C. B. Zeller

Strictly speaking, a retroactive amendment to cure a 401(a)(26) failure is a 1.401(a)(26)-7(c) amendment - however most of the same rules apply as under 1.401(a)(4)-11(g). So no - you can't do a -7(c) amendment to increase only an HCE, because the amendment has to be nondiscriminatory on its own, under the -11(g) rules. If there are no NHCEs that could be added to the plan (including possibly someone who has not yet satisfied age and service requirements), then your options are: 1. VCP (if they even approve it) 2. Retro amendment that increases the one HCE, plus enough NHCEs to satisfy coverage and nondiscrimination on its own All of this is assuming of course that the plan says a 401(a)(26) failure will be corrected by amendment. If the plan document has a fail-safe in place, then you have to apply the fail-safe.

No. There is a requirement under Title I of ERISA to file the 5500, but that only applies to plans which are subject to Title I. Plans which cover only the 100% owner of a business, or partners in a partnership, are exempt from Title I. The reason the 5500-EZ exists is for those plans to provide a report to the IRS, since the DOL reporting requirements don't apply. For the most part, a plan which can file 5500-EZ is the same thing as a plan which is exempt from Title I. That line has become a little blurred recently, but there is coordination between the IRS and the DOL here. A plan administrator is only required to file either the 5500-EZ or the 5500(-SF) for any given year.

Then what does "because we put the participant account" mean, and how is it relevant to whether you file SF or EZ?

2019 was the last year you could file a one-participant plan on SF. Starting in 2020 you could file the EZ electronically. If there is a non-owner participant in the plan, even if they are not active, then you can not file 5500-EZ.

I did not mean to suggest that March 15, which would usually occur in the 11th week of a calendar year, would be within the "first few weeks" of the year. My reference to the rule was in response to Nate S's comment. Is anyone aware of any situations in which the IRS, either in a ruling or on audit, found that compensation included under this rule was paid too late in the following year to count as the "first few weeks?" I think so. Paragraph (d) provides safe harbors for items included in compensation, and paragraph (e) provides timing rules. I read them as being taken in concert with each other.

Possibly the "first few weeks" rule of 1.415(c)-2(e)(2)?

This means amounts that could be distributed from the plan. For example, 401(k) deferrals if the participant is over age 59½. There are other distributable events for 401(k) deferrals, of course, but those would not generally be of much use in an IRR context. Terminated employees can't usually make rollover contributions, and hardship distributions aren't eligible rollover distributions, for a couple of examples. This is anything that couldn't be distributed from the plan, for example 401(k) deferrals or safe harbor contributions under age 59½ while still employed. If a plan allows a Roth conversion of amounts that are not otherwise distributable, then it has to retain the distribution restrictions that applied to the amounts prior to the conversion. That means, in most cases, the plan will have to track twice the number of sources that they had in the plan before. For example, now they have 401(k), 401(k) Roth conversion, safe harbor, safe harbor Roth conversion, profit sharing, profit sharing Roth conversion, etc. That might be the reason why a particular platform isn't supporting this type of conversion.

As long as the formula is written into the plan doc, then I agree you are good on the ADP and ACP safe harbor.

And you've determined that the spousal non-involvement exception does not apply? 414(b) says: The IRS has produced no such regulations. So, that leaves the allocation of the deduction to some reasonable interpretation of the taxpayer. Anything that's not abusive is probably ok, but I would encourage the sponsor to get advice from their accountant, if not an attorney, before doing anything questionable.

This will NOT meet the requirements for a discretionary ACP safe harbor match. It should be ok for a fixed match however.

Are A and B a controlled group?

The difficulty, of course, is that these elections are not being processed by fifth-graders - they are being processed by computers, which are only as smart as they are programmed to be. Computers generally process numbers as either integers or "floating point" numbers - essentially decimals. While it is certainly possible to program a computer to work with fractions, it is more complex than using integers or floating points, and the person developing the system (or more accurately, the person paying for the development of the system) may feel that benefit of the added precision gained by supporting fractional numbers is not worth it in terms of development and support costs. If we continue this line of reasoning, why stop at fractions? What if I want to designate 1/sqrt(2) of my account to my first contingent beneficiary, and 1-1/sqrt(2) to the second? This amount could be readily calculated, but it would not be reasonable of me to expect the software to support designations made in terms of irrational numbers. Instead I should be prepared to accept an approximation. An approximation that is accurate only to one part in a hundred (whole percentages) is not great, however the information provided in this thread seems to indicate that is uncommon among providers. More common seems to be precision to one part in a thousand (percentage with one decimal place) or one part in ten thousand (percentage with two decimal places), which while not perfect, is pretty good. For an account with a value in the millions, percentage with two decimal places means that the amounts will be accurate to within the hundreds of dollars.

Need an EIN. Do not use a SSN. From the instructions to 5500-EZ

Or better yet, if the tool were open source and its code could be examined and vetted by a range of industry professionals.

If the TPA works mostly (or exclusively) with plans on daily investment platforms, they may have little to no need to deal with 1099-Rs, as those would be handled by the platform. The TPA might be further insulated from the process if they utilize a service like Penchecks for distributions from non-platform plans.

I don't have any experience with this exact situation, but I wonder if you could accomplish it using an automatic contribution arrangement? Obviously it wouldn't meet the requirements for an EACA or QACA but that wouldn't be the goal anyway. The ACA would only apply to the employees acquired as part of this transaction, and the default contribution rate would be equal to the employee's last contribution rate in effect in the acquired company's plan. It might be an issue if anybody in the prior plan had a dollar amount election instead of a percentage of pay election, since I believe an ACA has to use a percentage of pay and not a dollar amount as the default. You might also have to generate customized notices for each employee to specify their default contribution rate. I think you would also be forced to put them in a QDIA, if that wasn't already the plan.

Same to you - enjoy your weekend! And don't forget, happy 48th birthday to ERISA!

Yes, but moreso that H is participating in the management of W's business. I think the fact that they have to have an accountant determine whether it is even reasonable for H's business to pay W's businesses is an indication that the two businesses may not be actually be managed independently. If the accountant determines it is not reasonable, then W's business would be providing services for free, which most rational business owners would not do, in the absence of some outside factors.

Best source I am aware of is the EFAST2 FAQ 33a: (https://www.dol.gov/sites/dolgov/files/EBSA/about-ebsa/our-activities/resource-center/faqs/efast2-form-5500-processing.pdf) I don't know if you are focusing solely on signatures required of the plan administrator but, the enrolled actuary's signature on the Schedule MB/SB must be manual, from the Form 5500 instructions: Now I have a question for you - let's say I load a PDF of a document onto my tablet, which has support for a digital pen, and I sign that document using the digital pen on the tablet. If the document requires a "manual" or "handwritten" signature, but states no requirement for ink-on-paper, does my pen-on-tablet signature satisfy that requirement?

Interesting, I will admit I have not scrutinized that section. Still, I wouldn't expect the DOL to have copies of returns that were filed solely with the IRS, for example a paper 5500-EZ that was mailed to Ogden. Is anyone familiar with the process to request a copy of a 5500-EZ from the IRS?

5500-EZ filers are exempt from the requirements of Title I, including the SAR and the public disclosure.

I agree that an individual does not have to be an employee in order to have ownership attributed to them or to have their ownership attributed to another person.

What was involved in checking? If you mean you looked them up on the DOL's website, you won't find them there, because it does not show 5500-EZs. If you believe that the Form 5500-EZ was not filed for one or more years for which it was required, then file the missing forms as soon as possible under the IRS relief program. https://www.irs.gov/retirement-plans/penalty-relief-program-for-form-5500-ez-late-filers

Your numbers appear to be off - 47% + 47% + 16% = 110%. Attribution under sec. 318, used for determining who is a 5% owner for HCE, Key employee, and RMD purposes, among other things, goes from grandchild to grandparent but not from grandparent to grandchild. In your example (notwithstanding the inconsistent total), both Grandma and Mom would be deemed 94% owners (since they are each attributed the other's ownership) and grandkid would be deemed 47% owner (attributed Mom's ownership only).

I think there is a good chance that this activity would invalidate the spousal non-involvement exception, and there would be a controlled group.