MWeddell

Yes, a distribution from a 403(B) contract may be rolled into an IRA.

The truncating or "rounding down" of the COLA figures was (1) determined by Congress (not the IRS) and set forth in the Code itself, and (2) designed to raise revenue not to be mathematically pure. Hence, it's not a very applicable precedent if we're ever in a position of trying to convince the IRS that truncating is a reasonable interpretation in a completely different context.

No, it can't be rolled over into a 401(k) plan, although Congress is thinking about changing it. The best cite I can give is Code Section 403(B)(8)(A) which basically says that the distribution from a 403(B) contract won't be taxable if it's rolled over into another 403(B) contract ("an annuity contract described in paragraph (1)" literally) or an IRA. Also, it won't fit the definition of a rollover in 415© so that it'll be subject to 415 limits when it comes into the 401(k) plan, if the plan even allows lump-sum after-tax contributions.

I agree there's no reason to reject the QDRO. Code Section 414(p)(7) contains a provision limiting the retroactivity of QDROs to an 18-month period, but it only applies to when payments are made, not the date used to calculate the portion assigned to the alternate payee. If the QDRO says payment isn't made until it's accepted by the plan administrator as valid (or the "earliest retirement age" if the plan doesn't allow immediate distributions to alternate payees), then it complies with the Code and the plan administrator must accept the QDRO as valid as long as it clearly specifies the manner in determining how much is assigned to the alternate payee.

You're correct, assuming that company A owns less than 80% of the joint venture. Note that if company A owns more than 50% of the joint venture, that both plans are aggregated for 415 purposes. Usually, the term "joint venture" implies an exact 50/50 split in ownership, so I'm probably being overly cautious to mention these concerns.

Tom Poje raises good points, that the client won't be able to get to $30,000 for each of the HCEs without a money purchase pension plan and that the minimum contribution to D won't work because of the top-heavy requirement. I'll retract my earlier suggestions along those lines. There's still the issue about whether excluding D and including G is desirable.

If your client truly says "I don't care about equity among the nonhighly compensated employees" then you'll be better off to include G and exclude D from the plan's coverage. Measured in terms of dollars, not % of G's or D's compensation, it'll be cheaper to include G and exclude D which will still allow you to pass the 70% ratio percentage coverage test. Of course D might decide to quit on the day other employees receive their profit sharing plan statements! This extreme situation points out that you almost always have to discuss with the client to what extent they want to tolerate inequity among NHCEs just to improve the testing numbers. Also, let's make sure these are new plans you're putting in, because otherwise you can't change eligibility and allocation conditions after the plan year ends. Profit sharing plan should be written so that each cohort of NHCEs born in a particular year is allocated a different contribution. The IRS has loosened up on the definite allocation formula requirement so that this type of plan should be able to get a determination letter. Putting 3% into a money purchase plan makes sense in the abstract, but raises administrative costs and probably is unnecessary. The 15% 404 limit applies to employees in the aggregate and probably won't limit you from putting in 18.75% for the two HCEs because you'll be able to put in a lesser amount for the NHCEs (who are younger). As a practical matter, put in a nominal amount, e.g. $50 for D, so you can use D's compensation when computing the 15% 404 limit. [This message has been edited by MWeddell (edited 03-11-99).]

Ask for a Summary Plan Description (abbreviated SPD) which will describe the eligibility, vesting, and benefit formula. Your husband will become vested, i.e. own the benefit, typically after 5 years of service. However, the statement that 80% of pre-retirement income will be provided (typically when including social security and all employer-provided retirement programs) should be treated with skepticism for someone who is hired at age 52.

It sounds like the definition of compensation you propose to use for ADP testing is not a 414(s) safe harbor definition of compensation, but you've done 414(s) testing to demonstrate that the definition is not discriminatory. If I've restated your situation correctly, Treas. Reg. 1.401(k)-1(g)(2)(i) makes it clear that you can use your plan's definition of compensation for ADP testing as well. There's a parallel citation for ACP testing in the 401(m) regulations.

Sorry I'm a little slow to respond, RMassa, to your question. The 403(B) is not aggregated with the qualified plan that holds the matching contributions. The whole idea of aggregation is not in ERISA. Instead, aggregation is a discrimination testing concept in the Internal Revenue Code and does not apply here. Aggregation for the limited purpose of 415 is covered in the prior posts. The 403(B) plan is subject to ERISA because by using the data to compute matching contributions, the employer is acting outside of the very limited role set forth in Labor Reg. 2510.3-2(f). The employer is acting more like a plan sponsor, not just a conduit for the annuity issuers. In addition, one could argue that the match makes the 403(B) participation no longer "completely voluntary" as stated in that regulation. It's not 100% clear, but I'd definitely urge my client to err on the side of treating the 403(B) as subject to ERISA even where the match goes into the qualified plan.

Not only is this legal, but it's a sensible plan design in some situations. It also improves the 415© situation because the 401(a) and 403(B) plans are subject to separate 415© limits. Another advantage is that the 401(a) plan can cover all employees, regardless of whether they contribute to a 403(B) contract or whether they contribute to a 401(k) plan for the for-profit affiliates. Even though the match isn't made to the 403(B) program, one ought to treat the 403(B) program as subject to ERISA now in my opinion.

The short answer is that you must use a 414(s) definition of compensation for the 3% nonmatching contribution to qualify as a safe harbor 401(k) / 401(m) contribution. Section IV(B) of Notice 98-52. W-2 taxable compensation plus any pre-tax deferrals such as 401(k) and 125 deferrals, automatically satisfies 414(s). If you want to modify it to exclude overtime, bonuses, and commissions, I think you can do that if you perform 414(s) testing and your definition of compensation passes that 414(s) testing as not favoring highly compensated employees. It's an interesting question because it could keep the cost of meeting the 401(k) safe harbor requirement down a bit lower if the employer passes the 414(s) test.

For testing for the plan year ended 12/31/1998, the plan is testing employees who otherwise could be excluded under IRC 410(a) separately. In other words, those with less than a year of service (including any entry date assumption) and those under age 21 are tested separately. An employee was hired April 1997, earning over $80,000 during the rest of 1997, and became eligible to participate in the 401(k) plan on 5/1/1997. The employee terminated employment February 1998. For the 1998 testing, is this person considered as having < 1 year of service because he terminated employment before 1 year of service, or is this person tested with the group of employees with at least a year of service because that's what he would have had at the end of the plan year? Alternatively, is either interpretation reasonable considering the vague regulatory guidance on this issue, as long as we're consistent for all employees in that plan? Yes, I realize the rules change effective for 1999, but my question relates to the 1998 rules.

HIPAAdrome is probably right, although it does depend on how the plan document is worded. Occasionally, we see profit sharing plans drafted to see "the annual contributions shall be 5% of pay, unless the employer declares a different amount shall be paid" so that broadening the group who receive the 5% of pay contribution after they've satisfied the conditions for the allocation isn't a problem. However, in the more typical situation where a discretionary amount is contributed and the plan just states how its allocated, amending the plan now for the 1998 allocation would violate 411(d)(6).

401(k) plans didn't exist in 1977. The employee probably contributed to a Section 403(B) contract, which often are called tax-deferred annuities. Most of those programs are administered directly between the annuity provider and the participant with the employer taking only a passive role. Thus, if the employee knows who the annuity provider was, she'd probably have better luck calling the provider directly rather than going through the old employer. She won't be able to roll over 403(B) money (if my guess is right) into a new employer's 401(k) plan or any other type of qualified plan. She may be able to roll it into an IRA, but many annuity providers place substantial penalties on distributions prior to a specified age. Good luck.

There's no reason that the stable value fund couldn't be one of the three core funds in an ERISA 404© set of funds. In order to cover the risk & return range that is appropriate for 401(k) plan participants, most observers conclude that one must have at least a money market, stable value, or possibly a short-term bond fund as a conservative option. You mention that all options will allow at least monthly changes. Note that the stable value fund may have to offer changes at least as frequently as any of your other investment options under Labor Reg. 2550.404c-1(B)(2)(ii)©(2)(ii). In other words, you can't allow participants to switch out of an equity fund on a daily basis unless they can go into an income-producing, low risk, liquid fund on a daily basis too.

I agree you've got to read the document and follow what it says. However, if a 401(k) plan, which is not a pension plan, offers annuity forms of payment, the spousal consent requirement and other QJSA requirements don't become effective until such time as the participant elects an annuity form of payment. If the plan's normal form of payment is a lump sum, not an annuity, then at the time the loan is taken you may not be subject to spousal consent rules even though the plan offers annuity options when a participant takes a complete distribution.

Most 401(k) surveys measure the situation statically, by looking for correlations between plans with higher participation rates and plans with various features. Basically a matching contribution and (to a lesser extent) offering participant loans increase participation and everything else, including number of investment funds, is irrelevant. The one survey I saw that tried to measure the situation dynamically, i.e. did the plan sponsor take action X and if so what was the effect on the participation rate, was a Foster Higgins survey on Savings Plans. They had the question in their 1994 survey but dropped it from their 1995 survey. The question was fascinating, but apparently the survey's authors didn't agree! I don't have any more recent surveys of theirs and don't know whether the survey still is conducted annually.

If the plan document (as amended by the end of the 1999 plan year) says that you're using the prior year testing method, what you've proposed is a reasonable interpretation of the guidance the IRS has given us in my opinion. Even better: you could test those employees who otherwise would be excludable under IRC 410(a) separately when computing the 1999 average percentages, resulting in no NHCEs in the 21+ / 1+ group, so there won't be any restriction on how much the HCEs defer until the year 2001.

T Hoffman states the rule correctly. See Treas. Reg. 1.411(d)-4(d)(8).

1) 20% withholding applies to all eligible rollover distributions, which includes most common withdrawals and distributions in the form of lump sum payments. One cannot waive that withholding. If a payment is not an eligible rollover distribution, the amount of withholding varies depending whether the payment is periodic or nonperiodic and the participant has the right to waive withholding. 2) One cannot take a hardship withdrawal of investment gains after 12/31/88. (This date might vary by up to six months for plans that didn't have calendar plan years.) One also cannot take a hardship withdrawal of "qualified" contributions designed to meet safe harbor contribution requirements or to improve ADP / ACP tests. Any other type of contribution has hardship withdrawal options to the extent specified in the plan document. There are regulations addressing hardship withdrawal of elective deferrals, which include several different options, which often are also used to defined what constitutes a hardship for plans that allow other types of contributions to be withdrawn due to hardships.

The IRS position in the 1995 examination guidelines for 415 is that the terms of the plan must preclude employer discretion. The plan(s) should state how a 415 failure is corrected and you have to follow what's in the plan document.

L CARUSI is absolutely right. What was I thinking yesterday?

Actually, Rev. Proc. 97-41, Sections 1.02(4) and 12 says that the deadline for amending 403(B) plans or annuity contracts to comply with SBJPA, GATT, and USERRA is the first day of the first plan year beginning on or after January 1, 1998. I wouldn't worry about missing that deadline though, based on verbal conversations I've had with IRS and DOL officials. [As usual, their verbal statements aren't binding, but it lets you know what their current views are.] The reason why qualified plans get a longer deadline for amendments is that Section 401(B) doesn't apply to 403(B) plans, so the IRS doesn't believe it has to right to extend a 403(B) plan remedial amendment period. The good news is that the Code doesn't require a 403(B) plan to be written anyway, so the IRS doesn't demand a written 403(B) plan during any audits. The only example in Rev. Proc. 97-41 deals with the annuity contract, not the plan document, being timely amended. So much for the IRS deadline. In short, there's no real deadline for amending a non-ERISA 403(B) plan because there's no requirement for a written plan document to begin with. What if we're dealing with an ERISA 403(B) plan? This doesn't make a difference to the IRS, but now we'd have to consider what the DOL would do. It's possible that the DOL could assert that the failure to timely amend a plan document violates ERISA 402(a)(1) and ERISA 404(a)(1)(D), but they've never been very rigorous about asserting that. I believe a strict reading of USERRA along with the SBJPA states that USERRA's veteran reemployment rights provisions should have been in the 403(B) plan document by 12/12/1996 (still assuming that we're dealing with a 403(B) program that is an ERISA plan, not the elective deferral only variety). However, the DOL doesn't have a hard and fast deadline so that if your amendment is done before the DOL were to ever audit your plan, it'd probably be fine. Think about how many years retroactively we made TRA'86 amendments to our clients' plans (with an IRS remedial amendment period but no parallel extension from the DOL) and the DOL never challenged the timeliness of those amendments. I'd tell your client to amend the plan ASAP, but it's unlikely that any harm will occur as a result of the late amendment.

Sounds fine, although I don't understand why you stated that new comparability doesn't work. One cannot impute permitted disparity for the 3% safe harbor contribution, but can still test it on an age-weighted basis. Given that the HCEs are younger than the NHCEs, it sounds beneficial to do that. Perhaps there are other facts that you didn't list in your post that caused you to conclude that new comparability wouldn't work, but I'd consider it.