Have clearinghouses become too big to fail?
Clearinghouses are now the central risk hubs of the global derivatives markets, and that's exactly what was intended. By moving most standardized derivatives into clearing, the reforms have addressed the problems that made the crisis so bad. But what if these new hubs run into trouble? Could the solution to the previous crisis become the source of the next one?
This is the problem that global supervisors are now tackling. Market regulators and central banks have embarked on a program to make sure that derivatives clearinghouses are prepared to weather any type of storm. They have raised the bar on financial resources, requiring clearinghouses to have the ability to manage the default of not one but their two largest members. They have expanded and enhanced the international standards for ensuring the resilience of clearinghouses, focusing on key issues in governance, stress-testing, margining systems, and the amount of money that the clearinghouses contribute out of their own funds to their default management resources. And they have issued guidance on recovery planning, the process by which a clearinghouse deals with losses or liquidity shortfalls and replenishes its resources.
The next phase of this program is resolution, i.e., developing a plan for determining when a recovery process is failing and the authorities should step in. The goal here is to maintain or restore the clearinghouse's critical functions, or shift those functions to another entity and wind the clearinghouse down in an orderly way. The Financial Stability Board, which coordinates policy on this issue at the global level, set out high-level guidelines in 2017. Now it is the responsibility of individual jurisdictions to implement them.
The Bank of England is a key player in these policy discussions. Since 2013, when the U.K. reorganized its supervisory structure, the Bank of England has been responsible for the oversight of financial market infrastructures, including clearinghouses, securities depositories and payment systems. That includes three U.K. clearinghouses that handle derivatives: ICE Clear Europe, LCH Ltd., and LME Clear.
All three clearinghouses are important to market participants in many parts of the world and the Bank has been a leader in the discussions on international standards organized by the FSB, the Committee on Payments and Market Infrastructures and the International Organization of Securities Commissions. The Bank was the co-chair, alongside the Federal Reserve in the U.S., of a group in CPMI-IOSCO that drafted an international framework for supervisory stress testing of clearinghouses. This framework, which was published in April, is intended to assess the systemic effects when multiple clearinghouses are responding to the same stress events, such as the default of a bank that is a member of multiple clearinghouses.
The Bank is also co-chair of the “derivatives assessment team” operating under the authority of the FSB that is reviewing the interaction of post-crisis reforms and the impact on the incentives for clearing. This review, which is expected to conclude with a final report at the end of 2018, is expected to address one of the biggest issues facing the industry, namely the increase in the cost of providing clearing services due to the implementation of the Basel III capital and leverage requirements.
Jon Cunliffe is the leader of the Bank's work in this area. He is the central bank's deputy governor for financial stability. This role, which he has held since November 2013, is responsible for the supervision of financial market infrastructures as well as resolution and international issues. He is a member of the G20 Financial Stability Board Steering Committee, the Bank for International Settlements’ Board of Directors and the European Systemic Risk Board.
Cunliffe has been involved in these issues for many years. Between 1990 and 2007 he held various posts at the U.K. Treasury related to financial regulation, the international financial system, and U.K. macroeconomic policy. From July 2007 to December 2011, he was the Prime Minister’s advisor on Europe and global issues and the U.K. Sherpa for the G8 and G20. In that role he was directly involved in the drafting of the post-crisis financial reforms that were agreed by the G20 leaders in 2009, including the clearing requirement for standardized OTC derivatives.
In recent years he has overseen the Bank's involvement in the first major cross-border tests of default management procedures. For example, last year three major clearinghouses – CME, Eurex Clearing and LCH Ltd – conducted their annual “fire drills” in parallel. That allowed their supervisors, including not only the Bank but also the Commodity Futures Trading Commission in the U.S. and Bafin and Bundesbank in Germany, to observe the drills and understand the interdependencies between the clearing members and the clearinghouses, a key source of concern for global supervisors.
In this interview, which was conducted in Washington, D.C. during the annual spring meetings of the IMF and the World Bank, Cunliffe talks about the original purpose of those reforms and the consequences, both intended and unintended. He emphasizes the success of those reforms in providing more “clarity” into the web of derivatives risk exposures across the financial system. He expresses a concern about the interaction of the leverage ratio with the clearing requirements and notes that international regulators have been asked to look at the leverage ratio to see if there are smart changes that can be made to take into account the segregation of client collateral.
He also provides his perspective on resolution for clearinghouses and the principles that supervisors should keep in mind if and when they decide to intervene. Above all, he says, supervisors should maintain the principle of “no creditor worse off.” Under this principle no creditor or shareholder shall incur greater losses than they would have incurred if the loss allocation arrangements in the clearinghouse’s rule book had been followed.
Recognizing that clearinghouses are effectively global infrastructures, he stresses that there needs to be a very high level of cooperation and information sharing among the supervisors. There need to be “rules of the road” about what happens in a crisis, and avoid a situation where a clearinghouse faces conflicting instructions from multiple supervisors, he says. That could actually change the clearinghouse from having a dampening effect on systemic risk to having a “risk transmission” effect on the system.
He concludes with some comments on what the Bank is doing to prepare for the U.K. withdrawal from the European Union, focusing on the treatment of clearinghouses while a new regulatory framework is established. Regarding third-country clearinghouses that want to continue offering services in the U.K. post-Brexit, he says that the Bank anticipates that any clearinghouse that has already met the requirements for authorization or recognition in the EU, provided they engage on the relevant process with the Bank, will be recognized in the U.K. The goal, he says, is to avoid any disruption to the provision of clearing services during and after the transition.
FIA: If you look back at the post-crisis financial reforms, and focusing on the parts related to derivatives, what were the benefits that have been achieved and what were the unintended consequences, either positive or negative?
That was part of the action plan to deal with the financial crisis. The reasoning behind it was that we’d seen two things happen in the crisis.
One, there was a web of bilateral derivatives contracts between key market players that we didn’t have good visibility of because of a lack of data available on financial transactions.
Secondly, the contracts were all on different [risk management] standards. Some of these had been margined, to some extent, and some not. When the crisis hit, it turned out that this exposed connections backwards and forwards between market participants.
In an atmosphere of panic, when credit dried up and nobody was quite sure who would be the next failure, people started margining very, very heavily. Margin calls acted as an accelerator of crisis. And because people couldn’t meet margin calls, the spiral carried on, they had to sell assets, and this led to fire sales which drove down asset prices.
So the prime driver was the idea that you could reduce that risk through central clearing in a number of ways.
One, you would be able to see what was happening. During the crisis, it was quite hard to see where the links were, and supervisors thought they could take action on one institution, and suddenly something somewhere else would get into trouble.
Two, you would be able to margin these contracts in a much more regular way, and the institutions that would be doing the margining and holding the collateral to absorb shock would not be the institutions actually involved in the transaction. They would have an interest in running a much safer service. They would take collateral from both sides, and they wouldn’t be affected by that sort of customer type of relationship.
There’s also a thought that this would be potentially more efficient. As everything would be together, people could net all the pluses and minuses, and it would only be the net exposures that they would have to cover. During the crisis, we saw some large margin movements going in different directions, and just the flow itself was costly. There’s an efficiency in risk management from netting positions across the system.
So that was the objective. I think it succeeded pretty well, actually. If I look at some interest rate derivatives, I think we’re up to between 60% and 65% that are now cleared. The big banks are holding twice as much collateral than they had before the crisis against those exposures.
We, the regulators, the supervisors, are beginning to get information as to where positions are.
We can observe the margining models. We have very detailed discussions with the clearinghouses on how their margin models work. If we think that’s not prudent, we can ask for changes.
Our calculations suggest that there are a trillion dollars more in collateral backing both cleared and uncleared OTC derivative products now than before the crisis. That, to me, is a success story.
You asked about unintended consequences. A lot of people say to me, “Ah, but you didn’t realize this would create a concentration of risk.” The answer is, of course we did realize that. That’s part of the object of the exercise. From a pretty dense opaque web of bilateral contracts we've gone to much greater clarity and simplicity. Of course, in the process we've created hubs [CCPs], so we have to make sure these hubs are safe. The fact that so much traffic is moving through these systems means that the regulation and the resilience have to be very strong. But the idea of concentration is not an unintended or manifested consequence. This was a part of the proposal. And it’s actually part of the reason why you can net, it’s part of the reason why you have visibility, and it’s part of the reason why you expect, in a crisis, these hubs will absorb stress.
I’ll give an example. When Lehman Brothers went bankrupt, within a relatively short period of time all their positions on LCH were hedged quickly and then auctioned off in a relatively smooth fashion with the whole process being completed within three weeks or so. On the uncleared side it took years to get the money back. So concentration, to some extent, can help.
Where we do perhaps see an unintended consequence is in the interaction between different parts of the reforms. The regulations we’ve made for infrastructure and the regulatory changes we made for banks have come together and produced effects that we weren’t necessarily anticipating. One key point is the way the leverage ratio, which binds on the clearing members, has affected the cost of clearing for clients. If we have now got a position where some people can’t access clearing because of another regulatory reform on the banking side, you have to say, look, we have two regulations which are both trying to achieve resilience, but perhaps they are having an effect where one may be undermined by the other.
FIA: Can this be addressed by regulators operating within a single jurisdiction, or will it require coordinated action at the level of the Basel Committee?
It’s got to be across the international community. There are some areas [of the capital standards] where there are national discretions. This is, I recall, not an area where there’s a national discretion. We all have to do it together. There is a group in the FSB now, the Derivatives Assessment Team, which is looking at the post-crisis reform package and its impact on incentives to clear. And also a group within the Basel Committee looking at the leverage ratio question specifically. The objective is not to weaken regulation or soften the impact of the leverage ratio. The question is, if this is client money, and it’s held separately, can you deduct it from the leverage ratio exposure, and what’s the most efficient way of achieving the level of resilience that you want to achieve.
I have concern when contracts that I would want to see in clearing are not in clearing because of the interaction with something else in the regulatory framework. The question to the international community is: is there a smart change we can make that leaves us with the same degree of resilience and protection on the bank side, but doesn’t disincentivise clearing.
FIA: Do you think the other central bank authorities recognize that issue?
We probably see it more closely because we have a very large amount of clearing activity in the U.K. Second, we and a lot of the international community, are, rightly, very cautious about changing the leverage ratio. But I think there is recognition that not everything works perfectly and a combination of changes might be having more of an effect than one would have wanted. The need is not to review the whole reform package in a huge holistic exercise, but rather, look at regulation area by area and consider, is it working as intended? Can we make it smarter?
I know it has been discussed in the Basel committee. I think people are starting to recognize it as an issue.
FIA: Let me ask about a different aspect of this issue. You're saying that yes, there will be concentration of risk in clearinghouses, but that’s a better outcome than the fragmented opaque system we had before. Well, the day may come when a clearinghouse runs into some sort of problem and might be on the verge of failure. What’s the status of the resolution regime for clearinghouses? Are we prepared for that kind of problem?
The reform program has come in stages. Bank capital, liquidity and clearing were initial stages. Resolution came along in a subsequent stage, and we're now starting to implement the resolution framework in the U.K. for the banking system. Resolution for clearinghouses is mainly in the policy design stage, not the policy implementation stage. There are some initial standards from the FSB. But there’s more that needs to be done.
A couple of points I’d make. First, resolution for clearinghouses and resolution for big systemically important banks both have the same objective, which is to protect financial stability and avoid having the taxpayer come in. But the similarity stops there. Clearinghouses are not banks. They don’t take prudential risk in the same way. In fact, they’re built not to take risk. They have to run a matched book. The business model of the clearinghouse is not the bank model of lending and risk taking. The model of the clearinghouse is to take no risk, but offer a service and charge for that service.
Secondly, the move to clearing was accompanied by strengthening the resources of the clearinghouses to absorb stress. First of all, there’s initial margin, which is collateral that wasn’t in the system before clearing. Then there are portions of the CCP’s own capital at risk and default funds that are there to deal with the failure of members. Then clearinghouses have the ability to make further cash calls, so they can call a reload of a default fund as well as, for some, the ability to haircut the variation margin gains owed to members and clients. And finally, CCP recovery plans provide for tearing up particular contracts or, at the very end, the tearing up of all the contracts.
So you could argue, as some do, that you don’t need resolution for clearinghouses. Unlike a bank, which absent resolution gets to insolvency and then you don’t know where it goes next, a clearinghouse has a rulebook that takes it all the way down to tearing everything up and going to extinction.
I don’t think that’s a convincing answer. If we’ve made the system depend on clearinghouses, the answer that we have established a predetermined, ordered way to end the clearing service in case of a crisis is not enough. The reason for having a resolution regime is to ensure financial stability and continuity of the CCPs critical functions, without putting taxpayers on the hook for losses.
If the default of a number of large clearing members leads to financial stress at a clearinghouse, I don’t think resolution for clearing is principally about whether we can restore solvency. It won’t be a problem of the solvency of the clearinghouse or the capital of the clearinghouse. It will be a problem of the default of the members and the obligations of the members, and how many times the clearinghouse can credibly go back to the members to reload the default fund or haircut variation margin if auctions of detailed positions are failing, without effectively killing any chance of continuity.
If you compare how far into the tail of distribution of unlikely events you have to go to exhaust bank capital, with all of the reinforcing that we have put in, against how far into that tail you have to go to exhaust all of the resources the clearinghouse has before it gets to tearing up contracts, the answer is you have to get much further into the tail to exhaust a clearinghouse’s resources.
To get to that situation, more than two of the main members will have to have failed. And I’m now thinking of a world not just in which these major banks have failed, but in which the resolution plans for these banks has been unsuccessful in ensuring that they can continue to meet their obligations to their clearinghouses. If so, this is a systemic crisis and we have problems right across the system. At that point, we need to ensure the clearinghouse can, actually, and in an orderly way, put its recovery plans into action. Because you might well have issues where the management can’t put all of those recovery plans into action because they’re quite difficult actions which may affect the wider stability of the system.
That, to me, is the reason why there needs to be a point where the authorities can step in. These may be litigation or management failure reasons why the clearinghouse finds it difficult to act promptly and clearly in such a situation. Or, the incentives for the members may be wrong. If the members of the clearinghouse don't participate in an auction because they think that once default fund resources are exhausted it will fall back to government to deal with, then knowing the resolution authority can come in earlier and, for example, require a partial tear-up may change incentives.
There’s one really big important caveat to all of this. If you think that, as the clearinghouse works through its recovery options in what would probably be a pretty bad crisis, at some point the authorities would have to step in while there are still some resources left in the system, then you need as well a very strong “no creditor worse off” protection for users. People go into clearinghouses because they know the rules in advance. They know what their losses are going to be if things go wrong. They are buying protection through derivatives and through putting them in clearing. If you ever get into tear up, what you are doing is tearing up people’s protection because the counterparty on the other side of the clearinghouse has defaulted.
We have strong “no creditor worse off” protection on the banking side. The resolution authority can’t just step in and decide it’s going to allocate losses in a very different way to the rule book for financial stability reasons. Or, if it does have to do that, then there is “no creditor worse off” protection so that people are compensated. We need similar protection for CCP resolution. Otherwise the danger is that the authorities step in and decide to allocate losses in a rather different way to the rulebook without the discipline of knowing that those who lose relative to the rulebook will be compensated.
The reason you use clearing is to protect yourself. In a way, the reason we mandated so much business to go through clearing is because of how it acts in a crisis as a systemic manager and dampener of risk. It is specifically there for those real stress situations. So if the authorities are going to step in and carry out the resolution, then they need to follow the rulebook and there needs to be protection for the CCP’s users.
FIA: You are in a particularly interesting position because the clearinghouses in the U.K. are important not only to the U.K. financial system but to other financial systems as well. How do you sort out who oversees what when you have these interconnections that go across borders? How do you ensure that you have the right level of oversight during the normal times, and you have the right powers available to you in crisis times, without ending up with the “too many hands on the steering wheel” problem?
The first thing to note is that cross-border risk is not limited to clearinghouses. This is only one albeit very important variant of a problem that we have to manage across the financial system. Unless we want our financial systems to be insulated completely from each other, we’re importing risk and exporting risk to each other all the time. So the clearinghouse issue, to me, is not some unique issue. The risk comes in a particular form, and it might come in a concentrated form very quickly. But it’s no different in principle than, for example, the fact that many European banks offer products and take exposures outside Europe in dollars. They have cross-currency risk, they have invested abroad, they have counterparty relationships, etc. They and their regulators have to manage this risk—and depend on risk management and regulation elsewhere. So I don’t think the cross-border risk from CCPs is a unique problem. People start to talk about it as though this is a unique financial stability risk and I don't think it is. We have to manage a range of cross-border financial sector risks and find ways to do that proportionately and effectively.
The next point is that, to achieve this, we have to have cooperation and common standards about what the rules are in the clearinghouse, and there has to be visibility. You need very high degrees of supervisory cooperation, information sharing, and rules of the road about what happens in times of stress. But a situation where a clearinghouse can’t act in stress because it gets two different instructions from two different supervisors, to my mind opens the possibility that we take something which is supposed to be a risk management machine in stress and turn it into a risk transmission machine.
The last thing I’d say is that, in a crisis, you have to allow the clearinghouse to take the actions that it is preconditioned to take to protect its members. You can’t ask it to forebear, because if you do that, you just transmit the problem right across the system. If a CCP resolution authority has to take such action, there needs to be, as I have said, a very strong “no creditor worse off” protection for the clearing members. One of the concerns that people have is, what would happen in a crisis if the clearinghouse puts a member firm into default? It might make the crisis much worse. A large part of this goes to the resolution agenda for banks. If there’s a credible bank resolution regime, and the resolution authority is taking the CCP member bank into resolution, then in resolution the clearing member should be able to continue meeting its obligations. If you’ve got a bank that’s in trouble and it goes into resolution, and it keeps meeting its margin calls because the resolution authorities have bailed in the capital to restore its solvency and ensure it has sufficient liquidity, then the clearinghouse shouldn't need to put it into default. If a CCP calls a default because a member has gone into resolution, then it can actually kill the possibility of a successful resolution. So all the parts of the system—the resolution authorities and the supervisors—need to work together in a crisis, domestically and internationally. There is no way to avoid that.
Up to now we’ve been talking here about something that is a theoretical issue. Hopefully it will never happen. But there’s something that’s not theoretical at all. Brexit is coming. People are wondering, how will this affect the oversight of U.S. or other clearinghouses that are operating in the U.K.?
We do not envisage there will be a big impact, if any, on non-U.K. clearinghouse operations in the EU post Brexit. There will have to be changes in regulatory authorities. When we leave the European Union, we will have to recognize overseas clearinghouses to operate with U.K. entities. At the moment, we don’t have the power to do that because the recognition of overseas clearinghouses is done for the whole of the European Union by ESMA [the European Securities and Markets Authority]. So we could not recognize CME, for example, to operate with U.K. entities until after we leave the European Union. We don’t have the powers to do that and we won’t be able to do that until the EU regulatory regime ceases to apply to the U.K.
We are very clear, however, about the need to minimize uncertainty and unnecessary disruption. So we’ve written to all of the third-country clearinghouses that are operating in the U.K., and to their supervisors, explaining our approach, which is as follows. First, when the EU regime ceases to apply to us, we’ll take the European Union rulebook onto the U.K. statute book. We’ll bring the European rules across. They will become U.K. law, not EU law. At that point, the Bank of England will become responsible for recognizing overseas clearinghouses.
Second, we will work on the presumption that, if clearinghouses have been authorized or recognized by the European Union, they should be recognized in the U.K. We will need to see the relevant information that they provided to EU authorities as part of the process. But the presumption is that if they are recognized or authorized now, they will be recognized when the EU regime ceases to apply in the U.K.
Next we will need a supervisory cooperation agreement with the home supervisor of the CCP. That’s what the European Union currently requires and what the U.K. regime will require once the EU regime no longer applies to us. The last thing is we won’t have the powers to do all of this until Brexit comes into effect and the EU regime no longer applies in the U.K. But we’re proposing that prior to that we ask firms and supervisors in home jurisdictions to get ready so that the process can happen very quickly. We want to engage with the firms to get the information so that, day one, when we leave, we can carry out the recognition so there is no gap.
So in writing to the firms we have made clear that, if you’re interested in operating in the U.K. after Brexit, engage with us now. The presumption is if you were authorized or recognized as a clearinghouse in the EU, that you will be recognized when we bring the law across, but we will need supervisory agreement with your supervisor. And we’ve written to the supervisors in a number of different jurisdictions explaining that.
There are two other considerations. One is the question of when the EU regime will no longer apply to the U.K.? There is now political agreement between the U.K. and EU that there will be an implementation period that will take us to the end of 2020. This will not be finally legally binding until the Withdrawal Agreement is ratified and signed. We have dealt with any uncertainty that could result from this by getting the Treasury to give us the fallback of an interim recognition regime. So that in the unlikely event that the EU regime is dis-applied in the U.K. before the end of 2020, we’ve got a fallback to ensure no discontinuation. This allows the Bank, the clearinghouses and the firms to plan on the basis that the EU legal regime will cease to apply to the U.K. at the end of 2020.
The second consideration arises from the government’s proposal to bring the European rulebook over to U.K. domestic law and the fact that the European rules on recognition of third-country clearing are in the process of being changed by the proposal that’s now in the European Parliament and Council. So we don’t know what European law will look like at the point the EU legal regime ceased to apply to us, because it’s “in flight” at the moment. To address this point we have made clear to firms, and more publicly, that our presumption is we will operate under the current rules as they are the only rules now in force in the EU.
You asked about U.S. clearinghouses. I've talked to [CFTC] Chairman [Chris] Giancarlo about this and explained to him the approach set out above: where jurisdictions and clearinghouses have been given equivalence and recognized by the European Union, our presumption is that we will bring that recognition across when we have the powers. The aim is to ensure that there won’t be any break in recognition or disruption. We work closely with the U.S. authorities on a range of cross-border and clearing issues and expect that strong collaboration to continue through the Brexit process.