Money markets are supposed to be resilient, but experts have grown wary after a series of disruptions in the past few years. Among them, two successive crises that rocked the US Treasury market within a six-month period in 2019 and 2020, and some temporary bedlam in the United Kingdom’s gilt market.
The search for a cause to these disruptions has led to an ongoing debate. Some argue that capital regulations, which impose balance sheet costs on banks, played a key role. Others emphasize the impact of regulatory requirements for intraday liquidity, which limit banks’ ability to act as lenders to funding markets in times of stress when central-bank reserves are scarce.
Or could both be at work? Adrien d’Avernas of the Stockholm School of Economics, Chicago Booth’s Quentin Vandeweyer, and MIT PhD student Damon Petersen, using a theoretical model, explore how both capital and liquidity regulations may jointly contribute to shocks. The interaction between the two types of regulations causes rate spikes in the “repo” market and fire sales in the US Treasury market, they find.
Government bonds are typically safe havens for investors during market crises. Because of these bonds’ inherent stability, their yields serve as benchmarks for transactions in both Treasuries and other securities and are the basis for the borrowing rates paid by consumers, businesses, and governments. A companion to the Treasury market is the market for repurchase agreements, or short-term secured loans, known as repos. In the repo market, the Fed is able to buy and sell Treasuries to redistribute liquidity between institutions that have it and those that need it.
The researchers modeled the disruptions in these two markets, homing in on regulations that were implemented following the 2008–09 financial crisis to maintain stability at large and systemically important banks. D’Avernas, Vandeweyer, and Petersen recognize that capital regulations make it costly for banks to maintain large balance sheets, which structurally pushed Treasury holdings to “shadow” banks such as hedge funds and other financial institutions that perform bank-like activities but without the same regulation. To magnify their profits from their Treasury positions, those institutions typically take on a lot of leverage by borrowing overnight in the repo market, thereby exposing themselves to rollover risk. If the overnight repo rate moves above the Fed’s interest rate, those institutions incur losses on their trades.
The ominous market behavior reflected a new form of inefficiency introduced as a safety measure after the 2008–09 financial crisis.
Why Was There a Treasury Bond Crisis amid the COVID-19 Stock Market Crash?Thankfully for hedge funds, however, traditional banks have remained an important player in this emerging ecosystem by acting as “lender of next-to-last resort” and lending their reserves in the repo market to shadow banks whenever necessary to avoid major disruptions. Crucially, banks have been able to do so because this asset swap does not increase the size of their balance sheets.
The intraday liquidity requirements imposed after the financial crisis changed this dynamic, the researchers explain, by tying banks’ ability to lend in the repo market during periods of stress to the quantity of reserves they hold. When a funding shock reduces the available reserves relative to the demand for repos from shadow banks, banks’ lending capacities are constrained and repo rates can spike. If the shock is severe enough and expected to last, shadow banks may respond by liquidating part of their Treasury portfolios to avoid paying high repo rates over an extended period.
This framework explains what happened in the US Treasury market in September 2019 and March 2020, the researchers write. The September 2019 disruption was driven by a tax deadline and was, therefore, perceived to be temporary. It drove repo rates above the interest paid on reserves, reflecting a scarcity of repo supply. Because shadow banks were expecting markets to calm down quickly, they preferred to pay a high repo rate for a short period of time rather than liquidate their Treasury portfolio. In contrast, in 2020, Treasury yields rose sharply while repo rates didn’t. Challenges associated with the pandemic were anticipated to be longer-lasting, so shadow banks, faced with potentially paying high repo rates for a while, dumped Treasuries and cut their losses.
The researchers write that their model helps identify the conditions that lead to repo-rate spikes, and sheds light on the role a central bank’s balance sheet plays in transmitting funding shocks to the Treasury market. US government debt has grown dramatically over the past two decades, and so have Treasury securities outstanding on the Fed’s balance sheet. A larger central-bank balance sheet, combined with an increased supply of reserves on commercial-bank balance sheets, encourages commercial banks to lend more in the repo market following a liquidity shock. A reduction in the size of the central-bank balance sheet exerts simultaneous pressure on both the demand and supply of repos, increasing the likelihood of disruptions, the researchers conclude.
D’Avernas, Vandeweyer, and Petersen emphasize that both the asset and liability sides of the central bank’s balance sheet affect markets and may independently drive market disruptions. The asset side determines the amount of Treasuries that shadow banks want to hold, whereas the liability side affects the capacity of banks to lend Treasuries.
Therefore, reducing the central banks’ balance sheet has a double impact on the likelihood of disruptions, write the researchers. And even passive adjustments to the Fed’s liabilities can reduce banks’ reserve holdings and exacerbate disruptions. All these insights could help policymakers trying to ward off market disruptions and maintain financial stability.
Adrien d’Avernas, Quentin Vandeweyer, and Damon Petersen, “The Central Bank’s Balance Sheet and Treasury Market Disruptions,” Working paper, May 2024.
A look at the market for these billion-dollar startups.
Death of the Unicorns
Analyzing both numbers and narrative context in company reports led to better price forecasts.
Large Language Models Can Improve Stock Market Forecasts
Research finds that banks prefer to issue mortgages for newer and more standardized housing stock.
Line of Inquiry: Anthony Lee Zhang on Why Buying Your Unique House May Be a ChallengeYour Privacy
We want to demonstrate our commitment to your privacy. Please review Chicago Booth's privacy notice, which provides information explaining how and why we collect particular information when you visit our website.