Streetwise Professor

News reports state that Geithner supports mandatory central clearing of OTC derivatives transactions.   His testimony is not quite so blunt, but could be interpreted that way:

Fourth, we should establish a comprehensive framework of oversight, protections and disclosure for the OTC derivatives market, moving the standardized parts of those markets to central clearinghouse, and encouraging further use of exchange-traded instruments.

I have argued against such a mandate here (“skinny” version), and here (“supersized” version).   Geithner is not specific about the organization of a clearinghouse, and in particular about the scope of products each would clear, but Darrell Duffie has argued that a specialized CDS clearinghouse would increase risk.   I’ve also laid out many of my arguments (that eventually showed up in the papers linked above) here on SWP–just type in “clearing” on the search bar if you’d like to find them.

Given the barrage of words I’ve already hurled at the subject, I’ll limit myself here to a bullet point list of the problematic issues:

• Risk pricing.   Dealers are likely to have better information about counterparty risks arising from the nature of the specific instruments, and especially from the balance sheet risks of counterparties.
• Netting is not necessarily a social benefit.   It redistributes wealth (in the event of a default) to clearinghouse members and from the other creditors of a defaulter, e.g., buyers of the defaulter’s commercial paper, its securities lending counterparties.   It is not immediately obvious that this allocation of counterparty risk is welfare improving.
• Clearing concentrates counterparty risk and redistributes it among the members.
• This concentration is especially problematic since derivatives counterparties have recourse to a dealer’s entire balance sheet in the event of a default.   In a CCP, recourse is limited to the capital of the CCP, plus any additional “good to the last drop” obligations of members to contribute additional capital in the event that the CCP funds are exhausted.   These funds appear small to the potential loss in the event of a major default.   As an example, CME clearinghouse can tap into $6 billion of capital if margins don’t cover a default loss.   (This includes CME stock deposited by clearinghouse members, valued at its current price too, which would likely fall in the event of a major default.)   The size of ICE Trust’s Guaranty Fund (the backstop monies that would be available in the event that margins don’t cover a defaulter’s obligations) is not publicly known (at least I have not been able to track down that number).   ICE has committed $10 million, with a commitment to raise it to $100 million.   I think members are obligated to kick in $20 million to the fund, but that could be scaled as position sizes increase.   I have not seen any indication that there will be “Maxwell House” (H/T Lucio) features in ICE Trust.   Anyways, we’re talking less than $1 billion here.
• CCPs have a more rigid, mechanical collateralization mechanism.   This could be a feature, but it could also be a major bug in times of systemic stress.   CCPs require posting of collateral in cash or close substitutes, sometimes intraday, and no later that the opening of business of the next day.   Dealers can be more flexible with collateralization, extending credit, or accepting a wider range of collateral.   Large market moves create large demands for cash in a CCP structure.   This can contribute to a liquidity crisis.   Note in October, 1987, many banks were unwilling to extend credit to fund margin calls.   Fed action alleviated a collapse of CME and CBT clearinghouses, but the margin induced demand for liquidity nearly led to a systemic collapse.   (Bernanke actually wrote a paper on this in the 1991 RFS).

I have seen no evidence that any of these points have been given serious vetting by the powers that be.