EMIR’s burning question: How risky are OTC derivatives CCPs?

By Dr David Murphy

No financial institution likes to remind its clients that it might fail, so it is no surprise that OTC derivatives clearing houses don’t heavily advertise their risks.  But they are risky.  A clearing house could fail in at least three different ways: 1) One or more clearing members could default, and the resulting losses could blow through the CCP’s financial resources; 2) The clearing house could suffer investment losses comparable to its capital; 3) It could have a large operational risk event which made it insolvent.

The risk of the first type of loss event is a counterparty credit risk question.  One tool that we can use to answer it is CDO pricing technology.  Here’s how it works:

  • The derivatives receivable that the OTC derivatives CCP has from its clearing members is viewed as the underlying collateral of a CDO.  Thus unlike a conventional CDO, where the collateral is bonds or loans, here it is in-the-money cleared portfolios.
  • The clearing members credit quality is modelled.  For instance, we might in the simplest instance use probabilities of default inferred from their CDS spreads and default correlations inferred from the price of some traded credit instrument such as the itraxx financials index.  This gives us a way of calculating the probability that multiple clearing members will default in some period of time.
  • We want to know not just what each clearing member owes today, but what they might owe at the point that their portfolio was closed out after a default, so we use some adjusted measure of exposure which captures these close out amounts.  Margin is then applied to reduce these exposures.
  • The exposures are combined with the clearing member default information to get a combined credit risk loss distribution for the CCP.  This allows us to estimate how probable a given level of losses over margin are.
  • The CCP is in stress if it suffers losses which exhaust its financial resources.  We can estimate how likely this is, too.  The resulting measure is a measure of the probability of clearing house default due to counterparty credit risk.

There are many choices in this process including how clearing member PD and default correlation are estimated; whether we included default fund assessment rights in CCP financial resources or not; how we calculate adjusted exposure.  All of these give rise to model risk in our estimates.  Still, the methodology is reasonable, and the resulting PDs are an interesting way of comparing the counterparty credit risk of different OTC derivatives CCPs.

David’s new book, OTC Derivatives, Bilateral Trading and Central Clearing: An Introduction to Regulatory Policy, Market Impact and Systemic Risk appeared from Palgrave Macmillan in the UK on the 7th August, and will appear in the US shortly.  This volume provides a comprehensive introduction to the bilateral OTC markets, the post-crisis regulatory reforms, and the new centrally cleared post-trade environment.  The post above is based on a part of chapter 8 where David discuss the risks which OTC derivatives CCPs run and how they can be modelled.  You can find the book at Amazon here.