Jeremy Grant has raised an interesting point in his Quick View in the FT Trading Room, where he discusses the debate in the US about whether central counterparties (CCPs) should have access to the Federal Reserve’s Discount Window in a crisis.
This debate brings into focus the curious No-Man’s Land occupied by CCPs. Just what kind of institution are they? And where do they fit in the financial regulatory structure?
In the US, some of them fall under the SEC (if their business relates to securities) and some fall under the CFTC (though in the case of CME’s clearing house, it is not actually a separate legal entity, but just a division of the exchange). The UK, unusually, has a specific category called “Recognised Clearing House” but in continental Europe the major CCPs, such as LCH.Clearnet SA and Eurex Clearing, are banks, more or less by definition. Try asking a German official how you can set up a non-bank central counterparty and he will have difficulty understanding the question. Everything is a bank. Well, actually not everything, since EMCF in the Netherlands had to set up a special voluntary arrangement to be supervised by the central bank and financial markets regulator, until the Dutch authorities develop a new legal framework for CCPs.
No wonder there is confusion about whether CCPs should have access to central bank facilities. Where they are constituted as banks, the answer is straightforward: as banks they naturally have access to central bank facilities. One can imagine this is why the ECB emphasises that a CCP operating in euros should be located within the euro area, since only banks within the euro area have access to ECB liquidity facilities (not that they want to say this publicly). For the non-bank CCPs, the position remains ambiguous.
In any case, what are the practical implications of the question? In other words, when would it actually matter? In business as usual, central banks only provide liquidity against eligible collateral. Thus, access to a central bank facility would help a CCP suffering a liquidity problem. Suppose it suffered a default from a clearing member, whose positions were adequately covered by margin posted at the CCP in the form of collateral, but for some reason the CCP could not quickly turn that collateral into cash. In those circumstances, the ability to use that collateral to obtain immediate liquidity from the central bank would clearly help the smooth functioning of the system and is not in any sense “bailing out” a CCP that got its risk management wrong.
Alternatively, suppose the CCP did get its risk management wrong and the defaulting member had not posted enough margin to cover its position, so the CCP was unable to make good its counterparties. In this case, access to normal central bank facilities would not help. Central banks do not lend without collateral in normal circumstances. It then becomes a question of how big a mess it is likely to be. As we saw in 2008, if the mess is big enough, central banks can be remarkably imaginative in finding ways to support institutions in difficulty (such as AIG), whether or not they had a priori access to central bank facilities. But they will never explain in advance how far they will go or when. Deliberate ambiguity on this question is how central banks give everyone else sleepless nights.