In some respects, the debate over clearing has been at cross purposes. Those who advocate clearing, including many in the banking community, often focus on the chaotic events that accompany the default of a large OTC counterparty, such as Lehman or LTCM; advocates of clearing argue that a central counterparty can mitigate the disruptions associated with a large failure. Those, like me, who are skeptical of mandates (but who are not averse to clearing if freely chosen by market participants, as has occurred in energy, freight forwards, and many interest rate swaps) focus on the incentive and information effects that affect behavior prior to a default.
I confess to being less than clear about the distinction, and therefore think it’s worthwhile to address these issues in more detail.
In essence, the distinction is between ex ante and ex post. I have focused on the ex ante incentives, whereas others have focused on ex post issues. Both are important. (This parallels distinctions that transactions-cost economists sometimes draw between agency theory and property rights models, which focus on ex ante incentives, and Williamsonian governance theories, which focus on the ex post management of contractual relationships.)
Of course, the ex ante and ex post issues are not unrelated. Ex ante incentives affect the positions and risks that are undertaken, and hence affect the likelihood of a major default. Resolution mechanisms affect the cost of dealing with default. The expected cost of clearing vs. bilateral mechanisms for allocating counterparty risk is, roughly, given by the probability of a major default (which depends on the ex ante incentives) times the cost of such a default (which depends on the ex post resolution mechanism). It is not evident a priori whether this product is larger under a cleared or uncleared system, hence the grounds for legitimate debate over the merits of clearing.
There is a colorable case that the combination of multilateral netting and the existence of a central counterparty that can coordinate the transfer and replacement of defaulted positions can reduce the cost conditional on a default occurring. Multilateral netting reduces the magnitude of the positions that need replacing. This reduces the stress on market liquidity resulting from a default. Moreover, clearing facilitates the transfer of customer positions to solvent clearing members, thereby avoiding the necessity of replacing these positions via market transactions, further reducing said stress. Moreover, the information that a CCP possesses about total positions, and its ability to coordinate the hedging/replacement of the defaulted risks can reduce uncertainty and mitigate price impact. (Although it should be noted that the experience of the CME with the Lehman problem, as documented by the Valukas report, and as suggested by what I have learned from informed sources that the LCH unwind of the Lehman positions was not as breezy as LCH has suggested demonstrate that this is at best a relative statement.)
Some research that grew out of the ‘87 Crash sheds light on this last issue. An interesting paper by Greenwald and Stein (JOB, 1988) shows that normal, continuous trading mechanisms can exhibit poor performance during periods of market stress caused by a large shock to the volume of transactions (especially when this shock is accompanied by an increase in fundamental uncertainty). In essence, there are execution price risks under these circumstances that create negative externalities. Potential replacement counterparties are reluctant to trade in these circumstances because of the extreme uncertainty about execution prices during periods of large volume shocks. This tends to reduce liquidity, which tends to exacerbate the execution price risk.
This means that the uncoordinated replacement of large numbers of defaulted positions by a large number of firms through the use of ordinary, continuous market mechanisms (whether OTC or exchange) can lead to substantial price changes that are not fundamentally driven, but are microstructural in origin. This can have further knock-on effects, as these (distorted) market prices affect collateral/margin calls, can induce asset fire sales, etc.
Thus, a plausible characterization of a key trade-off between bilateral and cleared structures is: (a) counterparty risks are more efficiently priced and shared in a bilateral setting, and hence moral hazards and adverse selection problems are less acute in that setting; (b) further, hub-and-spoke clearing networks create concentrated points of failure that are more problematic than more distributed (but still concentrated) bilateral networks; but (c) a cleared system reduces replacement cost risks/price impacts conditional on default.
Which raises the question: is it possible to obtain the information benefits associated with bilateral arrangements for some transactions, while mitigating the price impact of an uncoordinated replacement of defaulted positions? Can we have our cake and eat it too?
Put differently, clearing is a bundle of functions including inter alia: (1) the pricing of counterparty risks; (2) the mutualization of losses not covered by collateral; and (3) the management of risk associated with defaulted positions, and the replacement of these positions. I argue that for many transactions and transactors, bilateral mechanisms are superior for (1) and (2); I recognize that CCPs may do (3) better. Is there any way to get the benefits of (3), without incurring the disadvantages that CCPs arguably face with (1) and (2)?
I think that this could be possible. In essence, it would involve pre-commitment to a suspension of normal, continuous trading activities in the event of a default by a large market participant (a dealer, such as Lehman, or a big hedge fund, such as LTCM), and its replacement with an auction-type mechanism. (This is the essence of the Greenwald-Stein recommendation for “circuit breakers” that replace continuous trading with a call auction. The key difference is that the Greenwald-Stein circuit breakers are price-contingent, which has some problematic features, whereas what I am proposing is default-contingent.)
ISDA implemented a “Big Bang” auction protocol to facilitate the settlement of CDS positions in the event of the default in a named credit. Although as I predicted in my post on this subject some years back, this protocol has not eliminated all problems with pricing the debt of defaulted credits (as shown in a recent Markit Magazine article), it has greatly smoothed the process of handling CDS-triggering credit events. Perhaps something similar could be designed and implemented to deal with a big derivatives default.
ISDA could of course play a central role in shepherding the creation of such a mechanism. Similarly, regulators, notably the NY Fed, could also play a constructive role, perhaps in both the design and implementation. I am pretty sure that ISDA would be quite willing to engage in such a cooperative endeavor.
The process would of course be facilitated by the development of robust trade repositories that permit rapid determination of the positions that are defaulted against. I have supported the creation of such repositories (see my Regulation magazine piece) from the outset.
A potential structure
Just thinking out loud, here are some ideas of a potential structure.
First, a compression round whereby multilateral exposures are netted. (This would have implications for bankruptcy priorities and bankruptcy law; these implications would have to be weighed carefully and addressed).
Second, an auction round for replacement of positions. This auction round would follow a disclosure, based on the information in the trade repository, of all the positions, to the potential bidders.
The auction round raises several issues that need to be considered. Who would be allowed to participate? Permitting anyone to participate could just be a jump from the frying pan into the fire if some of the winning participants are themselves in dodgy financial condition. (This is a dilemma that CME faced in its Lehman auction, and likely explains why it limited participation in the auction of Lehman positions).
Another issue is that the number of positions to replaced can be immense, and highly heterogeneous, with many non-standardized contracts. Perhaps one way to deal with the latter would be to auction off shares of portfolios, e.g., USD interest rate (swap, swaption, etc.) positions; FX positions; equity derivatives; and so on. This would mitigate cherry picking/adverse selection problems and simplify the bidding. It would, however, mean the splitting of some individual contracts among multiple counterparties. (It should be noted, though, that many of the positions would not be cleared or even clearable in any event, so this issue is not eliminated by a clearing mandate.)
Yet another issue is that those assuming positions would be taking on counterparty risk. Thus, the repository would have to disclose the counterparties to the defaulted contracts to the potential bidders.
So, I realize that there are many thorny operational issues that must be addressed to implement this proposal. But some would exist under a clearing mandate because not everything would be cleared in any event. I further realize that FRBNY and the dealer community tried to do some of this on the fly in the “Lehman weekend,” and in LTCM years before. The Lehman attempt foundered (as Theo Lubke of FRBNY, and several bankers I spoke to, indicated at the ISDA AGM). But perhaps this is not surprising as extemporizing on the fly in those conditions faces faint chances of success, particularly given the lack of information available to the participants; a pre-planned mechanism that can rely on more complete information is almost certain to work better. I would also note that in a cleared market something like this will have to be in place in any event.
The logic behind my approach is pretty straightforward. Whenever things are bundled, the immediate question should be: can efficiency be enhanced by unbundling them? (A lot of what goes on in finance involves unbundling things and allocating the pieces in a value-enhancing way.) Sometimes you can’t: so be it. But sometimes you can.
Clearinghouses bundle counterparty risk pricing, counterparty risk management (including the collateralization mechanism), mutualization, position information collection, and default resolution. There is no logic that says that those functions have to be bundled. Repositories can collect and aggregate information, perhaps more effectively than CCPs (because they can incorporate information on non-cleared positions, and information on positions held across CCPs). CCPs are not always the best at counterparty risk pricing and collateralization mechanics. Mutualization can have some extremely problematic features. So why not an approach that unbundles, and allows specialization in these various functions?
Once repositories are created, the development of a robust, coordinated defaulted contract resolution/replacement mechanism would go a long way to improving the efficiency of the OTC derivatives market while permitting it to continue to do what it does best. This would be an easier process (though not a simple one) than what will be set in motion by the vast expansion of CCPs as contemplated in the pending legislation. CCPs will have to develop resolution procedures in any event. Moreover, it is desirable to develop procedures to deal with contracts that are not cleared (which will be the most challenging ones in any event). But force-fed CCPs will also have to grapple with challenging pricing, risk management, risk sharing, and governance issues as well. And if there are multiple CCPs, there will have to be some coordination of the resolution procedures. (If there isn’t there will be trouble.) So, a mandated and extensive expansion of CCPs will have to solve many more problems than just the resolution mechanism. So why not just focus on that?