The mandatory use of central clearing is a relatively new regulatory tool for mitigating systemic risk in the OTC derivatives space. It has brought the central counterparties (CCPs) to the forefront of financial market regulation. It is therefore crucial for CCPs to have comprehensive and effective recovery plans, relying on a wide range of recovery tools, to continue providing their critical services also in a very severe crisis without requiring the use of resolution powers by authorities. Obviously the stability of CCPs primarily depends on the establishment and implementation of robust risk management framework. But CCPs are also in a unique position to set up efficient and comprehensive recovery plans as they can base them on contractual arrangements with members. But CCPs need to have appropriate procedures for managing conflicts of interest between stakeholders - it is vital to involve all those who would bear losses in the design of the recovery plan. And on the regulatory side, we need to make sure our approach is cautious and non-prescriptive at this stage. I am confident that CCPs, authorities and stakeholders will work together constructively to meet the ambitious objectives set out in the recent CPMI-IOSCO report.
Ladies and Gentlemen,
Today, I want to share some thoughts with you on the recovery and resolution of central counterparties (CCPs).
The mandatory use of central clearing as a regulatory tool for mitigating systemic risk in the OTC derivatives space is a rather new feature in the history of financial market regulation. It has brought CCPs to the forefront of financial markets and financial market regulation hence it is crucial for CCPs to have comprehensive and effective recovery plans.
When we talk about “recovery” in relation to CCPs, we refer to the ability of a CCP to recover from a threat to its viability and financial strength so that it can continue to provide its critical services without requiring the use of resolution powers by authorities. Resolution, on the other hand refers to the orderly winding-down of an FMI, and would take place only if recovery fails or is not desirable.
Last month, the Committee on Payments and Market Infrastructures (CPMI) and the International Organization of Securities Commissions (IOSCO), published a report on “Recovery of financial market infrastructures”. Also last month, the Financial Stability Board published a report on the Key Attributes of Effective Resolution Regimes for Financial Institutions, which includes an annex dedicated to resolution regimes for FMIs. Together, these two reports are important milestones in strengthening the resilience of FMIs, and of the financial system more broadly.
Today, I will focus on the importance of CCP recovery and resolution, and the challenges faced by CCPs in designing appropriate recovery tools. I will also describe the specific role played by authorities during recovery situations, and mention the work that still needs to be done in the field of CCP recovery.
Recovery as part of CCP risk management
Before I explain why it is important for CCPs to have recovery and resolution tools, I would like to make it clear that these tools should only be seen as a way to complement regular risk management. In fact, recovery is the last step of a CCP’s risk management.
The regulatory requirements relating to the safety and efficiency of Financial Market Infrastructures defined under the CPMI-IOSCO Principles for Financial Market Infrastructures (PFMIs) (and national laws) are already fairly strict and conservative. Their aim, inter alia, is to ensure that recovery procedures will not need to be activated. In other words, if a CCP respects the requirements for risk management set out in the PFMIs (and national regulations) – which I should add that overseers and regulators are closely monitoring – this should, in principle, be enough to preserve the CCP and its functioning.
To ensure the PFMIs are properly and comprehensively applied, CPMI and IOSCO have set up an implementation monitoring procedure covering 28 jurisdictions. Two reports on the timeliness of the implementation have already been published and another work-stream will start shortly to assess whether jurisdictions have adopted legislation or policies to implement the PFMI (Level 1). The first assessments of the consistency and completeness of the Principles’ implementation (Level 2) for CCPs and TRs in the EU, Japan and US will be released in early 2015. Further L2 assessments will start at the end of the first quarter next year. In parallel, we have just initiated in all 28 jurisdictions an assessment of the implementation and application of authorities’ Responsibilities, with reports scheduled for autumn 2015. Moreover, preparatory work is now starting for a Level 3 assessment. It is anticipated that an assessment of the consistency of the outcomes of the application of the PFMIs to CCPs will start in autumn 2015. All these measures demonstrate that the regulatory community is extremely serious – and transparent - about the proper and equal implementation of the PFMI.
However, even if the safety rules under the PFMI are properly implemented, the unthinkable may still happen. Central banks and regulators have learned from the recent financial crisis that focussing only on “normal” risk management is not enough. In the event of a default of a major clearing member, a CCP might not be able to allocate all the losses and liquidity shortfalls that may occur. Therefore, each CCP needs to have a recovery plan for allocating any residual losses and liquidity shortfalls so that it can ensure continuity of its clearing services even when normal defences have failed. Without a robust recovery plan, the CCP might have to be resolved.
CCP recovery is therefore a crucial issue. One could even argue that continued provision of services is more important for CCPs than for other financial institutions. This is because CCPs, as for FMIs in general, are different to other financial institutions from a number of perspectives:
The use of a CCP is mandatory for certain products. So the relevant clearing services must be preserved under any and all circumstances.
CCPs are systemic risk managers at the core of the financial system. Their smooth operation is a necessary condition for many other types of financial services.
In most cases, CCPs have only very few - if any - substitutes or alternative providers.
The size and composition of CCP balance sheets is different – for instance, their balance sheets rarely include subordinated debt instruments. This makes bail in as a recovery or resolution tool difficult.
Designing recovery tools
Although setting up recovery plans will be a challenge for CCPs, they are in a unique and perhaps better situation to do so than other financial institutions such as banks. Banks also need such recovery arrangements. A key difference between banks and CCPs is that a CCP’s recovery plan can be based on contractual agreements with its members. Unlike banks’ protected depositors, they can be assumed to understand and manage their risk exposures. CCPs must have rules and procedures that allocate losses and liquidity shortfalls that are not covered by prefunded resources. From that perspective, CCPs are in a unique position in drawing on additional resources from members if and when needed. There may be cases when such rules and procedures go beyond the CCP members, but authorities should make sure in any case that they minimise losses imposed on innocent by-standers.
Simply having a recovery plan is not enough: CCPs also need to think about the optimal design and number of recovery tools.
As stated in the CPMI report I mentioned earlier, recovery tools should be comprehensive, effective, controllable, create appropriate incentives for risk management, and minimise negative impact.
The challenge is that no individual tool can equally meet all of these criteria because of trade-offs. For example, uncapped cash calls are comprehensive, but may create disincentives for central clearing. Variation margin haircutting can be effective, as it reduces pay-outs rather than requiring additional pay ins, but its systemic effects can be uncertain due to the way losses are allocated. Initial margin haircutting may also be considered as comprehensive, but on the other hand, it could increase the potential for contagion risk.
The CPMI-IOSCO report on the recovery of FMIs offers a thorough description and analysis of the various pros and cons of different recovery tools, as well as the trade-offs that exist between them. It is probably not possible at this point to identify a single specific recovery plan that is optimal and superior to any other plan. CCPs have a range of options to choose from. It will be important not to rule out, a priori, the use of some of these tools. Rather, it might be advisable for CCPs to have many different tools potentially available. This way, they will be able to choose the appropriate tools in a specific recovery situation.
One of the main challenges in drawing up recovery plans will be addressing the potential conflicts of interest between stakeholders.
Indeed, different loss allocation tools will have different effects on the distribution of losses. For example, rules on variation margin haircutting might be designed in a way that losses are passed on to clients, while cash calls might only have an effect on direct clearing members. Similarly, participants might not be willing to contribute to sharing losses which arise as a result of management failures.
It is therefore vital to involve those who would bear losses or liquidity shortfalls in the formation of the recovery plan. Mechanisms should be foreseen for involving all relevant stakeholders in the CCP’s decision-making process: this could mean establishing procedures whereby the CCP board formally consults direct participants, indirect participants and/or linked CCPs, as relevant, or having such relevant entities directly represented on the board, on the risk committee, or on other relevant committees. As opinions among stakeholders are likely to differ, the CCP should have clear processes for identifying the diversity of stakeholders’ views, and for managing any conflicts of interest between them.
In the field of recovery of FMIs, we are entering new territory. Therefore, authorities have so far taken a cautious approach to developing standards and regulation. The recently published CPMI-IOSCO report on the recovery of FMIs is not intended to create additional standards for CCPs or authorities beyond those set out in the PFMIs. Rather, the regulatory approach is non-prescriptive, and is aimed at providing guidance that should help FMIs to choose the appropriate mix of recovery tools. Over time, we may very well see the development of more prescriptive rules and regulations. A more prescriptive approach will in my view become inevitable if the current approach fails to prevent a race to the bottom between CCPs or across jurisdictions. The ongoing CPMI-IOSCO implementation monitoring process will be crucial to identify gaps and draw the lessons as needed.
Authorities are also aware of the need to ensure consistency between regulations affecting FMIs and those affecting banks. On one hand, FMI overseers and banking supervisors need to work closely together to ensure CCP recovery tools are consistent and not rendered ineffective by regulations affecting banks, or by the intervention of supervisors. On the other hand, overseers and supervisors need to make sure CCP recovery tools do not threaten the stability and soundness of financial institutions, and do not contradict the objectives of banking regulations. In this regard, CPMI and IOSCO have been working closely with the BCBS to ensure rules regarding FMI recovery and resolution are fully consistent with the preservation of the stability of participants. At the ECB, while respecting the separation between bank supervision and CCP oversight, we will strive to ensure consistent approaches and a common understanding of risks.
Managing a recovery situation
The main responsibility for designing, setting up and implementing recovery plans is with the CCPs themselves, in close co-operation with their members and other stakeholders who could potentially be affected. However, this does not discharge authorities from their own responsibilities, including in particular the review and assessment of CCP recovery plans, as well as the oversight of their implementation.
It is very difficult to predict the conditions that would require a recovery plan to be activated. There are many unknowns and uncertainties. It cannot be ruled out that a given crisis would be different from anything that has been experienced so far. So it is not clear if the recovery plan will work as intended. Some discretion is therefore needed both for the CCPs themselves and perhaps more importantly for the authorities, which cannot take a “hands off” approach during a crisis situation.
Authorities have a particular responsibility to do what is systemically needed in a crisis, and are better-placed to do this than CCPs given their broader perspective. For instance, authorities could be faced with a situation where the application of resolution procedures is preferable to the activation of a recovery plan. This means, of course, that much will depend on the willingness of authorities to do the right thing in the moment of crisis.
The recovery of CCPs is absolutely vital and hence it is of utmost importance that CCPs set up comprehensive and efficient recovery plans. However, there could still be a need to resolve a CCP if its recovery plan is not enough to address the particular crisis situation it is going through.
First of all, the attempt to recover may simply fail: ex-ante recovery plans may prove inadequate ex-post. Recovery plans have, fortunately, not yet needed to be implemented in real life, so it remains to be seen how they work in practice.
Most importantly, however, recovery is not a purpose in itself, but a means to preserve financial stability. From that perspective, resolution could be preferable over recovery if the application of a recovery plan is deemed to lead to undesirable results. For example, loss allocation tools could end up over-exposing certain groups of participants, such as buy side institutions or individual banks, depending on the scenario involved. Moreover, resolution may also become necessary if participants do not want to use the CCP any longer, and prefer instead to wind it down.
Designing effective resolution plans for CCPs is therefore also a key issue. The FSB’s report on Key Attributes of Effective Resolution Regimes for Financial Institutions is an important milestone in this regard.
The last point I wish to make today is to acknowledge that important work still needs to be done in the field of CCP recovery. The work on recovery has not come to an end with the publication of the CPMI-IOSCO report. Some further steps will be needed.
First of all, we need to make sure that the CPMI-IOSCO guidance on recovery is effectively and consistently implemented internationally. The CPMI-IOSCO report leaves a wide range of options for CCPs to meet the requirements put forward. We need to make sure that their implementation does not give rise to regulatory arbitrage and competitive distortions. In this regard, I mentioned earlier that CPMI and IOSCO are currently carrying out an implementation monitoring exercise for the PFMI: monitoring the implementation of the guidance on recovery and resolution will also be a priority.
We also need to better understand the systemic implications of pushing risk back to participants through allocation of uncovered losses and liquidity shortfalls. How will stress be transmitted through the system once a CCP’s pre-funded resources have been exhausted? An understanding of interdependencies, such as common participation, is particularly important. We also need to make sure we consider scenarios in which multiple CCPs enter recovery simultaneously. For instance, some clearing participants may be unable to fulfil all their obligations if multiple CCPs simultaneously ask for emergency cash calls and replenishments.
In this context, more thinking on CCP crisis scenarios could give us a somewhat clearer idea of the specific problems CCPs might have to face in a recovery situation, including the potential size of losses and liquidity shortfalls. To that end, deeper analysis of reverse stress testing to cover hypothetical portfolios and market conditions which go beyond ‘extreme but plausible’ would also be helpful. This is why CPMI-IOSCO have identified stress testing as a policy priority, and will soon launch a work stream to develop potential guidance regarding best practices and methodologies. The objective would be to ensure that all CCPs apply reliable and comparably robust stress testing frameworks.
Despite the work that still needs to be done to ensure the full effectiveness of CCP recovery and resolution, it is important to understand that CCPs are truly beneficial for the preservation of financial stability. Since the financial crisis, the systemic importance of CCPs has greatly increased, and there may be concerns within the market about their safety and soundness. Some of these concerns are legitimate, and regulators are taking them serious and aim at addressing them. However, it should be recalled that the alternative to central clearing is far more risky: the non-cleared world, which existed before the crisis, was dangerously opaque, did not allow for pooling of risks, and was inefficient in terms of collateral mobilisation. The assumption commonly held before the crisis that risk would be better managed if it were thinly spread across the system was simply proven wrong. Regulators are fully aware of the need to ensure the robustness of CCPs, and are determined to apply international standards such as the PFMI and the guidance on recovery and resolution in a consistent and rigorous way, consistent with bank supervision and leaving no room for regulatory arbitrage.
I am confident that CCPs, authorities and stakeholders will work together constructively to meet the ambitious objectives set out in the CPMI-IOSCO report.
Thank you very much for your attention.