Board of Governors of the Federal Reserve System

Thank you to Professor Scott for inviting me to join this discussion on cross-border resolution and risks of fragmentation. Like many of my international counterparts in the audience, I maintain a deep commitment to cross-border banking and efficient movement of capital and liquidity, which are important contributors to long-term economic growth. And it is with that commitment in mind that I have considered the topic of today's conference: "ring-fencing," beginning with some reflections on what the term means. Many here use the term to describe local capital and liquidity requirements, which are imposed ex ante on local subsidiaries and designed to protect those entities and their creditors from losses. The term is also used to refer to disruptive actions taken by host regulators to seize assets in the moment of crisis.¹ This type of ring-fencing occurs suddenly and unilaterally.

Both uses of the term are associated with the risk arising from the stress or failure of a global financial institution; however, whether ring-fencing as I first defined it--prepositioning--is helpful or harmful in minimizing this risk depends on one's perspective. The views of different stakeholders tend to vary depending on whether one is seeking to maximize efficient allocation of resources in good times or minimize losses in stress and, importantly, whether one is a home or host regulator.

Before the financial crisis, much of our collective orientation was on maximizing the efficient flow of capital across the globe. This should remain a paramount goal. Yet in the wake of the financial crisis, global regulators have understandably also focused on minimizing the cost of the failure of a global financial institution by mitigating the impediments to cross-border resolution. The single-point-of-entry (SPOE) and bail-in concepts hold particular promise for most large global firms. However, a successful SPOE resolution of a large global firm has not yet been attempted and will require close cooperation among a large number of stakeholders, including both home and host country regulators. This cooperation will be based on an understanding of separate and mutual interests, not on trust alone. So while SPOE creates a potentially workable framework for resolution, setting the conditions for cooperation is critical. I grew up among the ranches of the American West, where we lived by the motto taught to me as a young child: trust everyone, but brand your cattle. This is a theme that will run throughout my remarks today.

In addition to setting the stage for effective cooperation, I will focus specifically on the type of ring-fencing that will almost certainly undermine the successful execution of an SPOE resolution--namely the disruptive seizure of assets by host regulators in the moment of stress. As with other elements of our regime, I have been considering whether our current prepositioning requirements for domestic and foreign firms operating in the United States are both minimizing this risk and functioning in an efficient and transparent manner, and I will share some thoughts in that regard. Our vantage point in the United States as a large home and host regulator would counsel that it is sensible to find a middle ground and fine tune our approach as we learn more and global conditions evolve.

Resolution Considerations of a Home and Host Regulator
To enable cooperation and avoid a destabilizing seizure of assets by host regulators, I would submit that all jurisdictions must find a balance of flexibility for the parent bank and certainty for local stakeholders. Flexibility, or the ability to allocate capital and liquidity to different parts of the group on an as-needed basis, helps to meet unexpected demands on resources and reduces the risk of misallocation and inefficient use of resources. Certainty, or the local prepositioning of capital and liquidity to ensure a firm can satisfy local claimants under stressful conditions, helps to promote cooperation in the context of a cross-border resolution and avoid incentives for more drastic action by host authorities.

One's assessment of the optimal balance, as I alluded to earlier, can depend significantly on where one sits in the regulatory constellation. The home regulator, by nature, will logically prefer flexibility in a resolution; consolidated capital and liquidity requirements are most effective if resources can be freely allocated around the consolidated firm where and when they are needed. Flexibility also helps offset the uncertainty in forecasting the location within the consolidated firm where stress may arise. Yet I would also argue that home regulators should recognize host jurisdictions may take action to restrict the flow of resources, or worse yet, demand resources in the moment of crisis, even if the stress does not originally emanate from their location. Such actions tend to both limit the flexibility of the home regulator and undermine cooperation in times of stress.

The host regulator, by nature, will prefer the certainty that resources will be available to satisfy local customers and counterparties under stressful conditions. This is particularly the case if the risk of default or the potential local loss given default of a foreign firm is high, and the tolerance for a government-assisted intervention for foreign banks is low. However, the host regulator should also recognize that it is ultimately in its interest for the SPOE resolution of the foreign bank to be successful and, given the uncertainty of the circumstances or location of losses that emerge in an actual stress, adequate flexibility for the parent to deploy resources where needed is likewise in the host regulator's interest.

A global bank has other stakeholders who have preferences regarding flexibility or certainty and can take actions that can potentially destabilize the firm. For instance, parent company debt and equity holders may prefer less prepositioning, while local creditors and financial market utilities may prefer more. These other stakeholders can act in ways that can be destabilizing. Building a system that is transparent and is perceived by stakeholders as allocating losses fairly is key in this regard.

Finally, I would like to note that the considerations of policymakers in determining the right balance of flexibility and certainty may differ depending on whether the resource at issue is capital or liquidity. In resolution, the most important difference between capital and liquidity is the speed with which financial stress can appear. Liquidity needs are sudden and tend to manifest in all areas of the organization, and the consequences of not meeting liquidity demand--an immediate default on an obligation--can be grave. Capital needs, however, may be more localized and slower to evolve but are foundational to the execution of an SPOE resolution. It is unlikely that host regulators would be comfortable cooperating in an SPOE resolution strategy without some confidence in the viability of the entities in their jurisdictions.

U.S. Approach
Historically, the United States and the United Kingdom have been in a unique position of having large interests as both home country and host country regulators of internationally active banks. Soon, the European Union is likely to assume this privilege as well. We understand that any requirements we impose on foreign banks operating in the United States may well be imposed on U.S. firms operating abroad. In addition, we are operating under a veil of ignorance, as we don't know whether the next firm in distress will be a U.S. firm operating globally or a foreign firm with U.S. operations. This provides us with strong incentives to view the risks from both sides.

As the home regulator to U.S. global systemically important banking organizations (G-SIBs), we and the Federal Deposit Insurance Corporation (FDIC) have used the living wills process to set the expectation that a firm appropriately balance prepositioned and centrally managed resources.² This expectation is based on the premise that the optimal balance will depend on factors such as the firm's structure and the host jurisdictions in which the firm operates. As such, the Board and the FDIC have asked U.S. G-SIBs to analyze and anticipate capital and liquidity resources needed to ensure the continued operation of material entities in resolution. Regarding capital, the positioning of a U.S. G-SIB's internal total loss absorbing capacity (TLAC) should reflect a balance of certainty--prepositioning internal TLAC directly at material entities--and flexibility--holding recapitalization resources at the parent, known as contributable resources--to meet unanticipated losses at material entities.³

Regarding liquidity, the Board and FDIC expect a U.S. G-SIB to appropriately estimate and maintain sufficient liquidity for material entities, an expectation known as Resolution Liquidity Adequacy and Positioning, or RLAP.⁴ RLAP expectations are intended to be designed so that liquidity is not "double counted" among home and host jurisdictions, to provide transparency into the location of liquidity across the firm's material entities, and to ensure that liquidity can flow where needed with minimal potential disruption. The RLAP approach is aimed at ensuring that surpluses in one host jurisdiction generally are not relied upon to meet deficits in another host jurisdiction, given the confusion and vulnerabilities such reliance can cause in an actual stress. Specifically, a firm should be able to measure the stand-alone liquidity position of each material entity and ensure that liquidity is readily available either at the parent or at that entity to meet any deficits.⁵ As with capital, firms are expected to have a balance of prepositioned and centrally managed liquidity--specifically, by balancing the certainty associated with holding liquidity directly at material entities against the flexibility provided by holding high-quality liquid assets at the parent available to meet unanticipated outflows at material entities.

As a host regulator, our approach to local capital and liquidity regulations of foreign banks with large U.S. operations is motivated by the lessons of the recent financial crisis, where many foreign banks operating in the United States suffered severe stress and survived only with extraordinary support from the United States and their home country governments. In addition to increasing the resiliency of the U.S. operations of foreign banks, our approach also reflects both resolution and competitive equity considerations.

From a resolution perspective, our rules seek to ensure that there are sufficient resources in the United States today to ensure that we, as a host regulator, are well positioned to cooperate with a home country authority in the event a firm experiences material stress or failure. Our rules also ensure that we do not have a strong incentive to limit flows or seek additional resources in the moment of crisis, which could be highly destabilizing in a stress event.

From a competitive equity standpoint, we believe that U.S. subsidiaries of foreign banks should operate on a level playing field with their domestic counterparts. This is generally consistent with the long-standing treatment of large and complex bank and nonbank subsidiaries around the world.

As such, our rules subject a large foreign bank to the same capital and liquidity requirements as domestic bank holding companies by requiring the foreign bank to hold its U.S. subsidiaries through a U.S. intermediate holding company (IHC) and imposing capital and liquidity requirements to the IHC. At the same time, we have adjusted our approach for the U.S. branches of a foreign bank with a large U.S. presence in recognition that branches are subject to a narrower set of permissible activities and operate as a direct extension of the parent bank. For these U.S. branches, we have imposed local liquidity requirements in light of the liquidity vulnerabilities that many U.S. branches of foreign banks experienced in the crisis, but no separate capital requirements.

For IHCs that are subsidiaries of foreign-owned G-SIBs, the Federal Reserve requires such a firm to issue a minimum amount of loss-absorbing instruments to its foreign parent, known as internal TLAC, including a minimum amount of unsecured long-term debt.⁶ In the event that an IHC was experiencing significant financial distress, the internal TLAC could be used to replenish the IHC's equity and maintain its solvency. The U.S. implementation of internal TLAC is modeled on the internal TLAC framework developed by the Financial Stability Board (FSB), which includes a calibration of the amount of loss-absorbing resources that should be prepositioned in a given jurisdiction.⁷ Specifically, the FSB contemplates that internal TLAC requirements of a subsidiary of a foreign bank expected to be resolved through SPOE would be calibrated at 75 to 90 percent of the external TLAC requirement that would apply to the subsidiary if it were to be separately resolved. In implementing the TLAC standards in the United States, the Board calibrated the internal TLAC requirement for IHCs of foreign-owned G-SIBs at the high end of the FSB range, at around 90 percent.

Adjustments to the Current Regime
There are two principal points I have been making today. The first is that some amount of local capital and liquidity prepositioning can reduce the incentives for damaging and unpredictable seizures of resources by local regulators during times of stress--thus actually reducing the likelihood that improvised, beggar-thy-neighbor ring-fencing would frustrate completion of a successful SPOE resolution in the future. As we learned long ago out West, the branding of cattle creates the possibility of trust.

The second point, however, is equally important: the best prepositioning structure is not an eternal verity mathematically deducible from first principles, but it is instead a practical balance designed to promote cooperation among humans, and any such balance is likely to be improvable with experience, reflection, and debate. We are interested in views from the firms and the public on how the regimes can be improved, and we expect to invite public comment on our living will guidance for U.S. and foreign firms in the near future. In addition, we are currently weighing the costs and benefits of our current approach of directing firms to determine the appropriate amount of prepositioned capital and liquidity. We are also considering whether formalizing resolution capital and liquidity requirements through a rulemaking process would improve the predictability and transparency of our approach.

We continue to believe that the IHC and attendant requirements are appropriate for foreign banks with large U.S. operations. However, in light of our experience with these structures, I believe we should consider whether the internal TLAC calibration for IHCs could be adjusted to reflect the practice of other regulators without adversely affecting resolvability and U.S. financial stability. The current calibration is at the top end of the scale set forth by the FSB, and willingness by the United States to reconsider its calibration may prompt other jurisdictions to do the same, which could better the prospects of successful resolution for both foreign G-SIBs operating in the United States, and for U.S. G-SIBs operating abroad. Alternatively, it may be possible to streamline the elements of our resolution loss absorbency regime, which include both TLAC and long-term debt requirements. I will be recommending to my colleagues that we look closely at these possibilities in the coming weeks and seek comment on ways to further improve this framework.

Conclusion
We are committed to working with other jurisdictions to continue to build the foundation of the SPOE resolution framework. In addition to finding the appropriate balance of flexibility and certainty that I have discussed, we continue to advocate for increasing the standardization in the global implementation of the regulatory capital rules, improving host supervisors' transparency into the global liquidity and capital positions of a G-SIB on a consolidated and deconsolidated basis, and addressing impediments to a successful SPOE resolution. As with all regulations, we will be open to considering adjustments that would improve transparency and efficiency and will continue to reassess our regime as we make advancements in developing the cross-border resolution framework.

---

1. "Home regulator" refers to the prudential regulatory agency of the parent bank, and "host regulator" refers to the prudential regulatory agency of a foreign branch or subsidiary of the parent bank. Return to text

2. Section 165(d) of the Dodd-Frank Wall Street Reform and Consumer Protection Act requires a large firm to submit an annual resolution plan, or living will, that describes the company's strategy for rapid and orderly resolution under bankruptcy in the event of material financial distress or failure of the company. If the Federal Reserve and the FDIC determine that a firm's resolution plan is not credible or would not facilitate an orderly resolution under the U.S. Bankruptcy Code, the agencies, acting jointly, may impose more stringent prudential requirements on the firm until it remediates identified deficiencies. Return to text

3. Guidance for 2017 section 165(d) Annual Resolution Plan Submissions by Domestic Covered Companies that Submitted Resolution Plans in July 2015, available at www.federalreserve.gov/newsevents/pressreleases/files/bcreg20160413a1.pdf. Return to text

4. Id. Return to text

5. The liquidity position of an entity reflects high-quality liquid assets at the material entity less stressed net outflows to third parties and affiliates. Return to text

6. 12 CFR part 252, subpart G. Return to text

7. Financial Stability Board, "Principles on Loss-Absorbing and Recapitalisation of G-SIBs in Resolution" and "Total Loss-absorbing Capacity (TLAC) Term Sheet" (November 9, 2015), available at www.fsb.org/wp-content/uploads/TLAC-Principles-and-Term-Sheet-for-publication-final.pdf. Return to text