The Logic behind Clearing

The full text can be found here: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3046878

In this post, I will look at the logic of ‘clearing’. Clearing is a term used to achieve a variety of goals in economic and legal terms, all connected to the idea of ‘collapsing’ exposures that exist between a multitude of parties. The biggest headache is, however, caused by the fact that the economic outcome of a clearing process, that is the collapsing of multilateral exposures within an IT system, may not necessarily be matching the legal consequences. Hence, operations and legal enforceability may diverge – an immensely dangerous situation. I will, hence, attempt to map the different forms of clearing and describe their basic legal logic, to bring some clarity to this terminological thicket.

To take a simple example: A has a claim of 40 against C, C has a claim of 20 against B, and B has a claim of 20 against A.

  • The parties have the common interests to act efficiently and mitigate counterparty risk – to this end, they enter a clearing arrangement, see section a).
  • The operational outcomes of a clearing process need to be enforceable in legal terms, so that the operational situation is congruent with the legal situation at all times, see section b).

Efficiency and the operational side of clearing

While in principle no different from the introductory 3-party example, real-world scenarios of clearing arrangement involve dozens or hundreds of parties and a staggering number of transactions amongst them. Billions of payments, securities transactions, derivative contracts and other types of transactions are settled each day in a variety of clearing arrangements worldwide.

Central clearing is offered to the market as a service by infrastructure providers, typically against payment of a small fee per cleared transaction. The relevant central entities are called clearing system, clearing house, or central counterparty – whereas the first two expressions are general, the latter has a relatively clearly defined meaning, see below. The market employs the term of clearing first and foremost for a process of the central computation of net exposures between a multitude of parties, which involves, in their understanding, first and foremost IT-based data processing. That is the operational side of clearing, and, consequently, a clearinghouse or CCP is, in essence, an IT service provider. Clearing offers three main advantages in terms of efficiency, notably net settlement, settlement risk mitigation and counterparty risk mitigation (clearing houses and CCPs typically also offer add-on services).

Net settlement.  The separate and independent settlement of each transaction, from the transferor to the transferee (‘gross settlement’), can be very inefficient. Often, there are several intermediaries involved in a transaction between market participants. Hence, what appears to be one transaction in economic terms may consist of several, stacked transactions in operational (and legal) terms between the parties and their intermediaries and between intermediaries with other intermediaries. This is where intermediaries may use a clearing entity to reduce the number of individual transactions between them. To this end, relevant transactions of a clearing arrangement’s participants are not processed immediately at the moment when they occur but accumulated by over a period of time, say for one day. At the end of this ‘settlement cycle’, only the arithmetical net positions between participants are settled.

Referring back to the introductory example: instead of settling 3 individual transactions between A, B and C, one could consider the net exposures: A has a net claim of 20, C has a net obligation of 20, and B’s exposure is nought. As a consequence, only one payment from C to A in the amount of 20 is processed.

Settlement risk.  In a clearing arrangement, the relevant exposures of all participants are reduced and later settled at the same time. Hence, by definition, settlement risk cannot materialise.

Counterparty risk.  Using a clearing arrangement keeps net exposures small. Similar to bi-lateral set-off, clearing offers the advantage of combining rights and obligations at the moment at which they arise – the balance, or net exposure, is smaller. Hence, participants in a clearing arrangement can operate their accounts and risk management procedures on the basis of net exposures existing against the clearing arrangement in its entirety. Quite comparable to the situation of bilateral netting, a participant is only exposed to the net amount should another participant default.

Enforceability of the operational outcome

However, it makes all the difference whether net exposures calculated in operational terms under the arrangement reflect what participants would actually owe in the event of the insolvency of one of them. Or, in other words, what may appear a combined small net amount in operational terms requires legal enforceability. Otherwise, in the scenario of the insolvency of a clearing participants the remaining, solvent participants face uncertainty whether the exposure they assumed will be the actual exposure.

As a result of that concern, the data processing within a clearing arrangement would be useless and in fact dangerous (because it creates a futile appearance of net exposures) if not underpinned by a legal framework which comes to the same result. The law needs to attribute cleared rights and obligations in a way such that the relevant rights are legally attributed in accordance with the practical outcome of the IT process. It is not sufficient that the IT attributes an economic position to a person; that process needs to be given legal authority.

There are two key decisions in English law, British Eagle and Ansett. They both refer to the clearing arrangement for claims between airlines run by IATA. In the ubiquitous scenario where airlines offer flights on the basis of code sharing and similar subcontracting, typically one airline will sell the ticket whereas another airline will actually provide whole or part of the transport and associated services. Naturally, the former has to share the ticket income with the latter. The macro picture of this phenomenon looks like a highly complicated web of payment obligations between all airlines world-wide. It would make little sense to settle claims resulting from these arrangements on a passenger-by-passenger basis, or even on an airline-by-airline basis. Therefore, nearly all airlines are participants in the IATA clearing system. The latter clears all payment obligations stemming from ticket code sharing and similar arrangements. Payments claims are accumulated for one month. At the end of this period, those member airlines that are net debtors pay their net obligation to the clearinghouse. Those member airlines that are net creditors receive their net compensation from it. The sum of all payments to and from the clearing system is logically nought.

The danger of this arrangements is—obviously—that one of the participating airlines becomes insolvent at some point in time and that its insolvency administrator will not accept the net obligations between the insolvent and the other members but try to claim all outstanding gross obligations from the insolvent’s debtors while refusing to pay to the insolvent’s gross creditors—even if the obligations have already been entered into the clearing mechanism during the accumulation period. That is exactly what happened in the context of two airlines becoming insolvent, first British Eagle, and a couple of years later, Ansett. The cases were decided differently, at the surface of it because the terms of the IATA clearing arrangement had been changed in the meantime. However, the decisions don’t give much insight into the ‘why’. The legal issues are best considered from two different angles, not only in relation to these two decisions (again, they do not make that distinction clearly).

Without or with CCP

The first question is that of the means by which multilateral exposures are reorganised to become bilateral exposures (only bilateral exposures can be settled through a payment or delivery).

  • ‘True’ multilateral clearing (no CCP). The first method could be called ‘true’ multilateral clearing. It is based on reorganising claims immediately amongst participants. This is the scenario that the Court seemed to see in the IATA setting in British Eagle. In the above example, B’s claim against A in the amount of 20 could be attributed to C, and B would obtain consideration in the form of a payment claim in the amount of 20 against C. A’s and C’s payment claims result in a combined net payment claim A against C in the amount of 20. How the single exposures are reorganised in legal terms between participants depends on the available legal methods:
    • The reattribution of exposures could be based on assignments between participants. In this case, bilateral mutuality of exposures would be achieved, allowing for a later set off. This is only suitable in relation to simple claims, as more complex relationships, such as derivative contracts, typically cannot be assigned, as they contain rights and
    • Alternatively, the reattribution could be organised on the basis of novation. In this case, parties would agree that their original exposures cease to exist at some point and that relevant net exposures arise between them immediately afterwards. This method is suitable for simple claims but could also apply to more complex exposures, such as an entire derivative contract where both parties have rights and obligations (executory contract).
  • CCP clearing. In this scenario, all exposures between participants are reorganised so as to exist only between the clearing entity (which is called ‘central counterparty – CCP’ in this case) and the participants.
    • The CCP interposes itself in each transaction. A frequently used standard phrase (which is, however, imprecise) describes this process as the CCP becoming ‘seller to every buyer and buyer to every seller’. All exposures are re-attributed to the CCP. There are no direct exposures anymore amongst the CCP’s participants. In practical terms, the CCP and all of its participants agree in their framework contract that all transactions between participants are novated shortly after they have been concluded. Each transaction is reincarnated as two transactions involving the CCP. The content of the exposures does not change.
    • The existence of a central entity, regardless of what it is called, does not predetermine whether the clearing arrangement belongs to the first or second category. There might well be a central entity that only acts as payment agent, book-keeper and IT service provider for the participants, without itself taking a legal position in the cleared positions. If in doubt, only the interpretation of the contractual basis of the clearing arrangement can provide certainty as to whether a clearing arrangement belongs to the CCP or no-CCP type.
  • Hybrid arrangement. Also, in this case, the clearinghouse interposes itself by way of novation. However, this does not affect the entire legal relationships but only part of them. In particular, payment obligations could be assigned to or novated in the person of the clearinghouse, whereas other parts of the participants’ exposures (in particular, obligations other than payment, such as the performance of a service, or delivery) remain legally between the parties. At the end of the clearing period, net amounts are exchanged in relation to all payment obligations. This concept appeared to be underlying the Ansett However, again, the court was not very clear in its legal analysis.

Timing of enforceability

The second fundamental analysis concerns the point in time at which the clearing outcomes are not susceptible to being avoided in the event of the insolvency of a participant. The process of reorganising multilateral exposures starts with the entering of relevant positions into the common database, often called the ‘system’ (which in the EU means a designated system compliant with the relevant directive). This is the earliest point in time, in principle, at which the net exposures can possibly become enforceable.

This would be the preferred solution from the perspective of clearing participants – they could rely on the enforceability and organise their risk management accordingly right from the beginning. However, the clearing process ends only with the payment of the relevant net amounts from net debtors to the clearing entity, and from the clearing entity to net creditors. A court might consider this point in time the decisive moment at which the arrangement becomes enforceable and from which participants can rely on being exposed to net claims only (see the British Eagle case).

However, considering the effect of suspect periods and similar insolvency principles the amplitude of the problem becomes clear: a clearing arrangement is in danger of being avoided for quite some time. This would mean that cleared transactions need reversing – however, this is impractical (if not impossible) from the operational point of view, let alone the disastrous effects on risk management.

It is therefore crucial that clearing participants can rely on the enforceability of their clearing arrangement and know exactly the point in time at which enforceability arises.

Finality

That point cannot be defined easily in advance (‘ex ante’) – see the Ansett and British Eagle cases, above. The reason is that insolvency avoidance rules are complex and their application to some extent difficult to predict. Hence, even if parties invest considerable effort in drafting enforceable agreements, the ultimate proof of enforceability occurs in the form of a reality test – the decision of the insolvency judge. This situation would keep the risk management of clearing participants at the level of guesswork.

Given that clearing (and settlement) systems in the financial markets are amongst the fundamental infrastructures on which our economy is built, legislators in many jurisdictions, including the EU and all developed financial markets, found it appropriate to define directly in the law the point in time at which enforceability arises.

The basic mechanism works as follows: an early, clear moment at which acquisitions are enforceable is defined by the law provided that the clearing arrangement complies with a number of legal requirements, notably has binding internal rules that correspond with basic legal principles, such as priority of acquisition on the basis of chronology, non-discrimination, etc. In the EU, such system needs to be notified to the Commission. If a system complies with these requirements, the moment in time at which enforceability occurs is (a) the point at which positions are fed into the clearing system by its participants; or, (b) the point in time at which the clearing processes has commenced.

Courts are bound by these rules, and insolvency law cannot override them (they are, in fact, part of the insolvency law themselves). In the EU, these rules are laid down in the Settlement Finality Directive, which covers both types of systems, clearing (ie the calculation of participants’ positions) and settlement (the delivery mechanism employed to fulfil obligations).