Michelle W Bowman: Large bank supervision and regulation

Speech (virtual) by Ms Michelle W Bowman, Member of the Board of Governors of the Federal Reserve System, at the Institute of International Finance (IIF) event "In Conversation with Michelle Bowman", Washington DC, 30 September 2022.

The views expressed in this speech are those of the speaker and not the view of the BIS.

Central bank speech  | 
03 October 2022

Today, I would like to talk about the future of supervision and regulation of the largest banks, which changed significantly after the financial crisis 14 years ago and has evolved more gradually over the past 5 years. As the backdrop for this look into the future, it is important to recognize that this recent past for supervision and regulation has been a success, resulting in a banking system that is safer, stronger, better capitalized, and more resilient. This is particularly true for oversight of the largest banks, including global systemically important banks (GSIBs), and their central role in the financial system. As supervision and regulation have been refined in recent years, the largest banks have maintained high levels of capital, and their resilience has been repeatedly confirmed by both supervisory and real-life stress tests. Most notably, the U.S. financial system faced the onset of the pandemic in the spring of 2020, which disrupted financial markets and raised fears of a severe crisis. During this time, banks performed very well, continuing to keep credit flowing throughout the financial system as governments and central banks responded to the crisis, setting the stage for a rapid recovery from the sharpest economic contraction that the U.S. economy has ever experienced.

This outcome is a recognition, in my view, of the gradual and experiential approach to changing large bank supervision and regulation over the past several years. After the rush of regulation and the supervisory overhaul of large banks in the wake of the financial crisis, the Board of Governors took time to observe how the changes were working, and how things might be improved. Input was solicited from the public, and the changes made were incremental and carefully calibrated. I think the evidence is very clear that these changes have preserved and, in many cases, strengthened supervision and regulation, and that our gradual and evolutionary approach was wise. As I look ahead, that record of success is a testament to the progress we have made to date and argues for the same incremental approach to ongoing refinements in supervision and regulation.

I am looking forward to working with Michael Barr, the Board's Vice Chair for Supervision, on the dual goals of making the financial system safer and fairer, two objectives that I strongly support. In doing so, I am not opposed to changes that make sense, based on the experience we have gained from applying existing rules and approaches, or prompted by new and emerging issues. As always, we should ensure that any further changes yield significant improvement to safety and soundness at reasonable cost and seek to avoid approaches that fail to consider the tradeoffs between cost and safety.

In forming my judgments about whether proposed changes in regulation meet the standard I have just laid out, I will be guided by the four principles I described in 2021, outlining my perspective on bank regulation and supervision.1 I would like to briefly discuss these four principles, and then talk about how they have guided and will guide my thinking on a number of issues important to large bank supervision and regulation.

The first principle is that bank regulation and supervision should be transparent, consistent, and fair. Combined, these three elements, which we can think of collectively as due process, build respect for supervisory practices, and in doing so, make supervision more effective and encourage open communication between banks and supervisors. This principle applies equally to regulation. Supervision cannot replace-and should never supersede-rulemaking. Published regulations that have gone through the rulemaking process, with solicitation of public comment and bona fide engagement with the issues raised, are the best and clearest way for banks to understand the rules of the road and for the bank regulatory agencies to ensure banks satisfy safety-and-soundness objectives.

This brings me to the second fundamental principle for regulation and supervision: striking the right balance between ensuring safety and soundness, on the one hand, and promoting acceptable and manageable risk-taking, including encouraging responsible innovation. To put it simply, this means matching regulatory and supervisory requirements to the risks presented. For the largest banks, this naturally includes an increased focus on financial stability risks.

There are obvious risks from under-regulation, and it is those risks that were addressed in the wake of the financial crisis 14 years ago. But we sometimes overlook the significant costs to our economy, and risks to safety and soundness, from over-regulation, where rules are not designed and calibrated to address the actual risks. In a time of rising interest rates that could constrain credit, it is especially important to ensure that regulation and supervision not add costs and burdens for banks with little or no benefits to safety and soundness.

My third principle is that effective regulation and supervision needs to be efficient. Efficiency is key to effective regulation. In the design of a regulatory framework, there is flexibility in how to achieve a desired outcome, and there are often multiple approaches that would be effective in doing so. Once a decision has been made to regulate an activity, the next objective should be to ensure that the regulation achieves its intended purpose and that there are not more efficient alternatives that can yield those benefits at a lower cost.

My fourth and final principle is that regulation and supervision should serve a legitimate prudential purpose, like promoting safety and soundness, or reducing financial stability risk. After the last financial crisis, there was strong public support for enhancing regulation and supervision over the banking system, with a particular focus on the largest banks. And many of the steps taken after the last financial crisis have improved the resiliency of the U.S. financial system. While the need for robust regulation and supervision remains as true today as it was after the last financial crisis, regulation and supervision must also allow banks to continue providing credit and other financial products and services.

Collectively, these principles guide my thinking about the future of supervision and regulation. With that framework in mind, I would like to turn to some critical issues that are relevant to large banks.

Stress Testing

As I just discussed, a critical element of due process is that rules and supervision should be consistent among firms, and over time. This can be a challenge because of the variability in the business models of banks, especially among the largest banks. Each bank is different in terms of its balance sheet, business lines, and risk profile, so regulators must be vigilant that both regulations and supervisory practices are being applied consistently.

One area where this need for consistency is clear is in the Board's stress testing framework. Since 2013, stress testing has been used to assess banks' capital positions and determine whether they have sufficient capital to both absorb losses and continue lending during stressed conditions. This process has evolved considerably since its inception, and that evolution is important to ensure that stress testing continues to be relevant and effective.

Much of the stress testing framework is designed to encourage consistency-we have a common scenario design for firms, and similarly situated firms are subject to stress testing on the same frequency. However, the stress tests produce results that vary considerably from year to year due to how a specific scenario interacts with a specific firm's business model, and this volatility flows through to the stress capital buffers that apply to the largest firms. Although the stress scenarios are approved by the Board and change in some ways that are predictable over time or relate to changes in the underlying economy, how a scenario will affect a particular firm is not always predictable. These year-to-year variations are often not based in underlying changes to banks' business models and can create short-term challenges for capital management. There are likely many ways to limit this volatility while maintaining the value of the Board's stress tests, including by averaging results over multiple years.

As the stress testing framework continues to evolve over time, we should take into account what we learn from past tests, feedback from the public and the banks themselves, and ensure that the test evolves in a way that improves consistency and fairness over time.


Next, let's consider capital regulation. This is an area where requirements were quickly bolstered without extensive analysis in response to the 2008 financial crisis, resulting, in some cases, in redundant methods of calculating capital and demands that firms of all sizes and risk profiles comply with the highest requirements. The goal of efficiency dictates that over time this needs to be addressed. This is indeed what has happened over the past five years. In addition, the principle of balancing safety and soundness with the need for appropriate risk-taking is reflected in the extensive tailoring framework that the Board adopted over this time period, which carefully and deliberately matched regulation and supervision to the actual risks presented by different institutions. Often, the rules adopted immediately after the financial crisis applied a one-size-fits-all approach framed around the largest banks whose activities presented the most significant risks. This approach ignores the importance of bank size and business model. Obviously, expectations for the smallest banks with simple business models should not be equivalent to the expectations for large regional banks, or for large and complex bank holding companies engaged in significant securities or cross-border activities beyond taking deposits and retail lending.

As we look into the future to potential changes to the capital framework, including those expected under the Basel III endgame rulemaking, capital is a topic that is helpful to approach holistically. From my perspective, capital requirements should strike an appropriate balance for each relevant tier of firm, with requirements that appropriately address risks, including financial stability risks, while recognizing the costs of over-regulation. Calibrating capital requirements is not a zero-sum game, where more capital is necessarily always better. Regulation is not cost-free. Over-regulation can restrain bank lending, which becomes a burden for individual borrowers and a potential threat to economic growth.

Thinking about capital holistically also provides an opportunity to consider adjustments to the components of the capital requirements for the largest institutions, including the supplementary leverage ratio, the countercyclical capital buffer (CCyB), and, as I already discussed, the stress capital buffer, particularly where specific actions may have unintended consequences.

Since the onset of the pandemic, the banking system has seen a significant inflow of reserves due to the Federal Reserve's asset purchases in support of the economic recovery. For some firms, the influx of reserves resulted in leverage ratios becoming binding capital constraints, rather than serving as backstops to risk-based capital requirements. While these firms' leverage ratios may become less binding as the Federal Reserve reduces the size of its balance sheet and reserves are drained from the banking system, leverage ratios that discourage banks from intermediation in the Treasury market, or from holding ultra-safe assets such as Treasuries and reserves, can distort incentives and disrupt markets. Addressing these issues could improve market functioning and financial stability.

The CCyB is another component of the capital stack that deserves careful thought. In theory, the CCyB is a tool that could raise capital requirements in boom times to build resiliency and reduce capital requirements during times of stress to facilitate lending. In practice, the Board has not yet utilized the CCyB. While having releasable capital buffers shouldn't necessarily be ruled out, in my view, after a decade of stress testing and recent real-life stress experience, we have seen that the existing level of capital requirements has proven to be adequate for banks to deal with significant stress. Balancing safety and soundness with the need for appropriate risk-taking means that we should not simply assume that the further layering on of capital requirements, including through the application of the CCyB, would be beneficial.

The key is to strike an appropriate balance over time that addresses risks, including financial stability risks, without impeding the ability of the banking industry to extend credit and provide other financial services that are critical to our economy. The capital structure must also be predictable, to facilitate banks' longer-term capital planning, while preserving capital to allow firms to respond to unforeseen circumstances.

Bank Mergers and Acquisitions

I now want to turn to the review of bank mergers and acquisitions, specifically to discuss how the need for transparency, and to pursue legitimate prudential purposes, should guide our analysis of banking transactions under the established statutory framework.

The regulatory consideration of mergers is guided by the statutory factors prescribed by Congress-all of which are grounded in legitimate prudential purposes. The factors considered generally include the competitive effects of the proposed merger, financial and managerial resources, future prospects of the merged institutions, convenience and needs of the communities to be served, compliance with money laundering laws, and the effect of the transaction on the stability of the U.S. banking or financial system. 2 This analytical merger framework works best when it is accompanied by transparency, both in timelines and expectations, that allows firms to know and understand what is expected of them, and what they can fairly expect during the merger application process. The rules of the road should not change during the application process.

We should be vigilant to be sure that other factors, like the idea that mergers are harmful or that increased bank size is inherently problematic, do not infiltrate that statutory analytical framework. The analysis and approval of mergers and acquisitions should be based on the reality of how customers and the financial system would be affected. For larger banks in particular, the evolution of the merger review framework should also factor in the evolution of markets, industry, and customer preferences.

A merger can have a significant impact on local communities, in terms of the quality and availability of products and services. The effects of a merger can be beneficial to communities, enhancing the safety and soundness of a firm, and leading to significant public benefits. The consequences of getting these policies wrong can significantly harm communities, in some instances creating banking "deserts," especially in rural and underserved markets.3


Next, I would like to discuss resolution planning, or so-called living wills. 4 Each large banking organization is required to periodically submit a resolution plan to the Board of Governors and the FDIC, describing the organization's plan for an orderly resolution in the event of material financial distress or failure. The requirements for these plans are established by statute and regulation, with additional guidance published to give firms appropriate notice of regulatory expectations. For the U.S.-based GSIBs, resolution plans are also informed by other regulatory requirements, including the requirement that such firms issue a minimum amount of total loss-absorbing capacity, which includes both equity and long-term debt. 5 The requirements that establish and support the orderly resolution of firms are important to supporting the financial stability of the United States.

Of course, these requirements have evolved over time, most recently with a proposal intended to increase the efficiency of living wills by alternating between data-intensive full plans and risk-focused targeted plans. I expect that further evolution will be considered in response to ongoing changes in the financial landscape and the risks facing the largest firms. In doing so, I believe that fairness dictates that broad supervisory powers should not displace rulemaking.

In my view, the need for fairness and due process in resolution planning is particularly critical when it comes to considering whether and how to address concerns about the resolvability of regional banks. This question of fairness and due process is important, and it involves, among other things, a debate about the merits of a single point of entry resolution strategy on the one hand (more common among the GSIBs), and on the other hand, established bankruptcy and FDIC bank resolution procedures. This issue transcends particular firms, and particular transactions. It is an issue that affects a broad range of institutions of similar size. And policy actions in this area will require working with colleagues at other bank regulatory agencies and seeking public comment. Fairness dictates that this debate occurs in the arena of regulation, with all the appropriate due process protections that this entails, and not on an ad hoc basis for a single firm that chooses to make an acquisition subject to regulatory approval. If the regulatory framework for resolution needs to be improved, we should look at the framework, and identify and remediate any areas of concern. It is hard to understand why banks that choose to grow through acquisition should be subject to different resolution expectations than banks that grow organically. This strikes me as a clear example where requirements and expectations should only evolve through appropriate rulemaking processes, consistent with underlying law, in order to promote a level playing field.

Other Areas

There are a number of other areas where I think these principles can help frame a productive conversation about the future of regulation and supervision, including around banks engaging in crypto-asset-related activities, and improving the transparency of supervisory standards.

Another area where regulation and supervision continues to evolve is around banks engaging in crypto-asset activities. These activities raise a number of significant issues. When I think about the evolution of supervision and regulation of these activities, I ask myself whether the rules are clear in the current rapidly evolving environment, and whether the rules as they evolve are serving a legitimate prudential purpose. Banks seek to understand and comply with rules because, above all, they value predictability and consistency. When a bank understands the legitimacy of a rule and establishes internal incentives to comply with it, the bank itself becomes the strongest supervisory tool that there is. But sometimes, rules are difficult to apply. This can be due to quickly evolving technologies, particularly when it comes to digital assets, but it can also be due to a lack of experience with new rules, or when the rules are not that clear in a particular context. Banks should be able to know what the supervisory expectations are with respect to these new technologies in order to responsibly take advantage of them. The adoption and use of new technologies may present novel supervisory concerns, but the best way to address these concerns and encourage innovation is dialogue between bankers and supervisors before and during the development and implementation of those technologies.

I believe the goal with digital assets should be to match oversight to risk, and to provide clarity in supervisory expectations for banks seeking to engage in the crypto-asset ecosystem. As fluctuations in crypto-asset prices have shown, there clearly are material risks associated with these assets. However, it is also an area where there has been and continues to be intense consumer demand, and we should consider whether there is a stabilizing role for banks to play in intermediation, or ensure that the competitive landscape does not create a financial stability risk by pushing activities outside the banking system, as we have seen with the mortgage industry. To be effective in this space, any clarity regulators provide will need to recognize that this is not a risk-free activity, but I believe we should allow banks to participate as long as the risks can be identified and managed appropriately and responsibly.

Another way to promote consistency is to continue improving the disclosures around supervisory standards. While doing so improves transparency, it also improves fairness, another of my core principles. Like the due process protections enshrined in the U.S. Constitution and embedded in regulatory law, fairness is fundamental to the legitimacy and effectiveness of financial oversight, including supervision. In the context of bank regulation and supervision, fairness means being transparent about expectations, which should be clearly laid out in advance (and I want to emphasize that "in advance" part). Supervision should not be adjusted in specific situations to displace or alter regulations, or without appropriate notice and opportunity for public comment, and should be accompanied by clear communication with regulated firms. Where we have established precedents, we should respect them. Banks rely on our precedents in making their business decisions, so not respecting precedents can interfere with the ability of firms to plan and to fairly compete. If changes to precedents are appropriate, we should explain those intentions and employ a transparent and accountable administrative process to ensure fairness and appropriately implement the change.

Take, for example, the supervision criteria implemented by the Large Institution Supervision Coordinating Committee (the LISCC manual). Currently, these materials are not public. Making these materials public would not only improve transparency, doing so would also provide some assurance to the banks subject to them that they are being held to the same supervisory expectations as their peers over time. Without this clarity, it is far more challenging to build trust in this aspect of the supervisory process.

Improving transparency around supervisory standards promotes safety and soundness, both encouraging compliance, and limiting the role of formal and informal enforcement actions and penalties in addressing serious issues. In my view, success should not be measured by penalties or enforcement, but by how well banks are following the rules.

The principles I've articulated today reflect my approach to considering whether and how the regulation and supervision of the largest banks should evolve in response to changing economic and financial conditions. The regulation and supervision of financial institutions must be nimble to address new risks to safety and soundness and financial stability, but should always consider tradeoffs and potential unintended consequences, like increasing the cost of lending or pushing financial activities outside of the regulatory perimeter into the shadow banking system. I look forward to working with Vice Chair Barr, my fellow Board members, and colleagues at the other bank regulatory agencies, as we consider the evolution of supervision and regulation for the largest financial institutions.

1 See Michelle Bowman, "My Perspective on Bank Regulation and Supervision" (speech at the Conference for Community Bankers sponsored by the American Bankers Association, February 16, 2021). 

2 See 12 U.S.C. §§ 1828(c)(5), (11); 1842(c).

3 See Michelle W. Bowman, "The New Landscape for Banking Competition" (speech at the 2022 Community Banking Research Conference, St. Louis, MO, September 28, 2022). 

4 See 12 U.S.C. § 5365. 

5 12 C.F.R. pt. 252, subpart G.