A plan by global financial regulators to fix the mess created by the misuse of credit default swaps is flawed, says Darrell Duffie, professor of finance at Stanford GSB.
In a preliminary research paper, Duffie, and GSB doctoral student Haoxiang Zhu, conclude that the central clearing houses founded to rationalize the $27 trillion market for credit default swaps will not remove nearly as much risk as regulators might hope. What's more, despite a mistaken belief by some commentators, the clearing houses are unlikely to bring much needed transparency to trades of credit-default swaps, or CDS, says Duffie.
Credit default swaps are essentially insurance policies used to hedge risky bonds. Their misuse has been blamed for the near-collapse of American International Group (AIG) and the subsequent damage to the global financial system in an over-the-counter market, out of view and off the public record.
Because there was, until recently, no central clearing house for the CDS market, buyers and sellers have been unnecessarily exposed to the risk of default. A clearing house stands between buyers and sellers, ensuring that accounts are settled properly when trades are made, and that margin requirements have been met. In effect, the clearing house acts as a buyer to every seller and a seller to every buyer, reducing the risk of default by either counterparty, as participants in such trades are called.
Responding to pressure from regulators, dealers in Europe and the United States, agreed to the establishment of CDS clearing houses, and by early spring two had been opened and more are planned.
Duffie, a member of the Financial Advisory Roundtable of the New York Federal Reserve Bank, supported the establishment of a clearinghouse in testimony last year to the U.S. Senate Committee on Banking, Housing, and Urban Affairs. He still supports that idea, but maintains that the current implementation is flawed in several respects.
Although the worldwide market for credit default swaps is huge at $27 trillion, it has shrunk by more than 50 percent in the past year, and is too small—and the number of participating institutions is too small—for a clearinghouse that deals only in CDS to efficiently reduce counterparty risk, says Duffie. Instead, Duffie and Zhu suggest that the clearinghouse should clear a much larger fraction of trades made in the $500 trillion market for over-the-counter (off-exchange) derivatives.
"Our results make it clear that regulators and dealers should carefully consider the tradeoffs involved in carving out a particular class of derivatives, such as credit default swaps, for clearing," the research paper states. Here's why:
Banks reduce risk by trading across various classes of options, derivatives, and other financial instruments. Ultimately, positions between two counterparties tend to have offsetting exposures; some are of positive market value to a given counterparty, and others are of negative market value. These have a "netting effect," that is, only the net amount of market value is at risk in a default by one of the counterparties
Duffie and his co-author built a theoretical model to clarify an important tradeoff between two types of netting opportunities, "namely bilateral netting between pairs of dealers across different underlying assets, versus multilateral netting among many dealers across a single class of underlying assets, such as credit default swaps." The latter of these is the method by which the new clearinghouses will work.
Their model reveals that clearing only credit default swaps can actually increase the risk to the counterparties because the benefits of bilateral netting across asset classes is reduced in this case.
For instance, if Dealer A is exposed to Dealer B by $100 million on CDS, while at the same time Dealer B is exposed to Dealer A by $150 million on interest-rate swaps, then the introduction of central clearing for only credit default swaps increases the maximum loss between these two dealers, before collateral and after netting, from $50 million to $150 million. Additionally, CDS-only clearing would likely result in demands for additional, expensive, collateral to protect the two parties.
A CDS-only clearinghouse would work if the market were larger, say Duffie and Zhu. More precisely, their report finds that a dedicated central clearing counterparty [a clearinghouse] improves netting efficiency for these dealers if and only if the fraction of a typical dealer's expected exposure attributable to CDS is the majority of the total expected exposures of all remaining bilaterally netted classes of derivatives. In fact, the credit-default swap market is now too small to reach that threshold.
Making matters somewhat worse was the decision to establish multiple clearing houses. Having more than one reduces the netting effect even more, says Duffie, adding that each additional clearing house exacerbates the problem.
Even though the clearinghouse plan is flawed with respect to reducing counterparty risk, it has been suggested that establishing these new entities would at least add much needed transparency to the CDS market. Actually, the same level of information about CDS trades that would be available to regulators in a clearing house is already available through the Depository Trust and Clearing Corporation (DTCC). With or without a clearing house, there is no plan to reveal trades to the public. So, the stories of improved transparency are a red herring.
Public discussion, says Duffie, assumes that the clearinghouses would act like exchanges, such as the New York Stock Exchange, by systematically reporting all trades. "I'm sorry to disappoint, but most of the information about default swaps remains confidential even when cleared," he said during an interview.
Moreover, a clearinghouse can only clear standard transactions. But most of the credit default swaps initiated by AIG, are not standard, and would never have been cleared, even if a clearing house had existed years ago.
Presented in mid-February of 2009, the preliminary draft of the Duffie-Zhu paper is titled: "Does a Central Clearing Counterparty Reduce Counterparty Risk?" The work is something of a departure for Duffie, who says he rarely writes a paper to meet the immediate needs of a policy debate, but felt compelled to weigh in because of the critical nature of the discussion.
"During a research discussion over lunch, my co-author and I had a hunch that there was an important concept missing from the policy discussion. We could not confirm our intuition without building and solving a model. Once we did, it was obvious that we should present our results in a new research paper," he said in the interview.
Darrell Duffie is the Dean Witter Distinguished Professor in Finance at Stanford GSB and Senior Fellow, by courtesy, at the Stanford Institute for Economic Policy Research.