Magna Charta Webinars

W E B I N A R S

D E R I VA T E N R E C H T SPREKER PROF. MR. DR. M.G.C.M. PEETERS, BIJZONDER HOOGLERAAR DERIVATENRECHT UNIVERSITEIT VAN AMSTERDAM, ADVOCAAT NAUTHADUTILH 28 OKTOBER 2013 14:00 – 17:15 UUR

Magna Charta is onderdeel van de Academie voor de Rechtspraktijk Postbus 13346

|

3507 LH Utrecht

|

T 030 – 220 10 70

magnacharta.avdrwebinars.nl

|

F 030 – 220 53 27

ALUMNUS COLLEGES 6 COLLEGES OVER DIVERSE ONDERWERPEN

START 6 DECEMBER 2013 De sprekers: Actualiteiten Personen- en Familierecht (6 december 2013) door prof. mr. A.J.M. Nuytinck Actualiteiten Goederenrecht en Insolventierecht (13 december 2013) door prof. mr. drs. J.W.A. Biemans Actualiteiten Ondernemingsrecht (16 december 2013) door prof. mr. C.A. Schwarz Actualiteiten Verjaring en Verval (18 december 2013) door prof. mr. J.L. Smeehuijzen

4 PO

Waarheid in het materiële en formele strafrecht (19 december 2013) door prof. mr. M. Otte Actualiteiten Dagvaardingsprocedure (30 december 2013) door prof. mrr. M.J.A.M. Ahsmann Een gelimiteerd aantal van 30 personen kunnen deelnemen aan deze unieke colleges.

Vol is vol, dus schrijf u snel in!

Kosten: €125,- voor onze alumni.

Inhoudsopgave Mr. dr. M.G.C.M. Peeters Literatuur Peeters, Marcel, Derivaten: recht en risico's. oratie UvA, 2011

p. 4

Broekhuizen, K.W.H., en Peeters, M.C.G.M., 'Vervroegde beĂŤindiging van swaps', Tijdschrift voor Effectenrecht, 2000, pp. 79-84 (i.h.b. pp. 79-80)

p. 21

Peeters, Marcel , 'OTC-derivaten, beurseffecten en risico's', Onderneming & Financiering, 2005 (67), pp. 63-68 (i.h.b. pp. 63-68)

p. 27

Nijenhuis, A. en Verhagen, H., '"Netting": een beschouwing naar Nederlands recht' , De Naamlooze Vennootschap, 1994 (4), pp. 96-103

p. 42

Grove, Richard, 'Valuation in the Context of Derivatives Litigation', Capital Markets Law Journal, 2011, pp. 149-162.

p.50

Unidroit Principles on the Operation of Close-Out Netting Provisions, 2013 (i.h.b.

Introduction, Principles 1, 2, 6, 7 en 8 en de bijbehorende Key considerations en p. 65

Explanation & commentary) Aanbevolen, niet opgenomen in deze reader Broekhuizen, K.W.H., en Peeters, M.C.G.M., 'Aantekeningen bij het begrip Financieel instrument', Aantekening 221 bij art. 1:1 Wft, GS Toezicht FinanciĂŤle Markten, 2010 ISDA 2002 Master Agreement

3

Niet bestemd voor publicatie; niet verder verspreiden of citeren.

De definitieve versie wordt gepubliceerd op: http://www.oratiereeks.nl/

DERIVATEN: RECHT EN RISICO’S

Marcel Peeters

12 oktober 2011

4

Mevrouw de Rector Magnificus, Geacht Curatorium van de leerstoel Derivatenrecht, Geacht Bestuur van de Stichting Effecten, Markten en Regulering, Beste overige toehoorders,

Van harte welkom, allemaal. Het is mooi om zo'n plezierig heterogeen samengesteld gezelschap hier voor me zien. Maar naast al dat plezier nu enige ernst, want ik wil het vanmiddag hebben over massavernietigingswapens. Voordat u nu uw uitnodiging voor deze bijeenkomst gaat controleren en voordat een enkele scheikundige onder u denkt “Ha! Chemische derivaten!”, zeg ik maar meteen dat het toch echt over financiële derivaten zal gaan. “Financial weapons of mass destruction”, zo zijn deze derivaten ooit aangeduid door de bekende Amerikaanse ondernemer Warren Buffett. Ook andere kleurrijke en enigszins verontrustende omschrijvingen zijn zonder al te veel moeite te vinden, ook in serieuze literatuur: “the wild beast of finance” en “afschuwelijk complexe producten”. En dan is er dat verbazingwekkende getal van zeshonderduizend miljard dollar dat de totale waarde van alle uitstaande derivaten zou representeren. Een zes met veertien nullen, 600 biljoen / 600 trillion.

Inderdaad een hele massa, en één die zo ver buiten gewone denkkaders ligt dat dat op zich al verontrustend lijkt.

Wie voor minder opwindende en meer duidelijke, analytische omschrijvingen op zoek gaat in het nuchtere Nederlandse recht komt van een koude kermis thuis. Inderdaad staat in onze grote Wet op het financieel toezicht, de Wft, in het allereerste artikel een definitie van derivaten. Die is overigens niet als zodanig aangeduid, maar een onderdeel van de definitie van de meer algemene term “financieel instrument”. Die definitie begint met een paar onderdelen die onder meer betrekking hebben op klassieke effecten zoals aandelen en obligaties. Maar de laatste zeven onderdelen beschrijven derivaten.

Het is een lange opsomming van wat op juristen die zich niet bezighouden met deze materie moet overkomen als een rariteitenkabinet. Voor derivatenspecialisten rijst het beeld op van een porseleinkast waarin weliswaar niet een olifant maar toch een klein huisdier heeft rondgesnuffeld, dat wat dingen door elkaar heeft gegooid en ook het een en ander heeft gebroken. Die definitie, waarop ik straks terugkom, is dus niet erg behulpzaam voor wie op duidelijkheid en systematisering uit is. Eigenlijk is dat symptomatisch voor veel wettelijk derivatenrecht.

Misschien verrast dat u. Misschien had u verwacht dat derivaten een goede, gedegen wettelijke regeling kennen, alleen al vanwege hun kennelijk enorme kwantitatieve belang –

5

2

die 600 biljoen. Ik vind het niet zo verrassend. Ik denk dat het gebrek aan duidelijkheid en systematiek goed verklaarbaar is. Veel derivaten waar we tegenwoordig mee te maken hebben, zijn namelijk betrekkelijk nieuw en datzelfde geldt voor de derivatensector als onderdeel van de financiële sector. Natuurlijk, voor een deel is er niets nieuws onder de zon. Derivaten zoals opties en termijncontracten bestonden al in het klassieke oudheid en – dichter bij huis en in de tijd – in de Nederlandse Gouden Eeuw. In de Nederlandse Faillissementswet, die aan het eind van de negentiende eeuw is ingevoerd, stond van meet af aan een regel van derivatenrecht: artikel 38, dat over termijncontracten gaat.

Maar ongeveer dertig jaar geleden – en dat is toch vrij recent in historisch perspectief – begint een nieuwe ontwikkeling die je zonder overdrijving een derivatenrevolutie kunt noemen. Ik zal straks meer zeggen over de aard van die revolutie. Nu wil ik alvast kwijt dat die revolutie geen geïsoleerd fenomeen was, maar ingebed in een veel bredere ontwikkeling in het denken over risico’s en de wijze waarop in de praktijk met risico’s wordt omgegaan. In feite duurt die derivatenrevolutie, met al zijn revolutionaire ups en downs, nog steeds voort. Daarom is het volgens mij niet verwonderlijk dat er inconsistenties, lacunes en misslagen in wettelijk derivatenrecht zijn te ontwaren. Deze toestand weerspiegelt ook voor een deel de stand van zaken in de rechtswetenschap en de rechtspraktijk.

Deze staat van wet en recht in brede zin heeft ook de inhoud van de rest van mijn verhaal bepaald. Het vakgebied is in feite zelf het probleemgebied. Ik wil daarom proberen enig systeem in derivatenrecht aan te brengen, en derivaten zelf nader op hoofdpunten te analyseren. Als bijproducten van die analyse en systematisering komen dan vanzelf ook enkele van die gebreken van het bestaande recht aan de orde. Zeker vandaag vind ik echter belangrijker om vooruit te kijken en enkele onderzoekthema’s aan te stippen. Ik doe dat overigens primair vanuit privaatrechtelijk perspectief. Toezichtrechtelijke aspecten komen wel degelijk ook aan de orde, maar derivaten zijn op de eerste plaats een privaatrechtelijk fenomeen. Bovendien past de analytische aanpak van het privaatrecht beter bij de benadering die ik wil volgen.

Contractenrecht

Als ik privaatrecht zeg, bedoel ik op de allereerste plaats contractenrecht. Want derivaten zijn contracten. Dat is een simpele constatering, maar toch een die het waard is om benadrukt te worden. De contractuele aard van derivaten komt namelijk niet duidelijk naar voren bij de

6

3

derivaten waar de meesten van u waarschijnlijk wel enigszins bekend mee zullen zijn, namelijk beursgenoteerde derivaten, zoals aandelenopties – calls en puts.

Neem een calloptie Philips. Degene die gewoonlijk als de koper van zo’n optie wordt aangeduid, heeft het recht om op enig moment in de toekomst aandelen Philips tegen een vooraf vastgestelde prijs te kopen. Stel u wilt 10 van die opties kopen en u geeft uw bank opdracht om dat voor u “op de beurs” te doen. De bank voert uw order uit en er vindt dus een transactie op diezelfde beurs plaats. Wie heeft nu, als verkoper, aan u, de koper, die opties verkocht? “Geen idee” zegt u misschien en in wezen is dat het goede antwoord. De beurshandel en de afwikkeling van beurstransacties zit namelijk zo in elkaar dat die verkoper niet uw directe wederpartij is. De uitvoering van uw opdracht leidt niet tot één contract maar tot een hele keten van contracten, en in die keten is “het optiecontract” niet goed aan te wijzen.

De identiteit van de verkoper, die helemaal aan het andere eind van die keten te vinden is, zal u niet bekend zijn en is voor u ook irrelevant.

Het correcte zicht op een optie als contract kan verder worden vertroebeld door het gebruik van termen als “koper” en “verkoper” van een optie. Inderdaad worden die termen in de praktijk gebruikt, en dat zowel voor derivaten als voor aandelen. Maar het verschil is dat aandelen ook onafhankelijk van de handel bestaan. Juist daarom kunnen die op de beurs ook letterlijk gekocht en verkocht worden. Maar opties en andere derivaten komen pas dóór de beurstransactie zelf tot stand; ja, de transactie ís de optie. Er wordt dus in juridische zin niets gekocht en verkocht. Omgekeerd, wie die praktijktermen wel letterlijk neemt en het contractuele aard van een derivaat over het hoofd ziet, stopt derivaten al gauw in een keurslijf dat eigenlijk gemaakt is voor klassieke effecten. Dat heeft hier en daar in het toezichtrecht ook wel tot problemen geleid. Ik wijs bijvoorbeeld op de moeizame discussies (haarkloverijen zoals u wilt) over wie als uitgevende instelling van beursopties moet worden beschouwd (en dus als mogelijk prospectusplichtig).

De derivatenrevolutie die ik al even noemde, begon echter niet in de beurssector, maar met zogenaamde OTC-derivaten. OTC staat voor over-the-counter. OTC-derivaten zijn derivaten die niet op een beurs tot stand komen, maar over-the-counter in wat je als een traditionele contracteersituatie zou kunnen omschrijven. Twee partijen onderhandelen met elkaar; andere potentiële partijen kijken en doen in beginsel niet direct mee. OTC-derivaten worden privately negotiated, komen bilateraal tot stand in plaats van in de multilaterale beurshandel.

De derivatenrevolutie bestond er aanvankelijk nou net in dat er veel meer derivaten en ook geheel nieuwe derivaten in zo'n over-the-counter constellatie, en niet op beurzen, werden afgesloten. Het allereerste begin kan zelfs heel precies bepaald met een concreet OTC-

7

4

contract, namelijk de gecombineerde rente- en valutaswap tussen de IBM en de World Bank in augustus 1981. Bij dergelijke derivaten is het natuurlijk volstrekt duidelijk dat het derivaat een contract is.

Er is een aantal redenen waarom partijen niet via de beurs maar over-the-counter een derivatentransactie zullen doen. Een eenvoudige kan zijn dat het type derivaat dat partijen wensen, niet of nauwelijks als zodanig op beurzen worden verhandeld. Renteswaps zijn daar nog steeds een voorbeeld van. Het voornaamste motief is dat partijen in beginsel in zo'n overthe-counter situatie contractsvrijheid ten volle kunnen benutten, niet gebonden worden door de vaste contract specifications van beurzen.

De OTC-revolutie heeft er overigens toe geleid dat de OTC-sector als geheel al geruime tijd enkele malen groter is dan de beurssector. De verhouding tussen OTC en beurs (en handelsplatformen die daartussenin zitten) zal de komende jaren overigens waarschijnlijk veranderen. Zo heeft de G-20 in een reactie op de recente crisis afgesproken dat derivaten op meer transparante wijze moeten worden verhandeld en minder privately negotiated zouden moeten worden. Binnen de EU zijn er inmiddels concrete initiatieven genomen die daartoe moeten leiden. De handel zal zich verplaatsen naar beurzen, maar ook andere handelsplatformen, organized trading venues.

Dat betekent zeker niet dat OTC-handel geheel ongeorganiseerd en ongereglementeerd zou verlopen. Straks zal ik het nog over de bijzondere rol van derivatendealers in de OTC-sector hebben. Hier wil ik de aandacht vestigen op de aanzienlijke standaardisering van contractsvoorwaarden voor OTC-derivaten. Dat dergelijke derivaten “maatwerk” zijn, mag niet verhullen dat er ook veel is dat volstrekt standaard is. Anders zou de OTC-sector natuurlijk ook nooit die enorme omvang hebben bereikt. Verreweg het meest gebruikt worden voorwaarden opgesteld door de internationale branche-organisatie van de OTC-sector: ISDA, de International Swaps and Derivatives Association. Die documentatie beslaat meer dan duizend pagina's. Een omvangrijk begrippenapparaat en uitgebreide standaardvoorwaarden die partijen als bouwstenen voor hun contracten kunnen gebruiken. Voor de alledaagse internationale OTC-praktijk is de ISDA-documentatie eigenlijk het privaatrechtelijke derivatenrecht. Contractsvrijheid bestaat daarom in de meeste gevallen in het selecteren van geschikte bouwstenen en niet – om de metafoor voort te zetten – in het uit onbewerkte klei en cement optrekken van bouwsels naar geheel zelfbedacht ontwerp.

[Even terug naar die lacunes in wettelijk derivatenrecht. Inderdaad kent het BW geen regeling van, bijvoorbeeld, uitoefening van opties. Een echte lacune is dat naar mijn oordeel niet. De

8

5

markt, de derivatensector, zowel beurs als OTC, weet dergelijke min of meer technische onderwerpen meestal wel adequaat te regelen, al dan niet via trial-and-error. Ik denk bij dat laatste bijvoorbeeld aan de afwikkeling van kredietderivaten volgens physical settlement, zoals die aanvankelijk in de ISDA-documentatie was geregeld. Die regeling bleek onvoorziene effecten bleek te hebben, overigens al lang vóór de recente crisis.]

Risico-allocatie

Derivaten zijn dus contracten, maar wat maakt derivatencontracten nu tot derivaten? Heel algemeen gezegd, zijn derivaten financiële instrumenten die afgeleid zijn van – derived from – andere financiële instrumenten, of andere grootheden of variabelen. In het jargon zijn die grootheden de onderliggende waarden of underlying. Zo zijn aandelenopties afgeleid van de onderliggende waarde aandeel of een aandelenindex. Rentederivaten hebben als onderliggende waarde, u begrijpt het, de rente. Een renteswap is de afspraak dat de ene partij aan de ander zal betalen een vast percentage (een “rentepercentage”) over een rekenbedrag (bijvoorbeeld 100 miljoen). In ruil daarvoor betaalt de andere partij periodiek een variabele rente over datzelfde bedrag, bijvoorbeeld de zesmaandsrente op euro’s, die dan tijdens de looptijd inderdaad weer elke zes maanden opnieuw wordt vastgesteld.

Maar er zijn ook derivaten – termijncontracten, opties en swaps – op grondstoffen zoals olie en ijzer. Er zijn weerderivaten, waaronder de verschuldigde betalingen afhankelijk kunnen zijn van de temperatuur gemeten op een bepaald weerstation of de windsterkte. Nog andere voorbeelden zijn inflatieswaps en – ietwat luguber – death-rate swaps: derivaten met sterftecijfers als underlying. In feite zijn er geen juridische beperkingen, hooguit economische in de zin dat er er voldoende spelers in de markt moeten zijn met een economisch belang bij een bepaalde onderliggende waarde. Dat er amper beperkingen zijn, wordt ook geïllustreerd door de Wft-definitie van derivaat, die opsomming van alles en nog wat en die ook nog eens een open einde heeft.

Analytisch gezien brengt die Wft-definitie [en meer in het algemeen: concentratie op het type underlying] ons dus niet veel verder. Paradoxaal doet een bepaling die vrij recent uit het Burgerlijk Wetboek is verdwenen dat wél. Ik doel op artikel 1811 van Boek 7A BW dat over kansovereenkomsten gaat – pardon: ging – en dat je met enige goede wil een regel van derivatenrecht zou kunnen noemen. Het schrappen van dit artikel bij de invoering van het nieuwe verzekeringsrecht is overigens een van de misslagen waarover ik het eerder had. Bais en Jongmans hebben dat onlangs uiteengezet in het Tijdschrift voor Financieel Recht.

9

6

Het eerste lid van dat artikel 1811 gaf een definitie van “kansovereenkomst”, die eigenlijk heel goed uitdrukt waar het hier om gaat. Daarvoor moet je overigens wel door het negentiende-eeuwse Nederlands heenkijken:

“Eene kans-overeenkomst is eene handeling, waarvan de uitkomsten, met betrekking tot voordeel en nadeel, het zij voor alle de partijen, het zij voor eenige derzelve, van eene onzekere gebeurtenis afhangen.”

De moderne derivatenversie die ik ervan maak is:

“Een overeenkomst waarbij de waarde van de verschuldigde prestaties van ten minste een van de partijen afhankelijk is van een onzekere, economisch relevante gebeurtenis.”

Die wezenlijke eigenschap van derivaten, de wezenlijke rol die onzekerheid speelt, betekent dat derivatenrecht risicorecht is. Voor alle duidelijkheid: ik spreek hier over risico in de ruime zin des woords: positieve en negatieve risico's, goeie en kwaaie kansen, kansen “met betrekking tot voordeel en nadeel”, zoals artikel 1811 het formuleerde.

Kan het iets preciezer en informatiever? Jazeker, en daarbij kan het tweede lid van dat verdwenen artikel 1811 van pas komen. Dat somde een aantal kansovereenkomsten op, van verzekeringen tot overeenkomsten van spel en weddenschap.

Tot het wezen van verzekeringen behoort dat de verzekeraar negatieve risico’s van de verzekerde overneemt: het risico op het verloren gaan van een huis door brand, van een schip door schipbreuk, aansprakelijkheidsrisico’s. Dat is ook de strekking van het verzekeringsrecht, waaraan het zogenaamde indemniteitsbeginsel ten grondslag ligt. Zoals het BW het zegt: schadeverzekering heeft de strekking om vermogensschade te vergoeden die de verzekerde zou kunnen lijden. En: door een uitkering onder zo'n verzekering, bij realisatie van het risico, mag de verzekerde niet in een duidelijk voordeliger positie geraken.

Een verzekering dekt dus reële risico’s die de verzekerde loopt. Bij de kansovereenkomst van spel en weddenschap, zoals artikel 1811 die omschreef, is het precies omgekeerd: er is helemaal geen risico buiten het spel en overeenkomst en het is de overeenkomst die tezamen met het spel het risico voor de partijen in het leven roept.

10

7

Derivaten zitten hier tussen in. [Overigens heel letterlijk in dat tweede lid van artikel 1811.]

Het gaat wel degelijk om reële risico’s die onafhankelijk van de overeenkomst bestaan, zoals onzekerheden verbonden aan aandelenkoersen, rentestanden, grondstoffenprijzen, het weer en wat dies meer zij. Maar vereist is niet dat een van de partijen ook, afgezien van het derivaat, zélf risico's loopt op de onderliggende waarde. Net zoals bij spel en weddenschap kan het de overeenkomst zijn waardoor het onderliggende risico pas relevant wordt voor partijen.

Hoewel dus vaak over derivaten wordt gesproken in termen van risico-overdracht, is het nauwkeuriger om van risico-allocatie te spreken. Derivaten worden natuurlijk ook gebruikt om te beleggen en dat is er juist wel op gericht om “in een duidelijk voordeliger positie geraken”. Geen indemniteit maar opportuniteit, zou je kunnen zeggen.

Derivaten tussen verzekering en spel dus, maar ik laat spel en weddenschap nu verder voor wat het is, want deze overeenkomsten hebben niet tot uitvoerige juridische uiteenzettingen geleid. Overigens niet verwonderlijk omdat de wet vanouds aan dergelijke overeenkomsten de afdwingbaarheid ontzegde. [Of zoals het nog niet verdwenen artikel 1825 zegt: “De wet staat geene regtsvordering toe, ter zake van eene schuld uit spel en weddenschap voortgesproten.”]

Meer inspiratie kan de derivatenjurist ontlenen aan het verzekeringsrecht dat evenals derivatenrecht ook risicorecht is. Dat recht kent wél een uitvoerige en relatief moderne wettelijke regeling en natuurlijk een zeer omvangrijke jurisprudentie. Inspiratie uiteraard met inachtneming van de verschillen. Zoals gezegd, een typisch verzekeringscontract verzekert een specifiek, geïndividualiseerd risico-object. In derivatenterminologie: de onderliggende waarde, de underlying, van een verzekeringsovereenkomst is specifiek, geïndividualiseerd in relatie tot een van de partijen. Onder een typisch derivatencontract is dat anders: de onderliggende waarde is niet specifiek en geïndividualiseerd, maar juist generiek: de driemaandsrente, de AEX-index, de goudprijs (en niet dít gouden sieraad), gemiddelde sterftecijfers (en niet mijn of uw leven).

Zulke generieke risico’s zijn bijvoorbeeld minder gemakkelijk te manipuleren dan specifieke, geïndividualiseerde risico’s. Een verzekerde kan invloed uitoefenen op het verzekerde object – “zijn” verzekerde object – en de verzekerde risico’s, bijvoorbeeld door minder goed voor een verzekerd object te zorgen en risico's daardoor te laten toenemen (moral hazard).

Maar bij de rente, de AEX-index en sterftecijfers is moeilijker voor te stellen dat een individuele partij dat klaarspeelt – nog een luguber voorbeeld: je hebt een echt massavernietigingswapen nodig om het gemiddeld sterftecijfer van de Nederlandse bevolking te manipuleren.

11

8

Manipulatie van, of neutraler: invloed op de underlying, is echter ook bij derivaten zeker niet uit te sluiten. Generiek is immers niet hetzelfde als niet-manipuleerbaar. Een grote bank kan transacties doen in aandelen, die tevens de onderliggende waarde zijn van OTC-derivaten die die bank als dealer afsluit.

Een andere parallel met verzekeringsrechtelijke thematiek is te vinden in de situatie dat een bank dealer is in kredietderivaten op een bepaalde onderneming, en tevens bankier is van die onderneming. Uit die hoofde zal de bank bijzondere informatie over de kredietwaardigheid van die onderneming hebben.

Dergelijke

fenomenen

worden

overigens

op

enkele

punten

geadresseerd

in

standaardvoorwaarden die bijvoorbeeld in ISDA-documentatie zijn opgenomen. Dat geeft ook steun aan mijn idee dat het zinvol is om, onder meer aan de hand van vergelijkbare verzekeringsrechtelijke leerstukken, de privaatrechtelijke aspecten van die fenomenen te analyseren.

Terug naar de algemene karakterisering van derivaten als contracten van risico-allocatie. Ik wil de aandacht vestigen op wat ik het analytisch gebruik van derivaten zou willen noemen.

Derivaten stellen een belegger in staat om risico’s die vanouds als een ondeelbaar pakket werden beschouwd op te splitsen in afzonderlijke componenten. Neem het voorbeeld van het risico – het pakket risico’s – van het houden van een aandeel. Door middel van opties kan bijvoorbeeld het risico van koersdaling separaat van de aandelen zelf worden gealloceerd.

Maar het kan nog veel verder gaan. Door geschikte combinaties van opties is het bijvoorbeeld ook mogelijk om enkel het risico van een daling van minstens 5% maar niet meer dan 6% te accepteren of juist af te stoten.

Dat is volgens mij een van de cruciale vondsten van de derivatenrevolutie geweest: dat risico’s op te splitsen en te verhandelen zijn, los van de onderliggende waarden. Een belegger hoeft zijn aandelen Philips niet te verkopen om niet langer de risico's daarop te lopen, de kansen op koersverlies en koerswinst. Hij kan dat ook met derivaten doen en dan zelfs heel precies aangeven welke risico’s hij kwijt wil en welke hij wil behouden. Omgekeerd hoeft een belegger geen aandelen Philips te kopen om in Philips te beleggen; zij kan dat ook synthetisch doen met een aandelenswap. [Een onderneming die van haar bank geld geleend heeft tegen een variabele rente, kan die transactie effectief transformeren in een vastrentende lening door een geschikte renteswap aan te gaan. Amendement van de leningovereenkomst met de bank is daarvoor niet nodig.]

12

9

Complexiteit

Nu ik wat dieper op derivaten ben ingegaan, zou ik willen terugkeren naar die kleurrijke en ook verontrustende beschrijvingen waarmee ik begon. Allereerst deze: is een typisch OTCderivaat nou een “complex” contract? Mijn antwoord daarop is: nee, eigenlijk niet. Omdat het debat over deze vraag belangrijk is, moet het niet ontaarden in een welles-nietes. Daarom heb ik in de theoretische literatuur over complexiteit van systemen gezocht naar een min of meer objectief criterium van complexiteit. Vrij direct toepasbaar is het volgende criterium: kijk naar de lengte van een beschrijving van het fenomeen waarvan je de mate van complexiteit wilt vaststellen, naar een beschrijving in termen ontleend aan een gegeven begrippenapparaat.

Welnu, voor het merendeel van derivaten die in omloop zijn, zijn de beschrijvingen beperkt in omvang. U zult zich, naar ik hoop, nog mijn eerdere beschrijvingen van aandelenopties en renteswap herinneren. Die waren vrij kort en, naar ik hoop, toch vrij duidelijk.

Laat ik nog een voorbeeld ter adstructie geven, namelijk kredietderivaten, vaak als het complexe derivaat per uitstek gekenschetst. Neem een kredietderivaat, een credit default swap of CDS op Italië. Een gebruikelijke looptijd van zo'n CDS is vijf jaar. Gedurende die vijf jaar betaalt de ene partij – de koper van kredietbescherming of protection buyer – in beginsel elk kwartaal

een

premie

aan

zijn

wederpartij.

Blijft

Italië

aan

haar

rente-

en

aflossingsverplichtingen onder haar uitstaande overheidsschuld voldoen, dan is dat alles. Doet Italië dat niet (“Italië failliet”) , dan zal die wederpartij – de protection seller) het verschil tussen de nominale waarde (100%) en de marktwaarde van Italiaanse staatsobligaties moeten vergoeden. Doet Italiaans staatspapier bijvoorbeeld 40% na zo'n credit event en is de CDS afgesloten voor 10 miljoen, dan betaalt de CDS 100 min 40 procent = 60% van dat bedrag uit, dus 6 miljoen. Dat is alles en dat lijkt me toch in de kern een vrij eenvoudig contract.

Dit is een soort synopsis van een derivatencontract. Maar ook als je een typisch OTCderivatencontract in zijn geheel toetst aan de hand van het complexiteitscriterium – lengte van beschrijving in standaardbegrippen – dan blijft die conclusie overeind, want veel van die contracten, zeker als het om rentederivaten of kredietderivaten beslaan niet veel meer dan een aantal A4-tjes.

Natuurlijk moet je zeker bij juridische complexiteit rekening houden met de hoedanigheid van partijen. Beauty is in the eye of the beholder, en dat geldt evenzeer voor een eigenschap als complexiteit. Maar: de OTC-derivatenhandel is er een tussen professionelen voor wie deze beschrijvingen vrij eenvoudig te doorgronden en te bevatten moeten zijn.

13

10

Ik zou echter niet willen dat u nu het idee krijgt dat volgens mij complexiteit geen probleem is in de derivatensector en dat derivaten niet van die riskante producten zijn. Ik maak nog twee kanttekeningen bij complexiteit, die allebei raken aan naar mijn oordeel belangrijke onderzoeksthema’s.

Het eerste punt is dit. Dat afzonderlijke derivaten niet complex zijn betekent natuurlijk niet dat er in de derivatensector geen complexe structuren voorkomen. Ik beloofde u eerder terug te komen op een structuurkenmerk van de OTC-sector: de rol van derivatendealers, meestal grotere banken. De allereerste swap, die tussen de IBM en de World Bank is heel bijzonder omdat beide partijen zogenaamde eindgebruikers waren, endusers in het jargon. Dergelijke transacties komen niet vaak meer voor. Al vrij snel gingen swap dealers tussen de eindgebruikers in zitten. Juist omdat OTC-derivaten maatwerk zijn, is het namelijk lastig om twee eindgebruikers te vinden met twee exact tegenovergestelde risicowensen, die direct zouden kunnen contracteren. In plaats daarvan maken die dealers de markt, zij zijn de marketmakers van de OTC-markt. Een eindgebruiker contracteert dan ook meestal met een dealer.

En zo'n dealer dekt de risico's die hij onder die transactie accepteert af door op zijn beurt andere transacties af te sluiten met andere dealers of eindgebruikers. Die andere dealers doen weer hetzelfde, vaak vooral met weer andere (of dezelfde!) dealers. Zo ontstaat een netwerk van derivatentransacties tussen dealers. Het is alsof twee teams van eindgebruikers touwtrekken, maar niet met een strak touw, maar één waarin het midden van een enorme en complexe kluwen zit. Voor risico-allocatie tussen de eindgebruikers lijkt die kluwen niet nuttig; integendeel deze kan bijvoorbeeld systeemrisico's vergroten. Daarvoor zijn oplossingen bedacht en voor een deel al geïmplementeerd: multilaterale compressie en centrale clearing. Ook dat zijn overigens contractuele technieken – én, inderdaad, complexe technieken.

Dan het tweede punt. In vele gevallen waar derivaten als “complex” worden beschouwd, is er, als ik het goed zie, niet zozeer sprake van grote complexiteit, maar van grote risico’s die door het contract worden gealloceerd. Er zijn vele voorbeelden te noemen van eenvoudige derivaten waaronder er meestal kleine kansen zijn op zeer groot nadeel voor een partij. Bij het schrijven van aandelenopties wordt u als het goed is daarvoor door uw bank gewaarschuwd.

Als dat grote negatieve risico zich vervolgens realiseert, zal in de meeste gevallen, de partij die dat aangaat zijn grote verlies nemen. Maar soms niet, en komt het tot debat met de wederpartij. Of, het wel degelijk genomen verlies is catastrofaal, leidt tot de ondergang van de onderneming en er komt onderzoek. Een goed voorbeeld is AIG dat ten onder ging aan

14

11

overigens bijzondere kredietderivaten. Inderdaad massavernietigingswapens, al liep het door een bail-out niet op een complete slachting uit.

In een dergelijke gevallen – van debat, proces, onderzoek – kan dan makkelijk het idee opkomen dat die catastrofale contracten te complex waren om te doorgronden. Maar ik zou willen wijzen op ten minste de mogelijkheid dat het toch net iets anders zit. Dat niet het contract complex is, maar dat die partij of beide partijen niet in staat zijn om de risico’s correct te evalueren. Ik gaf zoëven het uitzonderlijke voorbeeld van AIG, maar het punt is van veel groter belang, ook in meer alledaagse situaties. Uit veel onderzoek is gebleken dat de mens, hoe rationeel en professioneel verder ook, in vele situaties niet goed in staat is om rationeel en prudent met kansen en kansberekening om te springen. Het bekende werk van Kahneman en Tversky biedt daar vele voorbeelden van. Wij mensen, ook u en ik, zijn allen kennelijk van nature in meerdere of mindere mate risicoblind. Dat is een constatering die zeker ook juridisch – privaatrechtelijk en toezichtrechtelijk – van het grootste belang is voor derivaten waarbij risico-allocatie centraal staat.

Waardering

Niettemin, een typisch derivatencontract is dus volgens mij niet zo complex. Eigenlijk is het nog eenvoudiger. Een derivaat verdeelt allerlei risico’s, allerlei kansen en kansjes op voordelen en nadelen tussen de partijen. Het resultaat van die allocatie kan worden samengevat in één enkele variabele: de waarde, de marktwaarde van dat contract. Dat is natuurlijk primair een economisch begrip, maar laat ik er meteen bij zeggen dat die waarde juridisch gezien in wezen niet anders is dan het contractsbelang. Nou wordt dit begrip in het recht meestal gebruikt wanneer het mis gaat met de uitvoering van een contract, bij ontbinding en schadevergoeding. In zo’n bijzondere situatie ziet het contractsbelang op de schade die een partij lijdt doordat het contract wordt ontbonden – beëindigd – en niet door beide partijen correct wordt nagekomen. Maar het begrip is ook nuttig in normale situaties, waarin contracten nog lopen en partijen aan hun verplichtingen voldoen. Zoals bij derivatencontracten.

Het contractsbelang, de waarde van een derivaat, kan zowel positief als negatief zijn. Heeft u bijvoorbeeld een termijncontract lopen waaronder u over veertien dagen aandelen Philips mag en moet kopen voor 15 euro per stuk en is de koers van Philips nu 20 euro, dan heeft dat contract voor u een positieve waarde – en voor uw wederpartij, de termijnverkoper, een tegenovergestelde negatieve. Zou Philips echter nu op 10 euro staan, dan heeft het contract

15

12

juist voor u een negatieve waarde. Immers, over twee weken moet u dan 15 euro gaan betalen voor iets dat op dit moment maar tien euro waard lijkt te zijn.

Ik kan het niet laten om heel even iets te zeggen over dat fabuleuze getal van zeshonderdduizend miljard dollar / 600 biljoen. Dat zou de totale waarde van alle uitstaande derivaten zijn. Maar dat is niet zo. Het is namelijk het totaalbedrag van de onderliggende waarden van al die derivaten en niet het totaal van de waarden van de derivaten zelf. Die 600 biljoen is mede daarom als zodanig irrelevant als maatstaf voor risico’s, met inbegrip van systeemrisico’s, die in derivaten liggen besloten. Is het dan beter om het totaal van alle marktwaarden van derivaten te nemen? Wel, als je dat zonder meer zou doen, en al die marktwaarden van al die derivaten van alle partijen zou optellen, kom je ongeveer op nul uit.

Elk contract is een zero-sum game, met tegengestelde waarden voor beide partijen, zoals ik net aangaf. Zo’n optelling komt neer op een grote verdwijntruc: zoëven hadden we nog 600 biljoen en nu niks. Ook dat is dus niet de oplossing en het goede antwoord ligt ook niet ergens in het midden tussen nul en 600 biljoen. Dit is niet de gelegenheid om op deze kwestie in te gaan. Neemt u van mij aan dat zinvolle schattingen van de omvang van de derivatenmarkt uitkomen op bedragen die verschillende orden van grootte kleiner zijn dan die 600 biljoen.

Terug naar mijn betoog over waardering en contractsbelang. Een bijzonder aspect van derivaten is dat hun waarde vrijwel continu door partijen gevolgd kan worden en dat partijen dat ook daadwerkelijk doen. Bij beursgenoteerde derivaten behoeft dat eigenlijk geen toelichting. De alledaagse betekenis van “beursnotering” is min of meer dat de koers van het betreffende derivaat steeds te volgen is. Maar bij OTC-derivatencontracten is het op zich opmerkelijk dat het überhaupt mogelijk is om deze te waarderen. Geen centrale, publieke prijsvorming, maar juist privately negotiated. Dat dat toch kan heeft veel te maken met de bredere ontwikkeling waarin de derivatenrevolutie is ingebed, bijvoorbeeld

op de

ontwikkeling van kwantitatieve modellen in de economie. Het bekendste voorbeeld is het Black-Scholes model uit 1973. Maar dat is overigens een ander verhaal.

16

13

Close-out netting

Ik ben redelijk uitgebreid op waardering ingegaan om een basis te hebben om tot slot een van de belangrijkste onderwerpen van OTC-derivatenrecht te bespreken: close-out netting. Kort iets over beide elementen van dat begrip.

Netting is een term die ik liever niet vertaal. In de praktijk blijkt namelijk dat vertaling meestal de onduidelijkheid vergroot. Waarom? Omdat netting vaak met verrekening wordt vertaald maar netting geen verrekening is, tenminste niet verrekening zoals geregeld in het BW en de Faillissementswet. Voor niet-juristen: verrekening ziet op het tegen elkaar wegstrepen van vorderingen en schulden die twee partijen over en weer hebben. Jansen heeft een schuld aan Peeters van 100 euro en Peeters een schuld van 80 euro aan Jansen. Het gemeenschappelijk deel, 80 euro, wordt door verrekening voldaan en Jansen hoeft alleen het verschil, 20 euro, aan Peeters te betalen.

Netting is dus geen verrekening, maar wat dan wel? Neem de OTC-standaardsituatie: twee partijen zijn in de loop van de tijd verschillende derivaten aangegaan. Naast die afzonderlijke contracten zullen deze partijen dan ook een raamovereenkomst, een master agreement hebben gesloten, die hun complete derivatenrelatie beheerst. Elke partij kan die complete relatie ook waarderen: namelijk door de positieve en negatieve marktwaarden van alle derivaten bij elkaar op te tellen, te salderen. Dat is netting. En net zoals voortdurend de waarden van afzonderlijke derivaten vastgesteld kan worden, geldt dat voor de waarde van de complete relatie.

Close-out is, in dit verband, de vervroegde beëindiging van al die uitstaande derivaten omdat een der partijen wanprestatie heeft gepleegd, failliet is gegaan of zich een ander zogenaamd event of default heeft voorgedaan. Alle master agreements geven de andere partij dan het recht alle uitstaande derivaten vervroegd te beëindigen – partieel te ontbinden in de zin dat alle vorderingen en schulden die in de toekomst – mogelijk! – opeisbaar worden. (In sommige gevallen gaat dat beëindigen zelfs automatisch.) Daarvoor komt in de plaats vergoeding van het totale netto-contractsbelang bepaald door netting.

Dat in het kort is close-out netting. Geen verrekening van schulden (want die zijn er na de close-out niet meer), maar berekening van een netto-waarde – met dank aan Faber die de tegenstelling verrekening/berekening in andere gevallen in zijn dissertatie hanteerde.

17

14

Gaat het om een faillissement, het event of default bij uitstek, en hadden de beëindigde transacties voor de wederpartij van de failliet netto een positieve waarde, dan zal hij dat bedrag claimen bij de curator. Is het negatief dan zal hij dat bedrag aan de curator betalen.

Daarmee is de kous af, dat is tenminste de bedoeling van de contractuele regeling.

De kernvraag is: werkt dit contractuele afwikkelmechanisme in faillissement of wordt het geheel of gedeeltelijk doorkruist door het Nederlandse faillissementsrecht? Het niet erg verrassende antwoord dat standaardopinies en de literatuur geven is: ja, close-out netting werkt ook in faillissement, en dat uiteraard met de nodige kwalificaties. Ik zal u niet vermoeien met de argumenten die voor die conclusie worden aangevoerd, maar volsta met een verduidelijking[, een speldeprik richting wetgever] en een lichte paradox.

De verduidelijking is dat die argumenten geen beroep doen – kunnen doen – op een algemene regeling van close-out netting in de Faillissementswet want die is er niet.

[De speldeprik is dat de wetgever in 2005 bij de implementatie van de Europese richtlijn over insolventie van banken in de fout is gegaan door de richtlijnbepaling over netting agreements te implementeren alsof het een bepaling over verrekening was (en dan nog een bijzonder type verrekening). Bertrams heeft indertijd in zijn bespreking van de implementatiewet al betoogd dat dat niet klopt maar tot een correctie heeft dat tot dusverre niet geleid.]

De lichte paradox is dat de verrekeningsregels toch ook wel een rol spelen in verband met close-out netting. Verrekening en close-out netting mogen dan verschillende mechanismen zijn, maar zij leiden wel tot hetzelfde resultaat, namelijk een enkele vordering op of schuld aan de failliet. De grenzen die door het verrekeningsregime worden getrokken kunnen daarom toch relevant zijn bij de beoordeling van de afdwingbaarheid van een nettingafspraak. Via contractueel netten zou een partij niet een resultaat moeten kunnen bereiken dat hem door het verrekeningsregime ontzegd wordt. Dat is tenminste het basisidee.

Ik illustreer dat idee kort aan de hand van het fenomeen novatie van derivaten, vrij gebruikelijk in de OTC-sector. Een dealer neemt een contract met een derde over van een andere dealer. Wat nu als die derde vervolgens failliet gaat? Wie een parallel met verrekening trekt, denkt dan aan artikel 54 van de Faillissementswet. Dit artikel blokkeert de verrekening van vorderingen en schulden die niet te goeder trouw zijn overgenomen vóór het faillissement.

Geldt iets dergelijks nu ook ten aanzien van close-out netting, met inbegrip van het recent genoveerde contract? En zo ja, hoe moet goede trouw van de overnemende dealer worden

18

15

ingevuld als het overgenomen contract op het moment van novatie een negatieve waarde had maar daarna door een positieve waarde heeft gekregen?

Dat zijn maar een paar van de vragen die nog steeds op zowel nadere analyse als wettelijke regeling wachten. Waren die vragen voor de kredietcrisis vaak nog theoretisch, de crisis heeft vele praktijkgevallen van close-outs opgeleverd en dat maakt de beantwoording ervan een stuk dringender. Ik hoop overigens zeer binnenkort in ieder geval aan die analyse een bijdrage te kunnen leveren.

Slotopmerkingen

Ik rond af. Ik zal dat niet doen door het voorafgaande samen te vatten. De aard van mijn betoog leent zich daar niet goed voor: het is een schets van hoofdlijnen vanuit vogelvluchtperspectief. Nog hoger dan een vogel en ik kom op de maan terecht en vanaf de maan zijn wij allen gelijk – en dat geldt evenzeer voor onze contracten, of het nou derivaten zijn of niet. Dat leek me niet de geschikte wijze om mijn pogingen tot analyse en onderscheid te besluiten.

Ik zou ook nog kunnen benadrukken dat de afgelopen veertig minuten allerlei interessante juridische problemen aan de orde zijn die het waard zijn om onderzocht te worden, maar dat heeft u natuurlijk ook wel begrepen en dat zal ik daarom ook niet doen. En dat vele van die problemen op het snijpunt van recht en economie ligt – daarvoor geldt hetzelfde.

Nee, veel zinvoller en vooral ook plezierig is dat ik tot slot een aantal mensen en instellingen kan bedanken.

Het College van Bestuur van deze prachtige en levendige universiteit, de Decaan van de Faculteit der Rechtsgeleerdheid en het Bestuur van de Stichting Effecten, Markten en Regulering wil ik danken voor het vertrouwen dat zij in mij hebben gesteld en, neem ik aan, stellen. Daarmee heeft u mij nog meer vertrouwen gegeven om van deze leeropdracht echt iets te maken.

Met het Stichtingsbestuur en in het bijzonder de vertegenwoordigers van NYSE Euronext in het bestuur ben ik het graag eens dat Amsterdam al jarenlang door de optiebeurs als de derivatenhoofdstad van Europa beschouwd mag worden. Dat ik in deze stad, dit Chicago aan

19

16

de Amstel, derivatenrecht mag doceren en onderzoeken, maakt mij zeer fier – om het nou eens op zijn Vlaams te zeggen.

De afdeling Privaatrecht A, mijn afdeling bij de Faculteit der Rechtsgeleerdheid, wil ik danken voor de warme ontvangst. Ik heb zelfs een eigen kamer op de evenzeer prachtige en levendige Oudemanhuispoort gekregen, een naar ik begreep zeer bijzondere gunst. Ik hoop dat wij, jullie en ik, nog veel mooi en opmerkelijk werk gaan doen op het terrein van het privaatrecht, al dan niet tezamen en al dan niet financieelrechtelijk. Ik zal de komende jaren overigens ook wel het een en ander over toezichtrecht te zeggen hebben, neem ik aan.

Ik ben ook advocaat, bij NautaDutilh. Nauta en ik hebben in de afgelopen jaren, als ik het neutraal formuleer, een nogal bijzondere relatie gehad, met aan beide kanten geven en nemen.

Dit laatste bedoel ik weer niet neutraal, maar positief en ik dank jullie hier vooral voor alle intellectuele giften. Ik beschouw de rechtswetenschap ook als een empirische wetenschap, dat zult u uit het voorafgaande ook wel hebben opgemaakt, en de mogelijkheid die Nauta mij heeft geboden om jarenlang in de empirie van de derivatenpraktijk actief te zijn, zal, naar ik hoop, ook in de toekomst nog wetenschappelijke vruchten afwerpen.

Dan verwacht u nu waarschijnlijk nog een persoonlijk dankwoord, maar dan moet ik u teleurstellen. Ik ben nu eenmaal zo iemand die bewust niet op Twitter en FaceBook zit omdat hij zijn persoonlijk leven en samenleven ook heel graag heel persoonlijk houdt. Zo’n dankwoord zal ik hier dus niet uitspreken maar op meer geschiktere plaatsen en tijdstippen.

Ik heb gezegd.

Niet bestemd voor publicatie; niet verder verspreiden of citeren.

De definitieve versie wordt gepubliceerd op: http://www.oratiereeks.nl/

20

17

21

22

23

24

25

26

bw 67 quark 4.0.qxd

14-11-2005

06:14

Pagina 63

M R . M ARCEL P EETERS P H .D.

OTC-derivaten, beurseffecten en risico’s

D

1

BIS (2005), tabellen 19 en 23A;‘volume’ is de notional amount.

27

63 Juridische risicobeheersing Een Master Agreement met aanvullingen

Een basisvoorwaarde voor adequate risicobeheersing is dat de rechten en verplichtingen van partijen duidelijk in een contract zijn vastgelegd. Zonder contract is de omvang van financiële risico’s onduidelijk en de beheersing ervan moeilijk of onmogelijk.

Op een markt met vele bilaterale onderhandelingen over vergelijkbare producten, zoals de OTC-markt, ontstaan vanzelf standaardcontracten. Op de internationale derivatenmarkt is zelfs één contract volstrekt dominant: de ISDA Master Agreement.Van de Master Agreement (hierna ook ‘MA’) bestaan twee versies, uit 1992 en uit 2002. De 2002 MA heeft de 1992 MA nog zeker niet verdrongen. Tezamen beheersen zij 90% van de OTC-derivatenhandel.

Termen met beginhoofdletters hebben betrekking op begrippen die in de ISDA Master Agreements of andere ISDA-documenten worden gedefinieerd.

Andere Engelse termen zijn gecursiveerd tenzij dat naar mijn oordeel al te pedant zou overkomen of het om citaten gaat. Afzonderlijke bepalingen zijn Sections, meestal afgekort tot ‘s.’. Omdat ik mij, zoals gezegd, beperk tot krediet- en vergelijkbare risico’s, is de volgende beschrijving van de Master Agreement selectief en is het meestal niet nodig op het onderscheid tussen de twee versies in te gaan. Een andere reden is dat in de Nederlandse literatuur al vrij uitvoerig aandacht is besteed aan de MA: zie bijvoor-

N r. 6 7 / o k t o b e r 2 0 0 5

Kredietrisico is een aan het contract intern effect: het betreft het exposure van de ene op de andere partij. Externe effecten kunnen eveneens grote financiële risico’s opleveren, wanneer zij de financiële verhoudingen, zoals in het contract neergelegd, verstoren of onduidelijk maken. De beheersing van kredietrisico wordt bemoeilijkt als door zo’n extern effect bijvoorbeeld niet meer duidelijk is wat partijen over en weer verschuldigd zijn. In het tweede deel bespreek ik de externe effecten van de beurs en bijbehorende systemen op OTC-derivaten die beursaandelen of -indices als onderliggende waarden hebben.Wat gebeurt er met een OTC-aandelenoptie als het onderliggende aandeel uit de notering wordt geschrapt? Wat als er door een storing van het computersysteem van de beurs op de uitoefendatum van de optie, geen koers van het aandeel tot stand komt?

In het geval van een beursoptie kan ervan worden uitgegaan, dat de beursregels zich over dergelijke

kwesties uitlaten, maar hoe regelen partijen een en ander in een OTC-contract? Ook in dit geval kunnen zij gebruikmaken van standaarddocumentatie van ISDA, de 2002 ISDA Equity Derivatives Definitions.

O & F

erivaten zijn belangrijke financieringsinstrumenten. OTC-derivaten, die bilateraal, over-the-counter, worden uitonderhandeld, zijn kwantitatief belangrijker dan beursderivaten. Eind 2004 was het wereldwijde OTC-volume bijna $ 250 biljoen tegenover bijna $ 46 biljoen voor beursderivaten.1 De veranderlijkheid van de waarde van derivaten en, zeker wat OTC-derivaten betreft, hun soms exotische vormgeving maakt dat de beheersing van kredietrisico’s een grote rol speelt.

Juridische technieken kunnen bijdragen aan het management van die kredietrisico’s en verwante risico’s. Een aantal van die technieken komt in het eerste deel van dit artikel aan bod aan de hand van het standaard kerncontract voor OTC-derivaten, de Master Agreement van de International Swaps and Derivatives Association (ISDA).

bw 67 quark 4.0.qxd

14-11-2005

06:14

Pagina 64

OTC-derivaten, beurs-effecten en risico’s

O & F

beeld Leeger en Renkens (2003), Leeger (2002) en Westrate (2001). Een uitvoerige Engelse gids bij de beide Master Agreements is Harding (2004). De 1992 Agreement kan overigens worden gedownload vanaf de ISDA-website, www.isda.org.

N r. 6 7 / o k t o b e r 2 0 0 5

64

De structuur van de Master Agreement is in wezen eenvoudig. Er is een kort artikel over de betaal- en leveringsverplichtingen van partijen, die uit hun afzonderlijke transacties voortvloeien; er zijn de gebruikelijke representations en dergelijke. Dan volgt de kern van de MA: ten eerste in Section 5 de Events of Default, die de wederpartij van de Defaulting Party het recht op beëindiging van alle transacties geven, en Termination Events, eveneens beëindigingsgronden, maar niet verwijtbaar aan een partij en beperkt tot Affected Transactions (die overigens alle transacties kunnen zijn).Ten tweede worden in Section 6 de gevolgen van uitoefening van een beëindigingsrecht geregeld. Deze bepalingen worden gevolgd door (standaard)voorwaarden die detailkwesties regelen, zoals de wijze waarop notices moeten worden uitgebracht. Section 14 ten slotte is belangrijk omdat daarin alle definities bij elkaar zijn gezet, waaronder enkele belangrijke contractsvoorwaarden.

Wie Master Agreement zegt, zegt ook Schedule.

Met deze appendix wordt de voorgedrukte MA afgestemd op de specifieke verhouding van partijen. De Schedule is daarom, ten eerste, de plaats waar partijen hun gegevens die relevant zijn voor (de afwikkeling van) hun transacties vastleggen. Interessanter zijn, ten tweede, de keuzen die partijen volgens de MA moeten maken. Op de derde plaats opent de Schedule uitdrukkelijk de mogelijkheid van amendering van definities uit Section 14. Echter, ook als de Schedule dat niet zo zegt, de contractsvrijheid staat partijen toe de voorgedrukte MA-tekst te wijzigen of bepalingen toe te voegen.

Niet ongebruikelijk is dat partijen ook een financiëlezekerheidsovereenkomst sluiten volgens een van de ISDA-modellen, bijvoorbeeld een Credit Support Annex. Daarmee is dan de bovenlaag van het OTCderivatencontract gereed – hoewel er nog geen enkele transactie (‘Transaction’) beschreven is en de bepalingen van deze bovenlaag ook te weinig specifiek zijn om een concrete transactie te beschrijven.

Vergeefs zal men bijvoorbeeld in de MA zoeken naar begrippen als Floating Rate Payer (voor een renteswap) of Exercise Period (voor een optie). In de MA 2 3

wordt ervan uitgegaan dat partijen transacties afsluiten en vervolgens betalingen en leveringen zullen (moeten) doen en dat daarbij het een en ander mis kan gaan, maar dat is eigenlijk alles.

Transactions worden vastgelegd in Confirmations die concrete gegevens bevatten, met name de economic terms: wie is de koper (Buyer) van de OTC-optie, wie is de verkoper/schrijver (Seller), gaat het om een Put of een Call, wat is de uitoefenperiode (Exercise Period), wat is de Premium, de Strike Price, enzovoorts. In een Confirmation worden ook product-specific definitions van toepassing verklaard die wél begrippen en bepalingen bevatten over specifieke derivaten. Naast de 2002 ISDA Equity Derivatives Definitions heeft ISDA ook andere definitieboekjes gepubliceerd, zoals de 2000 ISDA Definitions (voor de documentatie van rente- en valutaswaps) 2 en de 2003 Credit Derivatives Definitions. 3 Een volledig overzicht is op de ISDA-website te vinden.

De gehele contractsarchitectuur, inclusief (N) Transactions, kan er dan als volgt uitzien.

Kredietrisico en de single agreement

Dit is niet zomaar een structuur van afzonderlijke, samenhangende contracten; het is de bedoeling dat het geheel een single agreement vormt (s. 1(c)) en daardoor tot een reductie van het kredietrisico kan leiden. Een eenvoudig voorbeeld: tussen X en Y staan twee transacties uit: een swap die 20 in-the-money is voor X en een optie die X aan Y heeft verkocht en waaronder X dus (Y heeft de premie betaald) geen kredietrisico op Y loopt en die een waarde van 15 voor Y ofwel -15 voor X heeft. Het actuele kredietrisico dat X op Y loopt (‘exposure’) is, als de transacties afzonderlijk worden bezien, gelijk aan 20. Een single agreement impliceert echter dat beide transacties (en de MA met aanhang) één geheel vormen en dat hun waarden bij elkaar kunnen worden opgeteld. Het kredietrisico dat X op Y loopt, is dan 20 + (-15) = 5.

Vgl. Peeters (2001).

Zie De Vries Robbé (2003).

28

bw 67 quark 4.0.qxd

14-11-2005

06:14

Pagina 65

OTC-derivaten, beurs-effecten en risico’s

Bij derivaten zoals swaps moet een partij met een negatief exposure toch rekening houden met het risico dat dat exposure positief kan worden. Daarin verschilt een derivaat van, bijvoorbeeld, een lening.

Zo moeten banken onder solvabiliteitsregelgeving bij het actueel exposure een potentieel exposure optellen dat gelijk is aan een looptijdafhankelijk percentage van de notional amount. Indien een bank een renteswap van € 100 mln afsluit met een looptijd van 10 jaar en een initiële marktwaarde van nul, dan is toch sprake van een totaal exposure van € 1,5 mln, omdat in dit geval het percentage voor het potentieel exposure 1,5 is.4

Een stap terug: kredietrisico kan eenvoudig gereduceerd worden door geen transacties aan te gaan met partijen met een te lage kredietwaardigheid.

Lagere overheden zijn zelfs krachtens de wet verplicht alleen met wederpartijen met ten minste een A-rating of gelijkwaardige kredietwaardigheid te handelen. 5 De Staat gaat ‘vrijwillig’ nog verder en accepteert alleen wederpartijen met een rating van minstens AA-.6 Maar wat te doen bij een daling van de kredietwaardigheid van een wederpartij die aanvankelijk voldoende solide was en met wie gehandeld is? Geen nieuwe transacties aangaan is een deeloplossing, maar bij een groot bestaand positief exposure onbevredigend. De MA biedt geen algemene oplossing, maar een Additional Termination Event, dat aan de ‘voorgedrukte’Termination Events wordt toegevoegd, kan uitkomst bieden. Partijen kunnen in de Schedule overeenkomen dat indien de rating van een partij onder een bepaalde grens komt, de wederpartij van deze ‘Affected Party’ het recht heeft alle transacties te beëindigen.Wat te doen als er geen rating is voor die

4 5 6 7

Een collateralovereenkomst kan een effectieve manier zijn om kredietrisico te reduceren. Een positief exposure wordt in beginsel gedekt door collateral dat in de praktijk meestal bestaat uit cash en overheidsschuldpapier. Afwijkingen van dat beginsel zijn mogelijk, zowel naar boven (overcollateralisatie) als naar beneden. Dat laatste komt vaker voor doordat meestal gekozen wordt voor één of meer drempels die het exposure of de waarde van het te leveren collateral moet overschrijden voordat de out-of-themoney partij daadwerkelijk dient te leveren.

Naast of in plaats van collateral kan een garantie van een zogenoemde Credit Support Provider van de wederpartij zekerheid bieden. Aan de dealerkant ziet men dat wel wanneer deze een special derivatives vehicle is dat onderdeel is van een bankengroep. Een ander, naar men hoopt, goed gekapitaliseerd lid van die groep staat dan garant.

Settlement risico’s

Het settlement-risico van sommige betalingen wordt beperkt door s. 2(c). Als onder een Transaction op dezelfde dag beide partijen betalingen in dezelfde valuta moeten doen, worden die verplichtingen vervangen door betaling van het saldo door de partij die het grootste bedrag moet betalen. Bij swaps stuurt men er vaak op aan dat de periodieke betalingen op

DNB, Handboek Wtk, 4011 – 03.3.2.

Artikel 2 Regeling uitzetting en derivaten decentrale overheden (Stcrt. 2000, 251), zoals nadien gewijzigd.

Agentschap (2003), p. 31.

2000 ISDA Definitions, s. 15.1; vgl. Broekhuizen en Peeters (2000).

29

65 N r. 6 7 / o k t o b e r 2 0 0 5

Kredietwaardigheid

Collateral en andere zekerheden

O & F

Er is een tweede effect van de single agreement dat voor de beheersing van kredietrisico van belang is.

Indien een partij niet (volledig) voldoet aan een betalings- of leveringsverplichting onder een Transaction en zo een aanwijzing geeft van verhoogd kredietrisico, geldt dat als een wanprestatie onder het gehele contract en heeft de wederpartij het recht alle Transactions te beëindigen: zie het Event of Default Failure to Pay or Deliver van s. 5(a)(i) en s. 6(a).

partij? Met enige creativiteit en inzicht is het begrip kredietwaardigheid te objectiveren, bijvoorbeeld door het te koppelen aan financiële ratio’s en andere variabelen. Soms gaat een partij akkoord met een, naar het oordeel van de wederpartij, materially adverse change van haar financiële toestand als Additional Termination Event. De afhankelijkheid van dat grotendeels subjectieve oordeel introduceert echter onzekerheid (voor beide partijen) en dat gaat tot op zekere hoogte in tegen het primaire doel van het afsluiten van de MA. Een andere oplossing is om bij afzonderlijke transacties Optional Early Termination af te spreken. Zo kan men bepalen dat een swap met een looptijd van tien jaar na vijf jaar jaarlijks op een aantal data kan worden beëindigd met afrekening van de marktwaarde van de swap. 7 Desgewenst kan zo een deel van de relatie met de wederpartij beëindigd worden.

bw 67 quark 4.0.qxd

14-11-2005

06:14

Pagina 66

OTC-derivaten, beurs-effecten en risico’s

dezelfde dag opeisbaar zijn. Dat vergt overigens een zorgvuldige formulering van de Confirmation als de legs van de swap door verschillende definitions worden beheerst, zoals de total return swap (zie hierna onder ‘Equity Derivatives’). Het Herstatt-risico van quasisimultane betalingen in verschillende valuta is zo niet uit de wereld geholpen, maar opname van een escrow-bepaling in de Schedule kan uitkomst bieden.

Ten slotte, bij leveringen van effecten tegenover betalingen, dient waar mogelijk DvP, delivery-versus-payment, te worden toegepast, aldus ook s. 9.10 van de 2002 ISDA Equity Derivatives Definitions.

O & F

Default Under Specified Transaction

N r. 6 7 / o k t o b e r 2 0 0 5

66

Specified Transactions zijn derivatentransacties tussen partijen en hun eventuele Specified Entities en Credit Support Providers, die niet onder de Master Agreement tussen partijen vallen. Een Specified Entity van een partij moet in de Schedule worden opgegeven en is meestal een onderneming uit dezelfde groep. Een bank kan bijvoorbeeld eisen dat de moederonderneming van haar wederpartij X, MX, met wie de bank ook derivatentransacties afsluit een Specified Entity is. Het resultaat is dat een wanprestatie van MX onder een swap met de bank de laatste in staat stelt op grond van een Event of Default alle transacties met X te beëindigen. Het idee is duidelijk: als één onderneming in een groep in gebreke blijft, zit het waarschijnlijk ook niet goed met de kredietwaardigheid van andere groepsleden.

Partijen zijn vaak bevreesd om een eigen Specified Entity op te geven.Toch ziet men vaak een dealer dat wel, en geheel vrijwillig, doen. De reden is dat een eigen Specified Entity juist tot een beperking van kredietrisico op de wederpartij kan leiden. Neem twee derivatenvehikels X en X* van een bank, die transacties met Y doen. X doet rentederivaten, X* equity derivatives. Beide hebben een Master Agreement met Y. X* oefent een aandelenoptie uit, maar Y betaalt niet ondanks aanmaning, een Event of Default onder de Agreement met X*. Enkele dagen later dient X een betaling aan Y te doen. Als X* in de Agreement tussen X en Y niet is opgegeven als Specified Entity van X voor Default Under Specified Transaction, zal X die betaling moeten verrichten. Is X* wel zo’n Specified Entity, dan kan X betaling aan Y vermijden met een beroep op die Event of Default.

De eigen Specified Entities zijn als het ware de voelsprieten waarmee de financiële positie van de wederpartij gedeeltelijk wordt afgetast.

8

Cross Default

Gedeeltelijk wordt afgetast omdat Specified Entities slechts bij een beperkt aantal financiële transacties met de wederpartij betrokken zullen zijn. De MA biedt nog een mogelijkheid om geheel andere transacties van de wederpartij onder de Events of Default van de MA te brengen: Cross Default van s. 5(a)(vi).

Cross Default ziet op transacties van een partij met wie dan ook die betrekking hebben op Specified Indebtedness, gedefinieerd als verplichtingen met betrekking tot geleend geld. Deze definitie kan echter worden uitgebreid, bijvoorbeeld met derivatentransacties met derden8 of andere soorten transacties van enige omvang die de wederpartij afsluit.Wel verhindert een fikse Threshold Amount meestal om elke wanprestatie van de wederpartij onder Specified Indebtedness aan te grijpen om een beroep te doen op Cross Default. Daar staat tegenover dat niet alleen de omvang van de wanprestatie meetelt, maar ook de omvang van alle verplichtingen die daardoor opeisbaar worden. Is de Threshold Amount bijvoorbeeld € 10 miljoen, dan treedt een Cross Default in als de wederpartij bijvoorbeeld een rentebetaling van € 100.000 mist onder een lening met een hoofdsom groter dan € 10 miljoen en de leningvoorwaarden de hoofdsom dan vervroegd opeisbaar maken.

Section 2(a)(iii); No Adequate Assurances

S. 2(a)(iii) geeft de ‘onschuldige’ partij een opschortingsrecht bij een Event of Default, maar ook al bij een Potential Event of Default,‘an event which, with the giving of notice or the lapse of time or both, would constitute an Event of Default’. Een eigen betaling kan worden opgeschort wanneer de wederpartij niet betaalt op de afgesproken dag, ook al moet dan nog aanmaning plaatsvinden. Ook zoiets simpels als het niet tijdig opsturen van een jaarverslag – vaak een verplichting opgenomen in Part 3(b) van de Schedule – mag tot opschorting leiden; zie s. 5(a)(ii).

Als er een Event of Default is, kan de Non-defaulting Party bovendien op grond van s. 2(a)(iii) (be)denktijd nemen of zij wel of niet gebruik zal maken van haar recht om de transacties te beëindigen.

Maar wat kan een partij X doen als zij informatie over de wederpartij Y verkrijgt die erop duidt dat deze in ernstige financiële problemen verkeert, terwijl X over enkele dagen een grote betaling aan Y moet doen en de eerstvolgende betaling door Y pas over drie maanden gepland staat? De informatie over Y vormt geen (Potential) Event of Default zodat het

GDSC (2003),Annex B.

30

bw 67 quark 4.0.qxd

14-11-2005

06:14

Pagina 67

OTC-derivaten, beurs-effecten en risico’s

opschortingsrecht niet aanwezig lijkt. Moet X een betaling doen die een grote kans loopt in de aanstaande faillissementsboedel van Y te verdwijnen? In een dergelijke situatie kan de doctrine van adequate assurances te hulp komen. De modelbepaling van GDSC (2003) geeft aan op welke wijze: ‘A party fails to provide adequate assurances of its ability to perform its outstanding obligations to the other party (‘X’) under this Agreement on or before the second business day after a written request for such assurances is made by X when X has reasonable grounds for insecurity.’9

Een Event of Default geeft de ‘onschuldige’ partij, de Non-defaulting Party, het recht een Early Termination Date aan te wijzen waarop alle Transactions vervroegd beëindigd worden. Het eerste rechtsgevolg is dat alle toekomstige verplichtingen van partijen vervallen en worden vervangen door één betalingsverplichting van een der partijen. Het exposure wordt gefixeerd op de Early Termination Date. (Hetzelfde geldt bij beëindiging op grond van een creditrelated Additional Termination Event, zoals omschreven onder ‘Kredietwaardigheid’. Hierna ga ik voor de eenvoud uit van beëindiging na een Event of Default.)

Het tweede belangrijke rechtsgevolg is dat die ene betalingsverplichting in beginsel ziet op de totale waarde van de Transactions op de Early Termination Date. Hoewel er drie verschillende methoden ter bepaling van dat bedrag zijn – Market Quotation en Loss in de 1992 MA en Close-out Amount in de 2002-versie – gaan zij er alle van uit dat de Non-de-

9 10 11 12 13

Bij early termination zal de onschuldige partij toch vaak moeten betalen. Een simpel voorbeeld: X en Y zijn maar één transactie aangegaan, een call-optie. X betaalt de premie niet en aanmaning heeft geen effect. Er is nu een Failure to Pay or Deliver en Y kan de transactie beëindigen. De koers van het onderliggende aandeel en de waarde van de optie zelf kunnen echter inmiddels flink zijn gestegen, zodat de optie meer waard is dan de premie (plus rente). Zou Y gebruikmaken van zijn beëindigingsrecht dan is hij het saldo van die hogere waarde en de premie met rente aan X verschuldigd en dat terwijl alles begon met wanprestatie van X. Is het voor de Non-defaulting Party die out-ofthe-money is, beter om een beroep op zijn opschortingsrecht te doen bij een Event of Default in plaats van zijn beëindigingsrecht uit te oefenen? Cause célèbre is inmiddels Enron Australia v TXU Electricity,13 waarin TXU, na het Event of Default ‘Bankruptcy’

GDSC (2003), Annex A. Vgl. ISDA (2003a), p. 67-68.

Ik gebruik de term voor het gemak ook in het kader van de 1992 MA. Als onder de 1992 MA, zoals gebruikelijk, de Second Method van toepassing is: s. 6(e)(i).

[2003] NSWSC 1169 (24 December 2003).Vgl. o.a. Farrell (2005) en Weinstein et al. (2004).

31

67 N r. 6 7 / o k t o b e r 2 0 0 5

Beëindiging en afrekening

Afgezien van de Unpaid Amounts zijn er in de praktijk afwijkingen tussen de ETA en het exposure.

Onder Market Quotation is de ETA het gemiddelde van een aantal quotes van dealers voor de transacties.

Zo zal het echter exposure niet berekend worden in de dagelijkse gang van zaken. Onder Loss moet een partij een redelijke waarde voor de transacties berekenen en mag hij, zonder dubbeltelling, ook winsten en verliezen door funding en hedging meenemen.

Ook dat is niet de gebruikelijke manier om (dagelijks) de mark-to-market waarden van transacties te berekenen. Close-out Amount is een soort combinatie van Market Quotation en Loss. Kortom, er is sprake van enig wat wordt genoemd basis risk tussen het exposure en de ETA.

O & F

Is deze bepaling in de Schedule als extra Event of Default opgenomen, dan is er, zodra X het verzoek heeft gedaan, sprake van een Potential Event of Default ten aanzien van Y en kan X betalingen opschorten. Overigens geldt hier eenzelfde bezwaar als tegen de materially adverse change clausule: onzekerheid omdat dit Event of Default afhankelijk is van het redelijke, maar hoe dan ook subjectieve oordeel van een partij. ISDA heeft de bepaling kennelijk bewust niet willen opnemen, zelfs niet als keuzemogelijkheid, in de 2002 Agreement.10

faulting Party het nettobedrag bepaalt dat de vervroegde beëindiging hem kost of juist oplevert. Dat bedrag, vermeerderd met het saldo van de Unpaid Amounts (bedragen die op de Early Termination Date verschuldigd waren, maar nog niet betaald) is het compensatiebedrag, Early Termination Amount geheten in de 2002 MA (‘ETA’).11 Als de Non-defaulting Party in-the-money blijkt te zijn, dan zal de wederpartij dat bedrag moeten betalen, maar als hij out-of-the-money is, dient hij dat bedrag aan de Defaulting Party te betalen.12

bw 67 quark 4.0.qxd

14-11-2005

06:14

Pagina 68

O & F

OTC-derivaten, beurs-effecten en risico’s

ten aanzien van Enron, met een beroep op s. 2(a)(iii) zijn verplichtingen onder 78 transacties opschortte en door niet voor early termination te kiezen betaling van A$ 3,3 miljoen aan Enron wist te vermijden. De zaak is onderwerp van uitvoerige discussie en de conclusie lijkt te zijn dat TXU inderdaad het recht heeft deze strategie te volgen – en te blijven volgen (want TXU is niet van plan ooit early termination in te roepen). Er is wel gesuggereerd dat als de insolventie van Enron zou worden opgeheven, TXU zou moeten gaan betalen, maar dat lijkt onjuist. Bankruptcy is in de MA zo geformuleerd dat het niet door zo’n latere opheffing ongedaan wordt gemaakt. Ook in het optievoorbeeld geldt dat: Failure to Pay or Deliver houdt in dat Buyer niet binnen één (2002 MA) of drie (1992 MA) Local Business Day(s) na aanmaning betaalt. Of hij nu na vijf dagen of drie jaar alsnog betaalt, doet er niet toe: hij heeft niet tijdig betaald en dat Event of Default is niet ongedaan te maken.

N r. 6 7 / o k t o b e r 2 0 0 5

68 In feite realiseert zich het kredietrisico van de Defaulting Party op de Non-defaulting Party wanneer deze laatste een permanent beroep doet op zijn opschortingsrecht. De Defaulting Party blijft zitten met een oninbaar bedrag. Een algemene oplossing vergt amendering van de Master Agreement.

De 2002 ISDA Equity Derivatives Definitions (de ‘Definitions’) Een definitie van definition

Wat zijn nu die product-specific definitions, waarvan de Definitions een voorbeeld vormen? Het gaat zeker niet altijd om definities in de gebruikelijke betekenis.

Op de eerste plaats zijn er formele definities die inhoudelijk nietszeggend zijn. Een voorbeeld, met vele aan te vullen, is de definitie van Trade Date in s.

1.17: ‘ “Trade Date” means, in respect of a Transaction, the date specified in the related Confirmation.’

Zulke definities krijgen pas inhoud na invulling in een concrete Confirmation. Partijen nemen meestal, conform het jargon in derivatenland, als Trade Date de datum waarop zij de Transaction zijn aangegaan.14 Op de tweede plaats zijn er inhoudelijke definities: een omschrijving van een begrip dat in die betekenis in de gehele Definitions en in Confirmations gebruikt wordt, althans zou moeten worden. Aldus definieert s. 1.31 ‘Scheduled Trading Day’ als ‘any day

on which each Exchange and each Related Exchange are scheduled to be open for trading for their respective regular trading sessions.’ Ten slotte zijn er die definities die helemaal geen definities zijn, maar materiële contractsvoorwaarden, die aan partijen rechten of verplichtingen toekennen. Een eenvoudig voorbeeld is s. 2.4(a): ‘In respect of an Option Transaction, Buyer shall pay Seller the Premium on the Premium Payment Date.’ Bij dit derde type komt het, zoals in het vervolg zal blijken, vaak voor dat partijen kunnen kiezen uit een menu van alternatieve ‘definities’ om een bepaald aspect van hun rechtsverhouding te regelen. Soms geven de Definitions een fallback als partijen geen keuze maken.

Juist omdat de Definitions bepalingen van het derde type bevatten, fungeren zij als de voornaamste bouwstenen voor een (aanvullend) 15 derivatencontract. De Confirmations zelf kunnen betrekkelijk kort zijn en behoeven niet allerlei uitgeschreven bepalingen over rechten en verplichtingen van partijen te bevatten. Ter illustratie verwijs ik naar de model-Confirmations die bestemd zijn voor gebruik met de Definitions.16 Bereik van de Definitions

De Definitions zijn bestemd voor de documentatie van drie derivatentypen: opties, forwards en swaps.

Er zijn drie typen onderliggende waarden: een enkele Share, een enkele Index en een Basket (een mandje van hetzij Shares hetzij Indices). Shares, gedefinieerd in s. 1.14, zijn niet alleen aandelen maar kunnen ook andere securities zijn. Niet alle bepalingen van de Definitions werken echter zonder meer met andere effecten en daarom ga ik hierna van aandelen uit en bovendien, in het voetspoor van de Definitions, van aandelen die verhandeld worden op een Exchange.

Een Exchange is: ‘in respect of a Share [...] each exchange or quotation system specified as such for such Share in the related Confirmation, any successor to such exchange or quotation system or any substitute exchange or quotation system to which trading in the Share has temporarily relocated (provided that the Calculation Agent has determined that there is comparable liquidity relative to such Share on such temporary substitute exchange or quotation system as on the original exchange). 17 Analoog is Related Exchange gedefinieerd in s. 1.26: dit is een beurs waar

14 S. 3.7 van de 2000 ISDA Definitions definieert de Trade Date inhoudelijk als ‘the date on which the parties enter into the Swap Transaction’.

15 Want de Confirmation (met inbegrip van de Definitions) is weer onderdeel van de single agreement.

16 ISDA (2002b),‘downloadbaar’.

32

bw 67 quark 4.0.qxd

14-11-2005

06:14

Pagina 69

OTC-derivaten, beurs-effecten en risico’s

derivaten met betrekking tot het Share worden verhandeld.

De Definitions beschrijven voor elk type Transaction zowel volledig geldelijke afwikkeling (Cash Settlement) als afwikkeling (mede) door levering van aandelen (Physical Settlement), met uitzondering van Transactions met een Index of een Basket van Indices als underlying. Deze zijn altijd Cash-settled. Ook maken de Definitions het mogelijk om aan een partij het recht toe te kennen pas tijdens de looptijd een keuze te maken: de Settlement Method Election van s.7.1. Het recht moet uiterlijk op de prozaïsch, maar accuraat betitelde Settlement Method Election Date (s. 7.2) worden uitgeoefend.

X bezit 1 miljoen aandelen ABC met een actuele koers van € 12,50. De gemaakte koerswinst wil X veilig stellen, maar zonder de aandelen te verkopen, onder meer omdat hij op termijn, zeg over een jaar, weer ‘mogelijkheden’ ziet. Hiertoe gaat X een swap aan met Y met als ‘Equity Notional Amount’ de beginwaarde van de aandelen, € 12,5 miljoen, en een looptijd van een jaar. X zal elke drie maanden de koerswinst op 1 mil-

Hoewel hedging hierna slechts zijdelings aan de orde komt, speelt het bij equity derivatives een grote rol. In het voorbeeld van zojuist heeft X een perfecte hedge.Y, die een synthetische long positie in het aandeel heeft, heeft een aantal mogelijkheden om die positie af te dekken. Een mogelijkheid is short te gaan in de onderliggende aandelen door deze te lenen tot en met de einddatum van de swap en de geleende aandelen meteen te verkopen.20 Hoewel al dan niet hedgen op het eerste gezicht wellicht slechts een zaak van de betreffende partij lijkt, geven de Definitions toch een aantal mogelijkheden om problemen bij het hedgen (bijvoorbeeld verhoogde kosten), tot onderdeel van de Transaction te maken. Hebben partijen dat gedaan, dan kunnen die problemen leiden tot vervroegde beëindiging van de Transaction.21

joen aandelen ABC plus eventuele dividenden als ‘Equity Amount Payer’ aan Y betalen. Daartegenover betaalt Y op

Waardering

dezelfde data een eventueel koersverlies aan X én in elk geval

Met name bij Cash-settled Transactions draait alles bij de afwikkeling om de waarde van het/de onderliggende Share, Index of Basket op de relevante beurs of beurzen. Ik behandel hierna hoofdzakelijk Shares omdat de waardering van een Index en een Basket analoog verloopt. Het uitgangspunt is eenvoudig. De waarde wordt vastgesteld op een of meer Valuation Dates (s. 6.2). In het geval van een optie is elke Exercise Date 22 een Valuation Date. Voor forwards en swaps wijzen partijen Valuation Date(s) aan in de

een variabele driemaands rente over de Equity Notional Amount.19 Het resultaat is dat X een jaar lang geen risico meer loopt op ABC en rente ontvangt over zijn investering in ABC en dat Y juist een ‘synthetische’ positie in die aandelen heeft ingenomen.

Waarvoor kunnen dergelijke swaps dienen? Deze aandelenswap stelt X in staat om zijn winst op het

17 S. 1.25(b), hieronder nader te bespreken.

18 Vgl.Van ’t Westeinde (1999).

19 Voor de omschrijving van deze ‘floating leg’, moeten overigens niet de Definitions, maar de 2000 ISDA Definitions (onder meer bestemd voor fixed-for-floating swaps) gebruikt worden.

20 Vgl. ss. 12.9(a)(vii)-(viii).

21 Zie ss. 12.9(b)(iii)-(vi).

22 ‘ “Exercise Date” means […] the Scheduled Trading Date during the Exercise Period on which such Option is, or is deemed 33 to be exercised.’; s. 3.1(a). .

69 N r. 6 7 / o k t o b e r 2 0 0 5

OTC-opties en forwards op aandelen of indices kunnen voor gebruikelijke maar ook meer speciale doeleinden worden aangewend, zoals het hedgen van, of zekerstellen van winst op, exchangeable bonds of obligaties met een opbrengst die afhankelijk is van een index (of mandje van indices). Een ander voorbeeld is het hedgen van personeelsopties. Van een typische equity swap, de total return swap,18 geef ik een voorbeeld.

O & F

Equity Derivatives

aandeel voorlopig veilig te stellen, stemrechten te blijven uitoefenen en zonder transactiekosten verbonden aan de verkoop (en een latere heraankoop) van ABC toch geen enkel koersrisico meer te lopen.

Wederpartij Y zou een beleggingsfonds kunnen zijn, dat zo zonder funding in ABC kan beleggen (misschien in afwachting van een instroom van liquide middelen van beleggers waarna het fonds daadwerkelijk die aandelen ABC zal kopen of juist structureel synthetisch, zoals een hedge fund). Een andere equity swap is die waarin beide legs equity-related zijn, bijvoorbeeld de swap van de totale opbrengst van een mandje van aandelen tegen de opbrengst van een index. Zo kan degene die die aandelen bezit als betaler van de opbrengst van het mandje en ontvanger van de indexopbrengst zijn portefeuille synthetisch diversificeren.

bw 67 quark 4.0.qxd

14-11-2005

06:14

Pagina 70

OTC-derivaten, beurs-effecten en risico’s

O & F

Confirmation. (Een Valuation Date moet uiteraard een beursdag zijn zodat een aangewezen Valuation Date die geen Scheduled Trading Day is vervangen wordt door de eerstvolgende Scheduled Trading Day.) Omdat koersen gedurende de beursdag veranderen, dienen partijen ook een exact tijdstip te kiezen op de Valuation Date, waarop de waardering dient plaats te vinden: de Valuation Time (s. 6.1). De fallback is de reguliere sluitingstijd, de Scheduled Closing Time, van de Exchange.Verloopt alles normaal, dan bepaalt de Calculation Agent (zie hierna) de koers van het Share op de Valuation Time op de Valuation Date.

N r. 6 7 / o k t o b e r 2 0 0 5

70

Twee verstoringen worden door de Definitions impliciet opgevangen.Ten eerste, als de beurs op een Valuation Date sluit vóór de Scheduled Closing Time en de Valuation Time daardoor na de feitelijke sluitingstijd komt te liggen, dan wordt de Valuation Time gelijk aan de feitelijke sluitingstijd en kan waardering doorgang vinden, tenzij er sprake is van Market Disruption (zie hierna). Een tweede, ernstiger verstoring wordt opgevangen door het laatste deel van de definitie van Exchange en Related Exchange: de tijdelijke verplaatsing van de beurshandel (zie onder ‘Bereik van de Definitions’). Dit deel weerspiegelt de ervaringen in de periode direct na 9/11: een deel van de handel van de American Stock Exchange werd tijdelijk naar de NYSE overgebracht.23 De Calculation Agent

In vrijwel elke Confirmation van een OTC-derivatentransactie wordt een Calculation Agent aangewezen. In een transactie tussen een eindgebruiker en een bank of andere dealer is de laatste vrijwel steeds de Calculation Agent. Diens taak is meestal inderdaad beperkt tot het maken van berekeningen en het vaststellen van gegevens voor die berekeningen. Bij een renteswap stelt de Calculation Agent periodiek de nieuwe waarde van de variabele rente vast en berekent hij welke bedragen partijen aan elkaar verschuldigd zijn. Aanleiding tot discussies geven dergelijke beslissingen zelden omdat in de Confirmation (indirect) tot op de minuut is bepaald wanneer een bepaald type rente moet worden vastgesteld.Voor de hiervoor beschreven vaststelling van een aandelenkoers geldt hetzelfde.

Niettemin: ‘in many provisions throughout the 2002 Definitions, the Calculation Agent has broad discretion’. 24 Eén oorzaak is dat de Definitions, omdat zij bestemd zijn voor de wereldwijde OTChandel, slechts algemene regels kunnen geven die door de Calculation Agent nader moeten worden ingevuld.Verder gaat het onder meer om de reactie op een groot aantal mogelijke verstoringen van de normale gang van zaken waarvoor moeilijk specifieke regels gegeven kunnen worden. Toch zijn er soms alternatieve regels die partijen van toepassing kunnen verklaren die die uitkomst niet of minder afhankelijk maken van de beslissing van de Calculation Agent.

Voor het overige dient deze ‘in good faith and in a commercially reasonable manner’ te beslissen (s.

1.40). ISDA-documentatie bevat overigens geen formeel mechanisme om een dispuut tussen de Calculation Agent en een partij (in feite: diens wederpartij) op te lossen. Eindgebruikers die voldoende onderhandelingsmacht hebben – grote hedge funds, pensioenfondsen en andere institutionele beleggers – kunnen proberen een dergelijk mechanisme met de Calculation Agent overeen te komen.

Verstorende effecten van de beurs Disrupted Days

Allereerst zijn er verstorende effecten van de Exchange of de Related Exchange, die de normale wijze van waardering van een Share (zie onder ‘Waardering’) onmogelijk maken. De Definitions nemen een viertal gebeurtenissen samen onder het begrip Disrupted Day van s. 6.4: 1 de Exchange of Related Exchange blijft dicht op een Scheduled Trading Day; 2 Trading Disruption: schorsing of beperking van de handel in het Share op de Exchange of derivaten op dat Share op de Related Exchange in de periode van een uur voorafgaand aan de Valuation Time; 3 Exchange Disruption:‘any event […] that disrupts or impairs (as determined by the Calculation Agent) the ability of market participants in general’ om op de Exchange in de Shares te handelen of vergelijkbare problemen op de Related Exchange, eveneens in het uur voorafgaand aan de Valuation Time; 4 Early Closure: de Exchange of Related Exchange sluit vóór de Scheduled Closing Time, tenzij die

23 Rae (2003), p. 9.

24 ISDA (2003b), p. 9.

34

bw 67 quark 4.0.qxd

14-11-2005

06:14

Pagina 71

OTC-derivaten, beurs-effecten en risico’s

sluiting, kort gezegd, tijdig vóór de Valuation Time is aangekondigd.25

Een Index valt onder het Disrupted Day-regime indien er sprake is van een Disrupted Day voor effecten die ten minste 20% van het niveau van de Index bepalen.28 De Valuation Date verschuift naar de volgende Scheduled Trading Day tenzij dat ook weer een Disrupted Day voor de Index is. Dat kan zo doorgaan tot de achtste Disrupted Day na de oorspronkelijke Valuation Date. Op die dag stelt de Calculation Agent de waarde van de Index vast met een ‘good faith estmate’ voor de effecten ten aanzien waarvan er een Disrupted Day is.

Aanpassingen van Share Transactions

Naast deze marktverstoringen staan verstoringen van de voorwaarden van de Transaction zelf door handelingen van de uitgever, de Issuer, van het Share die, ‘in the determination of the Calculation Agent’, 29 een verwaterend dan wel concentrerend effect op de theoretische waarde van de Shares hebben. Voorbeelden van deze corporate actions 30 zijn opsplitsing dan wel samenvoeging van de Shares, bonusuitgiften, een buitengewoon dividend en inkoop van de eigen Shares; vergelijk de opsomming van Potential Adjustment Events in s. 11.2(e). Hier komt de ISDA-definitietechniek goed uit. Met drie woorden kunnen partijen in de Confirmation aangeven wat de Calculation Agent te doen staat: - is ‘Options Exchange Adjustment’ (s. 11.2(b)) gespecificeerd, waarbij partijen een Options Exchange kunnen opgeven (fallback: de Related Exchange), dan dient de Calculation Agent de aanpassingen van opties op het Share door de

25 S. 6.3. Naar de letter van de desbetreffende bepaling is er Early Closure als de aankondiging van de vervroegde sluiting na de Valuation Time plaatsvindt. Dat zal niet de bedoeling zijn.

26 S. 6.6(a)(ii)(B).

27 FMLC (2003), p. 10 met enige twijfel ten aanzien van 11 september 2001.

28 Vgl. Euronext (2005).

29 S. 11.2(a).

30 Vgl. Euronext.liffe (2004).

35

71 N r. 6 7 / o k t o b e r 2 0 0 5

Wat hier ook van zij, als een Valuation Date een Disruption Day is, schuift de Valuation Date op naar de eerstvolgende Scheduled Trading Day die geen Disrupted Day is. In het onwaarschijnlijke geval dat de volgende acht Scheduled Trading Days alle Disrupted Days zijn, is het niettemin die achtste dag waarop de ‘Calculation Agent shall determine […] its good faith estimate of the value for the Share as of the Valuation Time on that eighth Scheduled Trading Day.’26 (Bij wijze van illustratie: de Amerikaanse aandelen- en optiebeurzen bleven op 9/11 en daarna vier dagen dicht.27) Dat lijkt onbegonnen werk, maar een aantal factoren kan de Calculation Agent te hulp komen. Het kan, zoals gezegd, vooral de Related Exchange zijn die voor deze Disrupted Days heeft gezorgd. De Exchange kan op de verplichte Valuation Day open zijn en dat maakt het minder moeilijk voor de Calculation Agent om tot zijn ‘good faith estimate’ te komen. Bovendien vragen de Definitions niet om een pseudo-beurskoers, maar om een schatting van de waarde van het Share. De Calculation Agent kan daarbij rekening houden met transacties in de OTC-handel, die bij zo’n langdurige verstoring van de beurshandel ongetwijfeld in belang zal toenemen. Ten slotte betekent Disrupted Day op de Exchange niet dat er in het geheel niet gehandeld wordt in het Share.Als de Valuation Time 16.00 uur is en de beurs negen dagen lang gewoon open gaat

Overigens kan een optie wél worden uitgeoefend op een Disrupted Day; alleen zal bij Cash-settled opties de vaststelling van het bedrag dat Seller aan Buyer verschuldigd is, moeten wachten op de uitgestelde Valuation Day. De invloed van een Disrupted Day op een optie met Physical Settlement komt hierna onder ‘Physical Settlement en Disrupted Days’ aan de orde.

O & F

Bij Trading Disruption en Exchange Disruption is het oordeel van de Calculation Agent vereist dat de verstoring material is. Dat oordeel is echter niet nodig bij het gesloten blijven van de Exchange of Related Exchange en bij Early Closure. Het is echter mogelijk dat die gebeurtenissen niet steeds material zijn. Zo kan vervroegde sluiting van de Related Exchange, terwijl de Exchange een gewone Scheduled Trading Day heeft, betekenen dat de handel in aandelen niet substantieel verstoord is. Meer in het algemeen geldt dat een Disrupted Day geheel kan zijn toe te schrijven aan gebeurtenissen op de Related Exchange, terwijl de Exchange (ogenschijnlijk) normaal functioneert.

maar steeds rond 15.30 uur wordt geteisterd door computerproblemen (Exchange Disruption), zijn er toch koersen tot stand gekomen die een leidraad voor de schatting kunnen vormen.

bw 67 quark 4.0.qxd

14-11-2005

06:14

Pagina 72

O & F

OTC-derivaten, beurs-effecten en risico’s

N r. 6 7 / o k t o b e r 2 0 0 5

72

Option Exchange te volgen en de contractstermen van de Transaction dienovereenkomstig aan te passen – ook als deze niet een optie maar een forward of swap is. 31 De Calculation Agent bepaalt zelf welke contractstermen moeten worden aangepast. De wijzigingen treden op dezelfde dag in werking als de door de Options Exchange aangebrachte wijzigingen. In feite worden zo het rule book van de Options Exchange en de bijbehorende precedenten in de Transaction geïncorporeerd; - hebben partijen daarentegen voor ‘Calculation Agent Adjustment’ gekozen (s.11.2(c)), dan bepaalt de Calculation Agent allereerst of een Potential Adjustment Event inderdaad een verwaterend of concentrerend effect heeft en, zo ja, welke aanpassingen ‘appropriate to account for that diluting or concentrative effect’ zijn. Hierbij mag de Calculation Agent rekening houden met aanpassingen van beursopties op het Share, maar dat hoeft hij niet. Ook de datum van inwerkingtreding van de aanpassingen wordt door de Calculation Agent bepaald.

Merger Events

De Definitions kennen een aantal Extraordinary Events – externe effecten – waarvan er één door fusies en verwante gebeurtenissen gevormd wordt: Merger Event, dat in s.12.1(b) in een viertal typen is opgesplitst waarvan elk weer varianten heeft.

Gemeenschappelijk kenmerk is dat het Share als zodanig verdwijnt (behalve bij Reverse Merger) en dat Transactions met dat Share als onderliggende waarde dus niet zonder aanpassingen kunnen voortbestaan.Ter illustratie neem ik het gevalstype van een overname van de Issuer betaald met aandelen van de overnemer (‘New Shares’, s. 12.1(i)).Voor dit Sharefor-Share Merger Event moeten partijen uit vijf alternatieven kiezen (s. 12.2): - Twee keuzemogelijkheden zijn al behandeld: Options Exchange Adjustment en Calculation Agent Adjustment (met een kwalificatie van de laatste: als de Calculation Agent tot de conclusie komt dat geen enkele aanpassing commercially reasonable is, wordt de transactie beëindigd en is ‘Cancellation and Payment’ van toepassing, zie hierna); - ‘Modified Calculation Agent Adjustment’ verschilt in zoverre van Calculation Agent Adjustment dat de Calculation Agent in dit geval wel ‘adjustments to account for changes in volatility, expected divi-

dends, stock loan rate or liquidity relevant to the Shares or to the Transaction’ mag maken; - ‘Alternative Obligation’ houdt in wezen in dat de Transaction wordt voortgezet met de New Shares als onderliggende waarde. De economic terms van de Transaction worden gewijzigd aan de hand van de ruilverhouding tussen de beide aandelen en de Calculation Agent maakt indien nodig andere aanpassingen, met uitzondering van de aanpassingen die in de vorige alinea zijn geciteerd; - ten slotte is er ‘Cancellation and Payment’: de transactie wordt automatisch beëindigd op de closing date van de Merger Event. De bepaling van het compensatiebedrag bij beëindiging van de Transaction wordt onder ‘Cancellation Amount; compensatie voor Option Transactions’ besproken.

Delisting

Min of meer rampzalig voor het voortbestaan van een derivaat op een beursgenoteerd aandeel is Delisting: ‘the Exchange announces that pursuant to the rules of such Exchange the Shares cease (or will cease) to be listed, traded or publicly quoted on the Exchange for any reason (other than a Merger Event [...])’; zie s. 12.6(a)(iii). Niets anders dan beëindiging van de Transaction lijkt in zo’n geval uitkomst te bieden. Toch is dat niet noodzakelijkerwijs waarop de Definitions uitkomen. Uit twee alternatieven kan worden gekozen: - ‘Negotiated Close-out’ (s. 12.6(c)(i)): partijen proberen beëindigingsvoorwaarden, met name natuurlijk een bedrag ter compensatie van de in-themoney partij, overeen te komen. Als dat niet lukt, wordt de Transaction voortgezet ‘on the terms and subject to the conditions then in effect’, zij het dat elke partij een eventueel geldend Physical Settlement mag omzetten in Cash Settlement: het verwerven van Shares ter levering aan de wederpartij zal immers op zijn minst lastiger worden. Bovendien hoeft de Calculation Agent geen rekening meer te houden met Disrupted Days; de te waarderen Share toch niet meer op de disrupted Exchange genoteerd zijn. Het eigenaardige resultaat kan dus zijn dat als een swap op een delisted Share nog drie jaar te gaan heeft, de Calculation Agent drie jaar lang op, zeg, de eerste Scheduled Trading Day van elk kwartaal zijn ‘good faith estimate’ van de waarde van het aandeel om 16.00 uur moet geven; - ‘Cancellation and Payment’ (s. 12.6(c)(ii)) ver-

31 Euronext.liffe (2004) behandelt ook aanpassingen van futures (‘beursgenoteerde forwards’). Met die mogelijkheid houden de Definitions opmerkelijk genoeg in dit verband geen rekening.

36

bw 67 quark 4.0.qxd

14-11-2005

06:14

Pagina 73

OTC-derivaten, beurs-effecten en risico’s

schaft meer zekerheid. De Transaction wordt weer automatisch beëindigd: en wel reeds op de dag dat de Exchange aankondigt dat het Share uit de notering zal worden genomen. Het compensatiebedrag kan daarom waarschijnlijk mede op basis van een actuele beurskoers bepaald worden.

Index Adjustment Event

37

Physical Settlement

Physical Settlement vindt in beginsel plaats via een overeengekomen Clearance System, dat onder normale omstandigheden open is op, in typisch ISDA speak, Clearance System Business Days, ‘CSBDs’ (ss.

1.27 en 1.36).Voor een groot aantal termijnen in de Definitions is van belang hoeveel CSBDs gewoonlijk gemoeid zijn met de settlement van een transactie in de Shares op de Exchange door het Clearance System. Deze Settlement Cycle bepaalt bijvoorbeeld de dag van levering (in het algemeen: de Settlement Date) onder een Option Transaction met Physical Settlement. Heeft Buyer van een call-optie deze uitgeoefend op een bepaalde dag, de Exercise Date, dan moeten de Shares geleverd worden op de dag die één Settlement Cycle later valt dan die Exercise Date (s.

9.4(a)).

73 N r. 6 7 / o k t o b e r 2 0 0 5

32 Vgl. Euronext (2005).

33 ISDA (2003b).

Verstoring van Physical Settlement

O & F

Dit Extraordinary Event van s. 11.1(b) verstoort de normale afwikkeling van een Index-derivaat. Het valt in drie gevalstypen uiteen. De organisatie verantwoordelijk voor de Index, de Index Sponsor, kondigt een substantiële wijziging van de formule of methode voor de berekening van de Index aan, die niet in de regels voor de Index 32 zelf is voorzien (‘Index Modification’), hij heft de Index definitief op zonder een vervangende index te introduceren (‘Index Cancellation’) of hij ‘fails to calculate and announce’ de Index op een Valuation Date (‘Index Disruption’). Er is een drietal alternatieve reacties op Index Adjustment Event waaruit partijen in de Confirmation moeten kiezen: - opnieuw is ‘Negotiated Close-out’ mogelijk. Mislukt het beëindigingsoverleg, dan wordt de Transaction voortgezet ‘on the terms and subject to the conditions, formulas and calculation methods in effect as of any relevant time at which calculations may be made’. Ook nu is Negotiated Close-out problematisch. Enerzijds lijkt het vrij radicaal om bij Index Disruption meteen close-out mogelijk te maken terwijl het toch mogelijk is dat de Index op de volgende beursdag weer normaal gepubliceerd wordt (in de praktijk ziet men dan ook wel de clausule dat non-publicatie meerdere beursdagen moet aanhouden voordat er Index Disruption is).

Anderzijds, als partijen geen beëindiging overeenkomen, is er onder Index Cancellation juist geen sprake meer van ‘formulas and calculation methods in effect’, terwijl het economisch evenwicht door het gebruik van de nieuwe index bij Index Modification geheel verstoord kan zijn; - het tweede alternatief is ‘Cancellation and Payment’: de Transaction komt automatisch tot een einde door Index Disruption en Index Cancellation en bij Index Modification hebben beide partijen het recht de Transaction te beëindigen. Zoals gezegd is een (automatische) beëindiging in het geval van Index Disruption een radicale oplossing, ook in de context van de Definitions. Immers het ontbreken van een (zinvolle) waarde voor de

Index doordat de Exchange vele Scheduled Trading Days achtereen is gesloten, leidt niet tot beëindiging maar slechts tot Disrupted Days. Overigens lijken de Definitions op het eerste gezicht de samenloop van Index Disruption en Disrupted Day niet te regelen. De Index-regels kunnen bijvoorbeeld bepalen dat als de beurs in het geheel niet open gaat op een Scheduled Trading Day die Index niet wordt berekend. Moet nu het Disrupted Day-regime worden gevolgd of dat van Index Disruption? Ligt de oplossing daarin besloten dat het niet om samenloop gaat omdat de Index Sponsor in zo’n geval niet ‘fails to calculate and announce’ de Index?; - een beter alternatief lijkt ‘Calculation Agent Adjustment’, dat erop neerkomt dat de Calculation Agent doorgaat met het berekenen van de oude Index. De Definitions stellen wel de voorwaarde dat hij daarbij alleen de effecten gebruikt die onmiddellijk vóór het Index Adjustment Event deel uitmaakten van de Index. Zo zal er toch meestal een wijziging van een contractsvoorwaarde worden binnengesmokkeld omdat de samenstelling van vele indices periodiek wordt herzien (en gewijzigd). Definitions noch User’s Guide33 verklaren deze voorwaarde, maar wellicht is deze opgenomen om de Calculation Agent niet op te zadelen met ingewikkelde berekeningen om uit te maken of een effect al dan niet in de pseudo-Index moet worden opgenomen.

bw 67 quark 4.0.qxd

14-11-2005

06:14

Pagina 74

OTC-derivaten, beurs-effecten en risico’s

Settlement Disruption Event

O & F

Mislukt de levering van Shares op een Settlement Date door een Settlement Disruption Event, ‘an event beyond the control of the parties as a result of which the relevant Clearance System cannot clear the transfer of such Share’ (s. 9.8), dan moet op de volgende CSBD een nieuwe poging worden gedaan. Als een Settlement Disruption Event levering verhindert op de acht CSBDs volgend op de oorspronkelijke Settlement Date, moeten de Shares ‘in any other commercially reasonable manner’ geleverd worden als die mogelijkheid bestaat. Is dit niet het geval, dan wordt levering uitgesteld totdat het Clearance System weer werkt of de levering toch ‘in any other commercially reasonable manner’ kan plaatsvinden.

N r. 6 7 / o k t o b e r 2 0 0 5

74

Failure to Deliver

Dit Extraordinary Event geldt alleen bij Physical Settlement. Wanneer een partij met een leveringsplicht door illiquiditeit van de markt onvoldoende Shares kan verkrijgen, is dat de Failure to Deliver van s. 12.9(a)(iii) als de Confirmation deze bepaling van toepassing verklaart. Het gaat in feite erom of de illiquiditeit voor risico van beide partijen wordt gebracht of bij de ‘leverancier’ wordt gelaten. Als Failure to Deliver van toepassing is, is namelijk het niet (volledig) voldoen aan de leveringsverplichting geen Event of Default zoals s. 12.9(b)(ii) onnauwkeurig verklaart (want pas als volledige levering uitblijft na aanmaning en de toepasselijke grace period zou er een Event of Default zijn). De wederpartij heeft toch onmiddellijk een ietwat eigenaardig gestructureerd beëindigingsrecht. Allereerst mag hij dat deel van de Transaction beëindigen dat correspondeert met de verstoorde levering (dat is de gehele Transaction in het geval van een Europese optie; bij een swap kan uiteraard nog een deel van de looptijd resteren); heeft hij dat gedaan dan mag hij vervolgens één Settlement Cycle later het eventuele restant van de Transaction ook beëindigen. Beëindiging in één keer is naar de letter van s. 12.9(b)(ii) niet mogelijk, hoewel de User’s Guide anders suggereert.34 Physical Settlement en Disrupted Days

Het is niet geheel duidelijk of een Disrupted Day levering kan verstoren en tot verschuiving van de Settlement Date kan leiden. In Section 9.4 over Settlement Date wordt niet direct verwezen naar Disrupted Days, maar uitsluitend naar het Settlement Disruption Event. Op het eerste gezicht zou men ook

denken dat bij vele transacties met Physical Settlement omstandigheden op de Exchange of Related Exchange die valuation kunnen verstoren, er nu niet toe doen. Wordt een call-optie uitgeoefend dan is zowel het aantal te leveren Shares als de tegenprestatie, de Strike Price, onafhankelijk van de beurskoers van het Share. Hetzelfde geldt voor een Physicallysettled swap en de meeste forwards; zie ss. 9.2 en 9.3.

De User’s Guide stelt echter juist dat bij uitoefening van een optie op een Disrupted Day ‘execution of a trade to acquire Shares for delivery pursuant to the Option Transaction may be delayed until the Market Disruption Event has ended. [...] [T]his may have the effect of extending the Settlement Cycle for the Shares, which in turn would delay the Settlement Date.’ 35 In feite moet kennelijk in de definitie van Settlement Cycle worden ingelezen dat deze pas begint bij een trade in normale omstandigheden op de Exchange (en de Related Exchange?).

Cancellation Amount; compensatie voor Option Transactions Cancellation Amount

Vervroegd beëindigde forwards en swaps worden gewaardeerd volgens de methode van de 2002 Master Agreement: Close-out Amount is, als ‘Cancellation Amount’, in s. 12.8 vrijwel in zijn geheel overgenomen met grammaticale aanpassingen omdat het hier om de beëindiging van één transactie gaat. Doordat de bepaling is ‘overgeschreven’ – en ook elders in de Definitions geen beroep op de 2002 Master Agreement wordt gedaan – kunnen de Definitions overigens ook in combinatie met de 1992 Master Agreement worden gebruikt.

Toch zijn er inhoudelijke verschillen tussen Close-out Amount enerzijds en Cancellation Amount anderzijds. Ten eerste staat al bij voorbaat vast wie de Cancellation Amount berekent. De Determining Party wordt in de Confirmation aangewezen (s. 12.8(f)). Bij de meeste besproken Extraordinary Events ligt dat ook voor de hand omdat er geen ‘schuldige’ of affected partij is aan te wijzen. Minder duidelijk is dat bij Failure to Deliver en enkele niet behandelde Extraordinary Events met betrekking tot hedging, waar wel van een Affected Party zou kunnen worden gesproken. Failure to Deliver lijkt niet goed te zijn geregeld. S. 12.9(b)(ii) benoemt de wederpartij van de leveringsplichtige tot Determi-

34 ISDA (2003b), p. 57.

35 ISDA (2003b), p. 9.

38

bw 67 quark 4.0.qxd

14-11-2005

06:14

Pagina 75

OTC-derivaten, beurs-effecten en risico’s

ning Party. De verhouding tot s. 12.8(f) is niet geregeld, maar op zich zou de specifieke regel kunnen voorgaan. Echter, als partijen in de Confirmation als Determining Party de partij hebben aangewezen, die later juist de leveringsplichtige blijkt te zijn, zal déze keuze meestal voorgaan ten opzichte van s.

12.9(b)(ii).Vrijwel steeds bepaalt een Confirmation namelijk dat bij inconsistentie tussen de Confirmation en toepasselijke definities de Confirmation prevaleert. Ook de model-Confirmations van ISDA bevatten die bepaling.36

Bij vervroegde beëindiging van opties na sommige Extraordinary Events is een andere compensatiemethode dan Cancellation Amount van toepassing (s. 12.7(b)). Voorop staat dat partijen in onderling overleg tot een bedrag proberen te komen dat Seller aan Buyer dient te betalen (als de premie is betaald, is het altijd Seller die een bedrag dient te betalen).

Heeft dat overleg geen resultaat binnen vijf beursdagen, dan zijn er twee alternatieve compensatiemethoden die de Calculation Agent (niet de Determining Party) kan toepassen: ‘Agreed Model’ of ‘Calculation Agent Determination’. De laatste is eigenlijk geen methode want de Calculation Agent wordt alleen opgedragen het compensatiebedrag te bepalen. De enig relevante beperking hierbij is de algemene van good faith and commercial reasonableness (zie onder ‘De Calculation Agent’). Wellicht zou het strakker omschreven Cancellation Amount een betere oplossing zijn. Andere gevallen van vervroegd beëindigde opties vallen immers gewoon onder Cancellation Amount en de 2002 MA maakt geen enkel onderscheid tussen opties en andere derivaten bij de toepassing van Close-out Amount, waarop Cancellation Amount is gebaseerd.

Het alternatief voor Calculation Agent Determination is Agreed Model. Wie hier een wiskundig optiemodel verwacht dat de Calculation Agent moet toepassen, zal teleurgesteld (of juist gerustgesteld) zijn bij lezing van s. 12.7(b)(i): het gaat om niet meer dan agreed inputs die de Calculation Agent moet gebruiken bij de bepaling van de compensatie. 38 De kern van Agreed Model is veeleer dat de gebruikelijke waardering van een optie gecorrigeerd wordt voor de verandering in de volatiliteit van het onderliggende

36 ISDA (2002b), p. 1.

37 S. 12.8(a); cursivering toegevoegd. De gecursiveerde woorden vormen een ander subtiel verschil: in de Close-out Amount worden alleen de verplichtingen na de Early Termination Date meegenomen.

39

75 N r. 6 7 / o k t o b e r 2 0 0 5

Een laatste verschil is dat de waarde van de vervroegd beëindigde transactie moet worden bepaald ‘but for the Extraordinary Event’. In een enkel geval lijkt er een probleem te zijn. Zo is Delisting de aankondiging van een beslissing van de beurs. Het wegdenken van de aankondiging (but for) betekent echter niet dat de feitelijke delisting niet zal plaatsvinden.

Eigenaardige consequenties heeft de but for-clausule bij Failure to Deliver. Neem aan dat onder een forward partij X slechts 20% van de te leveren Shares op de Settlement Date heeft geleverd tegenover betaling van 20% van het verschuldigde bedrag door wederpartij Y.Y kan de volgende dag al aanwijzen als de dag waarop de transactie eindigt. De Cancellation Amount moet bepaald worden als de waarde van de verplichtingen die op en na die dag required zijn.Vanuit het afrekeningsperspectief van de MAs zijn die er niet want de enige Settlement Date was een dag eer-

Compensatie voor opties

O & F

Een tweede inhoudelijk verschil is dat in de Definitions niet duidelijk wordt geregeld wat er moet gebeuren met betalingen en leveringen die vóór de beëindiging hadden moeten worden verricht, maar dat niet zijn (Unpaid Amounts in de MAs). Hoewel het uitdrukkelijke verbod van de 2002 MA om deze Unpaid Amounts op te nemen in de Close-out Amount, niet in de ‘Cancellation Amount’ van s. 12.8 is overgenomen, lijkt het besloten te liggen in de formulering dat de vervangingswaarde of het economisch equivalent bepaald moet worden van de ‘payments and deliveries […] that would […] have been required on or after the date that the Transaction is […] terminated’.37 De Unpaid Amounts vallen misschien direct onder het regime van de Master Agreement, waardoor zij bijvoorbeeld aanleiding kunnen geven tot een Event of Default, hoewel deze oplossing niet steeds werkt (zie hierna).

der. Zouden Unpaid Amounts toch mogen worden meegenomen in de Cancellation Amount, dan gooit de but for-clausule enig roet in het eten: Failure to Deliver moet worden weggedacht, wat lijkt te betekenen dat X geacht moet worden wél volledig aan zijn verplichtingen te hebben voldaan, zodat hij niets meer heeft te leveren of te vergoeden aan Y. (Zelfs is verdedigbaar dat de betaling van Y van 20% geen onderdeel van Failure to Deliver is en dus niet mag worden weggedacht, zodat Y de resterende 80% aan X moet betalen.) Hier lijkt, kortom, iets fout te zijn gegaan.

bw 67 quark 4.0.qxd

14-11-2005

06:14

Pagina 76

O & F

OTC-derivaten, beurs-effecten en risico’s

N r. 6 7 / o k t o b e r 2 0 0 5

76

Share die veroorzaakt wordt door de enkele aankondiging van de Extraordinary Event (dat hier niet wordt weggedacht). Als voorbeeld is het geval te nemen van een call-optie op aandeel ABC. Hoe meer de koers van ABC varieert, hoe hoger de waarde van de optie. Een Merger Event vindt plaats: onderneming DEF kondigt een overnamebod aan en de geboden prijs is aanzienlijk hoger dan de koers van ABC. Er kunnen twee tegengestelde effecten zijn op de waarde van de optie: enerzijds stijgt de koers van ABC en dus de waarde van de call; anderzijds zal de koers wellicht niet veel meer variëren en in de buurt van de geboden prijs blijven, wat op zich een negatief effect op de waarde van de call heeft. Agreed Model impliceert nu dat Buyer ook gecompenseerd wordt voor het feit dat de waarde van de optie nóg hoger zou zijn geweest als de volatiliteit niet was afgenomen. Dat is niet zonder meer redelijk omdat men evengoed zou kunnen verdedigen dat uitzonderlijke gebeurtenissen voor risico van partijen moeten blijven. Mede daarom is Agreed Model apart opgenomen – zij het als keuzemogelijkheid – en niet onder Cancellation Amount gebracht, waar zo’n volatiliteitscorrectie pas een rol kan spelen als het duidelijk zou zijn dat deze deel van de contractsinhoud is.39 Afronding

Deze kan kort zijn. De ISDA-documentatie voor derivaten is een imponerende verzameling contractsonderdelen, geschikt voor vele typen OTC-derivaten. Dat neemt niet weg dat er voor partijen die een ISDA-contract willen sluiten ‘nog genoeg te doen’ is, mede uit het perspectief van beheersing van financiële risico’s. Immers, de Master Agreement moet in een Schedule worden ingevuld, aangevuld en meestal ook op punten gewijzigd, terwijl product-specific definitions zorgvuldig moeten worden bekeken om te bezien of zij zonder meer op een voorgenomen Transaction van toepassing kunnen zijn, of dat ook zij geamendeerd moeten worden. Zoals ISDA zelf aangeeft 40 zullen bij meer exotische derivaten ISDAdefinities als nuttig uitgangspunt kunnen dienen, maar zullen partijen en hun adviseurs toch uiteindelijk zelf dergelijke transacties juridisch moeten vormgeven.Ten slotte is er op punten aanleiding voor verbetering en amendering van, bijvoorbeeld, de Definitions, zelfs als zij worden ingezet bij standaardtransacties.

Literatuur

Agentschap ministerie van Financiën (Agentschap) (2003), Nederlandse staatsobligaties 2003, Den Haag en www.dutchstate.nl.

Bank for International Settlement (BIS) (2005), ‘Statistical Annex’, BIS Quarterly Review, www.bis.org.

Broekhuizen, K.W.H. en M.G.C.M. Peeters (2000),‘Vervroegde beëindiging van swaps’, Tijdschrift voor Effectenrecht, 2(5), p.79-84.

Euronext (2005), Rules for the AEX-index®, www.euronext.com.

Euronext.liffe (2004), Corporate Actions Policy, www.euronext.com.

Farrell, Scott (2005),‘Dispelling the three myths of Enron v TXU’, International Financial Law Review XXIV 5, p. 40-42.

Financial Markets Law Committee (FMLC) (2003), ‘Emergency power legislation’, www.fmlc.org.

Global Documentation Steering Committee (GDSC) (2003), ‘Global Documentation Steering Committee Recommendations to the 2002 Master’, www.newyorkfed.org/globaldoc.

Harding, P. (2004), Mastering the ISDA Master Agreements (1992 and 2002), 2nd edition, Harlow: Pearson.

ISDA (2003a), User’s Guide to the 2002 ISDA Master Agreement, New York: ISDA. ISDA (2003b), User’s Guide to the 2002 ISDA Equity Derivatives Definitions, New York: ISDA. ISDA (2002a), 2002 ISDA Equity Derivatives Definitions, New York: ISDA. ISDA (2002b), Confirmations for use with the 2002 ISDA Equity Derivatives Definitions, www.isda.org.

ISDA (2000), 2000 ISDA Definitions, New York: ISDA. Leeger, C.F. (2002), Swaps onder ISDAdocumentatie, Deventer: Kluwer.

Leeger, C.F. en N.C.J. Renkens (2003),‘De nieuwe 2002 ISDA Master Agreement: een vlaggenschip in revisie’, Tijdschrift voor Effectenrecht, 5(4), p. 64-77.

Peeters, Marcel (2001), ‘De 2000 ISDA Definitions’, Tijdschrift voor Effectenrecht, 3(8/9), p. 148-154.

Rae, Glen A. (2003), ‘The 2002 ISDA Equity Derivatives Definitions’, Futures & Derivatives Law Report, 23(3), p. 7-10.

De Vries Robbé, J.J. (2003), ‘Credit Derivatives Definitions 2003: two steps forward, one step back’,

38 Anders: Euronext.liffe (2004),Appendix 1.

39 Vgl. ISDA (2003b), p. 55-56.

40 Definitions, p.v.

40

bw 67 quark 4.0.qxd

14-11-2005

06:14

Pagina 77

OTC-derivaten, beurs-effecten en risico’s

Tijdschrift voor Effectenrecht, 5(3), p. 35-40.

Weinstein, Jeremy D., Bruce MacIntyre en William F. Henze III (2004), ‘Escape From the Island of the One-Way Termination: Expectations and Enron v.

TXU’, Futures & Derivatives Law Report 24(8), p. 1-9.

Westeinde, M.G. van ’t (1999), ‘Kredietderivaten en pensioenfondsen’, Tijdschrift voor Effectenrecht, 1(3), p. 56-62.

Westrate, H.W. R. (2001), ‘OTC-derivaten: Van ISDA Master Agreement naar Algemene Bepalingen Derivatentransacties ABN AMRO’, Tijdschrift voor Effectenrecht, 3(2), p. 31-38.

M a rc e l Pe e t e r s i s ve r b o n d e n a a n d e f o c u s g ro e p F i n a n c i a l S e r v i c e s, D e r iva t ive s & R eg u l a t o r y va n NautaDutilh te Amsterdam.

O & F

77 N r. 6 7 / o k t o b e r 2 0 0 5

41

42

43

44

45

46

47

48

49

Capital Markets Law Journal, Vol. 6, No. 2

149

Valuation in the context of derivatives litigation Richard Grove*

Key points

1. Introduction The financial crisis of the past few years has spawned an unprecedented number of disputes relating to swaps and other privately negotiated derivatives transactions. Many of these disputes have resulted in litigation which is now progressing through courts and arbitration tribunals around the world. In many, if not all, of these disputes, valuation of the relevant derivatives is a central issue.

Since swaps and other privately negotiated derivatives transactions are often bespoke transactions, their valuation presents a challenge. In contrast to exchange traded futures, which are also considered derivatives and for which a value can readily be determined by reference to the prices at which they trade on the relevant exchange, swaps and other privately negotiated derivatives are, in many instances, unique transactions. Swaps and other privately negotiated derivatives (which are sometimes called over the counter derivatives to distinguish them from exchange traded derivatives and henceforth will be referred to collectively as ‘OTC derivatives’) are also, as the terminology implies, not publicly traded transactions. This means that there is not often a comparable transaction from which a publicly available price can be observed.

My firm, Rutter Associates, has been called as an independent expert to address valuation issues in many of the disputes relating to OTC derivatives. Some of these disputes present legal issues as to the meaning of the language adopted by the International Swaps and Derivatives Association (ISDA) in its widely used 1992 and 2002 master agreements. However, interpretation of the ISDA language (a legal issue) is usually not the central issue in these disputes. Rather, the main issue in these disputes more often relates to questions of valuation (an issue of economics). Even when valuation is not the central issue, it is almost always an important ancillary issue. *Richard (Rick) Grove, Partner and Chief Executive Officer, Rutter Associates LLC, New York. Mr Grove wishes to thank Charles Smithson, Bob Selvaggio and Xiaojing Su of Rutter Associates, for their contributions. Ă&#x; The Author (2011). Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com

doi:10.1093/cmlj/kmr001

Accepted 25 January 2011

50 publication 25 February 2011 Advance Access

Downloaded from http://cmlj.oxfordjournals.org/ at Universiteit van Amsterdam on October 17, 2013

Litigation involving privately negotiated derivatives has increased dramatically in the wake of the financial markets crisis of recent years.

Valuation of the derivatives involved is a central issue in much of this litigation.

Since many privately negotiated derivatives are bespoke transactions, there is not likely to be a price that can be observed from a public trading market for purposes of valuing them.

As a result, valuation of privately negotiated derivatives often requires the use of an expert familiar with market standard methodology for determining their value.

150

Capital Markets Law Journal, 2011, Vol. 6, No. 2

This article will focus on the issue of valuation from an economic perspective and not a legal perspective. First, it will look at how valuation issues arise. Then, it will look at the principles that govern valuation exercises generally and the methodology that is used in the valuation of OTC derivatives specifically. Finally, it will address some special issues of interest that are worth noting in these valuation exercises.

2. Contexts in which questions of valuation arise Close-out value/damages Almost all OTC derivatives disputes involve the declaration of an early termination date.

Under the ISDA master agreements, the declaration of an early termination date entitles the non-defaulting party to calculate the value of the derivative as of the early termination date and thereby determine the amount that should be paid by the defaulting party to the non-defaulting party or vice versa.1 The 1992 ISDA Master Agreement allows the parties to select, at the time they enter into the Master Agreement, either of two methodologies for calculating the value of an OTC derivative.2 The first methodology is called ‘Market Quotation’ and requires the non-defaulting party to seek quotations from four market-makers for a transaction that would in effect replace the terminated derivative.3 The non-defaulting party may choose the day and time as of which the quotations will be obtained, but in choosing the day and time, the non-defaulting party is required to act in ‘good faith’. If the non-defaulting party receives three or more quotations, the highest and lowest quotations are disregarded and the arithmetic mean of the remaining quotations is deemed to be the value of the derivative in question unless the non-defaulting party is of the ‘reasonable belief’ that this is not a ‘commercially reasonable result’. If too few quotations (ie fewer than three) are obtained to produce a ‘Market Quotation’ or if the non-defaulting party determines that the result is not ‘commercially reasonable’, then the non-defaulting party is directed to use the second methodology.4 The second methodology is called ‘Loss’ and requires the non-defaulting party to determine its total losses and costs (or gain) in connection with the terminated derivative.5 The definition of ‘Loss’ lists several factors that the non-defaulting party may take into account, including ‘quotations of relevant rates or prices from one or more leading dealers in the relevant markets.’ However, ‘Loss’ does not provide for a formulaic damage calculation as is the case with the ‘Market Quotation’ methodology, but rather leaves it to the non-defaulting party to ‘reasonably’ determine its damages ‘in good faith.’

The 2002 ISDA Master Agreement replaces the two alternative methodologies set out in the 1992 ISDA Master Agreement with a single methodology known as the ‘Close-out Amount’.6 ‘Close-out Amount’ requires the non-defaulting party to determine the 1 2 3 4 5 6

1992 1992 1992 1992 1992 2002

ISDA ISDA ISDA ISDA ISDA ISDA

Master Master Master Master Master Master

Agreement Agreement Agreement Agreement Agreement Agreement

s s s s s s

6(a) and 6(e)(i) and 2002 ISDA Master Agreement s 6(a) and 6(e)(i).

6 (e)(i).

14 (definitions of ‘Market Quotation’ and ‘Reference Market-makers’).

14 (clause (b) of definition of ‘Settlement Amount’).

14 (definition of ‘Loss’).

6(e)(i).

51

Richard Grove Valuation in the context of derivatives litigation

151

amount of its losses or costs (or gain) that would be incurred in replacing the terminated derivative, or in providing for the non-defaulting party ‘the economic equivalent’ of the terminated derivative.7 The non-defaulting party is authorized to include ‘costs of funding’ and costs related to the termination or re-establishment of any related hedge so long as it does not double count any amount. Similar to the standard of conduct to which a non-defaulting party must adhere in determining ‘Loss’ under the 1992 Master Agreement, the non-defaulting party determining ‘Close-out Amount’ under the 2002 Master Agreement must ‘act in good faith and use commercially reasonable procedures in order to produce a commercially reasonable result’. In determining ‘Close-out Amount’, the non-defaulting party is required to use quotations for replacement transactions or relevant market data such as rates, yields, yield curves, volatilities, spreads, correlations and the like, unless such quotations or market data are not ‘readily available’ or their use would produce a result that is not commercially reasonable, in which case, the non-defaulting party may, but is not required to, use quotations or data from internal sources. The non-defaulting party may use its own pricing or valuation models if such models are used ‘in the regular course of business in pricing or valuing’ similar transactions between it and unrelated third parties.

Among the three ISDA methodologies, only ‘Market Quotation’ comes close to being a mechanical determination of value. Even in the case of valuations determined by ‘Market Quotation’, there is ample room for parties to dispute the result. For example, the defaulting party may argue that the non-defaulting party’s choice of the day and time as of which the non-defaulting party sought quotations was not chosen in ‘good faith’. For another example, if the non-defaulting party has determined that the results of the ‘Market Quotation’ determination are not ‘commercially reasonable’ and the non-defaulting party opts for the ‘Loss’ fallback, the defaulting party may dispute that conclusion.

If there is ample room for parties to dispute the results of a ‘Market Quotation’ valuation, there is even greater scope for parties to dispute the results of ‘Loss’ and ‘Close-out Amount’ valuations. The 1992 ISDA Master Agreement and the 2002 ISDA Master Agreement leave it to the non-defaulting party to calculate ‘Loss’ and ‘Close-out Amount’, respectively. Furthermore, in doing so, the ISDA agreements give an amount of discretion to the non-defaulting party to choose the factors it will consider, and the methodology it will use, in making its determination. However, both ISDA master agreements impose requirements of reasonableness and good faith on the non-defaulting party. These standards provide the defaulting party with a basis upon which it can challenge the determinations of the non-defaulting party.

Valuation as an issue that leads to dispute While most OTC derivatives disputes involve valuation issues relating to the calculation of damages once liability has been established, in some cases, the dispute itself has its origin in a valuation issue.

7

2002 ISDA Master Agreement s 14 (definition of ‘Close-out Amount’).

52

152

Capital Markets Law Journal, 2011, Vol. 6, No. 2

An example of a valuation issue that frequently leads to a dispute between parties to an OTC derivatives transaction can occur in transactions where at least one of the parties has agreed to post collateral to secure the performance of its obligations under the transaction. In these instances, the party posting collateral often posts a fixed ‘initial’ collateral amount and agrees to post additional ‘variation’ amounts of collateral if the value of the transaction moves in favour of its counterparty. Since the value of an OTC derivative will vary with the movement in the underlying markets upon which the derivative is based, the value of an OTC derivative normally changes every business day.

As a result, the ‘variation’ amount of collateral is normally recalculated every business day based upon the change in value of the derivative on that day. Depending upon whether the value of the derivative has moved against or in favour of the party posting collateral that party may be required to post additional collateral or may be entitled to a return of some of the collateral it has previously posted. While the party to whom collateral is posted normally has the right to determine the valuation of the derivative and the resulting required collateral amount, the party posting the collateral normally has the right to dispute the valuation and the consequent collateral calculation. In most cases of dispute, the parties are able to reach an amicable agreement. However, in some cases, especially in markets that are moving rapidly as was the case at times during 2007 and 2008, these disputes are not resolved amicably and result in one party or the other declaring an event of default and moving to terminate the derivative. In these cases, not only is valuation necessary to determine damages, but it is also critically important to the establishment of liability.8 Valuation as an issue that bears on another issue Valuation issues can also arise in relation to disputes about the inception of an OTC derivatives transaction. While most OTC derivatives are transacted on the understanding that each party is making its own determination as to the appropriateness of its entering into the transaction, it is not unheard of for a party to allege subsequently that it was not provided with proper disclosure or that the transaction was not suitable for it.

Whether such arguments, if proven, are grounds for allowing a party to walk away from a transaction is a legal issue beyond the scope of this article. Suffice it to say for our purposes, that if such an argument is to be considered, issues of valuation might come into play. For example, a party might argue that it was presented by its counterparty with incorrect valuation analyses, which induced it to enter into the transaction. Or a party might argue that, if it had understood that a transaction had a particular value at inception, it would not have entered into the transaction. If any of these arguments is to be entertained, then clearly the appropriate valuation of the derivative at the relevant points in time could have bearing on the outcome of the dispute.

8 It is worth noting that disputes can arise over the value of the collateral that has been posted as well as over the value of the derivative itself. While the most prevalent forms of collateral (cash deposits and government securities) are rarely, if ever, the subject of a valuation dispute, less liquid types of collateral are occasionally permitted and the valuation of such less liquid collateral can give rise to a dispute.

53

Richard Grove Valuation in the context of derivatives litigation

153

3. Principles of valuation First choice: observable prices from a liquid and transparent market Valuation is not a difficult exercise if the asset or position being valued trades with liquidity and frequency in an observable market. For example, the value of a share of a widely traded company, such as British Petroleum, and the value of a widely traded bond, such as the on-the-run 10 year United States Treasury Note, can both be readily determined by reference to the markets in which these instruments trade. This is true with respect to many exchange traded derivatives as well. For example, the Chicago Board of Trade (CBOT) listed future on the 10 year United States Treasury Note trades with sufficient liquidity and transparency that its value at any point during the trading day can be readily determined.

Markets that are liquid and transparent, such as those mentioned in the preceding paragraph, offer not only readily observable current pricing information, but also historical pricing data. So if one is looking to determine the value, as of a particular day in the past, of an instrument, which trades in a liquid and transparent market, data with respect to trades on that day can usually be obtained. In the case of instruments which trade on an exchange, such as the stock of British Petroleum or the CBOT 10 year United States Treasury futures contract, it is usually the exchange on which such instrument trades that stores and makes available this data. In the case of widely traded instruments, which trade in a non-exchange environment, such as the United States 10 year Treasury Note, other data providers often capture and make available historical trading data.

As noted above, OTC derivatives, in contrast to equities, government bonds and exchange-traded derivatives, are often bespoke structures, which are negotiated as private transactions. As a result, there is usually not a publicly observable trade in an OTC derivative from which a valuation of that derivative can be determined without some further computation being required.9 That is not to say that there may not be a wealth of current and historical data available regarding similar transactions. For example, current and historical data with respect to US dollar interest rate swaps are readily available.

However, the swaps for which this data are available will not take into account the bespoke features, such as notional amount, payment dates, etc., of every interest rate swap which one might need to value. The more bespoke the OTC derivative being valued, the less likely it is that one will be able to find current or historical pricing data for that derivative. Moreover, the available pricing data assumes that the contracting parties are 9 One of the current goals of policy-makers in the United States, Europe and elsewhere is to introduce greater price transparency into the OTC derivatives market. For example, the Dodd–Frank legislation, which was passed in the United States in 2010, mandates that certain derivatives that were previously transacted off of exchanges be transacted on exchanges and that other non-exchange traded derivatives be subject to reporting requirements that would include pricing information. The Dodd–Frank legislation leaves it up to the regulatory agencies in the United States to define these requirements. The agencies are currently drafting the relevant rules, which are expected to be promulgated sometime in the middle of 2011. Until these rules are published, the impact of the legislation will not be known for certain. However, these rules are not expected to apply to pre-existing OTC derivatives nor to new OTC derivatives that meet certain criteria. Therefore, it is likely that, for many OTC derivatives, it will continue to be difficult to find a publicly observable trade from which a valuation of that derivative can be determined without some further computation being required.

54

154

Capital Markets Law Journal, 2011, Vol. 6, No. 2

highly-rated dealers and does not account for the need to consider and price for credit risk when one party to an OTC derivative is of a lower credit quality than the other.

When relevant pricing information for a particular instrument is available from a liquid and transparent market, then this information would normally be the preferred source for valuing that instrument. However, when relevant pricing information is not available, then valuation will depend on more complex methodologies and data sources.

Principles for valuations where prices from a liquid and transparent market are not observable Three principles should govern the valuation of ‘hard-to-value’ assets and liabilities:10 1. Transparency: the methodology by which such a valuation is performed should be clearly stated and capable of being understood (and audited) by the audience for which the valuation is being performed. The methodology should be able to be described in a step by step outline. There should not be any unexplained ‘black box’ component to the methodology.

2. Consistency: to the extent possible, the same methodology should be used in similar cases. That is not to say that variations might not be required as a result of differences in the terms of the instrument being valued or the availability of data or for other reasons. However, valuations should be readily reproducible by third parties.

3. Independence: valuations should be conducted using methodologies and principles rooted in market practice and the academic literature. The party conducting the valuation should begin from this point and not refer initially to a desired outcome from which a methodology is then reverse-engineered to produce the desired outcome.

Valuations should be verifiable by disinterested third parties.

Basic methodologies for valuations where prices from a liquid and transparent market are not observable There are three primary methodologies for valuation11 in cases where prices from a liquid and transparent market are not observable:12 1. Relative valuation: the relative valuation methodology estimates the value of an instrument by looking at the price of comparable instruments relative to a common 10 Charles Smithson, ‘Valuing ‘Hard-to-Value’ Assets and Liabilities: Notes on Valuing Structured Credit Products’ (2008) 19 J Appl Finance 2.

11 While economists can identify three primary methodologies for valuation in cases where prices from a liquid and transparent market are not observable, the specific facts of a case could argue for a methodology that does not fall within one of the three primary methodologies outlined herein and that a court might accept as appropriate. Such an example can be found in one recent English case in which the court decided not to value a call option at the price at which it might have been sold, but rather to value the option by reference to the value it might have enabled its holder to realize by shorting the asset underlying the option and using the call option to cover the short position should the underlying asset increase in value (Maple Leaf Macro Volatility Master Fund v Rouvroy [2009] 2 All ER (Comm) 257). By citing this example, I do not mean to agree or disagree with the valuation approach adopted in this case. I merely present it as an example of how the facts of a case might present an opportunity to put forward a less common valuation methodology.

12 See Smithson (n 1) 2 and Aswath Damodaran, Investment Valuation:Tools and Techniques for Determining the Value of Any Asset (John Wiley and Sons, Hoboken 2002) 11.

55

Richard Grove Valuation in the context of derivatives litigation

155

variable.13 For example, the credit spread to be assumed in pricing the bonds of an issuer whose debt does not trade with liquidity can be derived from the credit spread implied by the prices of the bonds of similarly rated issuers whose debt does trade with liquidity.

2. Discounted cash flow valuation: the discounted cash flow methodology estimates the value of an instrument by determining the expected future cash flows generated by the instrument over its remaining life and discounting those future cash flows to their present value as of the date of the valuation.14 This methodology begins with a determination of the contractually specified cash flows. If the instrument includes any optionality such that any of these cash flows is subject to a contingency, then the contingent claim valuation methodology below would be employed. Otherwise, if there are no contingent cash flows and the amount and timing of the contractually specified cash flows can be determined with certainty, an appropriate discount rate is selected and the present value of each contractually specified cash flow is calculated and summed to determine the value. I will go into more detail on the discounted cash flow valuation methodology below.

3. Contingent claim valuation: if the instrument to be valued includes any optionality such that any of the contractually specified cash flows is subject to a contingency, then the future cash flows cannot be determined with certainty. In this case, an option pricing model has to be employed to calculate not only the amounts and timing of the future cash flows, but also the probability that they will occur. In a sense the contingent claim valuation methodology is really a special type of discounted cash flow methodology. Once the probability of the cash flows is determined using an appropriate option pricing model, then the resulting cash flows, as discounted by their probability, are present valued using an appropriate discount rate and summed to determine the value of the instrument in question. I will go into more detail on the contingent claim valuation methodology below.

The 1992 ISDA Master Agreement’s ‘Market Quotation’ methodology is essentially a relative valuation methodology. In the ‘Market Quotation’ methodology, valuation is determined by reference to quotations for a transaction that is comparable to the transaction being valued, albeit a transaction between different counterparties. The ‘Loss’ methodology in the 1992 ISDA Master Agreement allows the non-defaulting party to determine valuation by reference to a quotation for a relevant price and, thereby, permits, but does not require the use of relative valuation methodology. The ‘Close-out Amount’ methodology in the 2002 ISDA Master Agreement requires the non-defaulting party to determine valuation by reference to a market quotation for a replacement transaction unless a market quotation is not ‘readily available’ and, thereby, establishes a preference for relative valuation methodology. However, the definition of ‘Close-out Value’ in the 13 Relative valuation is also known as ‘Comparable’ valuation.

14 Richard A. Brealey, Stewart C. Myers and Franklin Allen, Principles of Corporate Finance (8th edn, McGraw-Hill/Irwin, New York 2006) 36.

56

156

Capital Markets Law Journal, 2011, Vol. 6, No. 2

2002 ISDA Master Agreement permits the use of alternatives to relative valuation methodology if a market quotation is not ‘readily available’.

So, relative valuation methodology figures prominently in both the 1992 and 2002 ISDA master agreements. However, in recognition that relative valuation methodology will often not be available, both the ‘Loss’ methodology in the 1992 ISDA Master Agreement and the ‘Close-out Amount’ methodology in the 2002 ISDA Master Agreement contemplate alternatives. The discounted cash flow valuation methodology and the contingent claim valuation methodology play important roles in calculating ‘Loss’ and ‘Close-out Amount’ in cases where relative valuation methodology is unavailable.

The following table illustrates the extent to which the ISDA master agreements rely upon these three valuation methodologies.

Methodology

Relative valuation Discounted cash flow valuation Contingent claim valuation

1992 ISDA Master Agreement

2002 ISDA Master Agreement

Market Quotation

Loss

Close-out Amount

Required Fall back to ‘Loss’ only if quotations are not available Fall back to ‘Loss’ only if quotations are not available

Permitted Permitted

Preferred Permitted if quotations are not available Permitted if quotations are not available

Permitted

4. Issues presented by the basic valuation methodologies Relative value methodology It is for good reason that the 1992 ISDA Master Agreement gives significant deference to market quotations, both in the ‘Market Quotation’ option where they form the basis of the valuation determination and in the ‘Loss’ option where they are a permissible basis for the valuation determination. Likewise, the 2002 ISDA Master Agreement directs the use of market quotations where they can be obtained. The premise underlying this preference for market quotations is that, if available, they are usually the best indicators of market value. Furthermore, if market quotations are available, the determination of the value of an OTC derivative is a pretty simple arithmetic exercise.

Although preferred, market quotations may not always be available. In some market environments, dealers may be reluctant to provide quotations. An example of just such a market environment occurred in the immediate aftermath of the Lehman Brothers bankruptcy in 2008. During the second half of September 2008, dealers needed to focus on mitigating the risk of their positions with Lehman Brothers, dealing with rapidly moving markets and preparing for the possibility that other financial institutions would collapse in Lehman’s wake. Not surprisingly, providing quotations for potential new transactions was not high on the list of most dealers’ priorities.

57

Richard Grove Valuation in the context of derivatives litigation

157

Even in more normal market environments, it may be difficult or even impossible to obtain market quotations. Some counterparties of lesser credit standing may not find any dealer willing to quote a price to them, especially for a transaction in which the dealer might be exposed to significant credit risk. Even if credit is not an issue, some transactions may be so customized or so complex, that other dealers may not be capable, or willing, to quote on them.

Even when market quotations can be obtained, there is some risk that the quotation process could have been manipulated. Most market participants pride themselves on their professionalism and would not even consider providing anything other than their true quotation for a given transaction. However, there is some risk that the party seeking quotations could tip off the party making the quotations that an off-market quotation would be acceptable because of the context. If manipulation is suspected, it may be in the interest of one party (or both parties) to the original derivative to check the accuracy of the market quotations using one of the other valuation methodologies (eg discounted cash flow or contingent claim).

While market quotations may be the preferred determinant of the contemporaneous value of an OTC derivative (eg immediately following the declaration of an early termination), market quotations are not generally used for historical valuations of OTC derivatives. For example, it would not be market practice to ask a dealer at a later date to provide a quotation as of a prior date, such as the date on which an OTC derivative was originally transacted. The concern in such an instance is that the dealer would know that it is not being asked to provide a trading price upon which it might be expected to transact. Thus, there is a greater likelihood that any quotation provided would not be a truly market quotation. Furthermore, it is not inconceivable that the benefit of hindsight might skew the dealer’s calculation of the quotation.

Thus, although market quotations may often be the preferred method for valuing OTC derivatives, there are many instances where market quotations will not do the job. This is why the ISDA master agreements recognize and permit the use of alternatives that depend fundamentally on different methodologies.

Discounted cash flow methodology Discounted cash flow methodology is a well known and widely used tool for valuing assets and liabilities.15 In order to value an asset or a liability, a valuation date is selected.

All future cash flows relating to the asset or liability, both positive and negative, subsequent to the valuation date are determined. To account for the time value of money (an amount today, under most circumstances, is worth more than the same amount in the future), an appropriate discount rate is applied to each future cash flow to determine the ‘present value’ of such cash flow as of the valuation date. The present values of all cash flows are then summed to determine the value of the asset or liability in question as of the valuation date.

15 Ibid 36; James C. Van Horne, Financial Management and Policy (4th edn, Prentice-Hall Inc., Englewood Cliffs 1977) 83.

58

158

Capital Markets Law Journal, 2011, Vol. 6, No. 2

At first blush, using discounted cash flow methodology to value OTC derivatives seems fairly straight forward. However, in most OTC derivatives transactions, at least one set of cash flows is determined by reference to floating market rates fixed at specified points in the future. In the case of a fixed for floating US dollar interest rate swap, for example, cash flows on the ‘fixed’ leg are determined by reference to a specified fixed rate multiplied by the notional amount of the transaction and are known at the outset, but cash flows on the ‘floating’ leg, which may be determined pursuant to a series of future US dollar LIBOR16 settings, cannot be known until each setting occurs. Fortunately, there is a solution to this problem. In order to determine the future cash flows for the floating leg, reference can be made to the forward curve, as of the valuation date, for the relevant floating rate, which in the case of my example is US dollar LIBOR. The actual LIBOR fixings in the future are quite likely to differ from the relevant points on the forward curve. However, as of the valuation date, the forward curve represents the market’s view of where LIBOR will be fixed in the future and, therefore, provides an appropriate basis from which to determine the floating leg cash flows that will be factored into the valuation.17 The forward rates are the ‘no arbitrage’ rates, which is to say that they are the rates at which market participants are indifferent between a future stream of payments and their present value and, therefore, they are the rates at which market participants can lay off risk.

Using forward curves to determine as of yet unspecified cash flows is a tool that can be used for many types of transactions in addition to the plain vanilla interest rate swap illustrated in the previous paragraph. For example, in a fixed for floating West Texas Intermediate (WTI) crude oil swap, reference can be made to the New York Mercantile Exchange WTI crude oil futures curve to determine the cash flows on the floating leg.

Similarly, forward foreign currency exchange rates can be referred to in transactions where cash flows depend on future foreign exchange rates.

It is important to remember that the forward curve data that should be used is the data that existed as of the valuation date. Subsequent market movements must be disregarded if one is to determine the value as of the valuation date. Some parties make the mistake of taking into account market information from dates subsequent to the valuation date. Doing so only serves to show how the value may have changed subsequent to the valuation date. All that is relevant in calculating a value as of the valuation date is information that was known as of the valuation date.

As long as forward curve data exists as of the valuation date, it can be used to determine cash flows that under the contract will be determined in the future. However, in some instances, relevant forward curve data may not exist. For example, the maturity 16 LIBOR is the widely used acronym for the London InterBank Offered Rate, a daily rate set pursuant to rules published by the British Bankers Association (the BBA). LIBOR is produced for ten different currencies and for 15 different maturities for each currency. LIBOR for each currency and maturity is compiled from a panel of quoting banks who submit the rate at which they believe they could borrow funds in the London interbank market by asking for offers and accepting funds at the rates offered.

LIBOR is a widely used benchmark in the financial markets.

17 John C. Hull, Options, Futures, and Other Derivatives (3rd edn, Prentice-Hall Inc., Upper Saddle River 1997) 123.

59

Richard Grove Valuation in the context of derivatives litigation

159

of an interest rate swap may extend beyond the length of the publicly available forward curve data for the rate by which the floating leg is determined. In such a case, one may need to determine the relevant forward rates by using econometric techniques to extrapolate the existing forward curve into the future to cover the relevant points.

Once the cash flows have been determined, an appropriate discount rate needs to be selected to present value each of the cash flows. Typically, the discount rate that is used is LIBOR because LIBOR is a measure of the approximate rate at which major swap dealers borrow in the interbank market.18 For US dollar OTC derivatives cash flows, the LIBOR rate used to discount the cash flows is normally constructed from a curve based on some number of months of eurodollar deposits (eg three months), followed by LIBOR futures out typically to two or three years and the swap curve thereafter.19 An adjustment needs to be applied to the futures contract so that the rates are all on a consistent basis.20 This curve results in zero coupon rates that can be determined for each date along the curve. Each cash flow is then discounted at the zero coupon rate that applies to the payment date of that cash flow to generate the present value of that cash flow as of the valuation date. The present values of all of the cash flows are then summed to complete the valuation.

The valuation resulting from the discounting exercise in the preceding paragraph would in effect be a ‘mid-market’ valuation. No bid–offer spread has been built into the valuation. A ‘mid-market’ valuation may be the desired goal in many cases. For example, the parties to an OTC derivative may agree to value the derivative at ‘mid-market’ for purposes of determining the collateral to be posted by one party or the other in respect of that derivative. Likewise, a ‘mid-market’ valuation may well be appropriate for purposes of assessing the value of an OTC derivative as of the original trade date.

However, the valuation of an OTC derivative by a non-defaulting party following a default by its counterparty should not be at mid-market. In the case of a default, if either the ‘Loss’ methodology specified by the 1992 ISDA Master Agreement or the ‘Close-out Amount’ methodology specified by the 2002 ISDA Master Agreement applies, the non-defaulting party may refer to market quotations for a replacement derivative in calculating its damages. Market quotations provided by dealers will not be at ‘mid-market’ because dealers will insist on both (1) adding, or subtracting, a bid or offer spread to the applicable mid-market rate in order to compensate the dealer for taking on the market risk of the transaction,21 and (2) adding, or subtracting, a spread to

18 Ibid 121–22.

19 For swaps that are collateralized, some dealers use a discount rate based on the Fed Funds overnight index swap (OIS) curve instead of LIBOR. The OIS curve normally produces a lower discount rate than the LIBOR curve. This lower discount rate may be appropriate given the reduced credit risk resulting from the collateral.

20 Futures prices require a ‘convexity adjustment’ because futures positions are margined daily while quotes for eurodollar deposits and swaps incorporate no margining.

21 Typically, a dealer will not actually bear the full market risk for the term of the transaction. Instead the dealer will usually hedge part, or all, of the market risk. The bid or offer spread is expected: (a) to cover the cost of the dealer’s hedge; and (b) provide the dealer with a profit.

60

160

Capital Markets Law Journal, 2011, Vol. 6, No. 2

compensate the dealer for the credit exposure that it would be assuming with respect to the non-defaulting party.

In order to take market risk into account in valuing an OTC derivative, the relevant bid or offer spread for market risk can be added to, or subtracted from, the appropriate discount rate that is applied to the leg (or legs) of the trade on which the spread is quoted. Bid and offer spreads are market data which can be obtained, with greater or lesser difficulty depending on the relevant derivative, both for current and historical points in time. Generally, the more liquid the market for an OTC derivative, the more likely it is that bid and offer spreads will be available. In the case of less liquid derivatives, the appropriate bid and offer spread may need to be determined using econometric techniques. For example, in the case of a plain vanilla US dollar interest rate swap of a fixed rate against a floating rate determined by reference to US dollar LIBOR, the bid and offer spread is normally obtainable. In order to value a US dollar interest rate swap, the discount rate used to present value the fixed leg of the swap would be adjusted by the bid or offer spread observable in the market as of the valuation date. It should be noted as well that it may be appropriate to adjust observable bid and offer spreads to account for other factors, including large size (where the transaction being valued exceeds the standard size transaction traded in the market) and illiquidity.

The credit risk to which a dealer is exposed when entering into an OTC derivative with another party is a function of several factors, most notably: (1) the dealer’s potential future exposure due to potential market changes that could result in an increase in the mark to market value of the derivative to the dealer; (2) the probability that the other party will default at any given point in time; and (3) the dealer’s expected recovery if the other party does indeed default. The interplay of these three factors is typically modeled by means of a simulation technique, such as a Monte Carlo simulation, the use of which will result in an expected credit loss amount from the transaction for which the dealer will require compensation for taking the risk of this loss. Such a simulation exercise requires as inputs (1) the forward curve and the implied volatility for the rate or price underlying the derivative, as of the valuation date, in order to determine the potential future exposure of the dealer and (2) the probability of default of the other party and the expected recovery in the event of such a default, again as of the valuation date.

The forward curve and implied volatility are typically obtained from observable market data. The probability of a party’s default and the expected recovery in the event of such a default can often be determined by reference to market data such as observable credit default spreads and/or bond spreads for that party. If such market data is not available, then it may be possible to use econometric techniques to estimate the probability of default and expected recovery.

Contingent claim methodology Many OTC derivatives have option characteristics such that the cash flows of one or more legs of the transaction are not known with certainty at the outset, but rather depend on 61

Richard Grove Valuation in the context of derivatives litigation

161

the occurrence of subsequent market performance. In valuing an OTC derivative with option characteristics, a simple (single) discounted cash flow methodology cannot be used due to the uncertainty of the cash flows. Instead, an option pricing model is needed.

As noted above, contingent claim valuation methodology is really a special type of discounted cash flow methodology. With the contingent cash flow methodology, an appropriate option pricing model is used to determine the amount and probability of the cash flows. Once the amount and probability of the cash flows have been determined, the resulting cash flows, as discounted by their probability, are present valued using an appropriate discount rate and summed to determine the value of the instrument in question.

There are several possible option pricing models that can be used to determine the amount and the probability of the cash flows of an OTC derivative with option characteristics. The most common choices of option pricing model are a Monte Carlo simulation or a lattice approach. The choice of model will depend on factors such as the nature of the relevant option, the availability of data and the style of the options (eg American, European or other).

Whichever model is chosen, a forward curve for the underlying rate or price as well as the implied volatility of the rate or price, both as of the valuation date, will be required.

The forward curve represents the market’s central expectation as to the future performance of the rate or price and the implied volatility represents the market’s expectation as to the extent to which the actual future performance might vary from the market’s central expectation, ie the forward curve. To plot the potential variation of the market’s future performance from the yield curve, the shape of the distribution must also be chosen. Common choices for the shape of the distribution include a normal distribution and a lognormal distribution. The choice of distribution will depend on factors such as whether one wants to set a limit on the distribution22 and the extent to which one intends to perform additional calculations with the resulting data.

It is worth noting two points on the subject of volatility. First, the preferred input is observable traded levels of implied volatility. Implied volatility is a market traded measure of expected future volatility of a relevant rate or price. Implied volatility can be extracted from observed prices at which options trade in the market. In some cases, implied volatility levels themselves may be quoted. In other cases, they may need to be backed out of prices at which the relevant options trade. Implied volatility should not be confused with historical volatility which is a retrospective measure of the volatility of a relevant rate or price. If implied volatility levels are not obtainable from the market, substitutes may be used. These substitutes, whether derived from historical volatility or from other sources, may or may not be good proxies for implied volatility. The quality of the valuation in these cases will depend on the reliability of such substitutes. Thus, it is preferable to use observable trading levels of implied volatility when available. Second, 22 If one is working with a rate or price which, for all practical purposes, ought not to move below zero, one could choose to use a lognormal distribution, which would not permit a value of less than zero.

62

162

Capital Markets Law Journal, 2011, Vol. 6, No. 2

implied volatility levels are most frequently available for at-the-money options. The implied volatility of in-the-money or out-of-the money options may, and often does, vary from the implied volatility of at-the-money options. If the options to be valued are in-the-money or out-of-the money, care must be taken to obtain the appropriate implied volatility level from the market, if available, or to adjust the implied volatility level observable for at-the-money options in an appropriate way.

Bid and offer spreads exist for implied volatility just as they do for most other market rates and prices. As is the case with discounted cash flow valuation methodology, a ‘mid-market’ valuation can be obtained by using traded levels of implied volatility.

However, if the purpose of a valuation is to calculate the damages to which a non-defaulting party is entitled following the default by its counterparty to an OTC derivative transaction, then it would be appropriate to take the bid implied volatility or offer implied volatility, as applicable, into account. As is the case with discounted cash flow valuation, it may be appropriate to adjust the bid or offer levels of volatility to account for a number of factors, including large size and illiquidity. Similarly, credit risk should be taken into account in contingent claim valuation just as would be the case in discounted cash flow valuation, as explained above.

5. Conclusion Disputes relating to OTC derivatives have been on the rise. In almost all of these disputes, valuation is an issue which arises in at least one, and sometimes more than one, context in the dispute. If the valuation of OTC derivatives were as simple as looking up a price in a stock table, there would not be much to discuss about valuation, nor would there be much of a need for valuation experts. However, OTC derivatives, by their very nature, are often bespoke transactions for which prices are usually not observable in the market.

Thus, valuation requires considerably more effort than simply looking up a price.

The valuation methodologies that can be used in valuing OTC derivatives have their roots in techniques that are well known to economists. In many respects, they follow well trodden paths. However, the complexity of some OTC derivative structures and a lack of data in some instances may require one to make choices when employing the standard valuation techniques. These choices may relate to such issues as the use of econometric techniques to extrapolate forward curves or other market-based data, the choice of an option model, the choice of an assumed distribution in a simulation and many others.

The need to make these types of choices and assumptions does not render a valuation impossible or less credible. However, the credibility of the valuation will depend on the basis for the assumptions and the choices made. The more it can be shown that the principles of transparency, consistency and independence have been adhered to in making these choices and assumptions, the greater the credibility of the valuation.

63

64

UNIDROIT International Institute for the Unification of Private Law

PRINCIPLES ON THE OPERATION OF CLOSE-OUT NETTING PROVISIONS

UNIDROIT

65

Suggested form of citation :

Principles on Close-Out Netting

Published by the

International Institute for the Unification of Private Law (UNIDROIT) Via Panisperna, 28 - 00184 Rome - ITALY

ISBN: 88-86449-26-7 Copyright ďƒ“ UNIDROIT 2013

66

FOREWORD It gives us great pleasure to present, on behalf of the International Institute for the Unification of Private Law (UNIDROIT), the UNIDROIT Principles on the Operation of CloseOut Netting Provisions which were adopted by the UNIDROIT Governing Council at its 92nd session (Rome, 8-10 May 2013).

These Principles are the outcome of two and a half years of intensive research and deliberations. The General Assembly of UNIDROIT at its 67th session (Rome, 1 December 2010) assigned the highest level of priority to the project of the development of an international instrument on netting. Following this decision, a Study Group of experts in the law of international financial markets was set up by UNIDROIT. It met three times between April 2011 and February 2012, to prepare a preliminary set of Draft Principles concerning the enforceability of close-out netting provisions. In order to ensure a balanced approach, UNIDROIT invited renowned experts to participate in this Study Group representing regulatory agencies, international organisations, legal practice and academia from jurisdictions which represent today’s international financial centres as well as developing countries.

At its 91st session (Rome, 7-9 May 2012), the UNIDROIT Governing Council discussed the Draft Principles as prepared by the Study Group and agreed to convene a Committee of governmental experts for further consideration and finalisation of the Draft. The UNIDROIT Committee of governmental experts on the enforceability of close-out netting provisions met for two sessions, in October 2012 and March 2013, approving at its second session a revised set of the Draft Principles on the Operation of Close-out Netting Provisions which was submitted to the Governing Council for adoption the following May. UNIDROIT would like to express its deepest gratitude to the Members of the Study Group and to the delegates to the Committee of governmental experts. Only their outstanding competence and personal commitment made possible the successful preparation of these Principles in such a short v

67

FOREWORD

timespan. Particular mention is deserved by Professor Stanislaw J. Sołtysiński (Poland), Member of the UNIDROIT Governing Council, who acted as chair of the Study Group, and by Ms Maria Rosina Vermaas (South Africa) and Mr Alexandre Pinheiro dos Santos (Brazil), who acted as chair of the Committee of governmental experts at its first and second session respectively.

A special word of thanks goes to Mr Philipp Paech (London School of Economics), who undertook the task of preparing the first study of this project and acted as Rapporteur both for the Study Group and the Committee of governmental experts, to the members of the Secretariat, in particular the Senior Officers Ms Frédérique Mestre and Ms Marina Schneider for the extensive work done in preparing the French version of the Principles, to the Associate Officers Ms Annick Moiteaux and Mr Ole Böger, who temporarily joined the UNIDROIT Secretariat in order to help with the completion of this project, and to Ms Isabelle Dubois and Ms Carla Milani, to whose secretarial support the project is greatly indebted.

Last but by no means least, UNIDROIT would like to express its deepest appreciation to the Bundesverband Deutscher Banken (Berlin) for the generous financial support provided for the research activities related to this project.

José Angelo Estrella Faria Secretary-General

Alberto Mazzoni President

vi

68

THE UNIDROIT GOVERNING COUNCIL (2009-2013)

Berardino LIBONATI † Alberto MAZZONI (since 2011) Chief Michael Kaase AONDOAKKA

President of UNIDROIT Nigeria

Hans-Georg BOLLWEG

Germany

Núria BOUZA VIDAL

Spain

Baiba BROKA

Latvia

Antonio Paulo CACHAPUZ DE MEDEIROS Sergio M. CARBONE

Brazil

Sergiu DELEANU

Romania

Michael B. ELMER

Denmark

Henry D. GABRIEL

United States of America

Ian GOVEY

Australia

Attila HARMATHY

Hungary

Arthur S. HARTKAMP

Netherlands

Monique JAMETTI

Switzerland

Ricardo Luis LORENZETTI

Argentina

LYOU Byung-Hwa

Republic of Korea

MO John Shijian

People’s Republic of China

Didier OPERTTI BADÁN

Uruguay

Kathryn SABO

Canada

Jorge SÁNCHEZ CORDERO DAVILA

Mexico

Rachel SANDBY-THOMAS

United Kingdom

Biswanath B. SEN

India

Stanislaw J. SOŁTYSIŃSKI

Poland

Itsuro TERADA

Japan

Daniel TRICOT

France

Ioannis VOULGARIS

Greece

Italy

vii

69

70

MEMBERS OF THE STUDY GROUP Mr Edosa Kennedy AIGBEKAEN — Director (Legal Services) and Secretary to the Commission, Securities and Exchange Commission, Abuja, (Nigeria) Mr Fernando R. de ALMEIDA PRADO — Partner, Pinheiro Neto Advogados, São Paulo, (Brazil) Mr Carl BJERRE — Professor of Law, School of Law; University of Oregon; - Private Sector Adviser - Uniform Law Commission, Eugene, (United States of America) Mr Diego DEVOS — General Counsel, Bank for International Settlements (BIS), Basel, (Switzerland ) Mr FAN Shilei — International Cooperation Department, National Association of Financial Market Institutional Investors, Beijing, (People’s Republic of China) Mr Ignacio GÓMEZ-SANCHA — Partner, Head of Capital Markets and Regulatory Finance, DLA Piper Spain, Madrid, (Spain) Ms Joyce HANSEN — Deputy General Counsel and Senior Vice President; Federal Reserve Bank of New York, (United States of America) Mr Holger HARTENFELS — Managing Director and Senior Counsel; Legal Regulatory & Governance; Deutsche Bank AG, Frankfurt am Main, (Germany) Mr Hideki KANDA — Professor of Law; University of Tokyo, (Japan) Mr Reginald Chukwudi KARAWUSA — Assistant Director; Head of Enforcement Division; Securities & Exchange Commission, Abuja, (Nigeria) Mr Rostislav KOKOREV — Deputy Director; Department of Innovation Development and Corporate Governance; Ministry of Economic Development of the Russian Federation, Moscow, (Russian Federation) Mr Hans KUHN — Director Legal and Property Services; Swiss National Bank, Zürich, (Switzerland) ix

71

MEMBERS OF THE STUDY GROUP

Ms LIANG Wenwen— Institute of International Law, School of Law, Wuhan University, Luojiashan, Wuhan, (People’s Republic of China) Mr LIU Yin— Associate of Market Discipline Department; National Association of Financial Market Institutional Investors, Beijing, (People’s Republic of China) Mr Klaus LÖBER — Head of Oversight Division; European Central Bank (ECB), Frankfurt am Main, (Germany) Ms Monica MARCUCCI — Legal Services and Law Studies Department; Senior Lawyer; Bank of Italy, Rome, (Italy) Mr Guy MORTON — Senior Partner; Freshfields Bruckhaus Deringer, London, (United Kingdom) Mr Philipp PAECH — Lecturer in Financial Law and Regulation; London School of Economics and Political Science (LSE); Department of Law; London, (United Kingdom),

Rapporteur

Mme Béatrice PASSERA — Adjointe de Direction; Banque de France; Direction des Services Juridiques - Service du Droit Communautaire et des Marchés Financiers, Paris, (France) Mr Stanislaw SOŁTYSIŃSKI — Member of the UNIDROIT Governing Council; SOŁTYSIŃSKI KAWECKI & SZLĘZAK; Of Counsel; Professor of Law, A. Mickiewicz University; Warsaw, (Poland), Chairman

Observers Mr Rogier WEZENBEEK — Principal Administrator; DirectorateGeneral for the Internal Market and Services; European Commission (EC), Brussels, (Belgium) Mr

Christophe BERNASCONI — First Secretary; Hague Conference on Private International Law (HCCH), The Hague, (Netherlands)

Mr Wouter BOSSU — Senior Counsel - Legal Department; International Monetary Fund (IMF), Washington, (United States of America) x

72

MEMBERS OF THE STUDY GROUP

Mr Edward H. MURRAY — Chair of the International Swaps and Derivatives Association Inc. (ISDA) Financial Law Reform Committee; Partner, Allen & Overy LLP, London, (United Kingdom) Mr Peter M. WERNER — Senior Director; International Swaps and Derivatives Association Inc. (ISDA), London, (United Kingdom) Mr José M. GARRIDO — Senior Counsel; The World Bank, Washington, (United States of America)

xi

73

74

TABLE OF CONTENTS

Foreword The UNIDROIT Governing Council (2009-2013) Members of the Study Group

v vii ix

PRINCIPLES ON THE OPERATION OF CLOSE-OUT NETTING PROVISIONS Introduction

1

Principle 1: Scope of the Principles

8

Explanation and commentary

8

Principle 2: Definition of ‘close-out netting provision’

10

Key considerations in respect of this definition Explanation and commentary ‘Close-out netting’ ‘Contractual provision’ ‘Obligations owed by the parties to each other’ ‘Occurrence of an event predefined in the provision’ ‘Reduced to or replaced by a single net obligation’ ‘Payable by one party to the other’ ‘Automatically or at the election of one of the parties’

10 11 11 11 12 14 14 17 18

Principle 3: Definition of ‘eligible party’ and related notions

19

Key considerations in respect of this definition Explanation and commentary Paragraph (1) - ‘Eligible party’ Paragraph (2) - ‘Qualifying financial market participant’ Paragraph (3) - ‘Public authority’ Principle 4: Definition of ‘eligible obligation’ Key considerations in respect of this definition Explanation and commentary

20 21 21 22 24 24 26 28 xiii

75

TABLE OF CONTENTS

General remarks on the scope of application and the minimum harmonisation approach under Principle 4 Paragraph (1) - ‘Eligible obligation’ Paragraph (1)(a)(i) - ‘Derivative instruments’ Paragraph (1)(a)(ii) - ‘Securities financing transactions’ Paragraph (1)(a)(iii) - ‘Title transfer collateral arrangements’ Paragraph (1)(a)(iv) - Contracts for the sale, purchase and delivery of certain assets Paragraph (1)(b) – Surety agreements and other personal security Paragraph (1)(c) – Master-master agreements Paragraph (2)(a) – Extension of the personal scope of application to other parties Paragraph (2)(b) – Extension of the substantive scope of application to other obligations Extension to loans and deposits Principle 5: Formal acts and reporting requirements Key considerations in respect of this Principle Explanation and commentary Paragraph (1)(a) – Formal acts Paragraph (1)(b) – Use of standardised terms of trade associations Paragraph (2) – Reporting requirements Consequences other than the restriction of the operation of a close-out netting provision or obligations covered by that provision: administrative, regulatory or penal sanctions Principle 6: Operation of close-out netting provisions in general Principle 7: Operation of close-out netting provisions in insolvency and resolution Key considerations in respect of Principles 6 and 7 Explanation and commentary to Principles 6 and 7 Systematic structure of Principles 6 and 7 xiv

76

28 29 29 31 32 34 34 35 35 36 37 40 40 41 42 43 44

45 46 46 48 49 49

TABLE OF CONTENTS

Principle 6(1) sentence 1 – Enforceability of a close-out netting provision according to its terms Enforceability of a close-out netting provision within and outside insolvency and resolution Principle 6(1)(a) – Additional enforcement requirements Principle 6(1)(b) – Invalid/unenforceable/ineligible obligation included Principle 7(1)(a) – Stay Principle 7(1)(b) – Cherry picking Principle 7(1)(c) – No conflict with equal treatment of creditors Principle 7(1)(d) – Suspect periods and zero-hour rules In relation to the close-out netting provision In relation to the obligations covered Zero-hour rules Safeguard against fraud or transactions with knowledge of a pending insolvency proceeding, etc.

Principles 6(2) and 7(2) – Exceptions Principle 8: Resolution of financial institutions Key considerations in respect of this Principle Explanation and commentary

50 51 54 55 56 56 58 59 59 60 61 61 62 63 63 65

xv

77

78

PRINCIPLES ON THE OPERATION OF CLOSE-OUT NETTING PROVISIONS Introduction 1.

Financial institutions and other financial market participants in their daily operations use a number of mechanisms designed to reduce their risk exposure. Amongst other things, they provide each other with security or collateral. In addition, they may agree that close-out netting will apply to the contracts into which they enter with each other. Both mechanisms, security/collateral on the one hand and close-out netting on the other hand, serve the same purpose, that is, to ensure that one party’s exposure to the other parties’ solvency and to changes in the value of the relevant assets is kept at manageable levels. Both mechanisms are capable of independently mitigating counterparty risk as well as market risk.

However, in practice, their functions are intimately linked: where collateral and netting mechanisms are used cumulatively, netting reduces exposure in the sense that much less collateral has to be put up. Taken together, security/collateral and close-out netting are one of the primary tools of risk management in the financial market.

2.

The notion of close-out netting is a relatively new addition to the legal terminology and it is not particularly well-defined.

These Principles follow a functional understanding of the term ‘close-out netting provision’ (see Principle 2 ), the main elements of which are as follows: a close-out netting mechanism comes into operation either by means of a declaration by one of the parties when a pre-defined event occurs, in particular default or insolvency of its counterparty, or it is triggered automatically when such an event occurs. The mechanism extends to a number, often hundreds, of outstanding transactions between the parties that are contractually included in a netting provision. Once the close-out netting mechanism is triggered, whether automatically or by means of a declaration by one party, generally all transactions that are covered by the close-out netting provision are terminated and a value is determined for each under a pre-defined valuation mechanism, which may also take into account, inter alia, the identity and credit standing of the party responsible for

79

2

PRINCIPLES ON CLOSE-OUT NETTING

this determination and any existing credit support and other material terms of the parties’ agreement. The sum value of all such transactions is then aggregated, resulting in a single net payment obligation. The net obligation remains the only obligation to be settled and is generally due immediately after being determined even though no debts may have been due and payable under the transactions covered by the close-out netting provision prior to the operation of the close-out netting mechanism.

3.

Broadly speaking, close-out netting is often understood as resembling the classical concept of set-off applied upon default or insolvency of one of the parties. Traditionally, the concept of setoff applies only to parties with mutual debts of the same kind that are already due and payable, and that are legally distinct. Whether set-off occurs by contract, by unilateral declaration by one party or by operation of law, the parties’ existing debts are set off against each other, such that the party with the smaller debt owes nothing, and the party with the larger debt owes only the difference between the two obligations. While overlap can occur between the concepts of set-off and close-out netting, these legal institutions are neither functionally nor conceptually identical and the latter mechanism encompasses additional elements, providing, for instance, for the netting of obligations not yet payable. Set-off as such is not addressed by the Principles.

However, set-off falls within the scope of the Principles where the parties to a close-out netting provision have agreed within that provision that their mutual obligations should be set off, or where the applicable principles of law provide for a set-off as regards the aggregation element of a close-out netting provision (see infra, paragraph 36). In both situations, the protection of the operation of the close-out netting provision afforded by the Principles is applicable.

4.

Close-out netting provisions are widely used in the financial market by private sector entities, in particular banks, but also private non-financial institutions. In the public sector, entities such as, especially, central banks and supranational financial institutions such as development banks make use of netting provisions. Close-out netting is typically applied to transactions such as derivatives, repurchase and securities lending agreements, and other kinds of transaction that tend to carry a high counterparty and/or market risk.

80

INTRODUCTION

3

Regulatory authorities (most recently, the Financial Stability Board and the Cross-border Bank Resolution Group of the Basel Committee on Banking Supervision) strongly encourage the use of such close-out netting provisions (alongside collateral) because of their beneficial effects on the stability of the financial system.1 The reason is that if, in the event of default, the counterparty market participants were to be required to calculate their claims on a gross basis instead of being creditors for the net amount only, the non-defaulting party might be exposed to levels of credit risk and market risk that are difficult to estimate and manage. The situation would be further exacerbated by the fact that there may be rapid changes in market values for the relevant types of transaction as well as uncertainty as to the risk of repudiation of contracts during the insolvency proceeding, against which the non-defaulting party might not be able to protect itself whilst being unable to terminate the contracts and re-hedge the position on the basis of its termination rights under the close-out netting provision.

Consequently, the operation of close-out netting provisions reduces the risk that the inability of one market participant to meet its obligations creates or increases financial difficulties for counterparties which could lead to a chain of failures of market participants (contagion effect or systemic risk).

5.

6.

These beneficial effects can be particularly evident in the event of the insolvency of a party. In that case, the use of closeout netting assumes that the legal effects stipulated to that end by the parties (the close-out netting provision) will be recognised by and be enforceable under the applicable insolvency law.

However, the current situation is that, even if some 40 jurisdictions2 recognise netting in insolvency, the extent to which

Bank for International Settlements/Basel Committee on Banking Supervision, Report and Recommendations of the Cross-border Bank Resolution Group, March 2010, Recommendation 8, p.

36 et seq.; http://www.bis.org/publ/bcbs169.pdf; Financial Stability Board, Key Attributes of Effective Resolution Regimes for Financial Institutions, October 2011, para. 4.1., http://www.financialstabilityboard.org/publications/r_111104cc.pdf.

2 According to a list regularly updated by the International Swaps and Derivatives Association (ISDA), the following jurisdictions have accommodated close-out netting in their law: Andorra, Anguilla, Australia, Austria, Belgium Brazil, British Virgin Islands, Canada, Colombia, Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Ireland, Israel, Italy, Japan, 1

81

4

PRINCIPLES ON CLOSE-OUT NETTING

they do so and the scope and legal effects of close-out netting provisions differ significantly. Furthermore, some jurisdictions do not clearly recognise netting, and the legal practice in such jurisdictions often resorts to the principles governing set-off, failing to recognise the fundamental differences between the two mechanisms. This global ’patchwork’ is unsatisfactory in crossjurisdictional situations, since it exposes the financial market participants’ risk management to unnecessary legal uncertainty and may even jeopardise it.3 7.

Close-out netting provisions have been effectively used as a risk mitigation tool in a bilateral context. However, there is some concern that, in times of pressure on financial markets, the enforceability of close-out netting provisions might increase the risk that counterparties of a distressed financial institution rush to exercise termination rights and close out their positions, thereby exacerbating systemic risk. While underlining the usefulness of close-out netting in general, regulatory authorities have contemplated the need for a brief stay on the netting mechanism in certain situations (especially in the context of resolution regimes) affecting a financial institution, so as to allow the resolution authority the time needed to decide whether and how to resolve an ailing financial institution in an orderly fashion so as to mitigate risks to financial stability.

8.

The emerging international regulatory consensus regarding the interplay between close-out netting and bank resolution is set out in the Financial Stability Board’s Key Attributes of Effective Resolution Regimes for Financial Institutions,4 which provide guidance as to how the regulatory intervention should be reconciled with the need of financial institutions and its regulators to rely on the enforceability of close-out netting for risk management and mitigation purposes. However, this newly Luxembourg, Malta, Mauritius, Mexico, New Zealand, Norway, Peru, Poland, Portugal, Romania, Russia, Slovakia, Slovenia, South Africa, South Korea, Spain, Sweden, Switzerland, United Kingdom and the United States. According to the same list, netting-friendly legislation is under consideration in the following jurisdictions: Argentina, Chile, Pakistan and Seychelles.

Source: http://www.isda.org/docproj/stat_of_net_leg.html.

3 Cf. for a detailed analysis UNIDROIT 2011, Study LXXVIII C - Doc. 2, 1st Part, in particular pp. 32 et seq.

4 Financial Stability Board, Key Attributes of Effective Resolution Regimes for Financial Institutions, October 2011, part 4, in particular para. 4.3.

82

INTRODUCTION

5

developing regulatory approach has to deal with a patchwork where the relevant legal mechanisms in which close-out netting is embedded are not compatible or comparable across borders.

Therefore, the sensitive connection between, on the one hand, regulatory measures such as stays on termination or portfolio transfers and, on the other hand, the essential insolvency and commercial law framework might fail in certain cases. Notably, stays might achieve a better cross-jurisdictional effect on the basis of harmonised legal principles. Equally, legal uncertainties arising in the context of asset transfers to domestic or foreign bridge banks can be more effectively mitigated on the basis of a more consistent international picture of the underlying commercial and insolvency law. This situation calls for a more harmonised and streamlined framework regarding close-out netting on which market participants and regulatory authorities can rely across all financial markets.5 9.

First steps have already been taken towards an international consensus on the legal cornerstones regarding enforceability of close-out netting provisions. The Geneva Securities Convention sets out an optional framework for the protection of collateral transactions. This protection extends to close-out netting provisions provided they are concluded as part of a collateral transaction. The Convention therefore contains a definition of close-out netting and a key rule on enforceability.6 10. Furthermore, close-out netting has also been recognised in the work of other international Organisations. Notably, the UNCITRAL Legislative Guide on Insolvency Law refers to the enforceability of close-out netting as a feature to be considered when designing insolvency law, and advises that close-out netting should be allowed under the applicable insolvency procedure.7 Moreover, the European Union Member States have implemented

5

Cf. for a detailed analysis UNIDROIT 2011, Study LXXVIII C - Doc. 2, 2nd Part, in particular pp. 68 et seq.

6 UNIDROIT Convention on Substantive Rules for Intermediated Securities, adopted in Geneva on 9 October 2009; in particular Article 31(3)(j) and Article 33(3).

7 UNCITRAL, 2004 Legislative Guide on Insolvency, Recommendations 7(g) and 101-107.

83

6

PRINCIPLES ON CLOSE-OUT NETTING

a partly harmonised legal framework for close-out netting provisions.8 11. The aim of these Principles is to provide detailed guidance to national legislators of implementing States seeking to revise or introduce national legislation relevant to the functioning of closeout netting. The Principles are designed to improve the enforceability of close-out netting, especially in crossjurisdictional situations, in order to provide a sound basis, in commercial and insolvency law terms, for risk management and mitigation by financial institutions and for the application of regulatory policies in the international context. The Principles are addressed to legislators and policy-makers of implementing States.

They are not intended to be chosen by private parties to govern a close-out netting provision.

12. There are several situations in which the law of a State which has implemented the Principles may govern the operation of a close-out netting provision. Generally, the courts of an implementing State might apply their own law as the law of the forum, especially in insolvency proceedings opened under this law. Furthermore, the protection of the operation of close-out netting provisions afforded under the Principles might be applicable even in proceedings conducted under the law of a State other than an implementing State. This may be the case, for instance, where a rule of private international law or international insolvency law of the forum leads to the application of the law of an implementing State, or the parties have chosen the law of an implementing State and the choice of law is upheld by the competent court. However, the Principles do not attempt to propose such rules to determine the law applicable.

13. Implementing States are free to choose any suitable method for implementing the Principles, be it the enactment of specific legislation, the application of general principles of law or the removal of restrictions to the enforceability of close-out netting provisions in the context of insolvency proceedings.

Implementing States are likewise free to extend their protection of the operation of close-out netting provisions beyond the minimum

8

Cf. UNIDROIT 2011, Study LXXVIII C - Doc. 2, p. 24-25 for a brief description of the various rules in place.

84

INTRODUCTION

7

scope of harmonisation as provided for under the Principles and may also in this regard choose any of the methods referred to in the preceding sentence (see infra, paragraph 16).

85

8

PRINCIPLES ON CLOSE-OUT NETTING

PRINCIPLE 1 Scope of the Principles (1) These Principles deal with the operation of close-out netting provisions that are entered into by eligible parties in respect of eligible obligations.

(2) Except as otherwise expressly indicated in these Principles, the term ‘operation’ encompasses the creation, validity, enforceability, effectiveness against third parties and admissibility in evidence of a close-out netting provision.

Explanation and commentary 14.

For the determination of the scope of the Principles,

Principle 1(1) refers to three core terms the content and meaning

of which are defined in other provisions of the Principles: ‘closeout netting provision’ (Principle 2), ‘eligible parties’ (Principle 3(1)) and ‘eligible obligations’ (Principle 4(1)).

15. Within the scope of application so defined, the Principles follow a broad and functional approach concerning the various aspects of the taking of effect of a close-out netting provision, as is evident from the broad scope of the term ‘operation’ as defined in Principle 1(2). All aspects, from creation and validity, to effectiveness against third parties, admissibility in evidence and enforceability, especially in insolvency, are covered by the Principles. Where the Principles seek to ensure the enforceability of close-out netting provisions, this concerns all these aspects, which can be summed up as the operation of the close-out netting provision. The term ‘operation’ of a close-out netting provision is therefore used throughout the Principles as a functional, shorthand reference to the various aspects described above and replaces the need to refer to the more specific terms which are not necessarily defined sufficiently precisely in an international context.

16. Concerning the limitations of the scope of the Principles, it should be noted that the Principles propose a minimum scope of harmonisation, where the enforceability of close-out netting provisions within the scope of the Principles should be ensured, whereas they acknowledge, without precluding further

86

PRINCIPLE 1 – SCOPE OF THE PRINCIPLES

9

harmonisation, that beyond this minimum scope, each implementing State may regard the enforceability of close-out netting provisions as an issue of its public policy. The Principles do not impose a maximum scope of harmonisation and therefore do not prevent or restrict an implementing State from having a legal framework (whether by means of legislation or otherwise) that goes beyond the Principles in the recognition of close-out netting provisions and the protection of their operation.

17. In view of the preceding paragraph, the Principles allow for a certain measure of discretion (see Principle 4(2) ) in determining the scope of netting-specific legislation or of the application of other general rules protecting the operation of close-out netting provisions beyond the minimum scope of harmonisation suggested by the Principles. Factors that each implementing State may wish to take into consideration in the exercise of its discretion notably include: -

the importance of protection against systemic risk and the relevance for the functioning of the respective markets of the use of close-out netting provisions as an instrument of counterparty risk management;

-

the relationship between close-out netting and the system of insolvency priorities in the implementing State in general;

-

the need for predictability of results and certainty in commercial transactions;

-

the general principle that the law should not treat similar situations differently without justification, and the specific principles against discrimination between domestic and foreign creditors in insolvency (see UNCITRAL Model Law on Cross-Border Insolvency, Art. 13); and

-

whether this discretion is exercised for the purposes of the application of all or only some of the provisions of the Principles, or for their application in general or in situations outside insolvency only.

18. The Principles leave it up to implementing States to consider the desirability of extending the scope of application of

87

10

PRINCIPLES ON CLOSE-OUT NETTING

their laws and regulations protecting the operation of a close-out netting provision to any specific parties or obligations beyond the minimum scope of harmonisation under the Principles. Likewise, the Principles provide guidance to national legislators only with regard to the operation of close-out netting provisions falling under the definition in Principle 2. This does not encompass certain other clauses that are often included in master agreements in relation to a close-out netting provision but do not strictly speaking form part of the latter (see infra, paragraphs 41 and 43).

PRINCIPLE 2 Definition of ‘close-out netting provision’ ‘Close-out netting provision’ means a contractual provision on the basis of which, upon the occurrence of an event predefined in the provision in relation to a party to the contract, the obligations owed by the parties to each other that are covered by the provision, whether or not they are at that time due and payable, are automatically or at the election of one of the parties reduced to or replaced by a single net obligation, whether by way of novation, termination or otherwise, representing the aggregate value of the combined obligations, which is thereupon due and payable by one party to the other.

Key considerations in respect of this definition  The definition of ‘close-out netting provision’ should be broad so as to encompass different types of provision which achieve a functionally identical result.  It should be neutral with respect to the various legal methods available to achieve the result that may exist in different jurisdictions and in different standard market contracts.  The definition exclusively relates to contractual close-out netting, including bilateral netting within the context of central clearing systems. It does not address close-out

88

PRINCIPLE 2 – DEFINITION OF ‘CLOSE-OUT NETTING PROVISION‘

11

netting to the extent that its functionalities are achieved under statutory provisions, nor does it address truly multilateral netting.

Explanation and commentary ‘Close-out netting’ 19. Close-out netting is best described in functional terms, i.e., by reference to a result. The process, in practical terms, is the following. A bundle of transactions with mutual obligations between the parties is contractually covered by a netting provision. Upon the occurrence of a predefined event, all outstanding obligations covered by the netting provision cease to be treated individually and their aggregate value is computed so as to result in a single net payment obligation. This obligation is owed by the party which is ‘out of the money’ to the party which is ‘in the money’. This obligation remains the only obligation (which, depending upon the terms of the relevant provision, may include incidental fees, costs or other expenses) to be settled and is generally due and payable shortly after being determined.

‘Contractual provision’ 20. This definition covers contractual close-out netting, as opposed to statutory rules that may achieve an identical or similar result.

21. In practice, a clause allowing for close-out netting between the parties may be included in standard master documentation, be part of a tailor-made framework agreement, or be an entirely selfstanding agreement. The Principles therefore refer to the term ‘close-out netting provision’ rather than to ‘arrangement’ or ‘agreement’, so as to encompass these various possibilities.

However, the term ‘close-out netting provision’ covers only those parts of an agreement that actually implement the close-out netting mechanism itself, and nothing else. Definitions, schedules and annexes that the parties may have related to their agreement are covered only to the extent that their content is necessary for the proper operation of the close-out netting mechanism. For instance, certain clauses (walk-away clauses, wait-and-see periods)

89

12

PRINCIPLES ON CLOSE-OUT NETTING

that are often used in connection with close-out netting provisions are not covered by the Principles (see infra, paragraphs 41 and 43).

22. Where the result of close-out netting is achieved through a combination of statutory rules and contractual provisions (e.g., the right to terminate is statutory, while acceleration, valuation and aggregation are arranged for contractually), the Principles only cover the contractual part. To the extent that the parties have relied upon the application of statutory rules, the operation of the close-out netting mechanism under those statutory rules is not an issue of the enforcement of the contractual agreement that would be covered by the scope of the Principles.

23. The internal rules of clearing, settlement and payment systems, as well as central counterparties are also contemplated by this definition. Despite the fact that they are usually approved by the relevant regulatory authority, the character of the relationship between the system and its participants is, or in any case is treated by this instrument as, one of commercial law (membership agreement, by-laws) as regards the treatment of the assets to be settled in the system. Thus, the close-out netting operation takes effect ‘on the basis of’ a contractual provision as envisaged by this definition.

24. The definition also covers contractual provisions that are not contained in the clauses of a single agreement, but in several interrelated arrangements, especially master-master agreements (to the extent that the underlying obligations covered by the various master agreements are eligible obligations under Principle 4, see infra, paragraph 81).

‘Obligations owed by the parties to each other’ 25. Transactions concluded between two parties may be settled either bilaterally, between the parties themselves, or through a central entity interposed between the parties. Close-out netting is equally important in both scenarios.

26. Bilateral settlement between the parties is the standard case and covered by the Principles.

27. The Principles also cover ‘central clearing’ mechanisms which are ultimately also built on bilateral relationships. Central

90

PRINCIPLE 2 – DEFINITION OF ‘CLOSE-OUT NETTING PROVISION‘

13

clearing is used as a collective term for the functionalities of central counterparties, net payment systems and certain clearing and settlement systems. Central clearing applies by virtue of contractual agreements between market participants or as a legal requirement. The arrangement usually works by interposing a central entity between the parties to every transaction that is concluded between the members of a clearing system, so that the central entity becomes ‘buyer to every seller and seller to every buyer’. In other words, the bilateral settlement obligations that exist between the system’s participants are entirely replaced by bilateral obligations between each participant and the central clearing entity. As a consequence, the net risk exposure is calculated on a bilateral basis, so that each participant’s exposure exists exclusively against the central entity. Thus, given that, from a legal point of view, central clearing breaks down into strictly bilateral relationships, considerations in respect of bilateral closeout netting generally apply to central clearing. This applies both inside and outside the insolvency of the participants and the system. Therefore, legal certainty also requires that the conversion of the original contractual relationships between the central clearing participants into bilateral relationships between each participant and the central clearing entity is insolvency-proof.

28. Truly multilateral close-out netting is not as common as bilateral netting. Under a multilateral close-out netting scheme, more than two parties compute their mutual exposure on a multilateral basis, employing functionalities similar to those used in close-out netting. A mechanism similar in concept to multilateral netting is sometimes used as a tool to circumscribe the exposure of one market participant vis-à-vis several other market participants, typically a bank managing its risk exposure under a single netting provision against several entities belonging to the same group of companies (hence this form of netting is also called 'cross-affiliate netting'). The recognition of a multilateral netting provision by the applicable insolvency law depends in part on whether the law is able to accommodate the lack of mutuality of the relevant obligations or on whether the ‘mutuality’ created through cross-guarantees, cross-collateralisation agreements or similar arrangements is recognised. Truly multilateral close-out netting is not covered by the definition in Principle 2.

91

14

PRINCIPLES ON CLOSE-OUT NETTING

‘Occurrence of an event predefined in the provision’ 29. The event that triggers the application of the netting provision (the occurrence of the event that is ‘predefined in the provision’) is commonly referred to in the relevant documentation as the ‘termination event’, ‘enforcement event’, ‘specified event’, or ‘default event’. Close-out netting can occur both in situations where both parties are solvent and in the event of the insolvency of either, since it is the parties to the netting provision themselves that determine the trigger for the operation of the mechanism.

This event may consist, for example, in one of the parties defaulting on one or more of its obligations, or in its filing for insolvency, in the appointment of a State administrator or a similar intervention by the public authorities, or in the opening of an insolvency proceeding or an administration or restructuring procedure. Often, parties additionally agree to include external circumstances as termination events, such as the objective impossibility of performing an obligation under one of the transactions due to a circumstance beyond the control of a party, or the downgrading of one of the parties’ credit rating following its merger with another company.

30. It is worth noting that the event triggering termination is determined, in certain jurisdictions, under the relevant legislation itself. In particular, the insolvency of one of the parties may lead to the termination of all open transactions by operation of the statutory law. Parties may supplement this statutory consequence of the termination event with additional contractual rules providing for other elements needed to achieve the result of closeout netting (cf. supra, paragraph 22). Such arrangements are likewise envisaged by the present definition.

‘Reduced to or replaced by a single net obligation’ 31. A close-out netting mechanism is commonly understood as resulting in a single payment obligation owed by the party that is ‘out of the money’ to the party that is ‘in the money’. However, a number of different functional steps can be used to achieve this result, and these can potentially be based on a number of differing legal concepts.

92

PRINCIPLE 2 – DEFINITION OF ‘CLOSE-OUT NETTING PROVISION‘

15

32. A netting mechanism generally involves several or all of the following steps: (i) termination of the transactions, (ii) acceleration of obligations, (iii) valuation of the transactions, and, (iv) aggregation to result in an overall net amount. The order of termination, acceleration, aggregation and valuation can vary according to the actual netting provisions. Not all netting provisions need all of these steps to achieve the functional result of close-out netting. Which elements are needed and used depends, rather, on the design of the relevant provision and the boundaries under the applicable law. Examples:  Termination of each transaction; valuation of each obligation; aggregation of all values to form one net payment obligation.  Acceleration of each transaction, valuation of each obligation, aggregation of all values to form one net payment obligation.  Termination of each transaction; valuation of each obligation; aggregation of all values to form one net payment obligation; acceleration of the net obligation.  Termination of each transaction; valuation of each obligation; creation of a new (immediately due and payable) payment obligation representing the overall value.  Etc. 33. These functional steps merely describe what happens in practical terms. The relevant close-out netting provision in combination with the applicable law needs to provide the necessary legal concepts, since the result (a single net payment obligation) is first and foremost a legal one. The legal concepts and terminology that underlie these steps differ, depending on the design of the netting provision and on the law applicable to it.

34. Termination is a term used to express the functional result of the relevant open transactions being put to an end. National laws achieve this result by legal mechanisms called cancellation, closeout, rescission, termination, etc.

35. Acceleration is a term used to express the concept that an obligation becomes due and payable before the contractually

93

16

PRINCIPLES ON CLOSE-OUT NETTING

agreed date; there might be other legal concepts and terms to achieve an identical functional result such as the replacement of the original and as yet unmatured obligation with a new obligation (‘novation’). It should be noted that, while the legal technique of novation is covered by the Principles as one of the methods that may be used to replace the original obligations with the single net obligation under the close-out netting provision, the type of transaction commonly referred to as ‘novation netting’ (or ‘settlement netting’) is not addressed by the Principles.

‘Transformation’ is another term that might be used in this context.

36. The aggregation element collapses all relevant transactions or the value resulting therefrom so as to produce a single obligation. This is functionally the same as a set-off of all due and payable obligations in the classical sense. Also, novation (i.e., the parties’ agreement that, upon termination of all open transactions, a new obligation arises representing the relevant aggregate value) is a suitable concept to achieve the effect of aggregation.

37. The valuation of the terminated transactions or the entire (aggregate) contractual relationship generally seeks to establish compensation for the party that was ‘in the money’. This means that the single net obligation is intended to ‘represent the value of the combined obligations’ that were covered by the close-out netting provision. Whereas the original obligations may include obligations for the payment of money or indeed any other performance (such as the delivery of securities or commodities), the resulting single net obligation must represent the ‘value of the combined obligations’ as an obligation for the payment of money or having a monetary value.

38. The Principles do not establish any specific requirements for the process of valuation. Usually (but not necessarily), valuation is effected by the non-defaulting party under a mechanism which has been pre-defined in the agreement. The parties are free to define the valuation mechanism and may use concepts such as replacement or market value or any other mutually accepted method of valuation.

39. As regards the substantive result of the valuation, Principle 2 requires that the resulting net obligation must ‘represent the aggregate value of the combined obligations’. No specific

94

PRINCIPLE 2 – DEFINITION OF ‘CLOSE-OUT NETTING PROVISION‘

17

additional criteria are laid down in the definition of the term ‘close-out netting provision’ and the implementing States retain some discretion as to the precise conditions to be fulfilled by a valuation mechanism in order to satisfy this requirement. Some legal systems, for instance, require an element of commercial reasonableness for the enforceability of close-out netting provisions. However, implementing States that currently do not apply such a commercial reasonableness requirement when determining whether the resulting net obligation can be regarded as representing the value of the combined obligations are not bound by the Principles to introduce such a criterion. Moreover, implementing States obviously remain free to recognise the enforceability of close-out netting provisions even if these are not covered by the definition of Principle 2.

‘Payable by one party to the other’ 40. Where close-out netting occurs in the context of the insolvency of one of the parties, and the net amount is positive for the other party, that party is paid from the insolvency estate and may therefore fail to recover some or the entirety of its claim, if unsecured. In the amount of this net sum, the position of the solvent party vis-à-vis the insolvent estate is no better than that of any other party: it needs to be secured in order to be certain of payment and the same requirements apply regarding the necessary proof of the claim. Where the net amount is positive for the insolvent party, as a rule the other party must pay the insolvency estate.

41. However, parties may have agreed on a clause that allows a non-defaulting party which is ‘out of the money’ to refuse payment to the defaulting party (‘walk-away clause’). Such clauses may have various justifications, for example, that a defaulting party should not benefit from its own default. However, not all jurisdictions permit such clauses because of their potential effect on systemic stability and for reasons of equal treatment of creditors. There is a regulatory debate as to whether they should be valid in the event of insolvency of the defaulting party. In order to take advantage of close-out netting for purposes of calculating required capital under the regulatory capital adequacy standards of the Basel Committee on Banking Supervision, parties may not

95

18

PRINCIPLES ON CLOSE-OUT NETTING

rely on a walk-away clause.9 This is a regulatory and policy decision on which the Principles do not reflect a position. Walkaway clauses are not immediately linked to the enforceability of close-out netting provisions, since they do not affect the steps of termination, acceleration, valuation and aggregation under a close-out netting provision as defined in Principle 2. Rather, they constitute a type of clause which merely affects the payability of the obligations covered by the close-out netting provision or of the single net obligation resulting from its operation. Thus, where parties have concluded a master agreement that contains a closeout netting provision and a walk-away clause, the walk-away clause falls outside the scope of the Principles, while the protection of the operation of the close-out netting provision as such under the Principles is not affected.

‘Automatically or at the election of one of the parties’ 42. Depending on the specific contractual agreement, close-out netting either occurs automatically, by operation of the contractual agreement (‘automatic termination’, which is not allowed in a number of jurisdictions), or it may occur at the discretion of the party which is not the party to which the predefined event relates. Close-out netting provisions employing either elective or automatic termination are covered by the Principles.

43. The extent to which the non-defaulting party should be able to suspend or wait for a period of time or for an indefinite period of time to exercise its rights to close-out is currently under discussion among market participants and regulatory authorities, particularly where the defaulting party is in resolution or insolvency. One concern is that a non-defaulting party which is ‘out of the money’ may be entitled, under the terms of the master agreement, not to make any payments to the defaulting party after the occurrence of the event of default even though the nondefaulting party refrains from terminating and exercising close-out.

Thus, the non-defaulting party may through inaction – by

9

Basel Committee on Banking Supervision, International Convergence of Capital Measurement and Capital Standards: A Revised Framework (2006) Annex 4 para.96(iii), available under http://www.bis.org/publ/bcbs128.pdf.

96

PRINCIPLE 3 – DEFINITION OF ‘ELIGIBLE PARTY‘ AND RELATED NOTIONS 19

refraining from the exercise of its rights under the close-out netting provision - avoid having to perform under its obligations to the defaulting party. Several different courts with jurisdiction over recent cross-border insolvency proceedings of large financial institutions have come to different conclusions in this respect. The Principles do not take a stand concerning these issues. The question of whether or not to impose a time limit for the exercise of the right to termination and close-out is regarded as a policy issue of the implementing State. Similarly, any suspension of the non-defaulting party’s payment obligations after the termination event has occurred, but before the actual termination itself, does not affect the operation of the close-out netting provision as defined in Principle 2 and is therefore an issue that is not addressed by the Principles.

PRINCIPLE 3 Definition of ‘eligible party‘ and related notions (1) ‘Eligible party’ means any person or entity, other than a natural person who is acting primarily for personal, family or household purposes, and includes a partnership, unincorporated association or other body of persons.

(2) ‘Qualifying financial means any of the following:

market

participant’

(a) a bank, investment firm, professional market maker in financial instruments or other financial institution which (in each case) is subject to regulation or prudential supervision; (b) an insurance or reinsurance company; (c) an undertaking for collective investment or an investment fund; (d) a central counterparty or a payment, clearing or settlement system, or the operator of such a system which (in each case) is subject to regulation, oversight or prudential supervision;

97

20

PRINCIPLES ON CLOSE-OUT NETTING

(e) a corporation or other entity that, according to criteria determined by the implementing State, is authorised or supervised as an important participant in the implementing State’s markets in contracts giving rise to eligible obligations.

(3) ‘Public authority’ means any of the following: (a) a governmental or other public entity; (b) a central bank; (c) the Bank for International Settlements, a multilateral development bank, the International Monetary Fund or any similar entity.

Key considerations in respect of this definition  The definition of ‘eligible parties’ determines and restricts the scope of the Principles, in conjunction with the definition of ‘eligible obligation’. Therefore, the application of the Principles to a legal relationship between two parties depends on whether both are eligible parties, and whether the relevant legal relationship represents an eligible obligation.  The minimum scope of harmonisation as suggested by the Principles reflects different types of risk that are relevant in the various situations in which close-out netting provisions can be applied. The different categories of eligible parties as defined in Principle 3 are the core factor for this distinction as applied under the minimum harmonisation approach of the Principles.  Principle 3 contains separate definitions for two subgroups of eligible parties both involving increased risk.

Paragraph (2) concerns ‘qualifying financial market participants’, i.e., parties where the enforceability or nonenforceability of close-out netting provisions may give rise to issues of systemic risk; paragraph (3) covers ‘public authorities’, where the enforceability or non-enforceability of close-out netting provisions may affect issues of public interest in the general sense, systemic risk or other. Under

98

PRINCIPLE 3 – DEFINITION OF ‘ELIGIBLE PARTY‘ AND RELATED NOTIONS 21

the minimum harmonisation approach taken by the Principles (see Principle 4(1) ), the scope of application of the Principles requires that at least one of the parties to the contracts from which the obligations covered by close-out netting provision arise must belong to one of these subgroups.  Apart from these sub-groups, the term ‘eligible parties’ is the general criterion for determining the personal scope of application of the Principles. It is defined in a broad and comprehensive manner, but acknowledges the need to protect natural persons acting primarily for personal, family or household purposes.  Implementing States are free to extend the personal scope of application of the Principles beyond the minimum scope of harmonisation as suggested by the Principles; this discretion is dealt with in more detail in the commentary to Principle 4, see infra, paragraphs 56 et seq.

Explanation and commentary Paragraph (1) – ‘Eligible party’ 44. The term ‘eligible party’ is defined in the broadest possible terms, reflecting the reasoning that it is well-nigh impossible properly to classify the different types of actor that might make use of close-out netting provisions in the financial or other markets.

45. The personal scope of application of the Principles covers both parties with legal personality and other entities, including trusts or partnerships, unincorporated associations or other bodies of persons, the latter being expressly referred to in paragraph (1).

All legal persons are covered by the definition in paragraph (1) and thus, reflecting the general systematic concept of Principles 3 and 4, ‘qualifying financial market participants’ and ‘public authorities’ as defined in paragraphs (2) and (3) are as sub-groups of ‘eligible parties’ always also covered by paragraph (1).

46. The only exception provided for in paragraph (1) concerns natural persons acting primarily for personal, family or household purposes. This exception reflects the concerns of many legal systems regarding specific protection for consumers in connection

99

22

PRINCIPLES ON CLOSE-OUT NETTING

with the operation of close-out netting provisions. Since the scope of the legal category of ‘consumer’ varies as between the different legal systems, however, the text of Principle 3 intentionally avoids the use of this term and the scope of the exception is defined in functional terms by reference to the primary purposes of the relevant transaction.

47. It should be noted that under the minimum harmonisation approach as suggested by the Principles, it is not sufficient for both parties to be eligible parties. Instead, at least one of the parties to the close-out netting provision must be a qualifying financial market participant or a public authority as defined in paragraphs (2) and (3). On the other hand, Principle 4(2) allows implementing States to extend the scope of their protection of the operation of close-out netting provisions beyond the minimum scope of harmonisation as suggested by the Principles. This may include the extension of the personal scope of application to situations where neither party to the close-out netting provision is a qualifying financial market participant or a public authority as defined in paragraphs (2) and (3) or even to parties not covered by paragraph (1) - see the commentary to Principle 4, infra, paragraphs 82 et seq.

Paragraph (2) – ‘Qualifying financial market participant’ 48. The term ‘qualifying financial market participant’ encompasses all those parties active on the financial markets in relation to which the non-enforceability of a close-out netting provision might not only raise concerns regarding the individual credit risk management, but also issues of systemic risk.

49. In the interests of legal certainty as to the personal scope of application of the Principles, the definition in paragraph (2) is not based upon a direct reference to a requirement of systemic risk, but instead, in sub-paragraphs (a) to (d), contains an enumeration of certain parties that are defined as being covered by the term ‘qualifying financial market participant’.

50. Sub-paragraph (e) introduces an element of discretion for the implementing State into the definition of the term ‘qualifying financial market participant’ by extending the definition to corporations or other entities that, according to criteria

100

PRINCIPLE 3 – DEFINITION OF ‘ELIGIBLE PARTY‘ AND RELATED NOTIONS 23

determined by the implementing State, are authorised or supervised as important participants in the implementing State’s markets with regard to contracts giving rise to eligible obligations.

The purpose of this provision is to include also certain nonfinancial corporate parties within the scope of the term ‘qualifying financial market participant’ under the law of each implementing State. The reasoning is that in some cases at least, certain parties participate in the relevant markets and make use of close-out netting provisions in a manner equivalent to financial entities and raising similar issues of systemic risk.

51. Sub-paragraph (e) does not specify any requirements as to the conditions, manner or content of the authorisation or supervision of such an important participant in the relevant markets. Instead, the intention is to allow implementing States to take into account the characteristics of their domestic markets in determining which entities should be allowed to take advantage of close-out netting provisions. Just as implementing States are free to extend the scope of application of the Principles under their own laws beyond the minimum scope of harmonisation suggested by the Principles by whatever measure, they may under sub-paragraph (e) follow both a general approach, authorising general classes of parties as being eligible for closeout netting (as a ‘qualifying financial market participant’ in the sense of this provision), or an entity-specific approach, authorising or supervising individual entities. Also, the decision as to whether ongoing supervision is required or whether a – general or specific – authorisation is sufficient, is left to each implementing State.

52. The implementing State may use various criteria in deciding whether or not to include certain entities within the scope of the parties covered by sub-paragraph (e), such as, for instance, the size of the corporation or entity (measured by reference to assets, revenues, or otherwise) or the scale of its dealings in the relevant markets. Establishing such uniform criteria in these Principles for all States would not be appropriate nor indeed feasible. However, the criteria chosen by an implementing State should offer all guarantees of legal certainty and predictability to counterparties while ensuring that systemicallyimportant entities are not excluded (see the general reasoning supra, paragraph 17).

101

24

PRINCIPLES ON CLOSE-OUT NETTING

Paragraph (3) – ‘Public authority’ 53. Paragraph (3) covers ‘public authorities’ as another subgroup of eligible parties. The underlying reasoning is that where one of the parties to a close-out netting provision is a public authority, the enforceability of the close-out netting provision is likely to raise issues of public interest. Often, this public interest might refer to issues of systemic risk; another specific concern might be the issue whether claims for the repayment of public funds made available to a private market participant can be protected in the event of the insolvency of the latter.

54. The aforementioned specific reasoning is especially evident in relation to the types of public authorities enumerated in subparagraphs (b) and (c), i.e., central banks, the Bank for International Settlements, a multilateral development bank, the International Monetary Fund, or any similar entity.

55. Sub-paragraph (a) extends the definition of the term ‘public authority’ to governmental or other public entities. These broad terms should be interpreted in the light of the objectives of paragraph (3), also taking into consideration the nature of the other instances of public authorities covered in sub-paragraphs (b) and (c). Thus, the term ‘public entity’ encompasses, at least in certain cases and depending upon the local framework of the law of the implementing State, private entities providing public services. However, minor public involvement in an entity would not necessarily be sufficient to regard that entity as a ‘public entity’.

PRINCIPLE 4 Definition of ‘eligible obligation‘ (1) ‘Eligible obligation’ means: (a) an obligation arising under a contract of any of the following kinds between eligible parties at least one of which is a public authority or a qualifying financial market participant: (i) Derivative instruments, that is to say, options, forwards, futures, swaps, contracts for

102

PRINCIPLE 4 – DEFINITION OF ‘ELIGIBLE OBLIGATION‘

differences and any other transaction in respect of an underlying or reference asset or a reference value that is, or in future becomes, the subject of recurrent contracts in the derivatives markets; (ii) Repurchase agreements, securities lending agreements and any other securities financing transaction, in each case in respect of securities, money market instruments or units in an undertaking for collective investment or an investment fund; (iii) Title transfer collateral arrangements related to eligible obligations; (iv) Contracts for the sale, purchase or delivery of: a. securities; b. money market instruments; c. units in an undertaking for collective investment or an investment fund; d. currency of any country, territory or monetary union; e. gold, silver, platinum, palladium or other precious metals; (b) an obligation of an eligible party (whether by way of surety or as principal debtor) to perform an obligation of another person which is an eligible obligation under sub-paragraph (a); (c) a single net obligation determined under a close-out netting provision entered into by the same parties in respect of obligations under subparagraph (a) or (b).

(2) An implementing State may elect to broaden the scope of paragraph (1) (a) in one or both of the following ways: (a) by providing that it is to extend to obligations arising under contracts between parties

103

25

26

PRINCIPLES ON CLOSE-OUT NETTING

neither of whom is a public authority or a qualifying financial market participant; (b) by providing that it is to extend to obligations not limited to those listed in paragraph (1); subject, in either case, to such limitations or exceptions as the implementing State may specify.

Key considerations in respect of this definition  While Principle 3 defines the notions of ‘eligible party’ and related terms that are relevant for the determination of the personal scope of application, Principle 4 is the core expression of the minimum harmonisation approach as suggested under the Principles. Under Principle 4(1), the scope of application of the Principles is restricted to obligations to which at least one public authority or a qualifying financial market participant is a party.

Principle 4(2), however, recognises that implementing States remain free to extend their protection of the enforceability of close-out netting provisions beyond the minimum scope of harmonisation suggested by the Principles.  Close-out netting would be possible from the perspective of the purely legal mechanisms involved in respect of all mutual contractual relationships the value of which can be expressed in an amount of currency. However, in the event of default of one of the parties, close-out netting offers special treatment of the non-defaulting party in relation to the insolvent’s general creditors. Therefore, under the minimum harmonisation approach suggested by the Principles, there is not only a limitation of the personal scope of application of the Principles as referred to in the preceding paragraph, but the substantive scope of application, i.e., the scope of eligible obligations, is likewise restricted to certain types of obligation. Therefore, there need to be elements justifying a contractual

104

PRINCIPLE 4 – DEFINITION OF ‘ELIGIBLE OBLIGATION‘

relationship being covered by a close-out provision. There are three such elements.

27

netting

 Single relationship: contracts entered into on the understanding that each as a practical matter affects the others should be covered. A first such case is the quasi ‘natural’ category of transactions in which the single relationship is directly implied. For example, swaps or repurchase transactions are entered into on the understanding that the mutual rights and obligations (which are legally distinct from one another) within a single transaction cannot be separated by the parties and should not be looked at separately in the event of one of the parties becoming insolvent (i.e., no cherry picking should apply in relation to only one leg of these transactions). In a second category of cases, this single relationship is wider and created contractually by the parties. However, given that close-out netting leads to special treatment in the event of insolvency, this contractual single relationship can only be established where there are good objective reasons to deal with a multitude of transactions on a collective basis. The main reasoning here is that it is more efficient for parties to monitor and manage their mutual risk exposure on the basis of an overall assessment of all transactions outstanding between them.  Rapid changes of value: A second justification for applying close-out netting to certain of the parties’ mutual rights and obligations stems from the fact that the volatility of the value of certain financial transactions would expose parties to considerable market and credit risk which they would have difficulty managing if they were not allowed to terminate such transactions upon the occurrence of one of the predefined termination events, in order to determine gains and losses and to re-hedge their portfolio.

Rapid and significant changes in the contract value might expose the non-defaulting party to a multiple of the anticipated counterparty and market risk which can no longer be appropriately hedged.

105

28

PRINCIPLES ON CLOSE-OUT NETTING

 Systemic risk: A third justification is the avoidance of systemic risk. This element flows partly from the second justification. In deteriorating market conditions, the ability to terminate contracts and thus to limit exposures is important in guarding against the situation where the failure by one of the parties to perform its obligations causes its counterparty likewise to become unable to perform its obligations vis-à-vis third parties.

Explanation and commentary General remarks on the scope of application and the minimum harmonisation approach under Principle 4 56. By defining the obligations eligible for inclusion in a closeout netting provision, Principle 4 addresses the substantive scope of application of the Principles. It should be noted that paragraph (1)(a), as the core element of the definition of the term ‘eligible obligation’, contains a reference to the definitions in Principle 3.

Therefore, in order to be eligible, obligations must not only belong to one of the types of obligation enumerated in subparagraph (a), nos. (i) to (iv), but the creditor and the debtor of the obligation must also be eligible parties as defined in Principle 3.

Additionally, one of them must be a public authority or a qualifying financial market participant. Under sub-paragraphs (b) and (c), there are equivalent restrictions of the personal scope of application on the basis of the references to the secured obligation (sub-paragraph (b) ) or to the obligations covered by the master agreements (sub-paragraph (c) ), see infra, paragraphs 80 et seq.

The scope of protection of the operation of close-out netting provisions is limited to such situations in which, at least on the basis of a generalised characterisation, the enforceability of a close-out netting provision not only raises issues as to the relevant parties’ individual credit risk management, but also concerns public interests such as, primarily, protection against systemic risk in financial markets (see supra, paragraphs 48 and 53).

57. Under Principle 4(2), however, implementing States may, at their discretion, extend the scope of application of their protection of the operation of close-out netting provisions to situations beyond the minimum scope of harmonisation suggested under the

106

PRINCIPLE 4 – DEFINITION OF ‘ELIGIBLE OBLIGATION‘

29

Principles, both in respect of the personal scope of application (eligible parties, Principle 4(2)(a) ) and the substantive scope of application (eligible obligations, Principle 4(2)(b) ). See infra, paragraphs 82 et seq.

Paragraph (1) - ‘Eligible obligation’ 58. The term ’eligible obligation’ is defined in paragraph (1) as covering three sub-groups of types of obligation. First, subparagraph (a) covers obligations arising under certain types of contract enumerated in that paragraph. In this regard, the term ‘contracts’ is to be understood in a broad sense, also including contracts that might be categorised as ‘commercial’ contracts. It is impossible to make a neat distinction between financial contracts, on the one hand, and commercial contracts, on the other hand.

For instance, futures and forwards are both used by industrial and commercial companies to hedge price swings in relation to raw materials, etc. Application of these rules to contracts entered into by energy traders, airlines and similar businesses, provided they fulfil the requirements regarding the personal scope of application of the Principles, would be beneficial as these face similar exposures to rapid price swings as those affecting financial firms.

In sub-paragraph (b), the scope of the term ’eligible obligation’ is extended to obligations securing another eligible obligation and in sub-paragraph (c), to obligations arising under master agreements covering other eligible obligations.

Paragraph (1)(a)(i) – ‘Derivative instruments’ 59. The term ‘derivative instruments’, refers to contracts in respect of an underlying or reference asset or a reference value.

The reference value may consist of rates or indices, or of any other measure of economic value, or of factual events. In today’s markets, the reference value usually consists of a rate, yield, price or index relating to interest rates, currencies, transferable securities, money market instruments, commodities, precious metals, credit risk, energy, emissions, economic or monetary statistics, actuarial or other insurance-related data, meteorological data, freight forward rates, bandwidth or property. However, other reference values are also conceivable. Regardless of whether the transaction is concluded in respect of an underlying or reference

107

30

PRINCIPLES ON CLOSE-OUT NETTING

asset or a reference value, the term ‘derivative instrument’ requires that the transaction is, or in future becomes, the subject of recurrent contracts in the derivatives markets. Thus, individual one-off transactions are not covered by paragraph (1)(a)(i).

60. Derivative instruments will typically fulfil both criteria (cf. key considerations, supra) for inclusion in the list of contracts.

First, two typical financial market participants such as banks, merchant banks, funds, insurance companies, etc. will ordinarily regard the multitude of their open derivative instruments with each other as a single relationship. Risk monitoring and assessment will usually be effected by the parties on an aggregate basis.

61. Derivative instruments also pass the test of the second criterion, i.e., exposure to considerable market and credit risk.

They are often highly volatile transactions with rapid and significant price movements. Rapid price movements combined with large outstanding counterparty credit exposures and transaction volumes could also increase the danger of systemic risk.

62. Financial markets subdivide derivatives contracts into a number of categories, notably options, forwards, futures, swaps, contracts for differences, and their respective subcategories. The boundaries between these categories are not always clear-cut.

Moreover, the list of derivatives categories can never be exclusive, in view of the need to cater for future market developments and differences in categorisation. Therefore, the underlying consideration is that the Principles apply to all derivatives covered by the definition in paragraph (1)(a)(i), regardless of which category market practice may attribute to them.

63. Depending on the provisions of the relevant contracts, derivatives can be either physically settled or cash settled. Both are included within the scope of the Principles.

64. For the purpose of the Principles, it is immaterial whether the relevant contracts are entered on-exchange or off-exchange, or whether they are settled ‘over-the-counter’ or through a clearing mechanism or central counterparty (n.b. that in the latter cases, a bilateral close-out netting provision between the central entity and the system participant emerges (cf. supra, paragraph 27)).

108

PRINCIPLE 4 – DEFINITION OF ‘ELIGIBLE OBLIGATION‘

31

Paragraph (1)(a)(ii) – ‘Securities financing transactions’ 65. Paragraph (1)(a)(ii) covers three methods of securities financing: repurchase agreements, securities lending agreements, and other securities financing transactions. In all three cases, transactions are covered if they have been concluded in respect of securities, money market instruments or units in an undertaking for collective investment or an investment fund.

66. A repurchase agreement is a combination of two processes simultaneously agreed upon between the same parties: first, the sale and outright transfer of an asset (e.g., a bond), and secondly, the subsequent repurchase and re-transfer of that same asset at a slightly higher price. This type of agreement is usually driven by cash needs, i.e., in functional terms, it has the same effect as a secured cash loan. The cost of financing (reflected, under a loan agreement, by the payment of interest) is here expressed in the price difference between the sale and repurchase legs of the transaction.

67. Securities lending entails that the securities are made available to the counterparty with a simultaneous agreement to retransfer or return these, or equivalent securities, at a predetermined point in time. The borrower must provide collateral (e.g., in the form of cash) to the lender for the duration of the arrangement. Securities lending is mostly driven by the borrower’s need for a certain types of securities.

68. In functional terms, the mutual flows of assets are identical for both types of transaction. Both types consist of a pair of reciprocal transactions. Although in both cases, each separate transaction could be regarded as legally independent, neither a repurchase agreement nor a securities lending agreement should be at risk of unbundling in an insolvency procedure. Therefore, a repurchase or securities lending agreement per se fulfils the first element of justification mentioned above (single relationship, first case).

69. The main example of another securities financing transaction that would be covered by paragraph (1)(a)(ii) is a margin loan. In much the same way as under the transactions referred to in the preceding paragraphs, under a margin loan money is advanced by a bank to its customer to purchase

109

32

PRINCIPLES ON CLOSE-OUT NETTING

securities on condition that the bank can subsequently regard these securities as collateral securing the loan. Again, the two prongs of such arrangements are (i) a flow of cash in one direction and (ii) the provision of rights over securities (collateral) in the other direction. The collateral can be provided under a title transfer arrangement or a non-title transfer arrangement (cf. infra, sub-paragraph (a)(iii) ), i.e., depending on the arrangement, ownership of the securities is transferred to the bank.

70. Where two parties have a multitude of repurchase, securities lending and margin lending agreements, these are usually closely interconnected as the cash and collateral flows are managed on an aggregate basis rather than separately. As a consequence, there is an objective reason for the parties to cover their mutual exposures flowing from these types of transaction by a close-out netting provision (single relationship, second case).

Paragraph (1)(a)(iii) – ‘Title transfer collateral arrangements’ 71. Paragraph (1)(a)(iii) extends the definition of the term ‘eligible obligation’ to obligations arising under title transfer collateral arrangements related to eligible obligations. There are title-transfer collateral arrangements and non-title transfer collateral arrangements. They differ as to their nature, and the analysis as to whether and to which extent they are suitable for inclusion in a netting provision differs accordingly.

72. Under a title transfer collateral arrangement, full legal title is passed to the collateral taker and the collateral provider receives a claim for transfer of the identical sum or asset at a later stage. No property interest is retained on the provider’s side. As a consequence, valuation and inclusion in the net amount of both legal positions are possible because there are claims for retransfer on both sides (a claim for repayment/retransfer of the value of the transaction, and a claim for retransfer of the collateral).

73. A non-title transfer collateral arrangement involves traditional security agreements such as pledge or charge. These are characterised by the fact that they are proprietary in nature and both the collateral provider and the collateral taker have proprietary interests in the encumbered asset. In particular, the collateral provider will usually retain legal title to the asset. This

110

PRINCIPLE 4 – DEFINITION OF ‘ELIGIBLE OBLIGATION‘

33

type of arrangement is not generally susceptible to close-out netting as commonly understood, since a proprietary interest cannot be combined with, or netted against, a claim of a monetary character. Instead, enforcement of the rights of the pledgee is by way of sale of the pledged property and discharge of the secured obligations out of the proceeds of sale. Where both close-out netting and a traditional security interest apply under the parties’ agreement, close-out netting operates under exclusion of the security interest. Rather, the security interest, in a second logical step, secures the net amount.

74. An important sub-category is the non-title transfer collateral arrangement which includes a right of use. In these cases, the parties agree, generally or in effect, that the proprietary right may be replaced, at the election of the collateral taker, by a right to the return of identical or equivalent assets. This is the case, in particular, where the agreement, sanctioned by the relevant law, permits the collateral taker to use the encumbered asset for its own purposes, in particular to ‘rehypothecate’ it, and subsequently to return not the same asset but an equivalent one.

In this instance, the residual property interest originally vested in the collateral provider may cease to exist and be replaced by a contractual claim for re-transfer or the equivalent thereof. As a consequence, again, there are claims of an obligatory nature on both sides. Therefore, such an arrangement, if allowed under the relevant law, is generally capable of being included in a netting provision, the effect of which is, broadly, that the pledged property remains with the pledgee but a claim equal to its value is included in the netting calculation. Principle 4(1)(a) does not cover non-title transfer collateral arrangements with a right of use, but implementing States may decide under Principle 4(2)(b) to extend the scope of the protection to such non-title transfer collateral arrangements.

75. As is the case of repurchase agreements and securities lending agreements, the separate obligations which constitute a title transfer collateral agreement should not be at risk of being unbundled (single relationship, first case). Likewise, collateral is managed on an aggregate basis. For this reason, a multitude of collateral arrangements between two parties should also be capable of being included in the scope of close-out netting.

111

34

PRINCIPLES ON CLOSE-OUT NETTING

76. It is important to note that repurchase, securities lending as well as title transfer-collateral agreements may be collectively managed and monitored from the perspective of counterparty risk.

Parties often cover all those types of transactions by a close-out netting provision.

Paragraph (1)(a)(iv) – Contracts for the sale, purchase and delivery of certain assets 77. Paragraph (1)(a)(iv) relates to contracts for the sale and delivery of certain assets against payment in so far as they are not covered by the definition of derivative instruments, in particular futures and forwards. For example, on the spot market, prices are agreed and paid immediately, whereas delivery occurs within a time frame of less than one month.

78. The relevant contracts are regularly entered into on the basis of a single relationship, and are subject to the same type of credit risk and change in value as other types of eligible obligation. In addition, they may be subject to settlement risk.

79. It should be noted that paragraph (1)(a)(iv) contains an exhaustive list of contracts giving rise to eligible obligations under this provision. This list does not, for example, include spot contracts for crude oil or transactions concerning emission allowances. Implementing States may, however, decide to extend their protection of the operation of close-out netting provisions to obligations arising from such transactions under paragraph (2)(b).

Paragraph (1)(b) – Surety agreements and other personal security 80. Paragraph (1)(b) ensures that not only the obligations of the (direct) parties to one of the contracts enumerated in paragraph (1) come within the scope of the Principles but also the obligations of other eligible parties that assume a liability of their own (whether by way of surety or guarantor, or as principal debtor) for the performance of the obligation of one of the parties to that contract. The most prominent such arrangements are guarantee and indemnity arrangements or other types of personal security that may exist in different jurisdictions, regardless of the wording employed. If the secured obligation is not an eligible obligation

112

PRINCIPLE 4 – DEFINITION OF ‘ELIGIBLE OBLIGATION‘

35

under paragraph (1)(a), the obligation of the third party security provider likewise cannot be eligible under paragraph (1)(b).

Paragraph (1)(c) – Master-master agreements 81. Paragraph (1)(c) provides that obligations arising as single net obligations under the close-out netting mechanism of a master agreement are likewise eligible obligations, thereby allowing close-out netting to take effect under master-master agreements, i.e., agreements under which there are several master agreements covered by another master agreement providing for the netting of the single net obligations arising under the covered master agreements (see also supra, paragraph 24). Such obligations arising under the master agreements covered by the close-out netting provision in a master-master agreement are eligible obligations under paragraph (1)(c) only to the extent that the covered master agreements have been concluded by the same parties and in respect of obligations covered by paragraphs (1)(a) and (b), i.e., eligible obligations between eligible parties, at least one of which is a public authority or a qualifying financial market participant, or obligations assumed by eligible parties under a suretyship agreement as security for such an obligation.

Paragraph (2)(a) – Extension of the personal scope of application to other parties 82. In suggesting a minimum scope of harmonisation, the Principles are not intended to restrict the possibility for implementing States to extend the scope of their protection of the operation of close-out netting provisions to situations not falling under this minimum scope. Paragraph (2)(a) addresses the extension of the personal scope of application of the protection under the Principles.

83. First, it is left to implementing States’s discretion whether to broaden the scope of paragraph (1)(a) by removing the limitation contained in paragraph (1)(a) that the obligations must arise under contracts to which at least one public authority or a qualifying financial market participant is a party.

113

36

PRINCIPLES ON CLOSE-OUT NETTING

84. Secondly, implementing States may decide to extend the scope of protection under the Principles to close-out netting provisions covering obligations arising from contracts that have been concluded by parties other than eligible parties in the sense of Principle 3(1). Such an extension might especially be useful in respect of natural persons that participate in the financial markets to a certain extent and in relation to whom it might otherwise be unclear whether they act for personal, family or household purposes. Each implementing State is free as regards the extent of the protection and in defining the relevant group of natural persons.

85. Generally, the Principles do not take a stand concerning the issue as to whether implementing States should, in the interests of the protection of the parties’ reliance on close-out netting provisions as a credit-risk management technique, decide in favour of such an extended scope of the enforceability of such provisions or not. The general criteria that could be taken into consideration in exercising this discretion are referred to above (paragraph 17). Implementing States are also free to decide upon the legal technique to be applied when extending the scope of protection of close-out netting provisions. Principle 4(2) allows the implementing States to specify any limitations or exceptions to such an extension of the scope of application of the Principles.

Paragraph (2)(b) – Extension of the substantive scope of application to other obligations 86. Paragraph (2)(b) provides that the implementing States may also extend the substantive scope of application of protection under the Principles. Implementing States may decide that their protection of the operation of close-out netting provisions under the Principles should, under their national law, also apply to obligations not covered by Principle 4(1).

87. The question as to whether loans and deposits should be included may be particularly relevant (see the following paragraphs). However, there might be other types of contract which States may decide to include, such as transactions concerning the sale, purchase or delivery of fungible commodities in general, non-title transfer collateral arrangements (see supra paragraph 74) or contracts for the clearing of obligations covered

114

PRINCIPLE 4 – DEFINITION OF ‘ELIGIBLE OBLIGATION‘

37

by Principle 4(1)(a). Implementing States may even decide to extend the scope to any contracts giving rise to a cash or physical settlement where both parties are public authorities or qualifying financial market participants or even to all types of contract in general.

Extension to loans and deposits 88. The inclusion of loans and deposits in the list of contracts is controversial because, although several arguments may be advanced for their inclusion, other aspects militate against such inclusion. From the outset, ‘consumer’ deposits and loans are left out of this discussion, since consumers are generally excluded from the minimum scope of harmonisation suggested by the Principles (see Principle 3(1) ) and may be included only at the discretion of the implementing State under Principle 4(2)(a).

89. Loans and deposits are closely related to one another from a functional perspective. Both are technically an advance of money (the principal) by one party to another, entailing a promise to return the principal at some point. Both generally, but not necessarily, carry the obligation to pay interest. A more superficial difference concerns the parties’ motivation. It is assumed that a borrower accepts the principal from the lender in order to satisfy its own funding needs, whereas the depositary rather plays the role of safe-keeper of the money in the depositor’s interest.

However, in practice, banks’ traditional sources of financing have been their clients’ deposits, a fact which rather blurs that distinction. From a functional and legal point of view, therefore, loans and deposits are akin to one another. From a regulatory point of view, on the other hand, deposits enjoy specific protection, most particularly the circumstance that deposits receive special support under national legislation and traditionally, only licensed credit institutions (‘banks’) are able to take deposits.

90. It might be argued that neither loans nor deposits pose a particular risk or threat to systemic stability that can be best prevented by the application of close-out netting. They are not necessarily subject to rapid changes in value or to the volatility of markets. They are not used for hedging but rather for funding and they are not traded in large volumes. However, a number of

115

38

PRINCIPLES ON CLOSE-OUT NETTING

factors suggest that the inclusion of loans and deposits might be worth considering in certain circumstances.  Loans mainly consist of a transfer and retransfer of cash.

This functionality is identical to the cash leg of a number of transactions used by banks and central banks, notably repurchase agreements, securities-lending agreements and cash-title transfer collateral agreements. The latter are all undoubtedly within the scope of close-out netting.

Carving out loans generally would mean that a clear distinction would have to be made between (ineligible) loans and the cash leg of the aforementioned (eligible) transactions. This might be difficult, particularly in a cross-jurisdictional situation, and thereby create legal uncertainty and provoke regulatory arbitrage. On the other hand, the presence of a non-cash leg in the eligible transactions might substantially mitigate the potential uncertainty and arbitrage.  Banks regularly post deposits with and give loans to one another. Such deposits may be very short term and thus quite volatile as a funding source, as volumes may change from day to day according to the relevant needs, and because they are often provided in different currencies. Such arrangements expose the parties to credit risk and market (currency) risk. Banks might wish to calculate their mutual risk exposure flowing from these operations on a net, rather than a gross basis.  Central banks take deposits from banks (in fulfilment of their minimum reserves policy) and extend loans to banks (in the framework of their monetary operations). A central bank will have an interest in being able to manage the risk exposure to each of the relevant banks on a net basis, i.e., in being able to apply close-out netting. Therefore, many central banks apply close-out netting to such loans and deposits.  Furthermore, the phenomenon of ‘cash-pooling’ benefits from close-out netting. Cash pooling occurs where member companies of the same group manage their cash reserves collectively. Typically, the positive credit balance of one member of the group is made available to

116

PRINCIPLE 4 – DEFINITION OF ‘ELIGIBLE OBLIGATION‘

39

any others members that are in need of cash, through a common master cash account held by the parent company. A deposit (alternatively: loan) arrangement comparable to a revolving account facility exists between each member of the cash pool and the parent company, under which mutual repayment obligations are expressed as a net credit balance. Legally, mutual payment obligations are not settled until the member in question exits the cash pool arrangement (despite the fact that the current exposure is expressed as a net balance).

However, the parties would not enter into such agreement if their exposure were not limited to the net exposure in the event of the counterparty’s insolvency. If the insolvency administrator were able to cherry pick those deposits/loans that were favourable to the insolvent estate, and if it could at the same time set aside those that were unfavourable, the risk to the solvent party would be considerably increased.

91. On the other hand, there are arguments against making loans and deposits eligible for close-out netting. In addition to the reasons articulated above:  including loans and deposits would mean that that part of a bank’s balance sheet that was subject to close-out netting would be considerably increased.  excluding deposits and loans from the scope of application of close-out netting would not necessarily mean that set-off was equally excluded. Many of the aforementioned arguments put forward in favour of the eligibility of loans and deposits for close-out netting could probably be addressed by set-off.

117

40

PRINCIPLES ON CLOSE-OUT NETTING

PRINCIPLE 5 Formal acts and reporting requirements (1) The law of the implementing State should not make the operation of a close-out netting provision dependent on: (a) the performance of any formal act other than a requirement that a close-out netting provision be evidenced in writing or any legally equivalent form; (b) the use of standardised terms of specific trade associations.

(2) The law of the implementing State should not make the operation of a close-out netting provision and the obligations covered by the provision dependent on the compliance with any requirement to report data relating to those obligations to a trade repository or similar organisation for regulatory purposes.

Key considerations in respect of this Principle ďƒ˜ Formal requirements that impinge on the legal enforceability of close-out netting provisions have considerable potential to create legal uncertainty in a cross-jurisdictional context. Accordingly, the operation of close-out netting provisions should not depend on requirements such as prior registration with a public register or notarisation. ďƒ˜ The operation of close-out netting provisions should not depend on the use of standard documentation so as to allow for tailor-made close-out netting provisions and framework agreements, for individual changes to existing standard documentation or for market-led changes of standard documentation itself. The regulatory framework may impose restrictions in this regard; however, these must not hamper enforceability in commercial and insolvency law terms.

118

PRINCIPLE 5 – FORMAL ACTS AND REPORTING REQUIREMENTS

41

 The reporting of data in relation to certain financial transactions to trade repositories and similar organisations is an important feature of the supervisory framework.

However, non-compliance with the duty to report such data should not entail the non-enforceability of the relevant contracts and the close-out netting provision which covers them.  The principle that the operation of close-out netting provisions should not be made subject to formal acts and reporting requirements does not restrict the implementing State’s power to provide for administrative, regulatory or penal sanctions for non-compliance with formal requirements and reporting requirements. Moreover, it is only the operation of the close-out netting mechanism that is not to be made subject to requirements of form. Where and to the extent that the parties include an agreement on the provision of security in the same contractual (master) agreement, form requirements of secured transactions law may apply, which can result in the ineffectiveness of purported security interests.

Explanation and commentary 92. The effect of non-compliance with formal requirements (in the broadest sense) needs to be considered carefully. Where such non-compliance entails invalidity or unenforceability of a contract, the legislator should always have regard to the fact that both parties to a contract are affected by this consequence. The effect of a considerable number of transactions and/or a close-out netting provision being unenforceable can pose a significant risk to one or both of the parties. In cross-jurisdictional situations in particular, at least one of the parties might be taken by surprise by that consequence. Thus, where the rules on formalities aim at promoting safe and sound market conditions, unenforceability will undermine rather than promote these objectives, and it might be better to settle for other enforcement measures, such as fines, personal liability of staff, withdrawal of license, etc., which can be imposed without creating additional legal uncertainty for the counterparty.

119

42

PRINCIPLES ON CLOSE-OUT NETTING

Paragraph (1)(a) - Formal acts 93. For the above reasons, in a cross-border context, any requirements as to formal acts other than writing (or equivalent forms) appear to create additional risk. The main, but by no means only examples of such formal acts are requirements of prior registration with a public register and notarisation. Other requirements such as a mandatory notice period or prior approval by a competent court arguably are more closely related to enforcement and are therefore dealt with infra, paragraph 118.

There are two strands of such potential risk caused by requirements of formal acts other than writing.

94. First, there is the general risk that, in a cross-border context, requirements of formal acts other than writing are liable to be misunderstood or mishandled from an operational point of view.

Such requirements might be overlooked, in particular as it cannot be excluded that different laws may be applicable within a single bundle of transactions covered by a close-out netting provision.

The necessary steps might not be carried out simply because of practical difficulties, such as language requirements.

95. Secondly, even if such formal requirements are initially complied with under the first law, any possibility of transferring a close-out netting provision (including the transactions covered) to a new, foreign entity would be in jeopardy since it is unlikely that the law of the acquirer would require compliance with exactly the same formal steps.10  This aspect is particularly relevant where a holding company re-integrates with a hitherto legally independent foreign subsidiary, in which case some or all contractual agreements entered into by the subsidiary might from that point on be subject to a different insolvency law, i.e., the law applicable to the parent company. It is unclear whether a contract transferred in this manner would be upheld in the event of the parent company’s insolvency if the formal requirements regarding the close-out netting provision differed.

10 Cf. UNIDROIT 2011, Study LXXVIII C - Doc. 2, p. 37 (Example 7), p. 71 (Example 17).

120

PRINCIPLE 5 – FORMAL ACTS AND REPORTING REQUIREMENTS

43

 It is equally relevant in the context of resolution powers concerning banks or other financial institutions, which usually include the possibility of transfer, by regulatory order, of part or all of the financial institution’s business to a second (solvent) institution. If the receiving second institution is subject to a different insolvency law, and if that law imposes formalities on close-out netting provisions, it is very unlikely that the formalities (if any) under which the close-out provision was originally entered into would suffice.

96. The registration of close-out netting provisions (and in some cases, the obligations covered by them) is required in certain jurisdictions as a condition for the creation, validity, enforceability, effectiveness against third parties, or admissibility of the close-out netting provision. In some cases, this requirement has a deterrent function against fraud, e.g., to exclude fraudulent backdating of close-out netting provisions prior, but close to insolvency. However, this means that all domestic and foreign parties, including those acting in good faith and in the absence of any fraudulent behaviour, as well as in the absence of insolvency of one of the parties, would be hit by the unenforceability of the close-out netting provision as a consequence of non-compliance with the registration requirement, e.g., due to a simple operational mistake. This situation might potentially create great legal uncertainty, and this is why registration should not be linked to the unenforceability of the close-out netting provision. However, there is nothing in the Principles to prevent courts from sanctioning fraudulent behaviour: Principles 6(2) and 7(2) leave open the possibility for the applicable law to treat close-out netting provisions as unenforceable as a consequence of fraudulent behaviour.

Paragraph (1)(b) - Use of standardised terms of trade associations 97. Another issue is the tension between close-out netting provisions contained in a standard master agreement and agreements between parties that wish to customise the close-out netting provision. If jurisdictions were to protect the enforceability of close-out netting provisions only where the latter are included

121

44

PRINCIPLES ON CLOSE-OUT NETTING

in standard documentation, individual amendments would imperil enforceability.

98. However, the relationship between two financial institutions can be quite an elaborate one and call for the master agreement to be customised to some degree. It is impossible to harmonise the extent to which such changes should be admissible, simply because there are too many different, individual situations.

Hence, the concept of only protecting the enforceability of closeout netting provisions that are part of standard documentation is not appropriate, particularly in a cross-jurisdictional context.

Paragraph (2) - Reporting requirements 99. In attempting to render the derivatives market more transparent, many jurisdictions have recently introduced or are about to introduce a duty to report data (parties, volume, type of transaction, date) relating to certain types of standardised derivatives to a trade repository. This measure serves prudential/supervisory purposes and the Principles do not intend to restrict the application of such regulatory requirements.

However, while other potential sanctions are not affected (see infra, paragraph 101), a failure to report should not have as a consequence the restriction of the operation of a close-out netting provision. Should reporting be a prerequisite for the enforceability of the close-out netting provision, any non-compliance would actually create risk, since it would endanger the enforceability of the risk management mechanism agreed by the parties in situations which the parties (and possibly also their regulator) might not have anticipated since the failure will, in most cases, be a consequence of unintentional operational failure. This result would be clearly disproportionate and dangerous.

100. This reasoning also underlies the extension of Principle 5(2) to the obligations covered by the close-out netting provision. The regulatory reporting requirements typically have the objective of ensuring transparency and monitoring the market in order to control risks that may be building up. If a failure to report an underlying obligation were to result in the ineffectiveness of this obligation, this sanction would create additional risks if both parties had relied on this transaction in their risk-management. In order to avoid such unforeseen risks it would appear preferable to

122

PRINCIPLE 5 – FORMAL ACTS AND REPORTING REQUIREMENTS

45

allow the parties to rely on the effectiveness of the transaction and on its valid inclusion in the close-out netting provision. Of course, this does not rule out any other consequences of the failure to comply with the reporting requirement concerning the obligations covered by the close-out netting provision.

Consequences other than the restriction of the operation of a close-out netting provision or obligations covered by that provision: administrative, regulatory or penal sanctions 101. The reasoning in the preceding paragraphs argues against the restriction of the operation of a close-out netting provision (and of the obligations covered by that provision in the situation described in paragraph (2) ) as a consequence of the failure to comply with formal and reporting requirements. Other sanctions, especially administrative, regulatory or penal consequences of such non-compliance, are not affected by Principle 5. It should also be noted that this Principle regulates the consequences of a failure to comply with formal requirements in relation to the close-out netting mechanism only. Where the parties agree on a close-out netting provision and on the provision of security in the same contractual (master) agreement, the validity of the security agreement may be subject to formal requirements of secured transactions law notwithstanding Principle 5. This may result, for instance, in the ineffectiveness of the provision of collateral as security for the obligations of the parties under the close-out netting provision.

123

46

PRINCIPLES ON CLOSE-OUT NETTING

PRINCIPLE 6 Operation of close-out netting provisions in general (1) The law of the implementing State should ensure that a close-out netting provision is enforceable in accordance with its terms. In particular, the law of the implementing State: (a) should not impose enforcement requirements beyond those specified in the closeout netting provision itself; (b) should ensure that, where one or more of the obligations covered by the close-out netting provision are, and remain, invalid, unenforceable or ineligible, the operation of the close-out netting provision is not affected in relation to those covered obligations which are valid, enforceable and eligible.

(2) These Principles do not render enforceable a close-out netting provision or an eligible obligation that would otherwise be unenforceable in whole or in part on grounds of fraud or conflict with other requirements of general application affecting the validity or enforceability of contracts.

PRINCIPLE 7 Operation of close-out netting provisions in insolvency and resolution (1) Subject to Principle 8 and in addition to Principle 6, the law of the implementing State should ensure that upon the commencement of an insolvency proceeding or in the context of a resolution regime in relation to a party to a closeout netting provision: (a) the operation of the close-out netting provision is not stayed;

124

PRINCIPLES 6 AND 7 – OPERATION OF CLOSE-OUT NETTING PROVISIONS

(b) the insolvency administrator, court or resolution authority should not be allowed to demand from the other party performance of any of the obligations covered by the close-out netting provision while rejecting the performance of any obligation owed to the other party that is covered by the close-out netting provision; (c) the mere entering into and operation of the close-out netting provision as such should not constitute grounds for the avoidance of the closeout netting provision on the basis that it is deemed inconsistent with the principle of equal treatment of creditors; (d) the operation of the close-out netting provision, and the inclusion of any obligation in the calculation of the single net obligation under the close-out netting provision, should not be restricted merely because the close-out netting provision was entered into, an obligation covered by the provision arose or the single net obligation under the close-out netting provision became due and payable during a prescribed period before, or on the day of but before, the commencement of the proceeding.

(2) These Principles do not affect a partial or total restriction of the operation of a close-out netting provision under the insolvency law of the implementing State on grounds which include factors other than, or additional to, those referred to in sub-paragraphs (c) and (d) above, such as knowledge of a pending insolvency proceeding at the time the close out netting provision was entered into or the obligation arose, the ranking of categories of claims, or the avoidance of a transaction as a fraud of creditors.

125

47

48

PRINCIPLES ON CLOSE-OUT NETTING

Key considerations in respect of Principles 6 and 7  Principles 6 and 7 aim at protecting the operation of closeout netting provisions from the effect of the application of domestic laws and regulations that may hinder the operation of close-out netting provisions whenever the application of such rules would be in conflict with the objectives of the Principles.  Close-out netting provisions should be enforceable between the parties and against third parties. In the event of the commencement of insolvency proceedings in relation to one of the parties, this includes the insolvency administrator and the general insolvency creditors of the defaulting party.  The operation of close-out netting provisions should be governed by the terms agreed by the parties, both before and after the commencement of insolvency proceedings and also in the context of a resolution regime. As a general rule, implementing States should not impair the operation of close-out netting provisions by imposing restrictions under national laws and regulations, neither in the form of general rules nor through specific rules directed against close-out netting.  However, close-out netting is not shielded against every rule of commercial or insolvency law. The demarcation between those legal rules that should not apply to closeout netting and other legal provisions that should continue to apply requires careful consideration, in relation to laws and regulations of a general nature as well as in relation to restrictions specific to situations of insolvency or in the context of a resolution regime. As a general rule, provided that the general requirements for the valid creation of contracts are fulfilled, the sole fact of entering into a closeout netting provision should not be subjected to additional conditions under contract or commercial law and likewise should not trigger the application of insolvency avoidance rules. However, if a situation involves qualifying elements (for example, fraud vis-à-vis other creditors), the relevant

126

PRINCIPLES 6 AND 7 – OPERATION OF CLOSE-OUT NETTING PROVISIONS

49

contract law and insolvency law tools (remedies for fraud, avoidance, actio pauliana) should continue to apply.  For purposes of international compatibility, a common standard in this regard is of utmost importance.

Explanation and commentary to Principles 6 and 7 Systematic structure of Principles 6 and 7 102. The common objective of Principles 6 and 7 is the comprehensive protection of the operation of close-out netting provisions, covering all aspects from the creation and formal validity (see also the specific rule in Principle 5 ), to the effectiveness against third parties, admissibility in evidence and enforceability, both before and upon the commencement of insolvency proceedings and within and outside the context of resolution regimes.

103. Principle 6 sets out the general rules on the protection of the operation of close-out netting provisions. This general standard of protection applies both before and upon the commencement of insolvency proceedings and within and outside the context of resolution regimes.

104. Rules that are specific to situations of insolvency and resolution regimes are contained in Principle 7. As is made clear in the text of the chapeau of Principle 7 by the words ‘in addition to Principle 6’, the protective rules of Principle 6 remain applicable even after the commencement of insolvency proceedings and in the context of resolution regimes and are complemented under Principle 7 by additional rules that are directed against a number of typical specific restrictions of the operation of close-out netting provisions in the latter types of situation.

105. Principles 6(2) and 7(2) both allow exceptions to the general rules on the protection of the operation of close-out netting provisions, allowing restrictions under the laws and regulations of the implementing State that do not limit the operation of close-out netting provisions as such, but that constitute requirements of general application or are triggered by the presence of other factors, especially fraud.

127

50

PRINCIPLES ON CLOSE-OUT NETTING

106. Principles 6 and 7 are not intended to impinge upon the particular rules applicable in the context of the resolution of banks or other financial institutions which, under certain conditions, may supersede close-out netting provisions. The relationship of the Principles to those rules is addressed in Principle 8, which provision takes precedence over the resolution-specific rules in Principle 7, as is made clear by the words in the chapeau of Principle 7 ‘subject to Principle 8’.

Principle 6(1) sentence 1 - Enforceability of a close-out netting provision according to its terms 107. Principle 6(1) sentence 1 is a ‘catch-all’ provision addressing all rules under national law that might potentially conflict with close-out netting provisions but should not (reservations apply, cf. infra).

108. The wording ‘enforceable in accordance with its terms’ is the core idea of the Principles. It relates to the challenge posed to close-out netting provisions by some quasi-universally recognised legal rules. The best example is probably the insolvency administrator’s right to ‘cherry pick’ (cf. infra), but there are others. However, the diversity of legal systems and of the rules within them makes it very difficult to find a general, international formula that precisely describes which insolvency or commercial law rules and principles cause problems. Such a description is possible only in relation to the most obvious rules, which are here captured under Principle 6(1)(a) and (b) and Principle 7(1)(a) to (d). However, as close-out netting provisions are embedded in commercial and insolvency law in much the same way as any other contract, many other legal obstacles are capable of rendering a close-out netting provision unenforceable. These are potentially numerous, but difficult to describe.

109. An important reason for this is that close-out netting is a new concept as yet not properly addressed in many jurisdictions, thereby forcing the courts to seek analogies to deal with this new matter.

110. A telling example of a conflict that might hamper the enforceability of close-out netting would be its assimilation to statutory set-off rights under commercial law and the resulting

128

PRINCIPLES 6 AND 7 – OPERATION OF CLOSE-OUT NETTING PROVISIONS

51

application of the requirements for set-off to close-out netting.

Despite the fact that statutory set-off is more limited than close-out netting, in the absence of any clarifying legal rule courts and insolvency administrators might apply its requirements by analogy to close-out netting provisions, thus potentially distorting the enforceability of close-out netting. In particular, set-off traditionally applies only to mutual obligations of the same kind that are due and payable and that are legally distinct; in some jurisdictions, set-off traditionally is restricted to obligations flowing from the same agreement, or that are very closely connected to each other (for the delimitation between set-off and close-out netting see also supra, paragraph 3). As these requirements will rarely be complied with by a close-out netting provision, there is a real risk that the provision will be unenforceable.

111. Similar impediments to the enforceability of close-out netting provisions might stem from their perceived similarity to such known concepts as, for example, novation, and the subsequent application of the enforceability requirements of a novation agreement to a close-out netting provision. However, as analogies like these are probably very diverse, there is a need for a ‘catch-all’ rule. This is why Principle 6(1) sentence 1 prescribes that a close-out netting provision, as defined in functional terms in Principle 2, should be generally enforceable.

112. It is obvious, however, that close-out netting provisions would never be allowed to trump certain other fundamental rules, such as the rules relating to misrepresentation and fraud to the detriment of the counterparty, its creditors or the insolvent estate.

In certain cases, the distinction may be quite difficult to make.

This is why Principle 6(1)(a) and (b) and Principle 7(1)(a) to (d) set out the most typical challenges to close-out netting provisions that should be disapplied in order to guarantee the enforceability of close-out netting, while exceptions to this general rule apply under Principles 6(2) and 7(2).

Enforceability of a close-out netting provision within and outside insolvency and resolution 113. The systematic structure of Principles 6 and 7 as set out above ensures that the scope of the principle on the enforceability of a close-out netting provision according to its terms as laid down

129

52

PRINCIPLES ON CLOSE-OUT NETTING

in Principle 6(1) sentence 1 covers situations both before and upon the commencement of insolvency proceedings and within and outside the context of resolution regimes.

114. The background is as follows. A close-out netting provision is a provision within a bilateral contractual relationship. Outside insolvency proceedings or resolution regimes, such a close-out netting provision rarely clashes with policy considerations so that there is scant reason to prohibit or limit its use. As a consequence, a close-out netting provision should generally be effective and enforceable as between the parties outside the context of insolvency proceedings or resolution regimes.

115. The role of a close-out netting provision in reducing counterparty and systemic risk becomes dominant in particular in the event of the commencement of insolvency proceedings in relation to one of the parties. However, rules of insolvency law that are generally intended to preserve the insolvency estate for distribution to creditors and to ensure equal treatment of the latter could potentially be incompatible with the essential features of close-out netting. One of the primary purposes of insolvency law is to determine the question as to which creditors’ claims should take precedence over which other creditors’ claims. Insolvency law traditionally provides for tools such as ‘cherry picking’ and avoidance of contracts to put its insolvency policies into practice (cf. infra), and the application of such rules may render close-out netting provisions meaningless. However, the enforceability of close-out netting is crucial not only outside insolvency but within insolvency and resolution as well, since the purpose of close-out netting is to reduce counterparty and systemic risk. If this is not recognised in the latter situations, the value of close-out netting provisions would be seriously disturbed. Accordingly, the purpose of Principles 6 and 7 is to clarify that the law should protect the enforceability of a close-out netting provision throughout its lifetime, namely both before and upon the commencement of insolvency proceedings and within and outside the context of resolution regimes.

116. For the purpose of the Principles, the definition of ‘insolvency proceeding’, i.e., which ‘insolvency’ procedures should accommodate close-out netting, should be very broad, and the Principles duly target the law governing a great variety of

130

PRINCIPLES 6 AND 7 – OPERATION OF CLOSE-OUT NETTING PROVISIONS

53

different procedures. Reference is made to Article 1(h) of the Geneva Securities Convention11: ‘insolvency proceeding’ means a

collective judicial or administrative proceeding, including an interim proceeding, in which the assets and affairs of the debtor are subject to control or supervision by a court or other competent authority for the purpose of reorganisation or liquidation. Both

judicial and administrative proceedings are covered, aiming at both liquidation and reorganisation, these terms being defined in the UNCITRAL Legislative Guide on Insolvency Law as follows:

’Liquidation’: proceedings to sell and dispose of assets for distribution to creditors in accordance with the insolvency law;12 ‘reorganization’: the process by which the financial well-being and viability of a debtor’s business can be restored and the business continue to operate, using various means, possibly including debt forgiveness, debt rescheduling, debt-equity conversions and sale of the business (or parts of it) as a going concern.13 117. As a factual situation of particular interest as regards the enforceability of close-out netting provisions, resolution regimes are expressly mentioned in Principle 7, emphasising that – subject to Principle 8 – the protection of the operation of close-out netting provisions under Principle 7 also covers the newly developed and still developing ‘resolution regimes for financial institutions’, as described in the Financial Stability Board’s Key Attributes of Effective Resolution Regimes for Financial Institutions, which – depending on the national legislation in the different jurisdictions – may or may not fall within the definition of the term ‘insolvency proceeding’ as referred to in the preceding paragraph. Under such resolution regimes, a national authority (typically the central bank or the financial services authority, or both) takes appropriate measures in respect of a financial institution that is no longer viable, such as, in particular, transferring the failed firm’s assets and liabilities to a bridge institution, overriding shareholders’ rights, conducting a ‘bail-in’, etc. It follows from the Key

11

UNIDROIT Convention on Substantive Rules for Intermediated Securities, adopted in Geneva, on 9 October 2009.

12 UNCITRAL, 2004 Legislative Guide on Insolvency Law, Glossary (w) (p. 5).

13 UNCITRAL, 2004 Legislative Guide on Insolvency Law, Glossary (kk) (p. 7).

131

54

PRINCIPLES ON CLOSE-OUT NETTING

Attributes, para. 4.1 that, first, the legal framework for close-out netting during a crisis should be clear and that close-out netting should be enforceable. Accordingly, the Principles should, in general, also apply to administrative procedures aiming at the resolution of financial institutions. Secondly, however, close-out netting should not hamper the effective implementation of resolution measures: in particular, the early termination of large volumes of assets under close-out netting provisions has the potential of undermining the effectiveness of the authority’s measures since such termination might occur before the appropriate measures can be taken. That is why the Financial Stability Board requires, inter alia, that the regulator be given the right temporarily to stay early termination and acceleration rights (Key Attributes, para. 4.3). This issue is addressed separately in Principle 8.

Principle 6(1)(a) – Additional enforcement requirements 118. Whereas Principle 5 deals with the formal acts which should not be required for the operation of a close-out netting provision, Principle 6(1)(a) relates to additional conditions for the enforcement of a close-out netting provision. The practical value and effect of close-out netting would be significantly diminished or even rendered void if the law were to impose any additional requirements as conditions for the enforcement of close-out netting provisions that went beyond those that the parties might have contractually agreed. In particular, the requirements traditionally imposed on the realisation of security interests such as pledges, charges and mortgages should not be made to apply to close-out netting. Such specific requirements may include, for example,  prior notice to the defaulting party that the close-out netting provision may be put into operation;  approval of the terms of the realisation or operation of the close-out netting provision by a court or other public authority; or that  the realisation be conducted by public auction or in any other prescribed manner; or that

132

PRINCIPLES 6 AND 7 – OPERATION OF CLOSE-OUT NETTING PROVISIONS

55

 the close-out netting provision be operated in a legally prescribed manner; or that  the close-out netting provision be subject to the requirements that may apply in the context of enforcing set-off.

119. It should be noted, however, that since the parties’ contract is based on contractual freedom, they are free to include any of the above or similar requirements in the close-out netting provision, if they so wish.

Principle 6(1)(b) – Invalid/unenforceable/ineligible obligation included 120. Another group of potential obstacles to the enforceability of netting provisions relates to the obligations covered. One or several of the obligations covered might flow from a particular type of transaction which is invalid, unenforceable or ineligible.

Since the close-out netting provision and all the obligations to which it applies are often regarded as one contract, general principles of commercial law could hamper the enforceability of the bundle as a whole. This might endanger the enforceability of the netting provision as a whole, i.e., with respect to all remaining obligations. A better solution would be to ensure that the netting mechanism is not affected in relation to the other obligations that are valid, enforceable and eligible.

121. An obligation is ineligible if it is not of a type covered by Principle 4. Ineligible obligations should simply be severed from the bundle of obligations covered by the close-out netting provision and continue their separate lives under the applicable regime of their governing law, whereas the remaining obligations can be netted.

122. Even if in principle eligible, an obligation may be unenforceable for various reasons. A prominent case relates to wagering or gaming prohibitions which might apply in relation to certain derivatives transactions in some jurisdictions. The fact that one or more of the obligations covered by the close-out netting provision are unenforceable should not have any impact on the netting of the remainder of the bundle of obligations covered by the close-out netting provision.

133

56

PRINCIPLES ON CLOSE-OUT NETTING

Principle 7(1)(a) – Stay 123. Insolvency rules often impose a stay on all transactions with the insolvent estate as from the moment of the commencement of the proceeding. Such a stay would traditionally also inhibit the operation of set-off and is sometimes expressly extended to netting mechanisms. The reasoning is that any further outflow of assets must be stopped and the insolvency administrator given the right to repudiate all unfavourable contracts. However, a stay imposed on the close-out netting of eligible obligations leads to a situation in which it becomes impossible effectively to manage the credit and market risk associated with the bundle of obligations covered.

During the stay, their value might fluctuate considerably and cause much greater potential damage to the solvent party than would have occurred had termination been possible at the moment of insolvency. Furthermore, from a conceptual angle, a stay appears unnecessary since the insolvency administrator should not have the right to choose among unperformed contracts (no cherry picking, cf. infra). This is why Principle 7(1)(a) suggests that the operation of the close-out netting provision should not be stayed upon the commencement of an insolvency proceeding in relation to a party to a close-out netting provision. It should be emphasised that this provision is only intended to address stays ordered as a general measure based upon the commencement of an insolvency proceeding alone, while a restriction of the operation of a close-out netting provision in more specific circumstances would be subject to the other provisions of Principle 7, notably Principle 7(2).

124. The prohibition of stays under Principle 7(1)(a) applies in the context of resolution regimes as well (cf. also supra, paragraph 117), but in this context it is subject to the exception in Principle 8, which accommodates stays that are necessary in the context of resolution of financial institutions.

Principle 7(1)(b) – Cherry picking 125. In an insolvency proceeding, the insolvency administrator or court may have the right to ’cherry pick’ from the insolvent party’s non-performed contracts. This refers to the right to require any counterparty to perform those contracts that are favourable to

134

PRINCIPLES 6 AND 7 – OPERATION OF CLOSE-OUT NETTING PROVISIONS

57

the insolvent estate while rejecting those that are unfavourable to it.

126. If it were possible to cherry pick among a netting set, the bundle of transactions would be disassembled and the solvent party would be required to perform its obligations under all the transactions that were unfavourable from its perspective, whereas the insolvency administrator would not perform the obligations under the favourable transactions – ultimately, the solvent party would be exposed to the full counterparty risk.

127. Cherry picking is essentially contrary to the characteristics of a single relationship as set out supra (cf. key considerations in respect of Principle 4 ). Furthermore, cherry picking disproportionately increases the counterparty risk for the other party and should not, therefore, be available to the insolvency administrator.

128. Those jurisdictions that accommodate close-out netting tend to solve the conflict between cherry picking and enforceability of close-out netting provisions by disallowing the selection of isolated obligations. The insolvency administrator may not demand performance under transactions covered by the close-out netting provision whilst rejecting any of the obligations owed to the other party. The right of the insolvency administrator, to the extent that this right exists under the applicable insolvency law, to reject all transactions covered by the close-out netting provision is not affected. Thus, either all transactions must be performed, or

none.

129. Obviously, the rule in the preceding paragraph is subject to the qualification that the underlying transactions may not be unenforceable for any other reason. The insolvency administrator is not prevented from demanding performance under other obligations covered by the master agreement merely by reason of having refused to perform an obligation that is not enforceable as such. In such situations, Principle 6(1)(b) protects the operation of the close-out netting provision only as regards the other obligations covered that are valid, enforceable and eligible (for a specific application of this rule as regards the time-based avoidance of obligations covered by the close-out netting provision, see infra, paragraph 141) Moreover, no performance may be demanded once the obligations covered by the close-out

135

58

PRINCIPLES ON CLOSE-OUT NETTING

netting provision are terminated under the close-out mechanism and reduced to or replaced by a single net obligation representing the value of the combined obligations.

130. The same principles apply where close-out netting provisions and their underlying obligations are again bundled by an ‘umbrella’ close-out netting provision (in practice, several master agreements are bundled by a ‘master-master agreement’, see supra, paragraph 24). To the extent that the various master agreements are validly included in the close-out netting mechanism provided for under the master-master agreement, the insolvency administrator should not be allowed to require performance on just one of them.

Principle 7(1)(c) – No conflict with equal treatment of creditors 131. This paragraph suggests that under the domestic law, the mere entering into and operation of a close-out netting provision as such should not constitute grounds for the avoidance of a closeout netting provision on the ground that it is deemed inconsistent with the principle of equal treatment of creditors of the insolvent estate by favouring one creditor to the detriment of the other creditors.

132. First, this rule is particularly relevant since the effects of a close-out netting provision often occur at the moment of, or shortly before, the insolvency proceedings are opened. As a consequence, conflicts with the so-called ‘anti-deprivation’ principle, the pari passu principle, or the unenforceability of ipso facto clauses might otherwise arise.

133. Secondly, this Principle addresses the concern that the mere inclusion of a close-out netting provision in the contractual documentation might be deemed inconsistent with the principle of equal treatment of creditors to the detriment of other creditors of the insolvent estate. In the absence of any qualifying facts, the conclusion of a close-out netting provision is neutral, as it is not clear which party, if any, will default. Furthermore, parties do not, at the time of entering into the close-out netting provision, know who will be ‘in the money’ or ‘out of the money’ at any given point in time in the future.

136

PRINCIPLES 6 AND 7 – OPERATION OF CLOSE-OUT NETTING PROVISIONS

59

134. However, the domestic law can impair the operation of a close-out netting provision should there be qualifying elements, going beyond the mere fact of entering into the close-out netting provision (cf. Principle 7(2), with references to some examples of such qualifying factors).

Principle 7(1)(d) – Suspect periods and zero-hour rules 135. National insolvency laws often contain rules avoiding (or allowing the administrator or court to avoid) the entering into agreements or the taking effect of close-out netting mechanisms, transfers, payments and provision of collateral which have occurred during a prescribed period prior to insolvency. Such a period is either defined as a fixed period prior to the commencement of insolvency proceedings (e.g., the three months preceding the date of commencement), or it may be defined by the insolvency court, counting, in particular, from the point in time when the over-indebtedness or similar indicator first occurred. The reasoning underlying such rules is to increase the pool of assets available for distribution amongst general creditors and to avoid unjustified preference of one or more creditors over the remaining creditors by ‘clawing back’ the relevant payments or property.

136. Neither the close-out netting provision nor, to the extent that they are included into the calculation of the single net obligation under the close-out netting mechanism, any obligations covered by the close-out netting provision should be subject to such avoidance rights.

In relation to the close-out netting provision 137. The close-out netting provision itself is protected under Principle 7(1)(d) against time-based avoidance in two separate respects. First, the mere fact that the close-out netting provision took effect, i.e., that the single net obligation representing the aggregate value of the combined obligations covered by the closeout netting provision became due and payable, during a suspect period should not justify any restriction of the operation of that provision. This principle is of key importance in protecting the enforceability of close-out netting provisions, since it is often the application for the commencement of insolvency proceedings

137

60

PRINCIPLES ON CLOSE-OUT NETTING

itself that triggers the close-out netting mechanism. This means that such provisions would be largely devoid of effect if they could be avoided on the grounds of the close-out netting mechanism having taken effect within the suspect period.

138. Secondly, in some jurisdictions, there might be uncertainty as to whether entering into a close-out netting provision during the suspect period might justify an avoidance action. Therefore, there is a risk that an insolvency administrator or court would attempt to halt, avoid or otherwise render unenforceable a closeout netting provision entered into during the suspect period.

However, the parties are in no position, at the time of entering into a netting provision, to know which of them, if any, might subsequently default. Equally, they cannot know with any certainty which party will be ‘in the money’ at the time of the potential subsequent default of one of the parties. Thus, entering into a close-out netting provision is neutral from the outset and equally beneficial or disadvantageous as to the risk borne by both sides. This situation is different from that of receiving payments or property or taking new or additional collateral, which decreases the credit risk of only one of the parties. As a result, entering into a close-out netting provision should not be subject to avoidance merely on the grounds that it took place before insolvency proceedings commenced.

In relation to the obligations covered 139. In so far as this concerns the operation of the close-out netting provision, this Principle also extends to the obligations covered by the close-out netting provision: To the extent that such obligations are included in the calculation of the single net obligation under the close-out netting provision, no obligation should be subject to avoidance solely on the grounds that it arose during the subject period.

140. The reasoning behind this Principle is that an insolvency administrator would typically avoid only those obligations falling within the suspect period and favourable to the solvent party. The result would be comparable to that described above (cf.

Principle 7(1)(b) – ‘cherry picking’). As a consequence, the solvent party would be burdened with a considerably increased credit risk which could not be foreseen at the time of entering into the

138

PRINCIPLES 6 AND 7 – OPERATION OF CLOSE-OUT NETTING PROVISIONS

61

contract or at any other point in time when the relevant obligation arose. On the other hand, the issue of the enforcement of the relevant obligation as such, where it is not included in the calculation of the single net obligation under the close-out netting mechanism, falls outside the scope of the Principles. Thus, where the solvent party does not trigger the close-out netting mechanism and does not rely on its rights under the close-out netting provision, the application of national insolvency avoidance provisions based on the fact that the obligation in question arose during a suspect period is not restricted.

141. It should also be noted that the limited protection afforded under Principle 7(1)(d) in respect of the obligations covered by the close-out netting provision has the consequence that where the national insolvency law allows such obligations to be avoided on the grounds that they arose during a suspect period, the insolvency administrator is not prevented, either under Principle 7(1)(b) or under Principle 7(1)(d), from refusing performance of those obligations while demanding performance of other obligations owed by the solvent party that are covered by the close-out netting provision.

Zero-hour rules 142. For the same reason, the enforceability of close-out netting provisions should not be impaired by the application of ‘zerohour rules’, i.e., rules that by way of legal fiction bring forward the commencement of insolvency proceedings to 0:00h of the day of the decision to open them.

Safeguard against fraud or transactions with knowledge of a pending insolvency proceeding, etc.

143. The aforementioned applies only to the extent that there are no other qualifying elements present (cf. the wording ‘merely because’). As a consequence, and in accordance with Principle 7(2), the law remains free to determine the consequences of fraud, knowledge of a pending insolvency proceeding at the relevant moment, or similar factors.

139

62

PRINCIPLES ON CLOSE-OUT NETTING

Principles 6(2) and 7(2) – Exceptions 144. Principles 6(2) and 7(2) refer to various national laws and regulations that are not to be affected by the Principles. These rules of national law are acknowledged as providing for exceptions to the general principle on the enforceability of a close-out netting provision according to its terms under Principle 6(1) sentence 1 and to the more specific sub-rules in Principle 6(1)(a) and (b) and Principle 7(1)(a) to (d).

145. Principle 6(2) addresses rules on fraud or conflict with other requirements of general application affecting the validity or enforceability of contracts and provides that neither close-out netting provisions nor, as a matter of course, the obligations covered by those provisions are exempted from the application of such general rules.

146. Principle 7(2) is directed more specifically at national insolvency law provisions which could restrict the operation of a close-out netting provision on the basis of factors other than, or additional to, those referred to in Principle 7(1)(c) and (d), such as knowledge of a pending insolvency proceeding at the time the close-out netting provision was entered into or the covered obligation arose, the ranking of categories of claims, or the avoidance of a transaction as a fraud of creditors.

147. Principle 7(2) provides a non-exhaustive list of examples of factors that might allow the operation of a close-out netting provision to be restricted under the insolvency law of an implementing State, but does not specify further conditions in this regard. Principle 7(2) is specific only insofar as it is expressly spelt out that the elements mentioned in Principle 7(1)(c) and (d) cannot justify the restriction of the operation of a close-out netting provision, not even on the basis of the rules mentioned in Principle 7(2), unless additional factors are present. Moreover, the nature of the examples given in Principle 7(2) and the nature of this provision as an exception to the general rule on the enforceability of close-out netting provisions suggests that there should be a rather high threshold for the restriction of the operation of a close-out netting provision under the national insolvency law. For instance, mere knowledge of a pending insolvency proceeding at the time the close-out netting provision

140

PRINCIPLE 8 – RESOLUTION OF FINANCIAL INSTITUTIONS

63

was entered into or the covered obligation arose, should not necessarily justify restricting the operation of that provision under the national insolvency law in cases where the parties pursued legitimate intentions.

148. It should be noted that Principles 6 and 7 are concerned only with such restrictions under national law as impair the operation of the close-out netting mechanism, as defined in Principle 2 as such. The Principles do not address the enforceability of other clauses that are merely ancillary to a closeout netting provision, such as walk-away clauses and provisions on wait-and-see periods (see supra, paragraphs 41 and 43).

Similarly, the rules of some legal systems which subject the operation of a close-out netting provision to standards of commercial reasonableness are not regarded as restrictions that might run counter to Principles 6 and 7 to the extent that the definition in Principle 2 covers only such close-out netting provisions under which the resulting single net obligation is calculated so as to ‘represent the aggregate value of the combined obligations’ (see supra, paragraph 38).

PRINCIPLE 8 Resolution of financial institutions These Principles are without prejudice to a stay or any other measure which the law of the implementing State, subject to appropriate safeguards, may provide for in the context of resolution regimes for financial institutions.

Key considerations in respect of this Principle  The Principles should assist in shaping domestic legal rules on close-out netting that also accommodate ‘resolution regimes’ for financial institutions. The current international consensus on the standards for such resolution regimes is laid down in the Key

141

64

PRINCIPLES ON CLOSE-OUT NETTING

Attributes of Effective Resolution Regimes for Financial Institutions developed by the Financial Stability Board.14  The first aspect is that the legal framework governing close-out netting should be clear and transparent and that close-out netting should be enforceable also within the context of resolution regimes for financial institutions. This aspect is covered by Principles 6 and 7.  The second aspect is that despite the above, close-out netting should not hamper the effective implementation of resolution measures. In particular, the competent authority should, under certain conditions and to a certain extent, have the right to delay the operation of a close-out netting provision by means of a stay of the termination or acceleration rights occurring under such provision. As such a right would be contrary to Principle 7(1)(a), there should be an express exception in this sense.  While stays are of particular relevance for the operation of close-out netting provisions, national resolution authorities might have a variety of other types of resolution powers besides stays (see the Financial Stability Board’s Key Attributes, part 3). The exception contained in Principle 8 is therefore extended to any measure which the law of the implementing State, subject to appropriate safeguards, may provide for in the context of resolution regimes for financial institutions.  ‘Appropriate safeguards’ refers to measures to protect the valid interests of the counterparty to the financial institution that is being resolved, while ensuring the effective resolution of the failed institution. The reference to ‘appropriate safeguards’ in Principle 8 is to be understood as subjecting the exception contemplated in this Principle to international standards on the appropriate treatment of close-out netting in the context of resolution regimes for financial institutions. Thus,

14

Cf. Financial Stability Board, Key Attributes of Effective Resolution Regimes for Financial Institutions, October 2011.

142

PRINCIPLE 8 – RESOLUTION OF FINANCIAL INSTITUTIONS

65

Principle 8 is oriented at and should be interpreted in the light of the Financial Stability Board’s Key Attributes

parts 4 and 5.

 It should be noted that international standards on the appropriate treatment of close-out netting in the context of resolution regimes for financial institutions and their implementation under national law are still evolving.

The policy underlying such resolution regimes is still subject to changes in the future and the references in Principle 8 and the commentary should therefore be understood as being dynamic and contingent.

Explanation and commentary 149. The Principles ensure the enforceability of close-out netting provisions – including within the context of resolution regimes for financial institutions (cf. supra, paragraph 117). However, the Cross-border Bank Resolution Group15 has shown that the unrestricted exercise of termination rights on the occasion of a financial institution’s entry into resolution proceedings, in particular the simultaneous close-out of high volumes, has the potential of harming the competent authority’s aim of orderly resolution of the relevant institution.

150. Principle 8 provides that stays or other measures restricting the operation of close-out netting provisions should be subject to ‘appropriate safeguards’, i.e., measures to protect the valid interests of the counterparty to the financial institution that is being resolved, while ensuring the latter’s effective resolution. The use of the term ‘appropriate safeguards’ is to be understood as a reference to the international standards with regard to special resolution regimes for financial institutions which are implemented under the various national resolution regimes and which include standards for the protection of close-out netting provisions in the context of such resolution regimes.

15

Bank for International Settlements/Basel Committee on Banking Supervision, Report and Recommendations of the Cross-border Bank Resolution Group, March 2010, Recommendation 9, p. 40 et seq.

143

PRINCIPLES ON CLOSE-OUT NETTING

66

151. The reference to ‘appropriate safeguards’ underscores the rule that close-out netting provisions should be generally enforceable and that a stay or any other measure taken by the relevant authority to restrict the operation of such provisions should be strictly limited to the effective implementation of the resolution regime. As regards such stays and the question as to whether the entry into resolution and the exercise of any resolution powers can be validly agreed in the close-out netting provision as a trigger for a close-out netting mechanism, the Financial Stability Board has set the current international standard for the interplay between resolution measures and appropriate safeguards in the Key Attributes part 4 and the relevant Annex IV, in particular by requiring that – - ‘subject to adequate safeguards, entry into resolution and the exercise of any resolution powers should not trigger statutory or contractual set-off rights, or constitute an event that entitles any counterparty of the firm in resolution to exercise contractual acceleration or early termination rights provided the substantive obligations under the contract continue to be performed’ (Key Attributes, para. 4.2); and - ‘should contractual acceleration or early termination rights nevertheless be exercisable, the resolution authority should have the power to stay temporarily such rights where they arise by reason only of entry into resolution or in connection with the exercise of any resolution powers. The stay should: (i)

be strictly limited in time (for example, for a period not exceeding 2 business days);

(ii) be subject to adequate safeguards that protect the integrity of financial contracts and provide certainty to counterparties (see Annex IV on Conditions for a temporary stay); and (iii) not affect the exercise of early termination rights of a counterparty against the firm being resolved in the case of any event of default not related to entry into resolution or the exercise of the relevant resolution power occurring before, during or after the period of

144

PRINCIPLE 8 – RESOLUTION OF FINANCIAL INSTITUTIONS

67

the stay (for example, failure to make a payment, deliver or return collateral on a due date)’ (Key Attributes, para. 4.3).

152. Reference is made to Annex IV to the Financial Stability Board’s Key Attributes for further details. ***

145

146