Mnuchin's Ultra-Long Bonds Threaten Stripping Business at BanksBy
A 50-year obligation would rival 30-year zero-coupon bonds
Strips’ bid-offer spread is wider than that on whole bonds
As Steven Mnuchin’s U.S. Treasury studies the merits of bonds maturing in 50 years or longer, there’s one group on Wall Street that’s hoping officials will close the books on the idea.
Ultra-long debt could very well save money for taxpayers in the long run. But it threatens dealers who run the $250 billion market for zero-coupon Treasuries, known as Strips, said five current and former dealers, who requested anonymity either because they’re not authorized to speak publicly, or because they didn’t want to be quoted discussing their competition.
Traders who specialize in breaking Treasury bonds into their component cash flows to sell individually have a relatively profitable niche in a business that has seen its margins squeezed by automation and alternative sources of liquidity. Dealers are the sole source of the securities, which are favored by pension funds and insurance companies.
That business could be jeopardized if the Treasury moves forward with issuing a bond maturing in more than 30 years, the dealers say. It boils down to liquidity: Buyers of Strips -- the acronym for separate trading of registered interest and principal of securities -- could switch to 50-year whole bonds, since they’d offer similar duration while proving cheaper to trade.
“Any time Treasury issues a new product, it’s competing with itself,” said Lou Crandall, chief economist at Wrightson ICAP, a research firm that focuses on Treasury finance. “A 50-year bond would be in direct competition with 30-year zeros.”
While historically low yields have curbed demand for zero-coupon bonds, the Strips market has swelled to a record size nonetheless as the overall Treasury market has grown. Bank of America Corp., Barclays Plc and Goldman Sachs Group Inc. were most frequently cited as trading partners for Strips by institutional investors surveyed this year by Greenwich Associates, said James Borger, managing director at the financial-services research firm.
Representatives of Bank of America and Goldman Sachs declined to comment. The view at Barclays is that 50-year bonds won’t greatly curb demand for long-term Strips.
Since his nomination as Treasury secretary, Mnuchin has expressed interest in adding an ultra-long security to the government’s offerings, which range from four-week bills to 30-year bonds, citing interest rates that remain near generational lows. Wall Street has resisted thus far: Treasury’s advisory panel of dealers and investors said in May that it doesn’t see evidence of strong or sustainable demand for such a security from foreign investors, U.S. pension funds or insurers.
To some market veterans, that conclusion sounds like dealers talking their book. They say an ultra-long Treasury bond could attract sufficient demand and achieve the department’s stated goal of obtaining the lowest funding cost.
An ultra-long Treasury “provides an alternative to a broken U.S. zero coupon market,” said Hirak Biswas, head of product and strategy at LiquidityEdge, a trading platform for Treasuries. Biswas was a long-end trader from 2001 to 2015 at primary dealers including Morgan Stanley and Credit Suisse, where he ran the Treasury desk.
Bid-offer spreads are 1 to 2 basis points for principal Strips and even wider for coupon-only securities, Biswas said. For the potential ultra-long whole bond, the spread should be less than a basis point, based on how 30-year Treasuries trade.
“You’d get your duration needs from something with significantly tighter pricing than the Strips market provides,” Biswas said.
One of the alternatives proposed by the Treasury advisory panel was issuing 50-year zero-coupon bonds. The other suggestion: revive the 20-year maturity, which dealers could strip and reconstitute since it falls within the current yield curve. The Treasury may provide an update at its next quarterly refunding announcement on Wednesday.
Adding a 50-year zero-coupon bond to the product line would allow Treasury to say it’s selling ultra-long debt, but it wouldn’t do much for the government’s cost of funding because principal-only securities are sold at a discount to face value, meaning they don’t raise much cash up front. A 50-year whole bond, on the other hand, would function like the department’s current offerings.
Barclays, one of the main trading partners in the Greenwich Associates survey, says a 50-year whole bond would likely siphon some demand for 30-year Strips, but isn’t worried about it.
“We don’t see long-end issuance as a zero-sum game for the U.S. Treasury,” said Chris Conetta, head of U.S. rates cash trading. “There’s inherent demand for Strips in that 20- to 30-year point, and that’s not going to change if the Treasury chooses to issue ultra-longs.”
It’s also not a sure thing that ultra-long bonds would trade at a tighter bid-offer spread than the current most-liquid zero-coupon securities, said Ian Dai, who heads Strips trading at Barclays.
Bracing for Change
For now, ultra-long bonds remain a cloud on the horizon. And, even if they’re issued, it’ll likely take a few rounds of sales to drum up sufficient liquidity.
A Treasury spokesman didn’t reply to an emailed request for comment on the potential impact on bond dealers from any decision to add an ultra-long maturity.
Yet with deliberations ongoing, dealers are bracing themselves. They’ve seen this movie before: The Treasury has changed up its slate of offerings more than a dozen times since 2000, discontinuing and reintroducing curve points amid perceived changes in investor appetite.
“When you create a market for new securities it brings in demand,” said John Fath, a Strips dealer at UBS from 1997 to 2008. “There’s no question there will be swapping from 30s into 50s. It’s a bona fide reason for concern.”
— With assistance by Saleha Mohsin