There is a popular aphorism that “when you find yourself in a hole, the first step is to stop digging.” With respect to public employee pensions, a growing number of policymakers are contemplating following that advice.
Across the country, defined benefit plans have unfunded liabilities estimated at $1 trillion to $4 trillion and annual servicing costs have more than doubled over the past decade. The recent bankruptcy filing in Detroit cited pension funding pressures as central to the city's fiscal woes, and public employee benefits played a role in the bankruptcy filings of Stockton, San Bernardino and Vallejo, all in California, and Central Falls, R.I. This has led governments to consider shifting new employees into more sustainable defined contribution, cash balance or hybrid plans. Logic and simple economics suggest that capping the source of unfunded pension liabilities would reduce risk and cost going forward.
But that's not what you hear from public defined benefit plans and the actuaries who work for them. They claim that shifting new hires to alternate plans actually increases costs; indeed, the more underfunded the current defined benefit plan is, the higher the cost of shifting to more manageable alternatives. This claim is nonsense, but it is nonetheless slowing reforms in jurisdictions in states such as Florida and Pennsylvania and cities such as Tucson and Los Angeles.
What are these “transition costs” that stand in the way of public pension reforms? One argument originates from now-defunct accounting standards issued by the Governmental Accounting Standards Board. These standards, it is claimed, obliged the sponsor of a closed defined benefit pension plan to more quickly repay, or amortize, its unfunded liabilities. Even if true, accelerating repayment would increase payments in the short term but produce even larger savings in the long term. Portraying this as a prohibitive cost is seriously misleading.
Moreover, these GASB standards never dictated policy, but were for disclosure only. States and localities can — and routinely have — set their own schedules for paying down pension debt. And since the amount of benefits owed, the timing of those benefit payments and the payer of the obligations is unchanged, there is no economic or policy rationale for altering amortization schedules simply because new employees earn benefits under a different system. In any case, GASB recently updated its guidelines, removing the wording that purportedly supported this claim and making the issue moot.
As the accounting argument for transition costs has been debunked, a seemingly more plausible investment-based claim has taken its place. Transition cost proponents argue that because a closed defined benefit plan has a finite life, the plan must hold progressively less risky and more liquid assets to ensure it has money on hand to make benefit payments as the fund winds down. And because less risky, more liquid investments have lower expected returns, higher contributions would be needed upfront. Thus, the claim goes, if you stop digging, the hole actually gets deeper.
A closed plan would need to hold more liquid investments at the very end of its existence, but this shift need not happen until the remaining asset base is very small. And since only a small portion of pension assets are truly illiquid to begin with, shifting to more liquid investments would have a trivial effect on overall plan cost.
The argument based on lower-risk investments has two flaws, both deriving from the dubious belief that defined benefit pension plans' long-term equity investment carries essentially no risk. If one accepts this belief, then there is no need for a closed plan to shift to low-risk investments. It can continue its current allocation and call on the general fund to make benefit payments in the short term. If long-term investment is truly riskless, any general fund transfers could eventually be returned in full.
A more rational view accepts that long holding periods — so-called time diversification — do not erase investment risk. Indeed, as Nobel laureate Paul Samuelson pointed out in the 1960s, the dispersion of possible portfolio values only grows larger over the long term. A pension plan holding high-risk investments thus imposes a contingent liability on future taxpayers to make good on guaranteed benefit payments should the assumed investment returns fail to materialize. Prices paid in private markets for “put options” protecting against market downturns show that the value of these contingent liabilities offsets the higher expected returns from risky investments. A closed pension plan that takes less investment risk imposes smaller contingent liabilities on future generations, and so the total cost of the plan remains unchanged.
Advocates of traditional defined benefit pension plans view these plans as virtual money machines that can generate high investment returns by holding stocks — and, increasingly, private equity, hedge funds and other alternative investments — while ignoring market risk. Not surprisingly, any shift away from this seemingly magical model is portrayed as imposing significant costs. Yet once we account for the full economic costs of defined benefit plans, in particular the large contingent liabilities that risky pension investments impose on future taxpayers and public employees, the supposed costs of shifting to more transparent and financially sustainable models are virtually eliminated.
Sometimes common sense is right: When you are in a hole, it really does make sense to stop digging.
Andrew G. Biggs is a resident scholar at the American Enterprise Institute, Washington. Josh McGee is vice president of public accountability at the Laura and John Arnold Foundation, Houston. Michael Podgursky is a professor of economics at the University of Missouri, Columbia.
This article originally appeared in the January 6, 2014 print issue as, "Transition cost not a bar to pension reform".