For corporate pension plan sponsors, it seemed that 2009 might go down as one of the most frustrating years on record. After weathering the storm of 2008, when the U.S. stock market declined 38.5%, 2009 was a lot better. Stocks made up more than half the loss, surging 23.4%. Good news—right?
"Not exactly," say Bart Pushaw and Alan Perry, consulting actuaries from Milliman's Dallas and Philadelphia practices, respectively. "While the 23% equity return was good news, the poor CFO is bound to get heartburn after learning that liabilities grew even faster than investment returns," says Alan.
Based on an analysis of the Milliman 100 Pension Funding Index, corporate pension funds (with an average equity allocation of approximately 50%) had a cumulative asset return of 13.2% in 2009. For the same period, the cumulative liability return increased 13.5%.
PPA mandates mark-to-market accounting
The jump in liabilities wasn't caused by a rash of employee retirements or layoffs. It was simply the result of an actuarial calculation mandated by the Pension Protection Act (PPA). As Bart explains, "Prior to PPA, liabilities were treated as a static part of the equation. The funding rate was self-determined, under the dominion of the actuary, and liabilities were guaranteed to grow only at that hurdle rate, which rarely moved."
PPA changed all that. Now, pension funds must value their liabilities using mark-to-market rules that tie liabilities to fluctuations in high quality corporate bond yields. As a result, sponsors need to model their liabilities because they may be even more volatile than the portfolio's investment returns.
That's what has caused the heartache for so many CFOs. Bart says, "For many plans, the net funded status came down. This created a cost control issue and a budgeting issue. The CFO has a full plate, piled high with both risk allocation and capital allocation problems."
Sophisticated asset-liability modeling needed
The solution lies in modeling assets and liabilities together. Asset-liability modeling (ALM) allows sponsors to see the impact of different strategies on the plan's funded status going forward. Then they can implement investment strategies to control their funding risk. These strategies are generally known as liability-driven investing (LDI). The most basic LDI strategy buys a portfolio of high-quality bonds with an average duration matching that of the liabilities. This way, the volatility of the investments closely matches that of the liabilities.
To cope with decreased expected returns (and increased pension expense) from the reduction in equity exposure, Milliman and others have created more sophisticated LDI strategies that, for example, use one part of the portfolio to immunize the liabilities and the other part to seek higher returns. The point is, the level of sophistication needed to manage a defined benefit plan's funding strategy increased dramatically as a result of PPA.
Interestingly, the computational power to handle the modeling for LDI strategies is not widely available. Alan observes that investment advisors have, historically, focused on assets, so many of them don’t have much experience in liability modeling. And traditional shortcuts no longer suffice. Alan explains: "Plan sponsors used to be able to get liability numbers from the actuary, and use them as baseline assumptions they could roll forward. Arguably, you could get away with that before PPA—but not anymore."
Milliman is fortunate that its ALM capability has always been robust. Long-time clients such as Axa and Metropolitan Life have extremely complicated pension plans that require detailed analyses of risk; as a result, Milliman developed systems that are capable of doing projections of full actuarial valuations under thousands of economic scenarios going forward 10, 20, 50, or even 100 years. This proved to be excellent preparation for the demands of PPA modeling.
LDI and endgame strategies
For all of its clients, Milliman provides a full range of ALM capabilities and LDI strategies that include the use of derivatives for hedging, risk budgeting, segment matching, and surplus management.
Also, because PPA requires plans to become fully funded over the next five to seven years, many plan sponsors are now trying to formulate an endgame plan. As Alan describes it, "We're helping to design investment strategies that allow sponsors to make a series of planned contributions while steadily de-risking the asset portfolio to protect their higher funded status. After that they never have to put in more money; either it's self sustaining, or they can terminate the plan and buy annuities.
That may sound like the end of the line, but Bart cautions, "even though the plan might be frozen, you can't just put it on the back burner and not worry about it. These things have a nasty habit of coming back to life when least expected. Continuing to use ALM is particularly important—to avoid surprises."