MWeddell

Your way is permissible. Other permissible methods (in my opinion -- the regulations are quite sparse on these details): Test 1: Group A Test 2: Group B Test 3: Group C Test 4: Group D Test 1: All matching formulas at least as favorable as A, which is Groups A, B, C, and D Test 2: All matching formulas at least as favorable as B, which is Groups B, C, and D Test 3: All matching formulas at least as favorable as C, which is Group C only Test 4: All matching formulas at least as favorable as D, which is Group D only

I'd guess that it is 9 replies (plus the original post for a total of 10 posts) that switches the status to a red "hot topic" icon.

I also would lean against it. I would have a hard time interpreting "requiring the plan to pass ADP after the shift" to include failing the ADP but correcting the failure through reclassifying deferrals as catch up contributions.

Another negative may be data collection. If you normally have access to demographics for just one portion of your controlled group (such as you manage benefits for a subsidiary of a larger company that has separate plans for other subsidiaries), realize that someone needs to look at the data for the whole controlled group to determine what the top paid 20% group cutoff is each year.

http://benefitslink.com/boards/index.php?s...2&st=0& Catherine has started a separate thread on her question here, so it makes sense for further conversation on her topic to be posted at the other thread.

If the amount contributed to the 401(a) plan is unrelated to whether or how much one contributes to the 403(b) plan, then the presence of the 401(a) plan is irrelevant to determining whether the 403(b) plan is subject to ERISA. If instead the amount contributed to the 401(a) plan depends at least in part on how much was contributed to the 403(b) plan (e.g. there are matching contributions), then my impression is that many practitioners would say that one no longer fits the safe harbor for making sure the 403(b) is non-ERISA, that it is prudent to assume that the 403(b) plan is subject to ERISA. I do not believe that view is unanimous however.

I question that, Tom. It seems to me that a plan is a safe harbor plan if and only if it meets the requirements specified in the 401(k) regulations, one of which is that the plan document specify which safe harbor it meets. It seems to me that potentially (but it'd have to more extreme than issuing the notice a little bit less than 30 days in advance) a plan could state that it is a safe harbor plan but that the proper correction for not following the plan document and not meeting the safe harbor requirements is to run the ADP/ACP tests.

12 months from the last day of the plan year, which is 9/30/2010. It seems to me that the merger creates a short plan year for the plan that was merged out of existence. The situation is ambiguous enough that you might wait to read other posters' opinions though.

I'd still be inclined to err on the side of not charging fees on the existing loans. Just because the loan agreement or promissory note is silent on the issue does not mean that the plan may unilaterally decide to charge fees whenever it feels like doing so. Asking participants with existing loans to sign an addendum consenting to loan fees does not sound like a practical suggestion to me.

That is correct. Roth 401(k) rules (despite the similar name) are largely separate from the Roth IRA rules. You may be smoking something despite being correct though!

There were some proposed regulations issued in May of this year that would allow the 2009 safe harbor nonelective contribution to be suspended in the middle of a plan year if there is a substantial business hardship. Even with that rule, you still have to give 30 days notice. There is another rule that allows one to wait until 30 days from the end of the plan year before deciding to fund the 3% SH QNEC but the original participant notice would have had to be worded to take advantage of this "contingent" safe harbor plan design.

Yes. The 410(b)(6)© transition period does not prematurely end unless there are material changes to the plan's demographics or the plan provisions. With that citation, you can look up the Code Section and brief regulation enforcing it yourself if you like.

Yes. I assume you are talking about a plan with a calendar plan year and the amendment is effective for the 2009 plan year.

If one is in a situation where a plan is mandatorily disaggregated for testing purposes into 2 plans (e.g. union and nonunion employees), then it is possible for only part of the plan to be a QACA. Other than those fairly limited circumstances, the plan document must specify the ADP/ACP testing method for the entire plan.

I'd write the same comment regardless of the number of highly compensated employees. Try to interpret the document so that operational practice is consistent with the document. However, if a correction is needed, the Self Correction Program requires full correction. The fact that there is only 1 HCE helps to identify that the operational failure, if there is one at all, is insignificant and therefore eligible for self correction. One still must fully correct any operational failures.

You have to correct if the operational practice varies from what is specified in the written plan document. It could be that the plan document is vague enough to permit this. If the plan states that an eligible employee may elect to contribute a percentage of "Compensation" and "Compensation" is defined to include the 401(a)(17) limit, then perhaps the party responsible for interpreting the plan document could conclude that it is interpreted to mean the first dollars earned for the plan year and that deferrals should stop once the 401(a)(17) limit is reached. If there is no way to interpret the plan document to be consistent, then under the Self-Correctino Program (assuming you meet the conditions for it), you must completely correct the operational failure. There is no limit on how many years one must correct.

The original provisions were fine: all participants are eligible at least as soon as permitted by IRC 410(a).

Another reason that age 59½ is selected for the stated age upon which vested employer contributions may be withdrawn is because by law the employee pre-tax deferrals cannot be withdrawn until that age. Picking age 59½ allows one to coordinate the withdrawal options more smoothly for all types of contributions.

Assuming the notice said that HCEs (in addition to NHCEs) were receiving the safe harbor match, then you could not give HCEs the match only 30 days after a new notice was issued. I don't see how you're allowed to retroactive take away all of the 2009 safe harbor match for HCEs.

I don't see a problem with adding the unwind withdrawal feature in the middle of a plan year. Is that really right that one can't create an EACA in the middle of the plan year? I understand that becoming a safe harbor plan must be done as of the first day of the plan year, which would cover your statement about QACAs.

I finally read the TIME article and thought that it was filled with misleading numerical comparisons. I think that the article's main point -- that 401(k) plans often don't generate sufficient retirement savings -- has some validity but this article doesn't inform us at all (in my opinion) of the magnitude of the problem.

Here is a citation: IRS Revenue Ruling 2004-12. It holds the following: So it depends on whether the plan document imposes any restrictions in in-service withdrawals or other distributions of rollover contribution accounts.

I don't work with SIMPLE plans, but is that allowed? If one has a regular 401(k) plan, can one then sponsor a SIMPLE IRA plan? I thought they were for employers who have never sponsored a qualified plan.

The definition of "other right or feature" in Treas. Reg. 1.411(d)-4 makes clear that the allocation conditions for receiving a contribution are only protected benefits that can't be amended retroactively once the allocation conditions are fulfilled. Hence, I agree with the prior posts. Beware that many plans contain exceptions to the "employed on the last day of the plan year" rule for those who die, become disabled, retire, etc., so the amendment couldn't take the match away from those employees. Lastly, if the match is completely discretionary anyway, you probably don't need a plan amendment at all. The employer may decide to just not fund any matching contribution.

I would add that if you do use the plan's definition of compensation, the one that excludes bonuses, for ADP / ACP testing, then you'll have to make sure that it is a 414(s) definition of compensation, which sometimes requires a 414(s) test.