Tilman v Unilever. CJEU supports choice of court in GTCs even if no possibility of click-wrap is offered.

Update 29 November 2022 see here for questions between Marco Farina and myself re the CJEU’s discussion 28-31 of the applicability at all of Lugano, in light of the Withdrawal Agreement.

The CJEU last week held in C-358/21 Tilman v Unilever, the context of which I reviewed here. Krzysztof Pacula has initial analysis here and also refers to the application of the consent for choice of court issues in Ebury Partners.

One of the parties’ (Unilever’s) GTCs  are contained on a website, and their existence is ‘flagged’ in the written main contrac, without there bring a tickable box that click-wraps the agreement. Does that suffice to bind the parties as to the GTC’s choice of court (in favour of the English courts)? Note the courts were seized pre-Brexit; the UK’s Lugano troubles are not engaged.

The CJEU answers exactly along the lines I suggested in my earlier post: no impeding of commercial practice; need for the contracting party relying on the clause to have drawn the attention to the GTCs; need for that clause to be durably consultable and storable; finally it is the national court’s task to verify  the formation of consent in these factual circumstances. That there is no box that can be ‘ticked’ is not conclusive [52].

All in all a welcome support for commercial choice of court.

Geert.

EU Private International Law, 3rd ed. 2021, Heading 2.2.10.

Ebuy Partners. Anti-suit viz Belgian proceedings re incorporation of e-mailed and /or hyperlinked general terms and conditions, with a serious miss on Rome I.

Ebury Partners Belgium SA/NV v Technical Touch BV & Anor [2022] EWHC 2927 (Comm) discusses ia whether choice of court and law included in general terms and conditions – GTCs, agreed (or not) by inclusion in email and /or e-mailed click-wrapeable hyperlink (this is a factual discussion), justifies an anti-suit injunction against Belgian proceedings.

Pre-Brexit such injunction would not have been possible. It has since of course been granted frequently; my most recent report of one was QBE Europe v Generali. Issuing an anti-suit post Brexit therefore is no longer surprising (commentators continue to suggest the EU should somehow shield EU proceedings from them). The application of the Rome I Regulation under retained EU law however does remain less discussed – and it is poorly executed in current judgment.

Anticipatory proceedings seeking a declaration of non-liability were launched unexpectedly (Belgian CPR requires no prior warning in any circumstance) in Belgium on 4 May 2022. The Belgian court later that month held that Ebury’s jurisdiction challenge  will not be dealt with separately, instead, as is standard, will be reserved for consideration at the same time as the merits.

The English proceedings were launched in July 2022. A critical question is whether Ebury can show, with a high degree of probability, that there is a jurisdiction agreement governing the dispute in question. Was the E&W jurisdiction clause contained in Ebury’s RA standard terms incorporated into the agreement between Ebury and TT? The factual circumstances are not conclusive, for there are suggestions of GTCS with choice of court sent by incorporation in an e-mail and /or by click-wrapeable  hyperlink similarly e-mailed.

The judge is correct to classify Rome I as retained law [83]. However the exclusion of choice of court agreements from that Regulation has somehow entirely escaped him and counsel, it seems.

Rather therefore than considering the issue under English conflict of laws (in EU Member States the issue is now subject to Article 25  Brussels Ia however that is irrelevant here), the judgment ventures into Article 10 Rome I’s putative law /von Munchausen /bootstrap principle, to identify English substantive law as the lex cause for the validity (including the issue of incorporation) of the choice of court. This leads after extensive discussion to a finding of incorporation under English law [102].

[103] ff Belgian law is signalled as a fall-back under Article 3(5) and 10(2) Rome I, however the judge essentially ignores that possibility (although he formally entertains it) by referring to a lack of indication on the facts that the counterparty agreed to the relevant clauses. He uses the ‘man on the Clapham omnibus’ formula to reach that conclusion: counterparty did consult or should have consulted the GTCs and there are no factual indications it disagreed with them. Conflicting Belgian law  expert evidence is not discussed.

Anti-suit was eventually granted.

If their apparent lack of raising the proper analysis (ie: no inclusion of choice of court) of the Rome I issue does not prevent defendants from appealing, they clearly should, to the extent the English conflict of laws approach to validity of choice of court, may lead to a finding of non-incorporation.

Geert.

Tilman v Unilever. A preliminary reference on flag-wrap B2B choice of court under Lugano.

A puzzling title perhaps I agree but let me explain. Thank you Matthias Storme for alerting me to the May 2021 preliminary reference by the Belgian Supreme Court, a reference now known at the CJEU as Case C-358/21 Tilman SA (of Belgium) v Unilever Supply Chain Company AG (of Switserland). Elucidation is asked of Article 23 of the Lugano 2007 Convention, the choice of court provision in the Convention.

The question referred, reads

Are the requirements under Article 23(1)(a) and (2) of [Lugano 2007], satisfied where a clause conferring jurisdiction is contained in general terms and conditions to which a contract concluded in writing refers by providing the hypertext link to a website, access to which allows those general terms and conditions to be viewed, downloaded and printed, without the party against whom that clause is enforced having been asked to accept those general terms and conditions by ticking a box on that website?

Article 23 Lugano 2007 is identical (mutatis mutandis: the only difference being that A23 Lugano refers to ‘States to the Convention’ instead of ‘Member States’) to the former Article 23 of the Brussels I Regulation, Regulation 44/2001.  A23 Lugano 2007 reads in relevant part

    1. If the parties, one or more of whom is domiciled in a State bound by this Convention, have agreed that a court or the courts of a State bound by this Convention are to have jurisdiction to settle any disputes which have arisen or which may arise in connection with a particular legal relationship, that court or those courts shall have jurisdiction. Such jurisdiction shall be exclusive unless the parties have agreed otherwise. Such an agreement conferring jurisdiction shall be either: (a) in writing or evidenced in writing; or (b) in a form which accords with practices which the parties have established between themselves; or (c) in international trade or commerce, in a form which accords with a usage of which the parties are or ought to have been aware and which in such trade or commerce is widely known to, and regularly observed by, parties to contracts of the type involved in the particular trade or commerce concerned.
    2. Any communication by electronic means which provides a durable record of the agreement shall be equivalent to ‘writing’.

The case at issue therefore does not question so-called ‘click-wrap’ consent to general terms and conditions – GTCs. These require the contracting partner to tick the relevant box which then ‘wraps up’ the agreement, including choice of court (and law). They were the subject of CJEU El Majdoub v CarsOnTheWeb. In that judgment, the CJEU held that in a B2B context, where the GTCs that have to be ticked can be saved and printed, they can be a ‘durable’ record of consent. (Not: consent itself: that is subject to a separate analysis, under the relevant lex causae, see below).

Rather, the title of this post calls the issue one of ‘flag-wrap’: one of the parties’ (Unilever’s) GTCs  are contained on a website, and their existence is ‘flagged’ in the written main contract. Does that suffice to bind the parties as to the GTC’s choice of court (in favour of the English courts; note the courts were seized pre-Brexit; the UK’s Lugano troubles are not engaged)?

The provisions on forum clauses are drafted in a way ‘not to impede commercial practice, yet at the same time to cancel out the effects of clauses in contracts which might go unread’ (Jenard Report), or otherwise ‘unnoticed’ (CJEU Colzani). The Brussels Convention and now the Regulation show great support for choice of court agreements and aim not to be as overly formalistic as the conditions imposed upon them.

Importantly, valid choice of court does require both a clearly and precisely demonstrated consent to be bound by choice of court and one or another Article 25-sanctioned form of expression of that consent. In Colzani the CJEU held [7]:

the requirements set out in Article [25] governing the validity of clauses conferring jurisdiction must be strictly construed. By making such validity subject to the existence of an ‘agreement’ between the parties, Article [25] imposed upon the court before which the matter is brought the duty of examining, first, whether the clause conferring jurisdiction upon it was in fact the subject of a consensus between the parties, which must be clearly and precisely demonstrated. The purpose of the formal requirements imposed by Article [25] is to ensure that the consensus between the parties is in fact established.

CJEU authority of Colzani and Coreck Maritime impose on the court the duty of examining ‘whether the clause conferring jurisdiction upon it was in fact the subject of a consensus between the parties’ and this had to be ‘clearly and precisely demonstrated’.

In practice, many courts conflate the check for consent with the check for expression of that consent and even the CJEU is not always clear in distinguishing it. In particular, absence of proof of any of the three possible avenues for expression of consent, included in Article 25(1) a, b or c, or then taken as an absence of consent, full stop. In Colzani, the CJEU held

[T]he mere fact that a clause conferring jurisdiction is printed among the general conditions of one of the parties on the reverse of a contract drawn up on the commercial paper of that party does not of itself satisfy the requirements of Article 17, since no guarantee is thereby given that the other party has really consented to the clause waiving the normal rules of jurisdiction. Where a clause conferring jurisdiction is included among the general conditions of sale of one of the parties, printed on the back of a contract, the requirement of a writing under the first paragraph of Article 17 of the Convention is fulfilled only if the contract signed by both parties contains an express reference to those general conditions.

The CJEU here, wrongly, seems to suggest lack of compliance with the expression of consent indicates a lack of that consent full stop.

Importantly, the CJEU in its rulings on what was then Article 23 and its Brussels Convention predecessor keeps utterly silent on national conditions relating to the actual formation or existence of consent. This, as regular readers of the blog will know, is at least for cases covered by Brussels Ia, subject to the lex fori prorogati, with renvoi, an issue which both national courts and the CJEU struggle with.

How then should the CJEU respond to the question (I asked my conflict of laws students at Leuven this question in a first exam on 18 June)?

Firstly, the Court should (and will) remind us of the Jenard /Colzani core instruction: the need to ensure consent is established, without being overly formalistic. Different from the context of the protected categories, there is no ‘weaker category’ to protect here.

Secondly,  there needs to be durability of the record of consent. That seems to be guaranteed here via the technicalities of the Unilever platform (downloadable GTCs) and in line with aforementioned CJEU Al Majdoub (the June students were not given technical details but should still flag durability).

Thirdly, despite the formal A23  requirement most probably being met, the consent requirement to me seems far from certain. In a click and wrap context ― lest there be issues of agency, duress, consumer protection laws etc. (in a context where the consumer title’s conditions are not met) which need to be held under the law applicable to consent ― the box ticking solidifies establishment of consent. In a mere flag and wrap context, that to me seems far less certain. If the reference were to a url where GTCs are properly and collectively displayed (if need be, updated with clear reference to chronology; see housekeeping), consent by an ordinary careful business (the proverbial (business)man on the Clapham omnibus). Yet if such as here, the link communicated in the formal contract refers to a platform where the  GTCs are not the first thing the contracting party sees, rather, where it is expected that that contracting party registers and /or downclicks, search and retrieve etc., that consent to me seems far less certainly established. [Again my students were not given the details on the platform which the reference includes, they did however have to signal the issue of consent).

Finally, under BIa, the lex fori prorogati, incl renvoi, would determine the above considerations of consent. Here, therefore, English law including its conflict of laws rules on choice of court. However seeing as the case is not subject to Brussels Ia, but rather to Lugano, the lex causae for consent will be an issue for the courts seized (here, the Belgian courts) to determine. Under the Belgian rules, this means application of Rome I (Rome I excludes choice of court agreements however Belgium’s private international law Act makes Rome I applicable even to carved-out contractual arrangements).

An interesting reference.

Geert.

EU Private International Law, 3rd ed. 2021, Heading 2.2.10.

 

The GDPR’s one stop shop principle put to the test in French Supreme Court confirmation of CNIL jurisdiction over Google Android case. The Court also rebukes the spaghetti bowl of consent ticking and unticking.

Thank you Gaetan Goldberg for flagging that the French Supreme Court has confirmed on 19 June last, jurisdiction of the French Data Protection Agency (‘DpA’), CNIL for issuing its fine (as well as confirming the fine itself) imposed on Google for the abuse of data obtained from Android users. The Court was invited to submit preliminary references to the CJEU on the one-stop shop principle of  the GPDR, but declined to do so.

Readers of the blog know that my interest in the GDPR lies in the jurisdictional issues – I trust date protection lawyers will have more to say on the judgment.

With respect to the one stop shop principle (see in particular A56 GDPR) the Court held at 5 ff that Google do not have a ‘main establishment’ in the EU at least not at the time of the fine complained of, given that the Irish Google office (the only candidate for being the ‘main establishment) at least at that time did not have effective control over the use and destination of the data that were being transferred – US Google offices pulling the strings on that decision. A call by the CNIL under the relevant EU procedure did not make any of the other DPAs come forward as wanting to co-ordinate the action.

On the issue of consent the SC referred to CJEU Cc-673/17 Planet49 and effectively held that the spaghetti bowl of consent, ticking and unticking of boxes which an Android user has to perform to link a Google account to Android and hence unlock crucial features of Android, do not amount to consent or proper compliance with GDPR requirements.

Geert.

Wiseley v Amazon: on consumer contracts, click-wrap and putative laws.

Thank you Jeffrey Neuburger for flagging Wiseley v Amazon in the US Federal Court of Appeal (9th circuit). Jeffrey has excellent overview and analysis so I will suffice with identifying a few tags: the issue of click-wrap agreements (when does one agree to GTCs contained in pop-ups and hyperlinks and the like); application of a putable law to a contract (the von Munchausen or ‘bootstrap’ principle); comparative dispute resolution law: how would EU law look at the issues? Have fun.

Geert.

 

Choice of court on the web . The ECJ on ‘click-wrap’ in El Majdoub v CarsOnTheWeb.

I have delayed reporting on judgment in Case C-322/14, Jaouad El Majdoub v CarsOnTheWeb.Deutschland GmbH, held 21 May 2015, for exam reasons. I reported earlier on the due diligence required of businesses when establishing choice of court through electronic means. The ECJ has now also had its say, in a case concerning a B2B contract for the purchase of a car. [Choice of court in a B2C context tends to be covered by the consumer contracts title hence is not at stake here. [Mark Young and Philipe Bradley-Schmieg review the relevance of the case for B2C contracts here].

Choice of court allegedly had been made in favour of the courts at Leuven, Belgium, in the vicinity of which the seller’s parent company has its head office. The buyer however sued in Germany, the domicile of the German daughter company (and of the buyer, a car dealer). Buyer claims that the  contract at any rate was with the daughter company, not the mother company, and that choice of court had not been validly made. He submits that the webpage containing the general terms and conditions of sale of the defendant in the main proceedings does not open automatically upon registration and upon every individual sale. Instead, a box with the indication ‘click here to open the conditions of delivery and payment in a new window’ must be clicked on (known as ‘click wrapping’).

In essence therefore the question is whether the requirements of Article 23(2) of the Brussels I Regulation (now Article 25(2)) are met only if the window containing those general conditions opens automatically, and upon every sale. That Article was added at the adoption of the  Brussels I Regulation, precisely to address the then newish trend of agreeing to choice of court (and indeed choice of law; but that is not covered by Brussels I) through electronic means.

The provisions on forum clauses in the 1968 Brussels Convention, Brussels I and the recast are drafted in a way ‘not to impede commercial practice, yet at the same time to cancel out the effects of clauses in contracts which might go unread’ (Report Jenard) or otherwise ‘unnoticed’ (the ECJ in the core case Colzani). the Report Jenard also notes that in order to ensure legal certainty, the formal requirements applicable to agreements conferring jurisdiction should be expressly prescribed, but that ‘excessive formality which is incompatible with commercial practice‘ should be avoided.

The first sentence of Article 25(1) discusses the parties ‘agreement’ as to choice of court. (It leaves a large array of national law issues untouched, such as consideration, mandate, 3rd party effect. etc. On some of those issues, see also Refcomp). The remainder of Article 25(1) concerns the possible formats in which agreement is testified. Article 25(2) (and 23(2) before it) accompanies Article 25(1) a’s option of having the agreement put down ‘in writing’.

In line with the requirement not to be excessively formalistic, the ECJ essentially requires that parties be duly diligent when agreeing to choice of court. If click-wrapping makes it possible to print and save the text of those terms and conditions before the conclusion of the contract, then it can be considered a communication by electronic means which provides a durable record of the agreement.

Note that the Court does not hold on whether the agreement is actually reached between the parties: only that click-wrap may provide a durable record of such agreement, where it exists. (One could imagine choice of court having been protested, for instance, or other issues of national law having an impact on the actual existence of the agreement. and one can certainly imagine a continuing discussion on what contract was concluded between what parties in the case at issue].

Geert.

%d bloggers like this: