A simulation analysis of systemic counterparty risk in over-the-counter derivatives markets

In this paper, we propose a simulation framework to assess systemic risk in over-the-counter derivatives markets. We incorporate credit valuation adjustment (CVA), a mark-to-market estimate of counterparty credit risk booked on a bank’s balance sheet, into an otherwise standard structural model of credit risk. In this model, banks optimally hedge CVA by trading a credit default swap (CDS). The model aims to capture a possible adverse effect called “CDS–CVA feedback loop” from CVA hedging, which could increase CDS spreads due to a lack of liquidity in CDS markets and even further increase CVA because CVA is valued using the default probability extracted from CDS spreads. In order to measure systemic counterparty credit risk, we aggregate CVA across banks and examine how the distribution of systemic counterparty credit risk changes depending on underlying model parameters. We document that the tail risk of CVA increases nonlinearly when the liquidity of CDS markets declines. As an extension, we also model cost of posting collateral and discuss the trade-off between reducing systemic counterparty credit risk, stability of CDS markets and collateral cost.