My 2 cents

It probably does not matter what the default beneficiary provision was in 2010, because if it was amended since then by the prototype sponsor, the 2010 provisions are no longer valid or enforceable. Sponsors of IRA prototypes have the authority to amend them in a number of ways without the consent of the account owners, and an IRA prototype that did not point to the spouse (or, if no spouse, blood relatives based on rank, without prejudice) as the default, in preference to the estate when the owner of the IRA makes no explicit designation, would seem to be in need of amendment. As noted by at least one poster above, the estate would usually be a poor choice for a default. The IRA owner does not have the power to enforce superceded provisions, especially provisions that had not been amended in a way to remove rights that the owner had had. The owner in this instance always had the ability to designate the beneficiary to his or her exact specifications but apparently did not do so. One presumes that there was a mention of the change (if there was one) in one of the mailings to the IRA owner (perhaps the one with the privacy policy which, be it noted, virtually nobody ever reads). As noted by a previous post, it is almost impossible for the IRA owner to prove that no notice was given, especially if the IRA sponsor has a record of having sent notices out.

1. Benefit accruals and the calculations underlying the actuarial valuation are two separate things. Inclusion in the actuarial valuation does not lead to benefit accruals and exclusion from the actuarial valuation does not lead to there being no benefit accruals. The accruals are based solely on the plan provisions. Remember, as it is a defined benefit plan, money does not need to have been set aside in order for a participant to accrue benefits. 2. Is it not the case that, for a defined benefit plan, "valuation date" as defined in the plan document is relevant only for top-heavy testing purposes, and isn't it generally defined as the last day of the plan year? 3. A substantial percentage of defined benefit plan actuarial valuations are performed on the first day of the plan year (certainly, no plan covering more than 100 participants can do otherwise). Only people who are participants on the first day of the plan year are generally assumed to accrue a benefit furing the year (for purposes of calculating the target normal cost for the plan year). If a plan allows people to become participants during the plan year, they would actually accrue benefits, at the very least, with respect to service on or after their plan entry dates if not for all service since hire, whether there was recognition of them in the actuarial valuation or not.

Medicare is not a retirement program, and so it does not really fit in to allow employers to integrate based on the Medicare taxes they pay. Technically, integration was not there to make up for the taxes not paid on higher earnings but to give credit for the impact on Social Security retirement benefits of the taxes that were paid. But wasn't the original post about additional Medicare taxes owed by the highly paid employees? Integration has never factored in the FICA taxes paid by the employees, which is appropriate. The argument was always that the employer was paying for some of the Social Security retirement benefits on a limited basis, and integration was intended to allow the employer to equalize the total being paid by the employer (on a limited basis) for those above the limit (counting both Social Security benefits and the retirement plan benefits). Remember, integration is a tool for employers, and employees are not employers, so taxes paid by employees that are not also paid by employers are irrelevant when considering what should be allowed for integration. Does the employer owe any extra Medicare taxes for its highest-paid employees?

Does such a distinction serve a rational purpose? How many highly compensated salaried employees (the terms "highly compensated" and "salaried employees" being distinct and not at all necessarily both true or both false) would fail to make such an election if required, and how many would elect only 3%? Wouldn't it help for testing purposes to automatically bring in as many non-HCE salaried employees (at a default rate of 3%) as possible? Is it possible that notifying the employees of such a policy would have negative effects on morale?

I did not mean to suggest what might be required - as noted by several, it could depend on plan language. I did mean to suggest that taking a loan would usually be preferable to any kind of withdrawal, since it would not be necessary to gross the amount up for taxes (or have to deal with covering the taxes some other way).

My understanding: Under $5,000, OK to put proceeds into default IRA. Over $5,000, pay to PBGC as missing participants. Participants under $5,000 whose whereabouts are known but who do not respond = OK to put in default IRA. Ditto for such participants if a "diligent search" has been conducted (whatever that means in the Internet Age).

1. The ABCD does not report violations of the funding rules to the IRS. And the confidentiality rules they follow extend, not just to the person reporting the apparent violation, to protecting the reported actuary from exposure (until or unless there is a finding calling for public reprimand, suspension or expulsion). Considering the length of time the person has been practicing (EA for 38 years, so in practice surely at least 40), if push came to shove, it is not unlikely that the actuary would find it better to resign his or her enrollment and retire from practice rather than force direct action by the ABCD (especially if adapting to PPA rules was too difficult for him or her to have pulled off). 2. Under Precept 13, the first thing to do would be to try to discuss the issues with the other actuary, although it is unlikely that a satisfactory result would be reached if the other actuary has similarly misperformed actuarial services for so many other clients. 3. MAAA (at least) would give the ABCD jurisdiction. 4. Self-correction would not be available or appropriate for incorrect funding determinations. Failure to meet the minimum funding requirements would not jeopardize plan qualification. Self-correction is for administrative issues. Anyway, didn't the original post say that the plan has been cleaned up and brought to where it should have been? 5. If the IRS caught on to the actuary, judging from Rball4's post, it would be easy enough for them to round up a list of plans affected by his or her practices.

1. Granted, if the plan was subject to PPA and the enrolled actuary was failing to apply the PPA rules correctly to the extent described, that would be pretty awful. Aren't there any automatic error checks that would result in such a Schedule SB being rejected? 2. Aren't ABCD inquiries supposed to be confidential unless the upshot is a public reprimand or suspension/expulsion of the actuary? Wouldn't that mean that the ABCD is NOT supposed to pass along information that an enrolled actuary is not properly following the applicable laws and regulations to the IRS or the Joint Board? Remember that the ABCD is under the jurisdiction of the actuarial organizations, not the government. If anything, if the IRS or the Joint Board were to determine that an enrolled actuary was filing Schedule SBs that were thoroughly defective, the IRS or the Joint Board might report the enrolled actuary to the ABCD. 3. Has any consideration been given to the possibility that the plan in question was operating under a delayed PPA effective date? 4. Technical point: It is the plan administrator who would have been failing to meet the requirements for annual funding notices and benefit restrictions, not the enrolled actuary. While input from the enrolled actuary is necessary to properly administer those requirements and the plan's enrolled actuary should be taking an active hand in prodding the plan administrator with respect to those requirements to the extent necessary, supplying annual funding notices and applying benefit restrictions are administrative actions. 5. Has your friend tried to contact the other enrolled actuary about the apparent problems? I think that contacting the other actuary, if possible, about the issues involved would be appropriate before reporting them to the ABCD. The other actuary might have a suitable explanation or volunteer to take corrective action.

Offering more questions than answers: Is the participant able to take a non-taxable loan (aren't they all, until there is a default on the repayment schedule?)? That would appear to take precedence over either in-service withdrawals or hardship distributions. The idea that a 60 year old participant has to spend funds earmarked for retirement on medical expenses not covered by insurance makes me sad. How will they be able to accumulate enough funds to make up the difference at this point? If the participant is over 59 1/2, whether the amount is taken as in-service or hardship, wouldn't it be taxable the same (ordinary income, no early distribution penalty)? Would there be a 20% withholding on a hardship withdrawal? Without a 20% withholding, how would the participant be able to afford the taxes on the distribution when they come due, presumably next year? Assuming that the account holds more than is needed to cover the medical expenses, would the in-service withdrawal require that the entire account be distributed? Would the portion not needed for that purpose be eligible for rollover, to minimize the tax impact?

I am shocked, shocked, to hear that people with Roth IRAs ever cash them in before the later of 5 years or attainment of age 59.5! When I read the initial post, I naively presumed that the hypothetical 50 year was not going to withdraw anything until eligible to do so without penalty. If they do wait until then, they don't need basis information, do they? So to what extent should Roth IRA providers build their systems to accommodate people who will be cashing them in prematurely?

Maybe I am just confused, but why would you need a "basis" for a Roth IRA? Aren't the withdrawals exempt from taxes? I thought that one needs to know the basis for a distribution to be able to exclude the portion that represents return of basis from taxation.

Question: Was the CPA firm that used plan assets to buy a building the plan sponsor or does the CPA firm consider itself entitled to spend the assets of plans sponsored by other organizations? To what extent would replenishing the plan's cash by taking a loan backed an asset of the plan help the situation? Have they considered selling the building (they did buy the building for its market value, right?) to replenish the cash? Isn't there a rule that the assets of the plan cannot be used to further the interests of any party in interest? The decision to invest plan assets in purchasing a building had to be made as a fiduciary decision, and should not have taken into account the personal interests of the sponsor or any of its owners/officers.

Not claiming to be an expert on this, but unless the parents of husband and sister are involved in the ownership, wouldn't there be no attribution of ownership to or from a sibling? Isn't attribution limited to direct descent (grandparents-parents-children), with no branching off to siblings, aunts or uncles, nephews or nieces etc.? So (if I understand the ownership and company names correctly) attribution of MW ownership might be as high as 48.5% to the husband and wife with the other 52.5% ownership being considered unrelated. RW is owned 50% by husband, 50% by sister and there would be no attribution to either of the other's ownership - they would be considered independently. MCDC would be at most 40% husband and wife, with the sister and two others owning 60% between them, with no attribution from the husband and wife to the sister or vice versa. If this is wrong, I stand ready to be corrected.

And if fails the test, the plan must be amended to fix things. Can't take corrective action unless the plan explicitly permits it to be taken, which would usually mean adopting an amendment changing the definition of compensation, at least for that one year.

Since you mention 417(e) I presume that you are talking about a defined benefit plan. I am assuming that you are asking about benefits actually payable under a lump sum option, not non-discrimination testing. Even if you are asking about testing, my recollection is that one normalizes the entire benefit, not just the excess portion under the permitted disparity rules. I am having some trouble coming up with any interaction between permitted disparity and 417(e). The sequence would be that you would calculate the accrued benefit (payable at normal retirement age under the plan's normal form, which is usually a straight life annuity). You would then apply the plan's provisions concerning adjustments for payment commencing at a date other than normal retirement age (if applicable) and under payment options other than the normal form (including the QJSA). Unless the plan defines actuarial equivalence using 417(e) (which appears not to be the case here), I don't think 417(e) comes into play for these adjustments at all, except for payment options that are not non-decreasing benefits payable for at least the life of the participant (i.e., a lump sum option). Unless you are talking about benefits large enough to be affected by the Section 415 limitations with respect to lump sums, the 7.5% interest rate would have no impact (given the current level of 417(e) interest rates) on the amount payable as a lump sum (although 7.5% might come into play if you are performing general testing calculations under IRC Section 401(a)(4)). Irrespective of whether part of the benefit is attributable to permitted disparity, the lump sum equivalent of the entire benefit must be no less than the value of the normal form benefit determined using IRC Section 417(e) interest and mortality. To the best of my knowledge, even if one is talking about 415 limitations, there would be no distinction between the adjustments with respect to the portion of the benefit based on the excess portion under permitted disparity and the remainder of the benefit. All calculations aimed at converting a life annuity to a lump sum should use the same rates for the entire benefit.

If the change is from FAS 132 (keeping unrecognized items unrecognized) to FAS 158 (having the net financial impact of the plan equal the assets minus the PBO), wouldn't you pretty much wind up at the same place (current net impact = assets-PBO) whether one restated the past few years or just made the jump now? Sure, prior years might look very different, but even then this year's AOCI less this year's prepaid (or plus this year's accrued) would equal the PBO - the assets, just as they would if only the current year was to be affected. Still don't really understand why the accountant would have continued to follow 132 and not 158. Wouldn't that be something like continuing to fund the plan using pre-PPA rules?

Wish I could! Haven't seen much material on the proposal, but it appears to be intended to serve as a step into the world of saving for retirement for people with limited ability to save, and that if it converts to an IRA once the balance gets to $15K, wouldn't it be expected that any further retirement savings would be pursued through an IRA? I am somewhat haunted by the comment made earlier "If your friend who is a business owner can attract and retain good employees without offering a 401(k), she's doing a lot of other things very well." I can remember a time when business owners actually had to strive to attract and retain good employees. I can remember a time when new college graduates could find work with greater career potential than bicycle messenger or convenience store clerk. I can also remember banks and gas stations giving away toasters and drinking glasses and other goodies to attract customers. Guess I'm showing my age!

Granted - FAS 158 requires inclusion of unrecognized gains and losses etc. in Accumulated Other Comprehensive Income, so that the net effect of sponsoring the plan on the financial position of the employer equals the difference between the PBO and the assets, which was not how things were handled under FAS 87. Wouldn't it be the case that if a company had been accounting for the plan under FAS 87, when FAS 158 became effective (years ago!), the company's accounting would have recognized that change? While there have been situations where the sponsor had not accounted for the plan under FAS 87/132R/158 at all but then started to do so, I cannot recall seeing anyone having gotten stuck in the middle, following 87 and 132R but not 158 for any length of time after 158 became effective. Other than nomenclature, has anyone noticed any meaningful distinctions between FAS 158 and ASC 715? Wouldn't a proper FAS 158 report only require some rewording of the text to become a proper ASC 715 report?

On a couple of occasions, clients for whom we had not been providing accounting information in accordance with FASB would ask us to start providing them information in accordance with the current accounting standards. In general, we would usually wind up using a current transition date. I don't think accountants usually require the virtually impossible task of trying to recreate accounting calculations for the past 25-30 years. Granting that it is possible that we have failed to take note of some aspects of changes in the accounting standards, wouldn't FAS 87 be roughly the same as FAS 87 as modified by FAS 132R as modified by FAS 158 as reworked by ASC 715? Sure, there is more done with getting unrecognized values into the sponsor's financial reporting and some added disclosure information and some differences in presentation, aren't the items to calculate and their usage pretty much unchanged? In going from FAS87/132R/158 to ASC 715, is there any material change?

Re: Larry Bird comments How long did you remain surprised that he was so good a pro basketball player? One season, one game, one quarter? Most of the surprises with respect to his pro basketball career after a very short time in the league were in connection with some of the remarkable plays he made. That he would be a very successful pro became obvious very quickly.

I did not mean to imply that one could roll over an RMD. The clear answer is no. My point was that an RMD is not supposed to be rolled over - the recipient is supposed to pay taxes on it, and any attempts to roll the money into an IRA and treat it as not currently taxable would be an error, at best (at worst, tax fraud). The impression I got from the original post is that the recipient wanted to roll the RMD over and thus further defer taxes, and it they did, is there a red flag on the 1099-R that would quickly reveal to the IRS that the amount reported on the 1099-R was an RMD? The 1099-R would presumably show the distribution as a normal distribution that was fully taxable (since there does not seem to be a specific code for RMDs from a qualified plan or a non-Roth IRA). Any rollover would have to be after receipt, not direct since (one presumes) the administrator of the plan would be well aware that part or all of the distribution being made was an RMD and, as such, not eligible for rollover. Does the IRA provider have any obligation to determine that a 60-day rollover being submitted was not an RMD before issuing a reporting form showing it as a rollover? I do agree with Bird that there may be some acts that go undetected. Does the IRS cross-check dependent Social Security Numbers with the Social Security Administration to catch people who make up SSNs so they can declare their pets as dependents? If the IRS chooses to look into the series of transactions involving the RMD and the rolloever, there would seem to be a good chance that the only way it would fly would be if it was declared as taxable income and an IRA contribution, offsetting at least some of the taxable income (but on different lines of the 1040). If the IRS became convinced that the recipient, contrary to the rules, knowingly rolled the money over to try to keep from paying taxes due, it might not go so well.

It is my understanding that an RMD = amount that must be taken as a taxable distribution in a year, but if it was rolled over, the taxpayer would not be declaring any of it as taxable for the year of distribution and thus paying no taxes on the RMD. How could the IRS not care?

I may be confused, but wasn't the whole idea behind there being such things as RMDs to force people to take taxable distributions? How would deferring receipt of pretax funds and trying to satisfy RMDs by taking (non-taxable) withdrawals from Roth accounts satisfy the law's objectives? Wouldn't the law call for separate RMDs from a regular IRA or 401(k) account and from a Roth account? Are there RMDs under Roth accounts? If so, wouldn't he have to take an RMD from the Roth plan as well as from the pre-tax funds in the 401(k)?

Wouldn't that be more appropriately addressed at the time the distribution was being made, rather than the beginning of the following calendar year? Perhaps your question is more along the lines of "Is there anything in the way a 1099-R is coded to clue the IRS in to the fact that the distribution, rolled over by the recipient, was not eligible for rollover?" If a participant receives a RMD and tries to avoid current taxation by rolling it over after the fact to an IRA, perhaps it would only come to light upon audit. In a more perfect world, the plan administrator, in making a required minimum distribution to a participant, would be aware that it is not eligible for rollover, and the communications to the participant would make that clear.

If one wishes to allow retirees to cash out, one must be willing to deal with the likely anti-selection in order to obtain whatever benefits may be had from such a derisking strategy. If one excluded people who had spurned a lump sum earlier, wouldn't that pretty much foreclose any such action if the plan offers a lump sum option (since it would presumably have been declined by all retirees)? I continue to be somewhat leery of such offers (in my comments in the other thread, I expressed concern as to how effectively the pros and cons of taking the plan up on such an offer can be communicated to older retirees), but at this time, it would be cheaper to cash the benefits out than purchase annuities if the plan is to be terminated in the near future. There are also potential issues if the current spouse is not the same person as the former spouse standing to receive death benefits under a QJSA (possibly even under a QDRO). Probably need to get consent from both (from all of them if the original retirement benefit was split by a QDRO, with the alternate payee as joint annuitant for part and the (now ex) second spouse as joint annuitant for the rest of the original retirement benefit)!