One is the loneliest number

With the 10-year Treasury well below 1.0% and pension discount rates continuing to hit new record lows, there are some uncomfortable questions many pension funds are (or ought to be) asking themselves. In the remainder of this piece, we outline areas for concern and implications for plan sponsors:

Necessity for convexity: Most pension plans have long duration liabilities, but these liabilities also have convexity. This means that as discount rates fall, the duration of liabilities will extend, and as discount rates rise, the duration of liabilities will contract. This is, of course, the exact opposite of what we would want, because it magnifies the liability increase when rates fall and dampens the liability decrease when rates rise. The chart below outlines how a $1bn pension liability today (as of March 5, 2020) might look at lower and lower discount rates. We see that as discount rates fall, both duration and convexity increase as the liability grows more and more sensitive. At a 0.0% discount rate, the DV01, the dollar value of a 1 basis point change in discount rate, is 1.6x the level today for the same liability.

Convexity in pension liabilities

Source: BAML AA Corporate Yield Curve, J.P. Morgan Asset Management. Calculations based on hypothetical pension liability with duration of 13 years as of 3/5/2020. All calculations assume a parallel shift in interest rates across the term structure.

Practically this means that discount rate changes can result in meaningful hedge ratio drifts, especially in a low rate environment. Plan sponsors should be vigilant about monitoring hedge ratio drift, especially if their strategy does not allow for flexibility around a target level.

Don’t forget about credit spreads: Pension discount rates have both a Treasury and high-quality credit spread component. Historically, high-quality credit spreads and interest rates have been negatively correlated, which has dampened the volatility of pension discount rates relative to Treasury yields. The chart below decomposes discount rate changes during some of the largest monthly declines in Treasury yields. We see that across most of these periods, spread widening provides some relief to falling yields, and that is certainly the case with the Treasury rally we’ve experienced year-to-date.

Discount rate changes during the largest monthly declines in Treasury yields

Source: BAML ICE Mature Pension Liability Index, J.P. Morgan Asset Management.

The downside of widening credit spreads is the potentially and often asymmetric impact on asset portfolios that often are down in quality relative to pension liabilities. Widening spreads also commonly coincide with sell-offs in risk assets, which is also the case with year-to-date experience. Mechanically, downgrades and credit losses are crystalized in the fixed income assets while the liability has immunity. We have been strong proponents of diversifying corporate credit exposure in the asset portfolio, which generally leads to improved funded status outcomes in a spread-widening environment and especially during a downgrade cycle.

Less than zero: What would happen if interest rates and possibly even pension discount rates fell below the zero lower bound? There are a range of impacts for plan sponsors to consider:

• Discount rates: ASC 715-30-35-44 describes the objectives of setting pension discount rates under U.S. GAAP to “measure the single amount that, if invested at the measurement date in a pension of high-quality debt instruments, would provide the necessary future cash flows to pay pension benefits when due.” If high-quality bonds produce a negative discount rate, can cash at zero yields satisfy this objective? In 2017, facing negative yield in the front-end of the JGB curve, the ASBJ issued guidance that permits a discount rate floor at zero. Would the same be allowed under U.S. GAAP?


• Interest cost: Interest on outstanding pension liabilities is a component of both pension expense and the change in liability over any given period. With a negative discount rate, would pension liabilities shrink with the passage of time, all else equal? Would interest on the liability actually be accretive to corporate earnings?


• Cash balance plans: Most cash balance plans utilize a Treasury-based interest crediting rate to grow account balances, sometimes with a minimum guaranteed crediting rate. How expensive is a promised 5.0% minimum guarantee on a 30-year Treasury bond yield when the market rate is below 1.0%? If the crediting rate is negative, and account balances start shrinking with the passage of time, how will that impact participant behavior? Finally, IRS regulations stipulate a “preservation of capital” rule for cash balance plans, meaning that the account balance cannot be less than the sum of principal/pay credits. How expensive is this option, which plan sponsors are implicitly short?


• Pension funding: Rising PBGC premium rates have increased the attractiveness of issuing debt to fund pension deficits. If companies can borrow at zero (Berkshire Hathaway just filed a pricing term sheet with the SEC for €1.0bn 5-year debt at 0.0%¹), will they be enticed to borrow to fund the pension?

¹  https://www.sec.gov/Archives/edgar/data/1067983/000119312520062545/d876702dfwp.htm

0903c02a82839234