The Federal Reserve Bank of New York issued a preliminary staff paper, titled “Does Central Clearing Reduce Counterparty Risk in Realistic Financial Networks?” authored by Rodney Garratt and Peter Zimmerman. The paper examines the effect of introducing a central clearing counterparty (“CCP”) on the expected net exposures of dealers, and also explores the effect of such introductions on the variance of these net exposures.
The authors find that a CCP is unlikely to be beneficial when the link structure of the network relies on just a few key nodes. In large scale-free networks in particular, a CCP will “always worsen” expected netting efficiency. Additionally, the authors find that CCPs can improve netting efficiency only if agents have some degree of risk aversion that allows them to trade off the reduced variance against higher expected netted exposures.
Consequently, when the number of asset classes is small in relation to the number of dealers, introducing a CCP is likely to reduce both the mean and the variance of net exposures. The authors speculate that this may explain why, in the absence of regulation, traders in a derivatives network may not develop a CCP themselves. For smaller core-periphery networks, the authors found that a critical size exists beyond which introducing a CCP is unambiguously good for netting efficiency.
Lofchie Comment: Despite the complexity of the math, or because of it, the authors thoroughly debunk the myth that central clearinghouses reduce risk. One problem with clearinghouses that the mathematicians ignore is this: because CCPs have the power to demand unlimited collateral from their clearing members, CCPs will drain liquidity from the financial markets in times of financial crisis. That is, the biggest risk is not that CCPs will fail; it is that they will survive by demanding so much collateral that they crash the rest of the market.