Stock markets shrugged off geopolitical concerns to make progress in April, with the MSCI World Index rising 1.3%. Concerns over the direction of oil prices weighed on the S&P/TSX index, keeping it to a 0.25% gain. However, the US 10-year bond yield ended the period at 2.28%, down almost 11bps.
Relative to some of the noises recently emanating from the Trump administration, one slightly mundane topic currently being discussed is the potential for the US Treasury to start issuing “ultra-long” bonds of 50 to potentially 100 years maturity, significantly longer than the current 30 year maximum. What are the practical considerations?
On the face of it, the appeal is obvious. Government borrowing costs remain very low, and extending the term of borrowings would lock those rates in for a longer time period, reducing the risk of having to refinance maturing bonds in a less appealing interest rate environment. Several European governments have already dipped a toe into this market for that reason. The US government has expressed a desire to make substantial investment in the nation’s infrastructure, long term assets that should ideally be financed by long term borrowings. In addition, the corporate sector, which has itself been borrowing over longer and longer time frames for the same reasons, would presumably benefit from a government security which would offer a reference interest rate for pricing their own debt. Finally, investors looking to hedge longer term liabilities could well appreciate the availability of an additional tool to enable them to do so more effectively or at lower cost.
The US Treasury has looked at this topic before, most recently in 2014, and passed on it as they were uncertain of the net benefits. Their approach to issuance is shaped by three underlying principles: getting the lowest cost of borrowing over time, maintaining a predictable schedule of debt issuance, and supporting the depth and liquidity of capital markets.
On the first principle, the argument rests on whether a new longer term bond will address an unmet need from investors (most likely pension funds with longer term liabilities) such that they will be prepared to pay a competitive interest rate relative to the current 30-year bond. The potential for cannibalizing demand for 30 year bonds would certainly need to be considered. Second, given the outlook for government borrowing over the long term, it is unlikely that this new instrument will affect their ability to maintain a schedule for issuing debt. The third consideration, similar to considerations on funding costs, means the Treasury would have to be conscious of the overall level of demand to avoid swamping the market, and overall manage supply and demand while avoiding negative impacts on demand elsewhere. The impact on dealer balance sheets could also be a consideration.
This may be a bit of a prosaic topic, but if implemented, an ultra-long term bond would offer both a new tool for managing government finances, which should be welcomed as offering more flexibility. The implications for some alternative asset classes currently used by investors to match long-term liabilities, such as infrastructure (although they also offer some inflation protection) will emerge should the Treasury decide to proceed. As for Cidel, we would welcome the arrival of another investing option to potentially help meet our clients’ goals.