The U.S. Treasury market, while considered the deepest and most liquid in the world, has proved vulnerable to serious disruptions in recent years, epitomized by the turmoil in March 2020, when intense selling pressure overwhelmed securities dealers’ balance sheets, causing volatility to spike, spreads to gape, and liquidity to evaporate. At times it was easier to sell investment grade corporate bonds than older, “off-the-run” Treasury bonds. While policymakers have focused on the Treasury market’s problems since March 2020, few substantive changes have been made. Here, we discuss what we believe are the most and least effective measures that policymakers can pursue to help the functioning of this important market.

What are we solving for?

First, it is important to define what policymakers should be solving for. As one of the largest participants in the Treasury market, PIMCO views market liquidity – the ability to buy and sell bonds efficiently, continuously, and economically – as that market’s biggest challenge. While there is ample liquidity when market conditions are good, that liquidity is fickle and disappears quickly under stressed conditions.

In short, the Treasury market functions spectacularly well – until it does not.

Remember that most Treasuries – and other bonds – are not traded on an exchange like shares of stock. Treasuries mostly trade “over the counter,” a transaction negotiated between dealers (i.e., large banks) and end-users (such as asset managers, sovereign wealth funds, or pension funds) – either electronically or, for some segments of the bond market, like off-the-run Treasuries, over the phone. Practically, this means that end-users largely depend on dealers to make markets in Treasury bonds. Usually, dealers are eager to make markets – to buy and sell Treasury bonds to end-users. But in times of stress, such as in March 2020, their willingness and ability often disappears. Put differently, the dealer community still controls the risk-transfer mechanism, particularly in off-the-run Treasury bonds.

Assuming policymakers’ primary goal is to avoid another market seizure, we believe they should focus on reforms that can improve liquidity and evolve the existing Treasury market structure. In that spirit, we advocate policymakers consider 1) broadening access to Federal Reserve (Fed) sponsored programs, such as the standing repurchase (repo) facility and bond buying programs, to inject liquidity into the market more directly and efficiently in times of crisis; 2) increasing dealer balance sheet capacity by tweaking existing bank regulations to allow dealers to make markets in Treasury bonds more readily; and 3) while seemingly quixotic, using the convening authority of policymakers (e.g., the Financial Stability Oversight Council, or FSOC) to help advance “all-to-all” trading – a system in which all market participants are able to trade with each other, bypassing the dealers in some cases. We also discuss efforts that policymakers are considering that we do not believe will bolster liquidity, including: 1) requiring cash clearing of Treasuries, and 2) instituting real-time reporting of Treasury transactions.

What can policymakers do to increase liquidity in the Treasury market?

1. Broaden access to liquidity facilities and Fed bond buying programs.


• Provide broader access to the Fed’s standing repo facility. In July 2021, the Federal Reserve created a permanent standing repo facility (SRF),^Footnote1 which allows primary dealers and certain depository banks to borrow against Treasuries and agency mortgage-backed securities (MBS) when liquidity conditions are stressed. Like other industry participants,^Footnote2 we believe establishing the SRF is not only an important step to mitigate the type of dysfunction experienced in March 2020, but also to increase Treasury market resilience.

Yet we believe the Fed could improve the effectiveness and efficiency of the SRF – and market liquidity and functioning generally – by broadening the types of investors that can access the facility. Indeed, rather than limiting SRF participation to primary dealers and a handful of depository institutions, policymakers should consider including asset managers, sovereign wealth funds, and other large institutions. Doing so is consistent with the Group of Thirty’s 2021 recommendations for improving market structure^Footnote3 as well as those by other experts, including former New York Fed President William Dudley. Dudley has argued that the SRF should open to a broad set of counterparties, and that doing so would address a potential repeat of primary dealers failing to lend the liquidity they receive from the Fed to other market participants that need short-term financing.^Footnote4

Dudley’s reference to the unwillingness (or inability) of primary dealers to pass through Fed liquidity is consistent with what we observed in March 2020. Early in the crisis, the Fed attempted to support the Treasury market by providing over $1 trillion of repo funding to primary dealers. Yet while other market participants struggled to secure funding, primary dealers used less than half of the Fed’s available funding, according to PIMCO estimates based on public data.

We believe any counterparty concerns that may arise from broader SRF access can easily be addressed by requiring commensurate haircuts – potentially depending on the type of institution, its size, and the scope of regulation to which it is subjected. Another guardrail could require clearing SRF transactions – which is already happening to some extent – and would subject the counterparties to requirements of the Fixed Income Clearing Corporation (FICC).^Footnote5 The Fed has widened its funding operations before: It expanded access to its reverse repo operations beyond its typical primary dealers beginning in 2019.^Footnote6

Another potential benefit of broadening SRF access: It may help decrease any potential stigma attached to using the SRF, which could encourage more banks and broker-dealers to sign up for the SRF.^Footnote7



• Allow a broader set of participants to access the Fed’s bond buying programs. Policymakers could also improve Treasury market liquidity, particularly in times of crisis, by expanding the eligible counterparties for the Fed’s bond buying program (known as quantitative easing, or QE). As we have written previously,^Footnote8 allowing large institutional market participants to sell Treasuries to the Fed directly – similar to the way they currently buy Treasuries directly from the government – could provide liquidity to the market directly instead of indirectly. Currently, the Fed buys bonds exclusively from a small list of banks (i.e., the primary dealers), but those bonds are usually sourced from asset managers and other end-users, such as large institutional pension plans and sovereign wealth funds. Allowing the Fed to skip the middleman and buy directly from end-users could enable it to inject liquidity more immediately and enact its policies more effectively.

Additionally, more participants selling Treasuries could lead to more competitive pricing for the Fed. Indeed, in March and April 2020, banks produced windfall trading profits, partly on the back of transacting with the Fed; if the Fed were able to buy from a broader set of participants, we believe it would be able to transact at more competitive levels.



2. Consider modestly recalibrating banking regulation. The benefits of the Dodd-Frank Act reforms on the banking system were clear in March 2020: Banks survived the market turmoil largely unscathed, protected by generally robust balance sheets, with high quality capital and plentiful liquidity. At the same time, however, their risk-taking capacity was significantly constrained even for Treasury bonds, which are backed by the full faith and credit of the U.S. government. As a result, we believe policymakers should evaluate some of the existing Dodd-Frank Act requirements with an eye toward achieving a better balance between market functioning and safety and soundness. We believe revisiting these rules is especially apropos given the doubling in size of the Treasury market since the Dodd-Frank Act was enacted.^Footnote9

At a minimum, as many stakeholders have recommended,^Footnote10 including the Undersecretary of Domestic Finance, Nellie Liang, while at the Brookings Institution,^Footnote11 we would encourage the Fed to make permanent the temporary changes it imposed in March 2020 to the supplementary leverage ratio (SLR): Exclude U.S. Treasury securities and reserves from the SLR calculation. We believe such a change does not compromise the stability and soundness of the banking system, but would allow banks to make markets more easily, especially in times of crisis.



3. Use the convening authority to advance all-to-all trading. Not only would we like to see broader participation in the Fed’s standing repo facility and bond buying programs, but we would like the entire Treasury market to move to all-to-all trading – a platform where asset managers, dealers, and non-bank liquidity providers are able to trade on a level playing field, with equal access to information. This is happening in some pockets of the bond market^Footnote12 and, of course, it’s the way equities are traded. Yet, for the vast majority of the bond market, including most parts of the Treasury market, liquidity remains intermediated, making the market more fragile, less liquid, and more susceptible to shocks. Even in those cases where the market has moved to all-to-all-like trading, it may be in name only: While certain hedge funds and professional trading firms have been allowed onto these platforms, often large asset managers and other institutional participants are excluded.

For this reason, we believe that policymakers could play an important convening role, bringing different stakeholders together to decide on the rules of the road for all-to-all trading. In our view, an effective all-to-all platform for Treasuries would function similarly to a utility and would 1) include all legitimate, professional market participants; 2) require that participants trade under the same rules with the same access to price, information, etc.; and 3) allow for total anonymity of all trades at the time of transaction, similar to the central limit order book (CLOB) rules that dictate the futures market. While it is theoretically possible that the market would shift this way organically, we are skeptical it will happen quickly, if at all, since the market has evolved little in decades. As such, we believe that policymakers, such as the Fed and Treasury (potentially in the context of the FSOC), could use their convening authority to bring different stakeholders together to discuss a framework and potential destination for all-to-all trading, pushing the bond market into the modern age, with deeper liquidity and greater resilience to financial shocks.

What efforts do we think will not help liquidity?

The Treasury Inter-Agency Working Group is also focused on areas that we do not believe would improve liquidity – and may inadvertently hurt it. They include: 1) required Treasury cash clearing and 2) real-time post-trade transparency of Treasuries.

1. Treasury cash clearing would have done little to help the March 2020 episode. One area of policymakers’ focus is the mandatory clearing of all Treasury securities (cash and repurchase agreements) through a Central Counterparty (CCP). While we see benefits to clearing repo, we don’t see those advantages in clearing cash Treasuries, nor do we think it will improve liquidity or prevent another March 2020 episode.

This is because Treasury security transactions do not have meaningful counterparty risk – the main issue central clearing can address. Treasuries typically settle quickly – the day after the trade date. While clearing could help avoid failed on-time settlements, doing so would not help with time-of-trade liquidity or mitigate market meltdowns such as those in March 2020 or September 2019. Additionally, existing accounting treatment for cash Treasury transactions already allows banks to net uncleared trades on their balance sheets, and thus would not necessarily free up balance sheet capacity, as it would for some other segments of the market.^Footnote13 Also, while clearing could theoretically bring new, smaller participants into the Treasury market, it could just as easily discourage participants from entering the market since clearing increases costs, compliance burdens, and operational complexity. While some have indicated that clearing could lead to more all-to-all trading – which is possible – it is not a necessary precursor.

Overall, while clearing Treasuries could improve certain facets of the market, we don’t think requiring it would improve market liquidity.

2. Post-trade reporting of Treasuries. Though theoretically beneficial, not all transparency is created equal and, in practice, it may have unintended consequences. Assuming all-to-all trading isn’t implemented across all Treasuries, we believe applying real-time post-trade reporting requirements to all bonds, including off-the-run Treasuries, would shrink liquidity meaningfully. That is because unlike a pure all-to-all format (such as in the equity market), most bonds are intermediated – bought and sold through a dealer at a negotiated price. Immediately disclosing information about the price and the type of transaction (e.g., dealer to client) allows for other participants to front-run the transacting dealer, decreasing the flexibility the dealer has to offload its risk.

In practice, dealers would simply charge more to make markets to compensate for the risk of front-running. Prices would likely rise, as would transaction costs for end-users, generally leading to less market making and less liquidity – exactly the opposite of the intent.

Conclusion

A well-functioning U.S. Treasury market is critical for global capital markets.  Given the Treasury market’s growth, the current structure leaves it vulnerable in times of stress to further bouts of the extreme price volatility seen in March 2020.  In our view, changes are urgently needed to lessen reliance on primary dealers to make markets, while increasing banks’ capacity to hold Treasuries. Without these changes, we believe Treasury market liquidity will again disappear during bouts of turbulence, ultimately leaving investors and the U.S. government exposed.