Given Brexit the ECB-wants-clearing-to-be-in-Eurozone thing has become important again — it had previously been rebuffed by the CJEU, albeit without going too much into details. It is often seen as an aggressive move to get the hands on a profitable business (and it might well be as far as politicians are concerned), but there are actually very good reasons why the ECB wants it that way.
Let’s start with a quick reminder what clearing actually is. Historically, clearing has been introduced as securities clearing, typically related to securities traded on an exchange. The issue that clearing was meant to address that transferring the ownership of a security wasn’t a straight forward task — not too long ago it still often involved physically exchanging one pile of papers (cash) against another one (the securities certificates). Now when you are trading on an exchange you are matched with a random person who wants to do the equal and opposite transaction at the same time. That’s good for liquidity, but as you have no control over who this person is you can’t really check whether this person can be trusted before you traded.
Exchanges addressed this issue by introducing clearing houses who’d step in between the two counterparties and guarantee their respective performance. The way it works in detail is that everyone trading on an exchange needs to go through a clearing member. This clearing member guarantees that the ultimate counterparty fulfils its obligations, ie either to come up with the cash, or to deliver the security certificate. To ensure that the clearing members does not default they have collateral (cash, interest bearing securities etc) on deposit with the clearinghouse. So if a customer agrees to buy a stock in the exchange and the counterparty does not deliver, worst case the clearing house purchases the stock on the open market at whatever price it is available, delivers it to the customer against the agreed payment, and takes the difference out of the collateral posted by the clearing member.
Derivatives and margining
A financial derivative is a contract to exchange cash flows based on the value of a financial indicator (the underlying). An example for a derivative is an interest rate swap where the underlying is a floating interest rate, for example LIBOR. A specific example would be a 10 year, €100m notional swap at 1%; here every year for 10 years one side pays fixed (ie €100m x 1% = €1m) and receives floating (ie €100m x LIBOR).
Initially a swap is typically worthless, in the sense that the fixed payment equals the expected floating payment. However, as rates move this is no longer that case. Assume the day after the above swap has been traded, rates move to 2%. Now the counterparty paying fixed is still paying 1%, but the floating leg is expected to be €2m, so the net payment is €1m per annum, or €10m in total, and slightly less in present value terms due to discounting.
Generally every derivative will over time have one side who is in-the-money and one who is out-of-the-money. If the latter defaults, the former incurs a loss equal to the present value of the derivative in question, and for some players — especially the large dealers — that can be quite a large sum. To mitigate the risk, derivatives are often margined. What this means is that the counterparties agree to exchange collateral equal and opposite to the net present value of the derivative, throughout the entire life of the contract. In the example above, the counterparty paying LIBOR would have had to post slightly less than €10m in collateral after the market move; in case of a defaults the other counterparty would have simply gone back to the market, and entered into a replacement contract, using the collateral received to compensate the new counterparty for the fact that the derivate is out of the money.
To come back to the example above: if rates move to 2% any new counterparty receiving only 1% fixed would expect to lose €10m over the life of the contract which means they’ll only do it if they receive the present value of those losses upfront.
Derivatives clearing is similar to securities clearing, except that the clearing house (in this case also known as central counterparty, CCP) inserts itself between the two derivative counterparties. In fact, in a process known as novation the original counterparties tear up their contract, and they enter into equivalent contracts with the CCP instead. This is an important point: even if the initial contract was between say Goldman and Morgan Stanley, once the novation has taken place there are now two different contracts, one between Goldman and the CCP, and one between Morgan Stanley and the CCP, but the initial contract has disappeared.
This is interesting because the CCP effectively consolidates those relationships. Assume there are three contracts between three counterparties that cancel out each other in a circular manner (eg Goldman pays floating to Morgan, who pays floating to Merril, who pays floating to Goldman). Once all three contracts are novated with the CCP they simply disappear.
Now CCPs are sitting between the market participants and they serve as a firewall in case a counterparty defaults. To mitigate credit risk they of course engage into margining as described above, and as privileged counterparties they are actually in a position to always receive a bit more than they’d need on a fair value basis, to ensure that if some counterparty fails to post margin after a move it is still protected.
The advantage of CCPs is that they can very effectively shield counterparties from each other’s defaults and therefore another Lehman moment can be avoided. The disadvantage is that if the CCP itself defaults the Lehman default looks in comparison like a period of exceptional calm.
The first thing to make sure for a safe clearing system is that CCPs have unlimited liquidity. Similar to a bank run, a CCP run is a very real risk: once people are worried about the soundness of a CCP (eg after a big market move) they might choose to delay or even completely avoid posting additional collateral, deciding that it is better to incur the additional cost of doing so than to lose the collateral itself. This of course does not absolve the CCP from posting collateral on those trades that moved further into the money, so unless it has a quasi unlimited credit line with the respective central bank it will miss posting payments, ie it will default.
Now the above scenario assumes that the CCP is illiquid but solvent which by design should always be the case. However, bad things happen. Firstly, we can have market moves that are so big that the counterparties’ excess collateral is all used up, and if those counterparties default the CCPs might default as well. Sizing of the excess collateral is therefore of utmost importance, and markets can probably not relied upon doing that (in competition between CCPs there will always be tendency reduce those excess margins, which impose a cost on their customers).
Also, we have seen repeatedly that systems meant to ensure compliance can fail (eg Barings, SocGen). For example, especially if margin is provided in terms of securities rather than cash, maybe the value of those securities is not what the CCP believed it to be, or even those securities are somehow not available, or pledged elsewhere and the CCP failed to notice.
So CCPs on their own are not really de-risking the overall market — all they do is to concentrate the risk onto themselves. That’s not necessarily a bad thing, even though ‘necessarily’ is doing a lot of work here — single points of failure tend to be hallmark of bad system design. It means however that CCPs must be closely supervised by the regulator who is in charge of systemic risk, and who was to make sure that
- default — even distress — of a CCP is only the remotest of remote possibilities, and
- there is a plan in case of CCP distress / default to get it back to work the next day, whatever it takes, lest markets seize up
Given all that I actually understand the even in the current context the ECB — which is responsible for systemic risk in the Eurozone — is not quite comfortable with Euro clearing being in London an therefore outside of its direct control. As an aside, not explicitly giving the ECB oversight of the CCPs — which are the most systemically important financial institutions in a financial system — when it got its systemic oversight responsibility was outright stupid by the EU.
After Brexit however — especially a Brexit where EU law no longer directly applies in the UK, and where the CJEU no longer has the ultimate say — it would be outrightly negligent to leave such a central part of the Eurozone financial infrastructure under third party oversight.