While hindsight is 20-20, as of this writing, the new issue appears well received by the market. With the 20-year bond back in the picture for the first time since 1986, we look at the implications for pension plans and relevance for liability-hedging portfolios before providing some commentary on pricing, liquidity and demand. Overall, we believe this issue will be a useful addition to pension portfolios in the longer term as it fills an important key rate to hedge liability risk. However, we suggest a wait-and-see approach in the near term given that initial pricing does not suggest a screaming buy.
Implications for Pension Plans
Pension plans typically use a combination of US Treasurys and corporate bonds to provide the bulk of exposure to manage against long-dated liabilities. Within the long-dated US Treasury bucket, new issuance has only been arriving in the form of 10-year and 30-year maturities for the past 14 years (the 30-year was just reintroduced in 2006). While there are other long-dated securities that might be considered (STRIPS, TIPS, mortgages, etc.), Treasury bonds typically provide cheap, liquid, reliable duration to help hedge the liability exposures.
KEY RATE DURATION: THE 20-YEAR GAP
As more and more plans consider fine-tuning their liability-hedging strategies by focusing on key rate durations (e.g., through “completion” mandates), the gaps between the 10-year and 30-year parts of the Treasury curve become more noticeable. Figure 1 shows the breakdown of outstanding US Treasurys by maturity year as of May 2020 (excluding the new 20-year bond). It is clear that there is a period of time in the 2030s where there is very little debt maturing. In addition, the only debt maturing in the 2040s is from old 30-year bonds rolling down the curve. These “off-the-run” bonds tend to be less liquid and consequently more expensive to trade compared to “on-the-run” new issues.
From a key rate duration standpoint, Figure 2 shows the details for the most recent on-the-run 10-year note and 30-year bond. Clearly, neither of these instruments could provide a meaningful hedge against the 20-year key rate point. In fact, the majority of the key rate exposure for each lands solely in the maturity bucket as the low coupons (relative to history) have less impact on earlier key rates.
A MISSING PIECE OF THE LIABILITY-HEDGING PUZZLE
We typically advise plan sponsors to focus on higher-level risk factors such as overall duration and credit spread risk until a plan is late in its lifecycle. However, we believe it is still worth exploring the key rate exposures in order to properly manage overall plan risk and determine when fine-tuning the hedge may be warranted.
In Figure 3, we show the key rate durations of a generic plan liability with a 12-year duration discounted using the FTSE Russell Pension Discount Curve. Of course the shape of liability cash flows can vary from plan to plan based on several factors, but for the generic liability below, more than one third of the duration exposure falls into the 20-year bucket. Regardless of the plan liability specifics, there is likely to be a significant amount of exposure to maturities between 15 and 25 years for nearly all plans, which should cause plan sponsors to evaluate the best hedging approach for this part of the curve.
Overall, faced with the importance of managing risk across the curve but with limited instruments, we conclude that liability-hedging programs must either:
In practice, most plans typically use a combination of all three approaches. As mentioned previously, we believe approach 1 is reasonable for many plans that are not nearing the end of a glide path or seeking to minimize overall plan risk. Approach 2 is also common given that the potential benefits of achieving a key rate match may outweigh the transaction costs of off-the-run bonds, which are likely to be small relative to total plan assets.
Approach 3 is more nuanced. We believe overlays are generally a valuable tool to consider as part of liability-hedging allocations. In particular, incorporating derivatives such as Treasury futures, interest rate swaps or total-return swaps can help improve the 20-year key rate hedge. However, even Treasury futures (likely the most common tool used by US plan sponsors) do not solve the issues of hedging the 20-year key rate.
Figure 4 shows the key rate durations of the cheapest-to-deliver bonds associated with the two longest Treasury futures contracts, the classic bond (associated with Treasurys maturing between 15 and 25 years) and the Ultra T-bond (25+ years). With 6.4 and 9.0 years of duration, respectively, at the 20-year key rate, these contracts offer significant improvement compared to physical on-the-run Treasurys. However, unlike physical Treasurys, using either contract will also add duration at either the 10-year or 30-year point. This may introduce additional complications such as shorting other contracts (and increasing the gross notional exposure) in order to isolate the 20-year point.
The key rate durations of the new 20-year bond are shown as of May 21, 2020, in Figure 5 below. This option would better isolate the 20-year key rate point and offer a useful addition to the set of instruments used for liability-hedging purposes.
As noted above, the duration profile of this new issue is likely to fill a crucial gap in the term structure with a highly liquid instrument. Therefore, we expect that there will ultimately be sufficient demand for the issue. In addition to natural demand from the pension community, we also expect demand from passive fixed income funds (which are mandated to track their benchmarks), and active fixed income funds (which are likely to utilize the issue as a placeholder and/or hedge for 20-year corporate supply). We believe this issue’s ability to fill the 20-year gap in key rate duration profiles will ultimately make it an appealing option for many fixed income investors, especially those with longer-term investment horizons.
However, as with the introduction of any new benchmark tenor, initial valuations must be compelling enough to entice investors to allocate capital from other outstanding issues or tenors. While markets can shift rapidly, at the time of this writing, we do not believe that the relative valuation of the new 20-year meets this criteria.
As of May 21, 2020, the implied yield on the new 20-year Treasury was about 6 basis points higher than the yields on outstanding off-the-run Treasury securities with 20 years to maturity. However, these issues are not an appropriate comparison for valuation purposes for several reasons:
The new issue has a much lower coupon (1.125% compared to 4.375% for the outstanding 20-year paper).
It carries a much longer duration. With a duration of about 18.4 years, the new issue is about 3 years longer than the outstanding “high coupon” 20-year paper (see Figure 6 below).
The outstanding Treasury securities that most closely match the duration of this new 20-year are those that mature in 2044. The implied yield on the new 20-year Treasury is about 15 basis points lower than the yields on these comparable-duration securities.
Given its current status as an on-the-run security, the new issue 20-year bond should trade with a bit of a liquidity premium relative to matched-duration off-the-run securities (especially after what the market experienced in March of this year, when off-the-run liquidity became highly impaired). But investors focused on liquidity could match the overall duration of the new 20-year with a barbell position utilizing highly liquid cheapest-to-deliver bonds into the current classic bond contract (February 2036 maturity) and the current Ultra T-bond contract (November 2045 maturity). Based upon current valuations, we estimate this barbell position would carry a yield advantage over the new issue 20-year of about 10 basis points and a convexity advantage of about 10%-15%. Of course this new 20-year issue could minimize curve risk for those targeting a pure 20-year key rate, but we believe that many investors—especially those with longer-term horizons—may not be immediately willing to give up so much yield and convexity in order to cover this curve risk, which they have managed around for many years.
To be sure, there are some idiosyncratic factors that contribute to the relative cheapness of the current outstanding 20-year paper that this new issue will not be subject to. The Federal Reserve currently owns nearly 70% (its self-imposed maximum position in any single issue) of all outstanding Treasury securities maturing between 2036 and 2042. These securities trade “cheap” as they are not likely to be included in any future Fed purchase operations. Conversely, should the Fed continue to conduct large-scale purchases of Treasury securities into the second half of this year, it may heavily target this new 20-year series out of necessity. So, the market appears to be assigning some “QE premium” to this new 20-year series. In our opinion, this is a short-term consideration that is subject to a good deal of uncertainty. We don’t believe this is the sort of consideration that will entice the longer-term investors, who we believe are critical to supporting this issue.
On net, we believe that at current valuations, this new issue 20-year stands out as quite rich relative to surrounding issues and the broader curve. Of course it is not at all unprecedented for highly liquid points to trade relatively rich, especially in the longer end of the Treasury curve. Classic bond contracts have provided a prime example of this in recent history, with the yield on the deliverable bonds often trading deeply below the yield on a barbell constructed form liquid points on the “wings” of the curve. But bond contracts have truly been an oasis in our view. They have served as the only truly liquid instrument between the 10-year and 25-year points on the Treasury curve, a broad and very volatile segment of the curve. While this new issue 20-year fills an important gap between the 15-year and 25-year points on the curve, we believe there is far less risk embedded in the term structure within this curve segment. Therefore, we believe the new issue 20-year should not be expected to carry anywhere near the premium that the bond contract has enjoyed.
We believe this issue will ultimately play an important role in fixed income pension portfolios over the longer term given its usefulness in plugging the elusive 20-year key rate of plan liabilities. However, we think it wise to wait for cheaper valuations before wholesale adoption into portfolios. We will be watching closely over the coming weeks as the market digests the new issue and pensions decide how to utilize it in liability-hedging portfolios.
This paper is provided for informational purposes only and should not be construed as investment advice. Opinions or forecasts contained herein reflect the subjective judgments and assumptions of the authors only and do not necessarily reflect the views of Loomis, Sayles & Company, L.P. Other industry analysts and investment personnel may have different views and opinions. Investment recommendations may be inconsistent with these opinions. There is no assurance that developments will transpire as forecasted, and actual results will be different. Data and analysis does not represent the actual or expected future performance of any investment product. We believe the information, including that obtained from outside sources, to be correct, but we cannot guarantee its accuracy. The information is subject to change at any time without notice.
This is not an offer of, or a solicitation of an offer for, any investment strategy or product. Any investment that has the possibility for profits also has the possibility of losses.
Commodity, interest and derivative trading involves substantial risk of loss.
Market conditions are extremely fluid and change frequently.
Past performance is no guarantee of, and not necessarily indicative of, future results.
LS Loomis | Sayles is a trademark of Loomis, Sayles & Company, L.P. registered in the US Patent and Trademark Office.