I am pleased to welcome you to the second conference on central counterparty (CCP) risk management organised by the ECB together with the Deutsche Bundesbank and the Federal Reserve Bank of Chicago. We are pleased that such a wide range of public and private sector stakeholders have joined us today to discuss the key risk management challenges in central clearing.
Considering the increased concentration of financial risk in CCPs, this is clearly an important issue. Since the 2009 G20 agreement to introduce mandatory central clearing obligations for over-the-counter (OTC) derivatives, we have observed strong growth in central clearing in these markets, in particular for interest rate swaps (IRS) and credit default swaps (CDS). Compared with 2009, when only around 10% of CDS and 37% of IRS were centrally cleared, these figures now stand at more than 50% for CDS and almost 80% for IRS.
The decision to launch mandatory central clearing was coupled with the premise that increased risk concentration in CCPs must be accompanied by stringent safeguards. And indeed, global standard-setting bodies have worked hard to further enhance the resilience, recovery and resolvability of CCPs.
But this progress is no reason for complacency. So far, the expansion of central clearing has taken place in the context of a favourable financial environment, supported by very accommodative monetary policies. But such benign conditions will not last forever. And, over time, market value corrections may test the defences of central clearing. Many regulatory reforms also still need to be fully implemented and in some areas of risk management, such as the management of extreme stress events, approaches are still evolving. Finally, the rapid evolution of centrally cleared markets and of the broader financial and technological environment poses further challenges.
Ensuring the safety and efficiency of central clearing is critical for the ECB given the Eurosystem’s role as central bank of issue for the euro. The large payment flows between CCPs and their participants mean that inadequate financial risk management of CCPs could transmit serious financial strains to institutions that are Eurosystem monetary policy counterparties. Interconnected payment systems and repo markets, which are essential for monetary policy transmission, could be equally affected. Furthermore, in situations of severe market stress, the Eurosystem may be called upon to act as a lender of last resort, which could expose us to significant risks.
Given the significant use of the euro by globally active CCPs, robust arrangements for cross-border cooperation among authorities are an overarching ECB priority with a view to ensuring central clearing robustness. Here, significant progress was made in Europe last year with the adoption of a new supervisory framework for CCPs in the revised European Market Infrastructure Regulation (EMIR 2). This framework requires non-EU CCPs that are critical for the EU to meet prudential requirements under the supervision of the European Securities and Markets Authority, with the involvement of the relevant EU central banks. This setup is essential in ensuring that appropriate safeguards for EU financial stability are in place, while risks of global financial fragmentation are minimised.
At the global level, however, cooperation is not yet at the level it should be. For instance, as the Financial Stability Board (FSB) highlighted in its 2019 Resolution Report,
no credible resolution plans are currently in place for any of the 13 major CCPs that are systemically relevant in more than one jurisdiction. Similarly, for most of these CCPs we have not yet even seen the adoption of a cooperation arrangement to support the necessary level of confidential information sharing.
Against this background, I would like to reiterate the call that was made at this conference last year to accelerate the development of cooperative arrangements at the global level. The financial community cannot afford to further delay the work on bringing cooperative oversight, crisis management and resolution planning in line with international standards, as this could undermine our ability to respond to a potential stress event affecting one of the major global CCPs.
However, cross-border cooperation among authorities is not enough. It needs to be complemented by coordination within the central clearing community –between banks, CCPs and their respective authorities – and this is the main point of my remarks today. Indeed, the interaction of the various stakeholders is attracting more and more attention, especially as we move towards thorny questions around the responsibilities of banks and CCPs for potential end-of-waterfall losses. It is also an important matter for the ECB in particular, given our dual perspective as banking supervisor and central bank of issue.
I will divide my remarks into three main parts. First, I will summarise why effective coordination between banks, CCPs and public authorities is essential for the safety of central clearing. Second, I will highlight what I see as the main shortcomings in this field. And third, I will suggest possible avenues for addressing them.
Importance of coordination within the central clearing community
Effective interaction between banks and CCPs is a prerequisite for safe and efficient central clearing, given that CCPs manage risks for banks rather than taking on financial risk themselves. As a result, the vast majority of CCPs’ financial defences are provided by their clearing members. Banks are also critical in providing access to global CCPs for smaller banks and non-bank clients and as service providers to CCPs. Interdependencies with banks arise not only within, but also across CCPs, given that major global banking groups are critical for several CCPs.
Finally, these interdependencies have further increased due to the concentration of OTC derivatives clearing in a small number of internationally active CCPs and banking groups.
The risk implications of these close ties between banks and CCPs are mutual, given the requirement for CCPs to always run a matched book and the principle of risk-sharing and loss mutualisation in CCPs. While clearing members post significant prefunded resources to CCPs to reduce risks arising from potential clearing member default, these defences may be insufficient for coping with extreme situations. In particular, if prefunded CCP margins and default funds are eroded, CCPs’ ability to recover their financial strength depends on the capacity of their clearing members to absorb large and unexpected losses on an ad hoc basis. This may be a challenge in situations of severe market stress, when banks may need to withstand credit and liquidity pressures from multiple sources.
Given the systemic risk concentration in CCPs, and the significant potential for contagion risks across CCPs and major banking groups, effective cooperation between banks and CCPs to reduce risks in central clearing is not only important for the private sector. It is equally important to any authority with a responsibility for banks, CCPs or financial stability.
Current shortcomings in coordination
Current shortcomings in coordination reflect two main weaknesses: diverging interests and knowledge gaps.
Diverging interests of banks and CCPs in CCP risk mitigation were underlined in the recent debate on CCP recovery and resolution. This is unsurprising, given the potentially very large unfunded payment obligations that may arise in such situations. While these diverging interests may be most visible at the end of the financial waterfall, they arise throughout a CCP’s lifecycle, and should therefore be addressed in a holistic manner.
Banks in particular have argued that their financial obligations in extreme events should be strictly limited, as they may otherwise be unable to manage their exposures. In addition, CCP capital and skin in the game requirements should be increased in order to provide for additional loss absorbing resources in CCP recovery or resolution and to incentivise prudent CCP risk management. Banks have also called for closer clearing member involvement in CCP risk management decision-making and enhanced CCP disclosures on margin and stress test methodologies to facilitate clearing member scrutiny.
Informal feedback from CCPs on these proposals has been negative. They have pointed out that, as it is the clearing members who take the decisions to engage in the financial transactions that are subsequently submitted to clearing, they are the ones who should be held responsible for possible losses. Moreover, higher CCP capital and skin in the game requirements would increase the cost of central clearing, which may conflict with the G20 objective of incentivising central clearing.
Overall, it seems that we are currently in a stalemate on this issue. This is a concern given that the private sector’s ability to fully absorb any losses in central clearing is critical to pre-empt risks for taxpayers and moral hazard.
The diverging interests I have just outlined are aggravated by serious gaps of knowledge. For one, the interaction of prudential requirements for banks and CCPs has not yet been sufficiently thought through. In particular, though banks may potentially face very large payment obligations in a CCP recovery or resolution event – as a result of, say, cash calls, potential haircuts of variation margin gains or tear-ups – they are currently not required to hold any capital against them.
Other concerns relate to the analysis of, and preparedness for, extreme stress events. Indeed, we are still at an early stage of identifying the scenarios that could lead to CCP recovery or resolution and quantifying respective funding needs. Stress events, which by definition would go beyond any “extreme but plausible” situation, are obviously not easy to describe precisely. We will never be able to anticipate all future events. Nevertheless, a credible range of recovery and resolution scenarios could and should be developed.
Another unclear aspect is how we can assess the potential systemic risk implications of certain CCP recovery tools, such as repeated cash calls, partial tear-up or variation margin gains haircutting. Such an assessment is critical in enabling a resolution authority to develop a credible CCP resolution strategy and in informing its decision on whether or not to place a CCP in resolution. A specific challenge in this context is the lack of granular data on central clearing interdependencies due to the strict confidentiality of data on the exposures of individual participants. While cross-CCP supervisory stress testing has proven helpful in gauging the systemic contagion risks arising from central clearing interdependencies, we have only seen a few such coordinated efforts so far. Similarly, crisis simulation exercises to test the operational preparedness of authorities for potential stress events have been very limited.
More also needs to be done to conceptualise non-default loss scenarios, reliable tools for absorbing the associated losses, and the roles and responsibilities of the various stakeholders. There are far fewer prefunded resources available for non-default losses than for default losses, so the risk of entering into CCP recovery or resolution for non-default related reasons may be higher than for default-related reasons. It is therefore critical to make progress in this area. In this context, we also need to reflect on how to ensure a fair loss allocation. While our choice can be guided by the basic premise that those taking the decisions should also be responsible for any losses, it is not always clear where we should draw this line. For example, the general principle that operational losses should be borne by CCPs may not hold in the case of contagion risks emanating from a clearing member that has suffered a cyberattack. In addition, given the small amount of CCP capital, it is uncertain whether comprehensive loss absorption could be ensured in all circumstances. Recourse to insurance or voluntary capital injections from parent companies may not be sufficient to ensure a reliable, timely and comprehensive response.
In the area of client clearing, we currently see very diverse arrangements for sharing information and involving CCPs in the relationships between clearing members and clients. This may result in uncertainties regarding CCPs’ ability to port client positions of a defaulting member in a timely manner, particularly as some clients hold very large directional positions. Given the growth of client clearing and the increased degree of client clearing concentration in a few major clearing members, this may expose the financial system to unacceptable risks.
Joining forces to enhance coordination on risks in central clearing
In order to address the problems I have just mentioned, action is necessary in various areas.
We need to improve the institutional setting for information sharing and coordination among CCPs, banks and public authorities. Progress in this direction would be important in both reducing knowledge gaps and identifying possible solutions reflecting the various stakeholders’ concerns. We need both public and private sector engagement to address the collective action problems that we face.
On the public sector side, banking supervisors need to be more frequently included in the regulatory dialogue between securities regulators and central banks. The interaction of prudential requirements for banks and CCPs should be further assessed, notably with regard to the treatment of client clearing, the conceptualisation of extreme stress scenarios and the identification of systemic contagion effects arising from CCPs’ deployment of their recovery tools. Looking forward, a discussion is warranted on robust approaches to meet funding needs in recovery and resolution.
Against this background, the ECB supports the idea of holding a dedicated workshop on credible CCP resolution funding involving the FSB, the Committee on Payments and Market Infrastructures and the Board of the International Organisation of Securities Commissions (CPMI-IOSCO), and the Basel Committee on Banking Supervision before the FSB finalises its guidance on the issue by the end of this year. Global standard-setting bodies could also provide a joint impetus to coordinated supervisory stress testing.
As for individual CCPs, we need to move beyond the purely administrative preparation of cooperation in stress events. We also need to consider concrete information needs and responsibilities in the event of a CCP crisis, thereby ensuring that we can respond promptly and effectively. Periodic crisis simulation exercises should be organised to test and advance our understanding. The ECB sees this as a short-term priority for CCP crisis management groups that should be pursued in parallel to developing resolution plans.
The private sector should also step up its efforts. In past years, both CCPs and clearing participants have made very helpful contributions to discussions on CCP risk mitigation. However, beyond their interaction in CCP risk committees, the various private sector stakeholders do not seem to be engaged in a structured dialogue. Rather than speaking about each other, private sector participants would do well to speak more to each other.
To this end, I would see significant benefits in a private sector-led standing forum for dialogue between the main industry associations of CCPs and banks. The ECB stands ready to help facilitate the launch of such an initiative. Besides focusing on governance, disclosure practices and client clearing arrangements in CCPs, such a horizontal forum could also work towards cross-fertilising approaches to developing stress scenarios that go beyond extreme but plausible events.
Finally, I also think that regulatory action might be helpful in a few targeted areas. For instance, this may be the case for non-default losses in central clearing, a topic on which current guidance is limited. Indeed, work on this is already underway at both CPMI-IOSCO and FSB level. We should also assess whether we may need more risk-sensitive approaches to the calibration of CCP capital to ensure comprehensive loss absorption. Client clearing may be another area requiring more consistent and transparent approaches, depending also on the findings of the ongoing work by the CPMI and IOSCO in this area.
Regulatory action should also be considered to address the diverging interests around the allocation of losses in CCP recovery and resolution. Given the high concentration of systemic risk in CCPs, a credible and effective framework and full stakeholder commitment are absolutely essential. In this context, I see the following guiding principles.
First, we must avoid leaving any private sector funding gaps, as they would result in unacceptable risks for taxpayers. CCPs must have arrangements in place to ensure comprehensive loss absorption in all circumstances, in line with the CPMI-IOSCO Principles for Financial Market Infrastructures and the related recovery guidance.
Second, we should acknowledge competitive pressures and diverging interests in central clearing. It is true that the growth of global OTC derivatives clearing has led to intensified oligopolistic competition among CCPs. Considering also that CCPs rely on fees in relation to volumes as their sole source of income, and bearing in mind the present low interest rate environment, it cannot be ruled out that some CCPs may search for higher yields through more risky investment practices, lower-cost risk management practices or insufficiently prudent “fit for clearing” checks. In a world of mandatory central clearing and an oligopolistic market structure, clearing participants may not necessarily be on an equal footing with CCPs when negotiating terms for participation and market discipline may not be optimal.
Third, CCP participants must continue to bear the bulk of default-related losses. We may well consider increasing skin in the game requirements for CCPs in the context of default loss absorption in order to strengthen incentives for prudent risk management, but we cannot turn the central clearing model upside down. The fact remains that the transactions brought to central clearing are initiated by clearing members. Against this background, clearing participants should also remain primarily responsible for any losses to pre-empt adverse risk management incentives on their side.
Fourth, given the role of CCP participants in absorbing at least the vast majority of default-related losses in central clearing, it may be appropriate to enhance their involvement in key CCP risk management decisions as well as improving transparency on CCPs’ key risk mitigants. Indeed, this could increase stakeholder ownership of risk management outcomes and would be in line with the basic principle that responsibility should be aligned with control. Initially, I would suggest that the financial industry comes together to assess whether individual CCPs have already established effective practices that could form a starting point for potential regulatory action.
Let me conclude. We have come a long way since the G20 agreement to increase mandatory central clearing obligations. However, critical gaps remain in cross-border cooperative arrangements for global CCPs as well as in coordination within the central clearing community.
Effective coordination among banks, CCPs and public authorities is necessary to manage risks in central clearing. There are significant shortcomings in this area, which cut across the various issues for discussion at today’s conference.
In order to improve coordination, we need to take action to find fair solutions for diverging interests and reduce knowledge gaps. To this end, both private and public sector stakeholders should contribute to developing an enhanced institutional setting for ongoing dialogue and coordination. In a few distinct areas, targeted regulatory action may also be needed.
We can make significant progress if we take advantage of synergies of expertise and adopt a systemic approach, rather than focusing on defending individual interests. Today’s conference offers the opportunity to better join forces along these lines.
Thank you very much for your attention