Ultra-Long U.S. Treasuries Are an Ultra-Long Shot
(Bloomberg Opinion) -- It didn’t take long for Wall Street to start talking about the prospect of the Biden administration issuing 50- or 100-year bonds. One of the first debt management topics studied by the prior Trump administration in 2017, new Treasury Secretary Janet Yellen stated in her confirmation hearing three weeks ago that she would be pleased to look again at the issue. Although a few years have passed and the fiscal outlook has worsened with significant additional Covid-19 relief spending on the horizon, the answer likely remains the same: ultra-long debt is not in the best interest of U.S. taxpayers.
In the context of 25% interest rates on personal credit cards and the double-digit rates of inflation seen in the 1970s, many believe it ludicrous that the government would not borrow money at historically low, single-digit interest rates for a length of time that few of us will ever outlive. Rather than incompetence, the Treasury’s decision to not issue ultra-long bonds reflects both the savvy of its debt management office and the sound advice of the Treasury Borrowing Advisory Committee – a group of senior financial market experts who serve their country by advising the Treasury Department.
The primary goal for the Treasury’s debt management office is to fund the government at the lowest expected cost over time. With $27.7 trillion of debt outstanding , annual interest expenses of about $375 billion and trillions of dollars in additional fiscal stimulus still needed to combat the ramifications of a global pandemic, it is critical that the Treasury not waiver from its main mandate.
The interest rates demanded for longer-dated bonds are almost always higher than those of their shorter-dated equivalents, for obvious reasons. Yields on 30-year bonds are currently about 1.82 percentage points more than those on two-year notes. That difference, or spread, is above the median of 1.22 percentage points since 1980. The spread between 10-year notes and 30-year bonds is about 0.79 percentage point, above the median of 0.35 percentage point over the long term. As such, in today’s environment it is still more cost effective for the Treasury to issue and refinance shorter-dated debt than to issue longer-dated debt.
Nevertheless, with longer-term interest rates near historic lows on an absolute basis, many, including former Treasury Secretary Robert Rubin, suggest the Treasury should act more opportunistically. Decisions to issue - or stop issuing - certain securities based upon a political appointee’s assessment of the level of interest rates is certain to open a pandora’s box. If that were to happen, the significant liquidity premium earned over the years through the Treasury’s “regular and predictable” approach to debt management would quickly evaporate, leading to higher relative borrowing costs across all of the Treasury’s debt issuances.
What of the risk for inflation to eventually drive short-term interest rates higher? Would the government then have surely wished that it had locked in the current low level of rates? If the Treasury’s debt managers were confident that a significant acceleration in inflation was coming, then issuing ultra-long bonds could make sense. However, the very fact that interest rates are so low largely reflects the broader market’s belief that it won’t happen.
Markets can be wrong, and investors make — and lose - tremendous sums of money trying to divine the message being sent by the markets. The Treasury’s debt managers are different in that they are stewards of both taxpayer money and the risk-free benchmark upon which nearly all financial assets are priced; it would define irresponsibility for them to gamble on where interest rates and inflation may be in the future.
Reducing the volume of debt coming due each week and the risk that an insufficient number of investors fail to show up to the Treasury’s debt auctions is the final argument put forth by many advocates for ultra-long bonds. Indeed, large-scale cyberattacks by foreign nations on the U.S. government’s infrastructure have become alarmingly frequent. What if the government couldn’t hold a normally scheduled debt auction for an extended period of time?
Refinancing or rollover risk is better mitigated by holding a prudent level of cash than by extending the maturity of the debt. With an average of about $350 billion coming due each week, the Treasury would need to issue an astronomical volume of ultra-long bonds to meaningfully reduce this risk. And, as highlighted by the Treasury Borrowing Advisory Committee , investor demand for ultra-long debt is much too limited and uncertain.
All this is not to say that there are no legitimate reasons to issue 50- or 100-year bonds. Ultra-long issuance would support both price discovery and liquidity in the corporate debt market. By providing this public good, the Treasury would support greater issuance of longer-dated corporate debt, thereby improving the capital structure and financial stability of the private sector. Pension plans and insurance companies would also benefit, with a larger investable universe to hedge their long-duration liabilities. These benefits, though, currently remain amorphous and small relative to the additional expense of issuing ultra-long debt.
The average over the last six months of 2020 was $367 billion.
This would meaningfully change should regulators require pension/life companies to more fully hedge their liability, as is required in the U.K. In this case, one could make a very strong argument that the Treasury would need to issue ultra-long bonds to meet the regulatory-created demand. As there is little to no talk of this happening,I didn’t think it worthwhile to raise beyond saying “currently.”
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
James Clark was the U.S. Treasury Department’s deputy assistant secretary for federal finance during the Obama administration.
©2021 Bloomberg L.P.