zimbo

I am struggling with a lump sum distribution methodology when calculated in partial years. I have a terminee who is age 47 and 7 months at the assumed 2/1/08 distribution date. The plan's NRA is 65. Plan does not discount pre-ret mort in PVAB. Do I use some sort of interpolated Ns when calculating my deferred temporary and deferred life annuities at the 2nd and 3rd segment rates, or can I simply interpolate the lump sums determined at whole ages 47 and 48, or in this new PPA world do I only use whole ages (nearest age or attained age)?

At the ASPPA meeting back in October, we were told that the IRS intends to issues regulatory "relief" of some sort to allow End of Year valuation dates to be workable under PPA. However, they would and could not issue such regs until the Technical Corrections Bill is passed since that Bill contained the authorization for the IRS to issue these regs. I still have not given up on my small plan EOY vals. Also, it was thought that the technical corrections bill might not be passed until early 2008.

In reviewing this Revenue Procedure I see that it is explicit in defining what it pertains to and clearly it does NOT mention 412 or 404. It does use the term "Fair Market Value" and it was that term that I was recalling without rechecking the references. Thanks for researching that. I guess I'm lucky I continued to use straight CVs. We also generally use the full cash accumulation value, rather than the cash surrender value in our valuations. I guess if one wanted to factor in the surrender probability in some acceptable fashion, one could possibly use the guidance in the 2005 revenue procedure where an average surrender rate over the next 10 years is factored in. But I would rather choose not to go there.

I think that the Revenue Ruling from 2005 is supposed to be used to determine market value of life insurance and that probably includes for minimum funding and maximum deductible limits as well. Admittedly, I believe that most administrators and actuaries have continued to use cash value or cash surrender value in their valuations, but if you check the ruling you will see that determining market value is far more complicated than that.

In reading PPA and then reading the QDIA Regs and the Auto Enrollment proposed Regs, some questions arise. Note that EACA= Eligible automatic contribution arrangement and QACA=Qualified Automatic Contribution Arrangement: 1. To take advantage of the ERISA preemption over state laws, and the 90 days distribution rules, and the 6 month failed ADP Test timing, a plan must meet the EACA rules, correct? 2. Are all QACAs also EACAs? Why would anyone have a QACA that is not a EACA? 3. Must a EACA use a QDIA (qualified default investment alternative)? If so, doesn't that effectively mean that all auto enrollment plans MUST abide by the QDIA requirements? 4. While it is true that a plan may not take distribution fees from a QDIA in the first 90 days after auto enrollment for those getting distributions upon "reversing" their enrollment, the auto enrollment proposed regs seem to allow distribution fees in exactly the same scenario. Does that mean that a plan CAN take distr fees if the auto enrollee had not used the default investment but instead had chosen a particular fund(s) after his enrollment? MANY THANKS

My only response is that under the regs, you cannot aggregate a safe harbor plan with a non safe harbor plan. But in 2008 if your plan became a safe harbor, I believe it could be aggregated with one or more of the safe harbor plans in the controlled group and perhaps together they could pass coverage testing.

I believe that tax filings are exempt from this law, so my concern is for whether our "normal" plan administration and/or recordkeeping could be affected by this and similar laws. Don't disagree with your comments, but I would love to see what the lawyers think. Stay tuned.

I like your answer but I'm not entirely persuaded. In theory, unique identifiers other than SSN could be developed without impeding the orderly administration of retirement plans. Is your opinion the general consensus among people who have studied these laws?

Back in 2006, New York State passed a law effective 1/1/2008 which restricts the use of Social Security Numbers by employers for many purposes. My understanding is that 6 or 7 more states, including Florida and California, have or are considering adopting similar legislation. Does anyone know whether this affects our operations as administrators, actuaries, and TPAs? For example, many plan administration systems use the SSN as an identifier for an employee or as an initial password for a participant's web access. Many daily administration systems such as Relius use these numbers for transmission and allocation of periodic contributions by payroll as well as for distributions and other purposes. Would these features run afoul of the new laws? There appears to be some relief when encryption is used. What about the transmission of employee census data requests and information received by e mail? By regular mail? Has anyone read or heard any commentary about these laws and whether or not they potentially affect our practice areas?

I don't believe this sort of amendment would be allowable under 411(d)(6). The basic principal is that by changing the allocation provision after the allocation date (in the case of your plan, this is the last day of the plan year), the allocation of the contribution (whether made during the year or after the year) would be lower for some, higher for others. Those getting a lower allocation would be violating 411(d)(6). I'm not sure how you would consider this to be an 11(g) corrective amendment under 410 or 401(a)(4) since there was no failure to begin with.

Except that it is not the benefit that was elected. The benefit elected is a lump sum and the maximum annual amount payable for a temporary period of time is that amount not in excess of a life annuity. As soon as the funding situation resolves, a lump sum is payable. Would your opinion change if the benefit elected was defined as 8 such installments followed by the actuarial balance payable in a lump sum in the 9th year (all of course subject to IRS limitations and restrictions)? Now the form of benefit is a series of installments payable over less than 10 years. If not, then why not?

I have a plan where the HCEs cannot receive a lump sum because the plan does not meet the 110% "Solvency" Test. An HCE has terminated and has elected to receive a lump sum but due to the restrictions he is receiving an annual installment equal to his actuarially equivalent life annuity. At such time as the plan meets the 110% rule (for purposes of this question, ignore the PPA lump sum restriction soon to go into effect), he will take a lump sum of the "actuarial" balance of his lump sum. I believe that the annual installment can be rolled over into an IRA. Admittedly this is a gray area. My logic is that he has not elected an annuity or a 10+ year installment payout. He has elected a lump sum and is settling for the greatest allowable annual installment until he can take the balance of his lump sum. Therefore, this is a periodic payment that qualifies for rollover. Does anyone agree with this position or, if not, why? Has anyone heard of this issue being addressed by IRS in meetings or other forums?

I seem to recall an IRS ruling that allowed this 25% limit on transferring excess DB assets to a Qualified Replacement Plan to be as high as 100%. This could allow you to siphon off more of the excess. The other rules pertaining to allocating over no more than 7 years are still in effect.

A rollover from a 403(b) to a 401(k) can be done (assuming there is a distributable event to receive the distribution from the 403(b)). A transfer from a 403(b) to a 401(k) cannot be done

I would be a bit leery of hanging my hat on a Revenue Ruling from 1953, although it can be valid. It just seems odd that a Defined Benefit Plan provides benefits based upon each individual participant's investment choices. This is a hybrid plan taken to it's ultimate conclusion and I believe it becomes more DC than DB. Also, I suspect that the market rate of return under PPA will be defined on a pooled basis and, if so, won't that render these extreme hybrids unworkable? Post PPA, you also have the issue of adverse selection whereby the participant can never lose money from one year to the next thereby encouraging aggressive investment choices. Also there are funding challenges since the CB accounts can never lose but plan investments (which might be mirroring these accounts) can lose. Also, should a plan terminate, I believe that the cash value benefits need to be recast based on a 5 year average return. How in the world would that work with a participant directed cash balance plan?

I have seen a few of these plans in action. There are a few firms I know of that aggressively market them. There is at least one major employer with many thousands of employees whose plan uses this concept. My current personal opinion is that they are not valid pre or post PPA. That is because a participant's ultimate benefit level is at least partially dependent upon the performance of the funds that the participant chooses. Thus you no longer have a definitely determinable defined benefit plan. Even when PPA defined market rate of return, it seems to me that the fact that the benefits are dependent upon investment performance chosen by the participant invalidates it as a Defined Benefit Plan. Finally, even if it were legal, how would we handle a decrease in account balance when a fund portfolio chosen by a participant loses money causing a decrease in the year to year value of his cash balance account. How does that NOT violate 411, especially pre PPA? Having said all of that, I know of no current or past IRS rulings one way or the other.

If a plan is effective 1/1/2002 and freezes benefit accruals as of 12/31/2006 (but is NOT terminated) and terminates as of 12/31/2007, what is the fractional reduction in the 415 dollar limit? Is it 5/10 or 6/10?

We have a 401K Plan with participant investment direction that is valued quarterly using "balance forward" calculation methodology. A bit old fashioned, but there are still some of these left. Would you agree that we need to add to the quarterly statement by describing the "investment limitations" i.e., that they can only change their investment allocations once per quarter and that any such change will be effective as of the first day of the following quarter? Or do you think I am reading too much into this PPA requirement?

I would add that this reg section only applies for determining which limitation year the contribution falls into. So, for a not for profit with relatively small allocations, it would not really matter if the contributions were made after that deadline. That is because the worst that might happen, as I see it, is that those allocations would be credited to the next limitation year and then if the next year's contributions WERE made within the deadline, you would effectively have doubled up for 415 purposes. Well, if the amount that you are doubling up is low enough, then you still would not have any 415 issues in most cases.

I agree with the Otherwise Excludable exception. Good catch. I was sort of giving the textbook answer rather than the better more practical response. Just make sure that your document makes direct or at least indirect reference to this "otherwise excludable option" if it describes the gateway contribution. Also to answer your last point, I believe the TH minimum is based on the entire year's comp regardless of the participant's entry date within the year and also regardless of the Otherwise Excludable option.

Assuming the plan is top heavy, the new entrant with less than 1 year of service will end up getting the 5% gateway because the TH minimum gets him to 3% and then he is considered "benefitting" in the plan and so is subject to the gateway which raises him to 5%. No additional contribution is required. Since that person is in the plan and benefitting, I would count comp towards the 25% limit.

I see your point in that the 401(a)(4) mentioned in that section only pertains if the matches don't meet the requirements of (a)(4)(iii). But, continue on in that reg section. Go to 1.401(m)-2(a)(5)(v). Doesn't that seems to say that matches that are fofeited are not taken into account in the ACP Test?

Thanks for your input. I still stand by my theory that you must first forfeit the impermissible match (due to a failed ADP TEST refund), then do your ACP using the "correct" match. My basis is 1.401(m)-2(a)(5)(i) which tells us not to include matching contributions that do not meet 401(a)(4) in the ADP Test. Perhaps there is more than one way to skin this cat?

Would be interested in opinions regarding the correct "ordering" for a plan that fails an ADP Test which then results in the need to forfeit some matching contributions to the HCEs who get refunds of deferrals due to the failed test. The real question is how to properly run the ACP test in that instance. It seems that some folks believe that you run the ACP Test without netting out any forfeited match and if it fails and if the refunds due to the failure exceed the amount of match to be forfeited due to the ADP Test refunds, then you are done. I believe that, as per the 401k regs, you must run the ACP Test net of the forfeited match and then, if it fails, make your refunds as necessary. The difference is in the amount of match that the HCE will ultimately receive either in the plan or as a taxable refund. I also believe that if the match has not yet been made at the time the ADP test is run, you must calculate the employer match based on the deferrals NET of any refund required due to a failed ADP Test. Do others agree with this procedure?