Michigan’s school pension system has already imposed $22.4 billion in unfunded liabilities on taxpayers, an amount that may be greatly understated given pension fund investment growth assumptions widely regarded as overly optimistic. 

A Senate-passed reform contains the underfunding problem and eventually eliminates it by closing the system to new employees, who would instead get 401(k) contributions. This is transformational, because it's the largest component needed to put the state on a glide path to eventually eliminating most government employee legacy costs  — a policy initiated in a major way by the 1996 Legislature, under the engaged leadership of Gov. John Engler.

In contrast, a House-passed alternative continues to create more long-term taxpayer liabilities every time a new school employee is hired.

In not embracing the opportunity represented by the Senate vote, it appears that key House members have bought into flawed analyses from state pension bureaucrats, whose position is also shared by the MEA teachers union.

Specifically, they claim converting to a defined-contribution retirement system will force the state to pay more than a billion dollars over the next few years to "catch up" on all those unfunded liabilities with a more front-loaded payment schedule.

This claim is not correct. Front-loaded amortization "transition cost" payments are not required under Government Accounting Standards Board rules. In June the standards board itself  published a statement that affirms this.

Legislators rightly care about those accounting rules because if the pension bureaucrats' interpretations were correct, the bond market might inflict higher lending costs on a state that doesn't frontload the amortization of the unfunded liabilities.

To address this concern, recent Mackinac Center articles have cited examples of states that closed their defined-benefit pension systems to new members without paying "transition costs," and if anything, were rewarded with improved credit outlooks due to the magnitude of that reform.

• Alaska started converting to a defined-contribution retirement system for government employees in 2005. Policymakers initially chose to pay those transition costs, but after one year they stopped, and reverted to a non-front loaded amortization schedule for paying down unfunded pension liabilities. The state's credit outlook did not suffer.
• Utah closed its open-ended defined-benefit pension system to new government employees in 2010, offering them instead either a defined-contribution plan or participation in a pension plan that caps employer costs. They did not pay any “transition costs.” The state assessed the unfunded liability amortization costs against total payroll, not just that of workers covered by the closed system, and as its own financial report stated: “Therefore, this law had no effect on this actuarial valuation.” Utah’s positive credit outlook was not changed by this action.
• Michigan closed the school pension system in 1995 to new employees at seven state universities which had been participants. Policymakers chose to amortize the unfunded liability over 40 years with no front-loaded “transition costs,” and assess each university its share on the basis of total payroll, not just the “covered employees” portion. The state did not suffer any bond market consequences.
• Michigan also closed its defined-benefit pension to all new state employees hired since 1997, saving taxpayers between $2.3 billion to $4.3 billion in unfunded liabilities since then. At the time, however, there were no unfunded liabilities, and so there were no “transition costs” from a front-loaded payment schedule.

To recap: House Republicans are basing their refusal to enact a transformational reform on an accounting rule interpretation that the makers of those rules themselves say is wrong, and on fears of credit rating consequences that specific real world examples show are misplaced.

It should be noted that government pension managers and school employee unions who support this claim have strong incentives for preferring to keep a defined-benefit system open and growing rather than closed and shrinking.

Nevertheless, some Republican House members continue to actively promote the claim. They defend their plan to perpetuate defined-benefit pensions by mischievously comparing the cost of unnecessarily high 401(k) contributions in the Senate-passed bill to the demonstrably inadequate pension fund contributions their own proposal relies upon.

This flawed apples-to-lemons comparison ignores a history of chronic underfunding to make the House plan appear “less costly.” It also ignores the fact that a pure defined-contribution system for new employees creates no new layers of long-term taxpayer liabilities to potentially underfund.

In addition, key House Republicans claim their plan is superior because it proposes prefunding health care benefits the Legislature has chosen to give school retirees (who are also eligible for federal Medicare at age 65). However, unlike monthly pension checks, the state has no obligation to keep providing a retirement health benefit that few outside of government receive.

Yet while prefunding is problematic for various reasons, it’s not a big deal compared to blowing the opportunity to close the defined-benefit pension system. If legislators feel money can be spared to prefund optional future health benefits, then doing so is tolerable as long as the pension system is closed.

Finally, policymakers are also concerned about retirement system assessments that are breaking the backs of local school district budgets. The House itself has offered one method to provide relief: Pay some of the costs directly out of the School Aid Fund, without first cycling the money through school districts.

This approach raises some questions, but isn't a “deal-breaker” if it helps realize the overarching goal in all this: Protecting taxpayers from unfunded liabilities by gradually getting Michigan schools out of the defined-benefit pension business.