Blogging Employee Benefits

October 31, 2006

Key ADEA Ruling from Divided En Banc 6th

Filed under: ADEA, Disability, Litigation — Fuguerre @ 5:37 pm

By demonstrating that a retirement plan’s disability benefits provision facially discriminates on the basis of age, the EEOC has established a prima facie case of age discrimination, according to a divided en banc 6th Circuit ruling reversing its original panel’s decision. (Equal Employment Opportunity Commission v. Jefferson County Sheriff’s Department, 03-6437) The full 6th further ruled that no evidence of discriminatory animus is required, since discriminatory intent was directly evidenced by the facially discriminatory nature of the plan. Perhaps somewhat less clear was the court’s rejection of its precedent in Lyon v. Ohio Education Assn and Professional Staff Union, by which earlier proceedings in Jefferson County had felt bound.

Prior to July 2000, a participant under the Kentucky Retirement System was ineligible to receive disability retirement benefits unless the individual was less than normal retirement age at disablement. After the current litigation commenced, an amendment modified the key condition to exclude from disability retirement any member who was eligible for an unreduced retirement allowance. The en banc panel found this plan design to be facially discriminatory on the basis of age in at least two ways: (1) Employees who remain in active service beyond a specified age are excluded from a particular employment benefit because of age; and (2) Even members who do become entitled to disability benefits receive lower benefits under the plan than similarly situated younger members.

Lyon‘s definition of a prima facie ADEA claim can no longer stand.

Previous rulings for this case had followed much the same path considered by the en banc panel, but had followed Lyon precedent, which had held the opposite of the two key rulings in the current case under arguably similar circumstances. Retracing steps all the way back to the origins of ADEA itself, the 6th now seems to see it differently, stating, “We believe that [OWBPA’s] legislative history is compelling evidence that when revising the ADEA in response to Betts, Congress intended to prohibit the very sort of age-based discrimination that the original panel [in the current case], bound by Lyon, condoned in this plan.” Yet although concurring with the majority opinion, Circuit Judge Rogers observes inter alia not only that it is unnecessary to overrule Lyon except to the extent that former precedent is inconsistent with the current holding, but that this decision leaves us with a fair degree of uncertainty, wryly quipping, “I would leave to future litigants the task of going through Lyon and identifying what survives and what does not.”

Dissenting opinion found relevant distinction between the Kentucky Retirement System design and the landmark Betts template, the crucial distinction being that Betts’ Ohio plan provided younger workers with a specific benefit that was withheld from older workers, whereas the Kentucky system’s design provides disability benefits that are intrinsically no different than normal retirement benefits. Perhaps the point at which one ought to commence Rogers’ task, then, might be to speculate which long-accepted normal retirement benefit designs could now come under renewed scrutiny, with an ever-evolving light?


October 26, 2006

415 and 401(k) COLAs Revisited

Filed under: 401(k), Pensions — Fuguerre @ 2:27 pm

So we have the official word from the IRS on next year’s 415 limits, as previously noted here. But what if you wish to project an advance estimate of a future threshold, such as we did back here? When I forwarded inquiries about the innards of the little spreadsheet I’d been given over to my pet actuary, I was sent a copy of this IRS information letter from 1993.

For current thresholds that employ the same methodology as Code §415(d), one needs monthly CPI-U results for the applicable base period and for the year preceding the year for the limit, the original level for the threshold for the year following the base year, a rounding multiple comparable to the numbers given in 415(d)(4), and the particular calculation methodology used by the IRS (including a curious truncate-round two-step in the middle of the calculation).

Let’s walk it through several instructive examples from this year’s edition –

  • Limitation on Annual Addditions to Defined Contribution Plan, Code §415(d)(1)(C).
    1. Base period is 2001. CPI-U index numbers for July, August, and September 2001 were 177.5, 177.5, and 178.3 respectively. The sum of those three numbers is 533.3.
    2. For the 2007 threshold, we need the comparable CPI-U numbers from 2006, which for July, August, and September 2006 were 203.5, 203.9, then quite surprisingly dropping to 202.9. The sum of those three numbers is 610.3.
    3. The ratio of the second sum to the first sum is something like 1.14438024 (i.e., 610.3/533.3, here carrying that result out to enough places to continue), but then here comes the odd little truncate-round two-step: (a) The 1.14438024 result is truncated to 5 decimal places, giving us 1.14438; then (b) That intermediate result is rounded to 4 decimal places, yielding 1.1444.
    4. For the threshold being adjusted, the original level in 2002 was $40,000, which is now multiplied by the truncated-rounded factor of 1.1444, giving $45,776, which doesn’t then require rounding to the nearest dollar since it is already an integer number of dollars.
    5. The truncating multiple for this particular threshold is $1,000. We round down, rather than to the nearest multiple, so dropping $45,776 down to the next lower $1,000 gives us $45,000 for the 2007 limitation on annual additions to a defined contribution plan.

    That drop in this year’s CPI-U index from August to September dropped the result here from my original estimate of $46,000 to the ultimate actual ceiling of $46,000, serving as warning to anyone attempting to play actuary or economist with future inflation expectations.

  • Limitation on Elective Deferrals, Code §402(g)(1).
    1. Base period this time is 2005, since EGTRRA hard-wired the recent years’ ceiling increases. CPI-U index numbers for July, August, and September 2005 were 195.4, 196.4, and 198.8 respectively. The sum of those three numbers is 590.6.
    2. For the 2007 threshold, we use the comparable CPI-U numbers for 2006 as described for the DC annual addition ceiling, yielding the same sum of 610.3 for the July-August numbers.
    3. Since we have a different base period, the ratio of the second sum to the first sum is of course different, looking something like 1.033355909, which truncated to 5 decimal places becomes 1.03335, which rounded to 4 decimal places turns into 1.0334.
    4. For the elective deferral threshold, the starting level is this year’s $15,000, which is now multiplied by the truncated-rounded factor of 1.0334, giving $15,501, which doesn’t then require rounding to the nearest dollar since it is already an integer number of dollars.
    5. The truncating multiple for the elective deferral limit is $500. Rounding $15,501 down to next lower $500 gives us $15,500 for the 2007 limitation on elective deferrals.

    We just barely made that increase! A smidge more of a drop in September’s CPI-U (or comparably different numbers in others of the 6 quarters used for the indexing), and we might have remained at $15,000 for 2007, instead of jumping up to actual ceiling of $15,500, echoing the warning expressed earlier.

  • Limitation on 401(k) Catch-Up Contributions, Code §414(v)(2)(C).
    1. Base period for this is also 2005, so the first step here repeats what we did for elective deferrals.
    2. Ditto.
    3. Ditto.
    4. But here, the starting level is this year’s $5,000, which is then multiplied by the truncated-rounded factor of 1.0334, giving us $5,167, which again doesn’t then require rounding to the nearest dollar since it is already an integer number of dollars.
    5. Although in 2006 the catch-up deferral ceiling is at one-third the level of the elective deferral ceiling, the truncating multiple is the same: $500. Rounding $5,167 down to next lower $500 leaves us at $5,000 for the 2007 limitation on catch-up contributions, and will likely continue to do so for several more years.

    In this instance, although the same warning about future projections, as expressed for the previous two calculations, still does hold (particularly when the unrounded amount nears $5,500 several years from now), for 2007 we were rather safe in expecting that number to remain the same as for 2006.

Actually, there’s no official word that the IRS is still using exactly the same methodology (updated to use 3rd-quarter comparisons and to step up in terms of the rounding multiples, both of which were introduced after the 1993 IRS letter), or is there any particular guarantee that future methodology will remain the same. For all years since 1993, however, I’m told that this methodology has precisely reproduced all of the official figures published by the IRS, even when there were close calls such as this years $15,501. So unless and until anything more formal is published, this looks good enough for my needs.

(P.S. – In case I haven’t made it rather obvious by now, I absolutely adore what Cornell has done with its U.S. Code!! And as far as I can tell so far, all without making the blunder that some websites make of wrecking all of the links you had previously made to their material. Quite commendable.)

October 25, 2006

Working on Solving the Annuity Puzzle

Filed under: Distributions — Fuguerre @ 7:04 am

As traditional pensions go through a sea change of global freezing from defined benefit pension plans to individual account defined contribution plans, three main issues have required attention: (1) participation rates of employees in defined contribution plans, which typically have been more voluntary in nature than their defined benefit predecessors; (2) investment risk on retirement savings, shifted to employees during their working life; and (3) mortality risk during former employees’ retirement years, the danger that individuals and their spouses will outlive their defined contribution plan accounts.

We’ve seen the first two of those issues addressed by Congress in recent years, most recently in the Pension Protection Act of 2006, which included measures aimed at facilitating automatic enrollment and enhancing participant investment education, among other provisions reflecting the DB-to-DC transmogrification. But as yet, the annuity puzzle – how to encourage the converse of lump sum distributions from a DB plan – exists only in the form of white papers and preliminary legislative proposals.

An excellent summary of the annuitization issues is presented in Lifetime Annuities for US: Evaluating the Efficacy of Policy Interventions in Life Annuity Markets, written by William M. Gentry and Casey G. Rothschild for the American Council of Capital Formation. The paper focuses on the potential value of two different porposed tax incentives designed to encourage voluntary annuitization of retirement savings: one, exclusion of a portion of annuity income from taxable income; and two, refundable tax credits for life annuity income. Although empirical analysis is nonexistent regarding annuitization response to tax incentives, the study’s benchmark results suggest substantial potential effect of the amount of voluntary annuitization in the U.S. economy.

Side note: Although this paper did not touch on the truly diabolical annuity puzzle that will be presented if the Social Security system is ever even partially privatized, let’s put that cousin of this paper’s annuity puzzle off to the side of our desk. Annuities serving Social Security privatization would likely be under a protected market, even if commercial annuities are involved, thus would involve far more mind-bending math formulas than this paper already has.

October 24, 2006

Oregon PERS Amendment Not Unconstitutional

Filed under: Litigation, PERS — Fuguerre @ 3:24 pm

No State shall . . . pass any . . . Law impairing the Obligation of Contracts.

That makes for a reasonably interesting blogging afternoon, to be able to quote from that legal document that starts out, “We the People of the United States, in Order to form a more perfect Union, . . .” In this case, from Art I §10 cl 1, the Contract Clause, sometimes raised by states’ public employees dissatisfied with changes made to a public employee retirement system. In a sense, like the ultimate authoritative version of what private pension plans face under IRC §411(d)(6), not only with respect to previously accrued benefits, but with respect to any future accruals or benefit rights to which the public employees felt they had been granted contractual rights by their state employer.

And on several notable occasions, the employees have prevailed, particularly when the issue of whether a contractual relationship existed was not the issue in dispute. Frequently, though, the employees run up against a long-standing presumption expressed by the U.S. Supreme Court in National Railroad Passenger Corp. v. Atchinson Topeka and Santa Fe Railway Co., quoting from its own earlier decision in Dodge v. Board of Education, maintaining that –

[A]bsent some clear indication that the legislature intends to bind itself contractually, the presumption is that “a law is not intended to create private contractual or vested rights but merely declares a policy to be pursued until the legislature shall ordain otherwise.

In Robertson v. Kulongoski (No. 04-35898), the 9th Circuit Court of Appeals today ruled that although the Oregon Public Employee Retirement System had been expressed in prior state litigation as “a contract between the state and its employees,” there exists no clear indication that the Oregon legislature intended to promise its employees a perpetual, immutable right to contribute to specific accounts under the terms of its retirement system. So although the U.S. Constitution prohibits a state from impairing its contracts, the “contours” of the pension contract between Oregon and its public employees may be open to amendment. Granting due deference to the Oregon Supreme Court’s detailed analysis in Strunk v. Public Employees Retirement Board [OR Sup.Ct., SC S50593, 3/8/05], the 9th Circuit has ruled that Oregon’s 2003 legislation modifying terms of OPERS does not violate the Constitution’s Contract Clause, affirming district court summary judgment in favor of the state. Although as previously blogged here, other aspects of the plan amendment remain in dispute.

Proposed New Accounting Standard for Social Insurance

Filed under: Social Security — Fuguerre @ 12:10 am

The Federal Accounting Standards Advisory Board has published Preliminary Views toward amending its Statement of Federal Financial Accounting Standards No. 17, Accounting for Social Insurance (included in the FASAB’s massive Statements of Federal Financial Accounting Concepts and Standards) to accelerate the government’s recognition of expense and liabilities for social insurance programs, including Social Security, Medicare, and Railroad Retirement. (FASAB Issues Preliminary Views Regarding Accounting for Social Insurance, News Release)

The Preliminary Views present two opposing approaches, a majority Primary View and a minority Alternative View:

  • Primary View – Expense would be recognized when participants in a social insurance program become fully insured, then as benefits increase due to additional work in covered employment by fully insured individuals. For example, under Social Security, expense would be reported for an individual upon completion of 40 quarters of work in covered employment, and a liability for accumulated unpaid expense would be recognized.
  • Alternative View – Expense and liability would continue to be measured and recognized according to current rules under SFFAS 17, upon satisfaction of all eligibility criteria for benefit entitlement, when benefits become due and payable.

Written comments are requested by April 16, 2007 (although the press release gives April 18 as the comment deadline).

October 19, 2006

Pension Villain’s Elegy

Filed under: Pensions — Fuguerre @ 11:57 pm

The risk’s not worth the burden. Time to freeze.
But not to worry: we have a great 401(k)!
This cut will benefit our employees.

FAS 158 gives our balance sheet the squeeze
While our cash projections wobble from PPA.
The risk’s not worth the burden. Time to freeze.

Our workers need to be their own trustees.
Just educate them; they’ll learn to find their way.
So this cut will benefit our employees.

Our old plan’s tangled up in legalese.
The DB pension system’s seen its day.
The risk’s not worth the burden. Time to freeze.

Our competition’s boosted their DCs,
And what works for Wall Street’s good for the U.S.A.
This cut will benefit our employees.

We know you thought we promised more, but please,
Eventually as a hybrid plan we may.
The risk’s not worth the burden. Time to freeze,
And this cut will benefit our employees.

October 18, 2006

2007 Pension Thresholds

Filed under: Pensions — Fuguerre @ 8:35 am

With the BLS reporting the September 2006 CPI this morning, I can finalize my previous estimates of COLAs for key pension thresholds. These numbers remain unofficial, since of course only the IRS has the authority to publish the official numbers. But relying on the same rounding conventions used by the agency for prior years, the numbers I expect to see are as follows –

            2007   2006
    Limitation on Elective Deferrals   §402(g)(1)   $15,500   $15,000
    Limitation on Catchup Deferrals   §414(v)(2)(B)(i)   $5,000   $5,000
    Limitation on Benefits   §415(b)(1)(A)   $180,000   $175,000
    Compensation-Based Limitation Increase Factor   §415(b)(1)(B)   1.0334   1.0383
    Limitation on Annual Additions   §415(c)(1)(A)   $45,000   $44,000
    Limitation on Compensation   §401(a)(17), 404(l)   $225,000   $220,000
    Highly Compensated Employee Determination   §414(q)(1)(B)   $100,000   $100,000

For the elective deferral threshold, the numbers came in just barely over the line needed for the increase to $15,500, with the September CPI-U at 202.9. At 202.8, the limit would have remained level at $15,000.

Older Posts »

Create a free website or blog at