Commentary

How Bad Investment Rate Projections Cause Pension Underfunding

Municipal discount rate assumptions prove to be costly

Michigan’s 100 largest cities and townships have over $5.4 billion in unfunded pension liabilities. The liabilities are the difference between what municipalities have promised to public workers as retiree benefits and what they have put aside to pay for them.

A main driver behind this future debt seems to be the “discount rate” – the investment rate cities assume when projecting costs far off in the future. Some cities assume extremely generous market returns, which mean they pay less in the short term but much more later when their assumptions turn out to be too high. But cities that use lower investment return assumptions tend to do a better job setting aside the money need to pay for pensions.

In Flint, the city’s defined benefit pension system is only 47 percent funded and has $285 million in unfunded liabilities, while the Grand Rapids defined benefit pension system is 97 percent funded and has only $29 million in unfunded liabilities (despite the city being nearly twice the population).

One difference between the cities is the assumed investment rate. Since at least 2006, the Grand Rapids government has estimated a discount rate between 7 and 7.5 percent. On the other hand, Flint’s government has estimated a discount rate of at least 8 percent since 2007.

At first glance, this appears to be a small one percentage point difference. But it is actually nearly 15 percent – which can add up to large differences in liabilities. For example, in the 2015 audit report for Grand Rapids, the city estimated a 7.5 percent discount rate in its general retirement system. If the 7.5 percent discount rate is achieved, then city will have $31 million in unfunded liabilities next year. However, if a discount rate of only 6.5 percent is achieved, then the city will face $84 million in unfunded liabilities.

By estimating a lower discount rate, Grand Rapids prepared for the uncertainty of investments by requiring more contributions from employers and employees, while Flint assumed optimal investment conditions for a decade straight. And they didn’t hit it. Flint’s five year return on investment amounted to 5.6 percent compared to Grand Rapids 7.2 percent.

Even though an 8 percent return on investment is very ambitious, policymakers still have an incentive to estimate high discount rates because it makes unfunded liabilities appear smaller in the short run. Over time, however, the unfunded liabilities can grow exponentially if investment returns are below expectations, and employers and employees are not paying enough to make up for the lower investment yields.

Michigan Capitol Confidential looked at the audits of all of Michigan’s largest cities and found a direct correlation between a city’s funded ratio and its discount rate. A funded ratio is a city’s assets divided by its liabilities. Municipalities that were at least 80 percent funded were assuming an average investment rate of just over 7.1 percent, while cities funded less than 80 percent were assuming an average discount rate of nearly 8 percent.

In 14 cities, taxpayers were on the hook for no pension debt because the employees instead participated in a defined contribution retirement system, which includes IRA’s and 401(k)’s. In the defined contribution system, retirees only get what they and their employers pay into it during their tenure, so there is no possibility for underfunding.

Defined benefit pension systems, on the other hand, have the possibility of unfunded liabilities because benefits are calculated based on years of service and salary at the time of retirement. Pension systems rely on politicians to accurately predict future returns and fully fund the system – which is almost never done.

Michigan Capitol Confidential is the news source produced by the Mackinac Center for Public Policy. Michigan Capitol Confidential reports with a free-market news perspective.