Roughly 1,400 multiemployer plans cover workers in an array of unionized industries, including retail, service, construction, manufacturing, mining, transportation, and entertainment.
Of the more than 10 million people who participate in multiemployer pension plans, approximately 1 million are in 100 plans that are projected to be unable to pay the full benefits that have been promised.
The PBGC multiemployer program, which pays benefits for those plans that fail, is in dire financial condition and is likely to exhaust asset reserves in about eight years.
Tackling the multiemployer pension plan issue will require solutions that focus on securing “legacy” pensions and also assuring a secure retirement system in the future.
There are only two ways to remedy the situation—infuse more money into the plans or reduce benefits.

1850 M Street NW
Suite 300
Washington, DC 20036
202-223-8196 |

Craig Hanna, Director of Public Policy
Ted Goldman, Senior Pension Fellow

© 2017 American Academy of Actuaries. All rights reserved.
Members of the Multiemployer Plans Subcommittee are: Josh Shapiro, MAAA, FCA, FSA, EA—chairperson; Jason Russell, MAAA, FSA, EA—vice chairperson; Christian Benjaminson, MAAA, FCA, FSA, EA; James Dexter, MAAA, FCA, FSA, EA; Aldwin Frias, MAAA, FCA, FSA, EA; Don Fuerst, MAAA, FCA, FSA, EA; Duke Gowen, MAAA, ASA, EA; Stan Goldfarb, MAAA, FSA, EA; Ted Goldman, MAAA, FCA, FSA; Eli Greenblum, MAAA, FCA, FSA, EA; Peter Hardcastle, MAAA, FSA, FIA, EA; Joseph Hicks, MAAA, FCA, MSPA, EA; Ethan Kra, MAAA, FCA, FSA, FSPA, CERA, EA; Brian O’Konski, MAAA, ASA, EA; Steve Rabinowitz, MAAA, FCA, FSA, EA; James Shake, MAAA, FCA, EA; Andy Smith, MAAA, FCA, ASA, EA; and Peter Sturdivan, MAAA, FSA, EA.

 For a print-ready PDF of this page, click here.

Overview of Multiemployer Pension System Issues

Of the more than 10 million people who participate in multiemployer pension plans, approximately 1 million are in plans that will be unable to pay the full benefits they have been promised under current projections. The Pension Benefit Guaranty Corporation (PBGC)—the government-sponsored program designed to backstop these troubled plans1—is likewise projected to be unable to pay all of the benefits that it guarantees, which are already typically much smaller than the underlying plan benefits. While there are no easy solutions, participants in these plans will not receive the full retirement benefits they expect, nor will they even receive the level of benefits guaranteed by the PBGC, if no action is taken. These benefit reductions could significantly affect the livelihoods of the retirees and their families who expect to rely on this income during their retirement years. This issue brief summarizes how this situation developed and options available to improve the security of existing benefits, and also discusses some ideas about how to maintain and strengthen the multiemployer pension system.

Roughly 1,400 multiemployer plans cover workers employed in an array of unionized industries, including retail, service, construction, manufacturing, mining, transportation, and entertainment. About 100 of these plans have been classified, in accordance with the Multiemployer Pension Reform Act (MPRA), as in “critical and declining” status—meaning they are projected to have insufficient assets in the fund to pay full benefits within the next 20 years. Other plans are already insolvent and receiving financial support from PBGC, and others still are likely to fail beyond the 20-year window for critical and declining plans.

Although the remaining multiemployer plans are currently projected to have sufficient assets to pay all benefits, they are not immune from the long-term risks that have driven other plans toward insolvency. The risks facing these plans are magnified by the fact that many are struggling to maintain their base of contributing employers, and few of them are able to attract new employers.

How the Current Situation Developed
The number of multiemployer pension plans grew dramatically after World War II providing retirement benefits to millions of employees by allowing groups of employers linked by trade, union, or geography to band together to offer collective pension plans that are bargained between labor and management. In theory, and often in practice, the combination of multiple employers participating in these plans provided stability and benefit security, because the decline of one employer would often be offset by the rise of another employer in the plan. Following several decades of success during which nearly all participants received their full benefit amounts from multiemployer plans, recent experience has demonstrated that there are limits to the stability and benefit security inherent in the current system.

Multiemployer Pension Plan Basics
Most multiemployer plans are administered and governed by a board of trustees, with labor and management equally represented. Contributions are collectively bargained, and workers often forgo some direct compensation or accept a reduction to another employee benefit in exchange for contributions to retirement income plans. In turn, employers are obligated to fund the benefits in accordance with their collective bargaining agreements as well as the Employee Retirement Income Security Act (ERISA). The plans must pay PBGC premiums for underlying financial support in the event of a plan failure. Assets are maintained in a qualified trust, and plan trustees retain investment professionals to assist with the management of the investment policy for the funds.

Employers are widely perceived to be solely responsible for risks associated with the delivery of the bargained benefits. However, in reality, the risks have always been shared between the employers and the employees through the bargaining process, as plans have traditionally responded to underfunding through a combination of contribution increases, reductions in future benefit accruals, and larger assessments on employers that withdraw from plans. While plan participants have often sacrificed both wages and future benefit accruals in order to improve plan funding levels, once a benefit was earned it was expected to be paid at retirement. This has generally held true, although there are several examples where PBGC has stepped in to provide financial support to allow troubled plans to pay benefits at the reduced level that the agency guarantees.

In spite of generally meeting the ERISA requirements, serious challenges have been emerging as plans have matured, and these challenges have been exacerbated by the recession of 2007–2009. The guaranteed benefits that PBGC expects to pay participants in troubled plans produce a liability of nearly $60 billion on PBGC’s financial statements.2

While different circumstances apply to each plan, many plans in critical and declining status share several attributes: 
  • Pension assets are invested in diversified portfolios. Plans have invested in diversified portfolios to try to achieve investment returns that can support higher benefit levels and lower contribution requirements than would be possible if the assets earned risk-free rates of return. These investment strategies, however, are not guaranteed, and plans need additional contributions or reduced benefits if the anticipated investment returns are not achieved.
  • Past surpluses led to benefit increases that were not sustainable. Funding pension plans using diversified portfolios will strengthen a plan’s funding status when investment returns are robust. These investment gains may be needed to offset losses when returns are weak. However, following the large asset gains in the late 1990s, many plans became significantly overfunded, and responded by increasing benefit levels or taking contribution holidays. Both the dynamics of the collective bargaining process and regulatory policies that were not conducive to maintaining overfunded plans contributed to this trend. These benefit increases ultimately became unaffordable for many plans when their assets declined dramatically in the subsequent decade.
  • Mature plans have fewer resources to recover from investment losses as the assets grow relative to the contribution base supporting the plan. In young plans, contributions are the primary source of asset growth and investment returns are comparatively small, while the opposite is true in mature plans. As the plan relies more heavily on investment returns, it becomes more difficult to make up for investment losses through additional contributions.
  • Fewer workers are employed in the industries sponsoring multiemployer plans. Some unionized industries have seen significant transformations over time. In some industries the workforce has shifted to more non-union employees as a result of restructurings or regulatory changes, while others have seen declines in the number of employees needed due to global competition, automation, or broad declines in the industry. A decline in the active workforce results in a diminished economic base for collectively bargained employer contributions.
  • Employers have exited multiemployer pension plans, either through bankruptcy or withdrawal, leaving unfunded obligations for the remaining employers in the plans. These obligations, often referred to as “orphan liabilities,” add to the maturity of the plan and subject the remaining employers to additional risks related to the funding of the orphan liabilities.

Typically, a combination of these factors has contributed to a plan that is projected to be unable to pay benefits. The differences in the circumstances between different plans can be important, but this overriding pattern carries important lessons for understanding the options to address the legacy problems and to preventing additional plans from entering critical and declining status.

There are two key challenges facing the multiemployer pension system. The first pertains to securing “legacy” pension benefits—those already earned by participants. The second challenge is to deliver retirement security for employees who continue to work in industries covered by multiemployer pension plans.

Addressing Legacy Pensions
If the accrual of future benefits in these critical and declining plans ceased today, the pension benefits attributable to past service would still present an enormous problem. This legacy problem has an impact on existing and future retirees, as well as the PBGC.

The PBGC Guarantee
Plans eligible for PBGC financial support are subject to the PBGC guaranteed benefit levels, which are generally relatively low (e.g., maximum payout for a full-career participant is approximately $13,000 per year) and are often much lower than the underlying obligation payable from a troubled multiemployer plan. Many participants will experience a significant benefit reduction even if PBGC is fully able to provide the guaranteed benefits, but the PBGC multiemployer insurance program is itself in dire financial condition, and is likely to exhaust its asset reserves in approximately eight years. To deliver the existing PBGC guarantees, it will take some combination of additional revenue or reduced claims from insolvent plans. Some of the measures discussed to improve PBGC finances—all of which would require enabling legislation—are summarized below.
  • Increase PBGC premiums. Premiums have already increased significantly and are scheduled to continue to increase. A potential concern with this approach is that insurance premiums are generally intended to pay for ongoing risks and not past losses. To the extent that higher premiums are viewed as paying for legacy liabilities and not future risks, they may drive healthy employers and employee organizations out of the system, making the increase self-defeating. Higher multiemployer plan premiums represent higher plan expenses, which would adversely impact the funded status of plans over time. While there is concern about any increase in premiums, the current and historical amounts may be perceived as relatively low.3 The multiemployer premium for 2017 is $28 per participant, which represents an average of approximately $0.05 per hour out of employees’ wage packages.4
  • Charge the specific industry. In some industries where restructuring has resulted in a considerable number bankruptcies among employers supporting the pension plan while new employers in the industry did not join the plan, it may be advisable to construct a specific industry tax or premium. This additional charge could be earmarked to pay for the orphaned liabilities left by the bankrupt employers. Another area of potential focus for a targeted charge is on industries for which the claims on the PBGC insurance fund are disproportionately large. However, in both cases, there may not be enough employers remaining in the industry, or the industry may not be healthy enough, to pay the amount needed. Charging the entire industry also means that employers that never participated in the plan would be paying a portion of the liability for the bankrupt employers that did participate in the plan.
  • Charge existing retirees. A modest payment collected from all existing retirees receiving multiemployer plan pensions could generate a significant amount of premium revenue, because the multiemployer system has matured and now has more retired participants than active participants. This premium, however, could face significant opposition, as retirees have never been directly charged for insurance on their pensions before, and many retirees live on fixed incomes with limited options to deal with unexpected reductions to their benefits.
  • Reduce the guarantee. Congress could reduce the guaranteed benefit level to align with the amount of premium resources available. But because the guarantee is already low, the resulting pension payouts may do little to help pensioners achieve financial security. Unlike reducing guarantee amounts for savings deposit insurance, where account holders could shift their assets, multiemployer pension plan participants could not take actions to secure their benefits if the pension insurance limit is reduced.
  • Provide financial support backed by the government. A financial commitment could be made from the general revenues of the federal government, a specific tax, or other taxpayer-supported funding sources. Under this approach, the solution is spread across a broader tax base, involving many taxpayers with little or no direct relationship with the struggling pension plans.
  • Combine PBGC’s multiemployer program with the single-employer program. The single-employer program is currently in a stronger financial position than the multiemployer program. However, this approach would generate potential inequities, as it adds new risks to single-employer plan sponsors and participants. In addition, there are fundamental differences in how the single and multiemployer programs operate, and combining funding could put stress on the single-employer system and further erode support for defined benefit plans.
  • Allow the PBGC to intervene early in troubled multiemployer plans. Under current law, PBGC does not generally provide any financial assistance, or reduce benefits to guaranteed levels, until a multiemployer plan is unable to pay full benefits. The PBGC multiemployer plan program could be aligned with the single-employer program, where PBGC has the authority to intervene long before plans actually fail. By identifying these plans before complete insolvency and reducing benefits to guaranteed levels sooner, PBGC’s limited resources could be conserved. The cost would be that participants in troubled multiemployer plans would experience benefit reductions earlier than occurs under the current multiemployer insurance program.

Other creative ideas may be developed to improve PBGC revenues, but the other potentially powerful approach to preserving PBGC guaranteed benefits is to improve the health of the plans so they don’t have to rely on the PBGC for assistance. So doing would require some combination of additional contributions to the plans, additional investment earnings, or reductions in benefits.

Legacy Liabilities in Critical and Declining Plans
In addition to increasing PBGC revenues, any actions that reduce claims from insolvent plans will help the PBGC to remain solvent, which would allow the PBGC to continue to support the legacy pension plans. Pensions are just one of the economic factors facing many industries. Thus potential actions must be considered in the greater context of a given industry and the possible implications not only to pensions, but to the economic challenges facing the industry.

In late 2014, the Multiemployer Pension Reform Act (MPRA) was passed. This legislation offers troubled plans the ability to reduce benefits to as low as 110 percent of PBGC guarantee levels for current and future retirees, if doing so is projected to achieve long-term solvency of the plan. Of the first 12 plans to apply to the Department of Treasury for benefit suspensions under MPRA, only one has been approved. The others were either denied by Treasury or the sponsors withdrew their applications, presumably expecting denials. The primary reason provided by Treasury for the denied applications was an assessment that the underlying actuarial assumptions were too optimistic. Because optimistic assumptions tend to reduce the level of benefit cuts, the application denials could suggest that plans have not proposed sufficiently large benefit reductions in their MPRA applications.

A successful application under MPRA must minimize the negative impact on pensioners while having at least a 50 percent chance that the plan will be able to deliver on the newly reduced benefit amounts. This balance is difficult to achieve, and even minor changes to assumptions can easily tilt the balance. The Central States Pension Fund was the first and most significant application that was denied. The Central States obligations are currently on a path to the PBGC within the next decade, and are by far the largest risk to PBGC solvency. However, even without Central States liabilities, the PBGC is still projected to become insolvent.

Under MPRA, the sponsor of a troubled plan may also apply to PBGC for financial assistance through a partition. In a partition application, a plan sponsor requests that PBGC immediately assume responsibility for paying the guaranteed benefits for a portion of the participant population. The participants included in the partition must be chosen such that the transfer would sufficiently strengthen the plan so that it is projected to avoid insolvency, but must also represent a savings to the PBGC when compared to allowing the plan to exhaust its assets. Partition is an option when benefit suspensions alone are insufficient to allow a plan to recover and must be requested in conjunction with an application for maximum permissible benefit suspensions. So far, no plan that has applied for a partition under MPRA has been granted one.

Two or more plans may also merge with financial assistance from PBGC. Both facilitated mergers and partitions require that the net result of the transaction is a reduction in PBGC’s deficit and that the assistance does not impair PBGC’s ability to assist other plans. PBGC’s funding distress is a barrier to the use of these tools, as the projected insolvency of the multiemployer insurance program in less than 10 years makes it difficult for partitions or facilitated mergers to generate savings fast enough to satisfy the non-impairment requirement.

None of the options made available by MPRA appear to be sufficient to address the payment of the legacy liabilities. Critical and declining plans are still heading toward insolvency, with participating employers, plan participants, and the PBGC bearing the primary financial risks. In addition, some plans already offer benefit levels at or below PBGC maximums, thus rendering MPRA unhelpful.

New potential solutions are clearly needed. What follows are some general approaches for discussion:
  • Separate the legacy commitments for the “orphan liabilities” relating to bankrupt and withdrawn employers from current employers that may now be shouldering the obligation these employers left in the plan. Segregating these obligations and finding a dedicated funding source helps compartmentalize the solutions to the legacy plan shortfall. There is also an inherent sense of fairness with this approach, and funding could come from a combination of the alternatives identified above for additional PBGC revenue.
  • Provide low-interest loans to troubled plans or employers. This approach borrows money at a low interest rate and invests the proceeds in plan assets. Loans could be made available either directly from the government or from financial institutions. If from a financial institution, the loan would likely require government support to encourage the private institution to offer the loan. These loans would provide a longer timeframe over which employers could spread the cost of funding the liabilities. Additionally, if actual investment returns exceed the borrowing rate, this approach could create the income needed to pay some or all of the promised benefits. However, if actual investment returns do not exceed the borrowing rate, the existing risks to employers, participants, and the PBGC may remain in force. In addition, the plan may be unable to repay the loan, creating a default risk to the provider or guarantor of the loans. The loans could be guaranteed by a governmental agency (e.g., Treasury or PBGC) that currently holds a portion of the risk. Alternatively, participants could absorb part or all of the risk by transforming their fixed monthly benefits to monthly benefits that vary based on actual investment returns. This strategy of converting to variable legacy benefits could also be deployed with many of the other solutions under consideration.
  • Increase minimum funding requirements for legacy obligations. Employers participating in poorly funded plans could be required to significantly accelerate the funding of these commitments. Employers, however, entered into collective bargaining agreements that call for specific contributions to these plans with an understanding of the existing specific limits and exposures. Charging them significantly higher costs for past benefits could cause some employers and covered employee groups to exit plans or push more employers into bankruptcy.
  • Strengthen rules to protect legacy liabilities with respect to withdrawal liability payments and bankruptcy laws. If employers can continue to withdraw and shift liability to the remaining employers, or become bankrupt and escape any withdrawal liability payments, troubled plans—as well as some plans that don’t appear to be troubled—could become even worse. However, giving plans a stronger claim on company assets would mean that other creditors have reduced claims, which could place further stress on the companies that contribute to the plans.
  • Take actions to promote the health of the general economy and, in particular, the affected industries. Economic growth in the industries that sponsor multiemployer plans could facilitate the funding of pension plans. A strong market could increase investment returns on pension assets.

All the revenue sources from the various stakeholders mentioned above in addressing PBGC’s shortfall could also play a role in strengthening individual failing plans. One question that needs to be answered is whether to focus efforts on the long-term sustainability of PBGC’s multiemployer program, or to focus on rehabilitating individual failing plans. Addressing only PBGC solvency would result in significant benefit reductions for many participants due to the low level of the PBGC guaranteed benefits. On the other hand, addressing individual failing plans could be more costly than assisting the PBGC, and could also raise equity concerns with respect to plans that have either already begun receiving PBGC financial assistance or have implemented benefit cuts under MPRA.

Assuring a Robust and Secure Retirement System in the Future
Creating a robust and secure retirement system for employees who participate in multiemployer pension plans is the second key objective. Multiemployer pension plans conveniently allow employees to accumulate and vest retirement benefits in one place, even if they work for a number of different employers across an industry over the course of their careers. However, to be sustainable, the risks for both employers and participants must be managed well. Employers must be assured that their financial obligations are relatively fixed and known. Participants need to be confident the benefits will be paid when due. Unless the multiemployer system finds better ways to manage risks, employers will continue to be reluctant to contribute to the plans, and employees will continue to be exposed to the possibility of significant unanticipated benefit reductions.

Defined contributions (DC) plans have historically played a largely supplemental role in the multiemployer system. The role of DC plans could be another consideration in the search for ways to deliver reliable and affordable retirement security to employees in multiemployer programs in the future. DC plans eliminate employer concerns about withdrawal liability and are highly portable, but they also place more responsibility and risk on plan participants for their own retirement security.

Approaches that have the potential to keep plans healthy and secure are discussed below:
  • Support risk-sharing benefit designs. Hybrid retirement designs that attempt to capture the best features of defined benefit and defined contribution plans have existed for many years, and new designs are emerging today. These designs, whether they are called variable, adjustable, or composite, have explicit mechanisms to adjust benefits (either automatically or based on trustee decisions) to maintain the funded status of the plan. They have the potential to improve the sustainability of the plans and reduce or eliminate withdrawal liability and the need for PBGC guarantees. Very few of these designs currently exist among multiemployer plans, in part because they don’t fit neatly into the current legal and regulatory structures for defined benefit or defined contribution plans. Some new structures or clarification of how to apply the current structures may be needed in order for these types of designs to be adopted more widely. In addition, more analysis would be needed to understand the potential impact on the long-term solvency of PBGC and the associated legacy plans if these designs become more available and popular.
  • Restructure withdrawal liabilities and bankruptcy laws to make it more attractive for new employers to join the system. The current approach of passing unfunded obligations from employers that have left the system onto employers that remain in the system is a significant barrier to attracting new employers into plans. Another concern is the 20-year cap on the annual withdrawal liability payments, which can lead to a significant gap between the actual amount an employer pays and the value of the underlying unfunded benefits. At the time an employer declares bankruptcy and withdraws from the plan, or upon a mass withdrawal of employers from the plan, there could be a combination of additional funding, de-risking specified benefits, prioritization in bankruptcy proceedings, and reducing benefits to keep the plan whole. Note that if pension obligations receive higher priority in bankruptcy proceedings, other creditors take on more risk. Protecting companies from being forced to fund the benefits earned by employees of other companies could address their concerns about participating in the multiemployer pension system.
  • Review funding and investment requirements. Funding and investment decisions could take into account plan maturity measures and the plan sponsor’s (or industry’s) ability to take risk. The current funding rules may not adequately take these factors into account. Funding measures that reflect the capacity of a plan to respond to adverse experience may lead to improved benefit security for participants, while stochastic projections that assign probabilities to a range of outcomes can help further quantify risk levels and drive decisions. Shortening the amortization periods used to determine the minimum required contributions could help keep benefits secure, but would also introduce more volatility. Enhancing the minimum funding requirements in these areas could result in improved retirement security. However, any changes to funding requirements need to take into consideration the collective bargaining nature of multiemployer plans and how employers and employees would respond.
  • Improve transparency with respect to benefit security risks. In any financial system, there is a relationship between the size of the assets backing a future payment and the level of risk associated with that payment. The benefits earned by most multiemployer plan participants are larger than what plans would have been able to pay if the plan assets had been invested in risk-free investments. Participants in these plans have been compensated for the investment risks that plans have taken (in the form of higher benefit levels and reduced contributions), but they have not been adequately informed about these risks. If participants understood in advance that the benefit levels they expect to receive are only possible due to investments in volatile asset classes, and that this means there is a possibility that benefits might need to be reduced in the future, the prospect of benefit reductions might not be as significant of a problem as it is today. 

Without any changes to the multiemployer system, over a million participants are at risk of not receiving the benefits they were promised by their plans or even the guaranteed benefit amount under the PBGC promise. There are no quick fixes that can eliminate this risk without some stakeholder carrying a burden, but the sooner there is action, the less painful the corrective measures will be.

1.  The PBGC sponsors two insurance programs, one for single-employer pension plans and one for multiemployer pension plans. There are material differences between the two programs. This issue brief addresses only the multiemployer program.
2. Based on the 2016 PBGC annual report: Annual Report 2016: Keeping Our Commitment to America’s Workers.
3. Because multiemployer plans cover workers in industries with a wide range of compensation levels, the impact of premiums on participants and employers varies considerably from plan to plan.
4. Based on a $28-per-head premium rate, and the assumption that 37 percent of multiemployer plan participants are active and that they will work an average of 1,600 hours per year. Note that the $0.05 figure is an average across the multiemployer system that will vary from plan to plan.