Absent deep benefit cuts, many union-sponsored multiemployer pension plans are likely to become insolvent even if they have access to subsidized loans, according to a new white paper published by the Pension Analytics Group.
The Pension Analytics Group bills itself as “a group of actuaries and economists who are concerned that the clock might run out before a viable solution for the multiemployer system’s funding problems can be designed and implemented.” They have published a series of white papers outlining the severe funding challenges faced by union pension plans in the U.S.
The group conducted its latest analysis of government-backed bailout loans in response to recent proposals from lawmakers and retirement industry professionals suggesting a long-term, low-interest-rate loan program could save the most troubled multiemployer pension plans without imposing undue hardship on participants, contributing employers, the Pension Benefit Guaranty Corporation (PBGC), the federal government, taxpayers or healthy plans.
Using its Multiemployer Pension Simulation Model, the Pension Analytics Group projects that about 200 multiemployer pension plans covering over 3 million participants will become insolvent over the next 30 years, and that the PBGC’s multiemployer guarantee fund will itself be exhausted by 2027.
“A variety of options are available to prevent the guarantee fund’s exhaustion, such as empowering plans to take stronger actions to avoid insolvency, reducing the level of the PBGC’s benefit guarantee, and increasing the revenue flowing into the guarantee fund through premium increases or by securing additional revenue sources,” the white paper says. “To this list, we can add a concept that has been circulating recently on Capitol Hill—subsidized loans provided to troubled pension plans by the federal government.”
As the paper explains, supporters of this approach contend that it buys time for struggling pension funds to get back on their feet. Eventually, it is hoped, a plan receiving a loan will regain its strength, pay off its loan and avoid insolvency.
For its analysis of whether this approach is likely to be effective, the Pension Analytics Group considers a relatively straightforward theoretical loan program, under which each troubled pension plan will be eligible to receive a one-time lump-sum loan equal to the plan’s funding deficit. To determine the deficit, plan liabilities will be measured at a 7% discount rate, and the loan interest rate will be fixed at 2%, which is below current Treasury yields. Among some other stipulations, the analysis assumes the term of each loan will be 20 years, at which time the entire amount of the loan must be repaid with interest.
“After simulating this policy option, we then compared it [with] a baseline scenario in which a loan program is not offered,” the white paper says. “To capture the broad range of possible outcomes for each plan, we used 500 trials in which asset returns were varied stochastically, and we restricted our analysis to those plans that we project will become insolvent within 30 years in the absence of a loan program.”
According to the analysis, across some 500 stochastic trials, the average total number of participants in plans projected to become insolvent is 3.1 million in the baseline scenario, and 2 million if the loan program is implemented.
“Thus, on average, the loan program prevents plans covering over 1 million participants from becoming insolvent,” the paper says.
Important to note, the results vary widely across stochastic trials, and in nearly a third of the trials (30%), the loan program has little or no impact on the number of plans projected to become insolvent.
“Considered in isolation, the loan program is expensive,” the paper says. “The net cost of the program—which we define as the present value of lending to plans minus the plans’ repayments, computed at Treasury discount rates—is an average of $56 billion across the 500 trials.”
One consistent positive outcome found across many scenarios considered is that lending programs would significantly reduce pressure on the PBGC.
“In some simulated cases, a plan pays back its loan and remains solvent in the long term, thus freeing the PBGC from the obligation to make assistance payments,” the paper says. “In other cases, a loan leads to a delay and/or a reduction in assistance payments. In 55% of the stochastic trials, the projected reduction in the present value of PBGC assistance payments exceeds the net present value cost of the loan program. While loan defaults occur, the cost of these defaults is offset by a reduction of PBGC assistance payments. Consequently, for these trials, the loan program leads to a reduction in the total cost of insolvencies—where we define ‘total insolvency cost’ as the sum of the present value of PBGC assistance and loan cash flows.”
According to the analysis, for the remaining 45% of the trials, the projected reduction in the present value of PBGC assistance payments is less than the present value cost of the loan program. For these trials, the loan program leads to an increase in the total insolvency cost to taxpayers.
“On average, across 500 trials, the loan program reduces the cost of insolvencies by about $3 billion; results, however, vary widely by trial,” the paper concludes.