Crowded Trades: Are Clearing Houses Immune From Systemic Risk?

Crowded Trades: Are Clearing Houses Really Immune Against Systemic Risk?The sudden decision by the SNB to remove the longstanding cap on the Swiss Franc against the Euro took markets by surprise, causing many casualties amongst the foreign exchange broker community. As stated by the Financial Times on January 19, “In one of the most damaging currency swings in the modern trading area, the Swiss Franc soared in value, leaving investment banks across the world with big losses and hitting foreign exchange brokers particularly hard”.

The dramatic repercussions from the Swiss franc volatility on many brokers sheds new light on the threat of systemic risk resulting from counterparty credit default.

In this post, I will highlight the key findings from a recent research about systemic risk for central counterparties. I will share my views about the business case for aggregate exposure and the technology requirements. Finally, I will explore the benefits that Central Clearing Counterparties (CCPs) could gain from relying on a new breed of systemic risk measures.

At first glance, it seems that the issues around the risk of clearing house default have been solved. The centrally-cleared model for derivative trading was mandated by regulatory bodies such as Dodd Frank and EMIR in order to precisely avoid the accumulation of exposures that could propagate credit losses through the financial system. It was enforced in order to tackle the shortcomings in the OTC bilateral system that accounted for the post-Lehman financial crisis. In this model, the CCP becomes the counterparty to each trade and acts as a trusted third party between two market participants. By requiring each individual member to post collateral in response to initial margin calls, CCPs provide for risk mutualization and greater resilience to shocks resulting from market instability. In fact, initial margin can be considered as the first line of defense for a CCP against the default of a clearing member.

Crowded Trades: Are Clearing Houses Really Immune Against Systemic Risk?

However, the similarity in roles between brokers on the retail FX market and CCPs on the wholesale derivative market raises the following question: how likely is it that what happened to brokers such as Alpari will also happen to large CCPs?

Recent academic research[1] published by Professor Albert J. Menkveld from the VU University Amsterdam, reopens the debate around the contribution of CCP default to systemic risk. During the “leading interdisciplinary forum on market microstructure” which we attended last December in Paris, Mr Menkveld emphasized that counterparty exposure now concentrates within the CCPs. He also shared his intuition that the hidden concentration risk associated with crowded trades may have been overlooked. Quoting his research, “Losses in members’ portfolios become correlated when their trades crowd on a single security or risk factor. If this factor experiences a large price shock, multiple members experience large margin calls simultaneously. If unable to post new collateral, the CCP might have to cover losses not on a single portfolio but on many portfolios at the same time”. In fact, this is exactly what happened to the FX brokers whose customers were heavily concentrated on the CHF/Euro positions.

This research clearly questions the methodologies being used to calculate initial margin. Traditionally, CCPs monitor their exposure by calculating initial margin requirements on a member-by-member basis. What this research tells us is that this approach fails to detect the risk of crowded trades. First, it does not measure the level of concentration of members’ positions on a specific risk factor. Secondly, it does not measure the degree of correlation between all members. Since crowded trades do not show up in individual members’ portfolios, a way of effectively assessing crowded risk could be to have visibility into the CCP aggregate exposure at a global level. The research paper unveiled that the CCP aggregate exposure becomes more volatile when its members’ trades crowd. This forms the basis for suggesting the use of other measures to complement current initial margin approaches. One of them is the “Crowd Index”, a measure to assess the extent at which trades crowd.

Crowded Trades: Are Clearing Houses Really Immune Against Systemic Risk?

Calculating the crowd index metric, let alone CCP aggregate exposure, is no trivial undertaking. In fact, it poses a clear data aggregation challenge due to data volumes, calculation complexity and the real-time factor. In order to obtain the aggregate exposure, the CCP needs to aggregate the individual positions of all its members at the same time. Considering the outstanding volume of trades cleared by CCPs for their members, this represents massive amounts of P&L stress test vectors to aggregate and derive with risk measures.

Needless to say, that technology plays a fundamental role. Unfortunately, many legacy systems are running the risk of collapsing under the strain of the data being processed and the complexity of the calculation that must be undertaken in real-time. In the light of this, a valid option for CCPs to consider is the adoption of an in-memory real-time aggregation platform that would deliver a new breed of counterparty risk measures such as the Crowd Index advocated by Mr Menkveld. At Quartet FS, we have helped several investment banks (and a large CCP) assess market risk exposures throughout the day. Our ActivePivot in-memory analytical platform has been natively designed to handle the aggregation of massive volumes of data to compute risk measures of similar complexity, based on P&L stress tests. Examples of these measures are Expected Shortfall, VAR, PFE, CVA. The Crowd Index is based on the same calculation pattern.  An embedded rule-based monitoring engine can automatically track these risk measures and generate alerts intra-day, thus providing effective management by exception.

I believe that there is a clear business case for CCPs to implement an aggregate exposure analytical solution. First, it would allow the CCP to detect if many of its members experience the same price shock simultaneously. Secondly, by asking members whose trades crowd to pay initial margin add-ons, the CCP would encourage them to clear their trades with another CCP. Not only would this reduce the crowded trade risk faced by the CCP itself, but it would also spread the risk across many CCPs, thus contributing to a more stable and resilient financial system as a whole.


1. “Crowded Trades: An Overlooked Systemic Risk For Central Clearing Counterparties“ – by Albert J. Menkveld – March 15, 2014

Like this blog?
Follow us on LinkedIn
to stay up-to-date with the latest blog posts!

This entry was posted in Finance. Bookmark the permalink.

Comments are closed.