Pension Rights Center Issues Public Comment on PBGC Rule

WASHINGTON, Aug. 13 -- Terrence Deneen, Michael Gordon fellow, and Norman Stein, senior policy consultant, at the Pension Rights Center, have issued a public comment on the Pension Benefit Guaranty Corporation rule entitled "Special Financial Assistance". The comment was written on Aug. 10, 2021, and posted on Aug. 12, 2021:

* * *

The Pension Rights Center ("the Center") submits the following comments on the PBGC's Interim Multiemployer Financial Assistance Regulation (hereinafter "IR"), published in the Federal Register on July 12, 2021. The Center is a nonprofit consumer organization that has been working since 1976 to protect and promote the retirement security of American workers, retirees, and their families.

We note our strong interest in the subject area of the agency's guidance. Since the early 2000s when some multiemployer plans began to experience signs of financial vulnerability, an important part of the Center's agenda has been to ensure the survival of the multiemployer pension system, which has served several generations of workers and retirees. The Center thus celebrated the passage of the Special Financial Assistance Program for Financially Troubled Multiemployer Plans (hereinafter referred to as the Butch Lewis Act),/1 as part of the American Rescue Plan Act of 2021, which promises to secure the benefit expectations of retirees and those approaching retirement, improve the financial health of the PBGC's multiemployer program, and insulate local economies and the national economy against the potential impacts of plan collapses./2

At the outset, we wish to emphasize that our criticism of parts of the IR is not a criticism of the good faith or competence of the agency staff who were required to develop and publish a detailed set of regulations implementing a highly ambiguous statutory scheme in an unrealistically short period. We also appreciate that PBGC had valid reasons to start from a cautious interpretation of the statute, given the artificially truncated period of time to vet applications, and the fact federal assistance must be paid in a lump sum payment that cannot be retrieved or corrected even if it develops that the federal assistance was improvidently granted.

Unfortunately, the IR does not implement the Butch Lewis Act in a way that reflects the intent or letter of the law Congress enacted earlier this year. The several components of the guidance are not properly coordinated with each other nor with pre-existing statutory requirements, and these disconnects will undermine the long-term financial stability of the relevant plans and ultimately may increase the long term loss to PBGC. Most important, if the IR is not modified to address the failure to fully fund benefit accruals earned by active workers after an application is approved, plans can be expected to fail no later than 2051, the year the guidance has engineered as the failure date for plans receiving aid. And even if we assume the IR framework is consistent with the statutory language, it is literally impossible for either plan actuaries or the PBGC (and its contractors) to calculate several of the required data sets relating to future contributions with any degree of reliability, creating an assistance calculation process that is essentially arbitrary. This cannot be what Congress intended and there are interpretations of the relevant portions of Butch Lewis that are at least as literally plausible as the IR and that do not invite the irrationality and likely perils of the IR./3

Our comments are organized into four substantive sections: (i) prohibition on retroactive benefit increases; (ii) calculation of award amounts; (iii) investment allocation and related issues; and (iv) restrictions on withdrawal liability. We note that our ordering of the issues does not necessarily comport with our sense of their relative significance (although we regard all of the issues as important./4

1. Post-Application Benefit Improvements.

The IR prohibits any plan that has received Federal Financial Assistance ("FA") from increasing benefits on a retroactive basis, regardless of future circumstances. In contrast, the IR does permit increases in future accruals under some circumstances, primarily that they be paid for by new contribution sources not contemplated by the application for relief. The prohibition on retroactive benefit increases and limitation on prospective benefit increases are in addition to the ordinary restrictions on benefit increases for plans in critical status.

Particular decisions on the nature of plan benefit increases when a plan has a new source of contributions adequate to fund the increase and can document that in a manner acceptable to the Corporation should be left to plan trustees and the collective bargaining process. At the least, the IR should recognize a procedure under which a plan may apply for an exception to the prohibition against retroactive benefit increases.

Our concern in this regard is, in part, that some multiemployer plans have been amended under the Pension Protection Act of 2006 to eliminate or reduce certain previously subsidized benefits under the plan. Over the years, we and the pension counseling projects for whom we provide technical assistance have learned of many people who retired at a benefit that was in some only cases less than half of what they expected. To absolutely prohibit a plan from restoring some or all of those benefit cutbacks no matter what the future brings is neither necessary nor appropriate. Moreover, if the plan demonstrably has the resources (especially because of new contribution schemes but really for any reason), the distinction between retroactive and prospective increase is irrational without further qualifications.

The IR should be amended at least to provide for some flexibility for retroactive benefit increases if future circumstances permit it without endangering the plan's ability to pay all benefits.

2. The Amount of Financial Assistance

The Butch Lewis Act provides that if a plan qualifies for financial assistance, the assistance "shall be such amount required for the plan to pay all benefits due during the period beginning on the date of payment of the special financial assistance payment under this section and ending on the last day of the plan year ending in 2051." The IR interpreted this to mean, in effect, the difference between a plan's projected benefit payments through 2051, including future accruals and future administrative expenses, less all plan resources projected to be available to the plan through the end of the 2051 plan year. Plan resources include all pre-application assets, all projected contributions to the plan through 2051, all projected withdrawal liability payments to the plan through 2051, and all earnings on those assets through 2051.

As has been repeatedly noted, the IR's approach to the calculation of the assistance amount--in particular its consideration of all future contributions as a plan asset--is deliberately designed to result in plan insolvency in the 2051 plan year./5

This is not what Congress intended nor is it what the statutory language requires./6

Also, the IR is inconsistent with the structure of the funding rules applicable to multiemployer plans; creates an arbitrary application process bereft of realistic measurements to determine assistance, indeed inserting an unsolvable logical bottleneck into the application process; and may result in some of the 18 plans whose MPRA applications were approved choosing not to apply for special financial assistance, a result that would undercut Congress's clearly stated intent to restore benefits suspended under MPRA. Moreover, the ultimate cost to the PBGC will almost certainly exceed the value of the reduction in assistance that the IR will produce. The IR is, in fact, a penny-wise, dollar-foolish construction of the statute.

The minimum funding rules for multiemployer plans generally require that plan contributions for each year be sufficient to cover normal costs for the year, plus amortize the excess of the value of plan liabilities over the value of plan assets over a 15-year period. The normal costs are intended to cover the costs of benefits accrued during the plan year, including benefits that will be paid after the 2051 plan year. The IR in effect usurps plan contributions that Congress intended to pay for post-2051 payments to pay down pre-application plan liabilities.

The statute does not dictate this result, as the preamble to IR seems to suggest, mistaking congressional silence on the issue as a clear signal that Congress wished to undermine the integrity of its multiemployer funding rules and doom all multiemployer plans qualifying for relief to insolvency no later than 2051. A congressional command to ensure payments to 2051 cannot reasonably be construed as a command to forbid payments in 2052 and after, and nothing in the legislative record supports this view. Yet the IR has turned the 2051 date into the actuarial equivalent of a guillotine.

Put another way, Congress intended that the amount of FA should cover 100% of the difference between the cost of benefits in pay status from 2021-2051 and a plan's "assets available to pay plan benefits" for the same period. Unfortunately, the IR in effect says otherwise, and creates a built-in shortfall between liabilities and resources, resulting in plan insolvency in 2051.

This result will also subject plans receiving relief to increased intergenerational tensions between older and younger participants and will encourage contributing employers to look for the exits, which the IR will in some cases further facilitate by the inadequacy of the conditions it has placed on the calculation of withdrawal liability.

We also note that there is no way an actuary--or anybody else--can gin up a credible present valuation of employer contributions over the next three decades. There are simply too many unknown variables--indeed, all of the variables are unknowable. Who can predict whether an industry will be wiped out by new technology or foreign competition or a new law? Who in 1990 foresaw Amazon and the bankruptcies of retail chains like K-Mart, Sears, Neiman Marcus and JC Penney./7

And how is the actuary to predict future legislative or regulatory changes that will impact a particular plan? And some public health officials have even suggested that the world might see a major pandemic in the 21st Century, which it ever happened could have an effect on future employer contributions.

The demand to "value" future withdrawal liability payments adds another layer of impossibility. At a minimum, this would require the plan to identify which employers will either go out of business or drop their collective bargaining obligation, the date those employers will do so (plan liabilities change, after all), how much the employer is able to pay, and whether a court or arbitrator will sustain the assessment.

We cannot suggest any way to make this process viable. Neither Generally Accepted Accounting Principles nor the funding rules of the Internal Revenue Code allow a plan to treat future contributions or withdrawal payments as an asset. Nor are they treated as a plan asset for plan withdrawal liability calculations. We might add that when the PBGC books liabilities for its annual financial statement, the agency does not simply use figures derived from its PIMS stochastic forecasts: agency staff scrutinize plans on an individual basis, and a given plan may change from "booked" to "unbooked" several times.

Moreover, the amount of future contributions and future withdrawal liability payments is itself related to the amount of financial assistance a plan will receive. If the IR will not ensure that a plan will survive beyond 2051, the support of current employees for plan participation may diminish. Workers (and their bargaining representatives) are not financial illiterates. Our limited review of internet postings related to ARA eligible plans disclosed that workers in a number of plans are already complaining that the IR guidance forces them to sacrifice tens, or even hundreds of thousands of dollars of contributions over their career with no benefits payable to them after 2051 in excess of PBGC multiemployer guarantees./8

Active workers may seek to abandon the plan as quickly as they can, because the value of their lost accrued benefit is only a fraction of the cost of several decades of contribution deductions. And simple economics will lead employers to a similar place, seeking to withdraw from plans.

These potentially corrosive pressures should, logically, substantially reduce an actuary's guess about the level of future contributions. But if such reduced contribution assumptions are accepted by the PBGC and result in an increased amount of financial assistance, the pressures that resulted in conservative estimates of future contributions will be less severe in reality and contributions will exceed the actuarial estimates.

As suggested by others, there is also the risk that some plans that reduced benefits under MPRA will choose not to apply for financial relief, deciding that plan solvency beyond 2051 is too valuable to active employees to surrender in exchange for restoration of benefits. While we hope that no plan would make such a choice, the possibility is difficult to dismiss out of hand, which would result in the continued often severe reduction in benefits of elderly Americans, benefits that the Butch Lewis Act was almost certainly understood to remedy.

The Center believes that a more plausible reading of the statute's instructions would respect the structure of the funding rules and count as a plan asset only the portion of a contribution that exceeds the normal cost of benefit accrual that will not be paid until after the 2051 plan year. Moreover, a close equivalent to this approach, which would also have the virtue of avoiding the essential arbitrariness of trying to predict future contributions, withdrawal liability, and new benefit accruals payable through 2051, would be to disregard as a plan resource post-application normal costs and also disregard all post-application benefit accruals as a benefit payable through 2051./9

3. Alternatives to Investment Grade Securities/Investment Allocations/Relationship Between Financial Assistance Funds and Overall Plan Portfolio

The preamble to the IR invites comments on the issues highlighted above. It is our belief that Congress earmarked investment of government assistance funds in specified investments to shelter such assets from not only investment risk generally, but also from market volatility, which can have extreme adverse impacts on plans whose active participant/retiree ratios are unfavorable. Insulation of government assistance against market risk and volatility would ordinarily suggest a conservative approach to the questions raised by the preamble: for example, that the investment allocation of the legacy plan asset portfolio be evaluated for prudence without consideration of the investment-grade fixed income securities in the segregated account for government assistance; timing restrictions on how quickly the government assistance portfolio can be used to pay benefits; only limited investment alternatives to investment grade securities for the government assistance; and perhaps special restrictive conditions on plans that have in the past taken on significant investment risk without measurably improving long-term return.

But given the interest rate/rate of return disparity in the statute and PBGC's view that it cannot interpret the statute in a way that harmonizes the two for purposes of calculating the amount of assistance a plan will receive,/10 we believe that plans must be provided greater investment flexibility with respect to investment government assistance amounts than would otherwise be consistent with congressional intent to limit investment risk.

We cannot offer specific recommendations, but instead offer a general principle that we think should inform specific decisions on the preamble questions. We believe that the animating principle should be to provide plans with sufficient flexibility to design a total portfolio that has an expected return equal to but not in excess of the statutory interest rate used to calculate the amount of government assistance. This approach would undoubtedly require more individuation among plans and more ongoing monitoring than is administratively ideal, but this seems the most protective path to allowing plans to achieve an expected rate of return that is consistent with congressional intent that assistance last through at least the 2051 plan year.

We also note that PBGC, the Department of Labor, or perhaps both, might consider guidelines on whether, how often and under what circumstances a plan can sell investment grade securities (or other assets approved by PBGC) from its general legacy portfolio to its segregated account for government assistance and visa versa.

4. Withdrawal Liability

The IR concedes that without the PBGC exercising its authority to place conditions on withdrawal liability, the Butch Lewis Act will reduce withdrawal liability obligations and encourage an employer exodus from a plan. This is because the infusion of lump sum FA will immediately lower the value of a plan's unfunded benefits, and consequently reduce an employer's withdrawal liability. Moreover, the ARA creates two additional perverse incentives: first, the most badly funded plans will have the biggest FA and the greatest percentage decrease in withdrawal liability, and second, the FA's immediate, temporary liability reduction encourages employers to withdraw sooner rather than later. And unions presumably would have only a weak interest in resisting, since retirees are taken care of for a lengthy time period, and active employees can accrue benefits in a new plan.

The IR attempts to counteract these incentives by tinkering with the actuarial valuation of a plan's UVBs. Under the 1980 Multiemployer Act, the plan actuary is supposed to use an array of assumptions that, in combination, reflect past experience and the actuary's best estimate of anticipated future experience. The IR specifies, however, that for withdrawal liability calculations only, the actuary must use the conservative, no-risk discount rates applicable in a mass withdrawal. This will increase the calculated present value of a plan's unfunded liability and thereby, it is hoped, either forestall employer withdrawals or at least extract a painful amount of withdrawal liability. We sympathize with this good intention, but have considerable doubt it will work, or at least work for all plans. For one thing, it is our understanding that at least one large plan already uses conservative assumptions for both funding and withdrawal liability calculations. For this plan, the requirement will have little if any impact on the calculation of a plan's unfunded liabilities. More generally, this special condition will almost certainly be the subject of court challenges, and several recent court decisions have overturned liability assessments that use the soi-disant Segal or "Blended rate" valuation rates, which use similar interest assumptions to value plan benefits not covered by plan assets./11

Even if the IR is held to be valid, we are unaware that PBGC has done any research to determine whether the requirement is sufficiently potent to ward off withdrawals. Indeed, we believe that the PBGC should consider adding at least three additional conditions to calculation of withdrawal liability: (1) non-consideration of special financial assistance in calculating withdrawal liability; (2) requiring the use of conservative assumptions for a five-year period after the special financial assistance is exhausted; and (3) requiring for a 15-year period that withdrawal liability be no less than it would have been as of the date a plan applied for special financial assistance.

Moreover, we suggest that the PBGC make appropriate and necessary changes in the IR to prevent further damage to local and regional construction industry plans. Under the special definition of a withdrawal in a construction industry plan a construction employer who ceases operations, or transfers operations outside the jurisdiction of its collective bargaining agreement, incurs no withdrawal liability./12

Obviously, an employer who decides that benefit costs have become too burdensome can just close the shop and be done with pension headaches. In theory, such dropouts should not affect contributions, because all construction is local, so if Corporation A drops out, Corporation B will expand its operations, or a new Corporation C may enter the plan. The reality is otherwise. Contrary to popular belief, many construction plans actually cover a small, discrete geographic area: Pipefitters Local 1 Plan covers county A, Pipefitters Local 99 covers neighboring county B. As the PBGC well knows, in many crafts, an employer who contributes to the Local 99 Plan in County B is allowed to perform work in County A, but contributes to the Local 99 Plan instead of Local 1. Thus, an employer can lawfully transfer operations to the next county, sign up to another plan with lower contribution requirements, and bid on exactly the same work in his former abode, incurring no liability. This phenomenon has already caused the demise of several plans and the IR will exacerbate the problem.

In summary, we respect the efforts and perspectives that went into crafting the IR, but nevertheless believe the IR includes the serious deficiencies discussed above. The suggestions we have made, if adopted, will better conform the PBGC guidance to the statutory language and congressional intent and will result in a healthier, more robust multiemployer system, a system that can be counted on for the long-term. If you have any questions, we would be pleased to respond.

Sincerely yours,

Terrence Deneen, Michael Gordon Fellow, Pension Rights Center

Norman Stein, Senior Policy Consultant, Pension Rights Center

* * *

Footnotes:

1/ Estil "Butch" Lewis was an Ohioan, a Vietnam Veteran, and one-time president of the Teamsters Local 100 Union in Cincinnati. Mr. Lewis became a pension activist after the passage of the Multiemployer Pension Reform Act of 2014 (hereinafter "MPRA") threatened massive pension cuts to participants in multiemployer pension plans. He organized and advocated on behalf of his fellow retirees and union workers and their families, working with public officials in developing a means of saving financially troubled multiemployer plans without devastating reductions in the benefits they had been promised and earned. Mr. Lewis passed away of a stroke in 2015, attributed in part to the strenuous nature of his organization and advocacy work. It was the Center's honor to work with him and his spouse Rita. The House version of the American Rescue Plan Act referred to the provisions of the multiemployer financial assistance program as the Butch Lewis Pension Plan Emergency Relief Act of 2021 ("Butch Lewis Act"), as we do in these comments.

2/ The Center opposed those parts of the Pension Protection Act of 2006, which allowed plans to impose substantial targeted benefit reductions on individuals who had done nothing wrong except rely on their benefit promises. It was among the early supporters of partition remedy legislation for multiemployer plans burdened by "orphaned" participants and in 2014 the Center actively opposed MPRA, warning that it would lead to disproportionately savage benefit reductions for retirees and those just approaching retirement, in other words, those for whom it was too late to make up the losses. And since the passage of MPRA, we have worked with grassroots groups of retirees, with organized labor, and with industry, toward passage of legislation that would restore the financial health of troubled multiemployer plans without devastating benefit cuts.

3/ We do not agree that PBGC's apres 2051, le deluge construction of the ARA is the most plausible reading of the statute, but our comments will assume that the IR and the Final Regulation will nevertheless survive the inevitable judicial challenge.

4/ For example, we discuss our concerns with the prohibition against retroactive benefit increases first because we suspect we will be among only a few commenters objecting to the prohibition and did not want those concerns lost by discussing them at the end of our comments.

5/ And this does not take into consideration the problem of the mismatch between the interest rate used to calculate assistance and the lower rate of return on the investment grade fixed income securities in which Congress intended plans to invest, a problem we will consider infra.

6/ If blind fidelity to the statutory language is the goal, we would argue that the PBGC should simply determine the present value of plan liabilities through 2051 (or perhaps only already accrued plan liabilities) and provide assistance in that amount, without considering the so-called, just invented concept, of plan resources.

7/ In the 1990s, PBGC treated the anticipated rental income from real estate leased to a JC Penney store as high as the earnings on a AAA bond. No longer.

8/ Moreover, given the accrual-based structure of the PBGC guaranty, large swathes of current active workers with vested benefits have already accrued enough service to obtain the full PBGC guaranty.

9/ We have not commented here specifically on the problems of the mismatch between the discount rate specified in the statute and the expected return on the investment grade securities in which special financial assistance must be invested. But we note that we agree with the view the statute can be interpreted reasonably in a way that corrects for this. We are not providing detailed comments in this regard because we understand that PBGC is aware of such arguments and despite the glaring illogic of a discount rate that exceeds plausible rates of return on assets (and will result in plan illiquidity before 2051) has apparently rejected them.

10/ And of course the larger the amount of financial assistance a plan receives, the more profound the implications of the mismatch become.

11/ This is a simplification, but we are preaching to the choir.

12/ ERISA Sec. 4203.

* * *

The rule can be viewed at: https://www.regulations.gov/document/PBGC-2021-0003-0001

TARGETED NEWS SERVICE (founded 2004) features non-partisan 'edited journalism' news briefs and information for news organizations, public policy groups and individuals; as well as 'gathered' public policy information, including news releases, reports, speeches. For more information contact MYRON STRUCK, editor, [email protected], Springfield, Virginia; 703/304-1897; https://targetednews.com