david shipp

Remember that section 204(h) requires a somewhat tortured sequence of events. Notice must be provided AFTER adoption of the amendment and not less that 15 days BEFORE the effective date of the amendment. (The amendment must have a prospective effective date.)

Be sure to issue timely 204(h) notice

Is something missing here? $13,750 is 25% of $55,000 which would make sense if A made a salary deferral of $3,000 in 1996 when 415 comp was reduced by salary deferrals. Otherwise, it would appear that the answer should be 25%of $58,000 which isn't what you indicate. I could be missing something, of course, but i can't see what (otherwise I guess I wouldn't miss it :-) )

1. Reduce gross comp by salary deferral since 404 still based on "net" comp. 2. Apply 401(a)(17). This is done after step one so A has net comp of 160,000. 404 "Net" comp = $303,000. 3. 15% of $303,000 = $45,450. 4. Subtract salary deferral and match. $45,450 - $22,500 = $22,950

reduce highest dollar contributor first and bring him/her down to next highest -reduce A by $900 reduce next two on a dollar for dollar basis until you either use up excess or get to the third person in line. - reduce A and B by 1/2 of remaining excess since that eliminates excess before getting to C. So - A's reduction = $900 + 1/2(4700 - 900) = $2,800.

Of total salary deferral of $9,500, only 1st 6% ($6,000) was matched. Returned excess was not matched so no need to forfeit a match.

Commencement of minimum distributions would generate an annuity start date. Spousal consent should have been obtained if QJSA requirements applied.

Tom, I need some help with option #2. My understanding is that use of a deferral compensation definition that does not satisfy 414(s) simply means that the deferral definition cannot be used in nondiscrimination testing. Since we appear to be dealing with salary deferral contributions only (at least that is all that is referenced), the consequence of failing 414(s) is that a different compensation must be used in determining deferral percentages. If the ADP test is failed as a result, the plan uses its standard correction mechanisms. I'm not sure how other issues such as ABT fit in.

I would respectfully disagree with RMassa. In example 3, the conclusion states that the example does not satisfy the matching limitations because more than 6% of compensation is matched - elective contributions up to 6% of compensation are matched PLUS up to 6% of voluntary contributions are matched. This results in a possible match on up to 12% of compensation. Example 4 shows that a 50% fixed match on up to 6% of compensation can be combined with a discretionary match of 50% on up to 6% of compensation (the same 6% of compensation) and still satisfy the ACP safe harbor when combined with a 3% nonelective ADP safe harbor contribution. (This design satisfies both the safe harbor requirement that matched compensation be limited to 6% and the requirement that any discretionary match be limited in amount to 4% of compensation.) These examples should show that the nonelective contribution doesn't reduce the permissible match and that the same compensation can be the subject of multiple matching formulas.

I think the issue is that you have to *test* with 414(s) compensation. If your deferral definition doesn't meet 414(s) you don't have to correct the compensation used, but your adp/acp testing results will be different than you would otherwise have anticipated since you have to use a different comp. for testing. If the plan fails the adp/acp testing, standard corrections should be made based on plan provisions. If this isn't the case or if there are other issues I'm sure someone will chime in.

Thanks to MWeddell for a well-reasoned explanation of his position. This is a gray area and one that is subject to different interpretations. The valid business reason factor gives sponsors the best "cover" if they choose to go this route. It is obviously the plan sponsor's ultimate decision and they need to decide whether the business risk is worth the benefit. I don't think this is an issue that one would resign an account over if the client chooses not to adopt a recommendation that expenses not be charged in this manner.

The home loan exception to the five year repayment rule provides that the five year limit "shall not apply to any loan used to acquire any dwelling unit . . ." (sec. 72(p)(2)(B)). Purchase of land to build on doesn't appear to fit the bill. For what it's worth, an National IRS representative also expressed the belief that a loan to cover the materials to build a house would not qualify for a longer pay-back period where the borrower was going to build the house himself. His reasoning was that the dwelling unit couldn't be acquired until it was built.

As Charlie alludes to, the issue here is whether "a significant detriment is imposed under the plan on any participant who does not consent to a distribution." This comes from 1.411(a)-11©(2) which indicates that a participants consent to distribution is invalid if a "significant detriment" is imposed on a decision to leave the money in the plan. The QJSA Exam Guidelines (which also cover 411(a)(11)) discuss the issue of significant deteriment and state that disparate treatment between former participants and participants may result in a significant detriment to former participants. The example given of a significant detriment describes a plan that allows participants the option of an immediate distribution on termination of employment, but requires those who decide to defer to wait until age 65 to receive their benefits (i.e., no interim elections to withdraw permitted.) I realize that this is not directly on point, but I think that a good case could be made that charging expenses to former employees but not current employees does create disparate treatment between the two groups. I also realize that a number of plans are already operated in this manner. My recommendation to clients, however, is NOT to adopt this practice since it not clearly acceptable.

Mike, you are, of course, correct. My y2k breakdown came 8 1/2 months early.

The following is from Sal Tripodi's WEB site (via BenefitsLink - What's New). It appears that the IRS is not buying the argument that 410(B) plan disaggregation for 401(k)/(m) plans means that a (k) feature started mid-year in an existing plan can be treated as a new plan for purposes of the requirements of Notice 98-52. The fact that the desired result can be obtained by establishing a separate, new 401(k) plan mid-year and then merging the plans at the beginning of the next year is a classic case of form over substance. ------------------ When can a safe harbor 401(k) arrangement be added to an existing non-401(k) plan? Consider the following example. A company maintains a profit sharing plan with a December 31 plan year. Effective July 1, 1999, the employer would like to start a 401(k) arrangement and wants to take advantage of the ADP safe harbor rules under section 401(k)(12) of the tax code and explained in IRS Notice 98-52. Can the employer simply amend the existing profit sharing plan to add a 401(k) arrangement effective July 1, 1999, and have the plan rely on the ADP safe harbor for the plan year ending December 31, 1999? IRS National Office unfortunately says NO, according to a internal memorandum. The added 401(k) feature would be subject to the ADP test for the 1999 plan year, but could be converted to a safe harbor 401(k) plan as of first day of the next plan year - January 1, 2000. If the employer wants to start a safe harbor 401(k) plan effective July 1, 1999, it must adopt a separate plan with a short plan year which starts July 1, 1999, and ends December 31, 1999. The two plans could then be merged effective January 1, 2000. The IRS is taking this position because Notice 98-52 requires the plan year of a safe harbor plan to be 12 months unless it is a new plan. Furthermore, notice to employees must be given no later than 30 days before the first day of the plan year (with the exception of the March 1, 1999, transition rule prescribed by Notice 98-52). It seems like an unnecessarily narrow position, but if the employer in this example wants to add a safe harbor 401(k) arrangement during 1999 it will need to take the separate plan approach. Note that the new plan would not have to reflect all the safe harbor rules, so long as the written notice is given to employees by the first day of the first plan year and the conforming amendments are adopted within the GUST remedial amendment period.

Take a look at prop. reg. 1.401(a)(9)-1 Q&A F-5(B) "The account balance is increased by the amount of any contributions or forfeitures allocated to the account balance as of dates in the valuation calendar year after the valuation date. Contributions include contributions made after the close of the valuation calendar year which are allocated as of dates in the valuation calendar year."

Despite my previous passionate responses arguing that the (a)(17) limit didn't necessarily apply with respect to determining deferrals (and matching contributions), I was apparently only half right based on the most recent pronouncement on the subject as contained in the 99 EA Gray Book. "QUESTION #30 Other DC Issues: Application of Maximum Compensation Limit In a 401(k) plan, does IRC Section 401(a) (17) preclude the following? A. Employee A earns $300,000 annually. He enrolls in 401(k) calendar year plan in August, after earning $175,000. He defers $10,000 for the balance of the year. B. Employee A earns $300,000 annually. He participates in a calendar year 401(k) plan making monthly deferrals of a flat dollar amount of 1/12 of $10,000 in 1998, even though his pay exceeded $160,000 before he was done making elective deferrals. C. Same as B, but deferrals are a percentage of pay (3.33333%). RESPONSE All of the above are acceptable, assuming the plan is not drafted in such a way as to prevent it. In situation C, for example, a plan provision permitting deferrals expressed as a percentage of compensation but not permitting deferrals expressed as a dollar amount could not accommodate deferrals on pay in excess of $160,000. Where the plan permits deferrals expressed as a dollar amount specified in the employee's salary reduction agreement, the reference to a percentage in the individual agreement is irrelevant." I think the comments regarding option C begin to make it clear that any *percentage* election can only apply to $160,000-worth of compensation. That circumstance can be avoided, however, by electing either a flat-dollar amount of deferral which is not tied to compensation level, or electing the appropriate % that gets one to the 402(g) limit based on the (a)(17) limit. A flat dollar election would allow a participant to spread the deferral over the full plan year. The percentage approach would obviously accelerate deferrals if comp exceeded the (a)(17) limit. If this is the case with deferrals, it should follow that the match is also affected where the overall match is a function of a % of compensation (e.g., 50% of first 6% of compensation). MWeddell's example above points out that the match in this type of formula will vary depending on whether the comp. limit is monitored. If not monitored, the match could exceed the level determined by limiting compensation to the (a)(17) limit. [i'm not sure it is a BRF issue as much as a failure to follow plan terms. In either case, however, the result is not good!] So, this is a long response to a short question, and I now think the answer is to impose the (a)(17) limit where participant elections or plan provisions deal in percentages of compensation. A favorite saying attributed to Yogi Berra: It isn't what we don't know that get us in trouble. It's what we know for sure that just ain't so.

See Notice 88-33 (the three notices that still govern are 87-77, 88-33 and 89-32) 88-33 states that 1099-Rs need not be furnished until the regular deadline for reporting, but that participants must be advised in writing, at the time of the distribution, of the year in which the income is taxable and that receipt of income taxable in a prior year will require an amended return if the return has already been filed.

Al, Can you share where the IRS made this view known (and who the spokesperson was?) Some are argueing that the interaction of two 98-52 requirements effectively preclude EXISTING PS plans from adding a 401(k) feature mid-year on a safe harbor basis. Section V(A) says the plan must satisfy the safe harbor contribution requirement for the entire plan year and Section XI(B) says you can't add a 401(k) feature retroactively. You can parse through the words and build a case for both sides. Any light you can shed on IRS comments would be helpful

Let's come at this from a different angle. The required beginning date for non-5% owners has changed such that no minimum distributions are required while such participant is still employed. The IRS, however, has determined that the old required begining date provision (age 70 1/2 regardless) resulted in a protected optional form of benefit (the "age 70 1/2 distribution option," even though it wasn't optional). When a plan changes from old to new RBD, the age 70 1/2 distribution option has to be protected for benefits accrued to the date of amendment unless certain conditions are met. One of the scenarios that allows a plan to avoid protecting the age 70 1/2 distribution option is the inclusion of an in-service withdrawal provision (e.g., at age 59 1/2) since such a provision effectively covers the age 70 1/2 distribution option. Sooo . . . a plan with an appropriate in service withdrawal provision can alleviate the 411(d)(6) issue, but it has nothing to do with required minimum distributions once the participant reaches his new RBD. For an explanation of the 411(d)(6) issues regarding the amendment of the RBD, see the preamble and prop.reg. 1.411(d)-4, Q&A 10.

Joel, I'll try one more time then I have to move on. It has been a pleasure debating with you, however. The basic model under both 401(a) plans and TSAs is that distributions are taxable. In addition, distributions are restricted unless certain events occur. To preserve the retirement nature of distributions, rollovers are permitted as an exception to taxability. Pre-UCA, in addition to requiring a distributable event for the intial ability to distribute, there were also "impediments" to rollovers in the nature of conditions which had to be met if a distribution was to be rolled over. Congress decided that it was not good pension policy to restrict rollovers. If an individual qualified for a distribution, why put roadblocks in his way if he wished to roll it over to preserve it for retirement. Hence, UCA removed most but not all rollover restrictions. You will recognize that this is a restatement of your "absurdity" and I think most will agree that it is not absurd from a pension policy standpoint. . . . Q. Does the participant INTEND to use the funds for retirement by maintainin and preserving the tax-deferred status of the funds? There are only two possible answers. Answer 1. no: Therefore, restrictions apply to the distribution. Answer 2. yes: Therefore, no restrictions apply to the distribution. . . Tax policy doesn't live on intentions. If the distribution of benefits is limited to specified events, the fact that an individual could thwart those restrictions by simply stating an intention to leave assets in an IRA after a rollover isn't good policy. Now, if you were to impose pension-like restrictions on the IRA, the tax policy arguement would go away (to be replaced by other arguements about access to benefits while still employed.) Joel, this is as lucid an arguement as I can marshall (and a darn good one if I do say so myself). If you don't buy this one, then I would have to say you would have gone down with the Titanic muttering "this ship can't sink." :-) Again, thanks for opportunity to discuss the issue with you.

Joel, I would suggest that operation as you propose would negate triggering events altogether. Once an amount is rolled to an IRA there are no restrictions on withdrawal. Part of the "deal" for tax qualified treatment of retirement plans is that the benefits are to be primarily for retirement, therefore there are restrictions on when benefits can be withdrawn. Allowing distributions at any time, simply because they are not taxable at the time of distribution due to a rollover, would open plan benefits to unrestricted access . . . simply roll a distribution over and then take it from the IRA. Your argument that a change in the conditions that apply in determining rollover eligibility also changed the availability of distributions under a qualified plan is incorrect. The restriction on distributions comes from reg. 1.401-1(B) for section 401(a) plans and (as you point out) section 403(B)(11) for TSAs. These provisions fit in with the tax "deal" and there is no way Congress is going to modify it in the way your suggest. It wasn't their intention at the time of the amendment and it won't happen in the future. The committee report for UCA talks about making rollovers easier, it doesn't deal with the issue of eligibility for distributions. The pre-UCA impediments to rollovers were removed to allow retirement benefits to be maintained as retirement benefits (i.e., rolled over) when distributed from a plan at permissible times. I've got no snappy close to this discussion, it's just a matter of statutory construction. I think we may have to agree to disagree.

Welcome back Brooksdaletsa. . . . the 102nd Congress wanted to eliminate the Code's restrictions for distributions to be afforded rollover treatment. . . This is a correct statement and Congress did, in fact, make rollovers easier. As we have discussed before, however, a distribution has to be permitted before it can be made. The easing of rollover rules did not, and was not intended to liberalize distribution restrictions. What specifically are you being prohibited from doing? If you could indicate what your ultimate goal is with respect to your TSA, the assembled knowledge on this site might be able to fashion a solution for you. [For the benefit of others, this is a continutation of a discussion held on the 403(B) board under the title "Rollover transactions] [This message has been edited by david shipp (edited 03-06-99).]

For HCE determination purposes, attribution is determined under Sec. 318. Family attribution goes to an individual's spouse, children, grandchildren or parents. No age limits, no requirement that the attributee have any other ownership interest.

Esopwizard, assuming plan termination and transfer to the remaining plan, based on the info provided, is there an advantage to terminating one plan over the other? (or is there a bias showing here? :-))