Randal Quarles: Getting it right - factors for tailoring supervision and regulation of large financial institutions

Speech by Mr Randal K Quarles, Vice Chairman for Supervision of the Board of Governors of the Federal Reserve System, at the American Bankers Association Summer Leadership Meeting, Salt Lake City, Utah, 18 July 2018.

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

Central bank speech  | 
19 July 2018

I want to thank the American Bankers Association for inviting me to speak.1 This is an era of relatively rapid evolution in banking regulation, an area of human endeavor that is not commonly known for its speedy metamorphoses. But in the time since I became the Vice Chairman for Supervision at the Federal Reserve, we have seen agreement on the final pieces of the international framework for post-crisis regulation-the so-called Basel III endgame. The Federal Reserve has issued a number of proposed rule changes that would improve our capital and stress testing regime. And in late May, the Congress enacted the Economic Growth, Regulatory Relief, and Consumer Protection Act (EGRRCPA), which, among other things, directs us to further tailor our supervision and regulation of large banks with more than $100 billion in assets. In other words, the Congress wants to see action and has, to a certain degree, specified some of the steps we need to take. How we respond to that task, especially as it applies to large banks, will be the focus of my remarks today.

But before I delve into those details, I want to provide an overview of what I hope you'll take away.

First, I want to underscore that I believe tailoring of financial regulation is good public policy. The Federal Reserve Board is a firm adherent of the recent legislation's underlying principle that regulation should be tailored to risk. To an extent, as I will later describe, this principle is already embedded in several aspects of our supervisory framework.

Second, the Federal Reserve will need to revise its framework to allow for a greater differentiation in the supervision and regulation of large firms. To date, our tailoring of regulations has been based largely--but not exclusively--on asset size, which reflects an unduly one-dimensional approach. We have been evaluating additional criteria that may provide for greater regulatory differentiation across large banks, and the recent legislation is consistent with the goals of that initiative. Specifically, the legislation recognizes that large banks have a variety of business models and risk profiles; supervision should be flexible enough to incorporate this heterogeneity.

Third, I believe we already have a good start on a path forward for tailoring regulation. In my remarks today, I will touch upon some potential factors we have identified for tailoring the supervision and regulation of large banks.

Tailoring: A Primary Component of Post-Crisis Supervision and Regulation

I think everyone in this room would agree that-while there are important ways it can be improved-the body of post-crisis regulation adopted by the Federal Reserve and its fellow banking agencies has, taken as a whole, clearly made the U.S. financial system stronger and more resilient. In implementing these reforms, the Fed sought to achieve two parallel goals: (1) promoting the safety and soundness of individual banking organizations and (2) enhancing the stability of the broader U.S. financial system.

A certain amount of tailoring was reflected from the beginning in how the Fed sought to achieve these goals. As the Board built its post-crisis framework, supervision and regulation were designed to increase in stringency in tandem with a firm's size and systemic footprint. This can be seen in stricter requirements in various elements of the regulatory capital framework that apply only to larger or more complex banks, including certain buffers and surcharges, the application of the supplementary leverage ratio, and the application of the qualitative objections as part of the Board's capital planning framework, among others. In April, the Board proposed the stress capital buffer, which would simplify its regulatory capital requirements for the largest banks by integrating the stress test results with the Board's non-stress capital requirements.

Further, the Board recognized that the failure of one of the largest banking organizations could create spillovers that would undermine U.S. financial stability and harm consumers and the broader economy. To offset this risk, the Board has required these firms to internalize the cost of their potential failure in a tailored manner that corresponds with their importance to the U.S. financial system. These efforts include the Board's capital surcharge for global systemically important banks (G-SIBs) and total loss-absorbing capacity requirements.

The EGRRCPA and Tailoring

But while the concept of tailoring is inherent in how the Board thinks about supervision and regulation, reasonable people can disagree on the sufficiency of tailoring to date. In my view, we've made a good start in improving the efficiency of our regulatory regime. But we still have more to do to streamline our framework in a manner that more directly addresses firm-specific risks.

The recent legislation requires us to reevaluate how we regulate banks that have between $100 billion and $250 billion in total assets. In particular, we need to make a tailoring-related decision in the near term: How will we decide which enhanced prudential standards should apply to which firms with total assets between $100 billion and $250 billion?

In applying enhanced prudential standards for firms with total assets of more than $100 billion, the Congress requires the Board to consider not only size but also capital structure, riskiness, complexity, financial activities, and any other factors the Board deems relevant. While we use similar factors to calibrate the largest firms' G-SIB surcharges, we have not used them more holistically to tailor the overall supervision and regulation of large banks that do not qualify as G-SIBs. Further, consistent with the legislation's tailoring requirements, the Board must proactively consider how firms with more than $250 billion in assets that do not qualify as G-SIBs may be more efficiently regulated by applying more tailored standards.

In conjunction with changing regulations, we also need to consider how such changes would be reflected in supervisory programs, guidance, and regulatory reporting. As supervisors, we need to balance providing appropriate relief to firms with ensuring that firms are maintaining resources and risk-management practices so they can be resilient under a range of conditions. We must also ensure we receive the right information in a timely manner so we can identify emerging risks.

A Path Forward

Banks with Assets between $100 Billion and $250 Billion

I want to spend the balance of my time focusing on the question I previously posed: On what basis will we decide to apply enhanced prudential standards to firms with total assets in the $100-billion to $250-billion range? The recent legislation directs the Board to consider factors other than size for differentiating supervision and regulation of large banking organizations.

Before I talk about the "other" factors, let me acknowledge the merits of size as one relevant factor to include on the list. We know that the effect of a large bank's failure on the economy is greater than when a smaller bank fails, even though the two banks might be engaged in similar business lines. The recent financial crisis in 2008-09 saw a much more severe recession than other financial crises, such as the savings and loan crisis in the late 1980s that preceded the relatively mild recession of 1990-91, and this appears at least in part related to the size of the affected institutions. In fact, empirical research done at the Federal Reserve shows that stress among larger banks does more significant harm to the economy than stress at smaller banks, even after controlling for the aggregate size of bank failures.2  We also know that larger banks are more operationally complex-even when not engaged in complex business lines-with a broader geographic scope and more layers of management than smaller banks. Therefore, it seems appropriate to me that a path forward for tailoring supervision and regulation of large banks should not ignore size, but consider it as one factor-although only one factor-along with other factors.

Before the enactment of the recent legislation, we had begun work on considering additional factors that capture large firms' degree of complexity and interconnectedness that--in conjunction with size--may provide a better basis for tailoring supervision and regulation than size alone. The systemic effect of a bank's failure or distress is positively correlated with that organization's business, operational, and structural complexity. Generally, the more complex a banking organization is, the greater the expense and time necessary to resolve it. Similarly, financial institutions may be interconnected in many ways, as large banks commonly engage in transactions with other financial institutions that give rise to a wide range of contractual obligations. Financial distress at a large bank may materially raise the likelihood of distress at other firms given the network of contractual obligations throughout the financial system.

So how do we gauge the degree of complexity and interconnectedness of large firms? Rather than formulating new and untested measures of these factors, I believe we would be well served to begin by looking to our body of post-crisis regulation as a source. I will highlight a few factors that already reside in various areas of our regulatory framework that I am considering.

The G-SIB surcharge indicators are likely to be a helpful source in this effort. For example, one factor from that framework that we might consider for purposes of tailoring is cross-border activity. This would measure assets and liabilities related to transactions with foreign banks, individuals, and companies, among others. This factor measures both a firm's complexity and resolvability, as foreign operations add operational complexity in normal times and complicate the ability of the firm to undergo an orderly resolution in times of stress.

Another factor from the G-SIB surcharge framework that could be useful is a firm's use of short-term wholesale funding, which may serve as a proxy for liquidity vulnerability. Historically, reliance on short-term, uninsured funding from sophisticated funding sources has created vulnerability to large-scale funding runs and increased risks related to financial sector interconnectedness. Specifically, this can lead to "fire sale" effects that may affect broader financial stability-which occurs when banks that fund long-term or illiquid assets with short-term deposits from financial intermediaries like pension funds and money market mutual funds need to rapidly sell less-liquid assets to maintain their operations.

Outside of the G-SIB surcharge framework, the Board has employed a measure of nonbank activities in certain rulemakings. These activities, which are conducted outside of a regulated depository institution, represent another source of complexity for large banks. For example, some nonbank entities engage in complex trading that is not permitted in depository institutions because of their risk.

Banks with Assets Greater Than $250 Billion

In addition to thinking about how we will tailor our regulation and supervisory programs for firms with assets between $100 billion and $250 billion, I believe the Board should also review the requirements applicable to the firms with more than $250 billion in total assets but below the G-SIB threshold. This review should ensure that our regulations continue to appropriately increase in stringency as the risk profiles of firms increase, consistent with our previously stated tailoring goals and the new legislation. The supervision and regulatory framework for these firms should reflect that there are material differences between those firms that qualify as U.S. G-SIBs and those that do not. For example, we know that non-G-SIBs with more than $250 billion in assets are generally less complex and less interconnected than U.S. G-SIBs and thus pose relatively less risk to financial stability. G-SIBs, on the other hand, have more complex activities, are more interconnected, and pose a far greater risk to financial stability should they fail. Yet at the moment, many aspects of our regulatory regime treat any bank with more than $250 billion in assets with the same stringency as a G-SIB. I believe there should be a clear differentiation.

Next Steps from the Federal Reserve

In my view, the Board should make it a near-term priority to issue a proposed rule concerning tailoring of enhanced prudential standards for large banking firms. This proposal, of course subject to notice and comment, would address our statutory obligations under the recent legislation by proposing to tailor enhanced prudential standards in a manner that recognizes relative complexity and interconnectedness among large banks. The statute sets an 18-month deadline for this regulatory process, but we can and will move much more rapidly than this.

In terms of capital requirements, both risk-based and leverage capital requirements should remain core components of regulation for large firms with more than $100 billion in total assets. Stress testing should continue to play an important role in assessing potential losses that large firms would suffer under a severely adverse economic scenario; the recent legislation recognized the importance of stress testing by requiring a supervisory stress test for these large firms. Therefore, the Board's proposed stress capital buffer, if finalized, would be critical for these firms.

However, we could consider a number of changes for less complex and less interconnected firms related to their capital requirements. For example, such firms, even if above $250 billion in assets, could have less frequent company-run stress tests. For those below $250 billion in assets, the statute requires supervisory stress tests to be conducted "periodically", which suggests the legislature wanted us to at least consider a rhythm other than annually. Additionally, less complex and less interconnected firms could be exempted from requirements to calculate risk-weighted assets under the models-based advanced approaches to capital.

I continue to strongly believe that liquidity regulation should be a primary component of supervision and regulation of large banks. We all saw the central role that liquidity risk played in the recent financial crisis. Minimum standardized liquidity measures and internal liquidity stress tests remain critical at firms with more than $100 billion in total assets. However, for less complex and less interconnected firms with assets greater than $100 billion, there may be opportunities to modify aspects of the standardized liquidity requirements as well as expectations around internal liquidity stress tests and liquidity risk management. Similarly, banks with more than $250 billion in assets that are not G-SIBs currently face largely the same liquidity regulation as G-SIBs. As I've said previously, I believe it would make sense to calibrate the liquidity requirements differently for these firms relative to their G-SIB counterparts.

Resolution planning is especially critical to ensure that the largest banking firms structure their operations in ways that make it more possible for them to be resolved upon failure without causing systemic risks for the broader economy. But most firms with total assets between $100 billion and $250 billion do not pose a high degree of resolvability risk, especially if they are less complex and less interconnected. Therefore, we should consider scaling back or removing entirely resolution planning requirements for most of the firms in that asset range. Further, we should consider limiting the scope of application of resolution planning requirements to only the largest, most complex, and most interconnected banking firms because their failure poses the greatest spillover risks to the broader economy. For firms that would still be subject to resolution planning requirements, we could reduce the frequency and burden of such requirements, perhaps by requiring more-targeted resolution plans.


In conclusion, I believe we have a unique opportunity to further tailor our supervision and regulation framework for large banks in a manner that allows us to be more risk-sensitive while still meeting our core goals of promoting safety and soundness and enhancing financial stability.

The recent legislation requires the Board to tailor its framework of supervision and regulation of large firms in a manner that continues to recognize size as one risk factor, but also more holistically incorporates other risk categories. In implementing this legislation, we should consider tailoring regulation further to take into account large banks' complexity and interconnectedness.

Of course, the details of how we implement a tailored framework will be subject to debate, and you can expect the Federal Reserve to be highly engaged in the public feedback process. Implementing the Economic Growth, Regulatory Relief, and Consumer Protection Act is a high priority for the Board, and we look forward to hearing the range of views as we make progress.

1  The views I express here are my own and not necessarily those of the Board of Governors of the Federal Reserve System.

2  Pending research to be published by Federal Reserve Board staff in the near future suggests that an increase in bank stress results in poorer economic performance in a way that scales with bank size. Put another way, staff estimate that stress at a single bank with $250 billion in deposits would result in roughly an 84 percent decline in quarterly GDP growth, while stress among five banks each with $50 billion in deposits would result in roughly a 24 percent decline in quarterly GDP growth. Although both scenarios have $250 billion in total failed deposits, the negative impact is greater when the largest bank is stressed.