Calculation Agent

From The Jolly Contrarian
Jump to navigation Jump to search
ISDA Anatomy™
incorporating our exclusive ISDA in a Nutshell™
Calculation Agent.png
Secret Co-Calculation Agent Apathy, in a still from Otto Büchstein’s under-appreciated 1952 Opco Boone noir The Co-Calculation Agent Who Loved Me
Resources Wikitext | Nutshell wikitext | 1992 ISDA wikitext | 2002 vs 1992 Showdown | 2006 ISDA Definitions | 2008 ISDA | JC’s ISDA code project
Navigation Preamble | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14

Events of Default: 5(a)(i) Failure to Pay or Deliver5(a)(ii) Breach of Agreement5(a)(iii) Credit Support Default5(a)(iv) Misrepresentation5(a)(v) Default Under Specified Transaction5(a)(vi) Cross Default5(a)(vii) Bankruptcy5(a)(viii) Merger without Assumption
Termination Events: 5(b)(i) Illegality5(b)(ii) Force Majeure Event5(b)(iii) Tax Event5(b)(iv) Tax Event Upon Merger5(b)(v) Credit Event Upon Merger5(b)(vi) Additional Termination Event

Index: Click to expand:

Comments? Questions? Suggestions? Requests? Insults? We’d love to 📧 hear from you.
Sign up for our newsletter.

Premium version of this page

Calculation Agent
/ˌkælkjʊˈleɪʃən ˈeɪʤənt/ (n.)
One who calculates things on behalf of contracting counterparties. In theory, under any kind of finance contract, but in practice, mainly in the ISDA and its extended fan-fiction universe (GMSLA, GMRA, DRV, FBF etc), and in the documentation of debt securities.

To be endlessly compared and contrasted with a “determination agent”, who determines things on behalf of contracting counterparties. Do “calculation” and “determination” differ? Not as far as this correspondent can see. You tend to say, perversely, that a Calculation Agent determines things, and a Determination Agent calculates things, but largely because elegant prose has a horror of repetition. But will that stop over-enthusiastic members of the bar waxing lengthily about how they do differ? it will not.[1]

In the ISDA

The ISDA Schedule has space to specify who the Calculation Agent should be but, curiously, gives scant hints as to what such an agent should do: the term isn’t otherwised defined — or even used — in either version of the ISDA Master Agreement. The role comes in to its own under the 1995 English Law CSA and the various definitions booklets ISDA has published. The Calculation agent can differ from transaction to transaction, and while guileless negotiation teams may therefore spend a great deal of energy haggling fruitlessly about who should be the Calculation Agent, and what rights the other poor sap should have to challenge its determinations, in practice it will be the dealer.

As well as in the ISDA Master Agreement, the term is defined separately in each definition booklet, giving everyone a nice opportunity for some clarifying hierarchy action.

Co-calculation agent

There’s an old saying:

A co-calculation agent is no calculation agent.

However superficially neat this might seem to the age-old valuation dilemma of who should price the trade, it suffers in one important respect: unless the parties agree on the determination, the parties — er — won't agree on the determination. And then what do you do?

Disputing a Calculation Agent’s determinations

One of the great old saws of negotiation in any capital markets transaction is what to do if you don’t like the number the Calculation Agent comes up with.

This springs from the ancient, primal fear that flutters in the breast of every buy-side legal eagle, and which is best articulated thus: All swap dealers are profit-obsessed predators. They eat their own young, so will hardly blanche about eating yours.

They will, thus, not pause to breathe before ripping clients’ faces from their skulls should the merest opportunity to do so arise. Derivatives, we know, are financial weapons of mass destruction™ even on a good day, so giving one of those dastardly dealers the unilateral right to determine values on the economic equivalent of an ICBM without any comeback would be insane.

Therefore, I must have a mechanic to dispute a calculation my dealer makes that seems “off”.

Now, to be fair, there was a time, in living memory, when swap dealers would rip off their clients’ faces at the merest opportunity, in some markets. “Cheapest to deliver” options in managed CDO portfolios spring unhappily to mind. Banks used to “prop trade” a lot more than they do now. The year of our lord 2006 was a wild time. There are different regulations now: capital rules, clearing obligations, prop-trading restrictions, and best execution obligations. Dealers play much more of an agency role now. They act like brokers ought to: they execute as riskless principals, they earn their keep from commissions and not by trading against their clients. In observable, liquid markets, clients can see for themselves whether they are getting good prices, and can take their business elsewhere if they are not.

Yet we are still beset by fear of dealer mendacity.

But even if they were justified, the dispute mechanisms our learned friends habitually confect boil down to seeking prices from other “reference dealers”. The exact method can be baroque: appeals to Law Society presidents, competing panels of reference dealers, fallbacks dealers, co-calculation agents, discarding outliers, splitting differences and so on, but each is predicated on the idea that a disinterested market participant — who is still, remember, a rapacious dealer, just one without a dog in the fight — will be less inclined to tear your face from the bone than the one with whom you have had a fruitful twenty-year relationship. This feels wishful.

As does the idea that a disinterested dealer will have any interest in offering a price at all. Why would it? What does it have to gain from reverse engineering a historical market value for a a security it will not actually get to trade? It will have its own legal eagles, they will be fearful, as all legal eagles are, and will worry about getting sued, or being joined in litigation. Their counsel will be to not get involved. So good luck with that leading reference dealer.

This is quite different from the case of your actual dealer. It will base its marks on where it has actually executed its own delta-hedge. In other words, these are prices at which it has actually traded, and for which it has off-setting liability. Why would it ever accept a hypothetical price from a disinterested third party over its own, actually traded, price? This not hypothetical: it stands to lose real money if it does.

All hail the commercial imperative

The real answer to the calculation agent dispute conundrum is the commercial imperative. Dealers trade countless swaps every day. Their business viability depends on satisfied existing customers coming back and placing more orders. Customers who have had their faces ripped off don’t do that.

This is basic business common sense. Still, we sense it will be a foreign country to buyside legal-eagles out there who will never be persuaded that swap dealers can behave like decent human beings, even if their own commercial interest encourages it. We know there are, in fact: their Byzantine valuation dispute mechanisms pepper ISDA portfolios from New York to Tokyo.

If you are one, here is a question: when did your client last actually invoke a dispute mechanism in anger?

Do write in and let us know.

See also


  1. Pedants will note the different roles played by the Calculation Agent and the Determining Party in the 2002 ISDA Equity Derivatives Definitions.