Buybacks, Market Functioning, and Treasury Predictability

U.S. Treasury market liquidity has received greater attention since March 2020, when the acute economic shock from the COVID pandemic drove large dislocations in the Treasury market, with similar-maturity bonds trading at vastly different yield levels.

The rapid growth in marketable debt in recent years has occurred amid regulations put in place in the wake of the 2008 global financial crisis (GFC) that have constrained primary dealers’ willingness and ability to warehouse securities during periods of stress.

When private intermediation has been insufficient to ensure Treasury markets function, the Federal Reserve has stepped in as a “market maker of last resort.” This means it purchases Treasuries and other assets – a process known as quantitative easing (QE) – in order to stabilize the market, with the view that later it will allow the securities to mature, reducing its balance sheet again (quantitative tightening or QT).

Concerns around moral hazard have reinforced the need for additional measures to ensure a well-functioning Treasury market. And the policy discussion has focused on a set of mutually reinforcing measures, including increasing dealer warehousing capacity through recalibration of leverage requirements, expanding central clearing, broadening market access and trading protocols, strengthening margining and settlement risk management practices, and so on.

The Treasury Department has also taken steps to support well-functioning markets for U.S. debt. Following consultation with the Treasury Borrowing Advisory Committee (TBAC), a group of private sector investors and institutions, Treasury launched its current buyback program in May 2024. Prior to the latest announcement, liquidity-support buybacks have been relatively small compared with the Fed’s QE operations, but they have provided a regular opportunity to sell older, less liquid off-the-run securities. Separate cash-management operations help smooth Treasury’s cash balance and bill issuance.

As a broader set of market participants and policymakers have discussed the implementation of these liquidity enhancement policies, the Federal Reserve, now led by Kevin Warsh, has focused on further reducing its footprint in the Treasury market. In the 16 April 2026 edition of Macro Signposts (“Why the Fed Could Shrink Its Balance Sheet Again (and Markets Might Not Notice)”), we argued that the Fed was laying the groundwork through regulatory changes to further reduce its balance sheet.

All of these important steps toward improving Treasury market liquidity have occurred despite growing questions around who (if not the Fed) would serve as the market maker of last resort.

If last week’s announcement was a signal that Treasury is willing to take on a greater role in ensuring stable market functioning throughout financial market and economic cycles, we believe that is a good thing. However, the questions that followed the announcement were less about Treasury’s commitment to market functioning and more about what Treasury’s actions signaled about its commitment to its long-held regular and predictable debt strategy.

It’s worth recounting some history to understand why “regular and predictable” is so important to Treasury market functioning and debt management.

The country’s fiscal deficit determines how much Treasury must borrow. Debt management determines whether that borrowing occurs through bills, notes, bonds, Treasury Inflation-Protected Securities (TIPS), or floating-rate debt. Because Treasury is the market’s largest borrower, its financing strategy matters. Since the 1970s, Treasury has increasingly organized issuance around a “regular and predictable” framework.

The modern doctrine can be traced to Paul Volcker’s 1972 speech, “A New Look at Treasury Debt Management.” Volcker, then the Treasury Department’s Under Secretary for Monetary Affairs, proposed placing more borrowing on a recurring schedule modeled on the then established bill-auction program.

Volcker’s speech was delivered as the fiscal deficit expanded in the 1970s, and Treasury conducted a series of tactical coupon offerings whose timing and maturity repeatedly surprised investors. A decisive lesson came in March 1975, when (as documented by New York Fed historian Kenneth Garbade¹) Treasury auctioned $1.25 billion of 15-year bonds at the same time a syndicate brought $600 million of AAA-rated corporate debt from a well-known U.S. company to market. The simultaneous offerings left the bond market in “disarray,” according to media reporting at the time.

By the early 1980s, Treasury had largely replaced tactical financing with recurring auctions of standardized maturities, reducing what had been elevated yield volatility on Treasury financing days.

Since the 1980s, according to Treasury debt-management framework updates, regular and predictable issuance has reduced the cost to Treasury to fund the government deficit through four main channels:

• Less supply uncertainty: Investors can reserve cash and dealers can manage inventories and hedges ahead of auctions.
• More liquid benchmarks: Repeated issuance and reopening concentrate trading, improve price discovery, and support hedging.
• A broader investor base: Investors can build recurring allocation processes around known supply.
• Less reliance on market timing: Treasury is too large and borrows too continuously to act opportunistically without affecting the market it is theoretically trying to time.

In a 2015 debt-management framework update,² Treasury reiterated that it was and is “too large to behave opportunistically,” that it is “not a market timer,” and that it does not react to current rate levels or short-term demand fluctuations. Instead, it seeks the lowest expected borrowing cost over time while managing risk and preserving market liquidity.

While regular and predictable is not the only driver of long-end term premiums, moving away from this strategy would likely come with higher issuance uncertainty and Treasury market volatility, and would tend to increase term premiums over time. It’s a strategy that directly goes against Treasury’s debt cost minimization goals.

The lessons of the 1970s argue against opportunistic issuance, while the lessons of the 2020 COVID period argue in favor of the importance of an official sector liquidity backstop. However, the question remains whether Treasury has the capacity to credibly provide this function.

Unlike the Fed, which can fund unlimited purchases of Treasury securities by creating reserves, Treasury buybacks are financed through additional bond issuance. In other words, Treasury can’t change total debt issuance, but it can change the mix of maturities and instruments it uses in order to provide liquidity in longer-dated maturities in times of stress. As a result, Treasury is constrained in its ability to finance long-dated issuance by its ability to issue T-bills, and it faces the risks of higher expected future funding variability associated with greater reliance on short-dated bills for broader financing purposes. (See the November 2025 TBAC discussion³ on optimal debt structure.) However, in practice, Treasury likely has some latitude to provide support to the market.

Currently, T-bills account for around 22% of total debt outstanding, according to Treasury reporting – a similar level to what prevailed prior to the GFC. If Treasury wanted to increase the T-bill share to 24% – which would leave it at the higher end of the pre-GFC range – it could buy $630 billion in the 10- to 30-year sector. While this isn’t unlimited, it does rival the size of Federal Reserve purchases of Treasuries during past QE programs. Treasury’s preference to purchase only securities that are offered cheap to the market (i.e., at yields above current market rates) could also limit the size of purchases while providing flexibility for increasing purchase amounts if market conditions warrant. This would both reduce the perceived stigma associated with the facility while limiting its size in normal times.

In the end, if Treasury market dysfunction were so severe that it became a systemic problem, we believe the Fed would work with Treasury to support the Treasury market, even if the Fed’s hurdle to enter the market is increasing.

Treasury buybacks are a useful tool in normal times to improve liquidity in off-the-run securities, while potentially acting as a backstop to market liquidity in periods of stress. While Treasury’s ability to act as the market maker of last resort has limits (unlike at the Fed), we do think a stable backstop liquidity facility – along with continued implementation of a host of other policies – can support the Treasury market’s perceived “safe haven” status. Treasury cannot control the broader fundamental factors that drive term premiums, but these policies should help reduce Treasury’s funding costs over time. The spread between Treasury yields and similar-maturity swaps suggests there is room to improve Treasury funding levels.

In terms of what last week’s announcement says about Treasury’s commitment to the “regular and predictable” framework, the market backstop could increase volatility in T-bill issuance, but we doubt Treasury is moving away from its policy of clearly telegraphing auction schedules further out the curve. As Treasury itself has said on multiple occasions, it is too big to time a market that it relies on for U.S. government funding. A regular, predictable, yet flexible framework that is well-understood by markets should serve everyone well.