The move from defined benefit (DB) pension plans to defined contribution (DC) has been ongoing for years in the US, both in the private and public sector. But more recently many state and local governments have adopted hybrid designs.
“There is a wide range of hybrid plan designs, each offering tradeoffs in terms of retirement benefits, risks, and costs,” says a new report1 published by the National Institute for Retirement Security (NIRS).
“Using the term hybrid as if it represents one design can obscure the important discussions that must occur around those tradeoffs,” the authors, Dan Doonan, NIRS executive director and Elizabeth Wiley, consulting actuary at Cheiron, point out.
“The new NIRS report helps taxpayers, plan sponsors, and members understand that there is not just a binary choice between a DB or a DC plan, but there are a lot of different combinations of DB and DC ideas,” says Becky Sielman, author of Milliman’s Public Pension Funding Study, which annually reports on the funded status of the 100 largest US public pension plans, and of a quarterly Public Pension Funding Index.
There is nothing such as “the best hybrid plan”, says Sielman, but she likes the horizontal hybrid model – with two modest DC and DB plans side-by-side – that during the past 10-15 years has been a frequently adopted model. “It is simple, easy to understand, and it works in sharing risks between the plan sponsor and the plan members, which is the main reason to move from a DB plan to a hybrid one,” says the Milliman consultant. “With a horizontal hybrid model fewer things can go wrong, even though the administrative burden gets heavier.”
Things did go wrong, with unintended consequences, when employees at one municipality had to choose between a DB or a DC option, says Sielman. “After the 2000 dot-com bubble burst, employees who had decided to leave the DB plan and move all their money to the DC plan suffered significant investment losses and asked to go back to the DB plan”, she says. “But that was impossible, and the employees were unhappy.”
A hybrid plan, that is a DB plan with some risk sharing, is a good way to help control taxpayer liability to from public pension plans, according to Donald Boyd, co-director of the Project on State and Local Government Finance, at the Rockefeller College of Public Affairs & Policy of the University at Albany, New York. He also co-authored Public Pension Risk-Sharing Policies: A Policymaker’s Guidebook.
“Public governments must understand that if pension plans invest in risky assets, you cannot make those risks go away,” says Boyd. “Risks about investment returns, longevity, inflation, and interest rates are there, and either the taxpayers or the funds’ members have to bear the risks’ burden.”
The Rockefeller College researchers used simulations to demonstrate how risk-sharing policies affect the trade-offs between contribution risk for employers and risks for plan members of lower benefits or higher contributions. “Actually most hybrid plans have not shifted a lot of the risk burden,” says Boyd. “Our simulations show that risk-sharing policies that only modestly adjust the structure of DB plans, such as contingent cost of living adjustments (COLAs) or contingent employee contributions based on the funded ratio or investment performance, have only a limited impact on reducing the contribution risk for employers.”
Protecting the public purse
The Wisconsin Retirement System (WRS) and the South Dakota Retirement System (SDRS) are two good examples of “more extensive risk-sharing arrangements that can significantly reduce employer contribution costs and risks, but at the expense of greater risks for plan members”, according to Boyd.
In Wisconsin, employees and employers share the actuarially determined contribution, which varies: it goes up if returns are bad. “This feature is an incentive to take less investment risks and, combined with other features of the Wisconsin plan, the results are positive: the WRS is virtually fully funded,” says Boyd.
In South Dakota, the SDRS board wanted to keep employers’ contributions stable and the plan fully funded. So they adopted a DB model with variable benefits: the COLA depends on assets meeting liabilities. If the Fair Value Funded Ratio (FVFR) is less than 100%, the maximum COLA will be the actuarially calculated percentage that brings the FVFR back to 100%; if that is not possible, then a corrective action plan is required.
Public pension funds that do not adopt risk-sharing and continue to invest in risky assets in an effort to reach high return assumptions – typically 7% – face great hazards. Sooner or later they will get into trouble that must be borne by either taxpayers or plan members, says Boyd.
In the private sector there is an increased interest in hybrid models, says Sielman. “In the past, almost all companies that adopted a hybrid plan chose the cash balance model, and then they froze it or terminated it,” she says. “But I’m starting to hear companies that are a little concerned about their employees not being able to retire because they don’t have enough retirement savings. Those companies may think of changing their benefit model.”
1 The Hybrid Handbook. Not All Hybrids Are Created Equal, NIRS, May 2021