“Retirement system too far in debt for reform”
This jarring headline, reporting the conclusions of a legislative study group in Virginia last year, crystallized the prime objection to pension reform nationwide. It is the pension world's version of Catch-22: even if the traditional defined benefit system is broken at its core, it's too costly to scrap; the best you can do is patch it up and carry on.
It might seem improbable that huge unfunded liabilities would be proffered as the reason not to adopt alternatives designed to prevent new unfunded liabilities. And yet, this claim has effectively stymied reform in state after state.
The public pension fund industry and its allies (public employee unions, money managers and some consulting actuaries) claim that closing an old DB plan and enrolling new employees in alternative plans requires an acceleration of payments to amortize the old plan's unfunded liability. This creates unpalatable transition costs, raising employer contributions in the short run.
Most commonly, the argument rests on an accounting rule of the Governmental Accounting Standards Board. GASB currently defines an annual required contribution, specifies its calculation and requires pension funds to report it. The ARC includes amortization. For ongoing plans, GASB allows funds to back-load the amortization, letting it rise with employee wages as a “level percent of payroll.” But if the old DB plan is closed to new entrants, the payroll of plan members shrinks and GASB requires a shift to “level dollar” amortization — no longer back-loaded.
The costs of accelerated amortization are readily quantified by pension fund actuaries, and the dollar amounts can reach billions for plans with large unfunded liabilities. As Keith Brainard, research director of the National Association of State Retirement Administrators recently put it, “Generally speaking, the more underfunded a plan is, the more expensive it is to try to switch.” Invariably, this deters lawmakers from fundamental reform.
Most actuaries know this argument is false, but the facts are not widely trumpeted. The concepts are arcane, so lawmakers often defer to the assumed expertise of the pension fund industry. Two facts are often glossed over, sowing great confusion.
The first point is simple: GASB does not determine funding policy and does not claim to. GASB sets standards for financial reporting.
Pension plans are required to report the ARC for comparison with actual contributions, but the contributions are set by each state's statutory authority — the legislature or the pension board. These authorities are not bound by GASB accounting rules in setting funding policy, and contributions often differ from the ARC. To be sure, large shortfalls below often are a sign of fiscal irresponsibility. But the specific GASB rule for amortization on closed DB plans is not compelling on policy grounds.
The second point: GASB's amortization rule assumes employer payments are based on the payroll of DB plan members alone. However, states can and do levy amortization charges on total payroll — old and new plans alike — and with sound justification. Employers are responsible for the unfunded liabilities of old members, regardless of what plan new members enter. The actuarial firm Cavanaugh Macdonald Consulting LLC has stated the case well:
“If the (amortization) payment is calculated using the total payroll of members in both the DB and defined contribution plans, the dollar amount of the payroll is the same as if the DB plan were still open. As a result, the (unfunded liability) is amortized at approximately the same rate of pay as would occur if the DB plan had not been closed to new hires.”
In short, the key to successful reform is to base amortization on total payroll, rather than payroll of the closed plan alone. There is no policy reason to change amortization, if the previous pace was considered sound. That is what reform-oriented states such as Utah, Rhode Island and Alaska have done, as they switched to DC, cash balance and hybrid plans. Other states considering reform might follow their lead, undaunted by the false claim that GASB rules require them to accelerate amortization payments.
In fact, GASB intends to drop the entire ARC construct, so the ARC amortization rule will likely disappear. Regardless, neither the rule itself, nor, more importantly, sound policy, requires accelerated payments.
The underlying problem with traditional DB plans is that benefits are not tied to contributions. Idiosyncratic formulas (“rule of 80,” “25 and out”) lead to chronic funding problems, distorted retirement decisions and gross inequities between mobile and career employees. Reforms that tie benefits to contributions address these problems. The obstacle is the purported link between amortization and future benefits.
How to structure amortization payments on the unfunded liability is a financial policy decision, not much different from the decision on scheduling ordinary debt service payments. How to structure future benefits is a separate policy decision.
Pension reform proposals should be evaluated on their own merits and not conflated with amortization schedules. Amortization pays for past debts; pension reform lays a path toward a responsible future.
Robert Costrell is professor of education reform and economics at the University of Arkansas, and fellow at the George W. Bush Institute. This article is taken from a longer paper that is available from the Laura and John Arnold Foundation