The Bank of England recently published a paper “Dear Prudence, won’t you come out to play? Approaches to the analysis of central counterparty default fund adequacy” (Financial Stability Paper No. 30 – October 2014) authored by David Murphy and Paul Nahai-Williamson. It is a timely and interesting look at how the central bank views CCP risks. And we think the Beatles reference was pretty funny.
There has been a fair amount of press recently on (the need for) CCP stress testing and additional capital. It is understandable that those who rely on CCPs would want default risk absorption buffers as deep as possible and their risk management process totally transparent. An argument put forth by JPMorgan recently was that the CCPs themselves needed to put more money in – their own capital – and not just rely on assessing the CCP’s members. This is similar to reports published by Blackrock and PIMCO advocating additional capital from the CCPs (aka “skin in the game”). See our post from last week “PIMCO and BlackRock on CCPs: unpacking the buy-side wish list”. This money would likely have to come from increasing their capital. We are reminded of the capital that LCH needed to raise and how it almost put a wrench in their merger deal with the LSE.
Back in October, 2011 we wrote a post “Bank of England’s Tucker: “There is a big gap in the regimes for CCPs – what happens if they go bust? I can tell you the simple answer: mayhem.” and it doesn’t look like much has changed. Our post from October 17, 2014 “Brilliant comment from the Chicago Fed’s McPartland on fantasy of CCP default procedures” drives the point home. But jumping to conclusions about what happens when a CCP defaults needs to be tempered by the adequacy of the protection put in place to absorb defaults in the first place. It is like panicking about the condition of the road without knowing if you are on a scooter or in a tank.
So what does the Bank of England have to say? From the introduction of the paper:
“…Our results give some reassurance in that we find that CCPs meeting the cover 2 standard are not highly risky provided that tail risks are not distributed too uniformly amongst CCP members. They do however suggest that CCPs and their supervisors should monitor this distribution as central clearing evolves….”
The paper looks at the risk management standard for CCPs: “cover 2” (from Principal 4 of “Principles for Financial Markets Infrastructure (Committee on Payment and Settlement Systems and Technical Committee of the International Organization of Securities Commissions (2012)). This has come to mean that a CCP should have enough loss absorption capacity to cover a simultaneous failure of their two largest members.
Note: While “cover 2” can be onerous in smaller emerging markets where centrally cleared derivatives might be concentrated in a handful of banks (we wrote about this in Finadium’s November 2012 paper “CCPs and the Business of Collateral Management”) “cover 2” it is about as close to a global norm as there is. That kind of agreement is pretty rare these days…
We should point out that “cover 2” refers to protection in a stress scenario. This is not to be confused with the risk absorption function of initial and variation margin. IM is there to cover adverse movements of a trade (or portfolio) after a counterparty defaults and variation margin payments stop (aka the “defaulter pays” model covering “ordinary conditions”). IM and VM are not likely going to be enough to protect against full-on major dealer defaults. Other parts of the risk waterfall are in place for that. They are typically some combination of the CCP’s capital (although the paper notes this is not a requirement in all jurisdictions), a member provided default fund, the ability to assess members to top up the default fund, and “…other arrangements, such as haircutting variation margin, to allocate losses. More of the CCP’s equity may be available to absorb losses too…”
The authors put together stress tests using “stressed losses over initial margins (SLOIMs)”. In other words various stress scenarios were constructed and portfolio values, less initial margin available, was calculated. Losses are listed by scenario and ranked by magnitude of loss. The maximum combined loss for the top two clearing members for a given scenario is calculated – and this is the amount of exposure that needs to be covered to satisfy “cover 2” utilizing default fund cash, capital, and other resources.
From the report:
“…There are therefore two principal drivers of the SLOIM: the scenarios used; and the portfolios that they are applied to. The former should be revised when new vulnerabilities become apparent, but they often do not change from day to day. The latter however are highly variable, and hence CCPs must ensure that their stress tests are genuinely stressful for the portfolios they clear, and then perform these stressful stress tests daily on each of these portfolios..”
This model is, on the surface, a little peculiar.
“…It is clear from the description above that the cover 2 requirement only depends on the largest two stressed losses over IM in the worst scenario. The risks of the clearing members who are not one of the two largest do not contribute at all to the requirement. This makes cover 2 a rather unusual risk measure: typically, adding incrementally more risk to an entity increases its measured risk…”
We won’t go into the machinations the authors went through, although it makes for interesting, if a bit wonky, reading. They look at three fundamental member risk distributions:
- Whale, where 40% of the SLOIM is held by one clearing member and the remaining follow a exponential distribution
- A purely exponential distribution, where a small number of clearing members a responsible for the majority of the total exposure
- Uniform distribution among clearing members
One interesting finding was that under the uniform distribution scenario “cover 2” might be weaker.
“…Certain distributions of risk among clearing members, such as the uniform distribution…give rise to situations where cover 2 is less prudent for CCPs with many clearing members. If these are found in practice, higher levels of financial resources may be needed to ensure clearing house robustness…”
The driver here was that the required “cover 2” is smaller when the risk distributions are equal. However the market doesn’t usually evolve into equal distributions – exponential distributions are more the norm.
The study suggested that perhaps the financial resources available to a CCP might want to go beyond “cover 2” and use a “cover all” metric and that both “cover 2” and “cover all” be made publically available.
“…Perhaps a simple backstop to cover 2 could be considered, such as demanding that the default fund in addition meets the requirement that it is larger than some fixed percentage of the ‘cover all’ requirement. One basic requirement for calibrating this percentage would be knowledge of the ratio of cover 2 to cover all, so a reasonable first step would be the disclosure of both measures by all systemically important CCPs…”
Finally the authors suggest robust monitoring infrastructure be implemented – including
“…monitoring the CDS spreads of clearing members and constructing a plausible default copula for them; ensuring that CCP default fund scenarios are sufficiently stressful without being implausible; and reviewing changes to the CCP counterparty credit risk distribution as business changes are all potentially insightful…”
This is an important look at how CCPs are risk managed. While somewhat theoretical, it can provide a good way to frame the CCP risk management debate.