ever amended, the new provisions only apply to participants with one hour of service on/after the effective date of the amendment unless the amendment states otherwise." Let us know what you find.

I think this might fall under 411(d)(6) protection under the optional forms section (just a starting point to look into further). §1.411(d)-4. Q-1: A-1: (3)(b) Optional forms of benefit (1) In general. The term optional form of benefit has the same meaning as in §1.411(d)-3(g)(6)(ii). Under this definition, different optional forms of benefit exist if a distribution alternative is not payable on substantially the same terms as another distribution alternative. Thus, for example, different optional forms of benefit may result from differences in terms relating to the payment schedule, timing, commencement, medium of distribution (e.g., in cash or in kind), election rights, differences in eligibility requirements, or the portion of the benefit to which the distribution alternative applies.

I am of the same opinion, but after a very brief limited search for a code/reg/otherwise that confirms this, I took the Benefitslink shortcut to ask. I will search further next week.

Under §1.411(d)-3(g)(6)(i), the term early retirement benefit means the right, under the terms of a plan, to commence distribution of a retirement-type benefit at a particular date after severance from employment with the employer and before normal retirement age (emphasis added). If a PS plan has an in-service distribution option at age 55, can this be amended to age 59.5 for all existing balances without violating 411(d)(6)?

Where the document does not specify, you must still comply with applicable law. The regulations interpret the law, and I think the regulations require the actuarial increase if the notice is not provided, but you may wish to seek a legal opinion in the matter.