Developing-country interests sidestepped in FSB proposals

Andrew Cornford runs the rule over proposals advanced by the Financial Stability Board concerning resolution procedures for and loss-absorption capacities of global systemically important banks.

__________________________________________________________________________________________

The proposals of the Financial Stability Board (FSB) on measures to end “Too-Big-To-Fail” (TBTF), through procedures for resolution of banks and strengthening balance sheets to protect taxpayers in respect of the global systemically important banks (G-SIBs), do not appear to cover or meet the need to counter systemic risks of developing countries and their institutions in cross-border insolvencies involving other SIBs.

One outcome could be that countries might decide to ensure subsidiarization as the corporate form for cross-border banks, since this choice may be more protective of a country’s interests owing to the way in which the losses of an insolvent cross-border banking group are assumed in the jurisdictions in which loss-making entities are legally incorporated.

Background

At the Pittsburgh Summit of the G20 major economies grouping in 2009, there was agreement that the FSB should propose measures addressing the risks associated with the operations and insolvency of systemically important financial institutions (SIFIs).

Amongst such institutions, special attention was directed at systemically important banks (SIBs) owing to the damage which the failure of such banks can inflict on economies and to the implicit subsidy which is entailed by the likelihood that in the event of insolvency or the threat of insolvency governments will have no option but to provide public funds to control such damage and the financial instability which it can cause.

The underlying rationale of the regulation of banks is to enable them to support productive economic activities, to prevent instability and crises with their social and economic costs, and – in the event of a banking crisis – to mitigate the risks and severity of systemic failure. This role holds for SIBs as well as for banks which are not classified as systemically important.

But SIBs have features which give the pursuit of these objectives by policymakers and regulators special importance. These features are their size and their inter-relationships with other financial and non-financial entities. They have taken on special importance in a world of cross-border banking. The cross-border dimension has complicated the regulation of SIBs since the regulations of more than one national regime are involved.

In a report of September 2013 (FSB, “Progress and Next Steps Towards Ending ‘Too-Big-To-Fail’”, Report of the Financial Stability Board to the G20, 2 September 2013), the FSB reviewed progress made since regulators had to confront the daunting lack of coherent planning, preparation and frameworks for dealing with the widespread post-2008 distress and insolvency amongst major banks.

The complexity of the resulting problems can be grasped from data concerning Lehman Brothers Holdings at the time of its bankruptcy in September 2008. The group had over 2,000 affiliates in a single integrated enterprise in over 40 countries (H.S. Scott and A. Gelpern, International Finance: Transactions, Policy and Regulation, nineteenth edition, Thomson Reuters/Foundation Press: 319). The failure led to a large number of judicial and administrative proceedings in many different jurisdictions. Moreover, it triggered default clauses in derivatives contracts which allowed Lehman’s counterparties the option of terminating contracts and seizing collateral, actions which were legal but complicated the task of administering the insolvency. The United States bankruptcy procedures for Lehman lasted three years, the plan for paying the firms’ creditors being confirmed only in December 2011.

Progress and outstanding issues

At the global multilateral level, the progress reviewed in the FSB’s report of September 2013 included a number of steps: the designation of  28 G-SIBs to which higher capital requirements will apply; recommendations as to tighter procedures for risk management and risk reporting; and agreement on measures regarding the improvement of the infrastructure for financial markets such as central counterparties for derivatives to reduce the risk of contagion throughout the financial system due to defaults by individual counterparties. (Central counterparties mutualize credit risks associated with contracts between bilateral counterparties by assuming one side of, and guaranteeing performance of, all open contracts.)

The report also drew attention to ongoing initiatives at the national or regional (EU) levels directed at structural reforms for the banking sector. Important issues which had still to be fully addressed were service-level agreements which assure the continuity of crucial services where chains of transactions involve insolvent firms and the staying of early termination of financial contracts in insolvencies which, although legal, can hinder the implementation of resolution strategies for cross-border banks.

Owing to their complexity and political sensitivity, two subjects considered by the FSB to be of critical importance to the regulation and resolution in insolvency of G-SIBs are requiring a predictably long process for the development of rules capable of achieving consensus among the G20’s membership: (1) resolution strategies and procedures for G-SIBs at the institution level, and (2) additional resources for the absorption of losses by a G-SIB sufficient to avoid a bailout with public funds.

The FSB’s recent work on these two subjects is the principal focus of what follows. Although the focus of the FSB’s work is on resolution and loss absorption, the proposals would also be expected to reduce the likelihood of distress and insolvency.

Key features of resolution

The essential elements of the FSB’s agenda regarding resolution were published in an FSB report of 2011 (FSB, “Key Attributes of Effective Resolution Regimes for Financial Institutions”, October 2011). The document, henceforth referred as the Key Attributes, was updated in October 2014.

The recommendations of the Key Attributes cover not only banks’ resolution strategies and procedures but also features of appropriate legal and institutional regimes within which the strategies and procedures can be successfully implemented. Thus, the Key Attributes cover the following: the powers of resolution authorities; the setting-off and collateralization of banks’ positions, and the segregation (and thus the protection) of individuals’ assets; safeguards of the interests of creditors; the funding of firms in resolution to avoid reliance on public ownership and public bailout funds; the legal framework for cross-border cooperation between resolution authorities in different jurisdictions; the establishment of crisis management groups for cross-border SIBs; institution-specific cross-border cooperation agreements; regular assessments by resolution authorities of the feasibility and credibility of G-SIBs’ resolution strategies; ongoing processes of planning for resolution and possibly recovery; and ensuring that there are no legal, regulatory or policy impediments to the cross-border exchange of necessary information between resolution authorities in different jurisdictions.

Alternative approaches to resolution

The first of the subjects discussed in this article starts from the choice of single point of entry (SPE) versus multiple point of entry (MPE) for resolution of G-SIBs. Resolution strategy thus involves a bank’s corporate form and its business model or strategy. The second subject (described in the following section) concerns the size and form of loss-absorbing capacity beyond the capital requirements additional to those of Basel III which G-SIBs must already meet.

Under SPE resolution (which generally presupposes a centralized model for a cross-border bank with a group consisting of branches), the authority of the parent institution manages the overall process of resolution. Under MPE resolution, the host supervisors of the jurisdictions where there are subsidiaries play key roles in resolution.

Although in the Key Attributes the FSB devotes greater attention to the mechanics of the SPE process, it avoids explicitly taking a position in favour of the SPE over the MPE process, thus acknowledging that the corporate form in which banking entities are given market access to different jurisdictions should allow for the policy choice of subsidiarization owing to the perceived need for national regulatory control.

Supporters of SPE resolution draw attention to what they consider to be the advantages of branches (the corporate form usually associated with this approach) from the point of view of both operational efficiency and the facilitation of resolution. The branch-based model is viewed as facilitating the flow of capital and liquidity in accordance with directives from an integrated organizational and risk management. Unsurprisingly, the managerial autonomy provided by this model means that it is favoured by major international banks and lobbies such as the Institute of International Finance.

Resolution, supporters of this model maintain, is facilitated by the SPE approach owing to the key role played by the authority of the parent entity. Thanks to the support of the parent entity all the constituent companies should remain solvent, their liquidity is assured by intra-group financial flows, and the critical functions of the group (i.e., systemically important financial services such as payments, settlement and clearing functions) continue to operate. Moreover, the SPE approach based on cross-border branching maximizes the likelihood of international cooperation.

As the Institute of International Finance, a lobby group for large banks, puts it in a recent report, “By requiring the local operation to subsidiarize, the host authority does extend its powers over it. However, the cost of this subsidiarization is that local creditors are potentially deprived of assets other than those of the local subsidiary” (Institute of International Finance, “Achieving Bank Resolution in Practice: Are We Nearly There Yet?”, Washington, DC, September 2014: 15).

The trouble with this portrayal of the SPE approach’s potential is that it is idealized. Such smooth cooperation between the entities – branches and parent institutions – of major groups does not correspond to much recent experience, where many of the steps actually taken were belated responses to crisis and required extensive intervention by national regulators.

Recent moves towards greater reliance on subsidiarization in several jurisdictions indicate that major national regulators are not convinced of being able to count on a favourable outcome for the SPE approach. According to this view, the high level of mutual trust required for the success of reliance on branch-based models during banking crises often does not exist. Moreover, clear legal obligations on parent institutions to holders of the liabilities of its branches are lacking.

The International Monetary Fund (IMF), while acknowledging the cost advantages of cross-border branching for some categories of banking group (in particular for those with primarily wholesale operations), has nonetheless drawn attention to the advantages of the subsidiary structure for the purpose of crisis management and resolution during banking crises (J. Fiechter, I. Otker-Robe, A. Ilyina, M. Hau, A. Santos and J. Surti, “Subsidiaries or Branches: Does One Size Fit All?”, IMF Staff Discussion Note, 7 March 2011: Chapter II).

In the IMF’s view, the ideal solution for tensions in cross-border banking relations consists of joint arrangements by the supervisors of parent and host countries, harmonized cross-border resolution regimes, and effective arrangements for burden sharing in stressed and crisis conditions. These are features which its supporters claim for the SPE approach. However, as the IMF notes, practical difficulties of cross-border cooperation during crises have led to the exploration by policymakers in several countries of the benefits of regulatory self-sufficiency of cross-border banks’ local operations, regardless of whether the institution’s business model is wholesale or retail banking.

Host countries’ authorities can ensure that local subsidiaries have sufficient capital and liquidity in the country, thus minimizing the possibility of risks to financial stability being imported from related banking entities abroad in distress and increasing the resources at the authorities’ disposal in cross-border insolvencies – if necessary through recourse to ring-fencing local banking entities.

The reduced freedom of management as to the location of capital and liquidity associated with subsidiarization can thus actually be an advantage and not a disadvantage as supporters of the branching model argue. Moreover, subsidiarization facilitates the selling-off by resolution authorities of a cross-border bank’s local constituent entities and the implementation of “living wills”, i.e., the recovery and resolution plans for the winding-down of systemically important financial groups if they fail.

Increasing loss-absorption capacity

Increasing the loss-absorption capacity of G-SIBs is the subject of a FSB consultative document of November 2014 containing detailed proposals (FSB, “Adequacy of loss-absorbing capacity of global systemically important banks in resolution”, Consultative Document, 10 November 2014). Noting that the FSB’s proposals would bail in – i.e., impose losses on – certain categories of a bank’s liabilities as part of increasing loss-absorbing capacity, some commentators apprehensively – but mistakenly in the context of the FSB’s plans – recalled losses imposed on depositors as part of measures taken in response to the economic and financial crisis in Cyprus in 2013.

This crisis followed Cyprus’s membership of the EU from 2004 and of the monetary union from 2008. These were accompanied by a doubling of the size of the financial sector between 2006 and 2011. The financial bubble, which included pouring of money into high-risk ventures abroad and extensive exposure to the Greek banking system, was eventually followed by collapse. The response of the Cypriot government, supported by EU finance ministers, initially included a tax on bank deposits including those of smaller savers. Amidst a storm of protest, which was not limited to Cyprus, the tax was rescinded in favour of the writing-down of large deposits through measures such as their conversion into equity. These policy interventions came with the price of a public-relations blow to belief in the safety of deposits in EU banks (D. Marsh, Europe’s Deadlock: How the Crisis Could Be Solved – and Why It Won’t Happen, New Haven and London, Yale University Press, 2013: 53-57).

The FSB’s proposals on loss absorption (which, as noted, are still at the consultative stage and are directed at TBTF banks) would not lead to the writing-down of retail deposits. Indeed, as explained below, one of the aims of the proposals is to maintain the banking system’s capacity for carrying out critical functions (defined earlier), which would not be compatible with a major disruption of retail banking.

This implies only that depositors are protected from losses in their role as depositors. To the extent that the resolution procedures and the loss-absorbing capacity for G-SIBs introduced in response to the FSB’s proposals do not cover losses in insolvencies, the losses would still be met by much the same group in their role as taxpayers. But the distribution of these losses would be for governments to decide.

In its report of September 2013, the FSB committed itself, in consultation with standard-setting bodies, to the preparation of proposals on the adequacy of the loss-absorbing capacity of G-SIBs in resolution (what it calls “gone concern loss absorbing capacity” as distinct from the “going-concern capacity” required to keep it out of insolvency).

The implication of this commitment was that the FSB was sceptical that the supplementary capital requirements prescribed for G-SIBs in Basel III would be sufficient to guarantee a resolution strategy which “mitigates risks to financial stability, ensures continuity of critical functions and minimizes taxpayers’ exposures to losses”, as the General Manger of the Bank for International Settlements has put it (J. Caruana, “How much capital is enough?”, address to the IESE Business School conference on “Challenges for the future of banking, regulation, supervision and the structure of banking”, London, 26 November 2014).

The FSB document of November 2014 covers proposals concerning the amount of total loss absorption capacity (TLAC) required, arrangements which will ensure its availability in the resolution of cross-border banking groups, the eligibility of financial instruments for inclusion in TLAC, the interaction of TLAC with pre-existing capital requirements, transparency, and restrictions on holdings of TLAC to avoid contagion risks if the holdings are exposed to loss during resolution.

The proposed minimum Pillar 1 TLAC requirement is in the form of a range of 16-20% of risk-weighted assets (RWA). The references to Pillar 1 and RWA are from Basel III itself and refer to a bank’s exposures to credit risk (through its RWA) under different headings (loans, mortgages, off-balance-sheet commitments, etc.), as estimated in accordance with Basel III’s rules, and to the minimum capital requirements corresponding to a bank’s total exposure. Pillar 2 TLAC requirements consist of TLAC above the Pillar 1 minimum which is set in accordance with discretionary rules of national regulators.

Under the FSB’s proposal, TLAC will include the principal capital instruments which count towards the fulfilment of Basel III minima but not the supplementary capital buffers (conservation, countercyclical, and designed to produce additional loss-absorbency for a set of G-SIBs designated by the FSB). The proposal also includes rules to ensure that all the entities in the banking group (parent holding company or companies and subsidiaries) will have adequate TLAC.

Instruments eligible for inclusion in TLAC must meet various legal conditions such as the following: the authorities must possess the legal powers to expose the TLAC-eligible liabilities to loss without the risk of a successful legal challenge or the possibility of compensation costs; and the authorities must be confident that holders of TLAC instruments are able to absorb losses during periods of stress in financial markets without spreading contagion or without the necessity of the allocation of losses to liabilities where the results could be the disruption of critical financial functions or significant financial instability.

Avoidance of contagion and of the disruption of critical financial functions implies, for example, that eligible TLAC must not include insured deposits, liabilities callable on demand without supervisory approval, and tax liabilities. Limitation of contagion is to be achieved through appropriate prudential restrictions such as dis-incentivizing the holding of TLAC issued by other G-SIBs. Meeting these conditions will require clarity about priorities in the order in which TLAC instruments will absorb losses in a resolution, and thus disclosure to holders and potential holders.

Concluding comments

This article has been principally concerned with only two of the many issues covered in the FSB’s coverage of measures to end TBTF. Procedures for the resolution of banks and the strengthening of their balance sheets to protect taxpayers in bank insolvencies are not of course limited to the banks classified as G-SIBs by the FSB, even though the systemic risks of failure on the part of these banks are particularly great.

As a result of the restrictions in the coverage of the FSB’s list, with the exception of a small number of Chinese banks (Agricultural Bank of China, Bank of China, Industrial and Commercial Bank of China Limited), institutions with parents in emerging market and developing economies (EMDEs) are not included, though of course the failure of an FSB G-SIB would be likely to have ripple effects in EMDEs.

Limitations on the direct relevance of the FSB’s proposals to EMDEs are acknowledged in the FSB’s proposals. Indeed, G-SIBs headquartered in such markets will not initially be subject to the Common Pillar Minimum TLAC requirement (FSB, “Adequacy of loss-absorbing capacity of global systemically important banks in resolution”: 14). This makes sense since the Chinese banks listed in November 2014 by the FSB among G-SIBs (Agricultural Bank of China, Bank of China, Industrial and Commercial Bank of China Limited) are state-owned so that their liabilities are not held by parties whose bailing-in will contribute to avoiding losses to taxpayers (who would anyway have to meet the costs of such a bank’s distress or failure).

Many of the guidelines proposed for G-SIBs would in fact be relevant in cross-border insolvencies of smaller banks from both EMDEs and advanced economies. In its report of September 2013 (FSB, “Progress and Next Steps Towards Ending ‘Too-Big-To-Fail’”: 16), the FSB also refers to the framework for the TBTF problem for D(domestic)-SIBs developed by the Basel Committee on Banking Supervision (BCBS, “A framework for dealing with domestic systemically important banks”, October 2012), whose rules are pertinent for banking sectors in EMDEs.

This framework would be “based on the assessment conducted by the local authorities, who are best placed to evaluate the impact of failure on the local financial system and the local economy”. An implication of this point is that “in order to accommodate the structural characteristics of individual jurisdictions, the assessment and application of policy tools should allow for an appropriate degree of national discretion”. While the framework for D-SIBs does cover higher loss absorbency, it does not include resolution procedures.

A consequence of the FSB’s focus on TBTF institutions in the documents discussed in this article is that its treatment of systemic risk is limited to that which results from the many forms of economic inter-relatedness of such institutions with other banks and with non-financial firms. But systemic risk is not limited to risks due to such inter-relatedness. A dissent to the report on the financial crisis of the United States Financial Inquiry Commission (National Commission on the Causes of the Financial and Economic Crisis in the United States, Final Report: 433) notes that systemic risk is also associated with situations characterized by the effects of correlated failures: “In a common shock, the failure of one firm may inform us about the breadth or depth of the problem, but failure of one firm does not cause the failure of another.” Correlated failures, such as those due to external payments-cum-banking crises, may frequently be more important sources of systemic risk in EMDEs than the inter-relatedness of large financial institutions.

Both sorts of systemic risk – those due to the inter-relatedness associated with TBTF institutions and those due to correlated failures – could present a threat to banks’ maintenance of critical functions, including functions on which retail banking depends. Hence, insured deposits and liabilities called on demand without supervisory approval are excluded from the bailing-in features of TLAC in the FSB’s proposals.

One feature of the FSB’s proposals is the way in which they evade a problem which has stymied previous initiatives to achieve agreement on procedures for the resolution of large cross-border banks. This problem concerns procedures for distributing losses between the different jurisdictions involved. The belief which was the source of these difficulties was that much of the losses would have to be met by public sector or cooperative institutions rather than the failed banks, and that agreement was thus necessary on how these losses would be distributed internationally.

The assumption of the FSB’s proposals would appear to be that the losses could be completely borne by resources available through the bailing-in of selected classes of holders of loss-absorbing liabilities. This may be over-optimistic, and even after the bailing-in there may remain significant losses which must still be distributed between different groups and jurisdictions.

The lack of such optimism among regulators helps to explain countries’ increasingly favourable view of subsidiarization as the appropriate corporate form for cross-border banks, since this choice can be protective of a country’s interests owing to the way in which the losses of an insolvent cross-border banking group are assumed in the jurisdictions in which loss-making entities are legally incorporated. (SUNS7967)                                                                          

Andrew Cornford is with the Observatoire de la Finance in Geneva.

Third World Economics, Issue No. 587, 16-28 Feb 2015, pp13-16