Final Thoughts

Designing employee pensions involves more than a traditional debate between defined-benefit and defined-contribution plans. Both types of plans have inherent advantages and disadvantages. For the record, defined-contribution plans have suffered asset downturns over the period studied as well. Any such losses are the responsibility of the individual participant, however, rather than current and future taxpayers as a group.

A complete analysis of the advantages and disadvantages of defined-benefit and defined-contribution plans in the public sector is beyond the scope of this brief. Nevertheless, it is reasonably certain that the MSERS defined-contribution plan has cost taxpayers less over the period studied than retaining this same group in the MSERS defined-benefit plan. The Legislature failed to make the annual required contributions to the defined-benefit plan even after the plan was closed, so it seems unlikely the Legislature would have made the larger annual required contributions necessary if the plan had continued to receive new entrants. Thus, continuing only with the defined-benefit plan would have likely placed that plan in worse financial condition than it exists in today; the truly debatable question is the magnitude of the additional unfunded liability.[*]

The calculations in this Policy Brief suggest that since the advent of the MSERS defined-contribution plan in 1997, Michigan taxpayers have saved approximately $167 million in lower pension normal costs and between $2.3 billion and $4.3 billion in lower unfunded liabilities. An additional and important advantage, though difficult to quantify, is the reduced political temptation to provide benefits whose costs are largely deferred to future generations. In other words, a defined-contribution plan is less prone to potentially harmful political interventions.

Of significant note, MSERS’ current and projected defined-benefit pension liabilities and related employer contributions are predicated on achieving an assumed 8 percent annual asset return over the long-term. The reasonableness of such an assumption could easily be debated and could well be the subject of a separate report. In fact, such an assumption was recently studied by Wilshire Associates, an independent international investment and consulting firm. The report, which studied 126 U.S. state pension plans (including MSERS, MPSERS and two other major Michigan government pension plans), concludes:

Using [our] return forecasts, none of the 126 state retirement systems are expected to earn long-term asset returns that equal or exceed their actuarial interest rate assumption.[14]

Wilshire further concludes that the median long-term asset return for the 126 state pension plans would be approximately 6.5 percent — 1.5 percentage points less than Michigan’s 8 percent return assumption.

The key point is this: If MSERS’ current actuarial valuations were to be recalculated using lower investment return assumptions, then the unfunded liability and annual required contributions for the MSERS defined-benefit plan would be higher. Thus, the cost savings calculated in this Policy Brief for switching new employees to the MSERS defined-contribution plan could be materially higher. The magnitude of the increase, of course, would depend on the precise return assumption used.

The nature and amounts of any future savings will depend on actual investment experience and other factors, including funding policies. Regardless, common sense and the calculations in this Policy Brief suggest that Michigan government should follow the demonstrated best practices of the private sector with regard to employee pensions. In the private sector, pension costs are now designed to be current, with no unfunded liability; predictable, with easily computed expenditures for coming years; and affordable, with annual costs between 5 percent and 7 percent of payroll.[†] The MSERS defined-contribution plan achieves these objectives and can thus serve as a model for reforming other government pension systems.

[*]  An analysis by the Michigan Senate Fiscal Agency, while recognizing the potential long-term financial benefit of transferring new public school employees to a defined-contribution plan, has nevertheless underestimated the total financial benefit that might be realized. Kathryn Summers-Coty, “Examining a Change from Defined Benefit to Defined Contribution for the Michigan Public School Employees’ Retirement System,”(Michigan Senate Fiscal Agency, 2009), Publications/Notes/2009Notes/NotesMarApr09ks2.pdf (accessed Aug. 24, 2010). The study focuses exclusively on comparing the employer contribution for a hypothetical MPSERS defined-contribution plan to the normal cost of the existing MPSERS defined-benefit plan, noting that the difference between the two is not particularly large. But as the author noted in an earlier Mackinac Center Policy Brief:

In fact, the normal cost of the program is only part of the annual cost; another portion is the annual payment on the unfunded liability. Hence, the normal cost does not represent the full cost of the plan. Indeed, if the normal cost were considered an absolute measure of the true cost of the MPSERS defined-benefit pension plan, the plan would not have an accrued unfunded liability of nearly $12 billion [now $17.6 billion].

See Richard C. Dreyfuss, “Michigan’s Public-Employee Retirement Benefits: Benchmarking and Managing Benefits and Costs,”(Mackinac Center for Public Policy, Oct. 25, 2010), 13-14, (accessed April 16, 2011).

[†] See Dreyfuss, “Michigan’s Public-Employee Retirement Benefits: Benchmarking and Managing Benefits and Costs,”(Mackinac Center for Public Policy, Oct. 25, 2010), 8-11, (accessed March 28, 2011).