LDI, Meet Your Little Brother: TDF

There is an uncanny spillover of liability-driven investing into defined contribution plans—and from an interest rate perspective, it's perfectly logical.

The most rudimentary tenet behind liability-driven investing (LDI) is, simply put, liabilities, says Kathy Lutito, CIO of CenturyLink Investment Management.

“LDI goes beyond investing in bonds,” she continues. “It’s about truly understanding a plan sponsor’s liabilities to define how assets should be managed.” Lutito, an early mover into LDI, began investing her company’s now-$12 billion fund in both liability-hedging and return-generating portfolios in the years prior to the financial crisis. Since then, significant changes have been made to accounting practices requiring plan sponsors to report pension plans’ funded statuses in their balance sheets, thereby jumpstarting the de-risking movement. And the numbers don’t lie: More than two-thirds of 119 plan sponsors studied in CIO’s 2013 LDI Survey said they currently implement LDI strategies.

However, with the retirement tide continually turning from defined benefit (DB) to defined contribution (DC) plans, an even larger trend is emerging: the confluence of thought behind both LDI and target-date funds (TDFs).

The concept of gathering and investing assets is strikingly and naturally similar in both systems, says Jodan Ledford, head of US solutions at Legal & General Investment Management America. The “comparable lifecycle approaches” between LDI strategies and TDFs are most apparent in the glide paths, he says. “As you get closer to fully funded in LDI, you’re going to look to lock down the risk—something you’d also see in a DC plan that is close to a terminal-funding position or as the participant nears the retirement date. They’re both de-risking, allocating more to fixed income, to ensure a more reliable drawdown with respect to taking benefits in retirement.”

Despite the uncanny spillover of LDI into DC plans, the dots haven’t been connected until recently, says David Eichhorn, managing director of investment strategies at NISA Investment Advisors. While LDI in the DB space is “focused on mitigating interest rate risk, there exists a similar problem for individuals with the responsibility to pay themselves income during their retirement. The liability is truly identical from an interest rate perspective. It’s perfectly logical.”

However logical the connection may be, most current TDFs—especially in their fixed-income portion—may not be de-risking as effectively as LDI programs are. According to NISA, fixed-income allocations that may seem low risk could actually introduce substantial volatility into retirement income as interest rates change. An answer to this problem could lie in a fundamental goal of LDI, the firm argues: building a stable retirement income by hedging risks and volatility. In this case, DC plan sponsors should better understand “the central role that interest rates play in determining how much retirement income a given account balance will be worth in the future.” The key lies in targeting interest rate sensitivity in the fixed-income portfolio, simply by matching the duration of the bond to that of the liability.

“De-risking pension plans has informed institutional investors in the DC space about the role of duration, especially the role long bonds can play in a diversified portfolio,” says Joe Nankof, co-founder and partner of consulting firm Rocaton. “Broadly speaking, target-date programs usually allocate to market-duration fixed income, something similar to the Barclays Capital Aggregate Bond Index. Today, relatively few TDF families make use of long bonds.” The role of matching bond duration to DB liabilities via LDI strategies is clear: According to CIO’s 2013 LDI Survey (Figure 1), the average fixed-income duration was slightly under that of the liability at 11.7 years to 13.5 years. Through the lens of LDI, duration targets for DC fixed-income allocations could be one’s life expectancy or mortality rate, Eichhorn says. 

CIO914 Chart - S&T Fixed Income and Liability


The firm’s calculations find that, in the run-up to retirement, each additional year of age signifies one year less in duration. Once the participant begins retirement, duration falls more slowly as cash benefits are distributed and shorter-duration cash flows are taken off the portfolio. However, NISA emphasizes that, before a retiree reaches the age of about 80, the target duration is greater than five years—and much greater for participants who are still working and accumulating assets—suggesting that “the typical ‘core’ or broad market-duration allocations within DC plans may fall significantly short of the duration required to hedge retirement income objectives.”

In determining how far into the future retirement spending is likely to occur, the St. Louis, Missouri-based firm notes that duration must take into account mortality assumptions. The new drafts of life expectancy rates and improvement scales (Figures 2 and 3) from the Society of Actuaries will soon become mandated standards for corporate pension funding ratio and lump-sum payout calculations, much to the displeasure of plan sponsors who will see liabilities rise. For example, a 65-year-old male, expected to live 84.6 years under the previous scales, now has a predicted lifespan of 86.6 years. The life expectancy of a typical woman of retirement age climbs from 86.4 to 88.8 years. NISA argues that the portfolio duration should be able to adjust accordingly to changes in longevity.

 CIO914 Chart - Male Annual Mortality 

 CIO914 Chart - Female Annual Mortality 

Climbing mortality improvement rates and the increased focus on decumulation of retirement assets may only add to the institutional challenges of understanding participants’ liabilities, longevity risk, and retirement income goals, Eichhorn says. While matching fixed-income duration to that of the liability does not hedge the risk of retirees outliving their retirement income, “by upping the duration in the portfolio, what the individual is de facto doing is hedging an annuity purchase,” he says. “Even if the individual never intends to buy an annuity, it means the retiree chooses to self-insure his or her own longevity, a strategy that allows [that person] to generate stable retirement income without bearing interest rate risk along the way.”

Rocaton’s Nankof agrees that this is an area where the spillover of LDI into DC plans gets interesting: “How retirement providers hedge annuity risks through investment strategies is very much akin to LDI strategies in pension plans. These are investment strategies for the long term that might have a glide path, but instead of providing retirement income for a group of individuals as a pension would, the focus is on one retirement income for one individual.”