The Impact of Rising Debt Levels and Constrained Dealer Capacity on Market Resiliency

• In this two-part blog series, we reexamine long-standing concerns about the capacity and resiliency of the U.S. Treasury market as government bond markets globally contend with rising yields, significant increases in issuance and renewed volatility. We also discuss the wide-ranging impacts that recent regulatory proposals, in particular the Basel III Endgame proposal, could have on these markets.
• In Part I, we examine how the rapid growth in U.S. government debt issuance and constrained dealer balance sheet capacity arising from bank capital requirements have combined to raise questions about the resiliency of the Treasury markets. Serious market disruption events, such as the 2020 “Dash for Cash” episode, as well as the growing risk premia that investors are demanding for holding Treasury securities, appear to support concerns about reduced market capacity and illiquidity during periods of stress.
• Given that Treasury debt issuance seems likely to continue to rise, policymakers have been actively exploring other ways to improve market resiliency. Unfortunately, regulators have not addressed the core problem of constrained dealer balance sheet capacity. In fact, recent regulatory proposals, most notably the Fundamental Review of the Trading Book (“FRTB”) and other trading book capital reforms contained in the Basel III Endgame proposal, would serve to further constrain dealer capacity, raise transaction costs, and potentially disrupt key parts of the market.

Background: Why are the U.S. Treasury markets important and how are they structured?

U.S. Treasuries are debt instruments issued by the U.S. government to finance its activities. Owing to the United States’ creditworthiness and status as the world’s leading economy, the U.S. Treasury market (comprised of the cash market as well as the repurchase agreement (“repo”) and futures markets) has been described as the “biggest, deepest and most essential bond market on the planet,”^[1] a fact that has allowed the U.S. government to finance its needs at a relatively low cost over time. Investors have historically viewed Treasuries as risk-free or near-cash assets i.e., safe haven assets that retain their value and can be easily sold during both normal and stressed market periods. Owing to their stability, U.S. Treasuries also often serve as benchmarks for other fixed-income securities and hedging positions; as a result, U.S. Treasury yields have an impact on the rates that consumers, businesses, and governments across the globe pay to borrow money. Moreover, the U.S. Treasury repo market is a key transmission mechanism for U.S. monetary policy, and is vital to the liquidity of the cash Treasury market. Put simply, the Treasury markets are the bedrock of the global financial system.

The Treasury markets operate through a dealer-based structure, with “primary dealers” – banks and broker-dealers that have been designated as counterparties of the Federal Reserve Bank of New York (“FRBNY”) –acting as the largest buyers in auctions of new Treasury debt and as market-makers or intermediaries in the secondary markets.^[2] In addition to their obligations to participate in all auctions of U.S. government debt, primary dealers are required to “[d]emonstrate a substantial presence as a market-maker that provides two-way liquidity in [the Treasury market], particularly Treasury cash and repo operations.” The obligation to support market liquidity extends not only to “on-the-run” securities (the most recently issued securities) but also to less liquid “off-the-run” securities that trade in the secondary markets.

In meeting these obligations, and in attempting to satisfy market and client demands, primary dealers are frequently required to commit a significant amount of capital. Principal trading activity in the “when-issued” market (i.e., securities that have been announced but have yet to be issued), during auctions, in the aftermarket of auctions, and in the secondary market requires these dealers to hedge their positions with other Treasury-backed products. By contrast, other market participants are not similarly bound by the market-making obligations that put primary dealers in the position of providing both buy and sell quotes on a more-or-less continuous basis. The ability of primary dealers to engage in this market-making activity supports the liquidity and overall functioning of all segments of the Treasury markets.

Why are there growing concerns about the capacity and resiliency of the Treasury markets?

As we discussed in a prior SIFMA blog, the resiliency of the Treasury markets has been called into question by a series of major market disruption events over the past decade, most notably the March-April 2020 “Dash for Cash” episode, in which market price volatility increased dramatically and/or where the depth of the market (i.e., the amount of liquidity available) decreased precipitously. While this type of disruption has not occurred since, the Treasury markets have continued to witness episodes of high volatility and trading volumes. As the Inter-Agency Working Group for Treasury Market Surveillance (“IAWG”) noted, Treasury market “volatility reached levels not seen since the 2008 Global Financial Crisis” in mid-March 2023 as difficulties in the regional bank sector led to a repricing of term risk-free rates, while daily, dealer-to-customer, and interdealer volumes all neared historic highs.^[3]

These market events and trends highlight two structural problems in U.S. Treasury markets: the growth of Treasury issuance as the U.S. government continues to borrow more, and the constrained ability of bank dealers to intermediate these markets. These problems are contributing to growing volatility and episodes of illiquidity, increasing costs for investors, and ultimately potentially contributing to higher borrowing costs for the U.S. government over the longer term. Although policymakers have taken some actions and issued proposals designed to improve market resiliency, these twin dynamics of ever-growing supply and constrained intermediation capacity are only likely to get worse in the coming years, particularly if recent banking agency proposals implementing the Basel III Endgame capital reforms, as well as changes to the way the Global Systemically Important Bank (“GSIB”) Surcharge is calculated, go into effect. These potential impacts will be discussed in Part II of this blog.

Growth of U.S. Government Debt

The dramatic growth in U.S. government debt over the past 15 years has represented a major structural change to the Treasury markets. Outstanding marketable Treasury securities stood at $25 trillion as of August 2023, which represents a 743% increase since 2000 (see Figure 1 below). Moreover, as federal budget deficits continue to rise, the average rate of Treasury borrowing is accelerating; excluding 2020 (when pandemic-related programs dramatically pushed up borrowing levels), the 2021/2022 average was 56% higher than it had been during the prior decade ($1.5tn versus $1tn in the 2010-2019 period). In 2023, borrowing levels are expected to be 200% greater than the 2010-2019 average, with the amount of new marketable Treasury debt likely to be issued exceeding $3tn.^[4] These trends are likely to continue, with the U.S. Congressional Budget Office (“CBO”) estimating that the total amount of Treasury securities outstanding will rise from 98% of U.S. GDP in 2023 to 107% in 2029.^[5] In sum, the Treasury securities market has grown massively in a relatively short period of time and is likely to continue growing rapidly over the coming years.

Figure 1: Treasury Securities Outstanding ($TN), 2000-2023

Source: SIFMA Research

Constraints on Dealer Balance Sheets

This growth in the size of the Treasury markets has coincided with a second structural change: dealer balance sheets have become increasingly constrained since the enactment of capital reforms in the wake of the Great Financial Crisis (“GFC”) of 2007-2008. The constraints on bank dealers’ ability to intermediate in these markets are a function of three post-GFC reforms.

First has been the implementation of the Supplementary Leverage Ratio (“SLR”) and the “enhanced SLR” (“eSLR”) for U.S. GSIBs. The SLR and eSLR create economic disincentives for bank dealers to hold low-returning assets such as U.S. Treasuries by assigning them an equal risk-weighting with far riskier and higher-returning assets (in contrast to risk-based capital requirements, which assign U.S. Treasuries a zero risk-weight).^[6] This difference in treatment can turn these leverage measures into binding constraints for some large dealer banks, placing constraints on bank dealers’ capacity to intermediate in the Treasury markets. Indeed, commentators with diverse perspectives on capital regulation agree that the SLR and eSLR have placed constraints on the ability of bank dealers to perform market-making activity in the Treasury markets, particularly during periods of stress.^[7] Recognizing this, the U.S. banking agencies decided to temporarily exclude Treasuries (and central bank reserves) from the SLR and eSLR calculation as part of their response to the pandemic-induced stresses witnessed in the March-April 2020 Dash-for-Cash episode.

The GSIB Surcharge, specifically the Federal Reserve’s preferred methodology for calculating the Surcharge, known as “Method 2” (as opposed to the “Method 1” approach under the internationally agreed Basel framework), has likely acted as a second constraint. Under this approach, Treasuries and Treasury repos are captured by multiple systemic indicator scores used to calculate the Surcharge^[8], which can lead to double or triple counting of Treasury assets and liabilities – a feature of the Surcharge that former Federal Reserve Governor Daniel Tarullo has acknowledged exists (though he contends that it is an intentional element rather than a “bug” of the metric).^[9] Moreover, the coefficients used to calculate the Method 2 Surcharge are set at a 2012-2013 baseline i.e., they are not adjusted for economic growth.^[10] Banks subject to the GSIB Surcharge therefore are likely to have a reduced incentive to grow their balance sheet with low-returning assets such as Treasuries beyond what is required under regulatory requirements such as the Liquidity Coverage Ratio (“LCR”), given that the result will inevitably be a higher capital penalty over time.^[11] As discussed in Part II of this blog, recently proposed changes to the GSIB Surcharge methodology may only compound this negative impact on Treasury dealer capacity.

Third, higher post-GFC capital requirements (including risk-based requirements) have generally constrained the growth in the size of dealer balance sheets, meaning that bank dealers have relatively less capacity than they once did to intermediate the growing inventories of U.S. Treasuries. Indeed, as Darrell Duffie notes in a recent paper, since 2007, the combined size of primary dealers’ balance has shrunk by a factor of four relative to the growth in U.S. Treasuries outstanding (see Figure 2 below).

Figure 2: Ratio of U.S. Treasuries Outstanding to Primary Dealer Assets, 1998-2022

Source: Duffie (2023)

There is strong evidence that this constrained balance sheet capacity relative to the growing size of the market has exacerbated illiquidity during stress events such as the March-April 2020 Dash for Cash. During that episode, bank dealers purchased large volumes of Treasuries that other market participants, such as funds, proprietary trading firms (“PTFs”), institutional investors, and foreign governments, were looking to sell, but their capacity to do so was nonetheless significantly constrained relative to the demand. Other actors that provided liquidity during normal periods, such as PTFs, withdrew from the market altogether, because they did not have the same incentives to make markets on behalf of clients as the primary dealers. As a Federal Reserve Bank of New York staff report shows, various measures of market liquidity (e.g., the bid-ask spread, depth from the interdealer market and the dealer-to-customer market) declined rapidly^[13], leading to an effective freeze-up of the markets until the U.S. Treasury and Federal Reserve began a series of interventions in the markets to restore stability.^[14]

Consequences of Reduced Capacity and Resiliency

All of this raises significant financial stability concerns given the central role of the U.S. Treasury markets in the global financial system. It could also have long-term consequences for the cost of financing the U.S. government’s debt. For U.S. Treasuries to maintain their global safe-haven status for investors, it is crucial that the markets have intermediation capacities that are resilient even during periods of crisis-level selling. If investors begin to question their ability to quickly liquidate Treasuries, then the risk premium they charge for holding those instruments is likely to go up.

Indeed, there is some evidence that this is already happening: risk-premium indicators such as credit default swaps (“CDS”), as well as the term premium associated with higher five-year real yields have generally increased beyond what would be expected given the current path of monetary policy.^[15] Similarly, the Federal Reserve Bank of New York’s gauge of the 10-year premium became positive in late September 2023, after having stayed negative for most of the past 7 years, reflecting not only an expectation that interest rates will stay higher for a longer duration, but investor concerns about growing U.S. budget deficits – and thus greater supplies of outstanding Treasuries that may be more difficult to sell – in the future.^[16] Ultimately if investor confidence about long-term capacity and resiliency of the Treasury markets begins to wane, then the risk premia (yields) they will demand to hold those instruments will increase, resulting in elevated costs of servicing an ever-growing volume of government debt obligations.

Conclusion

The capacity and resiliency of the U.S. Treasury market have been called into question in recent years as market disruption events and heightened volatility become more frequent. These events have highlighted two underlying structural problems: a rapid growth in Treasury issuance and the constrained intermediation capacity of dealers owing to bank capital requirements. While regulators can do little to address the growth in issuance, they can take actions that mitigate the negative consequences for financial stability and long-term debt financing by expanding capacity in the system. One way to do this would be to reduce constraints on the ability of bank dealers to intermediate in these markets through measured reforms to both the SLR and GSIB Surcharge, as well as targeted changes to other risk-based capital rules. However, recent actions by regulators, particularly the banking agencies’ Basel III Endgame proposal, are likely to do the opposite, further constraining dealer capacity, raising transaction costs, and potentially disrupting key parts of the market. We discuss these potential impacts in Part II of this blog.

—

Author

Peter Ryan is Managing Director and Head of International Capital Markets and Strategic Initiatives at SIFMA.

Footnotes

^[1] “The Joint Staff Report on the U.S. Treasury Market on October 15, 2014,” July 13, 2015. Available at: Joint_Staff_Report_Treasury_10-15-2014.pdf.

^[2] See Federal Reserve Bank of New York, “Primary Dealers.” Available at https://www.newyorkfed.org/markets/primarydealers.html.

^[3] Inter-Agency Working Group for Treasury Market Surveillance, “Enhancing the Resilience of the U.S. Treasury Market: 2023 Staff Progress Report,” November 6, 2023, p.3. Available at: 20231106_IAWG_report.pdf (treasury.gov).

^[4] SIFMA U.S. Treasury Securities Statistics. Available at: U.S. Treasury Securities Statistics – SIFMA – U.S. Treasury Securities Statistics – SIFMA.

^[5] Congressional Budget Office, “The 2023 Long-Term Budget Outlook,” June 2023. Available at: The 2023 Long-Term Budget Outlook | Congressional Budget Office (cbo.gov).

^[6] This issue is discussed in more depth in Peter Ryan and Robert Toomey, “Improving Capacity and Resiliency in U.S. Treasury Markets: Part II – Proposals for Reforming U.S. Treasury Markets, March 30, 2021. Available at: Improving Capacity and Resiliency in U.S. Treasury Markets: Part II – SIFMA – Improving Capacity and Resiliency in U.S. Treasury Markets: Part II – SIFMA.

^[7] See, for example, Nellie Liang and Pat Parkinson, “Enhancing Liquidity of the U.S. Treasury Market Under Stress,” Hutchins Center Working Paper #72, Brookings Institution, December 16, 2020. Available at: WP72_Liang-Parkinson.pdf (brookings.edu). The need for potential reform (albeit more modest than excluding Treasuries from the SLR altogether) has also been acknowledged by former Federal Reserve Governor Dan Tarullo, who one of the principal architects of the post-GFC prudential reforms. See Daniel Tarullo, “Capital Regulation and the Treasury Market, Hutchins Center on Fiscal and Monetary Policy, Brookings Institution, March 2023. Available at: Brookings-Tarullo-Capital-Regulation-and-Treasuries_3.17.23.pdf.

^[8] Those components of the systemic indicator scores include the size; complexity; and reliance on weighted short-term wholesale funding; and cross-border transactions.

^[9] Daniel K. Tarullo, “Capital Regulation an the Treasury Market,” Hutchins Center on Fiscal and Monetary Policy at Brookings, March 2023, p. 8. Available at: Brookings-Tarullo-Capital-Regulation-and-Treasuries_3.17.23.pdf. Note

^[10] For more on this issue, see Francisco Covas and Brett Waxman, “GSIB Method 2 Fixed Coefficients Must be Adjusted for Economic Growth,” December 4, 2020. Available at: GSIB Method 2 Fixed Coefficients Must Be Adjusted for Economic Growth – Bank Policy Institute (bpi.com).

^[11] This point is at least partially conceded by former Federal Reserve Governor Tarullo in his 2023 article. He notes that while “it is hard to know just how much bank holdings and intermediation of Treasuries would increase were the G-SIB surcharge to be reduced” that “nonetheless, there is surely something to the claim, especially if surcharges were to continue to rise based on balance sheet growth that roughly parallels economic growth (and thus, at least presumptively, does not necessarily increase in the systemic risk posed by that institution).” See Tarullo, p. 9.

^[12] Chart adapted from Darrell Duffie, “Resilience Redux in the U.S. Treasury Market,” paper presented at the Jackson Hole Symposium, August 25, 2023. Available at: DuffieJH-2023.pdf (darrellduffie.com).

^[13] Darrell Duffie, Michael Fleming, Frank Keane, Claire Nelson, Or Shachar, and Peter Van Tassal, “Dealer Capacity and U.S. Treasury Market Functionality,” Federal Reserve Bank of New York Staff Report, No. 1070, August 14, 2023. Available at: delivery.php (ssrn.com).

^[14] For more on this episode, see Peter Ryan and Robert Toomey, “Improving Capacity and Resiliency in U.S. Treasury Markets: Part 1 – Why is Reform Needed? A Brief History of Recent Market Disruptions,” March 24, 2021. Available at: Improving Capacity and Resiliency in U.S. Treasury Markets: Part I – SIFMA – Improving Capacity and Resiliency in U.S. Treasury Markets: Part I – SIFMA.

^[15] Steven Major, Himanshu Malik, and Guy Baverstock, “Bonds Are Not Potatoes 2: Revising the Impact of Supply on Yields,” HSBC Global Research, September 27, 2023.

^[16] Michael Mackenzie and Garfield Reynolds, “Treasury ‘Term Premium’ Guage Positive for the First Time Since 2021,” Bloomberg, September 27, 2023.

Related Resources

• Pennsylvania + Wall Blog

  Revisiting U.S. Treasury Market Capacity and Resiliency: Part II

• SIFMA Research

  US Treasury Securities Statistics

Treasury Market Reform Blog Series

• Pennsylvania + Wall Blog Series

  Part I: Why is Reform Needed? A Brief History of Recent Market Disruptions

• Pennsylvania + Wall Blog Series

  Part II: Proposals for Reforming US Treasury Markets

• Pennsylvania + Wall Blog Series

  Part III: Evaluating the Benefits and Costs of Central Clearing in the Treasury Markets