Bob R

Here's a very brief rundown of the nondiscrimination tests that apply to benefits. 1. There is an overall 25% test (referred to as the concentration test) that applies to all benefits available through the cafeteria plan. It's based on key employees under IRC Section 416 which isn't the same as HCEs. It's not clear, but I wouldn't include the premiums the employer is paying for the HCEs because that's not part of the cafeteria plan. But, you do aggregate the premiums paid by the NHCEs, along with all amounts contributed for dependent care and the health FSA. 2. For dependent care, there are two tests -- the 25% test mentioned by SLuskin in the prior message (this is based only on amounts contributed for dependent care), and, as you pointed out, the 55% average benefits test. The issue with the average benefits test is whether you include those eligible but who don't elect a benefit as a 0%. That brings the percentages for the NHCEs way down -- especially when you have a lot of NHCEs who don't even have dependents. This particular issue is one that is constantly being debated and for which there is no definitive IRS guidance on. 3. For the health FSA, as long as the same maximum is available to everyone, then there is not a discrimination problem. The test is based on availability, not utilization. Again, this is a brief rundown of the nondiscrimination tests that apply to benefits. You also have to make sure you satisfy the nondiscriminatory eligibility rules. In most plans that's rarely a problem. Providing all of the medical insurance premiums for HCEs only is not a problem due to the repeal of IRC section 89.

I don't think you can use Rev. Proc. 2000-20. All it states is that standardized plans can be designed to use the 410(B)(6)© transition rule. But, it's not automatic. And, since this isn't a change being made due to a GUST change in the law (nor is it integrally related to any GUST change in the law), I don't know of any authority to disregard the terms of the plan. In other words, this isn't something that you can apply in operation and then retroactively amend the plan at a later date.

Whether they have individually desgined plans depends on the language in the prototype plan. If the prototype plan was approved by the IRS and it doesn't automatically cover all members of the controlled group or affiliated service group, then it was mistake on the part of the IRS. But, adopting employers are generally able to rely on the terms of the prototype as approved. Thus, just because the IRS made a mistake doesn't mean the adopting employers have an individually designed plan. They are just following the terms of the plan as approved.

I don't think this any problem paying the fee from the plan assets. However, what's not clear is whether you must pay it from all of the assets or whether you can pay it just from the account of the individual you are searching for.

For whatever it's worth, I think 3 is the correct answer. There is nothing permitting you to disregard the terms of the plan. And, an amendment now would, in my opinion, be considered a significant change in the coverage. If both plans require 1 year of service as a condition of eligibility, then you might be in good shape. The reason is because I think service for both companies only needs to be taken into account after the date they became a controlled group. For example, employees of B will only be eligible under A's plan if they have 1 year of service. But, B didn't actually adopt A's plan so I would argue that they won't have a year of service for purposes of A's plan until they been employed as part of the controlled group for one year.

The comments in this thread clarify that no one knows what protection 404© will provide. Jon Chambers left out one alternative where an employer is concerned about potential liability due to investment direction. That alternative is don't provide for participant directed investments. I know that's easier said than done. It's like plan loans to participants -- most people think it's a bad idea for participants but we have to offer them because everyone else offers them. Likewise for directed investments. Most people don't think it's such a good thing for the majority of plan participants. Daily systems are certainly the way to go these days. But, look at the "educational" material handed out. "This is a great system because you can change your investments at any time. By the way, you really don't want to use this system. Market timing is baaaaad -- but in case you want to try it, we've given you the tool to do it." I like the mentally restarded competent mail clerk example. This is great for the "sympathetic plaintiff" example. This mail clerk brings an action against the plan for investment losses. The court goes through the laudary list of items needed to fall within 404© protection. OOps - the plan only complied with 4 out of 6 -- sorry trustees, you loose. Jeff V mentioned that the voice unit suggests that participants obtain a prospectus. I don't think that complies with 404© where a participant is first investing in a fund. Is that a material factor that can blow 404© protection? We won't know until the litigation starts. But, add a sympathetic plaintiff to the equation and who knows what a court will do.

I've always taken the position that the regulation is generally a "safe harbor" provision. In other words, the regulations state that you won't be deemed to have a CODA if the election is irrevocable, made at the time you are first eligible, and applies to all plans of the employer (existing and plans established in the future). But, it doesn't state that I automatically have a CODA if I have an election that doesn't satisfy the regulation. Rather, I think it's a facts and circumstances test to determine whether an individual is receiving compensation for electing out the plan. For partners, you must use the safe harbor because electing not to participate autmatically increases earned income. But, for a corporation, electing out doesn't automatically increase compensation. Having said that, if I were an IRS agent and I saw someone using an election that doesn't meet the "safe harbor," I'd hunt pretty hard for the smoking gun. Generally people, especially NHCEs, don't turn away a free contribution from an employer. But, there are legitimate reasons for electing out. For example, I've heard of situations where the only allocation is a small amount of forfeitures for a particular year. If the amount is allocated to an individual's account, the individual is an active participant and might be precluded from making a deductible contribution to an IRA. The bottom line is you have to be extremely careful in using elections not to participate. Plus, if NHCEs elect out, you will have the 410(B) coverage issues to deal with. If the election satisfies the safe harbor, then, as Tom pointed out, the employer would never be able to use a standardized prototype adoption agreement.

The relevant portion of Question 20 of the proposed regulations is below and, as pointed out, it limits loans to 2 per year. This wasn't in the final regulations because both the proposed and final regulations were issued on the same day. It's still possible these proposed regulations will be finalized. But, unless I overlooked it, there isn't anything in the proposed regulations stating that they can be relied on. Thus, I agree that right now there is no legal limit on the number of loans that can be made to a participant. (3) Multiple loans. For purposes of section 72(p)(2) and this section, a loan to a participant or beneficiary shall be treated as a deemed distribution if two or more loans have previously been made from the plan to the participant or beneficiary during the year. This limitation applies on the basis of a calendar year unless the plan applies this limit on the basis of the plan year or another consistent 12-month period.

Under the IRS rules (Rev. Proc. 2000-20 or the predecessor guidance from Rev. Proc. 89-9 or 89-13) there are basically 3 differences. However, depending upon the drafter of the adoption agreements, there could be other differences. For example, the use of the term "flexible" in the names of the adoptions agreements is a term added by the drafter -- it's not a term used by the IRS. The differences mandated by the IRS fall within 3 categories. The reason for the differences is because a standardized generally may not include any provision that has the potential of being discrminatory or violating the coverage rules. The 3 "general" differences are: - Who must be covered. In a standardized plan, the only class of employees that you can generally exclude are union employees and non-resident aliens. So, you can't exclude any other classes, nor any employees that are members of a controlled group or affiliated service group. - What compensation must be used. In a standardized plan, you must use a safe harbor definition of compensation. So, you can't exclude items such as commissions or bonuses. - Who must share in allocations. In a standardzied plan you can't require that a participant be employed at the end of the year or complete year of service to recieve an allocation or accrual. (You can still require 1 year (or 2 years with full vesting in a non-K plan)/ and 21 to become a participant. Those are the 3 primary differences. But, as mentioned, the actual adoption agreements that you have could have other differences depending upon how they were drafted.

The beneficiary can waive the annuity form of payment without the consent of the beneficiary's spouse. The QJSA rules only apply to participants - not to beneficiaries.

The non-key employees in the 401(k) plan are entitled to top heavy minimums, even though it has more liberal age/service than the mp plan. To be more specific, I'm presuming that key employees participate in both plans. If so, you have a required aggregation group and based on the facts you presented both plans are top heavy. Unfortunately, even though the government encourages liberal age/service in 401(k) plans (that's why you are able to permissively disaggregate or exclude NHCEs from adp testing if less than 1 yos/age 21), there is no relief from the top heavy rules. Since the money purchase plan already provides 3% to some of the key employees, you just need to provide the 3% top heavy in the 401(k) plan for those non-key employees who are not in the m/p plan -- assuming the plan documents provide that the top heavy minimum is provided in the m/p plan to the extent a non-key is in both plans. If any of the non-keys who are getting the top heavy in the 401(k) plan are HCEs (this is unlikely but it's possible), then you have some additional nondiscrimination testing to perform. That's because you could end up with a situation where some HCE non-key's get a 3% employer contribution in the 401(k) plan and some NHCEs get nothing if the only employer contribution to the 401(k) plan is the top heavy minimum. I know this is rather extreme, but when new prototypes are available for GUST, you will see language in adoption agreements stating that if both plans do not benefit the same participants and only one plan provides the top heavy minimum when someone is covered by both plans, then the uniformity requirement of the 401(a)(4) regulations may be violated.

Kip What I was attempting to state is that a health FSA and a self-funded health are both vehicles covered by the same Code Section -- 105(h). Thus, in a pure legal context, a health fsa IS a 105(h) plan. As you pointed out, there are numerous differences between what people in the industry, including myself, refer to as 105(h) plans and health FSAs. But, these differences are because of plan design, not because of any statute or regulation. A self-funded health plan has all of the features you mentioned because that's what the employer wants -- it wants coverage similar to an indemnity plan, HMO, PPO, etc. And that's when you need to apply underwriting rules and other factors (which would be required under 419 and 419A if the plan is funded). But, a health FSA is also a self-funded plan. We typically use the term health FSA when it's funded by employee pre-tax contributions and the maximum benefit is based on those contributions. For purposes of the Code, the employee pre-tax contributions are converted to employer contributions. That makes a plan funded by employer contributions, i.e. - a self-funded health plan. There is a risk of loss and as you pointed out, someone might even try to sell an employer stop loss coverage. See if the following example makes sense: Let's start a typical health FSA offered through a cafeteria plan that reimburses participants for all medical expenses. Could I amend that health FSA to require that there be a deductible? I think you could. (Of course participants will take that into account in determining the amount they want to contribute to the plan.) Next, could I amend the plan to require a 20% co-pay? Again, sure. Then, could I amend it to exclude coverage for eye-glasses and experimental procedures? Again, sure. The result is I've taken a classic health FSA and by merely changing the benefits offered, created a self-funded health plan. The point of my original message is that the health FSA was a plan subject to 105(h) from the very beginning. As far as use-it-or-loose it, I really don't see much of a difference between what employees contribute to a health FSA and an insurance premium. Think about this description of a health FSA when describing the use-it-or-loose rule to an employer: A health FSA is like an insurance policy except the participant determines the maximum benefit and the employer is in the shoes of the insurance company. If an employee has paid the "premium" for the month, then the employee can rack up expenses to the policy maximum. If the employee pays premiums for the year and incurs no expenses, the insurance company (i.e., employer) has a profit.

The key to the provision you referenced (which is at the bottom of this message) is that it applies to "matching contributions made at the employer's discretion." What that means is that if the employer has a discretionary match, as distinguished from an actual matching formula stated in the plan, then the total discretionary match can't exceed 4% of compensation. The net result is that a plan satisfies the ACP test if: 1) there is an ADP safe harbor contribution (in your example it's the safe harbor match) 2) matching contributions do not take into account deferrals over 6% of compensation and 3) if the plan has a discretionary match, then the discretionary matching component can't exceed 4% of compensation. For example, let's say the safe harbor match is 100% on deferrals up to 4% of compensation. This is fully vested and subject to the distribution requirements. That satisfies the first requirement. Then, let's say the plan provides -- in the document -- that there is another match equal to 200% (that's high but there's nothing preventing it) of deferrals up to 6% of compensation. That match is subject to a vesting schedule and employment at the end of year/1000 hours. That satisfies the second requirement -- i.e., the match doesn't take into account deferrals over 6% of compensation. It's worth noting that because of this 6% limit, if the safe harbor match was 100% on all deferrals (rather than 100% on deferrals up to 4%) you'd satisfy the ADP safe harbor but you wouldn't satisfy the ACP safe harbor. Last, the plan provides that the employer has the discretion to make an additional matching contribution. However, in applying that additional match, only deferrals up to 6% of compensation will be taken into account and the maximum match any participant can receive from this discretionary match is 4% of compensation. This match is subject to a vesting schedule and EOY/1000 hours. This satisfies the third requirement. A plan that includes these 3 matching contributions would satisfy the ADP and ACP safe harbor provisions. From Notice 98-52: A plan fails to satisfy the ACP test safe harbor for a plan year if the plan provides for matching contributions made at the employer’s discretion on behalf of any employee that, in the aggregate, could exceed a dollar amount equal to 4 percent of the employee’s compensation. This limitation on matching contributions made at the employer’s discretion does not apply to plan years beginning before January 1, 2000.

Actually, they are same. A health FSA is a permissible benefit that can be offered through a cafeteria plan only because it is subject to Code Section 105(h). If it wasn't a plan covered by 105(h), then there would be no provision in the code that would permit an employee to be reimbursed tax free. The confusion is really a matter of semantics and it just has to do with the design of the plan. Specifically, the benefits being provided. Most people refer to a 105(h) plan as a plan that resembles an indemnity plan (there is a deductible, co-pays, etc.). And, most people refer to a health FSA as one where benefits are based on an employees election w/out any co-pay, deductible, etc. Both plans are uninsured employer funded plans designed to reimburse employees for medical expenses. (For a health FSA offered through a cafeteria plan, the pre-tax employee contributions are considered to be employer contributions for purposes of the Code). Also, the use-it-or-loose-it rule applies to both health FSA's and the full self-funded health plan. Many self-funded health plans require an employee to pay a portion of the so-called "premium" just as with a regular indemnity policy. That employee paid portion can be paid pre-tax through a cafeteria plan -- it's generally structured through the premium payment portion of the plan. But, if the employee incurs no medical expenses, the self-funded health doesn't reimburse the employee just like an insurance company wouldn't reimburse a premium if someone didn't submit any reimburseable medical expenses. Likewise, the self funded health plan must pay up to the policy maximum (just like a regular insurance policy) if the "premium" is current. It doesn't matter that the individual only had the coverage for a short period of time.

I can't resolve the forfeiture issue because I'm not aware of any definitive guidance on the issue. But, see below for a portion of final 411(d)(6) regulations(the actual regulation is 1.411(d)-4) that were issued last September. Here's my take on the regulation: 1. This permits the transfer of a participant's "entire benefit." It doesn't refer to "vested" benefits. And, I think entire includes the vested and unvested portion. The reason I think this is because in situations where there has been an employment status change (e.g., union to non-union) service with the employer still counts for vesting so the participant's vested interest in the benefit must continue to increase. If you look at the part of the regulation below entitled "Circumstances under which transfers may be made" you will see that a change in employment status is one of the qualifying events. 2. It permits a voluntary "employee" transfer (not a rollover) to d/c plans of the same type. 3. If the participant agrees to the transfer, you are permitted to reduce or eliminate 411(d)(6) protected benefits. Requiring transfers to plans of the same types will protect all of the distribution restrictions that apply to a particular type of plan (e.g., money purchase plans will always be subject to the QJSA rules). 4. I think that this provision needs to be in the transferor plan to be used. The last part of the regulation (I apologize because it's not copied below) states that this right to a transfer is a 411(d)(6) protected benefit. Here's the relevant portion of the regulation (sorry about the format getting out of wack): (B) Elective transfers of benefits between defined contribution plans--(1) General rule. A transfer of a participant's entire benefit between qualified defined contribution plans (other than any direct rollover described in Q&A-3 of Sec. 1.401(a)(31)-1) that results in the elimination or reduction of section 411(d)(6) protected benefits does not violate section 411(d)(6) if the following requirements are met-- (i) Voluntary election. The plan from which the benefits are transferred must provide that the transfer is conditioned upon a voluntary, fully-informed election by the participant to transfer the participant's entire benefit to the other qualified defined contribution plan. As an alternative to the transfer, the participant must be offered the opportunity to retain the participant's section 411(d)(6) protected benefits under the plan (or, if the plan is terminating, to receive any optional form of benefit for which the participant is eligible under the plan as required by section 411(d)(6)). (ii) Types of plans to which transfers may be made. To the extent the benefits are transferred from a money purchase pension plan, the transferee plan must be a money purchase pension plan. To the extent the benefits being transferred are part of a qualified cash or deferred arrangement under section 401(k), the benefits must be transferred to a qualified cash or deferred arrangement under section 401(k). To the extent the benefits being transferred are part of an employee stock ownership plan as defined in section 4975(e)(7), the benefits must be transferred to another employee stock ownership plan. Benefits transferred from a profit-sharing plan other than from a qualified cash or deferred arrangement, or from a stock bonus plan other than an employee stock ownership plan, may be transferred to any type of defined contribution plan. (iii) Circumstances under which transfers may be made. The transfer must be made either in connection with an asset or stock acquisition, merger, or other similar transaction involving a change in employer of the employees of a trade or business (i.e., an acquisition or disposition within the meaning of Sec. 1.410(B)-2(f)) or in connection with the participant's change in employment status to an employment status with respect to which the participant is not entitled to additional allocations under the transferor plan. (2) Applicable qualification requirements. A transfer described in this paragraph (B) is a transfer of assets or liabilities within the meaning of section 414(l)(1) and, thus, must satisfy the requirements of section 414(l). In addition, this paragraph (B) only provides relief under section 411(d)(6); a transfer described in this paragraph must satisfy all other applicable qualification requirements. Thus, for example, if the survivor annuity requirements of sections 401(a)(11) and 417 apply to the plan from which the benefits are transferred, as described in this paragraph (B), but do not otherwise apply to the receiving plan, the requirements of sections 401(a)(11) and 417 must be met with respect to the transferred benefits under the receiving plan. In addition, the vesting provisions under the receiving plan must satisfy the requirements of section 411(a)(10) with respect to the amounts transferred.

As mentioned before, IRC Section 413© permits unrelated employers to adopt the same plan. That code section states that the exclusive benefit rule will not be violated and that all service with all entities adopting the plan will count. Also, as pointed out, you can't have one through a prototype. It is considered an individually designed plan so if you want a determination letter, you must file Form 5300. The user fee varies based on the number of employers that will be adopting the plan. As far as reporting, only one Form 5500 needs to be filed for the entire plan, but separate schedules are needed for certain portions of the form. (Prior to the 1999 forms, a separate Form 5500 was required for each of the employers adopting the plan.) As far as nondiscrimination testing, each company must run the applicable tests (e.g., 410(B) and ADP/ACP tests) separately. Lastly, if this is set up for a leasing organization, you get into another set of issues due to IRC Section 414(n). For example, that section provides that benefits providing by the leasing company are deemed to be provided by the recipient employer. Thus, deferrals of the leased employees might be tested both in the leasing orgs adp test and in the recipient employer's adp test.

I'm not aware of any specific IRS guidance on this. But, I agree that you can use the rollover to buy back the forfeited amount.

While surfing this site -- see the thread cited below about a spouse not consenting to a rollover (it's in this 401(k) section): http://benefitslink.com/boards/index.php?showtopic=9004 Just change the facts. Plan is amended to eliminate J&S. Participant takes the money and rolls over to IRA and spends money. Then gets divorced. Ex-spouse is not too happy. In this case $350,000 is potentially at stake. Is it ripe for a lawsuit?? While I think the spouse is out of luck, my opinion doesn't count. But, I'd argue that the qualified plan rules permit the amendment. And, Congress is aware that in non J&S plans, a spouse has no right to distributions made while participant is alive. That's why there was proposed legislation several years ago to require spousal consent for all plans.

In order to be consistent, if you use the restructuring, the bottom QNEC needs to made to people in the particular group that is failing (i.e., those over age 21/1 YOS). Alternatively, don't use restructuring and rerun the test. I presume you fail by an even wider margin. But, then you can make the QNEC to those under 21/1 YOS.

I don't have any experience with this. The argument makes sense. But, you might want to check out (I don't have the regulations handy right now) is the language in the 416 regulations that permits the use of the match to satisfy the top heavy minimum. The issue is whether it recharacterizes the match as something other than a match for all purposes of 401(a)(4) or just for nondiscrimination as to contributions. The reason that makes a difference is because of the benefits, features and rights (BFR) section of the (a)(4) regs. If the match is no longer a match for purposes of that section of the regs, then you might have a problem. The "rate" of a match is a must be tested under the BFR section of the regulations. If the match is no longer a match, then the HCEs may have a higher rate of match than the NHCEs becasue some of the NHCEs (the non-key HCEs) are not receiving the match. Just something to think about.

I think the regs do allow you to eliminate the J&S provisions, even though the J&S is currently the normal form of distribution. However, I have heard some people express concern about the elimination of the spouse's rights under the plan. Whether that will be problem may never be know -- until someone litigates it. All it takes is a sympathetic plaintiff. Participant withdraws all funds from the plan without spousal consent and spends it all. Then participant dies and widow or widower is in court because all of their savings have been spent. Who knows?

If it was a stock purchase, then I'm presuming that B formally adopted the plan of A. Once B adopts the plan, then I think all service with B would count. This would also be the case if A maintains the plan of B. You said that Plan B is being terminated. If there has been a stock sale that has already taken place, then it's too late. Once you own the stock of B, you are maintaining B's plan and all service is counted pursuant to IRC Section 414(a). If it was an asset sale, then there is no clear answer. Since B's plan will not be merged into A's plan, I think the former employees of A can be excluded because there date of hire with A would be the date of the acquisition.

I don't have the regulations in front of me, but the limit should be somewhere in 1.401(k)-1(d). FYI, the limit is the annual limit ($10,500 for 2001) reduced by the deferrals made by the individual in the prior year. So, if the person deferred $5,000 in 2000, then once the 12 month suspension period is over, the most that can be deferred for the remainder of the calendar year is $5,500.

The direct answer is there is no minimum level of participation that is required. But, the long answer as to whether you need to perform any tests for the plan in question depends. You indicated who is benefiting in the plan. For the tests that are based on utilization (such as the 25% and 50% test for dependent care), you don't need to perform any tests. But, there are nondiscrimination tests that are based on eligibility for coverage. For these tests, you have to look at who was offered the coverage and not who utilized it. So, if the only people who are even eligible to participate in the plan are key employees and NHCEs (and the class of people who are eligible is considered to be a resonable classification), then you'd automatically pass the nondiscriminatory eligibility tests. In that case, you wouldn't need to run any nondiscrimination tests on the plan at all.

When dealing with these types of expenses, a common approach that the IRS takes is the "but for" test. So, but for the pregrancy, would she have needed the medical equipment? I think the monitor (I presume it's some sort of fetal monitor) would be allowed. Reimbursing the cost of the thermometer would be a fairly aggressive position. I would not allow it to be reimbursed unless it is some sort of special thermometer for pregnant women.