The idea of a U.S. Treasury buyback program has major drawbacks for the U.S. Treasury, end-investors, and even the Federal Reserve. Indeed, buying back off-the-run Treasuries (OFTRs) to create jumbo on-the-run (OTR) benchmarks will inadvertently lead to further deterioration in the OFTRs, the sector of the market that accounts for 98% of all Treasury issuance.
The root of the issue with Treasury buybacks lies in the bifurcation of the market, with trading of OTRs dominated by principal trading firms (PTFs). Helping this constituency by bolstering the size of benchmark issues does not add to the stability of the overall market. In contrast, genuine long-term holders of the government’s debt are concentrated in the OFTRs. While these real money investors may use OTRs to adjust duration (or when they need immediate access to liquidity), portfolio re-balancing is often done in cheaper OFTRs.
Moreover, if OTRs are financed at negative rates, they become rich relative to everything else. This is a positive for the Treasury, because new debt issuance is completed at a lower cost to the taxpayer.
In October 2007, Ken Garbade and Matt Rutherford did an outstanding job of examining buybacks in a staff report titled, “Buybacks in Treasury Cash and Debt Management.” They noted that, “Treasury officials did not undertake any systematic ex post assessment of the costs and benefits of the buyback program,” and confirmed that debt management buybacks only change the form of Treasury indebtedness from OFTR to OTR debt, keeping the total indebtedness constant. But even more compelling was their statement, “there is … no reliable method for identifying the marginal consequences of additional supply for the size of the new issue premium and thus no reliable method for identifying the maximum appropriate scale of a strategic buyback program.”
Today, neither academics, policy-makers or regulators have access to a reasonable data set to apply to the OFTR market. This should not prevent them from understanding the gravity of the liquidity shift that is taking place. Further, sound bites and claims that buybacks lower borrowing costs should not be adopted as fact without in-depth exploration.
Buy-backs were first floated in 1998 after a period of budget surpluses to help the Treasury manage its debt. As the discussion continued into 1999, I was then Chair of the Government and Federal Agency Division of what became SIFMA. It was my signature on the letter to the Director of the Bureau of Public Debt at the Treasury on behalf of the dealer’s association that favored buybacks. Two years after the initial proposal, buybacks were launched and continued through 2002.
At the time of the program, total outstanding Treasury issuance was slightly over $4 trillion compared to over $13 trillion today. Billion dollar trades in aged securities (i.e., OFTRs) were commonplace. Over the course of the two-year buyback program, about 14% of the total amount of bonds publicly held at the end of 1999 was bought back. The program was tactical, in an era when dealers had seemingly limitless access to their balance sheets.
Today it is a very different marketplace. The United States is financing a significant deficit, underwritten in large part by the silent majority -- real money investors. Total outstanding issuance has tripled. OFTR liquidity is fractured. Indeed, where once billion dollar trades in OFTRs were common, today it is a strain to execute $150 million, and roughly 40% of Treasury securities with maturities of 10 years or greater are now held in the Federal Reserve’s internal SOMA portfolio, according to a 2015 report prepared by its markets group for the FOMC.
It’s now been two years since the Treasury asked the Treasury Borrowing Advisory Committee (TBAC) to explore the idea of a buyback program. Soon after, the Treasury released the TBAC minutes, seemingly as a ‘trial balloon’ to gauge reaction from the primary dealer community. TBAC argued at the time that buybacks were a good option to smooth peaks, manage cash balances, and capitalize on relative value. Since the OFTR liquidity problems are more severe today than in 2014, it’s quite possible that different conclusions would be reached had the analysis been conducted today.
In recent years, regulators are seen to have acted swiftly and in the best interest of the marketplace. However, both the Volker Rule and Basel III were put together with haste and with too many ‘cooks in the kitchen’. This led to mistakes, one of which was to cripple market liquidity. Continuing down this current path could trigger the exact result they meant to prevent: the next ‘Black Swan’ crisis.
For example, exempting the U.S. Treasury market from the Volker Rule without exempting Treasuries from Leverage ratio calculations is tantamount to giving someone a boat in the desert. Market makers cannot take risk without balance sheet. As a result, poor liquidity has left OFTRs in a state of dysfunction, even though they represent the vast majority of all issues outstanding. This has a direct negative impact for the holders of those securities: large mutual funds, pension funds, bank portfolios, and foreign central banks.
In a recent speech to the Mortgage Bankers Association, Nate Wuerffel, SVP and Director of Treasury Markets, described market liquidity as the cost of quickly converting a desired quantity of an asset into cash (or cash into an asset) at an efficient price. His definition incorporates four components. What is the transaction cost of executing a trade? How quickly can you get in or out of a position? What quantity can be bought or sold at a given price? And finally, efficiency. Are trades conducted at efficient prices?
Regardless of the measure, if securities that trade more frequently are associated with a greater degree of liquidity, then it makes sense that academics, regulators and policy makers alike should shift their focus to improving liquidity in OFTRs, and not be skewed by the volume or revenue generated by the 2% of issuance dominated by professional traders.
Is there another positive we are overlooking? Lou Crandall said, “dealers would be in a better position to provide liquidity to customers looking to sell off-the-run if there was a known buy-back schedule that would allow them to unload the position.” In February of this year, dealer inventories “ballooned” to $121 billion, but this only represents less than one-tenth of one percent of total outstanding issuance. During this time bid-offer spreads in OFTRs widened, because trading volume in these issues dried up long before the selling began.
The only way to tighten spreads and accommodate a large volume of transactions is to add balance sheet in combination with technology that allows disparate sources of liquidity to find one another. The constraints impairing the U.S. Treasury market must be loosened enough to allow relative value arbitrageurs to reenter the market and act as a line of defense; and loosened enough to allow dealers to increase their principal risk-taking activities in the U.S. Treasury market and encourage new platforms that allow the dealer community to take principal risk and act as agent. Do not assume PTFs will step in and magically fix the problem. Many who were active in sub-prime walked away during the collapse.
Ultimately, the responsibility for the smooth functioning of the marketplace is the Treasury’s responsibility, whose mission includes “protecting the integrity of the financial system.” This requires facilitating a fully functioning yield curve, not just for the 2% that is traded most actively, but also for the ‘forgotten 98%’ that is held in the hands of investors whose investment time horizon can fluctuate. Regularly scheduled trading auctions would be a good start.
Primary dealers, while continuing to take principal risk, should push for adoption of trading technology that can leverage dealer’s balance sheets on an agent-basis; i.e., connect disparate sources of liquidity that foster an ecosystem where both exist side by side.
Finally, the Fed should be encouraged to open up the SOMA portfolio in order to re-liquify the curve. The size of the OTRs do not matter because the market allows for negative financing rates. If owning in excess of 40% of Treasury securities with maturities of 10 years and greater has added to market dysfunction, the mistake should be corrected.
In an effort to curb a potential catastrophe, policymakers have historically adopted new approaches to restore integrity to the markets. Therefore, it is not unreasonable to suggest they get in front of the one materializing before our very eyes.
Treasury buybacks? Not so fast.
Susan Estes is CEO of OpenDoor Trading, LLC, a provider of an all-to-all marketplace for off-the-run US Treasuries. During her career, she has managed fixed income trading at several major financial institutions including Deutsche Bank, Morgan Stanley and Countrywide Securities. She was also a member of the Treasury Borrowing Advisory Committee under Federal Reserve Chairmen Greenspan and Bernanke.