MGB

It did not "replace" it. EGTRRA took some provisions from HR10 and some from the Senate pension reform bill. Although there were many similar provisions in each, there were many differences. About one-third of HR10 did not get into EGTRRA due to procedural issues. Only items that were specifically tax-related (in the eyes of the committee) were able to be put into this "tax reconciliation" bill. So, provisions such as reductions in PBGC premiums, easing of separate line of business rules, posting of SAR's electonically, etc. are still out there, unpassed, and are expected to be introduced as a mini-HR10 very soon. Current talk is to try and attach it to the minimum wage bill expected to be introduced later this summer.

One technical change to stephen's is the order of logic: The limit would be $40,000 or 100% of pay if less, plus $1,000. (So, there can be more than 100% of pay.) The catch up contributions are not subject to ANY limitation, testing, etc., including 415. However, elective deferrals are subject to a 100% of compensation limit themselves. So, if the person's compensation is $12,000 or less, then the maximum is 100% of compensation. (Andy's reference to $500 is applicable to SIMPLE plans only.)

This will depend on the how well this is described in the contract. Presumably, this is there...if they are just applying this as a rule without it being in the contract, you have something to fight over. If it was clearly stated in the contract, then you may not have an argument unless there is a law in your state overriding such a clause. Note that a recent court case (and appeals court) upheld a similar insurer's clause in a LIFE insurance contract. The clause was obviously written as a way to not pay when the "accident" was in conjunction with being over the "legal blood alcohol limit" in that state. However, the person died in bed from regurgitating and choking. Because the person's blood alcohol was over the legal limit for driving, the courts upheld the clause that no death proceeds would be made...even though the person did not die as a result of anything related to driving.

For those that don't understand the issue: There is a rule in the Senate that any tax law that decreases revenues in the future (i.e., has the potential to increase budget deficits - which is what we had when this rule was put in place), must have a "super majority" to pass. A super majority is 60 votes. It was perceived that Bush's plan would never have gotten the 60 votes in the 50-50 Senate. So, the Republican leadership went to the Senate Parlimentarian a couple of months ago and requested a ruling as to whether they could put Bush's tax cut proposals into a buget reconciliation bill (which only needs a simple majority to pass). Another feature of a reconciliation bill is that it may not be fillibustered - it can have a limited 40 hours of debate before a final vote. In a move that really ticked off the Democrats, he ruled this could be a reconciliation bill. However, it must be limited to the 10-year window that is used for scoring a reconciliation bill. What this means is that all provisions revert to current law in 2011. (I.e., all those tax rate reductions, repeal of estate tax, pension provisions, etc., revert to the law you have on your desk today.) In order for this to remain law, an actual tax law (with 60 votes) will need to be passed between now and 2011 in order to make these provisions permanent. My personal opinion is that the only way this will happen is if a lot of the tax reductions are repealed at the same time in order to get the 60 votes (Daschle has already indicated they will immediately move to begin this process). The last section of the bill is: "Section 901. Sunset of Provisions of Act (a) In General. All provisions of, and amendments made by, this Act shall not apply - (1) to taxable, plan, or limitation years beginning after December 31, 2010, or (2) in the case of title V, to estates of decedents dying , gifts made, or generation skipping transfers, after December 31, 2010. (B) Application of Certain Laws. The Code and ERISA shall be applied and administered to years, estates, gifts, and transfers described in subsection (a) as if the provisions and amendments described in subsection (a) had never been enacted." It is this last subsection (B) that most people haven't read. I would argue that this implies that DB valuations between now and 2011 can ignore the sunset provision and project out the higher DB limits ad infinitum. Of course, the IRS (particularly Jim Holland) could come up with a different interpretation.

Andy, I am not sure I would agree with Holland (but I am assuming he is right for the moment). I need to look at older laws and see what they did when they decreased limits. For example, when TEFRA reduced limits, were the reduced limits applied to limitation years that crossed '82-'83? If so, did they only apply to terminations after the effective date? Was the effective date written like it is now, or did it reference something else (like plan years beginning in '83)? I have not looked back at these yet, but probably will in the next few days. The idea is that whatever they did in reducing limits should apply in an increase in limits (I know...I a using logic here and sometimes logic doesn't apply to IRS views). If Holland is right, I don't agree with your conclusion that we have a calendar year limit in the future. The 1/1 effective date is only an issue with the $160,000 in 2002. He is saying that a limitation year ending in 2002 is subject to the $160,000, but only for terminations after the effective date (1/1/2002). But, that is only an issue this year. When the first CPI increase hits, it will be effective for the entire limitation year that crosses 2002-2003. This would not have been the case if the original Senate language had remained...then we would have been stuck with taxable years, i.e., calendar years, because we are talking about the taxpayer here, not the company.

Andy, I was working with the staffers that were doing the technical writing when it went to conference committee. In the Senate version, they used the terminology "tax years beginning in", with no reference to limitation years (which is only in regulations, not the Code). In the House version, it referenced "years", which is the same language used in all previous laws changing the 415 limits. When I pointed out the problems the Senate language would create (effectively negating the regulations use of "limitation years ending in"), they went back and changed the language. However, they only did it for DB limits and missed the DC change for this one year. The actual exchange we had late at night on the 22nd: Mark, We would love your thoughts about whether the attached proposed solution (to the issue you raised earlier today) staff has outlined as appropriate and whether we have missed any relevant issues. If there is any way you could get back to me early Wednesday, that would be extremely helpful. Thanks so much. (my comments went directly into the attachment) (from staff) Page 136, between lines 19 and 20 (and subsequent pages referenced below) Section 415(B) limit. The section 415(B) limit phase-in is based on “taxable years”; section 415 applies on the basis of “limitation years”, not taxable years. The phase-in could be revised to refer to limitation years or, as is often done, the reference could be simply to “years”. See, for example, the effective date of bill section 611 on page 146, lines 14-16. (my response) I agree with the above approach, given that previous laws making changes in the limits have referred to “years” for effective dates of new limits. This is very consistent and would be construed to mean what is intended. (from staff) Also, under present law, the section 415 limit increase that occurs in a calendar year applies to limitation years ending in that calendar year, not beginning in that calendar year. Accordingly, the phase-in should be based on years ending in the specified calendar years, unless that would affect the revenue estimate. (my response) The vast majority of plans use calendar year as the limitation year. This change would have a negligible revenue effect. (from staff) (A corresponding change would be needed to bill section 611(g) on page 146.) If the “beginning in” structure is used, the legislative history should clarify that old-law indexing should occur in 2002, but solely for purposes of plans with limitation years beginning in 2001 and ending in 2002. (my response) The “beginning in” would cause great hardship for administrative systems, plan documents, benefit calculation procedures, etc., that are all based on regulatory guidance (1.415-3(a)(2) and 1.415-6(a)(2)) that has been in effect since 1980. (from staff) Compensation limit. A similar issue arises on page 139, between lines 20 and 21 with respect to the compensation limit. The phase-in is based on “taxable years”; the 401(a)(17) compensation limit applies, however, on the basis of “plan years”. The complicating factor here is that, under the bill, the section 404(l) compensation limit incorporates the 401(a)(17) compensation limit by reference; the section 404(l) compensation limit applies on the basis of taxable years. Under the bill, the 401(a)(17) compensation limit is also incorporated by reference for purposes of Code sections 408(k) and 505(B)(7), which apply based on different types of years. Accordingly, again, the simplest solution may be to refer to “years” in the phase-in. (my response) I agree this is appropriate, although I am not sure this clears up the cross-referencing. (from staff) SIMPLE limit. The issue also exists with respect to SIMPLE plans (see page 145, between lines 2 and 3). The SIMPLE limits apply on a calendar year basis, not based on taxable years. So the phase-in should be based on “calendar years” or simply on “years”. (my response) I agree.

All plans of a controlled group must have the same limitation year. See 1.415-2(B)(1)(ii).

What is the exercise price? What is the current price of the stock? What are the prospects for the future price of the stock? How long are the options good for? If the exercise price is $X and the price is well less than $X and there is very little possibility that it will go up much, it is pretty much a worthless deal. On the other hand, if the exercise is, say, $10, and the price is $9, and you expect it to go to $20 or $30, then you could make good money on this. Multiply the exercise price by the shares...can you afford to buy that?

The suspension of benefits rules were in ERISA. DOL regs containing the notice requirements were issued in 1978, and amended in 1981. OBRA'86 (yes there was a tax law about a month away from TRA'86) changed the accrual rules after NRA. The requirement for an actuarial increase was from ERISA. The new rules allowed an offset to the actuarial increase (which applies when no notice is given) for the additional accruals. I'd say she should have received a notice at NRA, and if not, she is entitled to an actuarial increase (although mitigated by any additional accruals). If you want lots of details, see my speeches in the EA meeting transcripts for both 1990 and 1991 on this subject.

Sorry, I do not have an answer. But, I have had similar experiences. In the early '80s I consulted to a Fortune 100 conglomerate. A subsidiary of a subsidiary of a previously acquired company had a few hundred employees and decided to terminate their DB plan. Lo and behold, when the parent company and I stepped in to facilitate the termination, we found out it didn't have a few hundred...it had over 3000 employees eligible to participate (some with many years of service). All of these were illegal aliens. This was not just a case of their providing false information, the employer was knowingly doing this as an accomplice to the arrangements. Besides greatly ticking off the parent company (all of these companies are household names), they went ahead with the termination (originally very overfunded...now underfunded) and went ahead and included all of these people. Whether or not they had to or not was never questioned...they just did it. Of course, since then, we have had a lot of new legislation on green cards, employer responsibilities, etc., so I would imagine the legal environment of the employment relationship (and whether there is one) has greatly changed.

Getting back to the original question: Where is good background? The Society of Actuaries Study Note P-461U from 1996, "Mortality Tables for Pension Plans" is exactly what you asked for, discussing why you would use different tables and the characteristics of commonly used ones. To find it...??? If you know anyone that has taken these exams between 96 and 99, try to get a copy from them. Or, you may be able to order one directly from the SOA http://www.soa.org At the very least, you can contact their librarian and get a copy (copy fees). The problem is, I do not know if they even issue this note for current exams anymore. This was used under the last structure before reform in 2000. For the SOA's most current mortality table and a lengthy document on its creation (including comparisons to old tables) see the RP-2000 (RP is retirement plans, although it never mentions that fact in the report) report at:http://www.soa.org/research/rp2000.html Note that the Academy of Actuaries recommended last week to the IRS that this table (with different tables for blue/white collar) be used for current liability in the future.

I have not done the exact calc (other than a quick look at a spreadsheet with whole years). I would say PAX's numbers are correct, but reversed. The higher number should be the annuity starting immediately, the lower number is the deferred annuity.

Although now repealed, this is specifically why the government, during the 80s, added the 15% excess benefit excise tax (payments in retirement exceeding $150,000 per year). Because they couldn't stop you up front through existing 415 rules, they decided to catch you on the back side during payout. Some doctors/dentists/etc., were playing musical chairs every Friday working for each others' practices without having any ownership interest. They would be able to get a full Section 415 benefit from each.

Hans, I agree with you. I was not saying that the AE must be different from the variable rate, I was just pointing out that it "need not be" the crediting rate. (On a related issue, when the two are different, the IRS has been scrutinizing this as not meeting the minimum accrual rules of 411 due to a backloading effect...the additional variable rate credits above the AE rate are not earned in the accrued benefit until granted, which means there will be large accruals in later years when the balances get big.)

Keith, You are wrong about there being no requirement of defining an annuity. Cash balance plans must contain language that converts the account balance into an annuity at NRD for purposes of defining an accrued benefit. And, it must be in such a manner that precludes employer discretion (i.e., explicit actuarial factors). This projectionneed not be using the interest crediting rate (e.g., a variable interest crediting rate need not be what is used to project). That means that annual changes in a variable interest crediting rate would not cause the accrued benefit to jump around. Without such language, you do not have definitely determinable benefits. Your comment that the IRS has no problem with defining the accrued benefit as the account balance is the opposite of all of the information that comes from the IRS. However, that has not stopped numerous plans from attempting this, and they keep ending up in court. See IRS Notice 96-8.

At the ALI-ABA meeting, Mark Iwry and James Holland addressed the regulatory projects due out this year. They stated they expect to "finalize soon" the proposed regs on new comparability. That was the only "soon" comment, so it must be at the top of the list. (See BNA Pension and Benefits Daily, 4/13/01, or the weekly which will carry the same story.)

When I was employed at Mercer (benefits consulting), their 401(k)/profit sharing plan was 100% Marsh-McClennon stock (the parent company). No possibility of choosing anything else. I do not know if they still do that. Yes, it is OK to do this, but there is still the question of fiduciary responsibility in providing reasonable investments. During the years I was there, MM stock did not increase and paid very minute dividends. As a result, I never deferred more than the minimum required to get a match.

I am fairly familiar with the publicly available mortality tables and have never heard of a 70 GA. Are you referring to GA=Group Annuity? (These are usually referenced by the acronym GAM.) There was a published 71GAM based on mid-60's data, projected to 1971 (the same data was projected to create the 83GAM). Perhaps what you are looking for was created by someone from that same raw data and only projected to 1970. In that case, it is not a publicly available mortality table and you would need to know from the author how they constructed it, particularly which projection scale they used and whether or not any loads were involved. The 71GAM and 83GAM, although created from the same raw data, used different projection scales and different loads.

Harry, Your second statement of converting to another form of annuity is not exactly right. It is not a clean categorizing of annuities versus lump sum. With a low interest rate: Participant is helped if the new form (annuity or otherwise) has a weighted-average length of payout that is less than the original form. So, going from a deferred life annuity to an immediate SS leveling option is helped by a low interest rate. Also, going from a deferred life annuity to an immediate life annuity is helped by a low interest rate. The shortest payout period is a lump sum which is, of course, helped by a low interest rate. Participant is hurt if the new form has a weighted-average length of payout longer than the original form. So, converting from a life annuity to a J&S is hurt by a low interest rate. In the reverse situation, assume the normal form is J&S and they convert to a life annuity. They will be helped by a low interest rate. On the mandatory provision that the J&S be the most valuable form, it is my understanding that this is deemed to pass by virtue of using the plan's actuarial equivalence if they are a "reasonable" actuarial equivalence. The fact that 417(e) rates are greater or less than this is not part of the analysis. This has always troubled me with "simple" actuarial equivalencies (e.g., a stated percentage to switch from one form or another). I have always said they must be checked against some other reasonable actuarial equivalence (undefined in the Code) to make sure that any combination of ages will still produce the desired results. These desired results are two: The QJ&S must be the most valuable, and all alternative forms must be at least the actuarial equivalent of the normal form (using a reasonable actuarial equivalence to calculate this, not necessarily the plan's factors or any Code factors). Mark

There was only one table. It was the 71GAM with the percentage loading backed out, forecasted with a projection scale to another date...I think around '80. The official name included the word forecast. I am very surprised that a plan document would refer to a proprietary table only used by one consulting firm. It is not in the SOA site's table manager.

My last paragraph was twisted around...the higher the rate, the worse off the participant is. Sorry

Gary, I believe the SS leveling is considered a decreasing annuity because it is level until SS elig., and then "decreases" to the lifetime amount. It is that one step-down that causes it to be thrown out of the nondecreasing camp. If this is not true, then the discussion is moot. Harry, The higher the interest rate, the better off the participant is in going to a lump sum or other type that puts more of the cash flow into earlier years. A higher interest rate hurts the participant when going from life annuity to a J&S, because it is spreading it out further. When I referred to "someone game the system" I was referring to the plan sponsor, not the participant. I.e., this rule is to override the plan sponsor's natural intentions in using a more reasonable assumption than the one the IRS wants you to use.

There is a regulation (I think under DOL), but I don't remember where, concerning "reasonable administrative delays," which clearly states that you do not have to adjust with interest. I think there is a reasonability standard in there of 90 days. That has been around for a long time. Of course, I agree with the earlier post that the plan document language and/or precedent may bring on a different conclusion than the minimum required by law. [Note: My memory of this may be because of the following statement implying that an administrative delay is considered to have been paid on the annuity starting date, I don't remember now. This is 1.401(a)-20, Q&A A-10(B): "Administrative delay. A payment shall not be considered to occur after the annuity starting date merely because actual payment is reasonably delayed for calculation of the benefit amount if all payments are actually made."] However, we have a new proposed regulation a couple of months ago on "retroactive annuity starting dates." If the delay is due to the participant not providing the required J&S signature forms until after the annuity starting date, then you MUST adjust with "appropriate interest." That is undefined, but this proposed regulation is a part of 417, so it may be the 417(e) rates. However, it is also undefined as to how to apply it. For example, the policy reason for this clause is because retroactive annuity starting dates could be many years or decades (e.g., a lost participant is found). Could we interpret interest as an annual thing? In that case, anything under six months could possibly be rounded to zero? This all seems backwards. If the administrator causes the delay, there is no interest. If the participant causes the delay, there is interest. Go figure.

I'll add one more comment on the regulation. Recently, one of the cash balance court decisions concerned whether or not 411(B)(1)(H) applies. In that paragraph, it says the rate of benefit accrual cannot be reduced because of the attainment of any age. A similar statement is under the EEOC and ERISA. The question is whether a cash balance plan violates this because the rate of benefit accrual does decrease with advancing age. The court found that this paragraph does not apply because the title of the paragraph is "Continued accrual beyond normal retirement age." The court felt that the title of the paragraph carried as much weight as the language itself. Applying that logic to 417(e): The title of that subsection is "Restrictions on cash-outs." So, the question is whether or not the IRS has the authority to change it from "cash-outs" to "nondecreasing annuities" in their regulation. Besides the courts giving credence to the title, there is a separate issue as to whether the regulation is a reasonable interpretretation of the statute. That may be correct in this case, and therefore that may override the concern of the subsection title only referring to cash-outs. For example, could someone game the system so as to avoid the cash-out restrictions by calling a distribution a SS leveling option, or a two-time payment instead of a single payment? I just reviewed such a calculation. The person is near NRA and requested the SS leveling option. The accrued benefit is so low in comparison to the PIA that the leveling option creates a term-certain annuity for a very short period of time until NRA and no continuing life annuity after NRA. From a policy standpoint, shouldn't this fall under 417(e) cash-out rules? (By the way, this client is not using 417(e) rates on the conversion.)