Response to Financial Stability Board consultation on guidance for Central Counterparty Resolution and Resolution Planning
Norges Bank's letter of 20 March 2107 to the Financial Stability Board
Norges Bank commends the efforts of the Financial Stability Board in seeking to strengthen financial stability and building international standards for safe and efficient financial markets. We appreciate the opportunity to respond to this consultation on the proposed Guidance for Central Counterparty (CCP) resolution and resolution planning.
Our comments focus on two main areas: (i) resolution powers and loss allocation and (ii) the entry into resolution and coordination with other authorities during resolution planning.
1) Resolution powers, loss allocation and financial resources (sections 2 ,4 and 6 of the Guidance)
a. Variation Margin Gain Haircutting
Because ex-ante it is arbitrary and random which portfolio will be hit by Variation Margin Gain Haircutting (VMGH), it appears to be a “fair” tool. However, from a buy-side/sell-side perspective it may not be an entirely well-balanced recovery instrument. As clearing members, normally sell-side firms, tend to have portfolios that are more balanced and less directional than buy-side firms, usually clients, (ESRB, 2016) it seems that buy-side firms might be affected disproportionately by this measure in comparison to sell-side firms. Moreover, as buy-side firms are usually not members of the CCP’s risk committee, it will be more difficult for them to influence the development of recovery tools in general.
Norges Bank believes this aspect deserves further analysis to ensure that recovery tools are well balanced and applied non-discriminatorily across market participants.
b. Initial Margin Haircutting (referring to point 2.11 p.6)
Because both VMGH and assessment rights are commonly capped, one cannot exclude the possibility, however remote, that liquidation losses in fact exceed the impact of these recovery tools.
Another option might be for clearing members to contractually grant the CCP access to their initial margin once the end of the default waterfall is reached. This could be combined with a partial tear-up of contracts. Indeed, the CCP could take the required funds (in proportions from each clearing member that are to be determined such that incentives are aligned) and simultaneously reduce the notional value of all positions that correspond to the initial margin haircut. Such an access to the initial margin at the end of the waterfall combined with a partial tear-up could be implemented in several ways:
Instead of seizing the margin to cover residual uncovered losses, the CCP could perform a partial tear-up of the non-liquidated remaining part of the defaulter portfolio, and then mutualize losses on a broader clearing member base (Duffie 2015). Indeed it may not be entirely fair to allocate the losses solely to the counterparties of the defaulting trade, because due to trade compression cycles and further trading in the same standardized derivatives, it is not at all clear that the failing clearing member was the original trading counterparty. Moreover, concentrating the losses on a few such counterparties as opposed to mutualizing those increases the risk of default contagion to one of these selected counterparties.
At the same time, the risk of having Initial Margin seized at the end of the waterfall may incentivize clearing members to change their posted collateral cash/non-cash mix. This can have potential repercussions on CCP liquidity management and needs to be weighed against pros and cons of other end-of-the-waterfall measures (LCH, 2016; Cont 2017).
Norges Bank believes more research is warranted on this issue; in particular, case studies on actual CCP data could shed light on the incentives, efficiency and quantitative impact of one instrument versus another.
Overall, through access to Initial Margin at the end of the waterfall (i) losses can be fully allocated, (ii) the CCP returns to a matched book, (iii) losses are mutualized, (iv) the CCP can continue its operations, (iv) granting access to IM is likely less straining on the clearing member’s liquidity than it is to provide additional funds (e.g. voluntary contributions) to the CCP, and (v) no public funds are utilized (this may be the closest concept to a bail-in in the “CCP-domain”).
Norges Bank believes the application of Initial Margin Haircutting used in combination with a partial tear up should be given more thought as end-of-the-waterfall tool.
2) Entry into resolution and coordination with other authorities during the resolution planning (sections 3, 7 and 9 of the Guidance)
CCP default waterfalls are in principle designed so as to withstand the simultaneous default of the two clearing members to which the CCP has the largest exposures, usually G-18 dealers for systemic CCPs. Consequently, at the moment when a systemically important CCP were to be placed into resolution, it is plausible to assume that at least one or several of its clearing members will already have been put into resolution by their home resolution authorities. It is thus crucial to properly understand the interactions of CCP- and bank resolution schemes. (Referring in particular to “Cooperation between relevant authorities in the lead up to resolution, points 3.6 – 3.8 of the Guidance)
A highly relevant example of such interdependencies is the authorities’ power to temporarily stay early termination rights (referred to in point (vii) on p.15 of the Guidance), as implemented via the Bank Recovery and Resolution Directive or Dodd Frank for Qualifying Financial Contracts (QFCs). Similar proposals for moratoriums in the case of CCP resolutions are currently being drafted by the EU. Obviously, a dilemma arises between the desire to stabilize the dealer versus ensuring protection of the CCP and its other clearing members: “Should the CCP be stayed from liquidating the positions of a failing major dealer in order to improve the likelihood that the dealer can be stabilized? Or should the CCP be permitted to stop the potential growth of losses to its other systemic clearing members?” (Duffie, 2017).
Norges Bank holds the view that the interaction of temporary stays on early termination rights for dealers and FMIs needs careful and detailed consideration, and that further analysis and clearer guidance are warranted. Norges Bank further believes that such considerations highlight the importance of having a common goal (financial stability) and of ensuring a close cooperation in the planning phase as well as in the resolution phase itself between all authorities involved in both CCP and bank resolution.
Yours sincerely,
Torbjørn Hægeland
Executive Director
Norges Bank Financial Stability
Anna Grinaker
Director
Norges Bank Financial Infrastructure
References
- Financial Stability Board, Key Attributes of Effective Resolution Regimes for Financial Institutions, 2014
- Financial Stability Board, Essential Aspects of CCP Resolution Planning, 2016
- Financial Stability Board, Guidance on Central Counterparty Resolution and Resolution Planning,2017
- R. Cont, Central clearing and risk transformation, https://ssrn.com/abstract=2919260, 2017.
- D. Duffie, Resolution of failing Central Counterparties, http://www.darrellduffie.com/uploads/pubs/DuffieCCP-ResolutionJan2015.pdf, 2015.
- D. Duffie, RISK Magazine, http://www.risk.net/derivatives/3916441/ccps-should-prep-to-quash-sifi-swap-termination-stays
- ESRB, Shedding light on dark markets: First insights from the new EU-wide OTC derivatives dataset, https://www.esrb.europa.eu/pub/pdf/occasional/20160922_occasional_paper_11.en.pdf, 2016
- LCH.Clearnet, Recovery and Resolution – A framework for CCPS, http://www.swapclear.com/Images/framework%20for%20ccps.pdf, 2016.