MGB

In what context are you using "benefit commencement date?" "Annuity starting date" has legal definitions and shouldn't be used in other contexts to mean something other than the legal definition. It sounds like you are using benefit commencement date as something else, like a more generic term? Or, is this coming from some rule that you need a definition for?

Even though my title is "Director of Employee Benefits Research", my research unit can't take any credit for this. The SOA put out an RFP in 2001 for someone to do this. The winning RFP was from an investment consulting firm in San Francisco. Since then, Milliman USA bought the firm, so the copyright you see is ours, but my unit isn't involved in putting together this information.

RSNOW: It is irrelevant whether there is a prepaid or accrued pension cost on the balance sheet; that has nothing to do with whether or not an additional minimum liability is triggered. As was already stated, the "minimum liability" that must be shown on the books is just the (ABO - assets), but not less than zero. If there already is EITHER a prepaid or an accrued expense already on the books, then the "additional minimum liability" is the difference between the minimum liability and the accrued or prepaid. If there is an accrued and it is larger than the minimum liability, no adjustment is necessary; there is already enough of a recognition covering the underfunding (this is typically the case with unfunded nonqualified plans). When there is a prepaid on the books, the additional minimum liability is larger than the minimum liability because the prepaid was the opposite sign before netting them together. Note some important points that many don't realize: 1. Once the assets>ABO again, the additional minimum liability goes away. If there had been a prepaid before recognizing an additional minimum liability, it will reappear on the books (net of any net periodic pension cost in the intervening years). 2. There are two types of accounting regimes for income statements: one is based on "comprehensive income"; the other is based on "net income". For companies that use net income (most people are familiar with this approach), the changes in additional minimum liability each year do not go through the income statement and instead become an item in equity called "other comprehensive income." However, there are two situations that use comprehensive income. In that case, there is no such account as "other comprehensive income" and all changes in the additional minimum liability must go through the income statement as an expense. These two situations are: - Nonprofit organizations under GAAP accounting. - Insurance companies and HMOs reporting to state insurance departments under statutory accounting (Statement of Statutory Accounting Principles #8 covers pensions). (However, they are trying to change how this works, and last year some states allowed a divergence from this rule as a permitted practice.) Note that any reversal of the additional minimum liability under the comprehensive income regime becomes income in the year that it happens.

It is my understanding that it still must be deducted (in whichever year is appropriate), regardless of whether or not there is a zero or negative Schedule C income. You cannot carry over the deduction to a different year. However, the Schedule C loss may be carried over in certain circumstances.

Only 12 pages??? You must use very small fonts.

Yes, REA changed the 25 to 21, effective plan years after 12/31/84. There are also special rules for educational institutions: If the benefit is 100% vested after one year of service, then minimum age is 26. For pre-REA, the minimum age in this situation was 30. Special transition rules applied to both of these when changed.

The unrounded number for 2003 is 40,636, which is rounded down to the next 1000. That is 40,000. Given the CPI to date (through August, only September and October are unknown and will not affect these projections unless significant, in multiple percents, deflation or inflation occurs), the numbers for 2004 are: 401(a)(17) 205,000 415(b)(1)(A) 165,000 415©(1)(A) 41,000 414(q)(1)(B) 90,000 (no change)

By the way, for those that don't know my full name (Mark Beilke), I am doing the session with Ed Burrows on the future outlook for changes in funding (Ed) and accounting (me) requirements.

You are converting correctly, but not doing the CL correctly. The plan must define what the annuity amount is and then it is valued. The cash balance is irrelevant to the CL calculation other than to help define annuity amounts at various times in the future. Note question 4 from the 2003 EA Meeting. You use the plan's mortality for the CL calculation (from annuity back to date of decrement) if your assumption in the regular valuation is that everyone will take a lump sum at decrement. You use CL mortality prior to that when active (if applicable). Gray Book2003 -- Q&A-4 Funding: Application of Notice 90-11 to RPA ’94 Current Liability __________________________________________________________________ Copyright © 2003, Enrolled Actuaries Meeting All rights reserved by Enrolled Actuaries Meeting. Permission is granted to print or otherwise reproduce a limited number of copies of the material on the diskette for personal, internal, classroom, or other instructional use, on the condition that the foregoing copyright notice is used so as to give reasonable notice of the copyright of the Enrolled Actuaries Meeting. This consent for free limited copying without prior consent of the Enrolled Actuaries Meeting does not extend to making copies for general distribution, for advertising or promotional purposes, for inclusion in new collective works, or for sale or resale. QUESTION 4 Funding: Application of Notice 90-11 to RPA ’94 Current Liability When calculating current liability, Notice 90-11 requires the use of a qualifying current liability interest rate for purposes of calculating benefits in a form other than a non-decreasing life annuity. For example, assume a plan allows lump sum payments and the valuation assumes that participants take a lump sum upon retirement. For purposes of calculating the current liability the lump sum must be determined using the current liability interest rate – regardless of the rate specified in the plan for converting the plan’s annuity benefit into a lump sum. RPA ’94 added a required mortality table to the current liability calculations – currently the 1983 GAM male and female tables. For purposes of calculating the RPA ’94 current liability, must this required mortality table be used to determine the benefit in a form other than a non-decreasing life annuity? In the example above, would the lump sum valued in the RPA ’94 current liability be calculated using the current liability interest rate and the sex-distinct 1983 GAM table? RESPONSE No. The interest rate requirement in Notice 90-11 does not extend to the required mortality assumption, so that the lump sum valued in the RPA ’94 current liability should be based on the mortality table specified in the plan to convert the annuity benefit into a lump sum. Thus, the 417(e) unisex mortality table must be used for this purpose.

Under 412 and 404, it is possible to contribute to 412 in 2002 and deduct under 404 for 2003, so that you will not incur a carryover. In fact, as you state at the beginning, this is a classic EA-2 question. HOWEVER, there is a non-pension tax issue that most actuaries are unaware of. That is known as a "tax accounting method." You are not allowed to change your tax accounting method without prior approval from the IRS. In the normal course of changing a tax accounting method, they make you go back to prior years and refigure your taxes, as if you had always used the new method (this is similar to the approach for changes in GAAP accounting methods under APB 20). I doubt a recalculation would apply in this situation. In the past, I assume they have always deducted in X the contribution made in X+1 that has been designated for the X year. If you change this pattern, you have changed the tax accounting method. Of course, then you are in the position of probably wanting to change back to the old method in the following years, which creates a new issue. If you do not go through the filing for a tax accounting method change, you could very well be open for a problem under an audit. Of course, the only problem would be that the contribution gets reallocated to X and the 10% excise tax applies for one year and deducted in X+1. So, the only difference is the excise tax (plus interest and penalties if this is paid late). There are PLRs on this subject (most recent ones have focused on contributions to multiemployer plans).

As stated earlier, the key to applying a calculation one way or the other is the exact, specific language in the plan. Here is Holland's description of this issue (it is in the context of early retirement factors, but equally applies to any other calculation), from the 2002 EA Meeting, "Dialogue With the IRS." "MR. SEGAL: Okay. I'm going to ask a question that was sent in. The plan formula provides for a benefit of $200,000 payable at normal retirement age 65. The formula could be, for example, 100 percent of pay. Can the plan provide a payment of $160,000 at age 62? Let us assume the early retirement reductions were one-fifteenth/one-thirtieth. MR. WELLER: This is one of the questions that Mr. Holland always likes to answer, which is are you going to be able to withstand an ERISA lawsuit from the participant who says "The plan formula is $200,000. I want my plan formula. I don't care if your plan is disqualified for paying me my promised benefit." MR. HOLLAND: The long answer is that, first of all, we all understand you can't write a plan that will give you $200,000 at 65. Now the trick here is... MR. WELLER: You can't write a qualified plan. MR. HOLLAND: I'm making the assumption they want a qualified plan, but Mr. Weller is correct. But you can write...the trick in this question is the words "plan formula." Now the bottom line is you have to pay very close attention to your drafting. You can write a plan that says the benefit at 65 is $200,000 under some plan formula, but in no event can this benefit exceed what is allowed by IRC section 415. So the benefit at age 65 is really then the maximum allowed by Section 415. You will clearly have to do that to be qualified. Now let's go to early retirement. Now you're really talking about what is the interaction of this benefit formula with the section 415 limits? How do these two parts of the plan look together? There are two ways to write a plan. Let's just take this one-fifteenth/one-thirtieth early retirement reduction. There are two ways that I can write it. One, I could say I will take the plan benefit at age 65 and reduce it by one-fifteenth/one-thirtieth. That would mean I would take the section 415 limit at age 65 and reduce it by one-fifteenth/one-thirtieth. Another way to write your document is to have your early retirement benefit be the benefit described under Section X of the plan that has $200,000 in it, reduce that by one-fifteenth or one-thirtieth and then further constrain that to say this benefit cannot exceed what is allowed by Section 415. In essence then you would have a subsidized early retirement benefit versus what's payable at normal retirement age. That can be done and, in fact, in one of the first talks I ever gave on Section 415 probably almost 20 years ago, I did an illustration of how that could be done. I think it was ASPA in 1983. It was almost 20 years ago. Anyway, you really have to pay very careful attention to drafting. What you have to make sure is that your accrued benefit at any point in time is limited to what can be provided by Section 415. Otherwise you will run into the situation where you can't forfeit the benefit in order to meet Section 415 and you can't pay the benefit because it violates Section 415. So the trick is it's very careful drafting and this is something that would have to be done under plan terms. I've seen some people able to do it and I've seen that some people have not been able to do it. "

Bill and VTM, Surveys do not determine reaonability. What is reasonable for one plan may not be reasonable for another. Don't fall into the "lemming" trap of analysis like many others do. For discount rates, the "duration" of the liabilities is the key issue. If the plan is a "legacy" plan with primarily older, retired annuitants, the duration is very short. In looking at a yield curve, the discount rate should be low. However, if it is a primarily young, active participants, then the duration is very long and the discount rate should be at the high end of the yield curve. For a published yield curve that is designed to be used for this purpose (it has under 2% at short durations, up to almost 7% at the longest duration), see: http://www.soa.org/sections/pendis.html For expected return on assets, the key issue is asset mix. If the plan is invested primarily in money market funds, the expected return should be extremely low. On the other hand, with a high concentration of stocks, it should be fairly high. Note that the expected return is NOT what is expected in the next year...it is the annual yield expected over the long run, given the current asset mix. Depending on who the actuary is for the plan, intricate computer models are available for this analysis. Substantiating the selection of return based on these models is a common thing in the national actuarial firms, but much less common by local or regional actuaries.

Frank, correct. irc, The paragraph you are focusing on can only happen with a reemployment, not a continued working situation. This is describing what can be suspended (i.e., forfeited) when a person takes a subsidized early retirement benefit and later returns to work. It basically is saying the extra subsidy is suspendible and need not get an actuarial increase for later retirement. But, the actual normal retirement benefit underlying that cannot be suspended unless proper notice is given. There have been a couple of recent (last 2-3 years) court cases that also clarify this ER subsidy issue.

The discussion and reasoning of various issues was much more enlightening and interesting than the actual decisions. For example, they decided that there is no need to disclose market-related value of assets because if you take the expected return and divided it by the assumed rate of return, you can back into the numer. So much for transparency (which is what this project started out as). The biggest eye-opener was when the board literally threw a fit when the staff mentioned that actuaries would have to do some reprogramming of their systems to get projected cash flow of benefit payments for 10 or 20 years. They harped on this for about 10 minutes that anyone that uses a PV factor to apply to a benefit to get the PBO (or value at decrement), is not following SFAS 87. They were really upset to think anyone uses factors and not complete cash flow projections for determining liabilities. So, the need to reprogram systems will not be an excuse for not being able to report this information on 12/31/03. Additional decisions on disclosure of: Nothing new on multiemployer. No separate threshold requirement on OPEB -- same rules apply. No measurement date (they couldn't understand why anyone would want to know this); but if significant changes had occurred since the measurement date, then stating the measurement date would be a part of disclosing the significant changes. No breakdown by line item of where the NPPC ends up in the income statement (although a separate project due next year on performance measurement would break up the service cost and other items into operating income and financing cost). Forecast of benefit payments (maybe as long as 10, 20, or more years). Possible deletion of the reconciliation of assets and obligations from year to year, but asking in ED what items from the reconciliation should be kept (e.g., actual benefit payments and contributions for past year). Expected contributions for next fiscal year, broken down by minimum required and expected discretionary. If expected contributioins include items other than cash (e.g., the NWA situation with subsidiary stock), a disclosure as to how the value was calculated. Tabular presentation of the assumptions, breaking out which apply to the cost for the year and which apply to the end of year disclosure. Quarterly reporting of NPPC (this does not mean it is recalculated each quarter -- what it does mean is that the expected return on assets must be set at the beginning of the year, which it always was supposed to be anyway). The effective date will absolutely be for disclosures at the end of the calendar year. Note that there have been many, many more decisions besides these in earlier meetings, such as a breakdown of assets into broad categories with the expected return of each (the weighted average total must equal what is used for cost). Exposure Draft expected mid-September with an extremely short comment period and final statement applicable by year-end.

Disability payments based on age or service (e.g., tied to the accrued benefit) do not get any preferential tax treatment. If the disability benefit is purely based on salary, then it is possible to get tax relief under Section 104 of the Code. See PLR 200152017

I think this definitely requires a blackout notice. However, this would fall under the exception to the timing because it was an unforeseeable timing (up until the TPA pulled the plug, it wasn't for sure it would happen). The notice needs to be sent as soon as practicable once it happens.

mal, It sounds like you, or whoever is providing you this information, is mixing two different subjects together. The contribution rate (4.00/hr or 20 cents/hour) and the calculation of the benefit amount (90 per year of service) probably are not linked in anyway whatsoever. Dividing one by the other should not have anything to do with calculating a year of service. There should be a completely separate definition of what constitutes a year of service that is based on hours worked. Example: Let's say it is defined as 2200 hours in a calendar year. A person works exactly 2200 hours and there is a contribution of 20 cents/hour or 440 dollars. For this $440 they receive 90 per month at normal retirment age for life. There is no connection between the 90 per month at retirement and the 20 cents.

I usually listen to/attend all of the Board meetings, but was at a separate meeting yesterday trying to counter the Administration's proposal to rewrite the funding rules. You can listen to the actual Board meeting (75 cents a minute) up to 48 hours later, and don't need to listen to any subject other than the pensions (the details are on the FASB site). I plan to listen to it later today. BNA didn't give much detail this morning, only reporting that they decided no new information on multiemployer plans and that whether or not to apply all of the new rules to post-retirement medical should be left up to general materiality guidelines. It also said they are shooting for a mid-September release of the Exposure Draft and finalized and effective date prior to the end of 2003. The intriguing part (and why I want to listen) of the article is "The board took a last opportunity to revisit some of the decisions it had made previously on the project and reaffirmed nearly everything, but with some subtle changes, Proestakes noted." Given that the major changes are these earlier decisions, any change could be huge. Will know more by tonight.

The DOL said in an Opinion Letter in 1978 (I gave the exact reference in an earlier thread) that an SAR is required.

A change in the person signing the Schedule B is a change in enrolled actuary, period. It doesn't make any difference is they are in the same firm or not. The confusion arose (and is still out there) because of cryptic language in the instructions for 5500, Schedule C that says: "in case the service provider is not an individual, report when the service provider (not the individual) has been terminated." In order to properly understand this statement, one needs to realize that the "service provider" of an enrolled actuary is ALWAYS an individual. The firm that employs the enrolled actuary is NOT the service provider, the individual is. The firm is not the signator of the Schedule B, the individual is. On the other hand, the service provider of an audit is not the individual, it is the audit firm. An audit report is signed by the firm, not the individual. So, the statement in the instructions about changing individuals within the service provider can only apply to auditors. Jim Holland and others within the IRS also take the above stance. The Academy has repeatedly requested clearer rewording of this, but it hasn't happened yet. In your (1), this is a change in the service provider that must be reported on Schedule C. In your (1b), this depends on who signs the Schedule B after the shift. If the signature changes, it is a change; it is as simple as that. (Or, given that people's signatures can change, e.g., marriage, or for lack of writing for months while typing on a computer, the real issue here is "has the enrollment number changed that is listed on the Schedule B?") In your (2), this would not be a change in service provider.

I would add to the 401(k) side: Depending on the plan and its selection of investment options, you could get access to professional asset management/diversification at a very low cost. If the person is not an investment professional, making their own investment decisions in an IRA can be disasterous. Of course, they can pay for professional advice, but at this level of money, I doubt that it is worth it. "A Roth IRA is better yet, because your money can grow tax-free. Tax-free growth will almost certainly be better than tax-deferred growth, even if your initial contributions are tax-deductable, and this advantage will increase with time and with greater growth." This statement (from another post) is not universal. A person's individual situation (age, expected date of starting distributions, tax brackets now and expected in the future, etc.) must be modelled to determine which is better. And, although it usually isn't an issue, if the person has financial difficulties, there are different rules on protection of assets from creditors in bankruptcy between the different options.

I can't imagine why anyone would write a document in the second way, nor have I ever seen a document that does it that way. (I only work with individually designed plans, so I don't know what standard volume submitter language is.) The main thing here is "what does the document say?" It ought to say that the accrued benefit is limited to the 415 limit, not some projected benefit that has no relation to any rule in the law, accrual, benefit limits, etc.

k man, You are correct. No one ever has to file and if you do, you can file at any time. HOWEVER, if you have a problem with any language in your plan, you can only amend it prospectively. That means if that language was there during a past period, you have no ability to retroactively amend the plan and clean it up (other than through a correction program and pay the bucks), so you basically have a plan that is subject to disqualification retroactively if you haven't filed for a letter. That is where the remedial amendment period (RAP) comes in. When you are forced to amend a plan (e.g., a change in the law), each law contains a specific RAP time frame. Within that time frame, you can retroactively amend the plan and keep it qualified as long as the required amendments are due to the law changes that created the current RAP. AND, if you file for a determination letter, your RAP extends until the receipt of the determination letter. That allows the IRS to come back to you and ask for changes and you can still make the language changes and apply them retroactively. For the amendments required for GUST, that RAP is now running out for volume submitter plans. If they do not file before the RAP period is up, then any bad language cannot be corrected retroactively. They are considering extending the filing deadline under which you can still file for a determination letter and be able to maintain the retroactive application of any required changes in language. Now there has also been a new change in rules. Since EGTRRA, you can no longer wait until the end of the RAP for EGTRRA (and any new laws going forward) to be able to make the amendments. Now, there must be a good-faith amendment in the year a provision becomes applicable. There is still an EGTRRA RAP (I think it is the end of 2006) that applies to be able to clean up any final language based on guidance that is issued in the meantime.

They ought to also be aware of Code Section 4958, which imposes significant excise taxes on "excess benefit transactions" (i.e., total compensation packages for nonprofit executives). The connotation of the word "excess" here is "excessive" in relation to the rest of the world. These provisions are in the law to curb the abuses of exactly this kind of mentality in the nonprofit world.

The argument is extremely simple and based on an ambiguity that results in two opposite opinions. The problem is with the definition of "rate of accrual" because it is undefined in the law. If you believe that the only way to measure rate of accrual is in the context of an annuity payable at normal retirement age, then cash balance plans are age discriminatory. On the other hand, if you are allowed to define the rate of accrual as the addition to the notional account (i.e., the current value), then cash balance plans are not age discriminatory. It all comes down to that argument. There is no need for checking assumptions or methodologies used by this judge. Obviously, this judge decided the only way to measure it is the annuity at retirement age. The problem is that the law does not define rate of accrual, so the courts are left to making their own decision. The Treasury Department, as shown by their proposed regulations, believes that rate of accrual can be on the current value. And, except for a small handful of very vocal dissenters, most in Congress also feel that way. Note that in applying the ADEA to other benefit plans (e.g., health or life), there has always been the rule that discrimination may be measured in either one of two ways: either the cost of the benefit must be the same or the benefit itself must be the same. Using this logic (which shouldn't be used on a pension plan, but is being used here as an analogy), cash balance plans are not age discriminatory because they pass on the cost measurement.