As best we can tell, CCPs are still considered counterparties under Dodd-Frank 165. To vastly summarize the situation, the largest banks can have no more than 25% of their capital plus survey in exposure to one SIFI counterparty (or 10% in a few cases). This is a tricky situation for banks and CCPs alike. While we have confirmed that as of August 2012 CCPs were still considered a counterparty (as is the US government, other banks, etc.), there is a strong argument that CCPs should be excluded or given special treatment. Otherwise, a bank with a single large CCP exposure would have to shift that around across multiple CCPs trading the same product. In the search for efficient markets, that situation would be awfully inefficient and would detract from overall CCP use.
Here’s our argument for why CCPs shouldn’t be considered counterparties in the same way that bilateral counterparties are: the mutualization of risk at a CCP suggests that instead of one counterparty to default, a CCP would have to suffer multiple defaults to really default itself or perhaps need outside (government) support. Technically, it would be complex but possible to divide up the membership of a CCP to assess what portion of risk pertains to which entity and add that to a counterparty exposure figure. It’s not the CCP itself that has the ultimate exposure; its the underlying community of members.
The Basel Committee on Banking Supervision recognizes CCPs as a special category under Basel III. In July 2012 the BIS released “Capital requirements for bank exposures to central counterparties.” These rules said that “Where a CCP is supervised in a manner consistent with [the IOSCO/CPSS] principles, exposures to such CCPs will receive a preferential capital treatment. In particular, trade exposures will receive a nominal risk-weight of 2%. In addition, the interim rules published today allow banks to choose from one of two approaches for determining the capital required for exposures to default funds: (i) a risk sensitive approach on which the Committee has consulted twice over the past years, or (ii) a simplified method under which default fund exposures will be subject to a 1250% risk weight subject to an overall cap based on the volume of a bank’s trade exposures.” Basel III doesn’t go into the exposure limits – that’s a US phenomenon – but they take the specialness of CCPs into account.
These same Basel risk capital requirements are echoed in the US government’s big “Regulatory Capital Rules: Standardized Approach for Risk-weighted Assets; Market Discipline and Disclosure Requirements” issued in draft in June. There was no exclusion for CCPs under DF 165 though.
US lobbyists, never slow on the draw, have had their eye on the matter for some time. The most recent document we found was from August 2012, where the Financial Services Roundtable said “With respect to central clearing parties (“CCPs”)… we believe that the Rule’s approach of including exposures to all CCPs within the definition of credit exposure runs counter to the objective of encouraging the use of CCPs. We believe that, at the very least, the Rule should be modified to exclude exposures to “eligible” CCPs from the lending limit. The [Office of the Comptroller of the Currency] would have the authority to define what CCPs are “eligible.” By excluding exposures from CCPs (or “eligible” CCPs), the OCC would further incentivize banks to move trading activities into CCPs. This is consistent with the notion throughout Dodd-Frank that trading activities of banks should be conducted through CCPs.” Here’s the document. While we generally have some skepticism on the issuances of lobbyists, we fully agree with them here: CCPs really shouldn’t be in the same pot at banks when it comes to Dodd-Frank counterparty exposures. We’re not sure who defines eligible though.
This is a challenging issue and one that we hope will be dealt with rationally by US regulators. Until the matter is settled however it remains an important question to keep an eye on.
3 Comments. Leave new
Josh, good points made here.
I’d like to add mine, though, to stir up the discussion.
I am wondering if, with the increased emphasis posed on clearing via CCPs, we are actually increasing risk, by generating big concentration of exposures to the extent of the risks posed by big “nuclear plants”.
Considering “efficient” just what it is “cheap” is reductive, as you might simply underprice risk. Big fat CCPs seem to me very close to the idea of Soviet Union trying to exploit economy of scale and economic planning to run the economy. It was wrong for many reasons, but one in particular: the risks posed by wrong planned investment were massive and concentrated: the system was unable to self correct when a mistake was made. Having big CCPs that are “efficient nuclear plants” might seem appealing at a first look; but the failure (always possible) of such big institutions could be not bearable by the system itself and generate even more ex-post risk.
As the good Hayek and Austrian school pointed out, the failure of small entrepreneurs and enterprises along with competition in providing the same services and goods will ensure that the economy survive its inevitable exuberances, mistakes and corrections. Maybe involuntarily, but the Dodd-Frank act is actually trying to achieve exactly this.
Cheers,
Simone
Risk fundamentals indicate that CCPs are counterparties and do have exposures. Key points here are:
1. There are already strong incentives (funding costs and Basel III capital) for limiting exposure to CCPs.
2. Though portfolios facing CCPs are large in notional terms they will tend to be quite risk neutral for banks if usually client trades and corresponding dealer to dealer hedges are both cleared in the same CCP (resulting in mutual risk neutralization)
3. In a mature end state post DFA mandate transition there are likely to be more than one CCP per asset class with a signficant market share – fragmenting the total risk population per bank. This is likely to reduce the size of the largest CCP exposure in end state though there may be some upward and downward volatility during transition.
4. Properly set up – a bank is mostly not exposed to a CCP for client cleared portfolios it administers (the client is) only for the trading desks exposures.
There seems a strong possibility that the 25% limit applied to CCPs won’t have material impact in practice provided proper attention is give to the above factors by each bank.
Here’s a comment letter from the CME dated April 30, 2012 with their rationale for excluding CCPs from the exposure limit: http://www.federalreserve.gov/SECRS/2012/May/20120509/R-1438/R-1438_043012_107221_587030190198_1.pdf