The Court of Justice delivered last week (10 December 2020) its judgment in case C‑80/19 (A. B., B. B. v Personal Exchange International Limited), which is about Brussels I. The judgment is currently available in all EU official languages (save Irish), albeit not in English. Here is the French version (to check whether an English translation has finally been made available, just click on the link below and change the language version):
“L’article 15, paragraphe 1, du règlement (CE) no 44/2001 […] doit être interprété en ce sens qu’une personne physique domiciliée dans un État membre qui, d’une part, a conclu avec une société établie dans un autre État membre un contrat pour jouer au poker sur Internet, contenant des conditions générales déterminées par cette dernière, et, d’autre part, n’a ni officiellement déclaré une telle activité ni offert cette activité à des tiers en tant que service payant ne perd pas la qualité de « consommateur » au sens de cette disposition, même si cette personne joue à ce jeu un grand nombre d’heures par jour, possède des connaissances étendues et perçoit des gains importants issus de ce jeu ».
The Minerva Center for Human Rights at Tel Aviv University will host an international socio-legal (zoom-) workshop on 22-23 June 2021 to explore the impact of the Covid-19 crisis and its regulation on cross-border families:
Cross-border families (also known as transnational and globordered families) are a growing and diverse phenomenon. People around the globe create bi-national spousal relations, are assisted by cross-border reproduction services, or by a migrant care worker who provides care for a dependent family member. Likewise, families become cross-bordered when one of the parents relocates, with or without the child, or when a parent abducts the child. In addition, increasing rates of forced or voluntary migration create more and more cross-border families, with different characteristics and needs. While some kinds of cross-border families have attracted the attention of legal scholarship, other kinds are still neglected, and much is yet to be studied and discussed regarding the challenges embedded in the attempt to secure the right to family life in the age of globalization.
The global Covid-19 crisis provides more, and alarming, evidence of the socio-legal vulnerabilities of cross-border families. For example, bi-national couples are separated for long periods of time; intended parents are unable to collect their baby from the country of the surrogate; and families assisted by a migrant care worker, the workers, and their left behind families, are entangled in new complex relations of power and dependency. Likewise, the right to heath is at risk when a family member is denied treatment because of partial citizenship status, and questions such as the enforcement of child support across borders are even harder to address than in more peaceful times.
Crises, such as the Covid-19 pandemic, are often a methodological opportunity for socio-legal research. In many cases, a major social crisis shakes habitualization, and opens up taken for granted social scripts to individual and collective reflection. Likewise, such a crisis involves risk regulation and, in the current case, also plague governance—involving intense emergency regulative changes made by different nation-states that might both reveal and challenge deeply shared norms regarding familial rights and national interests. Hence, our current era lends itself more readily than stable, routinized periods to the investigation of current regulation, and the imagining of options for new regulation regarding cross-border families.
On 22-23 June 2021, we plan an international socio-legal workshop that will explore the impact of the Covid-19 crisis and its regulation on cross-border families. We hope to explore the ways Covid-19 restrictions affect cross-border families, and the role of the law, in different countries, in shaping this impact and in challenging it.
The questions during the workshop might include, but are not limited to:
Confirmed Keynote Speaker: Prof. Yuko Nishitani, Kyoto Univiertys Law School
The workshop will be conducted via Zoom, and is sponsored by the Minerva Center for Human Rights at Tel Aviv University. It will be open to the public, and hopefully, will set the foundations for further multinational research and collaboration.
We will give serious consideration to all high-quality relevant research, from any discipline. Work in progress is welcome, as long as the presentation holds new findings or insights and not only declaration of intent. Faculty members as well as independent researchers and advanced research students are welcome to submit.
The screening process for the workshop will include two phases:
Phase I – Abstract:
Abstracts should include:
Abstracts must be in English and be submitted to this email address: eynatm@media-authority.com
Deadline for submission: 28 February 2021.
Phase II – Summary:
Those who’s abstract will be accepted, will be notified by 31 March 2021 and will be asked to submit a 3-pages summary of their paper by 20 April 2021. Accepted papers will be presented at the workshop. Presenters are expected to take part in all the workshop’s sessions.
Academic Organizers:
Prof. Daphna Hacker, Law Faculty and Gender Studies Program, Tel Aviv University; Prof. Paul Beaumont, Law Faculty, University of Stirling; Prof. Katharina Boele-Woelki, Bucerius Law School, Hamburg; Prof. Sylvie Fogiel-Bijaoui, The College of Management Academic Studies; Dr. Imen Gallala-Arndt, Max Planck Institute for Social Anthropology; Dr. Sharon Shakargy, Faculty of Law, Hebrew University; Prof. Zvi Triger, Law School, The College of Management Academic Studies
On 18 and 19 March 2021, the third German Conference for Young Private International Law Scholars will be held at the Max Planck Institute for Comparative and International Private Law in Hamburg (via Zoom). The conference is dedicated to the question “PIL for a better world: Vision – Reality – Aberration?” The programme, as well as further information on how to register, can be found here. Twitter: @IPR_Nachwuchs
Ness Global Services Ltd v Perform Content Services Ltd [2020] EWHC 3394 (Comm) engages Articles 33-34 of the Brussels Ia Regulation, its so-called forum non conveniens light regime. I reported on it before of course, most recently re Municipio de Mariana in which the judge arguably failed to engage with BIa properly (making A33-34 a carbon copy of abuse and /or forum non arguments in my view is noli sequi).
Perform and Ness are UK-registered companies with offices in London. Perform are defendants in the UK action. Ness Global Services and its parent Ness Technologies Inc are defendants in parallel proceedings in New Jersey. Both sets of proceedings are based on the same facts and matters. These are said to constitute the basis for termination by both sides of a written agreement.
Ness argue application of A33-34 must be dismissed for there is non-exclusive choice of court in favour of England which, it argues, makes the A33-34 threshold very high. (The clause reads ‘”Governing Law and Jurisdiction. The Agreement shall be governed by and construed in accordance with the laws of England and Wales and the parties hereby irrevocably submit to the non-exclusive jurisdiction of the Courts of England and Wales as regards any claim, dispute or matter arising under or in connection with this Agreement.”)
Houseman J introduces BIa’s scheme clearly and concisely, using the excellent Adrian Briggs’ suggestion of there being a hidden hierarchy in the Regulation – which in my Handbook I have also adopted (clearly with reference to prof Briggs) as the ‘jurisdictional matrix’. Houseman J at 39 notes that non-exclusive jurisdiction is hardly discussed in the Regulation. and concludes on that issue ‘If the internal hierarchy is “hidden” then is fair to say that the concept of non-exclusive prorogated jurisdiction is enigmatic and elusive. It is The Scarlet Pimpernel of the Regulation.’ Later non-EJA is used as shorthand for non-exclusive jurisdiction agreement.
At 62 after consideration of the reflexive application of exclusive jurisdictional rules, including choice of court, the text of A33-34, and recital 24, the judge considers that the recital
focusses upon connections with the ‘first seised’ Non-Member State, rather than the ‘second seised’ Member State which is applying Article 33 or Article 34. This is conspicuous notwithstanding the fact that the jurisdictional gateway language presupposes some connection between either the defendant (domicile) or the circumstances of the case (special jurisdiction) and the ‘second seised’ forum. Further, there is no obvious room in this wording for accommodating or giving effect to a Non-EJA in favour of the courts of the latter forum, and no warrant for affording it the significance that it would receive under English private international law principles, as noted below. In contrast, the second paragraph of the recital appears to contemplate the conferral of exclusive prorogated jurisdiction (albeit reflexively) in favour of the ‘first seised’ Non-Member State, as noted above.
At 80, Houseman J emphasises that in his view the internal hierarchy of the Regulation (the matrix) has no direct role to play in interpreting or applying the gateway language in A33-34. Those articles are themselves part of such hierarchy and are themselves a derogation from the basic rule of domiciliary jurisdiction. He then refers in some support to UCP v Nectrus (reference could also have been made to Citicorp) to hold at 95 that
where Article 25 operates to confer prorogated jurisdiction upon the courts of the ‘second seised’ Member State, whether exclusive or non-exclusive, Articles 33 and 34 are not applicable. In such a case it cannot be said that the court’s jurisdiction is “based upon” Article 4.
A suggestion at 96 that in such case A33-34 can apply reflexively is justifiably rejected.
At 109 application of A33-34 had they been engaged is declined obiter as being not in the interest of proper administration of justice. At 107 mere reference, neither approving nor disapproving was made ia to Municipio de Mariana which effectively places the Articles on a forum non footing. At 112 it is held obiter
Without engaging in a full granular balancing exercise, given that this is a hypothetical inquiry in the present case, I am not persuaded that it is or would have been necessary for the proper administration of justice to stay these proceedings in favour of the NJ Proceedings. The parties bargained for or at any rate accepted the risk of jurisdictional fragmentation and multiplicity of proceedings by agreeing clause 20(f). That risk has manifested, largely through the tactical choice made by Perform to commence proceedings pre-emptively in New Jersey. The continuation of these proceedings, notwithstanding the existence of the NJ Proceedings, is a foreseeable consequence of the parties’ free bargain and a risk that Perform courted by suing first elsewhere.
An interesting addition to the scant A33-34 case-law, in an area this time of purely commercial litigation.
Geert.
European Private International Law, 3rd ed. 2021, 2.539 ff.
Application (dismissed) for a stay on the basis of Articles 33-34 Brussels Ia, 'forum non conveniens light'. https://t.co/gwl3B5y3hl
— Geert Van Calster (@GAVClaw) December 11, 2020
Circulation routière
Diffamation - Partie civile
Juridictions correctionnelles
Written by Anthony Kennedy, Barrister at Serle Court
This is the second online symposium on Private International Law in Nigeria initially announced on this blog. It was published today on Afronomicslaw.org. The first introductory symposium was published here by Chukwuma Samuel Adesina Okoli and Richard Frimpong Oppong. More blog posts on this online symposium will follow this week.
Introduction
Authority exists for the proposition that a creditor of a foreign judgment may bring an action at common law in Nigeria, by which action he, in effect, seeks recognition and/or “enforcement” of that foreign judgment[1]. The common law action has not been abolished by statute or disapproved judicially but, sadly, it is not widely understood or used by practitioners/courts in Nigeria. This is unfortunate, especially where the statutory mechanism[2] for the enforcement of foreign judgments is certainly limited but otherwise shrouded in confusion[3]. This paper argues for a reawakening of the common law action.
The construction placed on the statutory regime
It is impossible properly to assess the scope for the common law action in Nigeria without first addressing the statutory mechanism for the enforcement of foreign judgments. The common law action only works in the space which has been left for it by the applicable statutory regime. Moreover, tactically, judgment creditors are likely to favour registration of the foreign judgment under that statutory regime, where such registration is permitted, given the “better protection” which such regime affords them, at least theoretically, when compared with the common law[4].
With that in mind, the authorities yield the following propositions:
Difficulties generated by the legal profession’s approach to proposition 1
Proposition 1 holds, it is submitted, and, of itself, generates no difficulty for the continued existence and/or growth of the common law action. That said, the legal profession’s approach (and that of the courts) to proposition 1 has been problematic.
Two points are worth making here; both are demonstrative of problems which beset the current state of the law. First, insufficient attention has been paid to the consequence of proposition 1, meaning that its import has not been fully understood. This may, of course, be the result of practitioners and judges concentrating on establishing and endorsing proposition 1 (which process is still ongoing, given the difficult relationship between proposition 1 and proposition 3[9]). Even so, difficulty remains. By way of example: Section 3(1) of the 1922 Ordinance provides that an application for registration be made “at any time within twelve months after the date of the judgment, or such longer period as may be allowed by the court…” (italics added). Where one would have expected argument as to why the court should have, in the legitimate exercise of its discretion, extended the time within which the application could be made, one finds none[10]. A chance to establish when a judgment creditor might appeal to the court’s discretion[11], and, correlatively, when he might have to fall back on the common law action, has been missed.
The second point follows from the first. While, as noted, tactically less advantageous than registration under the statutory regime, Section 3(4) of the 1922 Ordinance allows a judgment creditor to bring an action at common law on the foreign judgment, rather than have it registered under the 1922 Ordinance itself[12]. In circumstances where courts have not heard substantial argument on the Section 3(1) discretion and/or have exhibited a hostile attitude towards extending the time within which to make the application for registration[13], one would have expected a much greater role carved out for the common law action; one remains disappointed. And doubly so because, while not free from all controversy[14], the common law action may be brought within a longer period of time than the 1922 Ordinance permits (if one discounts the fact that the court may, at its discretion, extend time thereunder). At a stroke – so long as the judgment debtor could demonstrate that the other requirements had been met[15] – reliance on the common law action would remove the judgment creditor’s need to act as swiftly as the 1922 Ordinance has been made to require[16].
Difficulties generated directly by proposition 3
The language in which Section 10(a) of the 1961 Act is couched gives rise to similar problems as those described when dealing with the first point under the previous sub-heading. Leaving those to the side, it is the second sentence of proposition 3 which poses the most significant risk to the continued life (such as it is) of the common law action in Nigeria. If “registration” is contemplated (or somehow required) when dealing with judgments from jurisdictions to which the statutory regime has not been extended, the common law action (which has nothing at all to do with “registration” of a foreign judgment) is rendered completely useless[17].
Several cases may be cited which combine to paint a rather gloomy picture in this regard. Teleglobe America Inc v 21st Century Technologies[18]is, as far as one can tell, the judiciary’s first (and so most egregious) brushstroke but others have since been added[19]. Taken collectively, they suggest that there is no room left for the common law action, even though there is Supreme Court authority which suggests that the statutory regime was not designed to kill it off.
To be sure, the statutory regime, properly construed, applies only to foreign judgments from a narrow field of jurisdictions. If this is thought to be a problem, the answer does not lie, it is submitted, in a distorted interpretation and application of that statutory regime. Supplementing (a narrow) statutory regime by allowing a judgment creditor to resort to the common law action makes sense: it recognises that the necessary reciprocity which underpins the statutory regime is absent in the majority of circumstances while, at the same time, preserving the judgment creditor’s ability to obtain the debt which the judgment debtor is said to owe, at least in circumstances where Nigerian legal policy (as set out in the rules which govern the common law action) thinks that he should. Judgments which treat this idea with kindness, or at least do not dismiss it out of hand, are to be welcomed[20].
Difficulties generated indirectly by proposition 3
If proposition 3’s formulation is the product of perceived problems either with the statutory regime or the common law action itself, that is most unfortunate. For, rather than alleviating those problems, proposition 3 rather ensures that they will continue, at least until action on the part of the legislator (which action appears to be some way off).
Removing the common law action from view means that the rules which govern that common law action cannot be changed by the judiciary (which change might allow certain kinks within those rules to be ironed out). To be sure, the common law world has not stood still in relation to the enforcement of foreign judgments: interesting questions remain to be explored. For instance, there is ongoing debate as to the circumstances in which a foreign court should be accorded international jurisdiction over the judgment debtor[21] and different views have been expressed regarding whether the common law may be used to enforce judgments from supra-national tribunals[22]. Consideration of these questions in Nigeria has been stymied by the side-lining of the common law action.
Perhaps even more importantly, with an eye to the future, deliberately obscuring the common law action prevents one from taking a clear view of the current Nigerian legal system, insofar as it relates to the enforcement of foreign judgments, and so, in turn, prevents an assessment of the merits of signing up to international projects, like the recent (and still draft) Hague Convention on Recognition and Enforcement of Foreign Judgments in Civil or Commercial Matters. The answer to such questions is of supreme importance if Nigeria wishes to attract (legal) business from the continent as a whole. Those answers must be reached using all of the information available, which is why the common law action must somehow be revived.
[1]Alfred C Toepfer Inc v Edokpolor (1965) NCLR 89. More recently, see: Wilbros West Africa v Mcdonnel Contract Mining Ltd (2015) All FWLR 310.
[2] The overarching statutory regime for enforcement of foreign judgments comprises the Reciprocal Enforcement of Foreign Judgments Ordinance 1922 and the Foreign Judgments (Reciprocal Enforcement) Act 1961.
[3] Olawoyin, Enforcement of Foreign Judgments in Nigeria: Statutory Dualism and Disharmony of Law (2014) 10 JIPL 129, 140.
[4]CSA Okoli and RF Oppong, Private International Law in Nigeria (Hart, 2020) 360.
[5]Macaulay v RZB of Austria (2003) 18 NWLR 282.
[6]Marine & General Assurance Company Plc v Overseas Union Insurance Ltd (2006) 4 NWLR 622; Grosvenor Casinos Ltd v Ghassan Halaoui (2009) 10 NWLR 309.
[7]Witt & Busch Ltd v Dale Power Systems PLC (2007) 17 NWLR 1.
[8]Teleglobe America Inc v 21st Century Technologies Ltd (2008) 17 NWLR 108.
[9]VAB Petroleum Inc v Momah (2013) 14 NWLR 284.
[10] By way of example, see Macaulay, supra fn. 5, and Marine & General, supra fn. 6.
[11] For a recent example of when an English court might be likely to exercise a similar discretion, see: Berhad v Frazer-Nash Research Ltd[2018] EWHC 1848 (QB).
[12] Though, where he does so, he is subject to a costs penalty where the conditions in Section 3(4) of the 1922 Ordinance are not satisfied.
[13] See the cases cited supra fn. 10.
[14] Compare, for instance, the competing views of the proper period of limitation as expressed by Olaniyan, The Commonwealth model and conundrum in the enforcement of foreign judgement regime in Nigeria, (2014) Commonwealth Law Bulletin, 40:1, 76, 88 (who seemingly advocates a standard 12 year time limit) and Okoli and Frimpong Oppong “Private International Law in Nigeria”(Hart Publishing, 2020), at 358-359 (who state that it depends on the state of Nigeria in which the action is brought).
[15] As to which, see Okoli and Frimpong Oppong, supra fn. 14 at 351-358.
[16]In this respect, Ogbuagbu JSC’s judgment in Grosvenor, supra fn. 6, at 334-335 is particularly disappointing. See: Okoli and Frimpong Oppong, supra fn. 14, at 373.
[17] See, to similar effect, Olaniyan, supra fn. 14, 88.
[18]Supra fn. 8.
[19] See, inter alia, African Reinsurance Corp v Gilar Cosmetic Store (2010) All FWLR 1194 (concerning a judgment from Liberia) and Obasi v Mikson Establishment Industries Ltd (2016) 16 NWLR 335 (concerning a judgment from Niger).
[20]Wilbros, supra fn. 1. Even there, however, Counsel took great pains to say that this was not an attempt to enforce a foreign judgment and the reasoning of the court in relation to that submission is not always easy to understand.
[21] Compare the position adopted by the Supreme Court of Canada in (originally)Beals v Saldanha 2003 SCC 72 and (more recently) Club Resorts Ltd v Van Breda [2012] 1 SCR 572 with that of the English Supreme Court in Rubin v Eurofinance SA [2012] UKSC 46.
[22] Compare the decision of the South African Constitutional Court in Government of the Republic of Zimbabwe v Fick [2013] ZACC 22 with that of the Ghanaian Supreme Court in Republic v High Court (Commercial Division) Accra, ex parte AG NML Capital and Argentina, Civil Motion No J5/10/2013. For a Nigerian perspective, see: Adigun, Enforcing ECOWAS judgments in Nigeria through the Common Law Rule on the Enforcement of Foreign Judgments (2019) 15 JIPL 130.
Responsabilité du fait des produits défectueux (Monsanto)
Directive (EU) 2020/1828 of 25 November 2020 on representative actions for the protection of the collective interests of consumers and repealing Directive 2009/22/EC was published on 4 December 2020 (OJ L 409/1).
It consists of 79 recitals, 26 provisions (some of them actually looking as recitals, and vice-versa), and two Annexes, the second being a correlation table of provisions – to my mind, a good lawmaking practice. It will enter into force on the twentieth day following that of its publication in the Official Journal of the European Union; transposition shall be ready by 25 December 2022; the national measures will apply from 25 June 2023.
The Key Features of the Directive in a NutshellThe Directive is based on Article 114 TFEU (see nevertheless Recital 76, arguing that the intervention of the EU is necessary due to the relevance of the cross-border element).
Having regard to its overarching objective, the Directive joins the group of measures aiming simultaneously at the protection of the consumer, the promotion of fairer competition and the creation of a level playing field for traders operating in the internal market (in other words, this piece of legislation has not been conceived only for the sake of consumers; therefore, it is here submitted that it should not be interpreted only with them in mind).
The Directive takes up the failures of precedent legal acts in relation to the enforcement of consumer law, particularly Directive 2009/22/EC on injunctions for the protection of consumers’ interests. As in other fields of EU law, it clearly endorses private enforcement.
According to the Directive, a consumer is any natural person who acts for purposes outside that person’s trade, business, craft or profession, independently of how she is referred to (data subject, traveler, retail investor, etc) in the legal act allegedly infringed.
The Directive covers infringements of the provisions of Union law referred to in its Annex I, to the extent that they protect the interests of consumers, and provided the natural person involved acts as such.
For those willing to get the whole picture as to the scope of the Directive it is worth noticing that it does not substitute the enforcement mechanism contained in the EU legal acts listed in Annex I (by the way: keep in mind that Annex is subject to be amended each time that a new Union act relevant to the protection of the collective interests of consumers is adopted). In addition, Member States are able to retain or introduce national legislation that corresponds to provisions of this Directive in relation to disputes that fall outside the scope of Annex I. My understanding is that this possibility relates only to subject matters, and not to the subjective scope of the Directive.
The impact of the Directive on national systems will depend on the Member State concerned. National procedural mechanisms for the protection of collective or individual consumer interests – where they exist – do not need to be replaced. It is for the Member States to decide whether the procedural mechanisms for representative actions required by the Directive are part of an existing procedural mechanism for collective injunctive measures or redress measures, or a distinct procedural mechanism.
What matters is that at least one national procedural mechanism for representative actions complies with the Directive in every Member State, and, as a consequence, at least one effective and efficient procedural mechanism for representative actions, for injunctive measures and for redress measures, is available to consumers in all Member States.
In terms of contents, the Directive does not address every aspect of the proceedings. Already whether these should be judicial or administrative, or both, is to be decided by the Member States considering the area of law or the economic sector at stake.
The provisions eventually adopted focus on legal standing (and its mutual recognition), remedies (injunctive relief, redress measures), funding, settlements, allocation of costs, information about representative actions, effects of final decisions, limitation periods, disclosure of evidence, and penalties.
It is for the Member States to lay down the rules complementing those of the Directive under the principle procedural autonomy, subject to the requirements of effectiveness and non-discrimination.
Aspects of the Directive of Interest for PILThe Directive does not intend to affect the application of rules of private international law regarding jurisdiction, the recognition and enforcement of judgments or applicable law, nor establish such rules (in other words, the well-known shortcomings of the existing PIL instruments are not remedied).
This intention has not prevented the lawmaker from shaping two cross-border categories: ‘cross-border infringements’, and ‘cross-border representative actions’.
The former covers ‘in particular’ the case of consumers affected by an infringement who live in Member States other than the Member State in which the infringing trader is established; what else is included is unclear. The latter designates the situation of a qualified entity bringing a representative action in a Member State other than that in which it is designated; conversely, if a qualified entity brings a representative action in the Member State in which it is designated, that representative action will qualify as a domestic representative action, even if it is brought against a trader domiciled in another Member State and even if consumers from several Member States are represented within that representative action. I would argue here that those categories have no meaning beyond the Directive itself; in other words, they should be accorded no significance in terms of application of PIL instruments.
Recital 22, according to which ‘It should be noted that Regulation (EU) No 1215/2012 does not cover the competence of administrative authorities or the recognition or enforcement of decisions by such authorities’, deserves a similar assessment. In my view, whether a representative action filed by an administrative authority falls under the scope of the Brussels I Regulation or not still depends on the autonomous characterization of the dispute as ‘civil and commercial’.
In addition to the clarification regarding PIL instruments, the following issues of interest for cross-border disputes are addressed in the Directive (not necessarily in the operational part).
Useful information for the courts – When bringing a representative action, a qualified entity should provide sufficient information on the consumers concerned by the representative action to the court or the administrative authority, thus allowing the court or administrative authority to determine whether it has jurisdiction and to determine the applicable law.
Useful information for the consumers – Member States should be able to set up national electronic databases that are publicly accessible through websites providing information on the qualified entities designated for the purpose of bringing domestic representative actions and cross-border representative actions, as well as general information on ongoing and concluded representative actions.
Legal standing criteria – For the purposes of cross-border representative actions, qualified entities should be subject to the same criteria for designation across the Union. Examples are listed – non exhaustively – under recital 25. Moreover, qualified entities that have been designated on an ad hoc basis are not allowed to bring cross-border representative actions.
Mutual recognition – Member States should ensure that cross-border representative actions can be brought before their courts or administrative authorities by qualified entities that have been designated for the purpose of such representative actions in another Member State. The identity of qualified entities enabled to sue abroad will be communicated to the Commission, who will compile a list and make it publicly available. Inclusion on the list serves as proof of the legal standing of the qualified entity bringing the representative action.
Opt-in – In order to ensure the sound administration of justice and to avoid irreconcilable judgments, where the consumers affected by an infringement do not habitually reside in the Member State of the court or administrative authority before which the representative action is brought, an opt-out mechanism is excluded regarding representative actions for redress measures. In other words, consumers have to explicitly express their wish to be represented in that representative action in order to be bound by the outcome of the representative action.
Cooperation and the exchange of information between qualified entities from different Member States is encouraged, in order to increase the use of representative actions with cross-border implications.
In what appears to be a rather straightforward extension of the Court’s earlier decisions in Cases C-498/16 Schrems and C-208/18 Petruchová, the CJEU held last week in Case C-774/19 Personal Exchange that a natural person contracting with the operator of an online gambling service remains a consumer in the sense of Article 15 of Regulation 44/2001 (Brussels I; now Article 17 or Regulation 1215/2012 (Brussels Ia)) even if they use it for many hours a day and make their living from it.
The question was referred in the context of a dispute between a user and the operator of a gambling website, concerning the alleged violation of certain rules by the user and the subsequent deletion of their account by the operator. The user, who had been playing about 9 hours of online poker per weekday for several years and won considerable sums of money from it, had brought an action in Slovenia although the contract, which the user had agreed to upon registration on the website, contained an exclusive jurisdiction clause in favour of Malta. As the defendant had not disputed that the service had been directed to Slovenia in the sense of Article 15(1)(c) Brussels I, the validity of the jurisdiction clause under Article 17 Brussels I (now Article 19 Brussels Ia) depended entirely on whether or not the user could still be considered a consumer in the sense of Article 15.
Although the Court begins its answer by the usual emphasis on the exceptional character and the resulting need for a narrow interpretation of Articles 15–17 (para. 24) and on the need to look at the purpose of the contract in question (para. 31), the rest of the decision consists entirely of a discussion of whether the claimant may have lost (“peut se voir refuser”) the quality of a consumer for the different reasons invoked by the defendant. According to the Court, this was not the case, neither in light of the sums earned by the claimant (paras. 33–36; as already decided in Petruchová) nor in light of their expertise (paras. 37–40; as already decided in Schrems). While the Court acknowledges the need for a dynamic interpretation of Article 15 (paras. 41–42), by which a party entering into a contract as a consumer could theoretically lose this status at a later point in the contractual relationship, the need for predictability prevents the consideration of the aforementioned factors.
In what Geert van Calster rightly describes as a confusing reference to substantive consumer law, the Court then also discusses to what extent the frequency and volume of the activity must be taken into account (paras 43–49). Seemingly seeking to align its decision with the one in Case C-105/17 Kamenova, where this factors had been considered, the Court qualifies it by holding that the claimant consumer does not lose this quality as long as they do not offer their services as a poker player for remuneration or formally register them.
L’article 7, point 2, du règlement Bruxelles I bis s’applique à une action fondée sur une allégation d’abus de position dominante et visant à faire cesser certains agissements mis en œuvre dans le cadre d’une relation contractuelle liant la société Booking.com à une société exploitant un hôtel.
Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer