Events of Default - ISDA Provision
2002 ISDA Master Agreement
A Jolly Contrarian owner’s manual™
ISDA Text: 5(a)
Related agreements and comparisons
Resources and Navigation
5(a)(i): The significant change between 1992 ISDA and 2002 ISDA is the restriction of that grace period from three Local Business Days to one. And a bit of convolutional frippery in introducing Local Delivery Days as well.
Compare also Failure to Pay under the 2014 ISDA Credit Derivatives Definitions, which is subtly different given the different purpose that it plays under a CDS.
5(a)(ii): Note the addition of Repudiation of Agreement to the 2002 ISDA. Common law purists like the JC will grumble that you don’t really need to set out repudiation as a breach justifying termination of a contract, because that’s what it is by definition but stating the bleeding obvious has never stopped ISDA’s crack drafting squad™ before. Also, an interesting question: if you do feel the need to provide for what is in essence an evolving common law remedy, then, to the extent your draw that remedy inside the cope of the common law remedy — or the common law evolves some new different and remedy that no-one had thought of before — then what? Section 9(d) has you covered. Woo-hoo.
5(a)(iii): A bit of pedantic flannel found its way into the 2002 ISDA — it captures not just the failure of the Credit Support Document itself, but any security interest granted under it, catering to the legal eagle’s most paranoid fears that a contractual right can have some sort of distinct ontological existence independently from the agreement which gives it breath and enforceable currency in the first place. But otherwise the same.
5(a)(iv): No change between 1992 ISDA and 2002 ISDA.
5(a)(v): DUST has been expanded in five significant ways by the 2002 ISDA. See the summary and general sections for details.
5(a)(vi): The 2002 ISDA updates the 1992 ISDA’s Cross Default so that if the combined amount outstanding under the two limbs of Cross Default exceed the Threshold Amount, then it will be an Event of Default. Normally, under the 1992 ISDA, Cross Default requires one or the other limbs to be satisfied — you can’t add them together. This was a bit of a snafu.
The two limbs are:
- a default under a financial agreement that would allow a creditor to accelerate any indebtedness that party owes it;
- a failure to pay on the due date under such agreements after the expiry of a grace period.
First, the 2002 ISDA has a slightly more specific concept of “insolvency”. In limb 4 (insolvency proceedings) a new limb has been included to cover action taken by an entity-specific regulator or supervisor (as opposed to a common or garden insolvency proceeding): If initiated by a regulator, the game’s up as soon as the action is taken. If initiated by a random creditor, the action must have resulted in a winding-up order, or at least not have been discharged in 15 (not 30) days.
About that grace period. Second, and unnervingly for those of little faith in their own accounts payble departments, the grace period in which one must arrange the dismissal of a vexations or undeserving insolvency petition (under 5(a)(vii)(4)) or the exercise of security over assets (under 5(a)(vii)(7)) is compressed from 30 days to 15 days.
5(a)(viii): ISDA’s crack drafting squad™ giveth and ISDA’s crack drafting squad™ taketh away. As it neatly excises one square of flannel here, it inserts another one, further itemising ways in which a company might reorganise itself, there. In practical terms — ones that might make a difference were they to be considered by the King’s Bench Division, that is — no real change between 1992 ISDA and 2002 ISDA.
Failure to Pay or Deliver under Section 5(a)(i) of the ISDA Master Agreement: where a party fails to pay or deliver on time and does not remedy before the grace period expires. The grace period for a 2002 ISDA is one Local Business Day; shorter than the three Local Business Days in the 1992 ISDA. This fact alone has kept a number of market counterparties on the 1992 form, nearly thirty years after it was upgraded.
There’s a technical funny due to the American habit of insisting on a pledge-only 1994 New York law CSA and then designating it as a Credit Support Document (against the hopes and dreams of ISDA’s crack drafting squad™ when it drafted the Users’ Guide, but still), and that is a failure to pay under an English law CSA is a Section 5(a)(i) Failure to Pay or Deliver, whereas a failure to pay under a New York Law CSA is a Section 5(a)(iii) Credit Support Default. Doth any difference it maketh? None, so far as we can see.
Funny old world we live in.
A failure to perform any agreement, if not cured within 30 days, is an Event of Default, except for those failures which already have their own special Event of Default (i.e., Failure to Pay or Deliver, under Section 5(a)(i)), those that relate to a pre-existing default (for example, default interest on unpaid amounts) and those that only bear on the “defaulting” party’s tax position, meaning that non-performance is punishment enough in itself (and does not affect the “non-defaulting” party) — that is, the non-compliant party will just get clipped for tax it could have avoided had it performed.
These are the boring breaches of agreement: those of a not immediately existential consequence to a derivative relationship (like Failure to Pay or Deliver, or a party’s outright Bankruptcy), which are not therefore hugely time critical, but which, if not promptly sorted out, justify shutting things down with extreme prejudice. Note that, while the 2002 truncates a bunch of other grace periods in the agreement (notably for a Failure to Pay or Deliver, and for discharging a Bankruptcy petition) it does not truncate the grace period for “boring” defaults, which stays at the glacially long 30 days.
All rendered in ISDA’s crack drafting squad™’s lovingly tortured prose, of course: note a double negative extragvaganza in 5(a)(ii)(1): not complying with an obligation that is not (inter alia) a payment obligation if not remedied within a month. High five, team ISDA.
New in the 2002 ISDA: repudiation of contract. Not actually breaching it, per se, but high-handedly saying that you are going to. In writing. Could you argue that by codifying that a repudiation must be in writing to count, in a counter-intuitive way this new clause dilutes the common law rules, rather than reinforcing them? A common law repudiation can, if clear enough, be oral, by conduct, body language, morse code, semaphore and so on,
So rather than empowering a Non-defaulting Party, the addition of a narrow definition of what counts as repudiation makes their avenue of redress that teeny bit narrower. Doubtful it has ever made a difference, but — well, they said that about LIBOR didn’t they.
Hierarchy of Events
Note that a normal Section 5(a)(ii)(1) Breach of Agreement that also amounts to a Section 5(b)(i) Illegality or a Section 5(b)(ii) Force Majeure Termination Event will, thanks to section 5(c), be treated as the latter, but a repudiatory Breach of Agreement under section 5(a)(ii)(2) willl not enjoy the same leniency. If you have repudiated your contract, the fact that there happens to be a concurrent Illegality — it is hard to see how a repudiatory breach could be an Illegality in itself — will not save you from the full enormity of section 5(a)(ii) Event of Default style close out.
Before you even put your hand up: no, a Credit Support Annex between the two counterparties is not a “Credit Support Document”, at least under the English law construct: there it is a “Transaction” under the ISDA Master Agreement that offsets the net mark-to-market value of all the other Transactions, so can’t be a Credit Support Document per se. It is different with a 1994 NY CSA — being a pledge document it is a Credit Support Document, but even there the Users’ Guide cautions against treating a direct swap counterparty as a “Credit Support Provider” — the Credit Support Provider is meant to be a third party: hence references to the party itself defaulting directly under a Credit Support Document.
Therefore, tediously — and we think it was avoiding precisely this tediosity that the Users’ Guide had in mind, but, best-laid plans and all that — there is an ontological difference between the mechanics of close out when it comes to a failure under a New York-law CSA when compared to non-payment under an English-law CSA.
A failure to meet a margin call under an English-law CSA or any of its modern English-law successors is therefore a Failure to Pay or Deliver under Section 5(a)(i) of the actual ISDA Master Agreement; a failure to post under a New York-law CSA is a Section 5(a)(iii) Credit Support Default.
Now, before you ask why anyone would ever do that, firstly let us say that far stupider things than that happen every freaking day in the negotiation of global markets documentation, and secondly that, for example, grace periods for regulatory initial margin may well be standardised — and dealers may not have the capacity or appetite to negotiate them tightly, given the paper war they will be in in any case — so you can quite easily see 2002 ISDAs with very brief grace periods, and Initial Margin CSDs with longer ones. So won’t this be fun when it comes to closeout.
The purist’s objection is that, since a representation is a pre-contractual statement which induced the wronged party to enter the contract and (ergo) was not, and could not be, itself, a contractual term at all — its bolt was shot, so to speak, before “minds met” — and, as such, one’s remedy for misrepresentation ought to be to set aside the contract altogether (ab initio, as Latin lovers — well, my one, at any rate — would say) voiding it on grounds of no consensus, and not suing for damages for breach of something which, by your own argument, never made it into the cold hard light of legal reality. The JC is nothing if not a purist. We feel that, as written, this provision is a bit misconceived.
Giving our friends at ISDA the benefit of the doubt we think ISDA’s crack drafting squad™ means “breach of warranty”, and were really just being loose with terminology. There again, unlike other, more fundamental obligations, misrepresentation as an Event of Default has neither a materiality threshold nor the accommodation to the wrongdoer of a grace period or even a warning notice, so perhaps not. Anyway.
The connoisseur’s negotiation oubliette.
Default Under Specified Transaction — colloquially, “DUST” — is often confused with Cross Default. In fact, they’re meant to be mutually exclusive. That won’t stop folks conflating them, though. Look, we all do it.
DUST is like Cross Default, but where Cross Default references indebtedness owed to third parties, DUST is all about non-“borrowing” style transactions — e.g., swap agreements, stock loans and repos, but only transactions between the two counterparties.
Changes from the 1992 Master Agreement
DUST overwent quite a makeover in the 2002 ISDA. For example:
Mini-closeout carveout: Defaults require the acceleration of just the Specified Transaction in question (for general defaults) but off all outstanding transactions under the relevant master agreement (for delivery defaults). This change was made with mini-close-out under repos and stock loans in mind — a concept which the stock loan market invented after the 1992 ISDA was published, so you can’t blame ISDA’s crack drafting squad™ for overlooking it at first — where delivery failures under are common and do not of themselves indicate weakness in the Defaulting Party’s creditworthiness.
Credit support failures covered: DUST under the 2002 ISDA can be triggered by default under a credit support arrangement relating to a Specified Transaction. These weren’t included for the 1992 ISDA DUST.
Shortened cure period: In tune with the general tightening up of cure periods — you know, we’re in a new millennium, computers work properly nowadays, and all that — the cure period for a failure to make a final or early termination payment on a Specified Transaction has been reduced from three days to one. This caused many a credit officer to sadly shake her head and refuse to move to the new agreement.
Repudiation evidence: Repudiation was modified to add the phrase “...or challenges the validity of ... after “... disaffirms, disclaims, repudiates or rejects ...” to reduce ambiguity as to whether a party’s action constitutes a repudiation. Also, we imagine, by way of stiffening the criteria for what counts as a repudiation, the 2002 requires written evidence that the repudiating party has an extended middle finger. This rules out being able to close out cornered hedge-fund managers, having been “brought to the negotiating table” by their fund’svproximity to a NAV trigger and who are not enjoying having their “feet held to the fire”, shouting “Well, bugger you, I shan’t pay, and let’s see how you like that” in the heat of the moment, when they really didn’t mean it, only to discover they had inadvertently repudiated a contract they were otherwise in perfect compliance with. Of course, no risk officer would dream of closing out an ISDA Master Agreement based on an intemperate oral communication, or the proverbial extended middle finger, for which she could not subsequently prove with fairly compelling evidence. But still.
Widened definition of Specified Transaction: The “Specified Transaction” concept has been broadened to include additional transaction types such as repos, and to include a catchall clause designed to include any future derivative products that have not been thought of yet.
Voltaire and DUST
In which ISDA’s crack drafting squad™ got bogged down in the weeds once in 1987, doubled down in their in-weed bogged-downness in 2002, and we’ve been dealing with resulting confusion ever since. A case of perfection being the enemy of good enough, as Voltaire would say, in the JC’s humble opinion, especially in these modern times where, thanks to compulsory daily zero-threshold variation margining, DUST is even more of a dead letter than it even was in the good old days. To our knowledge, no ISDA Master Agreement in history has been closed out using, exclusively, Section 5(a)(v).
That said, the 1992 ISDA version is a bit skew-wiff as regards mini-closeout, and you may find assiduous counterparties hungrily licking their lips at the prospect of a hearty negotiation about this bald man’s comb.
We are talking about other derivative-like transactions, between you and the same counterparty, where the counterparty presents a clear and present danger of blowing up, but where that behaviour has not yet manifested under the present ISDA Master Agreement, meaning you have no grounds to blow them up directly. So, you know, fairly implausible scenario, but still. You want to use the event arising under this other Specified Transaction to detonate the present ISDA. The squad breaks your ability to do so down in to four scenarios:
- Counterparty fails to pay amounts falling due before maturity on a Specified Transaction, and you accelerate that transaction, but not necessarily others under the same master agreement. Here the principle is that any obligation to pay a sum of money on time is fundamental, of the essence and speaks indelibly to a merchant’s credit, whether or not one accelerates other related Specified Transactions (though, actually, walk me through the scenarios in which you wouldn’t, or even weren’t obliged to?)
- Counterparty fails to pay amounts falling due at maturity on a Specified Transaction, so you can’t “accelerate” as such on that Specified Transaction, as it has matured, but you are still out of pocket and of a mind to press a big red button — though, again, curiously, only on this Specified Transaction and not the other outstanding transactions under the same master agreement, even though you could;
- Counterparty fails to deliver assets due under a Specified Transaction, and as a result you accelerate the Specified Transaction (1992 ISDA) or all Specified Transactions under the affected master agreement (2002 ISDA — the 2002 version being designed to carve out things like mini close-out under a 2010 GMSLA as these are not credit-related;
- Counterparty presents you an extended middle finger generally with regard to any obligation under any Specified Transaction, whether you accelerate it or not. Here if your counterparty is playing craziest dude in the room, it has committed a repudiatory breach thereby losing what moral high-ground it might otherwise stand on to expect you to follow form and protocol before closing it out.
Cross Default is intended to cover the unique risks associated with lending money to counterparties who have also borrowed heavily from other people.
Now, if — as starry-eyed young credit officers in the thrall of the moment are apt to — you apply this thinking to contractual relationships which aren’t “term loany” in nature — that is, that don’t have long spells where one party is deeply in the hole to the other, with not so much as interest payment due for months whose failure could trigger any acceleration — it will give you trouble. We go into this more in the premium JC.
Specified Indebtedness is generally any money borrowed from any third party (e.g. bank debt; deposits, loan facilities etc.). Some parties will try to widen this: do your best to resist the temptation. Again, more details on why in the premium section.
The Threshold Amount is a key feature of the Cross Default Event of Default in the ISDA Master Agreement. It is the level over which accumulated indebtedness defaults comprise an Event of Default. It is usually defined as a cash amount or a percentage of shareholder funds, or both, in which case — schoolboy error hazard alert — be careful to say whether it is the greater or lesser of the two.
Because of the snowball effect that a cross default clause can have on a party’s insolvency it should be big: like, life-threateningly big — because the consequences of triggering a Cross Default are dire, and it may create its own chain reaction beyond the ISDA itself. So expect to see, against a swap dealer, 2-3% of shareholder funds, or sums in the order of hundreds of millions of dollars. For end users the number may well be a lot lower (especially for thinly capitalised investment vehicles like funds — like, ten million dollars or so — and, of course, will key off NAV, not shareholder funds.
For those noble, fearless and brave folk who think Cross Default is a bit gauche; a bit passé in these enlightened times of zero-threshold VM CSAs but can’t quite persuade their credit department to abandon Cross Default altogether — a day I swear is coming, even if it is not yet here — one can quickly convert a dangerous Cross Default clause into a less nocuous (but still fairly nocuous, if you ask me — nocuous, and yet strangely pointless) cross acceleration clause — meaning your close-out right that is only available where the lender in question has actually accelerated its Specified Indebtedness, not just become able to accelerate it, with some fairly simple edits, which are discussed in tedious detail here.
The truncating the grace period from 30 days in the 1992 ISDA to 15 days in the 2002 ISDA has, in aggregate over the whole global market, kept many a negotiator in “meaningful” employment. It has also been a large reason why many organisations did not move to the 2002 ISDA and of those who eventually did — organisations whom you’d think would know better — then set about amending these grace periods back to the 1992 ISDA standard of 30 days or better still, insisted on sticking with the 1992 ISDA, but upgrading every part of it to the 2002 ISDA except for the Bankruptcy and Failure to Pay grace periods. A spectacular use of ostensibly limited resources, and an insight into whose benefit organisations really operate for.
Regional bankruptcy variations
The Germanic lands have peculiar ideas when it comes to bankruptcy — particularly as regards banks, so expect to see odd augmentations and tweaks to ISDA’s crack drafting squad™ standard language. Will these make any practical difference? Almost certainly not: it is hard to see any competent authority in Germany, Switzerland or Austria — storeyed nations all, in the long history of banking, after all — not understanding how to resolve a bank without blowing up its netting portfolio. Especially since Basel, where the netting regulations were formulated, is actually in Switzerland.
We have a whole page about Swiss Bankruptcy Language. True story.
The market standard “bankruptcy” definition
The ISDA bankruptcy definition is rarely a source of great controversy (except for the grace period, which gets negotiated only through custom amongst ISDA negotiators because, in its wisdom, ISDA’s crack drafting squad™ thought fit to halve it from 30 days to 15 in the 2002 ISDA.
So you have a sort of pas-de-deux between negotiators where they argue about it for a while before getting tired, being shouted at by their business people, and moving on to something more important to argue about, like Cross Default).
Otherwise, the ISDA bankruptcy clause is a much loved and well-used market standard and you often see it being co-opted into other trading agreements precisely because everyone knows it and no one really argues about it.
1987 ISDA and Automatic Early Termination
When a firm merges into, or is taken over by, another, some magical — or unexpected — things can happen. Not for nothing does the ISDA Master Agreement labour over the very description: that this might be a “consolidation, amalgamation, merger, transfer, reorganisation, reincorporation or reconstitution” — prolix even by the lofty standards of ISDA’s crack drafting squad™ — should tell you something. Generations of corporate lawyers have forged whole careers — some never leaving the confines of their law practices for forty or more years — out of the manifold ways one can put companies together and take them apart again.
Your correspondent is not one of those people and has little more to say about mergers, except that what happens to live contracts at the time of such chicanery will depend a lot on just how the companies and their assets are being joined together or torn assunder.
If the ISDA Master Agreement and its extant Transactions carry across — which, in a plain merger, they ought to — all well and good - though watch out for traps: what if both merging companies have ISDAs with the same counterparty, but on markedly different terms? Which prevails? Do they both? Which one do you use for new Transactions? This you will have to hammer out across the negotiating table.
But in some cases, Transactions might not carry across. Perhaps the resulting entity has no power to transact swaps. Perhaps it is in a jurisdiction in which they — or ISDA’s sainted close-out netting provisions, about which so many tears and so much blood is annually spilled — cannot be enforced. Perhaps the new entity just refuses to honour them.
Merger Without Assumption addresses all of these contingencies.
This is the clause that would have been covered by Section 5(a)(ii)(2) repudiation, had the resulting entity accepted the contract at all in the first place. It can be triggered if the resulting party repudiates any outstanding Transactions under the ISDA Master Agreement (or otherwise they are not binding on it); or any Credit Support Document stops working as a result of the merger.
- The JC’s famous Nutshell™ summary of this clause
- Events of Default vs. Termination Events showdown: why is there a difference between Events of Default and Termination Events, what is the difference, how will it affect me in practice, tell me more about this odd netherworld of Additional Termination Events, and is there a good way of describing both, or must I really stick with “an Event of Default or a Termination Event, as the case may be”?
- I know these sound like borrowing transactions, but they’re fully collateralised, and in fact aren’t.
- And — sigh — their Credit Support Providers and Specified Entities.
- Or — sigh — its Credit Support Provider or Specified Entity
- This is typically wide, though it excludes borrowed money — but check the Agreement!
- Your correspondent is one of them; the author of that terrible FT book about derivatives is not.
- This, by the way, is an ISDA In-joke. In fact, Cross Default is pretty much pointless, a fact that every ISDA ninja and credit officer knows, but none will admit on the record. It is the love that dare not speak its name.