Conveniently, the Sixth Circuit has issued another opinion illustrating this urge of the lower courts to find plaintiff remedies by creative applications of the law. The case, Bloemaker v. Laborers' Local 265 Pension Fund, No. 09-3536 (6th Circuit 2010), held that participants in defined benefit pension plans can assert benefit rights not created by the plan but reflected in written estimates of benefits based on an equitable estoppel theory.
The participant had asked for an estimate of benefits under a plan’s early retirement provisions. The estimate provided was incorrect, and obviously so to anyone who has experience with calculations of early retirement benefits. The participant then retired, in reliance on the estimate and received benefits for almost two years. At this point, the TPA noticed the error, and contacted the participant stating that (1) future benefits would be reduced, and (2) the participant had to repay the excess benefits actually received. The error was based on “a computer programming error.”
The opinion is poorly written, so much so that the operational test is stated in two similar but hardly identical formulations. In the first version, the Sixth Circuit said: “Under our precedent, the elements of an equitable estoppel claim are: 1) conduct or language amounting to a representation of material fact; 2) awareness of the true facts by the party to be estopped; 3) an intention on the part of the party to be estopped that the representation be acted on, or conduct toward the party asserting the estoppel such that the latter has a right to believe that the former’s conduct is so intended; 4) unawareness of the true facts by the party asserting the estoppel; and 5) detrimental and justifiable reliance by the party asserting estoppel on the representation.” The second version required “intended deception or such gross negligence as to amount to constructive fraud, plus (1) a written representation; (2) plan provisions which, although unambiguous, did not allow for individual calculation of benefits; and (3) extraordinary circumstances in which the balance of equities strongly favors the application of estoppel.” The Court treated these two formulations as being the same, and to maintain some semblance of order here the balance of this post will focus on the first because it is more detailed and less vague.
Under the first standard, the first and the last three items were pretty easily met by the participant’s complaint. The second, however, seems problematic. In fact, the complaint appears to have had only general allegations that the plan and the TPA knew the true facts. But what facts? Did the TPA know the benefit calculation was wrong? That is unlikely. Will the courts ultimately say that knowledge of the facts underlying the benefit determination is awareness of the true facts?
Also, the calculated early retirement benefit involved an actuarial reduction of the annual benefit of only 12.3%, a reduction that is pretty obviously too small. This raises a very real question as to whether there was actual reliance by the participant, whether any reliance was justifiable or whether the participant just decided to take advantage of the error. This fact question can be raised under either the fourth or fifth elements set out by the Sixth Circuit.
And, of course, it is still an open question whether the Supreme Court will, ultimately, allow an equitable estoppel claim. The Sixth Circuit makes all the right protestations to the effect that plans have to be in writing and that a determination that a participant has rights exceeding the terms of the plan may adversely affect other participants, but gives these considerations no effect.
Various Circuit Courts, but not all, allow some version of the equitable estoppel theory. Given that, the Supreme Court needs to step in again and shoot down this latest attempt to find remedies not provided in ERISA.
So how does this apply in a defined contribution context? Presumably, the theory could be applied based on (1) representations as to contribution rates, (2) representations as to the availability of various kinds of investments, (3) participant statements, and (4) whatever else is contained in a written communication.
Wednesday, June 9, 2010
Attorney’s Fees Awarded for Lawsuit Resulting in Remand to Plan Administrator
The Supreme Court, in Hardt v. Reliance Standard Life Insurance Co., decided that attorney’s fees can be awarded to a participant where a case was remanded to the plan administrator and the claim was then granted. The attempt to prevent the award was based on the premises that (a) the person requesting an award of attorney’s fees must be a “prevailing party” in the lawsuit, and (b) “a fee claimant is a “prevailing party” only if he has obtained an “enforceable judgmen[t]on the merits ” or a “court-ordered consent decre[e]”.
The logic of the decision was that the specific language of ERISA does not require “prevailing party” status, but only that the person seeking fees show “some degree of success on the merits”.
This rule is to be applied not just in determining whether a court has authority to make an award of attorney’s fees but also in reviewing that court’s exercise of discretion in making the award. This, in effect, overturned the existing five factor analysis applied by the lower courts. Described by the Supreme Court as “(1) the degree of opposing parties’ culpability or bad faith; (2) ability of opposing parties to satisfy an award of attorneys’ fees; (3) whether an award of attorneys’ fees against the opposing parties would deter other persons acting under similar circumstances; (4) whether the parties requesting attorneys’ fees sought to benefit all participants and beneficiaries of an ERISA plan or to resolve a significant legal question regarding ERISA itself; and (5) the relative merits of the parties’ positions.” The only limitations on the capacity of a court to award attorney’s fees are that the result has to be more that “trivial success on the merits” or a “purely procedural victory.”
This case is primarily being written up as to the first issue, whether the participant/claimant has to be a “prevailing party.” However, the lack of meaningful standards for reviewing the exercise of discretion by the court making the award is actually more troubling. Until the courts begin to establish a track record of making and reviewing awards and denials of awards under this decision, plans and plan administrators will live with uncertainty. Logically, since most cases outside of disability benefits are rejected, this broad ability to make awards could be advantageous. However, in the real world it is unlikely that plans will be awarded attorney’s fees at anything like the rate of awards to participants, because that second rejected factor will be applied, explicitly or implicitly.
The logic of the decision was that the specific language of ERISA does not require “prevailing party” status, but only that the person seeking fees show “some degree of success on the merits”.
This rule is to be applied not just in determining whether a court has authority to make an award of attorney’s fees but also in reviewing that court’s exercise of discretion in making the award. This, in effect, overturned the existing five factor analysis applied by the lower courts. Described by the Supreme Court as “(1) the degree of opposing parties’ culpability or bad faith; (2) ability of opposing parties to satisfy an award of attorneys’ fees; (3) whether an award of attorneys’ fees against the opposing parties would deter other persons acting under similar circumstances; (4) whether the parties requesting attorneys’ fees sought to benefit all participants and beneficiaries of an ERISA plan or to resolve a significant legal question regarding ERISA itself; and (5) the relative merits of the parties’ positions.” The only limitations on the capacity of a court to award attorney’s fees are that the result has to be more that “trivial success on the merits” or a “purely procedural victory.”
This case is primarily being written up as to the first issue, whether the participant/claimant has to be a “prevailing party.” However, the lack of meaningful standards for reviewing the exercise of discretion by the court making the award is actually more troubling. Until the courts begin to establish a track record of making and reviewing awards and denials of awards under this decision, plans and plan administrators will live with uncertainty. Logically, since most cases outside of disability benefits are rejected, this broad ability to make awards could be advantageous. However, in the real world it is unlikely that plans will be awarded attorney’s fees at anything like the rate of awards to participants, because that second rejected factor will be applied, explicitly or implicitly.
Another Attack on Administrator Discretion Turned Back
There is a pattern in case law development under ERISA of lower courts giving pro-employee results, only to be overturned on appeal, particularly to the Supreme Court. This pattern certainly extends to case law on the appropriate standard of review of decisions by plan administrators. Yet another instance of this pattern is reflected in the recent Supreme Court case of Conkright v. Frommert, No. 08-810 (Apr. 21, 2010).
In Conkright, a participant’s claim was denied on one basis. The participant sued and eventually the basis for denying the claim was rejected by the Circuit Court, which returned the case to the plan administrator for consideration of other interpretations. The plan administrator again denied the claim, based on another interpretation, whereupon the participant went back to the courts. The participant argued that, since the plan administrator had been wrong the first time, there should be a de novo review, rather than a review for abuse of discretion, and the District Court and Circuit Court agreed. The Supreme Court disagreed, holding that the decision of the plan administrator on remand was still entitled to the deferential abuse of discretion review.
There are three lessons here. First, procedurally, plans should push for remand after adverse court decisions. Second, decisions by plan administrators should include language stating, broadly, that the initial denial is based on the issues considered as sufficient, and that other issues may warrant a denial. Third, nobody should get too worked up about lower court decisions on ERISA issues.
In Conkright, a participant’s claim was denied on one basis. The participant sued and eventually the basis for denying the claim was rejected by the Circuit Court, which returned the case to the plan administrator for consideration of other interpretations. The plan administrator again denied the claim, based on another interpretation, whereupon the participant went back to the courts. The participant argued that, since the plan administrator had been wrong the first time, there should be a de novo review, rather than a review for abuse of discretion, and the District Court and Circuit Court agreed. The Supreme Court disagreed, holding that the decision of the plan administrator on remand was still entitled to the deferential abuse of discretion review.
There are three lessons here. First, procedurally, plans should push for remand after adverse court decisions. Second, decisions by plan administrators should include language stating, broadly, that the initial denial is based on the issues considered as sufficient, and that other issues may warrant a denial. Third, nobody should get too worked up about lower court decisions on ERISA issues.
Tuesday, April 27, 2010
FAB 2010-01: Further Thoughts
The issuance of FAB 2010-01 by the Department of Labor has raised a very important problem with respect to 403(b) plans. The language of FAB 2010-01 creates a risk that employers seeking to avoid the application of ERISA to their 403(b) plans may fail in that effort. This post addresses that specific problem from two perspectives, whether and to what extent the ruling is deficient and how to adapt to it.
The Problem
The problem arises from the way in which the response to Question 15 is worded. The FAB says: “Q-15. Would an employer exceed the ERISA coverage safe harbor limitations on employer involvement in 29 CFR 2510.3-2(f) if the employer hires a third-party administrator (TPA) to make discretionary decisions? Yes. The employer's selection of a TPA would be inconsistent with the safe harbor in 29 CFR 2510.3-2(f). The Department's FAB 2007-02 addressed the safe harbor conditions for tax-sheltered annuity arrangements to fall outside of ERISA Title I coverage, and specifically noted that the documents governing the arrangement could identify parties other than the employer as "responsible for administrative functions, including those related to tax compliance." As FAB 2007-02 further noted, the documents should correctly describe the employer's limited role and allocate discretionary determinations to the annuity provider or other responsible third party selected by a person other than the employer. Moreover, an employer may limit the available providers it will make available in its safe harbor arrangement to those where the 403(b) contracts or accounts or other governing documents prepared by the provider state that the provider or another appropriate third party is responsible for discretionary decisions related to loans and hardship distributions.” (Emphasis added.)
The key point here is that the employer cannot appoint anyone to make discretionary determinations. FAB 2007-02 limited the employer’s ability to make discretionary determinations, resulting in a market trend towards hiring third party administrators to manage 403(b) plans, but this new constraint in FAB 2010-01 makes such a decision problematic.
The Context
The issue arises under DOL Regs. 2510.3-2(f), which provides an exemption from ERISA for salary reduction-only 403(b) plans that have minimal employer involvement. The requirements, broadly speaking, are that (1) participation be completely voluntary, (2) all rights under any investment vehicle be enforceable only by the employee or beneficiary, (3) the involvement of the employer be limited to certain optional specified activities, and (4) the employer receive no compensation. It is the third of these requirements that is at issue, the scope of the employer’s involvement in the plan.
Under the Regulation, which has been around for a long time, the permitted employer functions are: “(i) Permitting annuity contractors (which term shall include any agent or broker who offers annuity contracts or who makes available custodial accounts within the meaning of section 403(b)(7) of the Code) to publicize their products to employees, (ii) Requesting information concerning proposed funding media, products or annuity contractors; (iii) Summarizing or otherwise compiling the information provided with respect to the proposed funding media or products which are made available, or the annuity contractors whose services are provided, in order to facilitate review and analysis by the employees; (iv) Collecting annuity or custodial account considerations as required by salary reduction agreements or by agreements to forego salary increases, remitting such considerations to annuity contractors and maintaining records of such considerations; (v) Holding in the employer's name one or more group annuity contracts covering its employees...(vii) After February 6, 1978, limiting the funding media or products available to employees, or the annuity contractors who may approach employees, to a number and selection which is designed to afford employees a reasonable choice in light of all relevant circumstances. Relevant circumstances may include, but would not necessarily be limited to, the following types of factors: (A) The number of employees affected, (B) The number of contractors who have indicated interest in approaching employees, (C) The variety of available products, (D) The terms of the available arrangements, (E) The administrative burdens and costs to the employer, and (F) The possible interference with employee performance resulting from direct solicitation by contractors.” The reference to paragraph (f)(2) permits the employee to designate the employer as the employee’s authorized representative in dealing with the investment vendor.
Then, the IRS issued regulations under 403(b) intended to make 403(b) plans more like 401(k) plans in their operation. Prior to this issuance, enforcement had been, to say the least, relaxed, but the 403(b) regulations initiated a dramatic wave of regulatory vigor. One of the new requirements is for there to be a written plan, which raised concerns among employers about the continuing availability of the exemption in DOL Regs. 2510.3-2(f). This concern was based on a misunderstanding of the exemption. The exemption assumes there is a plan, but deems the plan not to be maintained by the employer, which is a basic jurisdictional requisite of ERISA. However, the confusion made the exemption a focus of attention, and reasonably so. The problem is not the written plan document requirement, but the tax treatment of the plan. Under the 403(b) regulations, the tax effects of compliance failures by the plan were made explicit. These effects were, if anything, limited by the 403(b) regulations, but their issuance caused people to notice them.
The core requirements under the 403(b) regulations for a plan to be a 403(b) plan are at Regs. 1.403(b)-3(a). The relevant requirements here are that (1) the plan meet the general minimum distribution rules, (2) the plan comply with the direct rollover rules of 402(c)(8)(B), and (c) the plan limit contributions under 415. In addition, Regs. 1.403(b)-3(b)(3) states that: “For this purpose, a plan is a written defined contribution plan, which, in both form and operation, satisfies the requirements of §1.403(b)-1, §1.403(b)-2, this section, and §§1.403(b)-4 through 1.403(b)-11. For purposes of §1.403(b)-1, §1.403(b)-2, this section, and §§1.403(b)-4 through 1.403(b)-11, the plan must contain all the material terms and conditions for eligibility, benefits, applicable limitations, the contracts available under the plan, and the time and form under which benefit distributions would be made. For purposes of §1.403(b)-1, §1.403(b)-2, this section, and §§1.403(b)-4 through 1.403(b)-11, a plan may contain certain optional features that are consistent with but not required under section 403(b), such as hardship withdrawal distributions, loans, plan-to-plan or annuity contract-to-annuity contract transfers, and acceptance of rollovers to the plan. However, if a plan contains any optional provisions, the optional provisions must meet, in both form and operation, the relevant requirements under section 403(b), this section and §§1.403(b)-4 through 1.403(b)-11.” This language in fact expands the requirements listed at Regs. 1.403(b)-3(a) so as to require that, in order for any tax rule sought by application of 403(b) to be available, the plan must fully comply with 403(b). And that compliance is not just that the plan document be worded properly, but that there be operational compliance. Otherwise, the plan is not eligible under 403(b) and, for example, salary reduction arrangements are invalid.
This history creates a fundamental tension, between the requirements of the IRS and the DOL. For the two most common concerns in this area, loans and hardship distributions, the IRS requires that there be correct administration and the DOL precludes the employer from making the sorts of determinations required to ensure correct compliance. (There are other issues with similar problems, but this analysis can be applied to any of them.)
Now we get a little good news. The 403(b) regulations use the term “contract” to mean the terms of the plan and funding vehicles as related to one specific participant, and generally limit the sanctions for operational failures to the contract, to the specific participant with respect to whom the failure has occurred. The exceptions to this rule are (1) nondiscrimination failures, which, in a salary reduction-only plan, means the universal availability rule, and (2) failure of the employer to be an eligible sponsor. Thus, for example, an incorrect hardship determination as to Employee A will not cause the plan to fail as to other employees. This language reduces the scope of risk as to tax sanctions on the employer and employee. They are not, however, eliminated; operational failure as to Employee A cause deferrals for Employee A to fail.
The Errors in FAB 2010-01
The rules enunciated in FAB 2010-01 result from two basic errors in analysis. First, the nature of discretion is misunderstood. Second, the DOL creates an infinite regression of causes.
The FAB assumes that determinations related to loans and hardship distributions are discretionary. As to a standard loan provision, which requires only mathematical calculations to determine permissibility of loans, this is obviously absurd. Some loan provisions may have language that is not fully automatic (e.g., adopt a loan policy, flexibility in the duration of the loan), but these are easily eliminated by a simplification of the terms of the plan to make loan eligibility a fully mathematical exercise.
Hardship eligibility is a more difficult case. Hardship standards can involve factual determinations that seem to involve discretion and judgment and plan documents typically give the plan administrator discretion as to hardship determinations. However, the existence or nonexistence of a hardship is a factual matter, albeit a more difficult one than the math involved in loan eligibility. This is certainly true where the hardship is safe harbor-safe harbor, and true, but less obviously so, where hardship distributions are not limited to the safe harbors. If, then, the plan removes its discretion language, the administrator is making a nondiscretionary factual determination, subject to de novo review by the courts.
The infinite regression error is the equivalent of asking what caused the Big Bang. Positing, logically, that each event has a cause, any voluntary act that is a necessary condition to a discretionary determination is a cause of that determination and, in the logic of the DOL, a forbidden discretionary determination. Thus, under this logic, the decision to create a 403(b) plan is a forbidden discretionary determination, as is the designation of a person or persons to undertake the hiring of a TPA. There must be a line drawn somewhere that cuts off this logical chain, and DOL has not proposed any such line. (Given the logic of the DOL’s role as an enforcement agency, this is hardly surprising since it would reduce the number of persons who are deemed to be fiduciaries and therefore amenable to DOL sanctions and liability to plans and participants.) A logical place to draw this line is that decisions under the plan are subject to causal regression analysis, but that decisions as to the terms of the plan are not.
Now What?
Employers need to have control of their own tax compliance, as both DOL and IRS concede. The question is how to do so without subjecting the plan to ERISA. The following are possible approaches.
Take the FAB literally. Under this approach, the TPA would make no actual determinations of any kind. Instead, the terms of the plan would state that all paperwork must flow through the TPA, that the TPA would send that paperwork on to affected investment providers and that the TA could advise investment providers as to its opinion on the permissibility of the proposed loan or hardship distribution under 403(b) and the terms of the plan.
This structure takes advantage of Q-17 of FAB 2010-01, which reads as follows:
“Q-17. Would an arrangement that otherwise meets the terms of the safe harbor stay within the safe harbor if the written plan document required by Treasury Regulations under Code section 403(b) provides that salary deferrals will be discontinued to a provider that is not complying with Code requirements? Yes. If the purpose of the provisions for discontinuing a provider from offering products to participants in the arrangement is necessary to maintain tax code compliance, then including such provisions in the arrangement will not take it outside the safe harbor.”
The fact that the TPA considers the specific loan or hardship distribution to be not in compliance with 403(b), under this structure, does not mean that the loan or hardship distribution is not made. But, by providing a factual basis for cutting of future deferrals to the investment provider, it gives the TPA de facto veto power.
The second option is to redo the terms of the plan to remove discretion. This has two components. The first is to remove the normal discretion/standard of review language from the plan document. The second is to eliminate the need for complex factual determinations. Logically, either one of these would be sufficient, but doing both probably makes sense given the added costs of ERISA compliance.
The third approach is to appoint the TPA in the plan document. Essentially, the position here is based on a close reading of DOL Regs. 2510.3-2(f). That regulation presupposes a “program” that meets the requirements of 403(b), which obviously has to be created by action of the employer. Only then does it limit the employer’s involvement. Since the adoption of the 403(b) plan is the action creating or continuing the “program” that action is not subject to the requirements of the DOL Regulation. This would also tie in to the logic the DOL uses to determine what functions (grantor functions) cannot be paid for by the plan, rather than the employer. These generally include design analysis and documentation, and this is therefore a familiar division point or line to the DOL. As stated above, the reading is close, so that an agency determination by the DOL that it is incorrect would be given deference by a court (particularly because the exemption is purely created by the DOL and is not contained in any statute).
The fourth approach is to combine two or all three of the above. Given that the approaches do not contain any inconsistencies, this is the preferred option. Thus, a conservative approach would (1) rewrite the plan document to (a) eliminate the ordinary discretion/standard of review language, (b) designate the TPA as the administrator of the plan for nondiscretionary matters and as the employer’s representative for tax compliance, and (c) reduce all determinations of fact to their simplest levels through the use of safe harbors and hard rules, and (2) eliminate all investment providers that do make distributions or loans over the objection of the TPA from receiving future deferrals or intra-plan transfers.
NOTE: Hardships and loans are not the only areas in which these issues arise. They are the focus here simply for convenience.
The Problem
The problem arises from the way in which the response to Question 15 is worded. The FAB says: “Q-15. Would an employer exceed the ERISA coverage safe harbor limitations on employer involvement in 29 CFR 2510.3-2(f) if the employer hires a third-party administrator (TPA) to make discretionary decisions? Yes. The employer's selection of a TPA would be inconsistent with the safe harbor in 29 CFR 2510.3-2(f). The Department's FAB 2007-02 addressed the safe harbor conditions for tax-sheltered annuity arrangements to fall outside of ERISA Title I coverage, and specifically noted that the documents governing the arrangement could identify parties other than the employer as "responsible for administrative functions, including those related to tax compliance." As FAB 2007-02 further noted, the documents should correctly describe the employer's limited role and allocate discretionary determinations to the annuity provider or other responsible third party selected by a person other than the employer. Moreover, an employer may limit the available providers it will make available in its safe harbor arrangement to those where the 403(b) contracts or accounts or other governing documents prepared by the provider state that the provider or another appropriate third party is responsible for discretionary decisions related to loans and hardship distributions.” (Emphasis added.)
The key point here is that the employer cannot appoint anyone to make discretionary determinations. FAB 2007-02 limited the employer’s ability to make discretionary determinations, resulting in a market trend towards hiring third party administrators to manage 403(b) plans, but this new constraint in FAB 2010-01 makes such a decision problematic.
The Context
The issue arises under DOL Regs. 2510.3-2(f), which provides an exemption from ERISA for salary reduction-only 403(b) plans that have minimal employer involvement. The requirements, broadly speaking, are that (1) participation be completely voluntary, (2) all rights under any investment vehicle be enforceable only by the employee or beneficiary, (3) the involvement of the employer be limited to certain optional specified activities, and (4) the employer receive no compensation. It is the third of these requirements that is at issue, the scope of the employer’s involvement in the plan.
Under the Regulation, which has been around for a long time, the permitted employer functions are: “(i) Permitting annuity contractors (which term shall include any agent or broker who offers annuity contracts or who makes available custodial accounts within the meaning of section 403(b)(7) of the Code) to publicize their products to employees, (ii) Requesting information concerning proposed funding media, products or annuity contractors; (iii) Summarizing or otherwise compiling the information provided with respect to the proposed funding media or products which are made available, or the annuity contractors whose services are provided, in order to facilitate review and analysis by the employees; (iv) Collecting annuity or custodial account considerations as required by salary reduction agreements or by agreements to forego salary increases, remitting such considerations to annuity contractors and maintaining records of such considerations; (v) Holding in the employer's name one or more group annuity contracts covering its employees...(vii) After February 6, 1978, limiting the funding media or products available to employees, or the annuity contractors who may approach employees, to a number and selection which is designed to afford employees a reasonable choice in light of all relevant circumstances. Relevant circumstances may include, but would not necessarily be limited to, the following types of factors: (A) The number of employees affected, (B) The number of contractors who have indicated interest in approaching employees, (C) The variety of available products, (D) The terms of the available arrangements, (E) The administrative burdens and costs to the employer, and (F) The possible interference with employee performance resulting from direct solicitation by contractors.” The reference to paragraph (f)(2) permits the employee to designate the employer as the employee’s authorized representative in dealing with the investment vendor.
Then, the IRS issued regulations under 403(b) intended to make 403(b) plans more like 401(k) plans in their operation. Prior to this issuance, enforcement had been, to say the least, relaxed, but the 403(b) regulations initiated a dramatic wave of regulatory vigor. One of the new requirements is for there to be a written plan, which raised concerns among employers about the continuing availability of the exemption in DOL Regs. 2510.3-2(f). This concern was based on a misunderstanding of the exemption. The exemption assumes there is a plan, but deems the plan not to be maintained by the employer, which is a basic jurisdictional requisite of ERISA. However, the confusion made the exemption a focus of attention, and reasonably so. The problem is not the written plan document requirement, but the tax treatment of the plan. Under the 403(b) regulations, the tax effects of compliance failures by the plan were made explicit. These effects were, if anything, limited by the 403(b) regulations, but their issuance caused people to notice them.
The core requirements under the 403(b) regulations for a plan to be a 403(b) plan are at Regs. 1.403(b)-3(a). The relevant requirements here are that (1) the plan meet the general minimum distribution rules, (2) the plan comply with the direct rollover rules of 402(c)(8)(B), and (c) the plan limit contributions under 415. In addition, Regs. 1.403(b)-3(b)(3) states that: “For this purpose, a plan is a written defined contribution plan, which, in both form and operation, satisfies the requirements of §1.403(b)-1, §1.403(b)-2, this section, and §§1.403(b)-4 through 1.403(b)-11. For purposes of §1.403(b)-1, §1.403(b)-2, this section, and §§1.403(b)-4 through 1.403(b)-11, the plan must contain all the material terms and conditions for eligibility, benefits, applicable limitations, the contracts available under the plan, and the time and form under which benefit distributions would be made. For purposes of §1.403(b)-1, §1.403(b)-2, this section, and §§1.403(b)-4 through 1.403(b)-11, a plan may contain certain optional features that are consistent with but not required under section 403(b), such as hardship withdrawal distributions, loans, plan-to-plan or annuity contract-to-annuity contract transfers, and acceptance of rollovers to the plan. However, if a plan contains any optional provisions, the optional provisions must meet, in both form and operation, the relevant requirements under section 403(b), this section and §§1.403(b)-4 through 1.403(b)-11.” This language in fact expands the requirements listed at Regs. 1.403(b)-3(a) so as to require that, in order for any tax rule sought by application of 403(b) to be available, the plan must fully comply with 403(b). And that compliance is not just that the plan document be worded properly, but that there be operational compliance. Otherwise, the plan is not eligible under 403(b) and, for example, salary reduction arrangements are invalid.
This history creates a fundamental tension, between the requirements of the IRS and the DOL. For the two most common concerns in this area, loans and hardship distributions, the IRS requires that there be correct administration and the DOL precludes the employer from making the sorts of determinations required to ensure correct compliance. (There are other issues with similar problems, but this analysis can be applied to any of them.)
Now we get a little good news. The 403(b) regulations use the term “contract” to mean the terms of the plan and funding vehicles as related to one specific participant, and generally limit the sanctions for operational failures to the contract, to the specific participant with respect to whom the failure has occurred. The exceptions to this rule are (1) nondiscrimination failures, which, in a salary reduction-only plan, means the universal availability rule, and (2) failure of the employer to be an eligible sponsor. Thus, for example, an incorrect hardship determination as to Employee A will not cause the plan to fail as to other employees. This language reduces the scope of risk as to tax sanctions on the employer and employee. They are not, however, eliminated; operational failure as to Employee A cause deferrals for Employee A to fail.
The Errors in FAB 2010-01
The rules enunciated in FAB 2010-01 result from two basic errors in analysis. First, the nature of discretion is misunderstood. Second, the DOL creates an infinite regression of causes.
The FAB assumes that determinations related to loans and hardship distributions are discretionary. As to a standard loan provision, which requires only mathematical calculations to determine permissibility of loans, this is obviously absurd. Some loan provisions may have language that is not fully automatic (e.g., adopt a loan policy, flexibility in the duration of the loan), but these are easily eliminated by a simplification of the terms of the plan to make loan eligibility a fully mathematical exercise.
Hardship eligibility is a more difficult case. Hardship standards can involve factual determinations that seem to involve discretion and judgment and plan documents typically give the plan administrator discretion as to hardship determinations. However, the existence or nonexistence of a hardship is a factual matter, albeit a more difficult one than the math involved in loan eligibility. This is certainly true where the hardship is safe harbor-safe harbor, and true, but less obviously so, where hardship distributions are not limited to the safe harbors. If, then, the plan removes its discretion language, the administrator is making a nondiscretionary factual determination, subject to de novo review by the courts.
The infinite regression error is the equivalent of asking what caused the Big Bang. Positing, logically, that each event has a cause, any voluntary act that is a necessary condition to a discretionary determination is a cause of that determination and, in the logic of the DOL, a forbidden discretionary determination. Thus, under this logic, the decision to create a 403(b) plan is a forbidden discretionary determination, as is the designation of a person or persons to undertake the hiring of a TPA. There must be a line drawn somewhere that cuts off this logical chain, and DOL has not proposed any such line. (Given the logic of the DOL’s role as an enforcement agency, this is hardly surprising since it would reduce the number of persons who are deemed to be fiduciaries and therefore amenable to DOL sanctions and liability to plans and participants.) A logical place to draw this line is that decisions under the plan are subject to causal regression analysis, but that decisions as to the terms of the plan are not.
Now What?
Employers need to have control of their own tax compliance, as both DOL and IRS concede. The question is how to do so without subjecting the plan to ERISA. The following are possible approaches.
Take the FAB literally. Under this approach, the TPA would make no actual determinations of any kind. Instead, the terms of the plan would state that all paperwork must flow through the TPA, that the TPA would send that paperwork on to affected investment providers and that the TA could advise investment providers as to its opinion on the permissibility of the proposed loan or hardship distribution under 403(b) and the terms of the plan.
This structure takes advantage of Q-17 of FAB 2010-01, which reads as follows:
“Q-17. Would an arrangement that otherwise meets the terms of the safe harbor stay within the safe harbor if the written plan document required by Treasury Regulations under Code section 403(b) provides that salary deferrals will be discontinued to a provider that is not complying with Code requirements? Yes. If the purpose of the provisions for discontinuing a provider from offering products to participants in the arrangement is necessary to maintain tax code compliance, then including such provisions in the arrangement will not take it outside the safe harbor.”
The fact that the TPA considers the specific loan or hardship distribution to be not in compliance with 403(b), under this structure, does not mean that the loan or hardship distribution is not made. But, by providing a factual basis for cutting of future deferrals to the investment provider, it gives the TPA de facto veto power.
The second option is to redo the terms of the plan to remove discretion. This has two components. The first is to remove the normal discretion/standard of review language from the plan document. The second is to eliminate the need for complex factual determinations. Logically, either one of these would be sufficient, but doing both probably makes sense given the added costs of ERISA compliance.
The third approach is to appoint the TPA in the plan document. Essentially, the position here is based on a close reading of DOL Regs. 2510.3-2(f). That regulation presupposes a “program” that meets the requirements of 403(b), which obviously has to be created by action of the employer. Only then does it limit the employer’s involvement. Since the adoption of the 403(b) plan is the action creating or continuing the “program” that action is not subject to the requirements of the DOL Regulation. This would also tie in to the logic the DOL uses to determine what functions (grantor functions) cannot be paid for by the plan, rather than the employer. These generally include design analysis and documentation, and this is therefore a familiar division point or line to the DOL. As stated above, the reading is close, so that an agency determination by the DOL that it is incorrect would be given deference by a court (particularly because the exemption is purely created by the DOL and is not contained in any statute).
The fourth approach is to combine two or all three of the above. Given that the approaches do not contain any inconsistencies, this is the preferred option. Thus, a conservative approach would (1) rewrite the plan document to (a) eliminate the ordinary discretion/standard of review language, (b) designate the TPA as the administrator of the plan for nondiscretionary matters and as the employer’s representative for tax compliance, and (c) reduce all determinations of fact to their simplest levels through the use of safe harbors and hard rules, and (2) eliminate all investment providers that do make distributions or loans over the objection of the TPA from receiving future deferrals or intra-plan transfers.
NOTE: Hardships and loans are not the only areas in which these issues arise. They are the focus here simply for convenience.
BP Solutions Announced
PRESS RELEASE
BENEFIT PLAN SOLUTIONS
Benefit Plan Solutions announces its newest service offering, DC Solutions. DC Solutions is a technical consulting, problem solving and plan design service available on either an ad hoc basis or by subscription.
DC Solutions clients can submit issues or problems, ranging from plan or ERISA interpretation to plan design and correction of plan administration issues. BPS will provide, in response, analysis of the technical issues involved and alternative courses of action taking into account the client’s role and situation. All subscribers will receive sanitized versions of all submissions and the responses and all comments from any client will be fully shared.
Pricing for the service, on a subscription basis, is $500 per year or $49 per month. This price allows subscribers full access to all resources and products of BPS. These include a blog, a newsletter and a manual on Section 457 plans. An additional manual on Section 403(b) plans is scheduled for completion in May 2010. Nonsubscription services are available for individually quoted prices.
Benefit Plan Solutions is an independent retirement plan consulting firm, focusing on defined contribution plans. Services include plan design analysis, including integration of Section 401(k), 403(b) and 457 plans, plan administration support for sponsors, plan administrators and third party administrators, general consulting and small plan administration.
BENEFIT PLAN SOLUTIONS
Benefit Plan Solutions announces its newest service offering, DC Solutions. DC Solutions is a technical consulting, problem solving and plan design service available on either an ad hoc basis or by subscription.
DC Solutions clients can submit issues or problems, ranging from plan or ERISA interpretation to plan design and correction of plan administration issues. BPS will provide, in response, analysis of the technical issues involved and alternative courses of action taking into account the client’s role and situation. All subscribers will receive sanitized versions of all submissions and the responses and all comments from any client will be fully shared.
Pricing for the service, on a subscription basis, is $500 per year or $49 per month. This price allows subscribers full access to all resources and products of BPS. These include a blog, a newsletter and a manual on Section 457 plans. An additional manual on Section 403(b) plans is scheduled for completion in May 2010. Nonsubscription services are available for individually quoted prices.
Benefit Plan Solutions is an independent retirement plan consulting firm, focusing on defined contribution plans. Services include plan design analysis, including integration of Section 401(k), 403(b) and 457 plans, plan administration support for sponsors, plan administrators and third party administrators, general consulting and small plan administration.
Monday, March 29, 2010
Why I hate FAB 2010-01
The portions of FAB 2010-01 dealing with hiring a TPA are very poorly written and belie any knowledge of the real world.
The employer, in the plan document, is assumed to appoint a plan administrator of some sort. in the quoted language from FAB 2007-02. Reading the language narrowly, a problem arises only where the employer appoints a third party to make discretionary determinations. It is entirely possible, under the rubric of "administrative functions, including those related to tax compliance," to designate a TPA as plan administrator either in the plan document or otherwise. As to limitations on the TPA's authority:
1-As to loans, there are discretionary determinations only where there is discretion. If the eligibility and amount rules in the plan document are mechanical (i.e., amount, duration, etc. are formulaic), it is hard to see any discretion.
2-As to hardship, again there can be circumstances where no discretion is exercisable. E.g., the amount is deemed both an immediate and heavy financial need and necessary to satisfy financial need. However, there also can be circumstances under which there is discretion, as the DOL understands that term.
On the flip side, where a participant has more than one product vendor, it is possible for the participant to request loans/hardship distributions independently from each. This could easily result in excess loans or hardship amounts. Under the 403(b) regs. (i.e., the second sentence of Regs.1.403(b)-3(b)(3)(ii)), there must be someone performing those limitation functions.
The cautious approach would provide that the TPA simply collect and transmit information to the product vendors, giving the participant a single location for loan and hardship applications, coordinating so that maximum limits are not exceeded. and forwarding the results to the product vendors for discretionary determination. This much is not only permitted, but required.
In theory, there could be a less cautious approach that would attempt to draw a line, as to loans and hardship, between mechanistic determinations and discretionary ones. For example, a safe harbor/safe harbor hardship determination might be permitted, but others referred to the product vendors. Functionally, this runs into two problems. First, having ordinary TPA employees draw those sorts of lines will (not may) result in errors. Second, the product vendor has to make its own determination, under the terms of its own contract, as to whether or not the loan/hardship is permitted, and will not rely on the TPA unless the venodr has selected the TPA.
A third theoretical option is available. The plan document can designate a plan administrator (say, a corporate officer/nonparticpant) that can retain the services of a TPA. Unfortunately, this runs into the "what caused the big bang" problem. That is, appointing someone to appoint someone else to exercise discretion is, in the eyes of the DOL, a discretionary decision. So, probably a no-go here as well.
I am no philosopher or logician, but DOL seems to be unaware of two basic logical/linguistic fallacies.
First, they take the position that the kinds of factual determinations underlying loans and hardships are discretionary. Clearly, in the real world, this is not true all the time. And, even where there are factual issues, they are, in fact, factual. Thus, such determinations are inherently unlike decision such as whether or not to have a 403(b) plan. DOL appears to be confused by the routine language inserted in plans (broadest discretion, etc.) to get an abuse of discretion review of plan decisions into believing that all actions are discretionary. Is the determination as to whether an employer is of the sort that can offer the 403(b) catch-up any less discretionary than the determination that a person's house has burned down?
There is a pretty clear argument that this source of discretion can be eliminated simply by stating that the standard of decision for plan determinations other than as to tax compliance issues is nondiscretionary and the standard of review is de novo. And, in the kind of plan we are talking about (loosely connected products to which the employer makes no contribution), the kind of exposure to costs that motivates the seeking for an abuse of discretion standard of review is either reduced or absent. To the extent that the participant gets money, it is, after all, the employee's money. This is the fourth theoretical option.
The second fallacy is the illogic of infinite regression. Anything that is a necessary, or "but for," logical or sequential precondition to the determination of a discretionary determination can be caught under this fallacy. If the employer allows or requires a product vendor to make a determination, is that employer decision discretionary? Yes. If the plan permits or requires vendor determinations, is the employer's adoption discretionary? Yes. Once on this path, there is no place to stop.
The employer, in the plan document, is assumed to appoint a plan administrator of some sort. in the quoted language from FAB 2007-02. Reading the language narrowly, a problem arises only where the employer appoints a third party to make discretionary determinations. It is entirely possible, under the rubric of "administrative functions, including those related to tax compliance," to designate a TPA as plan administrator either in the plan document or otherwise. As to limitations on the TPA's authority:
1-As to loans, there are discretionary determinations only where there is discretion. If the eligibility and amount rules in the plan document are mechanical (i.e., amount, duration, etc. are formulaic), it is hard to see any discretion.
2-As to hardship, again there can be circumstances where no discretion is exercisable. E.g., the amount is deemed both an immediate and heavy financial need and necessary to satisfy financial need. However, there also can be circumstances under which there is discretion, as the DOL understands that term.
On the flip side, where a participant has more than one product vendor, it is possible for the participant to request loans/hardship distributions independently from each. This could easily result in excess loans or hardship amounts. Under the 403(b) regs. (i.e., the second sentence of Regs.1.403(b)-3(b)(3)(ii)), there must be someone performing those limitation functions.
The cautious approach would provide that the TPA simply collect and transmit information to the product vendors, giving the participant a single location for loan and hardship applications, coordinating so that maximum limits are not exceeded. and forwarding the results to the product vendors for discretionary determination. This much is not only permitted, but required.
In theory, there could be a less cautious approach that would attempt to draw a line, as to loans and hardship, between mechanistic determinations and discretionary ones. For example, a safe harbor/safe harbor hardship determination might be permitted, but others referred to the product vendors. Functionally, this runs into two problems. First, having ordinary TPA employees draw those sorts of lines will (not may) result in errors. Second, the product vendor has to make its own determination, under the terms of its own contract, as to whether or not the loan/hardship is permitted, and will not rely on the TPA unless the venodr has selected the TPA.
A third theoretical option is available. The plan document can designate a plan administrator (say, a corporate officer/nonparticpant) that can retain the services of a TPA. Unfortunately, this runs into the "what caused the big bang" problem. That is, appointing someone to appoint someone else to exercise discretion is, in the eyes of the DOL, a discretionary decision. So, probably a no-go here as well.
I am no philosopher or logician, but DOL seems to be unaware of two basic logical/linguistic fallacies.
First, they take the position that the kinds of factual determinations underlying loans and hardships are discretionary. Clearly, in the real world, this is not true all the time. And, even where there are factual issues, they are, in fact, factual. Thus, such determinations are inherently unlike decision such as whether or not to have a 403(b) plan. DOL appears to be confused by the routine language inserted in plans (broadest discretion, etc.) to get an abuse of discretion review of plan decisions into believing that all actions are discretionary. Is the determination as to whether an employer is of the sort that can offer the 403(b) catch-up any less discretionary than the determination that a person's house has burned down?
There is a pretty clear argument that this source of discretion can be eliminated simply by stating that the standard of decision for plan determinations other than as to tax compliance issues is nondiscretionary and the standard of review is de novo. And, in the kind of plan we are talking about (loosely connected products to which the employer makes no contribution), the kind of exposure to costs that motivates the seeking for an abuse of discretion standard of review is either reduced or absent. To the extent that the participant gets money, it is, after all, the employee's money. This is the fourth theoretical option.
The second fallacy is the illogic of infinite regression. Anything that is a necessary, or "but for," logical or sequential precondition to the determination of a discretionary determination can be caught under this fallacy. If the employer allows or requires a product vendor to make a determination, is that employer decision discretionary? Yes. If the plan permits or requires vendor determinations, is the employer's adoption discretionary? Yes. Once on this path, there is no place to stop.
Monday, June 15, 2009
Choosing between 401(k) and 403(b)
BNA has published a brief article on choosing between 403(b) and 401(k) formats for retirement plans at http://www.bnatax.com/tm/insights_kenty.htm. The summary is well written and brief, both of which are real virtues. I have only two comment.
First, one of the stated reasons for choosing 403(b) is the availability of investments other than annuities and mutual funds. This is certainly true, but warrants a qualifier. Since an employer can have all three of a qualified plan, a 403(b) and a 457 plan, this is not an either/or choice. For example, to the extent the pressure for other types of investments comes from a short list of employees, a complementary 457 plan would provide a pressure release valve (in addition to allowing adding the 457 limitation to the applicable variant of 415 and 402(g) limitations). Also, all such issues are irrelevant if the 403(b) would be a church plan.
Second, the 403(b) version of the overall 415 limitation is systematically higher than the 401(k) version. The primary reason for this is the use, in defining compensation, of the most recent full-time equivalent year of service rather that a 12-month measurement period.
First, one of the stated reasons for choosing 403(b) is the availability of investments other than annuities and mutual funds. This is certainly true, but warrants a qualifier. Since an employer can have all three of a qualified plan, a 403(b) and a 457 plan, this is not an either/or choice. For example, to the extent the pressure for other types of investments comes from a short list of employees, a complementary 457 plan would provide a pressure release valve (in addition to allowing adding the 457 limitation to the applicable variant of 415 and 402(g) limitations). Also, all such issues are irrelevant if the 403(b) would be a church plan.
Second, the 403(b) version of the overall 415 limitation is systematically higher than the 401(k) version. The primary reason for this is the use, in defining compensation, of the most recent full-time equivalent year of service rather that a 12-month measurement period.
Defective Contract Exchanges
The Segal Group issued a timely and well-written reminder of the June 30th deadline for defective contract exchanges from the 9/24/07 to 12/31/08 time period. It can be found at http://www.sibson.com/publications-and-resources/compliance-alert/archives/?id=1279.
Friday, September 5, 2008
Church Plan Effective Dates
The Pension Protection Act Blog has a concise, clear explanation of the effective dates of the final regulations for church 403(b) plans.
The post is at http://qualifiedpensionconsulting.com/ppablog/2008/08/21/final-403b-regulation-date-tricky-for-church-plans/
The post is at http://qualifiedpensionconsulting.com/ppablog/2008/08/21/final-403b-regulation-date-tricky-for-church-plans/
New Teacher Pay Rule from IRS
On July 21, 2008, in Notice 2008-62, the IRS created a new rule on public school teacher compensation. In a nutshell, the rule says that “recurring part year compensation” will not be treated as deferred compensation under Section 457(f) as long as “(1) the arrangement does not defer payment of any of the recurring part-year compensation beyond the last day of the 13th month following the beginning of the service period and (2) does not defer from one taxable year to the next taxable year the payment of more than the applicable dollar amount under 402(g)(1)(B) in effect for the calendar year in which the service period begins ($15,500 for 2008).” “Recurring part year compensation” is defined at Regs. 1.409A-2(a)(14) as follows:
“the term recurring part-year compensation means compensation paid for services rendered in a position that the service recipient and service provider reasonably anticipate will continue on similar terms and conditions in subsequent years, and will require services to be provided during successive service periods each of which comprises less than 12 months (for example, a teacher providing services during a school year comprised of 10 consecutive months), and each of which periods begins in one taxable year of the service provider and ends in the next such taxable year. The rules of this paragraph (a)(14) apply to a particular amount of compensation only once, so that an amount deferred under this rule may not again be treated as recurring part-year compensation for purposes of this paragraph and subject to a second deferral election under this paragraph (a)(14).”
The reason for creating this rule is to allow teachers and other eligible public school employees to annualize their compensation. Typically, this allows a teacher to receive pay in equal amounts from the beginning of a school year (say, September) to the beginning of the next school year (say, the following August). The unwillingness of the IRS to extend the rule to non-teachers, incidentally, serves as an example of the clout of K-12 teachers and otherwise has absolutely no justification. So far, so good.
There are two problems with this rule.
First, the very notion of regulating deferral of compensation within a year is bizarre. Note the limitation that the rule does not allow deferral to a new taxable year of the teacher. However, the rule not only limits how much may be deferred between the two years but how those amounts must be paid within the second year. The rule certainly could have been written more clearly in terms of the inter-year deferral, and in ways that do not regulate the payment structure in the second year. For example, the rule could have allowed deferrals of recurring part year compensation up to a limit of the lesser of the 402(g) limitation or the inter-year deferral amount, and then defined that second amount as the compensation that would have been paid in the first year minus the amounts actually paid. This would more clearly define the amounts limit the regulatory impact to inter-year deferrals and make the rule more easily applied.
Second, how does the Section 402(g) limitation get involved here? At no point does either 457(f) or 409A refer to 402(g) and neither of them contains the detailed language necessary to address an issue as narrow as annualized compensation elections. There is certainly no support in the statute for this add-on and, given the real-world reasons for and ubiquity of teacher annualization, I can think of no legitimate regulatory purpose for it. This is simply another example of the IRS making up the rules on the fly and of a disregard for the difference between interpretive regulations (in which the IRS is supposed to flesh out the rules in the statute) and legislative regulations (where the IRS is authorized to make up new rules).
At the least, the fact of application of the 402(g) limitation shows what deferral is being regulated, the inter-year deferral. The absurdity of regulating intra-year deferral supports this premise.
Of course, one of the problems in dealing with the entire non-401(a) compensation area is collateral implications. Absent the specific language of proposed regulations, it is difficult to resolve these issues, so the following comment are just tentative. What effect does this rule have on 457(b) and 403(b)? Or, for that matter, on 401(k)? The major possible issues are:
1. Section 401(a).
(a) The annualized compensation structure may affect plan operations. Assume the qualified plan has a compensation measurement year and a limitation year of 8/1-7/31. In that case, an annualization election will have the effect of shifting some compensation accrued during on compensation/limitation year back one year. This affects both the application of 415 and the compensation base for determining or allocating contributions. Setting aside the potential for a 415 violation as minor, the deferral could certainly affect a contribution of, say, 6% of compensation or a matching contribution of, say, 100% of the first 6% of compensation, particularly if the contribution rate varies between years. (And fixing that differential would involve individually designed plan-type features, the calculation of compensation separately for allocations and testing and the running of a general nondiscrimination test.)
(b) The annualization can reduce the ability of TPAs to catch unreported severances. Many TPAs rely on the cessation of compensation as a backup for employer reporting of severances. To the extent the amounts not yet paid on severance of an annualized teacher are not paid in a balloon payment, this backup will not work until the annualization period is completed.
(c) The annualization structure creates the potential for some of the compensation to be paid more than 2-1/2 months after severance, particularly if the “annualization” uses a 13-month period. Since the 415 regulations only allow the plan to take into account compensation paid before the later of 2-1/2 months from severance or the end of the limitation year, this could have a meaningful effect on severance. Using the same plan and assuming a 13-month annualization from 9/1 of the first year and termination on 6/1 of the second year, the 2-1/2 months expires on 8/15, while compensation continues to be paid through 9/30. Since the limitation year ends on 7/31, compensation paid after 8/15 is not taken into account under 415. The 9/1 and 6/1 dates are admittedly artificial, and the more likely scenario is 8/15 and 6/15, in which a 12-month annualization coincides with the end of the 2-1/2 months, more or less, so that the issue is minor, It can, nonetheless, cause violations.
(d) At no point does the rule say that the annualized amounts can or cannot be the subject of cash or deferred elections. Because of the 2-1/2 month limitation under the 415 regulations, it is possible that annualized amounts will not be eligible for deferral because the amounts would not be included in 415 compensation. Given that the employer and TPA may not have properly classified the employee as terminated, this is a specific risk of the annualized compensation structure that does not get fixed easily, with a higher risk of error when the fact pattern occurs than, say, a matching or employer contribution determination made after the end of the plan year.
(e) If the qualified plan has an active service requirement (say, employed on the last day of the plan year), the employee could lose a match or employer contribution as to the amount pushed into the plan year of termination. In addition, the last day rule may not be applied properly in the year of severance because of the continuation of compensation after the date of severance.
(f) The deferral limits are personal, and based on the taxable year of the employee. Thus, shifting within that year has no effect on how they are determined or applied. There is, of course, the possibility that changing elections will change deferrals (For example, increasing deferrals for the summer because of other compensation from summer employment is a possibility.); this simply afford more flexibility. The higher risk of administrative errors is a constant.
2. Section 403(b). The issues are essentially the same, except that the 415 limitation is not really relevant because the 403(b) rules are based on the employee's taxable year. As a result, the limitation that the annualization not defer to subsequent years of the employee prevents any non-technical 415 significance. In addition, most 403(b) plans will use a calendar year for all purposes to avoid the need to calculate the 415 limitation and other compensation-related items on separate bases.
3. Section 457. Ditto for 457. The limitations in 457 are based on taxable years of service providers and most are calendar year plans.
“the term recurring part-year compensation means compensation paid for services rendered in a position that the service recipient and service provider reasonably anticipate will continue on similar terms and conditions in subsequent years, and will require services to be provided during successive service periods each of which comprises less than 12 months (for example, a teacher providing services during a school year comprised of 10 consecutive months), and each of which periods begins in one taxable year of the service provider and ends in the next such taxable year. The rules of this paragraph (a)(14) apply to a particular amount of compensation only once, so that an amount deferred under this rule may not again be treated as recurring part-year compensation for purposes of this paragraph and subject to a second deferral election under this paragraph (a)(14).”
The reason for creating this rule is to allow teachers and other eligible public school employees to annualize their compensation. Typically, this allows a teacher to receive pay in equal amounts from the beginning of a school year (say, September) to the beginning of the next school year (say, the following August). The unwillingness of the IRS to extend the rule to non-teachers, incidentally, serves as an example of the clout of K-12 teachers and otherwise has absolutely no justification. So far, so good.
There are two problems with this rule.
First, the very notion of regulating deferral of compensation within a year is bizarre. Note the limitation that the rule does not allow deferral to a new taxable year of the teacher. However, the rule not only limits how much may be deferred between the two years but how those amounts must be paid within the second year. The rule certainly could have been written more clearly in terms of the inter-year deferral, and in ways that do not regulate the payment structure in the second year. For example, the rule could have allowed deferrals of recurring part year compensation up to a limit of the lesser of the 402(g) limitation or the inter-year deferral amount, and then defined that second amount as the compensation that would have been paid in the first year minus the amounts actually paid. This would more clearly define the amounts limit the regulatory impact to inter-year deferrals and make the rule more easily applied.
Second, how does the Section 402(g) limitation get involved here? At no point does either 457(f) or 409A refer to 402(g) and neither of them contains the detailed language necessary to address an issue as narrow as annualized compensation elections. There is certainly no support in the statute for this add-on and, given the real-world reasons for and ubiquity of teacher annualization, I can think of no legitimate regulatory purpose for it. This is simply another example of the IRS making up the rules on the fly and of a disregard for the difference between interpretive regulations (in which the IRS is supposed to flesh out the rules in the statute) and legislative regulations (where the IRS is authorized to make up new rules).
At the least, the fact of application of the 402(g) limitation shows what deferral is being regulated, the inter-year deferral. The absurdity of regulating intra-year deferral supports this premise.
Of course, one of the problems in dealing with the entire non-401(a) compensation area is collateral implications. Absent the specific language of proposed regulations, it is difficult to resolve these issues, so the following comment are just tentative. What effect does this rule have on 457(b) and 403(b)? Or, for that matter, on 401(k)? The major possible issues are:
1. Section 401(a).
(a) The annualized compensation structure may affect plan operations. Assume the qualified plan has a compensation measurement year and a limitation year of 8/1-7/31. In that case, an annualization election will have the effect of shifting some compensation accrued during on compensation/limitation year back one year. This affects both the application of 415 and the compensation base for determining or allocating contributions. Setting aside the potential for a 415 violation as minor, the deferral could certainly affect a contribution of, say, 6% of compensation or a matching contribution of, say, 100% of the first 6% of compensation, particularly if the contribution rate varies between years. (And fixing that differential would involve individually designed plan-type features, the calculation of compensation separately for allocations and testing and the running of a general nondiscrimination test.)
(b) The annualization can reduce the ability of TPAs to catch unreported severances. Many TPAs rely on the cessation of compensation as a backup for employer reporting of severances. To the extent the amounts not yet paid on severance of an annualized teacher are not paid in a balloon payment, this backup will not work until the annualization period is completed.
(c) The annualization structure creates the potential for some of the compensation to be paid more than 2-1/2 months after severance, particularly if the “annualization” uses a 13-month period. Since the 415 regulations only allow the plan to take into account compensation paid before the later of 2-1/2 months from severance or the end of the limitation year, this could have a meaningful effect on severance. Using the same plan and assuming a 13-month annualization from 9/1 of the first year and termination on 6/1 of the second year, the 2-1/2 months expires on 8/15, while compensation continues to be paid through 9/30. Since the limitation year ends on 7/31, compensation paid after 8/15 is not taken into account under 415. The 9/1 and 6/1 dates are admittedly artificial, and the more likely scenario is 8/15 and 6/15, in which a 12-month annualization coincides with the end of the 2-1/2 months, more or less, so that the issue is minor, It can, nonetheless, cause violations.
(d) At no point does the rule say that the annualized amounts can or cannot be the subject of cash or deferred elections. Because of the 2-1/2 month limitation under the 415 regulations, it is possible that annualized amounts will not be eligible for deferral because the amounts would not be included in 415 compensation. Given that the employer and TPA may not have properly classified the employee as terminated, this is a specific risk of the annualized compensation structure that does not get fixed easily, with a higher risk of error when the fact pattern occurs than, say, a matching or employer contribution determination made after the end of the plan year.
(e) If the qualified plan has an active service requirement (say, employed on the last day of the plan year), the employee could lose a match or employer contribution as to the amount pushed into the plan year of termination. In addition, the last day rule may not be applied properly in the year of severance because of the continuation of compensation after the date of severance.
(f) The deferral limits are personal, and based on the taxable year of the employee. Thus, shifting within that year has no effect on how they are determined or applied. There is, of course, the possibility that changing elections will change deferrals (For example, increasing deferrals for the summer because of other compensation from summer employment is a possibility.); this simply afford more flexibility. The higher risk of administrative errors is a constant.
2. Section 403(b). The issues are essentially the same, except that the 415 limitation is not really relevant because the 403(b) rules are based on the employee's taxable year. As a result, the limitation that the annualization not defer to subsequent years of the employee prevents any non-technical 415 significance. In addition, most 403(b) plans will use a calendar year for all purposes to avoid the need to calculate the 415 limitation and other compensation-related items on separate bases.
3. Section 457. Ditto for 457. The limitations in 457 are based on taxable years of service providers and most are calendar year plans.
Subscribe to:
Posts (Atom)