MWeddell

It's unlikely that the plan document allows the plan administrator to place different contribution limits on Division A HCEs versus Division B HCEs. To operate the plan in accordance with the plan document, you most likely need to amend the plan document.

If the plan allows loans, then the union participant must be allowed to take a loan or else you will have violated the prohibited transaction exemption that allows loans.

Code Section 415(e) used to include limits that affected both DB and DC plans, but it was repealed about 14 years ago. So, yes, the 415 rules are separate for the two plans you have. Whether the plans are aggregated for nondiscrimination testing purposes depends. They may be aggregated or may be tested separately.

This doesn't affect your main questions, but option 1 is not permitted. Plan A has a 1/1/2014 - 12/31/2014 plan year and nothing in the regulations permits you to break that up into two 6-month periods for ADP / ACP testing. If you go with option 3, then you might argue that BRF testing is performed with a 12/31/2014 snapshot date, in which case there are no BRFs that fail to be available to all plan participants. To do this, you would have to argue that 12/31/2014 is a representative date even though for half of the plan year, the BRFs were quite different. That's an aggressive enough position that I would warn the client to review it with legal counsel. You also could go with option 2, not aggregating Plan B with Plan A. That would more clearly eliminate the BRF testing issues and you likely could use the Code Section 410(b)(6)© merger & acquisition transition period to avoid any coverage testing difficulties.

First reaction is don't do business with any recordkeeper who is offering to backdate a document to solve a compliance glitch.

I was wrong. What the IRS says doesn't really matter, but the DOL has said that terminating a 403(b) plan does not jeopardize the non-ERISA status of the plan. See the last paragraph in the Analysis section of Field Assistance Bulletin 2007-2, located at http://www.dol.gov/ebsa/regs/fab2007-2.html. Maybe someone else wants to take a crack at the questions in the original post and get this thread back on track. I'll let someone else go first: I think my credibility may be shot for this thread!

The issue isn't whether the IRS lets the employer terminate a 403(b) plan. What I'm questioning is whether the employer who is not acting as a plan sponsor (which is why the 403(b) program is not a plan subject to ERISA) can terminate the 403(b) arrangement. I don't think the employer can do so, although confirmation from others on this point would be appreciated. Note that the 403(b) plan in Rev. Ruling 2011-7 was non-ERISA because it was a governmental plan, not because the employer was not acting as the plan sponsor. It is a different set of facts from your situation.

Why is the 403(b) plan not subject to ERISA? Is it a church plan? Alternatively, is it a non-ERISA plan because it only accepts employee pre-tax deferrals and the employer has a limited role? If it is the latter, then it is not sponsored by the employer. The employer may cease letting employees defer to it, but the employer does not have the power to terminate a plan that is not sponsored by the employer.

I used RIA Checkpoint to search for a ruling containing "11(g)" and couldn't find it. It seems to me likely that it wasn't published.

One cannot remove the annuity forms of payment from monies (and investment earnings on them) that came from a money purchase pension plan. The IRS has issued guidance addressing this very point.

The EPCRS doesn't prohibit you from using methods specified in the regulations at getting the ADP tests to pass. It only is relevant once there is an operational failure and you have exceeded the 12-month after the end of the plan year deadline in the 401(k) regulations for correcting a failed ADP test.

I don't usually go out of my way to advertise my own company's services, but if you're going to ask me a direct question, how can I refuse? I rejoined Towers Watson in January 2014, as you seem to know. We have been talking to clients about an analytical service called FiT Age, which calculates for each participant the age at which the participant is projected to be able to afford to retirement while maintaining his/her pre-retirement standard of living through his/her remaining life expectancy. By expressing the result as an age, not as a big lump sum amount of needed retirement savings, we think it is more tangible to employees and therefore more likely to trigger them to take action. A bigger difference is that it integrates wellness behavior, retirement savings behavior, HSA savings, retirement benefits, and participant demographics at the individual level. (Makes me wonder whether employer stock purchase plans will be incorporated at some point too.) FiT Age is really two services. First, an analytical service for employers that can be used to shape plan design and communications initiatives. Second, a participant-level communication that communicates FiT Age and then gives customized suggestions for how to improve one's FitAge. The exciting thing from participant's viewpoint is that often the most effective suggestion is not "save more and have smaller take-home paychecks" but instead can be save more efficiently.

Besides looking at the beneficiary designation portion of the plan, also look back in the miscellaneous provisions in the document. It is somewhat common to see a provision stating that if two persons die within 72 number of hours of each other, they are deemed to have died simultaneously.

So you would approve the owners request in my example? I would consider the issue ambiguous enough to refer back to the official plan administrator, even if the client was outsourcing administration of the hardship withdrawals to me based on the client's written policy. That being said, I would tell that the client that I would lean toward approving the hardship withdrawal request. This assumes the resources portion of the test was satisfied -- some of the comments above by others question that.

I would be less worried in your hypothetical than if the hardship withdrawal request were denied. If you are unsure, then advise the client now to consult with legal counsel and make clear the degree of uncertainty in your recommendation, not afterward.

I saw at a 2013 summer Vanguard conference statistics that surprised me regarding what a low percentage of participants were accessing Vanguard's website via smartphones and tablets. In other words, the speed of participants adopting those technologies specifically for retirement plans was slower than I would have thought. I'd guess that one could contact Vanguard to get their most updated statistics on that. I think the most important change is going to be individualized guidance / advice / account management. I know we think that's already happened, but there is more to come with the integration of HSA savings opportunities & individual tax consequences into the more common 401(k) material.

I would lean toward granting the request. In other contexts, e.g. the sale or purchase of a prinicipal residence, the tax code would include the swimming pool as part of the principal residence. Where the phrase is not defined in the hardship withdrawal regulations but the phrase "principal residence" does seem like a technical term and not just ordinary language, it seems most reasonable to me to like elsewhere in the Internal Revenue Code for what that phrase means. Note that this is a safe harbor event, an event that is deemed to satisfy the hardship distribution requirements. Hence, it does not really have to be an immediate and heavy financial need based on the facts and circumstances. For this reason, I find J Simmons' post to be unpersuasive: yes a pool is a luxury but that's not relevant in my opinion.

There have been plenty of articles over the years regarding making contributions to 401(k) plans from unused vacation pay. There are two methods: Employees choose whether to make the contributions or to receive them in cash. This is just another cash or deferred election, so the dollars are subject to 402(g) limits and ADP testing. Obviously look at the plan compensation definition in your plan document. Employers make a nonelective nonmatching contribution of unused vacation pay. Subject to 401(a)(4) general testing if any HCEs receive these dollars and one should question how likely it is to pass that test.

If the additional match provision already is in the plan document, then electing to use it doesn't seem too aggressive. But amending the plan to add a second match formula with its own allocation conditions seems like one is rather directly circumventing the prohibition against retroactively changing a 411(d)(6) protected benefit.

Yes, it is permissible. These are discretionary amendments, so the deadline is the last day of the plan year in which the amendments are effective. You'd have to made sure that the entry date and normal retirement age change were not retroactive cutbacks. There's been ample discussion on these boards about mid-year amendments for 401(k) safe harbor plans, but it sounds like that doesn't apply in your case.

Once the first pay period of 2014 has ended, then it is too late to change the conditions of how the matching contribution is allocated. Treas. Reg. 1.411(d)-4, Q&A-1(d)(8) fairly clearly provides that the allocation conditions are a protected benefit that cannot be changed retroactive to the beginning of the plan year. That said, one might be able to amend the document to add another discretionary employer contribution instead of changing the allocation conditions for the existing match. Seems awfully aggressive too -- tell the client to check with legal counsel -- but I can't say for sure that it doesn't work.

Roth contributions are after-tax, so they already are excluded from various safe harbor definitions of compensation.

What you'll need to do is to disaggregate those employees with < 1 year of service or < age 21 from the others. The otherwise excludable employees are the three part-timers, which is deemed to pass coverage because there are no HCEs in the group. For those with 1 year of service + age 21 at least, (2 NHCEs benefitting / 2 NHCEs total) / (1 HCE benefitting / 1 HCE total) = 100%, which is greater than or equal to 70%, so it satisfies coverage. If you didn't disaggregate, you'd have a ratio percentage of only 40%, forcing you to run the average benefit test.

For eligibility purposes, one can't exclude the prior service just because a qualified plan wasn't in place. Let us know whether the full-time HCE and two full-time NHCEs have a year of service and age 21. Sounds like the three part-time NHCEs have never earned a year of service.

You may run one coverage test for all employees because some of them are eligible on their hire dates. Alternatively, you may disaggregate the plan into those with < age 21 and < 1 YOS (optionally including an entry date assumption) and those above that threshold.