The Government Accountability Office (GAO) released a comprehensive report on defined benefit plan (DB) discount rates to inform the debate surrounding the appropriate interest rates used to value public plans. While the report does not provide definitive answers or give specific recommendations, its findings reveal an overall picture of the use of pension discount rates in the US.
The report identifies two primary approaches to determining a pension plan’s discount rate: the assumed-return approach and the bond-based approach. The former is based on the long-term average rate of return on the plan’s assets, while the latter is based on marker prices for fixed-income securities that are similar to the plan’s benefit promises. In the United States, public sector plans and private sector multiemployer plans use the assumed-return approach, while private sector single-employer plans use the bond-based approach. As a result, the two groups have potentially large differences in funded ratios and reported liabilities.
According to the experts interviewed by GAO, the appropriate discount rate to use depends on the purpose of measurement. The report lists five key purposes for which one might determine a discounted value of future benefits: (1) determining required contributions; (2) reporting plan liabilities to key stakeholders; (3) determining the funding amount needed to terminate the plan or to guarantee pensions to date; (4) expressing the value of participants’ benefits; and (5) determining optional lump sum amounts payable to participants in lieu of an annuity. The experts also identified several considerations in setting discount rate policy, including cost, risk, fairness, sustainability, transparency, and comparability. Many experts agreed that reporting multiple measures of plan obligations, using different discount rates, might be of value when there are competing purposes and considerations.
The GAO argues that historical returns have limited usefulness in acting as evidence for assuming a specific rate of return (8% for example), due to four main challenges. First, analysis of returns on overlapping rolling historical periods has significant statistical limitations. Second, historical returns vary with the time period used in the analysis. Third, actual returns would depend on plan characteristics and cash flows. Fourth, investment returns and plan benefit levels are not independent variables.
The report also finds that public plans in the United States generally use higher discount rates than plans in Canada, the Netherlands, and the United Kingdom.
To read the full report, go here.