September 24, 2019
Brian Rehling, CFA , Co-Head of Global Fixed Income Strategy
Charlotte Woodhams, Investment Strategy Analyst
Next-Gen Bonds, Next-Gen Networks
Will the U.S. Treasury issue an ultra-long bond?
The U.S. Treasury Department is again considering ultra-long-maturity bonds. While such issuance could lead to positive headlines, we do not believe that it would bring in substantial new pockets of liquidity for the U.S. Treasury.
- While the U.S. Treasury could issue a new ultra-long-maturity bond, we expect it to have minimal impact on funding the almost $16 trillion in outstanding marketable Treasury securities.
- In our view, a new ultra-long bond issuance would have little appeal to individual investors.
Ultra-long sound bites
Thirty year U.S. Treasury securities recently touched their lowest yield on record in late August (1.95%). With low long-term interest rates in the U.S., the Treasury Department is once again considering the issuance of an ultra-long-term bond. The issuance of ultra-long-dated debt would potentially tap into demand from longer-duration investors—while also locking in federal financing costs for generations to come. Previous feedback from the Treasury Borrowing Advisory Committee suggests that market participants have not provided “evidence of strong or sustainable demand for maturities beyond 30 years.”2 We do not believe that the Advisory Committee has changed its mind, but the political benefits of locking in low rates cannot be completely discounted.
Not the first time—and not the only ones
A number of countries around the world have experimented with issuing ultra-long-term debt. Mexico, Ireland, Canada, and France (among others) have issued ultra-long bonds. Even the U.S. issued a 50-year bond to help fund construction of the Panama Canal.
There is little doubt that the U.S. Treasury could issue 50-year or even 100-year debt. Yet, we are equally confident that an ultra-long bond could not be issued in a volume that would make even a dent in the almost $16 trillion in outstanding marketable Treasury debt. Ultra-long debt issuance really has not caught on and become a liquid market in any magnitude in these countries (or others). We are quite confident that if the U.S. decides to test the ultra-long-bond market, it will face a similar result.
Good public relations—but unlikely to solve the coming flood of Treasury debt
On the surface, locking in these “low rates” for multiple generations would appear to be a wise endeavor for a government to undertake. Without question, the sound bites around such an action would attract strong media attention. The U.S. is facing a meaningful increase in debt in the coming decade as Baby Boomers retire and begin to draw on Social Security and Medicare (for a more complete discussion of debt and deficits, please ask your investment professional for a copy of our report series titled “Paying America’s Bills”3).
While there is real demand for portfolio duration among a subset of investors today (mostly in the institutional pension and insurance space), there is risk that an ultra-long bond would not attract new sources of liquidity but would simply borrow from today’s 30-year Treasury market liquidity.
Locking in rates is not as attractive as it may appear (even at nominal cost)
We estimate that an ultra-long Treasury bond would require no more than 25 to 50 basis points in yield over a 30-year Treasury bond yield.4 Such a small increase seems to be a reasonable price to pay for 20-70 more years of rate protection for the Treasury. While the financing costs likely are nominal, these bonds would increase interest expense for the U.S. government. This is an expense that essentially would have no significant value to taxpayers for the next 30 years. In the near term, those extra costs would make federal budget expenses higher and increase the deficit.
If rates were meaningfully higher 30 years from now, this issuance could appear to be a wise step. It also is possible that rates will be meaningfully lower than today’s rates in 30 years. To illustrate this challenge, the U.S. Treasury issued a 30-year bond with a coupon of 7.5% in 1994—almost half of what the Treasury paid in 1980. Yet, it turned out that the low rates of the mid-1990s were rather expensive and hardly a good deal for taxpayers. In our opinion, the Treasury should not issue ultra-long bonds as an “interest-rate play.” Rather, we believe that new bond issuance should look to tap new pockets of liquidity for the Treasury.
If the Treasury was truly interested in locking in current financing rates, a more prudent route would be to extend the average maturity of outstanding Treasury debt. This could be done by issuing more 10- and 30-year debt. The Treasury could also consider the reintroduction of 20-year bonds.
In our opinion, ultra-long Treasury bonds would have little appeal for individual investors. Few individual investors have investment horizons that extend 50 years or more. Life insurance and pension funds would be the most likely buyers of ultra-long Treasury debt. Yet, we still see the Treasury market benefits as minimal.
One of the sticking points for long-duration investors is that a 50-year or even a 100-year bond would not drastically alter the duration profile of the 30-year Treasury bond issue. The duration of a 30-year Treasury bond today is around 22.5 years. A 50-year maturity would extend the duration to 30.5 years, and we believe that a 100-year bond would have duration of about 37.5 years. While an ultra-long bond could be a tool for investors to lengthen duration, we do not believe that there is enough differentiation between these investments to bring in a substantial new source of Treasury liquidity and demand.
Ultra-long Treasury issuance may offer good media or political sound bites. Yet, we believe that such issuance would have little impact on Treasury financing. If ultra-long Treasury bonds were to be issued, we expect that it would have an insignificant impact on overall Treasury market demand.