There have been a series of developments earlier this year calling attention to a crisis in multiemployer plan funding. These include:
The January 22, 2013 report to Congress on Multiemployer Pension Plans submitted by the Department of the Treasury, the Department of Labor (DOL), and the Pension Benefit Guaranty Corporation (PBGC) (the “Report to Congress”).
The January 29, 2013 FY 2012 PBGC Exposure Report.
The February 19, 2013 report by the National Coordinating Committee For Multiemployer Plans (NCCMP) Retirement Security Review Commission (“Solutions not Bailouts”).
A hearing held by the House Education and the Workforce Committee’s Health, Employment, Labor, and Pensions Subcommittee on March 5, 2013.
At the March 5 hearing, the Government Accountability Office submitted detailed testimony and stated that “[i]n a matter of weeks, we will be releasing a report that goes into greater detail about the issues [of multiemployer plan and PBGC multiemployer plan insurance system solvency] ….” So – there is more to come.
In this article we review the state of multiemployer plans as reflected in these reports and the hearing. We do not regularly cover multiemployer plan issues in depth. But many large corporate plan sponsors have some (typically limited) multiemployer plan exposure, and the state of the multiemployer plan system is so dire that it may have some effect on both budget deliberations (e.g., both the House and Senate 2013 budgets’ call for increases in PBGC premiums) and on how some policymakers view DB plans generally.
Our objective in this article is to describe – based on these recent reports – “what is going on with multiemployer plans” and to consider what risk it may present to sponsors. We assume that sponsors with material multiemployer plan exposure will already be familiar with these issues for the plans to which they contribute. For sponsors with no multiemployer plan exposure this article may still be of some interest, as the multiemployer plan world is struggling with issues (liability valuation, plan finance, PBGC support) that are also present (to a lesser degree) in the single employer DB world and may affect, e.g., policymaker deliberations about single employer plans. For sponsors with a non-material exposure to the multiemployer plans, this article may serve as a heads-up.
Multiemployer plans in perspective
Multiemployer plans (also known as Taft-Hartley plans) are plans sponsored by more than one unrelated employer under a collective bargaining agreement and administered by a joint union-management board of trustees. In this article we will focus on defined benefit multiemployer plans, not defined contribution plans or health and welfare plans.
Unless your company is in, e.g., trucking, coal, or one of the building trades (where these plans dominate), typical company involvement is simply paying a contribution (e.g., on a cents-per-hour basis) to a plan for each employee covered by a collective bargaining agreement under which the employee accrues benefits.
The Report to Congress states that multiemployer plan assets, as of 2010, were $366 billion. Total private sector employer-sponsored retirement plan assets (DB and DC) has been estimated (as of 2011) to be $6.263 trillion; so multiemployer plan assets represent a little more than 5% of total private sector retirement plan assets.
Multiemployer plan contributions — impact on earnings
In the past, companies typically only reported the cash cost of (generally, annual contributions to) multiemployer plans as an expense on their financial statements. But there is additional exposure: for instance, under ERISA, if you withdraw from a multiemployer plan, you may be liable to pay a share of the plan’s unfunded liability, which may include unfunded liability for employees of other, defaulting employers. (Withdrawal liability is discussed further below.)
In 2011, the Financial Accounting Standards Board (FASB) adopted an Accounting Standards Update that requires multiemployer plan sponsors to disclose:
The significant multiemployer plans in which an employer participates.
The level of an employer’s participation, including contributions and whether they represent more than 5% of total plan contributions.
The financial health of the significant multiemployer plans, including funded status, whether funding improvement plans are pending or implemented, and whether the plan has imposed contribution surcharges.
The nature of the employer commitments to the plan, including expiring collective-bargaining agreements and whether those agreements require minimum contributions.
Certain additional disclosures where the plan does not file a publicly available Form 5500.
The final Accounting Standards Update represented a watering-down of FASB’s original proposal, which was to require disclosure of estimated withdrawal liability.
In multiemployer plans, if an employer’s annual contributions are the ‘tip,’ withdrawal liability is the iceberg. Generally, under ERISA, a contributing employer is assessed withdrawal liability when it ceases making contributions under a plan, e.g., when it goes out of business or terminates (or does not renew) the collective bargaining agreement under which contributions were being made.
How much withdrawal liability a contributing employer has (or may have) is a very complicated question. ERISA provides (generally) that a plan may choose among four different methods for allocating a plan’s unfunded vested benefits (UVBs) to a withdrawing employer. Bottom line: withdrawing employers are generally allocated a proportional share of the plans total UVBs, which typically (in the current environment at least) include liabilities for participants who were never employees of the withdrawing employer.
Recent developments have increased withdrawal liability exposure significantly. Quoting from the Report to Congress:
The recent increase in plan underfunding has caused withdrawal liability assessments to soar. Also, because an employer’s annual withdrawal liability payments are based on the employer’s highest contribution rate in the last ten years, recent contribution rate increases have generally added to the amount an employer would be obligated to pay in annual withdrawal liability payments.
In the case of a ‘mass withdrawal’ — when all or substantially all of the employers withdraw from the plan — additional withdrawal liability is generally assessed on employers withdrawing in the year of the mass withdrawal or the previous two years, including a share of the plan’s remaining UVBs determined using (generally much more conservative) PBGC assumptions.
None of the foregoing adequately deals with the potential withdrawal liability risk to a contributing sponsor. There are many instances in which, for instance, the controlled group rules that apply to the assessment of withdrawal liability may produce (unpleasantly) surprising and non-intuitive results. And special rules apply in some industries, including the construction industry. An adequate assessment of a sponsor’s withdrawal liability exposure can only be made with the help of multiemployer plan experts, including counsel and actuaries with withdrawal liability evaluation expertise.
The problem of liability valuation
A critical problem for companies trying to understand multiemployer plan exposure is the different liability valuation rules that apply to multiemployer plans. The generally market-interest rate based rules that apply for accounting and funding for single employer plans do not apply to multiemployer plans. Instead, the liability valuation interest rate must simply be ‘reasonable.’ Current practice is, generally, to use a valuation assumption of around 7.5%; that is compared to, for instance, a valuation assumption under FAS 87 for single employer plans that is (for 2013) generally (for a plan of typical duration) in the low 4% range. So (some would argue) even multiemployer plans that are said to be ‘well funded’ may be in trouble if more ‘market-based’ assumptions are used.
The Report to Congress, in a number of places, restates multiemployer plan liability valuations based on PBGC assumptions, which are derived from annuity purchase rates and are generally regarded as very conservative. The following charts (based on data in the Report to Congress) show multiemployer plan funding for the period 2000-2010 based on those more conservative assumptions.
While status certifications in 2010 and 2011 show a clear improvement in plans’ funding status (between 2009 and 2011, critical status plans dropped from more than one-third to fewer than one-quarter of all plans, and green status plans increased from nearly one-third to 60% of all plans), it is difficult to distinguish the effects of funding relief from other possible sources of funding improvement.
Moreover, ‘improved funding’ results are based (generally) on a 7.5% valuation interest rate. Single employer plans (e.g., for accounting purposes) use a much lower rate and have had to contend with interest rates that have declined another 100-150 basis points since 2010.
Prior to the Pension Protection Act of 2006 (PPA), multiemployer plans had a lot of flexibility in how benefits were funded: past service liability was generally amortized over 30 years (pre-ERISA past service liability could be funded over 40 years); actuarial assumptions only had to be reasonable ‘in the aggregate;’ and a variety of funding methods could be used.
In PPA, Congress, to some extent, tightened multiemployer plan funding rules, e.g., reducing the past service liability amortization period to 15 years (for post-PPA past service liability) and requiring that interest rate assumptions be reasonable on a standalone basis. PPA also established rules for the mandatory remediation of plan underfunding. Each year a plan’s actuary must certify whether the plan is in critical status (‘red zone’), endangered status (‘yellow/orange zone’) or neither (‘green zone’). Endangered plans must adopt a funding improvement plan; plans in critical status must apply a contribution surcharge and may cutback certain ancillary benefits (e.g., early retirement subsidies).
But, in 2008 and 2010 Congress provided ‘funding relief’ which reduced the impact of the PPA rules. Quoting from the Report to Congress:
The Worker, Retiree, and Employer Recovery Act of 2008 … (WRERA) permitted multiemployer plans to elect a temporary forbearance from certain of the requirements of PPA. The vast majority of these plans certified to be in critical, seriously endangered, and endangered status in 2009 elected to defer actions otherwise required by their status certifications and/or to extend the time for demonstrating progress under their funding improvement or rehabilitation plans. And … [a]s permitted under PRA 2010 [the Preservation of Access to Care for Medicare Beneficiaries and Pension Relief Act of 2010], more than 700 multiemployer plans chose to amortize investment losses incurred in the 2008 market crisis over a 29-year period (nearly twice as long as otherwise required under PPA) and/or to lessen the impact of investment losses on the actuarial value of plan assets used to determine their future funding requirements and funding status.
Despite the substantial improvement in plan assets since the market crisis of 2008, however, certifications of plans’ funded status for the 2011 plan year — showing 60% of all plans to be in green status — likely overstate the extent of plans’ financial health. This is due to the significant effect of PRA 2010 funding relief, which increased plans’ funded percentages (e.g., by allowing plans to spread the recognition of asset losses over ten years) and delayed projected funding deficiencies (e.g., by extending certain amortization periods and reducing minimum required contributions).
What is the real unfunded liability?
Thus, an accounting policy that generally looks only at current cash cost, the use of a valuation interest rate that is substantially higher than the rates used to value single employer plans, and funding rules that delay recognition of losses all serve to obscure the true amount of multiemployer plan underfunding.
Multiemployer plan underfunding is highly concentrated in the construction and transportation industries. The following chart, from the Report to Congress, shows concentration of underfunding in 2010 (based on PBGC valuation rates).
Impending PBGC multiemployer system insolvency
The weakness of the PBGC multiemployer plan insurance system mirrors the weakness in multiemployer plan funding. The numbers here are particularly sobering. Quoting from PBGC’s 2012 Exposure Report:
Our multiemployer program ran out of money in 91% of [the 20-year] simulations. PBGC’s multiemployer insurance program became insolvent in 36% of simulations within the shorter FY 2013-2022 projection period.
This chart (from the FY 2012 PBGC Exposure Report) tells the story vividly:
PBGC multiemployer insurance system insolvency risk is so high that it dwarfs any conceivable premium flow. PBGC projects multiemployer premiums over the next 10 years to total less than $1.3 billion; new claims are projected to be $37.6 billion. To solve its problem with premiums would require a 3000% increase PBGC premiums.
All discussions of the multiemployer plan crisis inevitably involve a discussion of external, ‘macro’ issues: declining private sector unions; concentration in cyclical and declining industries; an upside down demographic pyramid. While it is true that these issues would not be relevant if multiemployer plans were well and conservatively funded, a funding crisis is a lot easier to solve if you have an expanding workforce and an expanding plan contribution and PBGC premium base. Most multiemployer plans do not.
What is going to happen?
If you read the NCCMP’s “Solutions not Bailouts” report, it’s clear that the industry understands there is a problem. That report proposes allowing multiemployer plans more flexibility in funding and in cutting benefits, together with some suggestions that (the title notwithstanding) look something like a (limited) bailout.
There are not a lot of options. All (the NCCMP and the Treasury, DOL and PBGC) seem to recognize that increasing contributions is a very limited solution. If contributions are increased to pay off, in effect, a legacy liability, a company contributing to a multiemployer plan becomes uncompetitive with companies that do not — it will either lose money, lose customers or lose employees.
Cutting benefits seems to be a particularly grim Plan B, especially for participants in or near retirement. Typically, these are not ‘rich’ plans, and PBGC guarantees for multiemployer plans are significantly lower than those for single employer plans. The only other alternative would seem to be some sort of bailout.
Effect on sponsors
As we said at the beginning, sponsors with material multiemployer plan exposure will already be familiar with these issues for the plans to which they contribute. For sponsors with ‘non-material exposure’ – limited contributions for small groups of employees – the multiemployer plan situation has reached such a critical stage that it is worth understanding, for these groups: (1) the financial condition of the relevant plan (based on realistic assumptions); and (2) the plan’s withdrawal liability and withdrawal liability rules.
Not all multiemployer plans are in funding trouble. Just because your company contributes to one does not mean there is a problem. But you should probably find out.
Effect on policymakers
There has developed in recent times a bias amongst certain policymakers against defined benefit plans. The multiemployer plan crisis is likely to reinforce that bias, making it more difficult (in some cases and to some policymakers) to argue for, for instance, funding relief legislation like we saw in 2010 and 2012.
In addition, increased PBGC premiums are now featuring in both sides’ (Republican and Democrat) budget proposals. As we have remarked before, curiously, the PBGC deficit (including the ballooning multiemployer plan deficit) is ‘off budget’ — it does not show up as part of the deficit. But premiums do count as revenues. Certainly, a multiemployer plan funding crisis provides a pretext for premium increases.
And, one conceivable bailout option is to merge the two systems — so far as we are aware, no one has suggested that. But it is an idea that, at some point, will occur to some multiemployer plan advocate.