Skip to content

The end of too-big-to-fail?

The bank resolution and resolvability agenda moves on apace.

The Financial Stability Board (FSB) is set to meet in November in Brisbane. At the top of its crowded to-do list is ending too-big-to-fail by creating global standards for the resolution of global systemically important financial institutions (G-SIFIs). Most of the building blocks are now in place. Developed markets now have (or, in the case of some EU Member States, will have from 1 January 2015) bank resolution statutes which provide the statutory underpinning for the resolution of a failed bank (or, in many cases, banking group) through the creation of a bridge bank or asset management structure, or by the bail-in of bondholders. Resolution planning and strategising proceeds apace, based on the FSB's single point of entry (SPE) and multiple point of entry (MPE) concepts discussed in the October 2013 Risk Note. Crisis management groups exist for the G-SIFIs. Recent experience, such as the case of Banco Espirito Santo in Portugal, demonstrates that resolution can be made to work for substantial banks. So what is still missing?

Loss-absorbing capacity – what, how much and where

The "more capital" mantra that culminated in Basel III has not gone away (the FSB is conducting a study on yet higher capital charges), but been supplemented by a call for more "loss-absorbing capacity". To the layman, the difference between the two may not be clear, but in an environment in which senior debt can be bailed in (written down or converted to equity) in resolution, such debt can be loss-absorbing – effectively a form of "tier 3" capital. The FSB will provide guidance on how much total loss-absorbing capacity (regulatory capital and bail-in debt) will be needed by G-SIFIs – this is likely to be in the order of 16%-20% of risk-weighted assets or double the leverage ratio requirement, whichever is higher. There will also be guidance on how loss-absorbing capacity should be distributed within banking groups. This ties in with discussions on the SPE resolution strategy and the allocation of rights and responsibilities between home and host state resolution authorities: the message will likely be that (as with regulatory capital) bail-in debt should be raised at the group level and downstreamed (and "trapped") at operating bank level within groups to preserve host state resolution authorities' ability to effect local resolution.

The guidance is likely to be followed by national regulators for domestically systemic banks. For major banks, this will fire the starting gun on raising loss-absorbing capacity prior to implementation (expected to be 2019). Holders of bank debt will need to reassess the risk profile of their holdings and pricing expectations in relation to new issuance, given the expectation that they will be functionally subordinated in resolution through the application of bail-in powers (a point reinforced by the forthcoming introduction of depositor preference in Europe). Legislators and regulators will need to bring their regulatory frameworks in line with the guidance – meaning the EU Bank Recovery and Resolution Directive (BRRD) will be outdated before it is implemented.

Cross-border recognition

The principal legal impediment to resolution remains cross-border recognition. Conceptually, this is a straightforward legal problem, which arises wherever a bank or group which needs resolving has a cross-border relationship. Consider a UK bank with a branch in Dubai. For it to be resolved successfully, the resolution proceeding in the UK (home state) needs to be recognised and upheld in Dubai (host state), as separate insolvency proceedings in Dubai are likely to frustrate effective resolution. If the UK bank has New York law-governed debt securities, the proceeding is also likely to involve a breach of their terms (by bailing wholesale creditors in to restore the solvency of the bank, or by transferring assets and liabilities from the bank to a bridge bank). So recognition is needed in New York, too – or else the debt will need amendment to provide recognition contractually. If the proceeding is not recognised then the bank may not resolve as intended, which could result in disorderly failure. The logical regulatory response is to ensure that resolution can be recognised in the host state, or restrict the activity so that it does not impair resolvability. Discretionary recognition will not be enough for this purpose, as the resolution authority will need certainty either way.

Considerable attention has been paid to ISDA master agreements, with a project ongoing to produce contractual submission to resolution proceedings in master agreements to which banks are party, with the intent that any solution adopted be carried forward into other financial agreements. ISDAs are merely the tip of the iceberg: the BRRD will require all non-EU liabilities of EU banks to include contractual recognition language from the beginning of 2016. But contractual recognition itself just scratches the surface of the issue. It arises in multiple contexts: foreign governing law of contracts and debt; foreign counterparties; foreign assets and liabilities; and foreign branches all present potential recognition impediments.

The lawyer's answer to these issues is easy: create a legal recognition regime, and the issue disappears. The BRRD achieves this inside the EU. Outside it, the political will to permit recognition remains low, however. Resolution powers enable the resolution authority to allocate losses around a failed institution (or group) so as to create winners (the transferees to the bridge bank, or creditors excluded from a bail-in) and losers. A "hard" (mandatory) recognition regime risks permitting a foreign resolution authority to prefer its local creditors over foreign creditors. Outsourcing the resolution of a major bank to a foreign sovereign while in the grip of a financial crisis may not seem a sensible plan, but it is the key concept that underpins SPE resolution. The public sector needs to decide whether it can live with the structure.

Other workstreams

The interaction between banks and systemic non-bank organisations – principally CCPs and insurers – remains a work-in-progress. The dovetailing of regulation of CCPs and other financial market infrastructure with bank resolution is outstanding (currently the rules of most CCPs could frustrate resolution proceedings), as is the vexatious question of how CCPs should be resolved. On a brighter note for the financial industry, calls for impact analysis on the flood of regulation now seem to be receiving a more favourable hearing, although conducting impact analysis on this topic itself looks highly complex.

A final area of change which arguably conflicts with the remainder of the resolution agenda is ring-fencing. An update on progress in the UK is set out at page 48. Much of the ring-fencing agenda runs contrary in spirit to the work of the FSB, as it presupposes standalone ring-fenced banks being run at arm's length and isolated from risk in the remainder of the group. This provides a timely reminder that it would be unwise to expect a joined-up approach to bank regulatory reform any time soon.

Legal and Regulatory Risk Note
United Kingdom