While public pensions and other post-employment benefits (OPEB) continue to garner attention in policy circles, little research has been done to quantify the cost impact these benefits have on multiple levels of government. Most studies focus on public pensions in states, ignoring OPEBs and municipalities. A recent report by Alicia Munnell and Jean-Pierre Aubry at the Center for Retirement Research at Boston College (CRR) provides a more comprehensive picture of those cost burdens not only on states but also on major counties and cities in the US.
Estimating the costs of those benefits, however, is not straightforward. Large portions of state pension liabilities are actually under the fiscal responsibility of lower-level counties and cities. Pension and OPEB costs are also sensitive to discount rates and amortization methods. To deal with these problems, the authors first reallocate the pension liabilities and OPEBs according to government responsibility instead of government administration, using the Annual Required Contribution data. They then estimate the pension and OPEB costs by applying a “conservative” 6% discount rate and a closed 30-year level dollar amortization period. Finally, they add debt service costs to pension and OPEB costs to arrive at the “total cost burden” for each state, and divide it by the respective government’s own-source revenue to obtain a relative cost measure. The report covers 50 states, 178 counties, and 173 cities.
The final results show that 9 states have pension costs exceeding 10% of own-source revenue and that number rises to 17 if OPEBs are included. The top five states with the highest pension and OPEB costs as a percentage of own-source revenue are Illinois, New Jersey, Connecticut, Hawaii, and Kentucky. Not surprisingly, those states also have the highest relative total cost burdens.
The picture is not pretty for counties and cities. Of the 50 largest counties in the report’s sample, 17 have combined pension and OPEB costs exceeding 20% of own-source revenue, and 25 have the combined costs exceeding 15%. Of the 50 largest cities in the sample, 19 have the combined pension and OPEB costs higher than 20% and a whopping 31 cities have the combined costs higher than 15%. Notably, six counties in California (Fresno, Sacramento, Kern, San Diego, Los Angeles, and Orange) have combined pension and OPEB costs greater than 35%. Chicago, Baltimore, Miami, Detroit, and Boston are the cities that have the highest pension and OPEB costs, at more than 30% of own-source revenue).
Those estimates may not fully reflect the true magnitude of those costs. Though the 6% discount rate used in the report is deemed “conservative” based on recent projections by investment experts about future market performance, that rate still derives from expected returns on risky assets in lieu of the risk of pension liabilities. This is at odds with financial economics that recommends using a bond-based discount rate for reporting purposes and a risk-free discount rate for funding purposes. The current yield on high-grade 20-year municipal bonds is about 3.5%, and the notional yield on 20-year Treasury bonds is currently 2.8%. Either of these rates is much lower than 6% and thus would reveal substantially higher pension costs.
Another problem is that the closed 30-year level dollar amortization period used in the report, while more fiscally tenable than the open level percent amortization methods that never pay off pension debt, is still too long compared to the actuarially prudent standard that advises an amortization period of no longer than 15-20 years. Adopting a shorter amortization period would too result in higher required contributions.
These caveats mean that the report’s results may underestimate the impact of pension costs on state and local budgets. According to this article by pension expert Andrew Biggs, if public pension plans followed corporate pension funding rules—which require a bond-based discount rate and a much shorter amortization period—pension costs as a share of state and government budgets would increase fivefold from the current 4% to 20%. Even if we set aside the academic debate about the proper discount rate, the kind of investments necessary to achieve the 6% assumed returns will generate a wide range of outcomes that are not meaningfully captured by the report’s estimates.
Overall, the CRR report provides valuable information about the current pension and OPEB cost burden on state and local budgets. Many large counties and cities have alarmingly high pension and OPEB costs relative to their own-source revenues. However, even those states and municipalities that appear to have the costs under control, per the report’s estimates, may need to reevaluate their pension plans under more rigorous standards, as previously discussed.