(September 20, 2012) – Traditional defined-benefit (DB) and defined contribution (DC) plans both fail to meet plan sponsors’ dual goals of adequacy and affordability, according to Keith Ambachtsheer, head of the University of Toronto’s Rotman International Centre for Pension Management.
“The time has come to stop defending the indefensible,” Ambachtsheer writes in a just-published article for the Rotman International Journal of Pension Management, which he also heads.
He argues that one of DB plans’ oft-touted advantages—that they provide simple and complete contracts between retirees, employees, and employers—doesn’t hold water: “The value of the guarantees embedded in DB contracts is often understated by using discount rates that embody the assumption that projected risk premiums will become realized risk premiums. In game theory terms, this is a mechanism for shifting wealth to current plan participants from whoever is underwriting the embedded payment guarantees. These “paper” wealth transfers are eventually realized through demands that benefits be increased in good times, and by enforcing the embedded payment guarantees unwittingly made by guarantors in bad times. The seriously underfunded condition of many public-sector plans today is finally forcing public-sector employers to recognize these fundamental design problems of traditional DB plans.”
Where DB plans by nature tend to become excessively expensive, he argues, DC plans tend to fall down on both the affordability and adequacy fronts.
The Teachers Retirement System of Texas (TRS), a $107.4 billion DB plan, came to the same conclusion about DC earlier this month. At the behest of the Texas Legislature, the fund studied a potential switch to DC and hybrid structures. According to the pension’s models, switching new members to either a self-directed or pooled defined contribution plan would increase unfunded liabilities by $14.7 billion, or 61%. “It is important to understand why the alternative plans modeled by TRS are more expensive than the current defined benefit plan to provide the same level of benefits,” TRS’ report explained. “The main reason for the expense difference is that most alternative plans do not generate the same level of investment returns as the defined benefit plan.”
While Texas Teachers’ doesn’t seem keen on restructuring any time soon, Ambachtsheer did come up with a set of design questions for modern employment-based pension plans:
1) How much is the employee/employer willing to pay to achieve that target pension?
2) What is a reasonable prospective net real investment return that can be assumed on a conservatively invested portfolio of retirement savings today? What additional reward for risk taking is it reasonable to project?
3) What respective lengths of the work and post-work periods should be assumed in funding the plan? How is retirement-age flexibility best built into the plan design?
4) To what degree, and how, should uncertainties about net real returns, inflation, and longevity be mitigated? If the plan offers guarantees, what are the mechanisms through which these guarantees are priced and enforced?
In conclusion, Ambachtsheer makes the point that many major pensions, and Texas Teachers’ would surely qualify, are well-governed, modern and high-performance organizations. These institutions, he argues, would be better off dropping the DC vs. DB debate and focus on crafting plan structures that “strike an intelligent balance between the dual goals of adequacy and affordability.”
Read Ambachtsheer’s whole article, titled “The Dysfunctional ‘DB vs. DC’ Pensions Debate: Why and How to Move Beyond It,” here.