It wasn’t long ago that the relationship between the funding of public pension funds and other postemployment benefits (OPEBs) and municipal creditworthiness went largely ignored. Not anymore--these obligations, which in many cases are seriously underfunded, are subject to various legal protections, and governments must pay them in the future.
At Build America Mutual (BAM), we are incorporating actuarial analyses of pension and OPEB liabilities into our review of every municipal bond we insure. We look at both the steps governments are taking to address shortfalls in funding, and metrics we have developed, based on actuarial calculations. Our ultimate credit decision many times rests on the results of our pension and OPEB analysis.
Can underfunded pension funds catch up? What are governments doing to confront the pension and OPEB underfunding issue?
In order to close a funding gap, an employer can take steps to increase plan assets, decrease plan liabilities, or a combination of the two. With respect to increasing plan assets, some employers are updating their funding policies to result in higher annual plan contributions. For example, the State of New Jersey has legislated a seven-year phased-in approach to funding the full actuarially determined pension contribution. Many governments are updating their actuarial assumptions, and notably, over the past several years there has been a trend towards adopting a lower investment rate of return assumption. And, especially since the Great Recession, many employers are reconsidering their asset allocations in order to (theoretically) maximize investment returns in line with their risk tolerance.
There are two basic approaches being used to reduce public sector pension and OPEB plan liabilities:
1. Cost shifting. Under this approach, an employer can shift plan costs to another government, or shift costs to employees. For example, a statewide teachers retirement system may require greater contributions from individual school districts, or a government may require employees to contribute a higher percentage of pay for their pension benefits. OPEB plans have been charging retirees more for their coverage, either through monthly premiums or increased deductibles or co-pays.
2. Plan design changes. Though cost shifting may be easier to implement, the only way to have an impact on the true economic value of pension and OPEB benefits is to change the ultimate benefits payable under the plan. There are many different possible changes to pension benefits, including: the benefit formula itself, plan eligibility, retirement eligibility, cost-of-living adjustments, vesting, pay averaging period, converting from a defined benefit to a 401(k)-like design, and others. The typical design change to OPEB plans relates to benefit eligibility. However, rarely have changes been made that affect the benefits of current employees and retirees.
How do we analyze pension and OPEB obligations? How do these obligations affect municipal credit analyses?
In our qualitative analysis, we take note of actions by governments, such as those described above, to determine whether the government is taking steps to confront the underfunding issue. We monitor the status of pension and OPEB reforms, and make judgments on whether we believe they adequately address the issuer’s underfunding on a long-term basis. We take into account currently effective reforms, and the likelihood that additional proposed reforms will be enacted into law, based on such considerations as statutory, contractual or constitutional protections of benefits. We also review the asset allocation of the pension and OPEB funds to see whether the funds are invested in risky asset classes, which could make the plan more prone to large investment losses in the case of another market downturn.
Quantitatively, it is important that the way we analyze the impact of pensions and OPEBs on a government’s creditworthiness be based on the fundamental principles of comparability and conservatism. The liabilities and cash requirements currently disclosed by governments in their audited financial statements are neither (rather, they are based on varying actuarial assumptions and methods that have tended to be quite aggressive). Therefore, we have adopted an approach to “normalize” both the unfunded benefit liabilities and the funding contributions.
This normalization includes use of: a standard investment rate of return assumption for liabilities (although the standard assumption is different for funded and unfunded plans); standard adjustment factors to convert liabilities and normal costs away from the return assumption used by the system actuary; market value of assets (as opposed to a smoothing approach); level dollar amortization of unfunded liabilities over a reasonable time period; and, minimum healthcare cost trend rates for OPEB liabilities.
The discount rate we have chosen for funded plans is 6%, and is based on a reasonable long-term expectation of investment returns on an average portfolio of plan assets, with a downward adjustment for conservatism. (Most public sector pension plans are valued using an investment return assumption of 7% to 8%.) Using a conservative long-term investment return assumption has an additional advantage in that it allows for comparisons across time periods, since this assumption would only be changed if emerging experience indicates that a new long-term assumption is warranted.
Once normalized unfunded liabilities and contributions have been calculated, they can be used in determining various ratios that indicate whether a government is carrying an excessive debt burden, and whether cash funding requirements are sustainable over the long-term (especially over the duration of the bond we are considering for insurance). Finally, the metrics we calculate based on normalized pension and OPEB data will not be fully comparable to generally available benchmarking data that exists for credit rating purposes. Therefore, we compare our metrics to a growing, internally-developed database geographically or by sector (i.e., cities, school districts, municipal authorities, etc.). Based on these comparisons, we can determine which governments appear to be carrying significant debt or cashflow burdens relative to others within a BAM-internal rating category, which in turn may have an impact on their overall credit rating result.
How has the actuarial analysis of pensions and OPEBs affected BAM’s willingness to provide its guaranty? Two examples:
Town of Reddink
Town employees participate in various defined benefit plans, depending on their job, including the State teachers plan, the Town’s collectively bargained Safety (police and fire) plan, and the Town’s plan for its general employees. On a BAM-adjusted basis, the combined pension liability is under 50% funded, and the OPEB liability is completely unfunded. BAM-adjusted annual contributions were three times the actual contributions in the most recent fiscal year.
Village of Hope Springs
Village employees participate in two pension plans: one for police and firefighters, and one that covers all other non-school personnel. The Village also sponsors an unfunded defined benefit plan for certain other Village employees. On a BAM-adjusted basis, the police and firefighter pension plan is about 15% funded, and annual contributions are more than twice the actual contributions in the most recent fiscal year. The pension plan for general employees is about 70% funded. All OPEB liabilities are unfunded.
Selected quantitative and qualitative factors are detailed in the accompanying graphic.
Conclusion: Based on BAM-adjusted metrics indicating the financial burden of pension and OPEB unfunded liabilities, and uncertainty surrounding the plan to confront the issue, BAM did not issue insurance on the Town’s bond issue. Though the Village’s metrics were indicative of an underfunding issue, BAM did insure the Village’s bond issue, because steps have been taken to confront pension underfunding, and the Village appears positioned to take further actions if necessary.
In conclusion, pension and OPEB obligations are an important factor we consider when we analyze the creditworthiness of municipal bonds for our guaranty. Our approach is to carefully consider these obligations from a comparable, conservative quantitative perspective, and also a qualitative perspective. The quantitative analysis provides us an understanding of the current state of the pension and OPEB burden. Though our quantitative analysis alone may indicate that some concern is warranted, the qualitative analysis may indicate that a government is taking the steps necessary to confront the challenge of underfunding.
Finally, we put our pension and OPEB analysis into the context of the overall credit analysis of the bond issuer. For example, there may be other government programs, such as state aid intercept programs, that provide a cushion against excessive debt burdens or budget strains. We may find that when combined with the government’s other debt burdens, the pension and OPEB liabilities may be manageable, even if on a stand-alone basis these liabilities seem high.