U.S. Department of the Treasury

As Prepared for Delivery

Introduction

Thank you to the Board of Governors, the IMF, and the World Bank for inviting me to speak today.

Over the past year, severe shocks hit a number of key institutions across global finance. Several large regional banks failed in the United States. The two Swiss global systemically important banks, or G-SIBs, announced an agreement to merge into one. One of the biggest cryptocurrency exchanges, an exchange with headquarters in The Bahamas, filed for bankruptcy.

The reach of these firms crossed international borders. My brief remarks today therefore ask, given these recent events, should the content and form of international financial regulatory cooperation shift?

U.S. Regional Bank Failures

In recent months, problems emerged at several U.S. banks. Silicon Valley Bank and Signature Bank faced a loss of confidence and funding and, to prevent impacts on the broader banking system, these banks were placed into FDIC receivership. First Republic Bank was closed by regulators and sold to JP Morgan.

The runs on SVB and Signature occurred at unprecedented speed.  SVB customers withdrew $42 billion of deposits in just one day, and the bank faced around $100 billion in withdrawals for the next day, before its ultimate closure.  Signature Bank entered receivership two days later, after depositors withdrew 20 percent of total balances. By contrast, depositors withdrew about $17 billion over eight business days during the run on Washington Mutual in 2008 – still the largest U.S. bank failure adjusted for inflation.  In May 1984, Continental Illinois saw six days of sizable withdrawals, but the peak one-day withdrawal rate was only 8 percent of deposits. 

As a recent Federal Reserve Board report has noted, a number of factors may have contributed to the speed of these runs.[1] The widespread use of social media let depositors share news of perceived problems more quickly. Fintech and online banking meant withdrawal requests could be submitted more quickly. A highly concentrated depositor base coming from related industries, a particularly notable characteristic in the case of SVB, may have led to a correlated need for withdrawals, covering a large share of the bank’s total deposits, of which a large proportion was uninsured. 

Disentangling these features, and generalizing from their coincidence, is a difficult task, but the speed involved in the episodes raises obvious questions for regulation, supervision, and financial policy. For example, key components of the post-crisis bank framework rest on the calibration of liquidity requirements to specific subsets of institutions, relying on certain assumptions regarding net stressed outflows. Do the scope and calibration of these rules need be adjusted, and if so, what would that involve? Must we change the way we monitor contagion in the banking sector? Should banks change the way they test and hedge their liquidity risk?

These events also raise obvious questions for deposit insurance frameworks. At its peak in 2021, the proportion of uninsured deposits in the U.S. banking system was 46.6 percent—higher than any point since 1949.[2]  At the end of 2022, around 90% of SVB and Signature Bank’s deposits were uninsured.  The FDIC has already compiled a set of options for deposit insurance reform. 

We clearly should pause and consolidate lessons learned. But need this imply a large pivot? International coordination and information sharing on supervision and regulation of the financial system occur to a large extent at the Financial Stability Board, or FSB, the standard-setting bodies, and regulatory dialogues. Collectively, they’ve focused substantially in recent years on mutual funds, hedge funds, pension and insurance companies, and various alternatives to bank financing referred to as non-bank financial institutions, or NBFI. With three of the four largest bank failures in U.S. history happening over a span of mere months, should we stop focusing so much on NBFI?

I don’t think so. Before the recent banking stresses, there was a widespread belief that NBFI posed important risks to financial stability. Given the shift in assets, if anything, there’s more cause to focus on it now, after those recent events, not less.

Over the 12 months ending in February of this year, deposits at commercial banks fell slightly, but assets in money market funds increased $157 billion or over 3 percent.  Since March, this growth accelerated because that’s to a large extent where investors moved their withdrawn deposits. Over the past 10 weeks, depositors appeared to shift over $460 billion into money market funds. 

And there are demonstrated risks in other areas of NBFI, including interest rate exposures and runnable liabilities at insurance companies, leverage strategies used by pension funds, and elements in how commodity markets function. Structural vulnerabilities related to liquidity mismatch and opaque leverage still remain, and at a time of tightening financial conditions, the consequences of those vulnerabilities are only growing. 

The NBFI sector is bigger and more interconnected than ever. The FSB set out an ambitious workplan to enhance the resilience of the NBFI sector in 2020, building on their earlier work on the topic, and other bodies have similarly done so. We of course should digest the new information we’ve learned from regional bank stresses, but NBFI warrants a continued focus in global discussions and coordination.

Global Systemically Important Banks

Other institutions came under acute stress this spring, including Credit Suisse, which had been suffering from a series of idiosyncratic problems for several years.  Credit Suisse, a G-SIB that was founded in the mid-19th century, eventually succumbed to these strains and was sold to the only other Swiss G-SIB, UBS.

The FSB, together with the Basel Committee on Banking Supervision and national authorities, has been identifying G-SIBs since 2011. One goal of such designations, together with various other post-crisis reforms, is to preserve market discipline in credit allocation, while minimizing spillovers upon failure. The resolution architecture is designed to make sure losses fall on the institution’s investors and, to the extent necessary, creditors. Not the taxpayer. 

But Credit Suisse was the first post-crisis failure of a global systemically important bank, and the actions taken to address that failure required support via a public backstop. The merger between UBS and Credit Suisse involved assistance from the Swiss government. 

To be clear, I strongly support the Swiss authorities in their decisive actions to shore up financial stability.  They offered a solution in a moment of acute stress, with short time frames and limited resources. I also note that the post-crisis reforms meant that policymakers and supervisors at that time of stress were dealing with a simpler institution, a “clean holding company” capable of supporting a pre-planned resolution if needed. And the existence of recovery and resolution plans can certainly have real value even if another arrangement ultimately emerges. Nonetheless, we should strive to better understand what the limitations on resolution were in this case, and to consider whether these limitations might apply more broadly.

Further, we should try to learn from the episode on cross-border communication, a critical foundation for success in coordinating a hypothetical G-SIB resolution. One of our most important tasks, outside of a crisis, is to lay the groundwork for rapid communication during times of stress.  It is important for all of us – including staff and political leadership – to know who our counterparts are, and to communicate with them regularly and openly, including planning together for what would be distinctly unpleasant hypotheticals.

Together with the FDIC, the Fed, and a number of supervisors including the SEC, we at Treasury conduct something called the Trilateral Principal Level Exercise, or TPLE. The TPLE connects staff and high-level leadership – including the Secretary – to their international counterparts in the UK and EU for resolution planning discussions, a key difference from crisis management group meetings. The recent episode with the Swiss G-SIBs confirms that such cooperation should not be limited to only a few large jurisdictions. Rather, G-SIBs’ headquarters span a wide geography, including Canada and Japan, and we should consider how to extend such coordination and discussions to include more countries that domicile G-SIBs.

Digital Assets and Emerging Market Jurisdictions

The regional banks and G-SIBs I just discussed are located in some of the most developed financial markets in the world. The United States and Switzerland are key participants in the FSB. The Basel Committee sits, of course, in Basel. But some key firms whose failures may come to represent important risks to global financial stability have now made their homes in non-G20 jurisdictions, or emerging markets and developing countries.

In November 2022, we saw the failure of FTX, once the third largest digital asset exchange by volume. FTX failed when it could not meet withdrawal requests on the order of $6 billion from its customers due to liquidity shortages and alleged fraud.  Digital lenders with significant exposures to FTX filed for bankruptcy, the price of FTX’s native coin FTT collapsed, and ultimately, the market value of crypto-assets fell by more than two thirds from its peak.  Luckily, FTX’s failure had limited spillovers to the traditional global financial system, but we should not assume that this will be the case next time.

FTX was incorporated in Antigua and Barbuda and headquartered in The Bahamas. Key entities in the digital assets ecosystem are located in jurisdictions that don’t participate in the FSB. And while the United States has a number of bilateral regulatory fora, such as the Financial Regulatory Working Group with the UK, or the EU-US FinReg Forum, we have had more limited opportunities for candid discussions on supervision and regulation with Caribbean nations, for example.

As technology or changes allow global financial risks to scale outside of traditional economic centers, we must think – and quickly – about how to invite these jurisdictions to join, and incent them to participate, in the global financial regulatory conversation. One promising avenue is via the FSB’s regional consultative groups or RCGs, which cover dozens of jurisdictions in six global regions. The RCGs typically meet twice a year, and consistent with the conclusions of an FSB review completed on the eve of the pandemic, more effort might be placed on raising the efficacy and profile of these meetings to broaden engagement and dialogue.

Unlike the historical development pattern for traditional finance, many emerging markets are at the forefront of digital asset development or adoption. Their policymakers must be included in discussions of the resulting implications for financial stability.

Conclusion

So, does the recent stress in regional banks suggest we need to think harder about some key elements in bank supervision and regulation, including deposit insurance regimes? Of course it does. But I think our focus on NBFIs has and will continue to serve us well, not least because that’s where many of the deposits ultimately went.

What are key lessons to learn from the public-backed merger of the two Swiss G-SIBs? It’s too early to tell, but one may be that more jurisdictions should be involved, with participants at the highest levels, in resolution planning discussions and exercises.

And finally, the fact that key nodes in the digital asset ecosystem are domiciled in emerging markets or developing economies adds even more reason to redouble our efforts to broaden participation in international regulatory discussions.

None of these three episodes, though difficult, spiraled into anything as serious as what was experienced in the 2008 global financial crisis. This may very well owe, in part, to the reforms implemented over the last 15 years and the venues for international cooperation and dialogue that we’ve built.  We should not miss the chance to keep up this good work.