Post-crisis bank resolution - what are the main challenges now?

Concluding remarks by Mr Agustín Carstens, General Manager of the BIS, at the 8th FSI-IADI conference on "Bank resolution, crisis management and deposit insurance", Basel, 2 February 2018.

Introduction

Ladies and Gentlemen

It is an honour for me to speak here today, at the end of an excellent FSI-IADI conference on bank resolution, crisis management and deposit insurance. You have already covered a lot of ground over the past three days, so let me be brief and offer some final thoughts.

This year's conference is very timely, as it follows the resolution of a few banks in the post-crisis regulatory environment. These cases teach us important lessons on how far the post-crisis reforms have enhanced loss absorbency and bank resolvability, including imposing losses on private stakeholders. They also show what remains to be done.

Since the Great Financial Crisis of 2007-09, a lot has changed in terms of regulation for financial institutions. The finalisation of the Basel III framework in December¹ last year is a landmark achievement for bank regulation, providing banks with certainty about the regulatory requirements applicable to them.

In the area of crisis management and resolution, efforts have focused on making banks resolvable and on moving away from bailouts. In order to address the too-big-to-fail problem, global regulators have developed frameworks for global systemically important financial institutions (G-SIFIs),² guidance on domestic systemically important banks (D-SIBs),³ the total loss-absorbing capacity (TLAC)⁴ standard and the Key Attributes of Effective Resolution Regimes for Financial Institutions (KA).⁵  National authorities have implemented these global standards in their jurisdictions, but not in all countries or in full.

With the bulk of regulatory changes broadly complete, the focus is now shifting to implementation.⁶  Some recent examples shed light on how far we have come, and what more needs to be done. Today's sessions covered some interesting cases from 2017. For instance, a major domestic bank was nationalised in Russia. While its failure had only limited system-wide consequences, the approach was closer to old-style bailouts, with significant public sector involvement. But as other cases discussed today show, even where bail-ins are part of the resolution framework, we still need to understand better how to operationalise them. In the successful resolution of a mid-sized Spanish bank, for example, some bail-in was included; but the strongest success factor was probably the existence of a credible buyer. In the liquidation of two mid-sized Italian banks, recourse to the domestic insolvency regime rather than the EU common resolution framework did not automatically prescribe bail-in.

Challenges

Implementing new regulation is always challenging, but especially so in the area of resolution. Let me elaborate. The Basel III framework is very broad, with many different elements. It will require consistent and timely transposition into national law, and very strong capacity building by both banks and supervisors.

But for the new rulebook on bank resolution, implementation is even harder. First, there is little real-life experience with the new resolution framework, which moreover has introduced major changes. Second, every resolution case is different, so learning-by-doing based on previous resolution experiences can only go so far. Third, cross-border cooperation is not only essential, but also particularly complex in crisis management, due to political sensitivities.

Against this background, the examples I mentioned earlier show that we still face a number of challenges before the instruments for resolving failing banks in an orderly way are fully operational. Let me highlight four key areas.

First of all, bail-in. The goal of the post-2007 reforms was to replace bailouts with bail-ins, and to make bailouts the exception in crisis management. However, this goal has not been reached yet, as the three examples above show. Why? The implementation of the tools for reaching this goal, TLAC and KA, is not yet complete in all countries, or consistent across countries. Where bail-in has been used, it has been subject to legal challenges. For now, bail-ins are also restricted by the limited stock of bail-inable debt, and market participants' understanding of the risks associated with it.

Making bail-ins work, without jeopardising financial stability, requires full and more consistent implementation of the relevant standards, clarity on cross-border implications, and higher loss-absorbing buffers in resolution.

The second major challenge is the resolution of cross-border banks. A precondition is to have resolution plans that are predictable, viable and supported by both home and (key) host authorities. The KA and "living wills" offer an internationally consistent blueprint for this, provided they have been implemented in all jurisdictions. But cross-border banks are inherently complex, so designing their resolution plans is not straightforward. Moreover, we have only an abstract idea of what may work - cross-border resolution still needs to stand the test of a real-life case. Beyond resolution plans, a panoply of ancillary conditions have to be worked out for resolution to work in a cross-border context. For instance, identifying the critical functions to be preserved in resolution; agreeing on triggers for entry into resolution; coordination on legal issues such as statutory stays on payments; and agreeing the amount and localisation of TLAC across home and host jurisdictions.

Resolving cross-border banks ultimately requires simplifying their structure enough that resolution can realistically be planned in advance, in a credible way, and that home and key host authorities have an agreed roadmap towards it, including the accessory legal provisions.

Two other challenges also deserve attention. As will not surprise you in the context of this conference, one is coordination among supervisors, resolution authorities, central bankers and deposit insurers. Coordination among many parties can be complex, especially across borders. For instance, for a number of global systemically important banks (G-SIBs), their home authorities have not yet put in place cooperation agreements.⁷

Developing solid cooperation agreements already in good times, including crisis simulation exercises, is key. Clear upfront coordination agreements also help to foster market confidence.

Finally, the challenge of the provision of liquidity in resolution. Certain resolution solutions, such as open-bank resolution, take longer than a weekend, and even the post-resolution "good bank" may not have access to market funding from the start.⁸  Funding can help maintain critical operations. But how can we set public backstops in this case, and would they be needed on top of bail-ins? The biggest challenge here is to ensure that retaining the flexibility of extending funding in resolution does not come at the cost of reintroducing moral hazard.

Conclusion

Let me conclude. This conference has provided an overview of the standards in crisis management and resolution as well as of the challenges so far and those to come.

A lot of ground has been covered, but we clearly need to road-test the new standards and requirements. Time will tell which ones work best.

In the meantime, implementation of the standards will continue to teach us important lessons, and it is key that authorities carry forward their cooperation, nationally and internationally, and further strengthen their resolution frameworks.

Events like today's conference convened by the IAID and the FSI can only help to support the development of best practices, so allow me to congratulate both bodies on the success of this occasion.