Janus Henderson: Treasury’s ultra-long bonds have no natural demand

US Treasury Secretary Steve Mnuchin reiterated last week that he’s gauging the appetite among investors for ultra-long US government bonds. He may be disappointed.
Although Mnuchin has said that sales of 50-year and 100-year bonds “absolutely make sense,” he told Bloomberg News on 20 June that the US government would only embark on such a programme if it’s not a one-off, but becomes part of the Treasury’s regular debt issuance. The longest maturity currently offered by the Treasury is 30 years.
The idea of locking in low long-term interest rates is obviously appealing to the US government. However, the internal working group formed by the Treasury to study demand for debt with such maturities may end up reporting back that the idea is a non-starter. There’s not much point issuing such debt if no one wants to buy it, and the market is limited in the US.

The natural demand for ultra-long bonds comes from buyers who need to hedge ultra-long liabilities, and they generally are defined benefit pension plans. While 96% of assets under management were in defined benefit plans in Japan and 82% in the UK in 2016, the amount of money in such plans, both traditional and hybrid, was only 40% in the US, with the remainder in defined contribution plans, according to Willis Towers Watson’s Global Pension Assets Study 2017. 

The decline in defined benefit in favour of defined contribution looks to be systemic and irreversible. Only 5% of Fortune 500 companies offered the option of a traditional defined benefit plan to new hires in 2015, with the average size of such plans near the bottom 10th percentile by assets, according to Willis Towers Watson. Further, only 33% of defined benefit plans, which includes hybrid and traditional plans, were open in 2015, with 63% frozen or closed, further limiting the market for long-duration assets as the liabilities in the closed or frozen plans have likely shortened substantially.

Trends indicate that proportionally fewer traditional plans versus hybrid plans remain open. If true, this further shortens duration-matching requirements, because hybrid plans pay a lump sum on retirement, creating shorter duration liabilities compared with traditional plans that act as annuities.Because most of their pension assets are in defined benefit plans, ultra-long bonds have a market in the UK and Japan. The UK has £219.4 billion ($276.8 billion) of gilts maturing after 2047, while Japan has 17.1 trillion yen ($153.3 billion) of debt outstanding with maturities of more than 30 years, according to data compiled by Bloomberg. In Australia, where just 13% of pension assets are held in defined benefit plans, there is no sovereign debt with a maturity beyond 2047.

With defined benefit plans in the US small and shrinking, and likely holding mainly legacy assets because most of them are frozen or closed, the natural pool of ultra-long bond buyers is drying up and is unlikely to be replenished. As a result, such long duration bonds are likely to suffer from limited liquidity.

So while the idea of locking in low-coupon borrowing for multiple decades appeals to Mnuchin and his apparatchiks (officials) at the Treasury, he may well find that the response to such an opportunity from US investors is a resounding “Meh”.

First published on Bloomberg View on 26 June 2017: https://www.bloomberg.com/view/articles/2017-06-26/treasury-s-ultra-long-bonds-have-no-natural-demand

Republished with permission.