The stock market is high, unemployment is low, but many multi-employer, union-sponsored pension plans are hopelessly insolvent and facing their own financial crisis. So is the government’s program that guarantees those pensions through the Pension Benefit Guaranty Corporation (PBGC). “Insolvent” means that while they have not yet spent their last nickel of cash (although that day is coming), their liabilities are vastly greater than their assets, and all the liabilities simply cannot be paid. In short, many multi-employer pension plans are broke and so is their government-sponsored guarantor. Unsurprisingly, the idea of a taxpayer bailout arises, although its proponents do not wish to call it a bailout.

The PBGC’s multi-employer program has a net worth of a negative $54 billion, according to its September 30, 2018 annual report. It has assets of only $2.3 billion, and liabilities of $56 billion—it has $24 in liabilities for each $1 of assets. And this striking deficit only counts the probable losses for the next ten years, not the unavoidable further losses after that. PBGC estimates the total unfunded pension liabilities of the multi-employer plans at $638 billion. Making financial promises is so much more enjoyable than keeping them.

One of the causes of these deficits is the government guarantee itself, which can induce these pension plans to make bigger pension commitments than they funded or can fund, reflecting the expectation of a taxpayer bailout. This displays the moral hazard of getting the government to guarantee pensions, an unintended but natural risk of creating the PBGC in the first place.

The deficits in the insolvent pension plans and in the PBGC are facts. We know for certain that losses which already exist will necessarily fall on somebody. On whom? That is the question. From where we are now, there is no possible outcome in which nobody loses.

The Employee Retirement Income Security Act of 1974 (ERISA), in establishing the PBGC, specified that it would never take any money from the Treasury. As the PBGC annual report explains, “ERISA requires that PBGC programs be self-financing.” Whoops. Furthermore, “ERISA provides that the U.S. Government is not liable for any obligation or liability incurred by the PBGC.” Should we ever believe such protestations? The same provision was made for the debt of Fannie Mae and Freddie Mac, but they got bailed out anyway.

Last year, Congress set up the Joint Select Committee on Solvency of Multiemployer Pension Plans to figure out what to do. The name was nicely diplomatic, since the core issue was rather the “Insolvency” of these plans. The special committee held hearings and did its best, but disbanded without issuing its required report.

Now it inevitably occurs to many politicians that there should be a bailout to benefit the pensioners, unions, employers, and the PBGC, while moving losses to the taxpayers. A bill to this effect, the “Rehabilitation for Multiemployer Pensions Act,” was introduced this year and is headed to a mark-up in the Ways and Means Committee of the House.

Suppose you have decided that a taxpayer bailout is less bad than having pensions cut, unions embarrassed, employers faced with unaffordable pension contributions, and watching the PBGC’s multi-employer program head to default. How would a bailout best be structured? I suggest the following essential points:

Congress should be honest about what it is doing. You can’t think clearly about the principles and effectiveness of bailouts if you don’t face up to the fact that you are designing a bailout.

Congress should adapt for use in this case a globally tried and true method for dealing with hopelessly insolvent financial entities: the Good Bank/Bad Bank structure. This structure should be required for any pension plan receiving appropriated taxpayer funds in any form.

A fundamental principle is reform of the governance of bailed out entities. Those who ran the ship on the rocks should not be left in command. They should not be in charge of spending the money taken from other people to make up their deficits.

The Good Bank should contain what has a high probability of being a successful, self-sustaining entity going forward.

The Bad Bank should contain the deficits and unfunded obligations from past unsuccessful operations, plus the bailout funding. It should be run as a long-term liquidation. It will make clear the real cost of the bailout and dispense with the need for further bailouts in the future.

The Good Bank should begin and continue on a fully funded basis. The required contributions of the employers should be determined as a mathematical result of the committed pensions, not be a result of bargaining subject to the moral hazard of the government guarantee. This calculation should use the discount rates required for single-employer plans. Employers should have to book as their own liabilities their pro rata share of any underfunding which might occur. Finally, data and reporting should be revised to be made timely and more transparent.

The Bad Bank should have whatever assets, if any, are left over after forming the Good Bank, all the plan’s pension commitments already made but not funded, the obligations of employers for contributions to those commitments, and the bailout funding. It should purchase high quality annuities to meet its pension obligations, not try to run risky asset portfolios. In time, it would disappear, with remaining funds, if any, returned to the Treasury.

The Good Bank should be governed by a board of independent directors with fiduciary responsibility for the good management of the plan.

The Bad Bank should be run by a government-appointed conservator.

If you are going to have a bailout of the insolvent multi-employer pension plans, a Good Bank/Bad Bank structure along these lines would be highly advisable.

Image credit: szefei