In September 2012, the Society of Actuaries (SOA) released the results of research indicating that recent U.S. mortality improvements have outpaced expectations. The “Mortality Improvement Scale BB” report was authored by the Retirement Plans Experience Committee (RPEC), a subcommittee of the SOA whose primary directive is to research mortality experience among U.S. retirement plans. The RPEC also publishes the mortality tables and mortality improvement scales that are in frequent use for actuarial valuations of pension plans in both the public and private sectors.

Currently, the RPEC’s most recently published mortality tables in the United States are the RP-2000 tables. These tables contain expected rates of mortality by age, and are mandated by the IRS for use in U.S. corporate pension valuations. The RPEC had also previously published Scale AA, a scale of mortality improvements intended for use with base mortality tables to reflect the expectation that mortality is improving over time. However, the most recent research shows that even greater actual mortality improvement has been occurring than was predicted by Scale AA.

Additionally, the RPEC research has revealed that age alone may not be the best predictor of mortality improvement; some birth groups may experience rates of improvement that differ from other birth groups, or entire populations may experience a boost in longevity that is due to medical achievements such as antibiotics, regardless of age. For reasons such as these, the RPEC believes the best predictors of mortality improvement may be tied to both age and calendar year. This would create the need for a two-dimensional table of mortality improvement rates, which some current actuarial software may not be able to utilize without modification.

To that end, the RPEC has released an interim improvement scale, known as Scale BB. Scale BB, although not itself two-dimensional, incorporates information from the fully two-dimensional rates developed by the RPEC. It also reflects the stronger improvement patterns seen in recent years. Although the impact of using Scale BB in actuarial valuations will vary by pension plan based on plan provisions, maturity, and other factors, Scale BB is expected to result in higher liabilities as it reflects increasingly lengthy lifetimes for benefit recipients. The RPEC study estimates that switching from Scale AA to Scale BB might increase liabilities between 2% and 4%, based on some sample plan analysis.

There is ongoing debate in the actuarial community as to the use of Scale BB, particularly centering around some assumptions used in the creation of that scale. The RPEC anticipates the release of final mortality tables and a final improvement scale by the end of 2013 or early 2014. Therefore, whether or not the use of Scale BB becomes widespread, increases in mortality improvement will likely be reflected within the next several years for many pension plans.

To read the entire SOA report, click here. For more information on actuarial mortality assumptions, click here.

In July 2012, Moody’s released to subscribers a “request for comment” draft of proposed adjustments to be made by Moody’s to the pension data reported by U.S. state and local governments. These adjustments, if adopted, could influence local government credit ratings. Moody’s has indicated in this draft report that state credit ratings might not be impacted by the changes; however, the proposed adjustments would still be made and considered for state plans.

The proposal would involve several key changes to be made to the reported liabilities of state and local government pension plans. These would include adjusting liabilities to reflect a 5.50% discount rate using a uniform assumed liability duration of 13 (an assumption which would result in an approximate 13% increase in liabilities for every decrease in discount rate of 1%); using market value of assets rather than any smoothed value reported by the plan; amortizing any resulting unfunded actuarial accrued liability over a 17-year period on a level-dollar basis; and allocating liabilities for cost-sharing plans according to proportionate share of total plan contributions.

Moody’s has stated that these adjustments, if adopted, “would likely result in rating actions for those local governments where the adjusted liability is outsized for the rating category,” and where the plan has not shown the ability to increase funding or otherwise respond to shortfalls.

However, some actuaries and plan sponsors have voiced concerns that these adjustments may be too broad to appreciably improve metrics for individual plans. In particular, using a duration of 13 to adjust liabilities may cause some very mature plans to appear more sensitive to changes in the discount rate than they actually are, while some younger plans may in fact be more sensitive to changes in discount rate than a duration of 13 would predict. Additionally, amortizing the resulting unfunded actuarial accrued liability (based on market value of assets) over a 17-year period on a level-dollar basis is likely to show significantly higher contribution amounts in early years of the unfunded liability amortization, because most systems use a level percentage of pay methodology, which results in increasing contribution dollars over time.

The deadline for comment on the Moody’s proposal was August 31, 2012; however, no final Rating Implementation Guideline has yet been published, so it remains to be seen to what extent any final guideline will mirror the draft proposal. Follow Moody’s here.

Milliman last week released its first Public Pension Funding Study. The purpose of the study was to independently measure the aggregate funding status of the 100 largest public pension plans in the United States, responding to statements made by some sources that public pension liabilities are being significantly underreported.

The study found that, in aggregate, the independently calculated actuarial accrued liabilities of these plans were only 3% higher than the liabilities reported by the plans. While the weighted average investment return assumption being used by the plans was 7.80%, the independent calculations performed in the study suggested a weighted average rate of 7.55%, or a decrease of just 25 basis points from the assumption the plans are actually using. This is in line with the recent trend of plans lowering their investment return assumptions to reflect reduced capital market assumptions since the economic downturn. Interestingly, many plans in the study were actually using a more conservative approach than the independent calculation showed, with 33 of the 100 plans having an investment return assumption lower than the independently calculated rate (and therefore reporting greater actuarial accrued liabilities than the study independently measured for them).

The aggregate funding ratio of the 100 plans, using a market value of assets (MVA) (i.e., the value of the assets with no deferral of gains or losses) and the actuarial accrued liability calculated using an independent measure of investment return rates, was 67.8%. This compared to an MVA-funded ratio of 69.8% as reported by the plans. This indicates that the plans in aggregate are reporting a reasonable estimate of their liabilities. On an actuarial value of assets (AVA) basis, the independent funded ratio measured in the study was 73.0%, compared to 75.1% as reported by the plans.

Although much negative attention has been given to the state of pension plans in the United States, the findings of the Milliman study indicate that on the whole the plans are aware of their funding shortfalls, and are using reasonable investment return assumptions to perform measurements of their liabilities. It should also be noted that many public retirement systems and their plan sponsors are additionally addressing funding needs by adjusting benefit levels; a recent study by the National Conference of State Legislatures found that 43 states have enacted significant pension reform legislation for at least one statewide retirement plan from the beginning of 2009 through the end of 2011.

Clearly, significant additional funding is needed to improve the long-term financial solvency of public pensions; however, based on Milliman’s report, most plans appear to be adhering to a high level of transparency and appropriate calculation in the reporting of pension funding.

More information, including the complete study, can be found here.

In August 2012, the Governmental Accounting Standards Board (GASB) published final accounting regulations for U.S. public pension funds. The new rules will be effective beginning in 2013 for pension plans and 2014 for employers.

At the heart of the new GASB accounting rules is a fundamental separation between funding and accounting. Under the previous rules, plans calculated a GASB annual required contribution (ARC). The ARC is frequently used as a funding measure due to its consistency with certain actuarial funding principles. Plans are not required to contribute the ARC; however, any shortfall or excess contribution is accumulated from year to year into a number called the net pension obligation (NPO). The NPO appears in the notes to financial statements.

Under the new GASB rules, no such metric as the ARC will be specified. Rather, a balance sheet liability and an income statement expense will be calculated. The liability, referred to as the net pension liability, will closely resemble the number currently referred to among actuaries as a “market value of assets unfunded accrued liability,” with certain restrictions imposed on the calculation methodologies (including the discount rate). Disclosing the volatile net pension liability as a line-item on the balance sheet will be a big change for many employers. The income statement expense number, referred to as pension expense, will be designed to meet accounting definitions of period-specific accrual, and will likely be subject to high volatility because of the specific calculation instructions in the new GASB rules. GASB has made it very clear that neither the net pension liability nor the pension expense numbers should be used for funding purposes.

In the absence of any useful accounting metric for funding policy guidance, it will fall to other entities to develop any across-the-board funding standard in the public pension community. This represents an opportunity to enhance the understanding of what it means to fund plans responsibly; the GASB ARC was considered by many actuaries to be insufficient for long-term plan funding. Actuarial entities such as the Conference of Consulting Actuaries (CCA) and the California Actuarial Advisory Panel (CAAP) are currently in the process of deliberating on guidance for funding of U.S. public pension plans. Ultimately, the funding policy decision for each plan will remain, as always, with plan sponsors, retirement system boards, and legislative bodies.