Template:M detail 2002 ISDA 6(a)
“by not more than 20 days’ notice”
What is the significance of the maximum notice period of 20 days that one may use to close out the ISDA Master Agreement? Poor defaulted Counterparty is in pieces, on its knees, bleeding out, but really, as long as it gets some notice, does it really care how much? Surely, the longer the period, the more hope you have? While the agreement remains in termination but un-terminated, en route to that crater in the ground but not there yet — the chance remains, however remote, that things will come right; that you, its counterparty, will see sense, or unexpectedly discover the one compassionate bone in your body that, until now, has gone wholly unnoticed and, in a cloying bout of clemency, will change your mind and withdraw your notice of termination? Well, a little hedge fund can dream, can’t it? So why deprive it, and yourself, of that option?
I teach you the Children of the Woods
Now, this is deep ISDA lore. It is of the First Men[1] — yea, even the Children of the Woods. As such — since they didn’t have a written tradition back in 1986 and legends were passed down orally from father to son[2] and much has been lost to vicissitude and contingency — it is not a subject on which there is much commentary: That dreadful FT book about derivatives sagely notes that, usually, much less notice is given than 20 days (I mean, you don’t say) but doesn’t give a reason for this curious outer bound, in the same way it doesn’t give a reason for much else in the ISDA Master Agreement despite costing a monkey and that being its express purpose. Nor, for that matter, does the official ISDA User’s Guide to the 2002 ISDA Master Agreement.
One is just expected to know. Yet, in point of fact, no-one seems to. And no-one wants to risk looking stupid by asking, right? Well, companions, just not knowing is not how we contrarians roll. We like looking stupid. Compared with plain old ignorance, it speaks to having at least put in some effort, even if wasted: noble but futile toil is flattering in some lights. So, in the absence of a credentialised, plausible reason,[3] let us speculate.
“Not more than 20-days notice” does not impose a cliff edge
The first thing to say, is the Early Termination Date is not the date on which you must pay an Early Termination Amount nor, for that matter, even have calculated it: it is the date by reference to which you must calculate an Early Termination Amount. Thanks to the good graces of Sectrion 6(d), provided you so as soon as reasonably practicable, you have time to calculate your values diligently and properly.
And even that “by reference to” is heavily qualified: Close-out Amounts[4] are intended to be determined “as of” the Early Termination Date, being the date designated in your Section 6(a) notice which had to be within 20 days of that notice. Now that makes it seem like you are facing a rather untimely cliff-edge if you can’t practicably close out your whole hedge book in 20 days, but note:
Each Close-out Amount will be determined as of the Early Termination Date or, if that would not be commercially reasonable, as of the date or dates following the Early Termination Date as would be commercially reasonable.[5]
This is very important. This means[6] you don’t have to liquidate a portfolio in its entirety within 20 days, or even take the values as of that Early Termination Date. If you can, you should — but it may well not be commercially reasonable — or even possible — to. The Lehman insolvency took months to unwind. Note also that commercial reasonableness is viewed from the Non-Affected Party’s perspective. It is not a licence to do whatever the hell you want — but the court won’t second guess prudent application of your own models.
Therefore no-one has any reason to wait any days, let alone 20
So this leaves the mystery of why a party designating an Early Termination Date, for any reason, wouldn’t just designate it now — and for those peculiar types who don’t want to do that, why there should be this arbitrary long-stop of 20 days.
Remember the ISDA Master Agreement was invented by banking folk: people who who view the Cosmos chiefly through the prism of indebtedness[7]. A lender whose borrower has defaulted will not dilly dally: she will bang in a default notice and seize whatever assets she can get her hand in poste haste. I lend, you owe. I don’t muck about. Breakage costs on a loan are easy to calculate and they are not especially volatile. There is nothing to be gained by waiting around: interest continues to accrue, at a penalty rate: the longer I take to terminate my exposure, the larger it is likely to be.
But, but, but. ISDAs are different. They are not, principally, a contract of indebtedness, and while a large uncollateralised mark-to-market exposure[8] is economically the same as indebtedness, the contract is bilateral, these days almost always fully collateralised, and who is indebted at any time is dependent on the net exposure: it can and does swing around.
Also, the mark-to-market exposure on swap transaction is a wildly volatile thing: With a loan, less so: you know you have (a) principal, (b) accrued interest and (c) break costs — the last of which might be significant for a long term fixed rate loan,[9] but generally will pale in comparison to the principal sum owed.
So a swap counterparty who terminates might be out of the money, and disinclined to terminate just now, hoping that a more benign market environment might be just around the corner to dig it out of its hole so that when it does pull its trigger, the Close-Out Amount will be favourable. This is still taking quite the market punt on a bust counterparty — by means of a European option[10] — of course, and not the sort of thing a prudent risk manager would do,[11] but I don’t suppose banking folk can be expected to have understood this in 1986.
Actually, even that makes little sense, since such a counterparty wouldn’t be obliged to close out at all, but could just suspend its obligations under Section 2(a)(iii) — something which it can (or could, at any rate, when the notice period was devised, in 1987) do indefinitely. To be sure, a 2(a)(iii) suspension is just that — a suspension; should one come eventually to terminate the Transaction, those as-yet unperformed obligations will come back to haunt you as Unpaid Amounts, but at least here you retain control of the process and timing of close-out: it is an American option, not a European one. If you see the market moving against you, you can cash in your chips. So, ask yourself which is a bigger punt: that, the mark-to-market value you determine in 20 days — in a market that is likely to be a flaming wreck, by the way — better suits your book than the one you can actually trade on today, or on any day between now and that distant Early Termination Date?
So we get back to an alternative, disappointing explanation: It is just flannel.
- ↑ I know, I know — or women, but that spoils the Game of Thrones reference, you know?
- ↑ See footnote 1 and/or get a life.
- ↑ And we have done our due diligence, you know: in coming to this conclusion the JC has consulted Magic circle law firm partners, managing directors, inhouse GCs and even a former general counsel of ISDA, all of whom swore me to secrecy but were as nonplussed as, let’s face it, you are about this baffling clause.
- ↑ In this essay, as elsewhere, I use “Close-out Amount” as a generic term to refer to the amount determined to be payable under a terminated Transaction, whether documented under a 2002 ISDA or a 1992 ISDA. I know there is no such thing as a “Close-out Amount” under a 1992 ISDA. Just go with me on this, ok?
- ↑ This is in the definition of Close-out Amount (2002 ISDA) and Loss (1992 ISDA). Curiously, Market Quotation in the 1992 ISDA does it slightly differently, saying “The party making the determination (or its agent) will request each Reference Market-maker to provide its quotation to the extent reasonably practicable as of the same day and time (without regard to different time zones) on or as soon as reasonably practicable after the relevant Early Termination Date.” We guess that gives a bit of flexibility, but is not quite so clear-cut. We suppose the point is that the Non-Affected Party can presumably hit the prices offered by the Reference Market Makers — making the enormous assumption any will actually provide a price — and so isn’t subject to any market risk; which is good. But on the other hand, block-trading a huge portfolio on an arbitrary day you had to set because of the random requirement for “not more than 20 days” is hardly calculated to help the Defaulting Party. You would like to think common-sense would prevail for those dinosaurs still on the 1992 ISDA, who are using the Market Quotation concept. Then again, the fact that they are still on a 1992 ISDA twenty years after it was superseded suggests somewhat that common sense may be lacking somewhere in the relationship.
- ↑ Arguably unless you’re on a 1992 ISDA and using Market Quotation — see the footnote above.
- ↑ Hence, a Cross Default clause in the ISDA Master Agreement. Well — can you think of another reason for it?
- ↑ Such as the sort you could have if it were 1987 and the credit support annex hadn’t been invented.
- ↑ But are there such things in this day and age? Serious question.
- ↑ A correspondent writes pointing out — quite correctly — that, once an Event of Default has happened, the option to send a 6(a) Notice is American. So it is — thanks to Section 2(a)(iii), a potentially open ended one — until you convert it into something like a European option by sending the notice and specifying a date in the future on which it takes place.
- ↑ The silly FT book is right about this, to be fair.