Risk: A number of regulators have published rules on central clearing – the Dodd-Frank Act was passed into law in July 2010, the final version of Emir has been published, and regulators in Canada, Hong Kong, Singapore and Australia, among others, are consulting on the issue. There appear to be several areas of difference in those rules – in particular, some regulators have proposed a location requirement for local currency derivatives. What implications does this have?
Stephen O'Connor: There is academic research out there that essentially says one clearing house per asset class is the most efficient and systemic risk-reducing model – that should be the aim. If you get to the situation where each country has its own clearing house for products in its own currency, then that introduces massive inefficiencies, both from a risk perspective and from a capital and liquidity consumption perspective. It will also have a knock-on effect on liquidity in those markets, so it becomes much more expensive to trade. As an association, we would prefer to have one clearing house per asset class. However, we are respectful of local regulatory needs and wishes, so I think we’ll have to strike a balance.
Guillaume Amblard: Today, LCH.Clearnet clears 17 different currencies. The systemic risk and the liquidity drain on the industry would increase quite a lot if there were 17 different central counterparties (CCPs). If you have a proliferation of CCPs, you start creating a very fragmented, dislocated market. If you cannot net risk from one CCP to another, it will result in a greater liquidity drain, greater risk concentration and less transparent markets. People sometimes mix up the notion of CCPs and trade repositories. At the end of the day, it is not so important – except maybe for national pride – to have a CCP in your country. What is crucial is for regulators to have access to the risk, so it is important for every regulator around the globe to have access to a trade repository. In that sense, the consultation paper by the Monetary Authority of Singapore is quite welcome, because it recognises the risk of fragmentation and shows openness to the idea of a global CCP, as long as the regulator has access to a trade repository. That is why the process of education is so crucial.
Stephen O'Connor: I would agree 100%. The data repositories should be seen as a giant spreadsheet that regulators can access and sort by any row and any column to see exactly who is doing what in their currency. That is an enormously powerful tool, which doesn’t necessarily have to accompany a clearing house. The other thing is that the global clearing-house model at LCH.Clearnet has been tried and tested during the Lehman default. If that book was spread around 17 different clearing houses in different countries, with longs and shorts in different places, it would have been chaotic from a close-out perspective.
One clearing house per asset class is the most efficient and systemic risk-reducing model – that should be the aim
Risk: Nonetheless, it does seem as though regulators are moving towards a scenario where there are domestic CCPs. Assuming that is the model, how can this be dealt with? Is interoperability the answer?
Eric Litvack: The short answer is: not exactly. In theory, interoperability gets you there, but it has never really been done – not with derivatives. It would be very complex to have a framework in which the different CCPs interact and offset risk between each other. The problem is, you run the risk of creating a tremendous amount of complexity, where the whole system effectively becomes subject to its weakest link. So I would be reticent to put too much faith in interoperability, which is unproven at this stage. That is reflected in the text of Emir, which has pushed all consideration of interoperability for derivatives into the future, because it is just too complicated at this stage.
Michele Faissola: All of these proposed clearing houses at a national level are start-ups, so you need to start thinking about different legal systems, different regulatory systems and different operational systems. The complexity you might end up with is such that the original G-20 desire to reduce risk is actually going the opposite way – definitely from an operational perspective. These are very large, very complex new companies starting from scratch. I think we need to be careful and very sensible.
As an industry body, we clearly welcome some sort of competition. At the same time, there has to be a discussion about how we regulate the systemically important CCPs and we need to ensure there is consistency in risk management.
Guillaume Amblard: It is quite important to harmonise all the different risk and control procedures across CCPs. Fundamentally, these are systemic entities, and each one needs to be controlled in a very industrial fashion. One thing to realise is that interoperability is not an easy process. It took five years for the cash equities market to establish interoperability, so it doesn’t happen overnight. Interoperability will probably also create an extra liquidity drain, because CCPs will have to charge extra protection to take the risk of other CCPs.
Stephen O'Connor: I see this fragmentation as a backward step for systemic risk reduction. If you think about the drivers for clearing in the first instance, the market was a web of dealers each dealing with its own clients – it was a tangled plate of spaghetti. One clearing house takes that mess and puts it into a hub-and-spokes system, so each market participant does not contaminate other market participants if it defaults. Moving to a system where you have multiple, fragmented CCPs with interoperability goes back towards the tangled web approach.
Interoperability also gets very complicated when you start to consider CCP resolution, because clearing houses will potentially be the biggest counterparties of other clearing houses. They will probably be in different jurisdictions, so resolution becomes very challenging.
Risk: There are a lot of models up for discussion, and the European Central Bank, for instance, has suggested that clearing houses should be domiciled in the country of the currency in which they clear. That seems to suggest the end of multiple-currency clearers such as LCH.Clearnet, which clears 17 currencies. What is the industry response?
Stephen O'Connor: I think this is a tricky question. CCP resolution is a hot topic right now, and there is a natural hesitancy on the part of central banks over whether they should be lenders of last resort to clearing houses. I think efforts should be made to avoid that – essentially, the members of the clearing house need to work with clearing houses to come up with an end-game that doesn’t call for central bank liquidity. At that point, you are not tied to the jurisdiction of a clearing house.
Michele Faissola: It is a controversial point right now. The UK has taken this issue to the European court, and we shall see how this develops. In my mind, if we think the CCP is the most systemic entity, then there has to be some connectivity to a lender of last resort. The jurisdiction angle might create a further complexity but if the CCP is well designed and well capitalised and the clearing members have high minimum requirements, then you have a system that can work. One of the issues that needs to be addressed is how CCPs are regulated and the identity of the clearing members, because that is the first line of defence.
Guillaume Amblard: I think we all agree we should limit the number of CCPs. I would agree we need to ensure all the clearing members have enough capital and that the CCPs are very solid on a stand-alone basis and have very solid risk management procedures. At the end of the day, it is important for the market to think CCPs are safe. As a last resort, central banks should be there in the background, but we need to make sure we don’t have to rely on them. Central banks and regulators need to communicate with each other globally and we need to make sure that internal politics doesn’t create systemic risk.
Stephen O'Connor: I would add that, in a world where there is an international clearing house that dominates a certain asset class, the central bank for that country doesn’t necessarily need to be the one to step in if the robust processes and capitalisation fail for some reason. Some international solution is probably more appropriate in that case.
The week on Risk.net, November 25-December 1, 2016Receive this by email