Belgarath

But there's a change in the Medicare surtax, right? Without looking it up, I seem to recall it was something like 2.9% on earned income up to 200,000, and 3.8% on earned income over that? GROAN - at just a quick look, the amount of income subject to this extra tax will be dependent upon filing status? So it looks as though we'll have to know this in order to calculate S/E contribution/deductions? I sure hope I'm wrong about this... See IRS form 8959.

I don't know how you EA's keep this garbage straight! How do the Multiple Annuity Starting Date regulations enter into this? It seems to me, from looking at Sal's book, that there isn't even an approved methodology? In essence, it appears that this would have to be taken into account when making any future distribution to him, for 415 purposes - is this separate from your earlier comments or was the MASD stuff implicit in the "taking into account" prior distributions? Retirement keeps looking more attractive. However, I think the virtues of poverty may be overrated, so I think I'll stick it out for another decade or so...

Thanks Tom. Any thoughts on the DOL PT issue? After letting all this percolate overnight, I feel like a pretty reasonable argument can be made for the "B" interpretations, although I'd feel much more comfortable (being of a generally conservative nature on such issues) if there were no restrictions on Roth accounts that are different than on the non-Roth. I do believe that some of these restrictions come from funding companies, rather than the TPA's/employers. I know that at least one of them won't allow participant loans from Roth accounts, and I believe that, at least earlier when Roth accounts were "newer" that some wouldn't allow it due to accounting/basis tracking problems.

I'm seeing a lot of plans that preclude in-service withdrawals or loans from Roth accounts, but allow them from pre-tax deferrals. I'm wondering what y'all think about this. All opinions appreciated! First, the in-service withdrawal is an "optional form of benefit and the loan is a "right or feature." Purely from the standpoint of nondiscrimination testing, do you believe: A. That nondiscrimination testing is required, because of the Roth restriction, or, B. That nondiscrimination testing (in the absence of other restrictions) will pass since the decision to defer on a pre-tax or Roth basis rests solely with the participant, and if loans/withdrawals are important to them, they have an unfettered right to do pre-tax deferrals, and this would therefore satisfy both current and effective availability? (And I recognize that effective availability is subjective at best.) C. Or, some other opinion? Second, with regard to Prohibited Transaction issues with the DOL regulations under 2550.408b, since loans must be available on a "reasonably equivalent" basis, I think the same general thought process applies. Do you believe: A. The lack of availability from Roth accounts makes the plan fail the "reasonably equivalent" test, and thus Prohibited Transactions become an issue, or B. As in B above, due to the fact that participants CHOOSE pre-tax or Roth, that this satisfies the "reasonably equivalent" requirement? C. Or, some other opinion?

I think GMK hit the nail on the head. "It will depend on the make up of your workforce." With some employers, particularly smaller employers that pay relatively well, I've seen pretty much 100% start deferring the extra in this type of situation. Other employers may see little change. And sometimes it depends upon the skill and presentation of the employer and investment people. Sure wish MY employer would institute this kind of a match...

I probably won't even ask the question right, but here goes, using, as you will see, grossly hypothetical numbers just for the sake of illustrating the concept. I'm used to exposing my ignorance when it comes to DB calculations. Suppose you have a sole prop, who had a defined benefit plan that was terminated in, say, 1999. At the time, his income was much highr than now, and he had accrued a benefit that was close to the 415 limit. Let's just say that his 415 limit at the time was 1,000 per month, or $12,000 per year, which equated to a lump sum of $100,000. His accrued benefit was $950 per month, or $11,400 per year, leaving him under the 415 limit by $50 per month, or $600 per year, and his lump sum payout was $95,000. This $95,000 lump sum amount was paid to him and rolled to an IRA. Now fast-forward to 2013. He is still a sole prop, and has to be considered in a DB plan due to controlled group/minimum participation. His salary is now quite low. When calculating his 415 limit under the new plan, I know that the old plan must be taken into account for 415 purposes. If the new plan 415 limit, (assuming he never had a prior plan) based upon age, salary, participation, etc., is, say $150.00 per month, or $1,800 per year, how do you calculate his 415 limit in the new DB plan? Is it only $50 per month, because his "old" 415 limit was higher than what his new limit would be, and there was only $50 per month available under the prior plan? Or is it done in some other manner - for example, since his prior benefit was higher than the current 415 limit based solely upon his current income, is his 415 limit in the new plan zero? Or is something altogether different from either answer? Thanks!

Hi Tom - interesting point. But for simplicity and using the morre common situations, lets say that the SH is either the 3% nonelective or the usual SH match. At least in my experience, this would typically be true. I think it would be very unreasonable to impose sanctions in such a situation. But...

Didn't say it was any different...just that this general line of thinking has been advanced in some quarters to rationalize the IRS stance on the subject of mid-year amendments in safe harbor plans.

I have heard discussion of this with regard to Safe Harbor matching plans, to the effect that, "if the participant knew in advance that loans would be allowed, it might have had an effect on whether, or how much, they deferred. So adding it mid-year isn't permissible." I think there are very few people who don't think the IRS stance on this whole subject is generally unreasonable, but as Tom points out, how much risk do you (or your client) want to accept? Certain amendments I believe are less risky than others. For example, a plan excludes a class of NHC. They decide mid-year that they want to amend the plan to allow them to enter. I have a hard time believing that the IRS would impose sanctions on a plan for this - BUT, who knows?

This is an interesting question, and I haven't attempted to analyze it in great detail, for the simple reason that (so far) all such clients have deducted for the prior year. I don't think the ability to deduct for a prior year is in question, if contributed within the 404(a)(6) period - that's moving away from the original question. The question was, what compensation is used for nondiscrimination testing if the DON'T deduct for the prior year. To take it another step, and solely for purposes of discussion to illustrate the point, suppose the client is not on extension, so the contribution must be made by 4/15 to be allowed as a deduction for 2013 under 404(a)(6). Further suppose it is a fixed formula, REQUIRING an employer contribution for 2013. The client makes the contribution on 4/30. So the client CANNOT deduct it for 2013, and it is not "allowed" under 401©(2)(A)(v) for 2013. But under 415, this contribution is made timely, and is ALLOCATED for 2013. The plan definition of compensation is "earned income" for the plan year (2013) for self-employed individuals. Although I'm not sure the statutory/regulatory/plan language necessarily contemplates such a situation (but it very well may, in something I haven't considered!) I lean toward taking a "reasonable" approach, and to me, if the contribution is allocated for 2013, you would do the earned income reduction to arrive at the appropriate allocation and compensation for testing purposes. However, I could be convinced otherwise by someone who has looked at this in greater depth and can provide a better argument! I'll be interested to see what people think.

IMHO, you would lose reliance. You might be able to get away with it by having it detailed in the SPD, if the plan language is silent or ambiguous on the time limit, but this seems like quite a stretch to me. Is this because of the Heimeshoff v. Hartford Life & Accident Insurance Company decision? I was wondering about this myself, as to whether plans would be amending to put in a time limit, but since DC docs are in final stages of getting IRS approval, I don't see this as an option likley to be offered in the near future for pre-approved documents. Perhaps more for very large plans that are custom drafted anyway, or of course one could always amend the pre-approved plan and file for a d-letter - I don't know how important this item is for most employers.

I'm confused. How is a transfer of surplus assets (and assets in a DB plan are not "allocated" to individual participants) directly to a replacement plan considered as being reportable to a participant or participants on a 1099? I don't see this as being reportable on a 1099 at all. Nothing is being distributed or is distributable to participants.

I've heard of this issue being raised on audit, but I'm not familiar with any notice such as you describe. In the audit situations, I've not heard of a penalty being imposed - they just tell the fiduciary to get bonded. That may change at some point. As an aside, a plan isn't exempt from the bonding requirements just because it is the first year. In that case, it needs to be estimated. See Dol Reg 2580.412-15(b), and FAB 2008-4.

Question - you don't specify if there are any common low employees who will participate? If this is a two-person only corporation, then they will both have to participate to satisfy 401(a)(26). Since most DB plans being set up these days seem to be for very small businesses, I always ask this question.

In a recent conversation with someone fairly high up at the DOL, this question of splitting into two plans came up as a tangent to the question actually being discussed, and they indicated that they would look at this "very hard." Stopped short of saying definitively one way or the other what their position would be. Personally, I wouldn't recommend it, and would tell a client to seek the advice of ERISA counsel before doing it. I'm not saying it isn't legitimate, just potentially a risk.

Thank you for your thoughts. Since the Loan Policy specifies "termination of the Plan" rather than "termination date," what is your thought on the position that a plan isn't considered "terminated" until all assets are distributed? A plan, particularly a DB plan which files for a D-letter, might not distribute assets for a year or two. Would you still maintain that loans are due and payable on the termination date? Thanks again.

A slight twist upon this question. Plan termination date is 12/31/2013. Assets likely won't be distributed until March or April of 2014. In the meantime, the Loan Policy states, "A loan, if not otherwise due and payable, is due and payable on termination of the Plan..." It seems to me that this is ambiguous enough to be interpreted either that loan is due and payable as of 12/31, or payments can and should be made until assets are distributed. Personally, I feel like the second interpretation is more technically correct. Any thoughts either way?

Maybe I'm misunderstanding the proposed loan program, but I'm having a hard time seeing why this is perceived as a potential Nondiscrimination testing problem. The 50% up to $30,000 applies to both HC and NHC. If a HC wants to take a $35,000 loan for a purpose OTHER than a residential loan, he can't do it. The 50% up to 50,000 increased limit for residential loans is available for both HC and NHC. If a NHC has an account balance of only $40,000, the NHC can still take a residential loan for $20,000. If their account balance is $80,000, a $40,000 residential loan is available to them. Same applies to HC. The only way you would have a problem is if residential loans are not allowed for amounts of less than $30,000, but are available for loans in excess of $30,000. THEN you'd have a problem. I don't see any problem here. (although I think it is a ridiculous provision, but that's a personal preference)

Austin - did you ever get an answer on this?

See 2002-21 for some useful information on PEO plans. http://www.irs.gov/pub/irs-irbs/irb02-19.pdf

I think this will help you out. http://www.dol.gov/ebsa/faqs/faq_dfvc.html

Yes, the beneficiary can generally exclde the Taxable Term Costs from the otherwise taxable amount over and above the "net insurance proceeds." So in the example used - of the $40,000 CSV, if $5,000 represented TTC, the taxable distribution would be $35,000.

Is it an employer not subject to ERISA? (public school, perhaps?) If not - that is, if it is an employer otherwise normally subject to ERISA, and there are employer contributions, it is subject to ERISA - you don't have to waste time and effort looking any further for potential traps.

Here's an interesting situation. You have two spouses, each of whom have their own sole proprietorship business. NO other employees. Spouse A who makes scads of money is setting up a DB plan. They are a Controlled Group, or I guess technically a Common Control Group, which has the same effect. Due to the minimum participation rules, spouse B who makes very little money must be included in the plan, since there are only the two of them. We're having a discussion - one seems to be telling us that some of the costs for the benefit of one spouse must be allocated to the other – rather like a 50/50 partnership. I don’t think this is necessarily required – it seems like it should be two entirely separate calculations – one for each sole prop, based solely upon their own Schedule C income, as they are separate entities, even though part of a controlled group. Am I missing something here? Are you aware of anything saying that two such spousal sole props in a controlled group are treated as a partnership for cost/deduction purposes? I know there is “joint and several liability” under IRC 412(b)(2), but in the absence of any regulations, wouldn’t it be permissible/reasonable to simply allocate as per their own calculation based upon their own individual Schedule C’s? Alternatively, is it reasonable for her to pay the entire contribution, as there is joint and several liability?

"I think in an ACA you only have to tell them once you are going to do it. With EACA, you have to give a notice every year. EACA has to be for the entire year, ACA can be implemented at any time. The Sponsor may not want the headache of processing the 90-day distributions." A couple of observations, FWIW. First statement - getting hyper-technical, the notice requirement only applies to "covered employees." So initial notice is required, then annual notice is required to covered employees - and I maintain that applies to an ACA or a EACA. Now, depending upon how the plan is written, anyone there who already made an election may not be a "covered employee" for purposes of the notice requirement. Personally, I think this is a potential administrative and compliance nightmare, and I'd provide an annual notice to everyone, whether technically required or not. Second statement, generally agree, although a EACA may be implemented mid-year if it applies ONLY to new employees after the effective date of the EACA. Third statement - absolutely agree, although the return of contributions is optional under a EACA, so avoiding this doesn't mean you must implement an ACA instead. As an aside, an ACA requires a QDIA, whereas an EACA does not. (I don't know if this was intentional or an oversight) - but personally, I can't quite see why a fiduciary would choose NOT to utilize the QDIA for fiduciary protection,but again, there may be lots I haven't thought of...