To the list of defined contribution strategies that attempt to mimic the reliable payout of a traditional pension, add a new one from Russell Investments: long-duration bonds.
The idea — as with annuities and their newest variations — is to create a program that will stretch 401(k) assets to last through all the years after retirement. “If plan sponsors are truly thinking about defined contribution plans as retirement income vehicles, not just savings vehicles, then there’s a case for extending the duration of bonds,” says Josh Cohen, Russell’s defined contribution practice leader.
Russell doesn’t have an actual product yet, but Cohen says he and his colleagues are talking with plan sponsors “about the research, trying to get our clients and the industry thinking in terms of retirement income.”
Right now, the typical bond duration in a 401(k) target-date fund is four and a half years, according to Cohen. However, taking the same factors used for creating annuities — including projections of longevity, interest rates, and spending — he says a newly retired 65-year-old would need a bond with a duration of eight and a half years to have a sufficient payment stream.
The simplest way for a plan sponsor to do this, Cohen continues, would be to use Treasury STRIPS (separate trading of registered interest and principal securities) for the fixed income portion of their target-date funds. The reason for using STRIPS, in which the interest payments become their own separate securities, is that their duration and maturity are the same, unlike with traditional bonds, says Daniel Gardner, a defined contribution analyst at Russell. And the reason for using target-date funds, Cohen adds, is the same reason more and more plans are turning to pre-programmed asset management of all types. “This is another way to embed something that would be a complicated concept.”
Other than substituting one kind of bonds for another, Russell wouldn’t alter a target-date fund’s overall glide path, or asset allocation. Thus, a typical fund would have a significant chunk in equities. Why not put the whole caboodle in bonds, the way some defined benefit plans use the tactic of liability matching to align their pension payout stream with a staggered series of bond maturities? Most employees probably haven’t saved enough for retirement and will need the kick of stocks’ higher returns, Gardner explains.
Of course, the glide paths can begin 40 years before retirement, depending on when the person first invests. By Russell’s calculations, that would mean that a 2050 target-date fund would need bonds with a 48½-year duration, which don’t exist. Thus, the firm says that the funds would probably have to start out with regular bonds and switch to longer-duration types 15 years before the target retirement date.
With interest rates near zero, this might seem like a lousy time to be suggesting a product that locks people into long-term rates. However, Gardner says that if rates do go up, “there is reason to believe that the greatest interest-rate movement would occur on the short end, so it might not be as detrimental as you might think.”
Gardner argues that this concept has several advantages over the annuity-type “lifetime income” products that insurance companies and other vendors have been avidly pitching to plan sponsors. The bonds have more liquidity, and their payment stream wouldn’t be tied to the solvency of one insurance company for decades to come. “Holding a diversified bond portfolio, that risk goes away,” Gardner says.
That may be. The annuity lookalikes have been slow to catch on, and certainly, it would be nice for plan sponsors to have a choice. But at least products like Prudential’s “income flex target” and MetLife’s “personal pension builder” actually exist. So if employers want to help their employees tight now, they will probably have to settle for flawed annuities.
Or, of course, keep their defined benefit plans.
