RDP 2015-02: Central Counterparty Loss Allocation and Transmission of Financial Stress 1. Introduction

Since the global financial crisis, the G20 has overseen an ambitious program of regulatory reform in financial markets. One goal of the reform program is to ‘make derivative markets safer’ by reducing interconnectedness, improving counterparty risk management and increasing transparency. An important step towards meeting this objective is the 2009 commitment by G20 Leaders that ‘all standardized OTC derivative contracts should be … cleared through central counterparties [CCPs]’ (G20 2009). Further, policymakers have developed standards that require current and potential future counterparty exposures to be collateralised where contracts cannot be centrally cleared.

The resulting increase in the importance of CCPs in OTC derivative markets is well documented. It has been noted by many commentators that, given their central role, CCPs could be a channel for the transmission of financial shocks.^[1] In this paper, we build on the conceptual framework of Heath et al (2013) to gain a better understanding of how the potential for transmission of stress can be mitigated by the risk management and loss allocation arrangements established by CCPs.

In contrast to Heath et al (2013), we use actual rather than simulated data on derivative positions, as well as banks' Tier 1 capital and liquid asset holdings. The analysis in Heath et al highlights the importance of considering a network that extends beyond the ‘core’ of the financial system. Reflecting this, our sample, which is based on the data collected for the Macroeconomic Assessment Group on Derivatives (MAGD) coordinated by the Bank for International Settlements, includes the 41 largest bank participants in global OTC derivative markets in the fourth quarter of 2012. This extends well beyond the core of 16 highly interconnected dealer banks to also include banks that have fewer counterparties. Using the available data, we consider market participants' positions in five OTC derivative asset classes (interest rates, credit, currency, commodities and equity) and explicitly model exposures and collateral requirements under scenarios with different clearing configurations.

We apply a variant of the methodology in Heath et al (2013) that simulates extreme changes to OTC derivative prices and directly traces the propagation of contagion through the system. This analysis supports the view that a well-designed CCP operating in accordance with international risk management standards can be a source of stability in the system, rather than a source of instability. Fundamental to the analysis is the observation that a CCP cannot generally be a trigger for initial stress in the system. A CCP does not typically take on discretionary risks, only assuming financial risks that arise from the positions it clears for its participants. A CCP seeks to maintain a balanced position at all times and is exposed to potential stress only if one or more of its participants default.^[2]

Consistent with international standards, to limit the propagation of stress in the event of a participant default, a CCP collects initial margin from each participant to cover at least 99 per cent of potential price changes in the products that it clears. It also maintains an initial buffer of pooled resources to ensure that it could withstand the default of the participant (or, for larger CCPs, the two participants) to which it has the greatest credit exposure in extreme, but still plausible, market conditions.

If, however, either the CCP experienced multiple participant defaults or the market conditions prevailing at the time of these defaults were more extreme than the scenarios considered when calibrating its additional resources, the CCP's available financial resources could be exhausted. To ensure that it could still meet its obligations to non-defaulting participants in such a scenario, the CCP would allocate any uncovered losses to its participants. One way to do this would be to ‘haircut’ variation margin that participants were owed. While such loss allocation could be a channel for transmitting stress to participants, our analysis demonstrates that even in a range of very extreme scenarios, any such losses would be sufficiently widely dispersed that stress would be well contained.

After beginning with some background and relevant literature in Section 2, we turn in Section 3 to the key inputs to our analysis. In particular, we describe the dataset, the scenarios under consideration, and some of the key exposure metrics used in the analysis. Section 4 presents the analysis of contagion by modelling the design features of CCPs and the propagation of extreme shocks through the financial network. The results of this section show that, while introducing CCPs creates critical nodes in the financial network, if they are designed and operated in accordance with international standards, they can be expected to increase stability and reduce the propensity for contagion. Section 5 considers the policy implications, and Section 6 concludes.

Footnotes

See, for example, Pirrong (2011). [1]

The notable exceptions to this general observation are the general business risks that a CCP assumes and the risks associated with its reinvestment of cash collateral. International standards place tight limits on a CCP's discretion in these activities. [2]