Liability-driven investing: matching the assets to the promise

A frozen pension plan with no new accruals, no new entrants and no change in the benefit formula can still see its funded status move five points in a quarter. The usual explanation, that markets were volatile, is only half right. The other half is that the plan's assets and its liability respond to different things, and nobody designed them to move together.

Three liabilities, one set of assets

A corporate plan sponsor lives with at least three measurements of the same promise. ASC 715 discounts the projected benefit obligation using a yield curve built from high-quality corporate bonds, matched to the plan's own expected benefit payments. The minimum funding rules under Section 430 use segment rates that are smoothed and, since 2012, held within a corridor around a 25-year average, so they move slowly. A termination or settlement measurement uses what insurers charge for annuities and what Section 417(e) requires for lump sums, which track market rates with little smoothing.

The accounting liability is the one most CFOs watch, and it is the most rate-sensitive. For a typical frozen plan the duration is somewhere between ten and fourteen years, which means a half-point drop in corporate yields adds roughly five to seven percent to the obligation. A portfolio that is 60 percent equities does not move with corporate yields. When rates fall and equities are flat, the deficit widens with no change in the plan at all.

That gap shows up in places the finance team cares about: the pension line on the balance sheet, other comprehensive income, the year's pension expense, the next contribution requirement and the PBGC variable-rate premium, which is charged at $52 per $1,000 of unfunded vested benefits, subject to a per-participant cap that is indexed each year.

What a liability-matched portfolio does

Liability-driven investing builds part of the portfolio to behave like the liability. In practice that means long-duration fixed income chosen so that its interest rate sensitivity tracks the plan's own cash flows, usually a mix of long corporate bonds, long Treasuries and Treasury STRIPS, sometimes supplemented with interest rate swaps or futures when the sponsor wants more rate exposure than the physical bonds provide.

The central measure is the hedge ratio: the dollar duration of the fixed income portfolio divided by the dollar duration of the liability. A plan at 30 percent still carries most of its interest rate risk. A plan at 80 percent has removed most of it, and from that point funded status is driven mainly by the return-seeking assets and by the behavior of credit spreads.

Credit spreads deserve their own line. Because the accounting liability is discounted on corporate yields, it falls when spreads widen. A hedge built entirely from Treasuries will not capture that, and funded status can move even when the Treasury curve is still. Most programs hold a meaningful share of long credit for this reason, and accept that corporate bonds carry default and downgrade risk that the liability does not. Curve shape matters as well: a liability with payments concentrated twenty years out is not hedged by bonds with a duration that happens to match on average. Key rate durations, measured at several points on the curve, are the practical check.

Glide paths

Few sponsors move to a full hedge at once. The usual approach is a glide path that raises the hedge ratio and lowers the return-seeking allocation as funded status improves. A plan at 85 percent funded might hold a 50 percent hedge; at 95 percent, 70 percent; at 105 percent, 90 percent or more. The triggers are the plan's own numbers, so when the market gives the plan a good quarter, part of the gain is locked in rather than left at risk.

A glide path has two failure modes. One is that the triggers are hit but nobody acts, because the committee meets quarterly and the market does not wait. The other is that the path was set once and never revisited, so a plan that changed its endgame is still following the old route. Both are governance problems rather than investment problems, and both are avoidable.

Hedge toward which liability?

This is the question most often skipped. A sponsor planning to run the plan for decades (sometimes called hibernation) should hedge the accounting liability and hold enough return-seeking assets to cover expenses and PBGC premiums. A sponsor planning to buy annuities within a few years should hedge toward insurer pricing, which has its own duration and spread behavior. A sponsor planning a lump sum window should hold short-term liquidity for the payouts and understand that 417(e) rates reset annually. These targets differ, and a portfolio built for one will not serve the others well.

Questions to settle before starting

• What is the endgame? Hibernation, a lump sum window, a partial annuity purchase or a full termination each imply a different target portfolio and a different hedge.
• What hedge ratio today? Current funded status, the sponsor's capacity to contribute and how much balance sheet volatility the company can absorb all bear on it.
• Rates, spreads or both? The answer follows from the discount curve the plan uses and the sponsor's view on credit risk.
• Who watches the triggers? A glide path works only if someone checks funded status at least monthly and has authority to rebalance.
• How will the hedge be measured and reported? The committee should see the hedge ratio, the key rate exposures and a funded status attribution each quarter, not only the asset return.

What changes in the committee room

Without a matched portfolio, the annual pension review is often a report of what happened: rates fell, the deficit grew, contributions will rise. With one, the review is about decisions the committee can actually make: how much risk to keep, how quickly to de-risk and when the plan will be ready to settle.

We project the liability cash flows under each measurement basis, test hedge ratios and glide paths under interest rate, spread and equity scenarios, and present the trade-offs to the committee on one page, with the detail behind it.

This article is general information about the topics discussed and is not actuarial, legal or tax advice for any specific plan or employer.