The ISDA Master Agreement 1992 (ISDA 92) is one of the most widely used standard-form commercial agreements in the world. Judicial decisions as to its proper construction have potentially very far-reaching consequences. Recent cases in the Commercial Court have engaged in the contentious debates regarding two important issues:
- the operation of the netting provisions of ISDA 92 where a party is in Default (the Netting Issue); and
- the consequences of an Event of Default on future obligations to pay, and the calculation of Loss following Early Termination (the Once-and-For-All Issue).
The cases present contrasting views that cry out for clarification by the Court of Appeal.
Starting Point: Marine Trade
The decision of Flaux J in Marine Trade SA v Pioneer Freight Futures Co Ltd BVI  EWHC 2656 (Marine Trade) staked out the battleground of the debates. Very briefly, the claimant and the first defendant were parties to 14 Forward Freight Agreements (FFAs), which constituted Transactions governed by an ISDA 92 between the parties (as supplemented and amended by Forward Freight Agreement Brokers Association (FFABA) 2007 terms).
In January 2009, Pioneer was affected by an Event of Default. Neither party terminated the agreements. The Settlement Sums for that month were, approximately, US$7 million in favour of Marine Trade, and US$12.1 million in favour of Pioneer. The question arose whether those sums should be netted under s 2(c) of the ISDA 92.
Flaux J held that the netting provision did not apply. The Judge’s reasoning focused on the wording of s 2(c), which provided that “amounts [which] would otherwise be payable” could be netted. Pioneer was in Default; therefore it did not satisfy the condition precedent in s 2(a)(iii); therefore the Settlement Sums otherwise due to it were not ‘payable’, in the sense that they were not due and owing for immediate payment; and therefore those sums were not eligible for netting under s 2(c). Flaux J could “quite see the commercial sense of being able to insist on ‘gross’ payment by a Defaulting Party”.
The conclusion that a Settlement Sum was not ‘payable’, where a party was in default on the relevant date, had further consequences. Where a party does not satisfy the condition precedent in s 2(a)(iii) at the relevant date, the obligation of its counterparty to make payment does not arise at all: it is not merely suspended, and therefore cannot be ‘revived’ should the default be cured.
Both of Flaux J’s conclusions have striking commercial consequences: using the facts of the case as an illustration, Pioneer would be required to pay the $7million sum it owed Marine Trade in full, without netting it off against the much larger sum owed to it in return. And it could never recover that much larger sum, even should it cure its default. (Flaux J did not discuss the scenario of an Early Termination, which has been addressed in subsequent cases.)
The case was not appealed.
Subsequent Debate of the Once-and-For-All Issue
The most recent edition of Firth’s Derivatives: Law and Practice suggests that Flaux J’s decision on this issue would lead to “extremely uncommercial result[s] … which … cannot have been intended” and “paradoxical result[s]”. For example, a party might be forever deprived of a payment in circumstances where it bears no responsibility for a default – as in the case of the presentation of a winding up petition by a vexatious litigant. Other market commentary echoed those concerns.
In Lomas v JFB Firth Rixson  EWHC 3372 (Ch) (Lomas), Briggs J took the opportunity to consider the Once-and-For-All issue. Briggs J was sympathetic to Flaux J’s reasoning – acknowledging that it was “more consistent with the language of s 2(a)”, simple and certain. Ultimately, however, he reached the opposite conclusion.
Briggs J focused on the commercial effects of the alternative constructions. He considered that to hold that the payment obligation does not arise at all would produce “a pointlessly draconian outcome” where the default was minor and fleeting; and particularly where there is only a Potential Event of Default (which is sufficient to render the condition precedent un-met). He also found it relevant that, on Early Termination, all Unpaid Amounts (including those which would have been payable but for an earlier Default) became owing. If the effect of s 2(a)(iii) truly was to prevent a payment obligation from arising in the event of default, it seems counter-intuitive for the obligation to spring up on Early Termination.
Briggs J reaffirmed his conclusion on the Once-and-For-All issue in Lehman Brothers Special Financing Inc v Carlton Communications Ltd  EWHC 718 (Ch) (Carlton).
The Once-and-For-All Issue was next substantively discussed by Gloster J in Pioneer Freight Futures Company Ltd (in liq) v TMT Asia Ltd  EWHC 778 (Comm) (TMT Asia No 1). Very briefly, Pioneer and TMT Asia had entered into a number of FFAs, governed by ISDA 92 and, variously, the provisions of FFABA 2005 and 2007. Pioneer – as we have seen – found itself unable to meet its obligations in late 2008/early 2009; and then the market turned in its favour. Gloster J was posed several preliminary questions including, relevantly, what sums should be included in the calculation of Loss following Automatic Early Termination.
TMT argued that, because Pioneer was in Default and therefore did not meet the condition precedent in s 2(a)(iii), any obligation on its part to pay Pioneer never arose. It argued that the phrase “any payment … required to be made (assuming satisfaction of each applicable condition precedent)” in the definition of Loss only applied to payments actually required to be made where in fact the conditions precedent were met; it did not deem those conditions to have been satisfied when in reality that were not. Therefore any such sums should not be included in a calculation of Loss.
Gloster J began her analysis by concluding that the language used could not sensibly be interpreted in the way proposed by TMT: the words “assuming satisfaction” clearly required an analysis of an artificial hypothesis. They were not linked to what had actually happened between the parties. If it were, in most cases a Non-defaulting Party would never owe anything to a Defaulting Party following Automatic Early Termination, because it would be able to argue that the Defaulting Party did not meet the condition precedent at the relevant times. This would render a choice of the Second Method and Loss next to worthless.
The commercial purpose of s 2(a)(iii), according to Gloster J, is to “mitigate counterparty credit risk during the currency of what may be numerous swap transactions under the umbrella of ISDA 92 and while they remain open.” It fulfills this purpose by suspending the payment obligation (where the Defaulting Party may not be able to meet its own payments). It does not suspend the debt obligation (or prevent it from arising). Instead, it substitutes an accounting procedure – where respective debts are totted up on paper – while suspending any obligation to make payment.
Upon Early Termination, that ongoing counterparty credit risk disappears: the ongoing relationship between the parties is at an end. The calculation of Loss, therefore, would naturally include all of those debts accrued on paper – to be reconciled in the ‘wash out’. According to Gloster J, “Commercially, this all makes sense.”
Gloster J turned to the decision in Marine Trade: she reasoned that the Judgment failed to recognise that there were two separate obligations, and that the condition precedent in s 2(a)(iii) was only applicable to the payment obligation: it did not prevent the debt obligation from accruing. The “natural reading of [s 2(a)(iii)] envisages that once a condition precedent is fulfilled, the obligation to pay revives”. The debt obligation is not revived: it was there all the time. Gloster J also agreed with Briggs J’s reasoning in Lomas.
Subsequent Debate of the Netting Issue
Marine Trade met with intense criticism on this point. In the second edition of Henderson on Derivatives, its conclusion on the Netting Issue is floridly described as “astonishing”, “arbitrary” and “bizarre”. The core of Henderson’s analysis is that Flaux J did not acknowledge that the obligation in s 2(a)(i) (and the conditions precedent to that obligation, in s 2(a)(iii)) is expressly said to be “subject to the other provisions of this Agreement” – including the netting provisions. Henderson argued that there was no justification for Flaux J giving ‘primacy’ to s 2(a)(iii).
In both Lomas and Carlton, Briggs J did not follow Marine Trade on the Netting Issue, on the invitation of all the parties.
Although the Netting Issue did not arise directly in Pioneer Freight Futures Company Ltd v Cosco Bulk Carrier Company Ltd  EWHC 1692 (Comm) (Cosco), Flaux J seized the opportunity to reinforce the views he set out in Marine Trade. His Honour analysed other usages of the word “payable” in ISDA 92, and again concluded that only sums which were “immediately enforceable obligation[s] to pay” could be netted pursuant to s 2(c).
Following delivery of the judgment in TMT Asia (No 1), TMT sought to amend its defence to raise the Netting Issue (on its case, following Flaux J’s conclusion on the issue would reduce the sum it was required to pay Pioneer by some $10 million). After the new point was fully argued, the parties settled. Nonetheless, Gloster J chose to deliver her judgment on the issue, characterising it as “in effect … an advisory opinion”: see Pioneer Freight Futures Company Ltd (in liq) v TMT Asia Ltd  EWHC 1888 (TMT Asia (No 2)). The Judge’s reason for adopting this unusual course included the uncertainty in the market caused by the Marine Trade decision, and that she had “firmly reached the opposite conclusion” to Flaux J.
Gloster J began by considering the ‘landscape’ of ISDA 92 and the other agreements as a whole:
“One does not have to approach the canvas of the relevant FFABA 2007 terms and ISDA 92 very closely to see that the broad commercial scheme of the instruments … is that aggregate or gross amounts in respect of all Transactions between the parties subject to the Master Agreement are to be netted off against each other … the scheme was intended to be a cohesive one, governing the entire period of the relationship of the parties.”
The Judge referred to the commercial purpose of s 2(a)(iii) – as elucidated in TMT Asia (No 1) – and held that that purpose would be ‘wholly undermine[d]’ if a non-defaulting party could insist on payment on a gross basis. Such an interpretation would be “wholly contrary to the ethos of ISDA 92” and would “emasculate the netting provisions … in the very circumstances where they may be most needed.”
Further, there was no utility in focusing on the word ‘payable’ where that word “can mean many different things in many different contexts”; it certainly did not require the interpretation given to the clause by Flaux J. Gloster J also picked up Henderson’s point that the payment obligation in s 2(a)(i) is expressly subject to the other provisions of ISDA 92 – including the netting provisions. There was no reason, therefore, for the condition precedent in s 2(a)(iii) to limit the scope of which sums could be netted pursuant to s 2(c).
In Marine Trade, Flaux J concluded that:
- Where a Defaulting Party did not meet the condition precedent in s 2(a)(iii), its counterparty never comes under an obligation to make payment; and
- Sums nominally owed to a Defaulting Party are not available to be netted pursuant to s 2(c).
Both of these conclusions proved controversial – academic commentators and, more importantly, subsequent judicial decisions, have fundamentally disagreed with them. Nonetheless, Flaux J reinforced his decision on the Netting Issue in the very recent case of Cosco.
The stage is set for the Court of Appeal to reconcile these conflicting views. Lomas, Carlton and Cosco have been appealed. They have been consolidated – together with a fourth ISDA-related case, Britannia Bulk plc (in liq) v Pioneer Navigation Ltd  EWHC 692 – and the hearing is scheduled for 14 December 2011. The commercial consequences of these cases will be far-reaching, and the market will be watching closely.