g8r

We may get them in another week or so. Last year Rev. Proc. 2003-85 came out on Nov. 19th. http://www.irs.gov/pub/irs-drop/rp-03-85.pdf

DeePA, You are correct. For prototype plans, elective deferrals can only be withdrawn on account of a hardship that satisfies the safe harbor conditions in the 401(k) regs. For individually designed plans, including volume submitter plans, the IRS doesn't "currently" mandate that the plan be drafted to include the safe harbor conditions. There is a proposal to change this when volume submitter plans are updated for EGTRRA, but that's just in a draft procedure and will hopefully be dropped before it is finalized. For unrestricted amounts (regular P/S contributions and matching contributions that aren't QMACs), a facts and circumstances test can be used in both prototypes and volume submitter plans. And, the facts in circumstances is not the same standards found in the 401(k) regulations (1.401(k)-(d)). Those rules apply to facts and circumstance for distributions of elective deferrals. Regular p/s and matching contributions are not subject to that regulation. Instead there was an old revenue ruling that is very vague as to what constitutes a hardship. For that reason, many plans are drafted by applying the 401(k) facts and circumstances provisions to these other amounts. Remember that QNECs, QMACs, Basic or Enhanced ADP safe harbor matching contributions and nonelective safe harbor contributions (i.e., the 3% safe harbor), can't be withdrawn for a hardship.

As a last resort, have him read the standardized document he is (or intends on using). You have to follow the terms of the plan you adopt, regardless of what the IRS allows or doesn't allow in a prototype. That's what reliance is all about. If you're actually drafting a prototype for submission to the IRS, have him put the provision in the plan. He can argue with the IRS reviewer (and maybe you could listen in to get a good laugh).

It's not a problem if you can pass nondiscrimination. Of course trying to figure out the nondiscrimination rules can be a challenge. Personally, I'd just use a 410(b) coverage test approach comparing the HCEs and NHCEs in each separate plan (or each separate "component" plan). No, there isn't any guidance specifically allowing you to use such an approach -- but I don't see how the IRS could ever challenge it.

Just like a 401(k) plan, you can have different eligibility conditions/entry dates for different benefits. There's nothing wrong with having a 90 day wait for health insurance and a 1 year wait for the health FSA (or an entry date that is the first day of the plan year following date of hire). You can design the plan in any manner that is nondiscriminatory and just because you have different conditions for different benefits isn't per se discriminatory. As far as pro-rating the health FSA limit, it can be done (subject to nondiscrimination rules -- which wouldn't be a problem if it's only being pro-rated for mid-year entrants). However, I don't see many plans using pro-ration. I've seen numerous plans with a longer waiting period and annual entry for the health FSA.

I agree - you need to be careful where someone changes classifications (but that could be handled in the drafting of the plan). And, to expand on GBurns comment, you should avoid basing it on age or service. The following is from the nondiscrimination section of reg. 1.105-11 which deals with nondiscrimination as to benefits under a self-funded health plan (which would include a health FSA). While this regulation is arguably old and outdated (and doesn't adequately deal with techniques such as creating separate plans), I'd try to find some classification that is nondiscriminatory and is not based on age and/or service. _____________________ However, any maximum limit attributable to employer contributions must be uniform for all participants and for all dependents of employees who are participants and may not be modified by reason of a participant’s age or years of service.

1950 - Not to beat it to death, but what you are referring to in T.D. 8794 is below. It cites Q&A 2 as authority for being able to increase from $3,500 to $5,000. So, I think it actually reenforces the interpretation of the regulation that you can modify the cash-out rules without violating 411(d)(6). It would be nice if the IRS were to reiterate the rule in their upcoming guidance. But, Ieven if they don't, I feel confident that 1.411(d)-4 allows a modification of the cash-out rules. G. Benefits Protected from Reduction or Elimination Section 411(d)(6) provides, in general, that a plan shall be treated as not satisfying the requirements of section 401(a) if the accrued benefit of a participant is decreased, or an optional form of benefit is eliminated, by an amendment of the plan. Section 1.411(d)-4, paragraph (b)(2)(v) of Q&A-2 provides that a plan may be amended to provide for the involuntary distribution of an employee's benefit to the extent such distribution is permitted under sections 411(a)(11) and 417(e). In accordance with that provision, a plan may be amended for plan years beginning on or after August 6, 1997, to permit the involuntary distribution of an accrued benefit using a cash-out limit of $5,000, with respect to benefits accrued before the amendment was adopted and effective. Such an amendment is permitted even if the plan, prior to amendment, did not permit involuntary distributions (as well as if the plan permitted involuntary distributions if the present value of the participant's benefit did not exceed the prior cash-out limit of $3,500). Such an amendment will not violate the anti-cutback rules of section 411(d)(6).

No, there aren't any regs out on this.

I don't have my markup the law with me. But, you may want to look at the definition of "qualifying dependent." If someone is not a "qualifying child" (e.g., b/c of the age limit), the person might still be a "qualifying dependent" if those requirements are met (and the compensation threshold to be a qualifying dependents does not apply to IRC 105).

I agree. I don't think the cited regulation provides ample room to modify the plan's cash-out rules w/out fear of being locked into a particular position. I have had the same issue regarding the spouse of a deceased participant and it seems to me that the automatic cash-out rule would apply. But, it's less than clear. On the loan, I'll have to look at the guidance. But, it seems that the rule applies when there is a distribution that is eligible to be rolled over and there may be some wiggle room to argue that the loan can't be rolled over. I know it can be rolled to a qualified plan but it can't be rolled over into an IRA and that may be enough to argue that the automatic IRA rollover rule doesn't apply to the loan.

For whatever it's worth, I think the loan is taken into account (the note held by the plan is an asset).

It's interesting that the cited article only refers to 2 institutions that may be willing to accept the rollovers. I know that Penchecks has been accepting these for years, and I had heard (just a rumour) that Fidelity was considering it (but possibly only for their current clients). As far as plan amendments, we don't know yet what the IRS is planning. Absent any guidance, it appears that an amendment would be needed (it would just need to be a good-faith EGTRRA amendment). But, I wouldn't do anything yet - I'd wait to see what they intend on issuing (good-faith language, a 402(f) notice, etc.). If you want to eliminate cash-outs or reduce the threshold to $1,000, I think an amendment would also be needed. Presumably, the standard for that would also be a good-faith amendment.

Agreed. 106 is for premiumes and 105 is for benefits received.

On the BNA material, I agree that this could be considered a copyright infringement. And, it is a form, not language for the SPD, although much of it could be put into the SPD. As far as whether the plan should provide the specific %, I'd opt to put it in the plan to be safe. It's unlikely the % will change very often. From a qualification perspective, it's a matter of how strictly the IRS wants to apply the definitely determinable benefits. That seems to depend on who you ask at the IRS.

Here we go again. Yes, there is a difference in pre-tax vs. non-taxable. But, the end result is the same - which is what the initial question was. How can you provide insurance coverage for employees (here it's a select group) on a tax "preferred" basis. I used preferred so you don't have to get hung up on the technicalities. There are only 2 ways: 1. An employer can "pay" the premium and employees have an exclusion from income under IRC 105 (and this is explained below in more detail - bear with me). 2. An employee can pay the premium (with after-tax dollars) and claim a deduction (subject to the 7 1/2 % threshold). Option 2 is not a good solution - the compensation is subject to payroll taxes (e.g., FICA, FUTA) and you have to meet the 7 1/2% threshold. So, the question is how do we fit under option 1. And, let's put $$ to it. The employer will only pay a total of $100,000 in comp and benefits and the premium is $1,000. The direct method - employer pays the entire premium to the insurance company. Employee gets $99,000 as cash and employer pays $1,000. As pointed out earlier, an employer can pick who it will pay premiums for. Section 89 is long gone so the only issue is state insurance laws. Also, as pointed out, it doesn't have to be a group or employer sponsored policy. Rev. Rul. 61-146 allows the policy to be an individul policy. The "reimbursement" method - Employee pays premium with after-tax dollars. Employer then "pays" or "reimburses" employee for the premium. Thus, the individual is getting additional funds from the employer over and above the regular pay that was already taxed to the employee. But, this additional reimbursement is excludible from gross income - per Rev. Rul. 61-146. Here, the employer pays the employee $99,000, the employee pays taxes and pays the premium (so the premium is paid with over tax $). The employer then pays (reimburses) the employee an additional $1,000 for the premium but this additional $1,000 is not subject to taxes. The "cafeteria plan" method. As we know, this just converts what would have been taxable compensation to a tax-free employer provided benefit (i.e., a benefit the value of which is excluded from income because it becomes an employer provided benefit excludible under 105). Employer offers $100,000 and employee elects to reduce that through the cafe plan by $1,000. So, employee gets $99,000 net. Employer takes the $1,000 that the employee gave up and pays the premium -- using either of the above two methods (direct payment or reimbursement method). I don't think it's rocket science here. Perhaps the bigger issue is why do you even need a cafeteria plan? You only need it if you want to give employees a choice. If the IRS can prove that you gave the employees a choice of $100,000 in cash or $99,000 in cash plus a $1,000 fringe benefit (don't get hung up on the terminology), then the $1,000 is taxable b/c the employees had constructive receipt.

As long as the HDHP is insured rather than self-funded, there isn't a discrimination problem.

I agree with B2kates. For insured benefits, an employer can decide which employees it will pay the coverage for. However, you may find that state insurance restrict your ability to do this - that would obviously vary from state to state. This has been discussed in prior threads, but you could have a premium payment account under a cafeteria plan. There are some non-tax issues that arise (such as HIPAA) so you have to be very careful doing this. The other wrinkle is that you only want this to be available to managers. If they are HCEs, then you're going to have a potential discrimination problem.

One point regarding the DOL guidance on missing participants in terminating DC plans. It only applies to amounts not subject to the J&S rules. So, for amounts of $5,000 or less, the new gudiance should be followed (noting that the IRA rollover is the DOL preferred method). For amounts in excess of $5,000, you're still in the same old quandry that everyone has been dealing with for years. On the automatic IRA rollovers, eliminating mandatory cash-outs is certainly a consideration. But, unless you think a plan will be continued forever (I know that's what you are supposed to intend when establishing a plan), you are just delaying the inevitable. And, many years down the road it may even be harder trying to locate a missing participant.

I don't think it's specifically addressed. It would be best to have all of them sign - but I suspect that's not the answer you want to hear. But, I could certainly make some arguments that only one needs to sign the certification (the best argument being that one entity has the authority to sign on behalf of the other entities).

Nope. See the IRS web site: http://www.irs.gov/retirement/article/0,,id=122798,00.html

There is no problem being covered by 2 HDHPs (as long as each satisfies the definition of a HDHP). Having double coverage is not much different than having double coverage under a low deductible plan. Once you hit the deductibe (whatever it is), you may end up with 100% coverage by having both plans. If a person only has a HDHP plan (or plans), then you look at the HSA rules. I don't have the dollar limits handy. But, you look at the first individual. If he/she only has individual coverage, you base the HSA limit on the deductible (using the lower deductibe of the 2 policies). If he/she has family coverage, then you use the deductible under that policy. But, that higher limit must be split among the family members. They can split it however they want - but the total HSA contributions as a family can't be more than the deductible under the HDHP (or if less, the applicable HSA dollar limitation).

Unfortunately, there just any helpful guidance on how to apply the nondiscrimination test under IRC 105(h). The regulations leave a lot to be desired. Personally, I think you hit the nail on the head -- you stated that the plan would pass the coverage test. Those of us who are familiar with the retirement plan rules think that's all you need to worry about. If you can pass coverage by excluding the group, then you're fine. Another way to look at it. Suppose you provide the excluded group some benefits, but not the full benefits being provided to everyone else. How would show that the plan is nondiscriminatory? I'd argue that you use the same coverage test. Break it up into 2 "hypothetical" plans and see if you could pass coverage. If so, then I think you pass nondiscrimination as to benefits. That would be the case if you provide them with a smaller benefit, or no benefit at all.

Actually, it's not clear that you have to pro-rate the comp limit. You have to pro-rate it if there is a determination period of less than 12 months. And, only counting compensation while a participant is not a short determination period. So, if compensation is defined as compensation paid for the calendar year, you don't have a short determination period. And, arguably, if you exclude compensation while not a participant, then no one has compensation prior to the effective date of the plan, but you don't have to pro-rate the dollar limit. As for top-heavy purposes, compensation is 415 compensation. Compensation is based on the limitation year and an initial short plan year doesn't create a short limitiation year. Rather, the regulations only address a short limitation year when there is a change in limitation years. You don't have a change here. So, I think the full year counts (and for 415 comp, you can't exclude compensation prior to date of entry). The minimum gateway can be based on compensation while a participant.

Sorry for the delay in responding. I don't check these on a regular basis. Thanks KJohnson, oriecat and calcu for providing the links, etc. I was remiss in pointing out the items all of you raised. As far as being able to run an individual premium through the cafe plan, I'm confident it's allowed and will address (hopefully) GBurns' issues below. There are 2 fundamental issues being raised. First, KJohnson pointed out quite a few issues - HIPAA, COBRA, nondiscrimination, etc. I have similar concerns and that's why, even thought it's allowed, I wouldn't normally advise someone to do it. If I'm not mistaken, some or all of the Blue Cross companies won't accept premium payments for individual policies if the employer pays the premium b/c of HIPAA. And, I don't know how they would know if an individual were reimbursed but they may ask on their applications. Under HIPAA - if 2 people with the same employer have the same policy, then the 3rd person can't be denied coverage. These issues arise b/c amounts that go through a cafe plan are treated as "employer" provided benefits for purposes of the Code. They aren't generally treated that way under ERISA (they are employee contributions). It's similar to the treatment of elective deferrals made through a 401(k) plan. But, that's irrelevant - the point is that you need to watch out for these issues b/c they can be a problem. The second fundamental issue is that a cafe plan must generally be for employees. But, it's o.k. for others to benefit. So, is paying insurance that covers just the dependent, and not the employee, a violation of the 125 rules? I think Harry Becker has waffled on this one when referring to spouses. Personally, I think it should be o.k. b/c the employee is clearing benefiting by having coverage for a dependent. As to GBurns' concerns -- I agree that it's very confusing. The issue I THINK you are raising is that if the emplyer reimburses the employee for the premiums (that were paid with after-tax dollars), then isn't that an FSA? And, the regs prohibit the reimbursement from an FSA. I have the same concern (even though I don't think it's a problem). If you look at the stuff everyone cited above -- including the IRS training manual - they allow the payment or reimbursement of individual premiums. I know this is preaching to the choir. But, on what basis can the IRS say that you can't do it? The 61 ruling says it's allowed under 105. 125 basically allows ANY benefit under 105 to be made available as a choice of cash or a tax-free benefit. So exactly what give the IRS the authority to restrict an otherwise permissible benefit under 105. There's no specific prohibition in the statute (such as the prohibition for long-term care insurance) and it's not a carryover of benefits issue. And the 61 ruling didn't say the reimbursement was a self-funded health plan subject to the provisions of 105(h) -- which arguably is why you can't reimburse premiums from a health FSA -- b/c it would violate 105(h). Not to harp on this, but don't you find it troubling that they are setting forth what can be reimbursed under a health FSA in 125 regs rather than 105(h)? There are some provisions in the code that enable the IRS to craft rules through regulations. I don't see any such provision in 125 so they should be limited to interpreting 125, 105, etc. not creating additional prohibitions on what would otherwise be permissible benefits. And, I think they sort of acknowledge this - look at the training manual and Q&A 7 where is says that the prohibition of payment of ins. from a health FSA doesn't prevent the payment through the normal operation of the plan. The IRS seems to think a separate premium payment account is allowed. And, here's another twist. Suppose I have a health FSA and rather than ask for a reimbursement I request that the plan pay the premium directly to the carrier. Is that o.k.? If reimbursing a premium pursuant to 61-146 creates an FSA, then arguably having the premium paid directly from a true FSA to the carrier shouldn't violate the prop. 125 reg. The bottom line is it doesn't make much sense and it seems, to me that the Treasury and IRS may be taking inconsistent positions. I don't like premium payment accounts b/c of the other issues. But, many of us feel that it's allowed for the reasons cited in prior posts and this has been confirmed (informally) by Harry Becker on a consistent basis for many years. I think there is enough "official" guidance to support it - the statute and the 61 ruling and the 125 reg might not contradict that entirely b/c it allows premiums to be paid through the normal operation of the plan. If anything, I'd argue that the prop. 125 reg is invalid rather than arguing that a premium payment account creates an FSA that is subject to that prop. reg. What I really wish is that the Treasury clear this one up....

Last I heard, they were with OMB (office of management budget). I don't know how long it will take for that office to release them, but that's probably just a clerical type process (i.e., no review of the substance of the regs). So, they should be out very soon.