Client Clearing and Market Structure in CDS Markets

Disclaimer: The views and opinions expressed are those of the authors and do not necessarily represent official positions or policy of the Office of Financial Research, the U.S. Department of the Treasury, or any of the other institutions with which the authors are affiliated. 

Introduction 

Central clearing has become a cornerstone of global financial markets in the years since the 2007–09 financial crisis. In a centrally cleared transaction, a central counterparty (CCP) interposes itself between the two trading parties, thereby eliminating the risk that they fail to meet their contractual obligations to one another. Much of the extant academic literature studying the effects of central clearing has focused on clearing members, typically large dealer banks, that maintain direct access to CCPs. Significantly less work has centred on nondealer clients such as hedge funds and asset managers, though these institutions now account for the majority of the risk managed by CCPs in many markets. 

In a new research paper presented at this year’s WFEClear conference, I sought, together with my coauthors Robin Lumsdaine and Mark Paddrik, to understand the economic drivers of client clearing and its consequences for market structure and stability. We examine these topics using confidential transaction-level data on credit default swaps (CDS) from the Depository Trust & Clearing Corporation (DTCC). We find that client clearing facilitates netting and promotes dealer competition but also introduces operational fragilities during periods of stress. These findings carry direct implications for ongoing policy debates surrounding the expansion of clearing mandates, including in U.S. Treasury markets.

The Structure of Client Clearing

To clear CDS transactions, clients rely on members to act as "clearing agent" intermediaries that collect margin and effectively guarantee their obligations to the CCP. This structure, known as the agency model, is the dominant form of client clearing in U.S. futures and swaps markets, and has been proposed as a template for U.S. Treasury clearinghouses as well. 

Client clearing presents distinct trade-offs for market participants. For dealers, acting as a clearing agent generates operational and capital burdens, but doing so may strengthen relationships with clients. For clients, clearing adoption can be prohibitively expensive, particularly for smaller or less active traders, though it mitigates counterparty risk and may improve access to liquidity. Understanding how these competing influences shape trading behaviours is important not only for market participants but also for regulators. 

Incentives to Clear: Netting, Pricing and Regulatory Reform 

The first part of our analysis examines why clients and dealers choose to clear. A key benefit of central clearing is that it enables multilateral netting - the consolidation of offsetting contracts across counterparties - which reduces gross exposures and can lower margin requirements and regulatory capital charges. We provide direct empirical evidence that dealers clearing a greater share of their single-name CDS transactions maintain lower gross exposures relative to their trading volumes. 

Given that clearing reduces dealers' margin and capital costs, they have an incentive to encourage clients to use CCPs by offering more favourable pricing on cleared transactions. Our regression analysis confirms this, as clients receive better pricing terms on centrally cleared trades than on comparable uncleared transactions. 

Not all clients, however, choose to embrace clearing. The introduction of the Uncleared Margin Rules (UMR) in 2016, which increased the cost of bilateral trading, prompted many smaller clients to exit the single-name CDS market entirely rather than incur the fixed costs of CCP participation. For the subset of clients that adopt clearing, we find they are more likely to clear a given trade when their dealer counterparty is riskier and when doing so would reduce their net exposure at the CCP. 

Market Structure: Dealer Competition and Network Effects 

The second part of our analysis investigates how the growth of client clearing affects market structure. We find no evidence of new dealer entry following clearing adoption, but we find that clients forge additional trading relationships with incumbent dealers and reduce counterparty concentration once they begin to clear. These results provide concrete evidence that clearing enhances competition in over-the-counter derivatives markets, a finding that prior work has only been able to support indirectly. 

We also examine the strategic incentives underpinning dealers' provision of clearing agent services. Acting as an agent generates spillover benefits for market making activities, as clients are significantly more likely to trade with a dealer with whom they have established a clearing agent relationship. Dealers that provide clearing services are thus able to capture a greater share of client transaction volume. Furthermore, we find that dealers have greater pricing power on trades where they serve as both the transacting dealer and the clearing agent. 

Fragility in Client Clearing: Agent Concentration and Stress 

Finally, our paper examines the fragilities introduced by the client-agent intermediation structure. Despite the breadth of their dealer trading networks, most clients rely on a single clearing agent. This concentration reflects the fixed costs of establishing clearing agent relationships and the margin efficiency gains from clearing through a single member. Clients that do maintain multiple agent relationships tend to be larger and more sophisticated. They prefer clearing agents with stronger credit quality, suggesting operational concerns in the event of member distress. 

We use the sudden collapse of Archegos Capital Management in March 2021 as a natural experiment to assess the effects of member constraints on client clearing access. Credit Suisse suffered losses of $4.7 billion due to the event, making it the most severely affected clearing member institution. Though these losses did not originate from its clearing business, the amount of client margin handled by Credit Suisse dropped sharply. We show that this decline did not simply reflect voluntary client reallocation, as clients that relied heavily on Credit Suisse for clearing services reduced their overall cleared positions. The reduction is most pronounced for clients with fewer clearing agent relationships, suggesting that clients with multiple agents were able to replicate their positions using other members. This heterogeneity highlights the value of agent diversification as a hedge against member-specific distress. 

Conclusions 

Our findings have important implications for the design of client clearing frameworks. As clearing mandates expand into new asset classes, the frictions we identify become increasingly consequential. The concentration of clearing services among a small number of dealers, the prohibitive fixed costs faced by smaller clients and the operational dependencies created by agent relationships all represent potential points of fragility in a system designed to reduce systemic risk. Facilitating broad and resilient client access to CCPs will be critical to realising the stability benefits that central clearing is intended to deliver.