Posts Tagged Choice of law
Apple v eBizcuss. Wahl AG on choice of court, anti-trust (competition law; clarifying CDC) and ‘corresponding relationships’.
Those of us who are familiar with the issue of multilingualism and international courts, will enjoy the discussion of contractual terms in Wahl AG’s Opinion in C-595/17 Apple v eBizcuss. Not only does the issue entre around the precise implications of the wording of a choice of court provision. The Opinion (not yet available in English) also highlights the difficulty of translating the original English of the contractual term, into the languages at the Court.
Current litigation is a continuation of the earlier spats between Apple and eBizcuss, which led to the Cour de Cassation’s 2015 reversed stance on the validity of unilateral choice of court – which I discussed at the time.
The 2002 Apple Authorized Reseller Agreement (in fact the 2005 version which applied after continuation of the contract) included a governing law and choice of court clause reading
„This Agreement and the corresponding relationship between the parties shall be governed by and construed in accordance with the laws of the Republic of Ireland and the parties shall submit to the jurisdiction of the courts of the Republic of Ireland. Apple reserves the right to institute proceedings against Reseller in the courts having jurisdiction in the place where Reseller has its seat or in any jurisdiction where a harm to Apple is occurring.” (emphasis added)
Footnote 3 displays the translation difficulty which I refer to above: parties disagree as to the translation of the contractual clause in French: applicant suggest this should read „et la relation correspondante”, defendant proposes „et les relations en découlant”. The AG suggest to include both for the purposes of his analysis „Le présent contrat et la relation correspondante (traduction de la requérante)/et les relations en découlant (traduction de la défenderesse) entre les parties seront régis par et interprétés conformément au droit de l’Irlande et les parties se soumettent à la compétence des tribunaux de l’Irlande. Apple se réserve le droit d’engager des poursuites à l’encontre du revendeur devant les tribunaux dans le ressort duquel est situé le siège du revendeur ou dans tout pays dans lequel Apple subit un préjudice.” In Dutch: „De door partijen gesloten onderhavige overeenkomst en de bijbehorende betrekking (vertaling van verzoekster)/de hieruit voortvloeiende betrekkingen (vertaling van verweerster) tussen partijen zullen worden beheerst door en worden uitgelegd volgens het Ierse recht, en partijen verlenen bevoegdheid aan de Ierse rechter. Apple behoudt zich het recht voor om vorderingen jegens de wederverkoper aanhangig te maken bij het gerecht in het rechtsgebied waar de wederverkoper is gevestigd of in een land waar Apple schade heeft geleden.”
This translation issue however highlights precisely the core of the discussion: ‘the corresponding relationship’ suggest a narrow reading: the relationship corresponding to the contractual arrangements. Infringement of competition law does not correspond, in my view. ‘La relation correspondante’ displays this sentiment. ‘(L)es relations en découlant’ suggests a wider reading.
In 2012 eBizcuss started suing Apple for alleged anti-competitive behaviour, arguing Apple systematically favours its own, vertically integrated distribution network.
The Cour de Cassation had rebuked the Court of Appeal’s finding of lack of jurisdiction. In its 2015 decision to quash, (the same which qualified the Court’s stance on unilateral jurisdiction clauses) it cited C-352/13 CDC, in which the CJEU held that choice of court clauses are not generally applicable to liability in tort (the clause would have to refer verbatim to tortious liability): the specific para under consideration is para 69 of that judgment in CDC:
‘the referring court must, in particular, regard a clause which abstractly refers to all disputes arising from contractual relationships as not extending to a dispute relating to the tortious liability that one party allegedly incurred as a result of its participation in an unlawful cartel’.
At issue in Apple /eBizcuss is essentially what kind of language one needs for choice of court to include infringement of competition law (for Dutch readers, I have an earlier overview in Jacques Steenbergen’s liber amicorum here).
Wahl AG emphasises (at 56) that it would not be in the spirit of Article 25 Brussels I Recast (which he analyses in extenso in the previous paras) to require parties to include the exact nature of the suits covered by the choice of court agreement. He is right of course – except those suits in my view do need to be contractual unless non-contractual liability has been clearly included: that in my view is the clear instruction of the CJEU in CDC.
The AG then continues the discussion (which will be redundant should the CJEU not follow his lead) as to whether the clause covers both follow-on (a suit for tort once a competition authority has found illegal behaviour) as well as stand-alone (private enforcement: a party claiming infringement of competition law in the absence of an authority’s finding of same) suits. He suggests there should be no distinction: on that I believe he is right.
I reported earlier on Sulamerica and the need properly and preferably, expressly to provide for choice of law vis-a-vis arbitration agreements, in particular vis-a-vis three elements: lex arbitri, lex curia, lex contractus. In Shagang the High Court added its view on the possible relevance of a fourth factor: the geographical venue of the arbitration, and its impact in particular on the curial law: the law which determines the procedure which is to be followed.
Atlas Power Ltd -v- National Transmission and Despatch Co Ltd  EWHC 1052 is another good illustration of the relevance (but in practice: rarity) of the proper identification of all four factors.
Bracewell excellently identify the four take away points from Atlas Power:
- It is the seat of arbitration that determines the curial law of the arbitration, not the governing law of the contract.
- (To English Courts) the choice of the seat of arbitration is akin to an exclusive jurisdiction clause in favour of the courts of the place designated as the seat of the arbitration having the supervisory role over the arbitration.
- The English courts can and will use their powers to grant anti-suit injunctions to prevent a party from commencing foreign proceedings in breach of an arbitration agreement.
- Complex drafting increases the risk of satellite litigation and the accompanying delay and expense.
The core point which Atlas Power illustrates is that specific identification of arbitration venue, curial law, lex contractus and lex arbitri is best done in simple terms. Overcomplication, particularly variance of any of these four points, is a truly bad idea. Specifically: the arbitration clause in the contracts between the parties (text from Bracewell’s overview)
- Started by providing that the “arbitration shall be conducted in Lahore, Pakistan”.
- Then stated that if the value of the dispute was above a certain threshold or fell within a certain category, either party could require that the arbitration be conducted in London.
- Finally, the clause provided that, notwithstanding the previous sentences, either party may require that the arbitration of any dispute be conducted in London, provided that if the dispute did not satisfy the threshold or category requirements set out earlier in the clause the referring party would pay the costs of the arbitration incurred by the other party in excess of the costs that would have been incurred had the arbitration taken place in Pakistan.
Various procedural events led to Phillips J essentially having to decide: whether the parties had validly and lawfully chosen London as the seat of the arbitration (answer: yes); and whether, in light of Pakistani law (which was the law governing the contracts), the choice of London as the seat of arbitration did not result in the English courts having exclusive supervisory jurisdiction with the effect that the courts of Pakistan had at least concurrent jurisdiction (answer: no, for this would result in an unsatisfactory situation where more than one jurisdiction could entertain challenges to an award)
Variation of any litigation relevant articles really does open all sorts of cans of worms.
Dana Gas v Deutsche Bank et al. Islamic financing. Interest v usury (riba). Depecage, von Munchausen and overriding mandatory law. Partial unenforceability. All in the face of anti-suit.
In  EWHC 2928 (Comm) Dana Gas v Deutsche Bank et al., Leggatt J treats his readers to a concise insight into islamic finance (particularly in para 10) which he needs to do to inform readers of the essence of the case. The operation essentially involves raising investment (with a view to restructuring), organised by the main agreement (of the ‘Mudarabah’ type), subject to UAE law, and supported by a purchase undertaking of the same date, subject to English law. The set-up therefore evidently is not one of dépeçage per se (this would require one and the same agreement being subject to different laws) however it comes close.
Inevitably following unfavourable market conditions, an anti-suit injunction was sought and obtained in the UAE, followed however by English proceedings which required the aint-suit to be lifted – something which Dana Gas did not succeed in as a result of shareholder opposition. The English proceedings were effectively saved from collapse by the involvement of a third party, BlackRock, who as a non-party to the UAE sharia proceedings, were not bound by the anti-suit injunction. The somewhat complicated result is that the English proceedings really can only limp along.
Dana Gas seek confirmation that the transaction is unlawful and all the relevant contractual obligations are unenforceable as a matter of UAE law. Leggatt J with neither emotion nor hesitation refers essentially to Rome I’s universal application: the Mudarabah agreement is subject to UAE law and he is happy to assume it is invalid under UAE law – hence not enforceable by an English court. See in this respect Article 10(1) Rome I.
That however leaves the viability of the purchase undertaking. (at 46) The fact in and of itself that the contract or its performance would be regarded as invalid or unlawful under the law of some other country than England (for example, a country where one of the parties is domiciled or carries on business) is generally speaking irrelevant (reference is made to Kleinwort, Sons & Co v Ungarische Baumwolle Industrie AG  2 KB 678.
At 48, Dana Gas sets out its case for unenforceability of the purchase agreement under English law. This includes reference to ordre public but also inevitably an attempt to ‘contaminate’ the purchase agreement with the Mudarabah agreement. Leggatt J justifiably turns this around: at 54: it is apparent from the purchase agreement’s terms that the risks against which the Purchase Undertaking is intended to protect the Certificateholders include the risk that the mudarabah and the transaction documents governed by UAE law will turn out to be invalid. That is why they needed to be separated. (In that respect merging the two agreements into one and applying dépeçage might give even stronger force to this argument: however I do not know whether under UAE law such construction would be acceptable).
Further arguments swept aside, the Court turns to ordre public.
Dana Gas nb had employed both ordre public and, earlier Article 9(3) Rome I: overriding mandatory law: a rare treat indeed. Relevant English precedent is Ralli Brothers: Ralli Brothers v Cia Naviera Sota y Aznar  1 KB 614: an English court will not enforce an obligation which requires a party to do something which is unlawful by the law of the country in which the act has to be done. Rome’s Article 9(3) operates in a similar context. However Dana Gas later abandoned that claim for (at 80) those rules of law are only applicable if and in so far as the obligations in question have to be performed in the UAE – quod non.
A switch was then made to ordre public, now with Foster v Driscoll  1 KB 470 as leading precedent. However, here too, it is only if a contract has as its object and intention the performance in a friendly foreign country of an act which is illegal under the law of that country that the contract will be considered (at 82 in fine) contrary to English public policy.
Conclusion: the Purchase Undertaking is valid and enforceable.
Without claiming anything near proper competence in Islamic finance law, it would seem that Dana Gas does not introduce new principles in that area. However in diligently applying conflicts analysis, Leggatt J in my view does practice a great service: he re-emphasises the need for parties clearly to identify locus implementi: the place of performance of an obligation. When obligations are marked out for a seperate lex causae, such clear identification of place of performance will insulate them from collapse.
(Handbook) of Private International Law, 2nd ed. 2016: essentially, almost every section of Chapters 2 and 3.
It is one of the pinnacle theories of conflict of laws and when first introducing students to it, they almost invariably respond glassy-eyed. Renvoi has an unlimited ability to surprise parties and courts alike. It is best excluded, either by Statute, or by the parties, but frankly to be on the safe side: always and everywhere best by both. (Lest there are well considered arguments not to do so in a specific instance. As readers of my book know, the Brussels I Recast provisions on renvoi for choice of court (complicating less fori prorogati) is not such an instance: Handbook 2016, p.128-129, Heading 126.96.36.199.2).
At issue in Dankor [Dancor Construction, Inc. v. FXR Construction, Inc., 2016 IL App (2d) 150839] was the choice of court and governing law clause cited by the court at 44:
“The parties agree that this agreement was executed in Kane County, Illinois and shall be governed by the law of the State of Illinois. Any claims, lawsuits, disputes or claims arising out of or relating to this agreement shall be litigated in Kane County, Illinois.”
This clause could be a boilerplate or midnight clause except those routinely do exclude renvoi. ‘The law of the State of Illinois’ in the clause would then be followed by ‘excluding its choice of law rules’ or something of the kind. Why it was dropped here is entirely unclear. As Clifford Shapiro writes ‘So what happens when an Illinois general contractor fires a New York subcontractor who was working on a New York project under a subcontract that required Illinois law to apply and litigation to take place in Illinois? Unfortunately for litigants, what can happen is nearly three years of jurisdictional litigation in both New York and Illinois, and then dismissal of the Illinois case less than 60 days before trial with an order directing the case to be re-filed in New York.’
As the court notes (at 69) choice of court and choice of governing law are separate issues (for that reason they are als best dealt with in clearly separated contractual clauses). Relevant precedent for the validity of the former is Rieker 378 Ill. App. 3d 77, 86 (2007). Applying Rieker, and following Section 187(2) of the Restatement (Second) of Conflict of Laws, the Court held (reference is best made to Clifford’s summary or to the judgment itself) that New York law applied to the validity of the clause, leading to its being void: New York law mandatorily prohibits application of another State’s law or litigation outside of the State for New York construction projects (Illinois incidentally has a mirror provision).
Need one say more? Renvoi is always best excluded. It would not necessarily have made this clause enforceable: ordre public discussions could always still be raised. However it sure as anything would have made the validity of the clause much more likely.
(Handbook of) EU Private international law, 2nd ed. 2016, Chapter 1, Heading 1.4).
Thank you Cozen O’Connor for alerting me. California’s Senate Bill 1241 was signed into law at the end of September. It will apply to employment contracts entered into, modified, or extended on or after 1 January 2017.
The Bill will feature in a forthcoming article that I am co-authoring with Jutta Gangsted. I have not (yet) studied the preparatory work in detail however the Bill immediately calls for comparative analysis with the EU’s’ approach to this particular ‘protected category’: what is a labour (employment) contract; how does ‘primarily resides and works in California’ compare with ‘habitually carries out his work’ and ‘domicile’; when exactly is a contract ‘modified’ (on this see for the EU, Nikiforidis). The starting point of both the California and the EU rules is the same: employees cannot be considered to really consent to either choice of law or choice of court hence any clause doing same will be subject to mandatory limitations.
(Handbook of) European Private international law, 2nd ed. 2016. Chapter 2, Heading 188.8.131.52, Chapter 3, Heading 3.2.5.
Commission effectively supplements Rome I using the posted workers Directive. Defines ‘temporary employment’ as not exceeding 24 months.
Update 31 May 2017 A quick note by way of interim update: the proposal is stuck in Parliament (awaiting committee decision).
Thank you Fieke van Overbeeke for pointing this out to me. The EC have proposed to amend the posted workers Directive, to address unfair practices and promote the principle that the same work at the same place be remunerated in the same manner.
The amendment essentially relates to Article 8(2) of the Rome I Regulation, which partially corrects choice of law made in the context of contracts for employment. The proposal amounts to Union harmonisation of the concept ‘temporary employment’, as one not exceeding 24 months.
The proposal, if adopted, would insert an Article 2a in the posted workers Directive, 96/71, as follows:
Posting exceeding twenty-four months
1. When the anticipated or the effective duration of posting exceeds twenty-four
months, the Member State to whose territory a worker is posted shall be deemed to
be the country in which his or her work is habitually carried out.
2. For the purpose of paragraph 1, in case of replacement of posted workers
performing the same task at the same place, the cumulative duration of the posting
periods of the workers concerned shall be taken into account, with regard to workers
that are posted for an effective duration of at least six months.
Recitals 6-8 give context:
(6) The Rome I Regulation generally permits employers and employees to choose the law
applicable to the employment contract. However, the employee must not be deprived
of the protection of the mandatory rules of the law of the country in which or, failing
that, from which the employee habitually carries out his work. In the absence of
choice, the contract is governed by the law of the country in which or, failing that,
from which the employee habitually carries out his work in performance of the
(7) The Rome I Regulation provides that the country where the work is habitually carried
out shall not be deemed to have changed if he is temporarily employed in another
(8) In view of the long duration of certain posting assignments, it is necessary to provide
that, in case of posting lasting for periods higher than 24 months, the host Member
State is deemed to be the country in which the work is carried out. In accordance with
the principle of Rome I Regulation, the law of the host Member Sates therefore applies
to the employment contract of such posted workers if no other choice of law was made
by the parties. In case a different choice was made, it cannot, however, have the result
of depriving the employee of the protection afforded to him by provisions that cannot
be derogated from by agreement under the law of the host Member State. This should
apply from the start of the posting assignment whenever it is envisaged for more than
24 months and from the first day subsequent to the 24 months when it effectively
exceeds this duration. This rule does not affect the right of undertakings posting
workers to the territory of another Member State to invoke the freedom to provide
services in circumstances also where the posting exceeds 24 months. The purpose is
merely to create legal certainty in the application of the Rome I Regulation to a
specific situation, without amending that Regulation in any way. The employee will in
particular enjoy the protection and benefits pursuant to the Rome I Regulation.
It would obviously be attractive to ensure the same rule is verbatim included in a future amendment of the Rome I Regulation.
(Handbook of) EU Private International Law, 2nd ed 2016, Chapter 3, Heading 3.2.5.
Banco Santander Totta: the High Court upholds snowball interest rate swaps under English law. The ‘purely domestic contracts’ provision of Article 3(3) Rome I is not engaged.
Update December 2016 the Court of Appeal has confirmed in  EWCA Civ 1267.
A longer title than readers are used to from this blog. However judgment itself is also an unusually long 163 pages. In  EWHC 465 Banco Santander Totta, the High Court was asked whether snowball interest rates swaps in loan agreements between a Portuguese Bank and four Portuguese public transport companies, should be declared invalid under Portuguese ‘mandatory’ law, applicable by use of the corrective mechanism of Article 3(3) Rome I.
The Transport Companies do not assert that BST wrongly advised them to enter into the swaps, or misrepresented the swaps to them. Rather, defences raised by the Transport Companies are that:
(1) under Portuguese law, each company lacked capacity to enter the swaps which are therefore void; this is on the basis (among other reasons) of an assertion that the swaps were speculative transactions; this defence applies regardless of the law applicable to the swaps; it is common ground that, if correct, it is a complete answer to the claim;
(2) although English law governs the Master Agreements, this is subject to Art. 3(3) of the Rome Convention; this provides that where all the elements relevant to the situation at the time of the choice of law are connected with one country only, the choice does not prejudice the application of rules of the law of that country which cannot be derogated from by contract (“mandatory rules”). Portuguese mandatory rules apply to the swaps, giving rise to two defences: a) under rules dealing with gaming and betting and ordre public, the swaps are void for being unlawful “games of chance”, alternatively speculations; b) seven of the nine swaps are liable to be terminated under rules dealing with an “abnormal change of circumstances” (which termination takes effect as though the swaps were void); this is on the basis that since 2009 (following the financial crisis), the reference interest rates relating to the swaps (EURIBOR and LIBOR) have been close to zero (and remain so at the time of this judgment);
(3) in presenting the swaps to the Transport Companies, the bank acted in breach of its duties under provisions of the Portuguese Securities Code which implement relevant European Union legislation; these apply to the bank as a financial intermediary and relate to the protection of the legitimate interests of the Transport Companies as clients, and to conflicts of interest; the breach is said to entitle the Transport Companies to damages thereby extinguishing their liabilities under the swaps.
Blair J reviews precedent (European (limited, mostly related to the preparatory works), English and Portuguese (likewise limited) and decides against the engagement of Article 3(3). I will not regurgitate all of the analysis: readers are best referred to the judgment, in particular p.65 onwards, and the decision at 411, where Blair J concludes
because of the right to assign to a bank outside Portugal, the use of standard international documentation, the practical necessity for the relationship with a bank outside Portugal, the international nature of the swaps market in which the contracts were concluded, and the fact that back-toback (sic) contracts were concluded with a bank outside Portugal in circumstances in which such hedging arrangements are routine, the court’s conclusion is that Art. 3(3) of the Rome Convention is not engaged because all the elements relevant to the situation at the time of the choice were not connected with Portugal only. In short, these were not purely domestic contracts. Any other conclusion, the court believes, would undermine legal certainty.
The latter element is quite important. Referring in particular to Briggs (at 374), the Court holds that the uncertainty of the rule of Article 3(3) should lead to its narrow interpretation. I agree. With party autonomy the core consideration of the Regulation, standard recourse to Article 3(3) [or 3(4) for that matter) under the pretext for instance of a general campaign against fraus legis is most definitely not warranted.
Permission was granted to appeal the issues on the Rome Convention (thank you to Ali Malek QC for pointing that out).
(Handbook of) European Private International Law, 2nd ed. 2016, Chapter 3, Heading 3.2.8, Heading 184.108.40.206