The 2002 ISDA Master Agreement Isn’t a Contract-Drafting Masterpiece

Recently someone reminded me of the ISDA master agreement, the widely used form of master contract for over-the-counter derivatives transactions. It was last updated in 2002, and after I found a copy of the 2002 version online here, I had a look at it.

My conclusion? The drafting is pretty ragged. No surprise there—I’ve looked at a few examples of model contract language promulgated by trade groups, and they’ve all been varying degrees of not-so-great, no matter how widely used. (For example, go here for my take on the American Arbitration Association’s standard arbitration clause.)

So below are a few comments based on my cursory review of the first four pages of the master agreement, out of 28 pages (not including attachments). These comments just scratch the surface. (For my analysis of another usage employed in the ISDA master agreement, see this post.)

Categories of Contract Language

The word will is sometimes used where I’d use the simple present: “The terms defined in Section 14 and elsewhere in this Master Agreement will have the meanings therein specified for the purpose of this Master Agreement.”

The first subsection of section 2 (Obligations) is given the subheading “General Conditions.” But the first two provisions under it are obligations, so that isn’t a helpful subheading.

Amounts are referred to as being “payable.” When considered in isolation, it’s not clear whether the party in question has discretion to pay those amounts or is obligated to pay them.

See section 2(c) for an example of awkward use of the passive voice: instead of “The election may be made … ,” I’d say “The parties may so elect …”

Defined Terms

For purposes of integrated definitions, in the 2002 version the defined-term parenthetical is routinely not placed at the end of the definition. For example, see the three defined-term parentheticals in the first sentence of the introductory clause:

have entered and/or anticipate entering into one or more transactions (each a “Transaction”) that are or will be governed by this 2002 Master Agreement, which includes the schedule (the “Schedule”), and the documents and other confirming evidence (each a “Confirmation”) exchanged between the parties or otherwise effective for the purpose of confirming or evidencing those Transactions.

Here’s the next sentence in the introductory clause: “This 2002 Master Agreement and the Schedule are together referred to as this ‘Master Agreement’.” It’s redundant, as is the defined term “Master Agreement,” in that the previous sentence specifies that the master agreement includes the schedule.

The reference in the introductory clause to “those Transactions” is illogical—there aren’t any other Transactions. I’d delete the word “those.”

In section 2(c), the defined term “Multiple Transaction Payment Netting” is created by means of neither an integrated definition nor an autonomous definition. That’s unorthodox.


Section 3 says that specified representations “will be deemed to be repeated by each party on each date on which a Transaction is entered into and, in the case of the representations in Section 3(f), at all times until the termination of this Agreement.” I like this use of “will be deemed,” which matches exactly use of that phrase in the representations lead-in that I recommend in The Structure of M&A Contracts. But saying that a representation will be repeated “at all times” is very odd.


The layout is based on a bastard hybrid of hanging indents and first-line indents.

Section 2(c) features dangling text and an unenumerated paragraph, both MSCD no-nos.

In section 2(d), a separate sentence, with its own set of tabulated enumerated clauses, it tacked on to the last of a set of tabulated enumerated clauses. Instead, the period at the end of the last of a set of tabulated enumerated clauses should be followed only by a  new section or subsection.

Drafting As Writing

The “and/or” in the introductory clause (“entered and/or anticipate entering into”) isn’t great.

Don’t beat about the bush. Here’s the current section 1(c):

All Transactions are entered into in reliance on the fact that this Master Agreement and all Confirmations form a single agreement between the parties (collectively referred to as this “Agreement”), and the parties would not otherwise enter into any Transactions.

Here’s my version (limiting my changes):

This Master Agreement and all Confirmations form a single agreement between the parties (collectively referred to as this “Agreement”).

Such is used instead of this, that, these, and those. That’s a hallmark of needlessly legalistic prose.

Abstract nouns bad, verbs good. For example, see section 2(b): “by giving notice to [read notifying] the other party …”

Section 2(d)(ii) contains the following: “… if Y has failed to comply with or perform any agreement contained in Section 4(a)(i), 4(a)(iii) or 4(d)).” I’d say “… if Y has failed to comply with section 4(a)(i), 4(a)(iii) or 4(d)).” Why both “comply with” and “perform”? And why refer to as an “agreement” an obligation contained within a given provision?

Section 3(a) uses the subheading “Basic Representations.” It’s unhelpful to offer opinions regarding the complexity of, or importance of, a set of provisions.

Section 3(e) and 3(f) refer to representations as being “accurate and true.” I’m sympathetic to use of “complete and accurate,” but not “accurate and true.” I’d use just “accurate.”

About the author

Ken Adams is the leading authority on how to say clearly whatever you want to say in a contract. He’s author of A Manual of Style for Contract Drafting, and he offers online and in-person training around the world. He’s also chief content officer of LegalSifter, Inc., a company that combines artificial intelligence and expertise to assist with review of contracts.

13 thoughts on “The 2002 ISDA Master Agreement Isn’t a Contract-Drafting Masterpiece”

  1. Mostly good comments, Ken. I was drawn to your edit in the last block-quoted excerpt — your version starts out, “This Master Agreement and all Confirmations form a single agreement between the parties ….” QUESTION: Would you have completely omitted the reliance language from the original? That would be a pretty significant change to the agreed substantive provision; most style editors would regard that as above their pay grade.

    • You get two lashes with a wet noodle for uncertain etiquette! First, are you the arbiter of what is good and not good? And second, by calling me a “style editor” I think you’re significantly understating the scope of the guidelines in MSCD.

      Regarding the reliance language, when I sign a contract, I’m relying on a whole bunch of things in the contract. No purpose is served by singling one or more of them out.

      • Ken, as to whether you’re a “style editor,” all three editions of your book HAVE been titled “A Manual of Style,” after all. Plus, you seemed to have made the (wise) decision not to claim substantive expertise in all of the vast array of subject areas that contracts address; it’s doubtful anyone who made such a claim would be believed.

        I apologize if I committed a breach of etiquette, but I don’t think I did. The arbiter of what’s good and not good in a given situation has to be the individual lawyer exercising professional judgment in representing his or her client. You seem to have eliminated the reliance language from the IDSA form language without explanation. That’s a significant substantive change. From my days as a litigator I can absolutely understand why a lawyer might want the reliance language in there for use as a sound bite before a judge. Evidently the ISDA drafting committee (or at least a majority thereof) fell into that category. So you might consider whether you’re setting your own professional judgment above that of the IDSA drafting committee — again, without explanation.

        • D.C.: I did my best to make my points without sounding like a turkey. I hope I was at least marginally successful!

          In used the phrase “manual of style” in the title of MSCD to suggest that as with other kinds of writing, contract drafting would benefit from a set of guidelines. Of course, the focus isn’t what to say in a contract, but how to say whatever you want to say. I think that my omitting the reliance language falls within that scope.


          • We’ve about exhausted this discussion topic, Ken, but see my comment to ‘AWrightBurke’ about statements of reliance. Personally, I would start with the (rebuttable) presumption that the lawyers on the ISDA drafting committee knew what they were doing and had a reason for including the reliance language. Deleting that language might be entirely appropriate, but certainly it would make sense to know the drafters’ reason(s) for including it before doing so.

            Anyway, enough of that; I hope you and your family have a happy Easter!

  2. Screwing up my courage after the noodle-lashing you gave D.C. Toedt, I’d like to comment on “will be deemed,” as in your comment as follows:

    Section 3 says that specified representations “will be deemed to be repeated by each party on each date on which a Transaction is entered into and, in the case of the representations in Section 3(f), at all times until the termination of this Agreement.” I like this use of “will be deemed,” which matches exactly use of that phrase in the representations lead-in that I recommend in The Structure of M&A Contracts.

    I haven’t read your treatment of “will be deemed” in The Structure of M&A Contracts, but I criticize the usage on a couple of grounds.

    First, it’s passive without a by-agent. Who is to do the deeming? If the usage is meant to guide the behavior of a reviewing judge, it’s diplomatic to the point of non-binding. If it’s meant to bind a party, more aggressive language is appropriate (see below).

    Second, I thought MSCD style calls for “deem” to create a legal fiction. Why should the application of representations (or should we now be saying “statements”?) to each Transaction involve setting up a legal fiction? Just make the statements directly applicable to each Transaction.

    It’s not possible to redraft the provision fully without knowing a little more context, such as the consequences of falsity of the representations at the specified times, but here’s a “deem-free” attempt that assumes that (1) the specified statements are the same ones that speak as of the date of the contract, and (2) the remedies for falsity of the statements at later times are the same as those for falsity at inception (a cross-reference might be useful):

    “If [specified statements] are false when a Transaction takes place, the consequences will be the same as if the statements were false on the date of this agreement. If the statement in section 3(f) is or becomes false before the termination of this agreement, the consequence will be the same as if the statement was false on the date of this agreement.”
    Given your sharp treatment of poor D.C. Toedt, I hope you won’t be opining on whether the above comments are good or not good. (Kidding!)
    Also, I think you must accept the description of “style editor” with an asterisk to a note that good contract style is not the same as individual style. To paraphrase John Lennon on working class heroes, “A good style editor is something to be.”
    Finally, I think a generous reading of D.C. Toedt’s remark about the omission of the reliance language being a matter of substance and not style would call forth the answer that it is indeed a matter of style as MSCD uses the word, because the reliance language is, if I may put it this way, “redundant of what’s implicit,” in that parties’ reliance on the terms of the contract, like mutual consideration, goes without saying, so “style” counsels not saying it at all, and certainly not saying it here and there, like those contracts that scatter language of agreement randomly.

    • A. Wright: If you don’t mind, I won’t take your “deem” bait. This issue is discussed at some length in “The Structure of M&A Contracts,” and I’m too lazy to hash it out again here. Ken

    • A. Wright, in U.S. law the presence or omission of an express statement of reliance (or, more commonly, non-reliance) can sometimes have substantive significance. For example, an express statement of non-reliance on extrinsic representations can defeat a claim of fraudulent inducement, whereas an expansive entire-agreement clause — which to my mind _implies_ non-reliance — typically will not have that effect. Or at least so say many U.S. courts.

      Moreover, as I replied to Ken, litigators like it when the contracts they are litigating include favorable sound bites that they can quote in a brief, or in open court to the judge or jury. Some contract drafters sniff at such sound bites as being inconsistent with stylistic purity codes. That seems a bit narrow-minded, given that the _ultimate_ goal of any contract is to put the client in the best position possible in the event of litigation.

      • Include me among those drafters who aren’t inclined to defer to rhetorical emphasis. Furthermore, I suspect that most who indulge in it aren’t aware that it constitutes rhetorical emphasis.

      • I think the reliance language must be about rescission. If the Transactions were considered separate contracts, it might affect the netting and collateralisation provisions and therefore completely throw off all the risk calculations made by the banks in entering into the Transactions. They could find themselves hugely exposed to a counterparty when they never expected to be, with the result being general panic and chaos.
        So my guess is that banks want to say “if we can’t net across transactions, we want to unwind the whole thing entirely, including transactions already closed out”. They just pull out of the risk entirely. This may be for want of another obvious remedy if the Transactions were deemed to be individual contracts.
        I could be wrong, but that is my first guess. I’m sure the ISDA drafting committee will tolerate something that is arguably redundant if there is any chance that it might make a difference to something like this. Again, I don’t usually take this stultifyingly conservative sort of line, but better safe than sorry in this case.

        • Agreed. Section 1(c) underpins the single agreement concept and forms the basis for netting.

          Without the “redundant” language, it opens up the possibility of a counterparty asserting that the ISDA agreement was more for a matter of convenience rather than intending to enter into a netting agreement for transactions – this could strike at the very heart the single agreement concept and certainty should the relationship have to prematurely. When it comes to disclaiming liability for a derivative transaction, one would have to attack any kink in the armour in order to set it aside.

          One also cannot disregard that the ISDA Agreement is used globally and parties could agree to non-exclusive jurisdiction. Courts which have little to no experience in the realm of derivatives may sometimes be unfairly sympathetic to the non-banking counterparty – the language has to deal with the intention of the parties as unequivocally as possible.

          The ISDA has been drafted to assume the worst (whether in terms of commercial viability or ethics) of either party. In light of the technicalities, how litigation of derivatives has evolved, how transactions are entered into and managed operationally, the ISDA Agreement, no matter how convoluted it seems, is drafted as it is for a reason – it is not perfect but it’s well thought out in its drafting.

          Of course, if anyone can propose digestible language for the even more technical parts of the ISDA Agreement (say section 6(e)) without the loss of certainty, that is something I would like to see!

  3. The 2002 ISDA Master Agreement is very closely based on the 1992 version – they really only made commercial changes – and in 1992 drafting was even more paleolithic than it is now. Considering that, and considering it was drafted by a committee – and a committee of New York firms at that – I think the ISDA Master Agreement is mercifully well-drafted.

    There is no prospect of any updated version of the ISDA Master Agreement being much different. There are billions (possibly trillions) of dollars hanging on the wording in this agreement, because most OTC derivative contracts use it. It must be some of the most “valuable” contract language in the world by that measure. (I can’t think of many other things in that league – US Treasuries and other government bonds, perhaps.)
    I know it is anathema to say this here, but litigation has decided what many of these clauses mean and everyone is happy that those things are settled. Huge swathes of it could be improved, but I am certain they will just let sleeping dogs lie when it comes to the drafting rather than risk increasing the uncertainty over what any of it means. I wouldn’t really support that stance for any other piece of contract drafting, but it is a braver man than I that would start amending the ISDA Master Agreement for style.


Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.