Remarks By Assistant Secretary for Financial Institutions Graham Steele at the Americans for Financial Reform Education Fund

As Prepared for Delivery

Introduction

Good afternoon. It is nice to see so many of you.  I appreciate the opportunity to be a part of this event, and I am thankful to Lisa for the invitation to be here today.  I am fortunate to have a long history with Americans for Financial Reform, having been present at its founding in the wake of the 2007-2009 Global Financial Crisis. I would also note that last Friday was the thirteenth anniversary of the enactment of the Dodd-Frank Wall Street Reform and Consumer Protection Act.

Today, as the Assistant Secretary for Financial Institutions at the Treasury Department, my portfolio includes developing the Department’s policy views on banks, credit unions, insurance, consumer protection, access to capital, and financial sector cybersecurity.  I will focus my remarks on the recent banking sector turmoil that began in early March.  I will offer some observations about what led to the bank failures and some considerations that may be relevant for any future steps that may be taken to help prevent future banking stress.^[1]

Response to the Banking Turmoil

In early March, the rapid failure of Silicon Valley Bank (SVB), followed two days later by the failure of Signature Bank, created the potential for significant impacts to the broader banking system and U.S. economy.

SVB’s failure was caused by a bank run.  As SVB experienced a decrease in deposits from late 2022 into early 2023, the bank did not have sufficient cash on hand to meet those demands, leading it to sell its portfolio of available-for-sale securities, generating a $1.8 billion after-tax loss.  Concurrent to its announcement of the sale of securities, the firm also indicated it would seek to raise additional private capital, but it had difficulty doing so and quickly abandoned its efforts.^[2]  Depositors came to believe that the bank would have difficulty meeting withdrawal demands, motivating them to join the withdrawal line before the bank ran out of cash or failed.  The result was a depositor run of historic proportions, with over $140 billion in deposits—roughly 85 percent of the bank’s deposit base—leaving or attempting to leave SVB over the course of two days.^[3]

In response to these events, and concerns about potential contagion, Treasury, the FDIC and the Federal Reserve took decisive and forceful actions to strengthen public confidence in the U.S. banking system and protect the American economy.  The steps we took were not focused on aiding specific banks, classes of banks, or specific groups of depositors.  Our intervention was necessary to protect the broader U.S. banking system.  Our actions increased the stability and soundness of our banking system while protecting depositors.  Management, shareholders, and debtholders were not protected.  In May, First Republic Bank was closed by regulators and sold.  As Secretary Yellen has already noted, the resolution of First Republic was more of an aftershock of the March developments than a sign of any shift in the fundamental health of the banking system.^[4]

As the Secretary has stated, we believe our actions reduced the risk of further bank failures that would have imposed losses on the Deposit Insurance Fund.^[5]  Since then, we have seen the stabilization of deposit flows.  The Federal Reserve’s Bank Term Funding Program is working as intended to provide additional liquidity to complement the discount window.  That said, we continue to closely monitor the banking system and work with the banking agencies to ensure that we are attentive to remaining risks, including the continuing need to properly manage interest rate risk.

With the events of the past few months fresh in mind, I would like to offer some observations on the preliminary lessons learned from this episode.

Underappreciated Risks of Large Regional Banks

In February, I participated in a panel discussion at the OCC’s symposium on bank mergers—just a month prior to the SVB failure—where the discussion centered on how to evaluate the financial stability risk from bank mergers, including whether the failure of a large regional bank would affect U.S. financial stability.^[6]  In hindsight, those questions were more germane than many realized, occurring shortly before the second, third, and fourth largest bank failures in U.S. history by assets, accompanied by two systemic risk exception determinations.

Assessing the financial stability implications of a bank’s failure requires understanding multiple dimensions of that bank’s potential systemic importance.^[7]  One way to measure the financial stability risk posed by a bank is to focus on the expected impact of its failure using ex-ante metrics that assess direct and indirect exposures through risk indicators like size, interconnectedness, and the provision of critical services.  This expected impact method is the predominant approach to measuring a firm’s financial stability impact.  Using this method, SVB’s systemic importance score was 17 basis points as of year-end 2022, well below the 130 basis point threshold that would qualify a bank as systemically important.^[8]  

A second way of assessing financial stability risk is through the actual or ex-post impact of a bank’s failure, for example from deposit runs on institutions with similar characteristics and resulting costs to programs like the Deposit Insurance Fund.  This measurement approach provides a more complete picture but can only be assessed with certainty after a bank’s failure.    

Another important consideration in evaluating the impact of a bank’s failure is the resulting implications for its customers and, more broadly, the communities and regional economies in which it operated.  I have heard firsthand about the impacts of SVB’s failure on its investors, depositors, business customers, and affected communities in Northern California.  While it was encouraging that a buyer emerged shortly after SVB’s failure, which has facilitated the transition for most of SVB’s customers, there have been tangible disruptions to the industries and communities for which SVB served as an important financial intermediary.

As Secretary Yellen has noted, the U.S. economy benefits from our broad and diverse banking system.^[9]  Banks that serve specific regions and industries can often provide higher-touch and more customized products and services than global systemically important banks.  After a regional bank’s failure, the demand for these services may remain.  While there may be alternative bank and nonbank providers, underwriting and other business processes may take longer and become more cumbersome.  At the same time, there may be less innovation in developing novel or customized products.  As a result, certain parts of the startup and small business ecosystem may receive less attention and, by extension, capital to grow and expand.  The substantial value that regional banks provide during normal times means that, conversely, regional bank failures can also impose substantial costs on regional economies.^[10]  

Observations and Preliminary Lessons Learned

Much of the post-mortem examination of recent events in the banking sector, particularly of the SVB and Signature failures, has focused on institutional mismanagement and supervisory failures.  But blaming these two factors—the individual firms’ inadequate risk management and the lack of effective supervisory oversight of liquidity and interest rate risk—is both true and insufficient.  Regulation also plays a vital role.  Regulation is the mechanism that the banking agencies can use to impose and enforce requirements that are substantive, standardized, and transparent.  Appropriately calibrated regulation establishes clear expectations for supervisors and enforcement staff and provides them with institutional tools to take action against noncompliant firms.    

As was demonstrated by the spirited discussion at the OCC symposium, the tailoring framework that weakened, and in some cases removed, regulatory and supervisory requirements for large regional banks was undergirded by a belief that the failure of those banks would not have broader systemic effects.^[11] The systemic impacts of a bank’s failure can be difficult to anticipate and can vary greatly depending on a range of factors, including the ex-ante and ex-post factors I described earlier and the broader economic backdrop.^[12]

These issues are not new.  Since the 2007-2009 global financial crisis, policymakers have debated where and how to set the thresholds for banks to be subject to more stringent regulations—known as enhanced prudential standards (EPS)—designed to both reduce the likelihood of bank failures that could affect U.S. financial stability and to reduce the systemic effects of any failures that occur.  The Dodd-Frank Act, as originally passed in 2010, applied the EPS at a considerably lower level than it is today. This was, in part, to avoid the implication that the application of these standards meant that all banks above this threshold are considered systemically significant in their own right—and, by extension, would be treated as Too Big to Fail.^[13]

In light of the lessons that we have learned, and in some cases re-learned, about the inherent uncertainty in predicting banking panics, I want to highlight three relevant principles that I believe were reinforced by the events that occurred this spring.

First, there is a role for simplicity and complementarity in regulatory design.

While the banking agencies work towards modifying their regulatory frameworks, the importance of strong bank capital requirements remains top of mind. As other have noted, one factor motivating the depositors’ run on SVB was a concern about its solvency, particularly the risk that the unrealized losses on the firm’s securities holdings were larger than the firm’s equity.^[14] This loss of confidence underscores the importance of credible and robust capital standards and prompt regulatory intervention.

In a period of financial stress, the market looks to simple measurements to assess a bank’s solvency.  As they did in the global financial crisis of 2007-2009, market participants focus on the most loss-absorbing capital—known today as Common Equity Tier 1, or CET1—and the most risk-insensitive capital requirement—the leverage ratio—as the most salient indicators of a bank’s health.^[15]

The market has now become focused on the unrealized losses in banks’ available-for-sale and held-to-maturity securities as important metrics to assess a bank’s solvency.  This highlights the disconnect between useful measurements of a bank’s health and the calculation of bank capital requirements.  Today, capital requirements do not account for unrealized gains or losses on held-to-maturity securities for any firm, and even unrealized gains or losses on available-for-sale securities are reflected in capital requirements only for the largest banks, those in Category I and Category II of the tailoring framework.  Recognizing unrealized gains and losses on its available-for-sale securities in its CET1 capital would have reduced SVB’s regulatory capital by $1.9 billion in the fourth quarter of 2022, potentially causing it to raise capital sooner.^[16]  Federal Reserve Board Vice Chair for Supervision Barr has indicated that the Board will revisit the ability for banks with $100 billion or more in assets to opt out of accounting for unrealized gains or losses on available-for-sale securities in their capital requirements.^[17]

It is also worth noting that the recent stresses were touched off by interest rate-induced losses in banks’ securities portfolios, including highly-rated securities with very little or no credit risk.  This demonstrates both the importance of less risk-sensitive measures in addition to risk-based ones and highlights the importance of evaluating the safety and soundness of banks from a variety of credible perspectives.^[18]  No single measure will comprehensively capture the full range of potential risks, underscoring the need for a belt-and-suspenders approach to capital regulation.

Second, the effectiveness of regulations depends on their embedded assumptions.

Banking rules rely on various assumptions regarding credit risk, liquidity needs, and broader macroeconomic indicators using projections based upon historical performance.  This underscores the limitations and imprecision inherent in any regulation, as well as the importance of model risk management.

For example, liquidity standards and supervision, along with effective liquidity risk management by firms, are a critical component of bank regulation and supervision.  Shortcomings in liquidity risk management, particularly in the case of SVB, were clearly a catalyst in recent events.  As a result of the tailoring changes to bank regulations, banking organizations like SVB with between $100 billion and $250 billion in assets generally no longer have any standardized liquidity requirements applying to them, as then-Federal Reserve Governor Brainard noted in her statement accompanying her vote against the final tailoring rule in 2019.^[19]  The Federal Reserve Board’s report on SVB’s failure noted that SVB would have been below the full liquidity coverage ratio (LCR) requirements for some time.^[20]  Had the full LCR applied to the firm, the regulatory framework would have triggered a requirement for the bank to promptly provide the Federal Reserve with a remediation plan to address its shortfall and importantly provided an early warning signal of its risky liquidity profile.^[21]  Again, more timely action may have resulted in a different outcome.

It is important to emphasize how relatively new prudential liquidity standards are.  Prior to the implementation of the Basel III accord, U.S. banks did not have any standardized requirements for liquidity risk.^[22]  These recent experiences suggest that depositors’ relationship to their banks is changing and that runs are much faster and more significant.  As the Federal Reserve’s report noted, the level and speed of withdrawals by SVB’s uninsured depositors went far beyond the firm’s own assumptions under its internal liquidity stress test (ILST).^[23]  There are many reasons why this might be the case, including concentrated depositor networks and the ease with which any depositor can initiate a transfer of funds. In light of the remarkable scale and scope of deposit outflows, Vice Chair Barr has stated that the Board should re-evaluate the current run-off assumptions for uninsured deposits in the standardized liquidity rules and in a bank’s ILST.^[24]

Recent changes to FDIC rules and the agencies’ LCR rule have also made it easier for banks to accept reciprocal deposits and effectively excluded some bank-fintech partnerships from the brokered deposits rule.^[25]  Reciprocal and brokered deposits may warrant greater attention now that they are playing an increasingly important role in bank funding structures in light of the recent events.^[26]  This recent episode can help to inform a broader consideration of how well the standardized liquidity frameworks are performing and if further refinements would be appropriate.^[27]

Third, we must be mindful of the gaps in the regulatory framework.

As the Fed’s report noted, SVB grew rapidly—nearly tripling in size over a two-year span^[28]—but there was a delayed application of the tailored EPS for Category IV firms.  Currently, Category IV firms are subject to the supervisory stress test every other year. As a result, stress test results—and the stress capital buffer, the associated regulatory requirement—may come too infrequently to address risks at rapidly growing firms.  

To illustrate the point, SVB’s total consolidated assets first exceeded $100 billion in December 2020.  However, the firm did not exceed the $100 billion in average total assets threshold until June 2021—at which point, under the tailoring rule framework, it became subject to Category IV standards.^[29]  SVB would have first been subject to the stress test in 2024, with the stress capital buffer requirement from the test applying in the fourth quarter of 2024. ^[30]  In other words, there would have been a four-year lag between the bank reaching a particular size and the application of capital requirements commensurate with that size.

More timely application of the EPS could address the increased risks that arise from rapid asset growth by ensuring banks have appropriate capital and liquidity.  For example, it is worth considering whether the capital requirements contained in the existing tailoring framework accurately reflect the risks posed by banks in Categories II, III and IV—particularly as they increase in asset size without exceeding other thresholds contained in the tailoring framework.  The capital requirements for such large regional banks are relatively static, meaning that many of the largest non-GSIB domestic banks have the same risk-based capital requirements as much smaller and less complex banks.^[31]  Because of the important role that large regional banks play in the economy and in their communities, it is critical that they have capital commensurate with their risk.

It is also noteworthy that Signature and First Republic operated without bank holding companies and were therefore not subject to the holding company regulatory regime, including regulatory requirements like the Fed’s supervisory stress testing framework.^[32]  It is not obvious why the application of critical regulatory frameworks should distinguish between institutions with a holding company structure and those without one—particularly for large regional banks with assets primarily held in their depository institutions.  

Conclusion

In closing, it’s important that we try to learn the right lessons from the banking turmoil of recent months.

In 2008, the U.S. government was forced to arrange fire sales of failing regional banking organizations, including Wachovia and Washington Mutual.  The regulatory and supervisory regimes developed in response to the financial crisis helped to ensure that the recent banking stress has thus far been limited to only a few firms.  The broader banking system has also generally weathered the recent volatility in the crypto-asset markets, largely due to the banking agencies taking a careful and cautious approach to these activities and responding to emerging risks in a timely and well-founded manner.^[33]

That said, these recent events have made clear that the risks of some banks and activities were underappreciated.  In that sense, complacency was a contributing factor in the failures of SVB, Signature, and First Republic.^[34]  As Secretary Yellen has noted, regulatory requirements have been loosened in recent years and it is therefore appropriate to assess the impact of these deregulatory decisions and take any necessary actions in response.^[35]  The President has urged the banking regulators, in consultation with the Treasury Department, to consider a set of reforms that would reduce the risk of future banking crises.^[36]

The banking agencies have recently made commitments to propose the final elements of the Basel III reforms in the near term,^[37] and to propose additional revisions to their prudential frameworks soon.^[38]  Depending on what the agencies ultimately propose, this series of reforms could both complete work that began in the wake of the global financial crisis and respond to the specific vulnerabilities highlighted by recent events.^[39]

Treasury, across administrations, has long supported the efforts of the U.S. banking agencies to implement the Basel Accord in the United States in a manner consistent with the unique structure of our domestic banking system to further safety and soundness and promote domestic financial stability.  In the wake of the 2007-2009 global financial crisis, Secretary Geithner remarked that the original Basel III package was critical to making the financial system more stable and more resilient.^[40]  Further, after the final package of Basel III reforms was agreed to in 2017, Secretary Mnuchin stated that the reforms would standardize the approach to capital regulation, improve the quality and consistency of bank capital requirements, and help level the playing field for U.S. banks.^[41]  It is important to note that any modifications to prudential regulations would be subject to notice-and-comment rulemaking and would include an appropriate transition period.^[42]

A few weeks before SVB’s failure, I observed that strong regulation and consumer protection are vital to ensuring that our financial system can serve as a source of stability and equitable growth.^[43]  Appropriate regulation helps to preserve banks’ viability and competitiveness, enabling them to weather disruptions resulting from declines in asset values due to factors like changing interest rates.  This instills confidence in customers and depositors, even during times of economic uncertainty and financial stress.  While there is a fair amount of work to be done to respond to recent developments in the banking sector, the good news is that some of this rethinking is already underway.

Thank you again for having me, and I look forward to the panel discussion.