Written by Madeleine Petersen Weiner/Marc-Philippe Weller

In this article, we critically assess the question of where to locate the “event giving rise to the damage” under Art. 7 Rome II in CO₂ reduction claims. This controversial – but often overlooked – question has recently been given new grounds for discussion in the much discussed “Milieudefensie et al. v. Shell” case before the Dutch district court in The Hague. In this judgment, the court had to determine the law applicable to an NGO’s climate reduction claim against Royal Dutch Shell. The court ruled that Dutch law was applicable as the law of the place where the damage occurred under Art. 4 (1) Rome II and the law of the event giving rise to the damage under Art. 7 Rome II as the place where the business decision was made, i.e., at the Dutch headquarters. Since according to the district court both options – the place of the event where the damage occurred and the event giving rise to the damage – pointed to Dutch law, this question was ultimately not decisive.

However, we argue that it is worth taking a closer look at the question of where to locate the event giving rise to the damage for two reasons: First, in doing so, the court has departed from the practice of interpreting the event giving rise to the damage under Art. 7 Rome II in jurisprudence and scholarship to date. Second, we propose another approach that we deem to be more appropriate regarding the general principles of proximity and legal certainty in choice of law.

1. Shell – the judgment that set the ball rolling (again)

The Dutch environmental NGO Milieudefensie and others, which had standing under Dutch law before national courts for the protection of environmental damage claims, made a claim against the Shell group’s parent company based in the Netherlands with the aim of obliging Shell to reduce its CO₂ emissions. According to the plaintiffs, Shell’s CO₂ emissions constituted an unlawful act. The Dutch district court agreed with this line of reasoning, assuming tortious responsibility of Shell for having breached its duty of care. The court construed the duty of care as an overall assessment of Shell’sobligations by, among other things, international standards like the UN Guiding Principles of Human Rights Responsibilities of Businesses, the right to respect for the private and family life under Art. 8 ECHR of the residents of the Wadden region, Shell’s control over the group’s CO₂ emissions, and the state’s and society’s climate responsibility etc. This led the district court to ruling in favor of the plaintiffs and ordering Shell to reduce its greenhouse gas emissions by 45% compared to 2019.

In terms of the applicable law, the court ruled that Dutch law was applicable to the claim. The court based its choice of law analysis on Art. 7 Rome II as the relevant provision. Under Art. 7 Rome II, the plaintiff can choose to apply the law of the event giving rise to the damage rather than the law of the place where the damage occurred as per the general rule in Art. 4 (1) Rome II. The court started its analysis by stating that “climate change, whether dangerous or otherwise, due to CO₂ emissions constitutes environmental damage in the sense of Article 7 Rome II”, thus accepting without further contemplation the substantive scope of application of Art. 7 Rome II.

The court went on to find that the adoption of the business policy, as asserted by the plaintiffs, was in fact “an independent cause of the damage, which may contribute to environmental damage and imminent environmental damage with respect to Dutch residents and the inhabitants of the Wadden region”. The court thereby declined Shell’s argument that Milieudefensie’s choice pointed to the law of the place where the actual CO₂ emissions occurred, which would lead to a myriad of legal systems due to the many different locations of emitting plants operated by Shell.

2. The enigma that is “the event giving rise to the damage” to date

This line of reasoning marks a shift in the way “the event giving rise to the damage” in the sense of Art. 7 Rome II has been interpreted thus far. To date, there have been four main approaches: A broad approach, a narrower one, one that locates the event giving rise to the damage at the focal point of several places, and one that allows the plaintiff to choose between several laws of events which gave rise to the damage.

(1.) The Dutch district court’s location of the event giving rise to the damage fits into the broad approach. Under this broad approach, the place where the business decision is made to adopt a policy can qualify as a relevant event giving rise to the damage. As a result, this place will usually be that of the effective headquarters of the group. On the one hand, this may lead to a high standard of environmental protection as prescribed by recital 25 of the Rome II Regulation, as was the case before the Dutch district court, which applied the general tort clause Art. 6:162 BW. On the other hand, this may go against the practice of identifying a physical action which directly leads to the damage in question, rather than a purely internal process, such as the adoption of a business policy.

(2.) Pursuant to a narrower approach, the place where the direct cause of the violation of the legal interest was set shall be the event giving rise to the damage. In the case of CO₂ reduction claims, like Milieudefensie et al. v. Shell, that place would be located (only) at the location of the emitting plants. This approach – while dogmatically stringent – may make it harder to determine responsibility in climate actions as it cannot necessarily be determined which plant led to the environmental damage, but rather the emission as a whole results in air pollution.

(3.) Therefore, some scholars are in favor of a focal point approach, according to which the event giving rise to the damage would be located at the place which led to the damage in the most predominant way by choosing one focal point out of several events that may have given rise to the damage. This approach is in line with the prevailing opinion regarding jurisdiction in international environmental damage claims under Art. 7 Nr. 2 Brussels I-bis Regulation. In practice, however, it may sometimes prove difficult to identify one focal point out of several locations of emitting plants.

(4.) Lastly, one could permit the victim to choose between the laws of several places where the events giving rise to the damage took place. However, if the victim were given the option of choosing a law, for example, of a place that was only loosely connected to the emissions and resulting damages, Art. 7 Rome II may lead to significantly less predictability.

3. Four-step-test: A possible way forward?

Bearing in mind these legal considerations, we propose the following interpretation of the event giving rise to the damage under Art. 7 Rome II:

First, as a starting point, the laws of the emitting plants which directly lead to the damage should be considered. However, in order to adequately mirror the legal and the factual situations, the laws of the emitting plants should only be given effect insofar as they are responsible for the total damage.

If there are several emitting plants, some of which are more responsible for greenhouse gas emissions than others, these laws should only be invoked under Art. 7 Rome II for the portion of their responsibility regarding the entire claim. This leads to a mosaic approach as adopted by the CJEU in terms of jurisdiction for claims of personality rights. This would give an exact picture of contributions to the environmental damage in question and would be reflected in the applicable law.

Second, in order not to give effect to a myriad of legal systems, this mosaic approach should be slightly moderated in the sense that courts are given the opportunity to make estimations of proportions of liability in order not to impose rigid calculation methods. For example, if a company operates emitting plants all over the world, the court should be able to roughly define the proportions of each plant’s contribution, so as to prevent potentially a hundred legal systems from coming into play to account for a percentile of the total emissions.

Third, as a fall-back mechanism, should the court not be able to accurately determine each plant’s own percentage of responsibility for the total climate output, the court should identify the central place of action in terms of the company’s environmental tort responsibility. This will usually be at the location of the emitting plant which emits the most CO₂ for the longest period of time, and which has the most direct impact on the environmental damage resulting from climate change as proclaimed in the statement of claim.

Fourth, only as a last resort, should it not be possible to calculate the contributions to the pollution of each emitting plant, and to identify one central place of action out of several emitting plants, the event giving rise to the damage under Art. 7 Rome II should be located at the place where the business decisions are taken.

This proposal is discussed in further detail in the upcoming Volume 24 of the Yearbook of Private International Law.

(This post is written by Chen Zhi, a Ph.D. candidate at the University of Macau, a trainee lawyer in Mainland China)

The arbitration-favored policy has been adopted by many jurisdictions across the world in recent years, as the support of arbitration by local judiciaries has been viewed as an important standard for gauging the business environment of a jurisdiction.  While the decision of Morgan v. Sundance Inc. rendered in May 2022 by the Supreme Court of the USA illustrates that arbitration-favored policy has its boundary, this seems a trend emerging from the laws and legal trends in other jurisdictions.

Summary of the Fact

This case concerned a class action initiated by a former employee, Morgan against Sundance Incorporate (the owner of a Taco Bell franchise restaurant, hereinafter “Company”) regarding the arrear of overtime payment in the context of Federal law of the USA.

Albeit there was an arbitration agreement incorporated in the contract between Morgan and the Company, the Company failed to raise any motion about the arbitration agreement at the outset and defended as if the arbitration agreement did not exist.

Nearly 8 months after the commencement of the litigation, the company raised jurisdictional objection by invoking the omitted arbitration agreement and filed the motion to compel arbitration under the 1925 Federal Arbitration Act (hereinafter “FAA”). Morgan argued that the Company had waived the right to arbitrate. By measuring the case against the standard for the waiver as set out in the precedent of the Court of Appeal of Eighth Circuit, the court of first instance ruled in favor of Morgan and rejected to refer the case to arbitration.

Nonetheless, the Court of Appeal of the Eighth Circuit had adopted the requirement for waiver based on the “federal policy favoring arbitration”. Under the new requirement, Morgan shall furnish the proof showing prejudice incurred by the delay, and overturns the trial court’s decision thereby.[i] The case was subsequently appealed before the Supreme Court of the USA.

Supreme Court’s Decision

It is not surprising that lower courts in the USA have been consistently adopting specific rules for arbitration in the name of the arbitration-favored policy, which is contradictory to the proposition of the Supreme Court.[ii]

In the Morgan case, the Supreme Court holds that the Appeal Court of the Eighth Circuit has erred in inventing a novel rule tailored for the arbitration agreement, and reiterates that the arbitration agreement shall be placed on the same footing as other contracts. In the unanimous opinion delivered by Justice Kagan, the Supreme Court explicitly states that:

“Accordingly, a court must hold a party to its arbitration contract just as the court would to any other kind. But a court may not devise novel rules to favor arbitration over litigation.”  [iii]

In this regard, the arbitration agreement shall not be distinguished from other types of contracts in the context of Federal Law, under which the prejudice will generally not be asked about in the assessment of waiver. By Stripping off the requirement of prejudice, the Supreme Court remands the case to the Court of Eighth Circuit for reconsideration.

The Supreme Court does not delve into the jurisprudence behind arbitration-favored policy but simply states that the purpose of this policy is to make arbitration agreements as enforceable as other contracts, but not more. [iv]

The Main Concern of Morgan v. Sundance Inc.

In the context of American law, the grounds for equal treatment emerges from Section 2 of the 1925 Federal Arbitration Act, which stipulates that an arbitration agreement is valid and enforceable unless the grounds for revocation of any contract as set out in law or equity were found. Against this backdrop and in collaboration with the drafting history of the enactment of the Federal Arbitration Act, the Supreme Court has set out the basic principle that the arbitration agreement shall be placed on the same footing as other contracts, by which the arbitration-favored policy does entitle a higher protecting standard for arbitration agreement, as stated in Granite Rock Co. v. Teamsters:

“[…]the ‘policy’ is merely an acknowledgment of the FAA’s commitment to overrule the judiciary’s longstanding refusal to enforce agreements to arbitrate and to place such agreements upon the same footing as other contracts.”[v]

Through the decision in the Morgan case, the equal treatment principle is recapped and stressed, by which the arbitration-favored policy creates no new rules tailored for waiver of arbitration clauses under the legal framework of the USA.

The Complexity of Arbitration-favored Policy and the Boundary

Recent years have witnessed state courts’ preference to embrace the notion of “arbitration-favored policy” or “pro-arbitration policy”. Nonetheless, the arbitration-favored policy is a sophisticated and vague concept without an agreed definition worldwide. In principle, this policy flows from the well-recognized characteristics of international commercial arbitration such as autonomy, expediency, efficiency, and enforceability across the world. As per the analysis of Prof. Bremann, there are at least 12 criteria for gauging the arbitration-friendliness policy.[vi]

Likewise, Justice Mimmie Chan at the Court of the Instance of Hong Kong SAR fortifies 10 pro-arbitration principles employed by courts in Hong Kong towards enforcement of arbitration awards in the case of KB v S and Others, which sets up relatively high thresholds for parties to challenge arbitral awards in the enforcement stage, as the Chan J. highlights: (1) the courts’ reluctancy to looking to the merits of the case, (2) challenger’s duty to make a prompt objection against any alleged irregularities under the bona fide principle and, (3) the court’s residual discretion to enforce the award albeit the statutory grounds of rejection has been made out.[vii] Similar principles can also be extracted from decisions by courts in other jurisdictions like Singapore. [viii]

In the author’s view, these considerations for arbitration-favored policy can be distilled as the following four limbs:

(1) adherence to the parties’ autonomy to the largest extent,

(2) promoting the fairness and efficiency of commercial arbitration,

(3) minimizing the judicial interference throughout the arbitration proceedings, including the stages before and after the issuance of the arbitral award, among others, refraining from conducting the review on the merits issue of the case unless in exceptional circumstances and nullifying arbitral award based on trivial errors,

(4) providing legal assistance to arbitration proceedings for the promotion of fairness, expediency and efficiency (i.e., auxiliary proceedings for the enforcement of arbitration agreement and award, issuance, and execution of interim reliefs, taking of evidences).

As to the field of arbitral jurisdiction, the arbitration-favored policy always takes the form of the validation principle, where at least four scenarios are present in legal practice:

First, when confronted with the issue of the law governing arbitration agreement, and more than one laws are relevant, courts are required to apply laws that are in favor of the effectiveness of the arbitration agreement, either by virtue of statutory regulations[ix] or provided as one of the considerations in judicial practice.[x]

Second, courts are declined to intervene in the dispute over arbitral jurisdiction before the decision of the arbitration tribunal is rendered, as a result of the negative effect of the competence – competence principle to ensure the integrity and efficiency of arbitration proceedings.[xi]

Third, the invalidity of the matrix contract does not necessarily negate the arbitration agreement incorporated therein as per the widely-accepted separability doctrine.[xii]

Fourth, the courts will interpret in a manner that is likely to give effect to the arbitration agreement, particularly where the arbitration agreement is pathological in form or substance.[xiii]

At least one of the aforesaid scenarios emerges from legislation or judicial practices in jurisdictions featuring or advocating arbitration-favored policy, in which courts are always inclined to refer the case to arbitration. Nonetheless, the arbitration-favored policy does not mean that the court will give effect to the arbitration agreement unconditionally. The aforesaid Morgan case demonstrates that arbitration-favored policy has boundaries in the context of American law, taking the form of the equal treatment principle.

The boundary of arbitration-favored policy also emerges from laws and legal practices in other jurisdictions, as representative examples, the BNA case by the Court of Appeal of Singapore, the Kabab-Ji case by the Supreme Court of the UK, and the Uber case by the Supreme Court of Canada will be further illustrated below:

BNA Case

In this case, at issue before Singaporean courts was the law governing arbitration agreement, where the parties had designated PRC law as the governing law of the contract and expressly set out the term “arbitration in Shanghai” in the arbitration clause. The plaintiff objected to arbitral jurisdiction after the commencement of arbitration proceedings before the tribunal and subsequently resorted to courts in Singapore for recourse against the tribunal’s decision ruling that the arbitration agreement was valid under the laws of Singapore.

The plaintiff contended that the laws of China shall be applied, while the respondent argued that the arbitration clause in dispute was alleged to be invalid under PRC law, and submitted that the Singaporean court shall apply laws that are more in favor of the effectiveness of the arbitration agreement under validation principle hence the governing law shall be the laws of Singapore. The Singapore High Court applied Singaporean law and the dispute was filed before the Court of Appeal of Singapore.

The Court of Appeal opines that the validation principle can only be taken into consideration when there are other laws that can compete with PRC law to be the governing law of arbitration clause,[xiv] as all factors point to China as the proper law and Singapore was not the seat in the context of Article 10 of International Arbitration Act, this case shall be given to Chinese courts to decide.[xv] Therefore, the Appeal Court overturned the controversial decision by the Singapore High Court which determined Singapore as the seat by twisting the meaning of arbitral seat.[xvi]

Per the decision in the BNA case, the validation principle is only applicable where some prerequisites are met. While parties expressly reach an intention likely to negate the arbitration agreement without other competing factors, the court shall not rewrite the contract to nakedly validate the arbitration agreement.

Kabab-Ji Case

In this case, a Paris seated tribunal decided to extend the arbitration agreement to Kout, the parent company to the signatory which had been actively engaging in performance and re-negotiation of the contract in dispute, while not being a signatory to the contract. The tribunal’s decision was under the scrutiny of judiciaries in the UK at the enforcement stage.

Unlike the scenario in the BNA case, there were two competing factors regarding the determination of the proper law of arbitration agreement in Kabab-Ji: laws of England as the designated laws governing the main contract and the laws of France as the lex arbitri fixed in the contract. While the French laws turn out to be more in favor of the effectiveness of the arbitration clause, the Supreme Court of the UK rejected enforcing the arbitral award for lack of valid arbitration agreement via the application of English law as the proper law of arbitration clause. The court stresses in the decision that the validation principle does not apply to issues concerning the formation of a contract, and hence this principle was not relevant in deciding the issue of non-signatory.[xvii] And departing from the validation principle as set out in its precedent.

Per the decision of the Supreme Court of the UK, the extension of the arbitration agreement to non-signatory pertains to the formation of an arbitration agreement rather than the interpretation of the contract, which is contrary to the approach employed by French courts over the same case scenario. The decision in the Kabab-Ji case has given rise to controversies, as a commentator pointed out, the English court may be criticized for stepping over the line.[xviii] Nonetheless, the decision of Kabab-Ji is to some extent in line with the stringent attitude toward the non-signatory issue of arbitration agreement that judiciaries in England have consistently taken.[xix]

Uber Case

The dispute arose out of the putative employment relationship between Heller, a delivery driver, and UberEATS, a Toronto-based subsidiary of Uber. During the litigation, UberEATS filed a motion to compel arbitration by invoking the arbitration clause embedded in the boilerplate service agreement between Uber and all drivers who sign in for service of Uber.

The Supreme Court of Canada finds the arbitration clause unconscionable based on two main findings: (1) inequality of bargaining power between Heller and Uber, (2) improvidence produced by the underlying arbitration clause. The court stresses the fact that according to the arbitration clause, arbitration proceedings shall be administered under the Rules of Arbitration of the International Chamber of Commerce, which requires US$14,500 in up-front administrative fees for the commencement of the putative arbitration proceedings. Also, Amsterdam shall be the place of arbitration per the arbitration clause, hence further fees for traveling and accommodation will be incurred thereby. The court ruled that the arbitration clause was invalid and rejected to compel arbitration.[xx]

The judgment also discusses the arbitration-favored policy contention, stating that arbitration is respected based on it being a cost-effective and efficient method of resolving disputes.[xxi] By this logic, arbitration clauses creating a hurdle toward cost-effective and efficient resolution of disputes will not be safeguarded albeit the arbitration-favored policy is applicable.

The Uber case illustrates that different values may at odds with each other in the application of arbitration-favored policy, hence trade-offs will be presented before decision-makers. As discussed by Prof. Bremann, one given policy or practice may be pro-arbitration in some respects while anti-arbitration in other respects, further, the implication of arbitration-favored policy may also be detrimental to policies extrinsic to arbitration.[xxii] In the Uber case, two kinds of conflict are present simultaneously, first, upholding the effectiveness of the underlying arbitration clause may be detrimental to the policy for the protection of those who are vulnerable(trade-off between arbitration-friendly policy and extrinsic policies), second the enforcement of alleged parties’ autonomy taking the form of “arbitration administered by ICC in Netherland” is likely to be detrimental to the expediency and efficiency nature of arbitration(trade-off between arbitration-favored policy and extrinsic).

The answer to the said trade-offs remains unresolved, as there is no agreed standard by far, and courts in different jurisdictions can be divergent on this issue. As a prime example, while there is a discrepancy regarding the number of tribunal members between the rules of the arbitration institution and the arbitration clause, where the former provides a mandatory sole-arbitrator regulation for consideration of expedition and efficiency, the latter had designated a three-member-tribunal, the court of Singapore upheld the preemption of arbitration rules over the arbitration clause,[xxiii] while Chinese court once ruled in favor of the arbitration clause and rejected to enforce the award rendered by the sole arbitrator.[xxiv]

Takeaways for China

The arbitration-favored policy is a complicated notion that includes a myriad of separate and to some extent, conflicting considerations. In a general sense, courts embracing arbitration-favored policy are reluctant to negate the arbitration agreement. However, there are some exceptional instances where:

(1) the vindication of the arbitration agreement will produce prejudice to other values that are extrinsic to arbitration, such as the rule of law principle, the consistency of legal practice, policies for the protection of vulnerable parties, etc., like the situations in Morgan case and Uber case, and,

(2) the interpretation or implementation of the arbitration clause will undermine other considerations among the arbitration-favored policy, for instance, while the enforcement of the arbitration clause can be low-efficient and costly, or the validation principle may be contrary to the parties’ true intention, like the situations in BNA case and Kabab-Ji case.

Therefore, every jurisdiction shall tailor the arbitration-favored policy for its legal system and meet its own needs, instead of employing a dogmatic understanding of the policy.

Like other rising economic bodies like India,[xxv] China is also moving toward a jurisdiction that is “arbitration-favored” under the Belt and Road imitative and the blueprint for the construction of the Guangdong- Hong Kong- Macao Greater Bay Area. Against this backdrop, judiciaries are taking more liberal approaches that are tended to give effect to arbitration agreements that are likely to be considered invalid previously, particularly in disputes regarding the choice of law issue and the substance of the arbitration agreement. [xxvi]As to the formal requirement of arbitration agreement, the Supreme People’s Court also made a great leap in dispensing with the stringent approach by acknowledging the effectiveness of an arbitration clause as set out in a draft contract not being signed by neither party, based on the findings that the parties have discussed and finalized the arbitration clause in the draft of the contract during the negotiating phase.[xxvii]

Moreover, the Draft Revised Arbitration Law released in late July 2021 provides more liberal approaches for the validity of arbitration agreements, which includes:

(1) the recognition of ad hoc arbitration agreement in foreign-related disputes,

(2) the relaxing requirement for a valid arbitration agreement, where parties’ failure to designate a sole arbitration institution does not negate the arbitration agreement,

(3) the promulgation of extension of the arbitration agreement to non-signatories in some types of disputes, and

(4) the adoption of a new framework of competence-competence principle that is more in line with the international framework as set out in UNCITRAL Model Law.[xxviii]

These attempts have been heatedly debated and are by and large arbitration-favored and laudable by lifting the unreasonable hurdles for the autonomy, expediency, and efficiency of arbitration. Nonetheless, recognizing the validity of arbitration agreement is not the sole consideration, lawmakers, judiciaries, and other participants in commercial arbitration of Mainland China will confront trade-offs during the law-making and implementation of the rules under the arbitration-favored policy. As a corollary, an arbitration agreement can be safeguarded to the extent it is in line with the basic principles that are placed at a higher level.

[i] Morgan v. Sundance, Inc. 596 U. S. ____ (2022) (Supreme Court of USA, decided on 23 May 2022).

[ii] Amicus brief of Law Professors in Morgan v. Sundance, Inc., 596 U.S. ___ (2022), pp. 11- 12, available at https://www.supremecourt.gov/DocketPDF/21/21-328/207550/20220106140817376_Morgan%20amicus%20brief%20final.pdf last visited on 21 November 2022.

[iii] Morgan v. Sundance, Inc. 596 U. S. ____ (2022) (Supreme Court of USA, decided on 23 May 2022).

[iv] Ibid.

[v] Volt Information Sciences, Inc. v. Board of Trustees of Leland Stanford Junior Univ., 489 U. S. 487(1989), as quoted in Granite Rock Co. v. Teamsters, 561 U. S. 287, 302 (2010) (Supreme Court of USA, decided on 24 June 2010).

5 These considerations are: (1) to what extent does it render international arbitration economical in term of time or cost? (2) to what extent does it ensure consent to arbitrate and enhance the scope for party autonomy? (3) to what extent does it effectuate the likely intentions or expectations of the parties? (4) to what extent is it consistent with the lex arbitri or the institutional rules chosen by the parties? (5) to what extent does it, consistent with party intent, enable the tribunal to exercise sound discretion and flexibility on matters of arbitral procedure? (6) to what extent does it ensure the independence and impartiality of arbitrators? (7) to what extent does it protect a party’s right to be heard? (8) to what extent does it promote accuracy in the administration of justice? (9) to what extent does it minimize, to the fullest extent reasonably possible, the intervention of national courts in the arbitral process? (10) to what extent does it help ensure that the resulting award will be an effective one? (11) to what extent does it enable the resulting award to withstand challenges in an annulment or enforcement action? (12) to what extent does it expand the categories of legal claims treated as arbitrable? See George A. Bermann, What Does it Mean to be ‘Pro-Arbitration’?, Arbitration International, Volume 34 (2018), p. 343.

[vii] KB v S. and Others, [2015] HKCFI 1787, para.1(Hong Kong Court of First Instance, decided on 15 September 2015).

[viii] China Machine New Energy Corporation v Jaguar Energy Guatemala LLC, [2020] SGCA 12, para. 87(The threshold for the finding of breach of natural breach for the purpose of vacating arbitral award is a high one and can only be crossed in exceptional cases.) (Appeal Court of Singapore, decided on 28 February 2020).

[ix] See Article 178 (2) of Private International Law of Switzerland (“As regards its substance, the arbitration agreement shall be valid if it conforms to the law chosen by the parties, or to the law applicable to the dispute, in particular the law governing the main contract, or to Swiss law.”).

[x] Enka Insaat Ve Sanayi AS v. OOO Insurance Company Chubb, [2020] UKSC 38, para. 97 (Where the clause in question is an arbitration clause, because of its severable character its putative invalidity may support an inference that it was intended to be governed by a different law from the other provisions of the contract […]) (Supreme Court of the UK, decided on 9 October 2020). See also BCY v. BCZ, [2016] SGHC 249, para. 74 (“[…] governing law of the main contract should only be displaced if the consequences of choosing it as the governing law of the arbitration agreement would negate the arbitration agreement even though the parties have themselves evinced a clear intention to be bound to arbitrate their disputes.”) (Singapore High Court, decided on 9 November 2016).

[xi]Article 16(1) of UNCITRAL Model Law on International Commercial Arbitration(“[…] an arbitration clause which forms part of a contract shall be treated as an agreement independent of the other terms of the contract.”)

[xii] Tomolugen Holdings Ltd and Another v. Silica Investors Ltd and other appeals, [2015] SGCA, para. 60 (Singapore court should adopt a prima facie standard of review when hearing a stay application) (Court of Appeal of Singapore, decided on 26 October 2015). Fiona Trust & Holding Corp. v. Privalov [2007] UKHL 40 para. 13, (Arbitration clause shall be construed in accordance with the presumption that parties are likely to have intended any dispute arising out of the underlying contract to be decided by the same body.) (House of Lords of the UK, decided on 17 October 2007).

[xiii] “[W]here the parties have evinced a clear intention to settle any dispute by arbitration, the court should give effect to such intention, even if certain aspects of the agreement may be ambiguous, inconsistent, incomplete or lacking in certain particulars…” Insigma Technology Co Ltd v Alstom Technology Ltd [2009] 3 SLR(R) 936, para. 31, as quoted in HKL Group Co Ltd v Rizq International Holdings Pte Ltd, [2013] SGHCR 5, para. 13 (Singapore High Court, decided on 19 February 2013). See also ?????? and ??? v. Ace Lead Profits Ltd and another, [2022] HKCFI 3342? para. 53 (Arbitration clause is not nullified by the non-existence of putative arbitration institution) (Hong Kong Court of First Instance, decided on 4 November 2022).

[xiv] BNA v. BNB and another, [2019] SGCA 84, para. 95. (Court of Appeal of Singapore, decided on 27 December 2019).

[xv] Ibid., at para. 102.

[xvi] See BNA v. BNB, [2019] SGHC 142, para. 101(agreement referring to Shanghai instead of PRC is not a reference to seat) (Singapore High Court, decided on1 July 2019). Ironically, contrary to the plaintiff’s assertion and the Singapore court’s wariness, the validity arbitration clause at issue was subsequently confirmed by the Chinese court following the conclusion of judicial review proceedings before the Singapore Court of Appeal, as set out in Daesung Industrial Gases Co Ltd v. Praxair (China) Investment Co Ltd (2020) Hu 01 Min Te No.83 (Shanghai No.1 Intermediate People’s Court, decided on 29 June 2020). See also José-Antonio Maurellet, Helen Shi, et al., PRC Court Confirms Validity of “SIAC-Shanghai” Clause, available at https://dvc.hk/en/news/cases-detail/prc-court-confirms-validity-of-siac-shanghai-clause/ last visited on 21 November 2022.

[xvii] Kabab-Ji SAL v. Kout Food Group, [2021] UKSC 48, para. 51 ([Validation principle] is not a principle relating to the formation of contracts which can be invoked to create an agreement which would not otherwise exist.) (Supreme Court of UK, decided on 27 October 2021).

[xviii] Andrew Tweeddale, The Validation Principle and Arbitration Agreements: Difficult Cases Make Bad Law, The International Journal of Arbitration, Mediation and Dispute Management, Volume 88, Issue 2 (2022), p. 248.

[xix] The restrictive approach emerges from the Peterson Farms v. CM Farming Ltd., [2004] EWHC 121, as cited in Andrea Marco Steingruber, Consent in International Arbitration (UK: Oxford University Press, 2012), p. 156.

[xx] Uber Technologies Inc. v. Heller, 2020 SCC 16, paras. 93 – 94(Federal Supreme Court of Canada, decided on 26 June, 2020).

[xxi] Ibid., at para. 97.

[xxii] George A. Bermann, What Does it Mean to be ‘Pro-Arbitration’?, Arbitration International, Volume 34 (2018), pp. 343-353.

[xxiii] AQZ v. ARA, [2015] SGHC 49 (2015) (Singapore High Court, decided on 13 February 2015).

[xxiv] Noble Resources International Pte Ltd v. Good Credit International Trade Co Ltd, (2016) Hu 01 Xie Wai Ren No. 1 (Shanghai No.1 Intermediate People’s Court, decided on 11 August 2017).

[xxv] Like India, see Aditya Singh Chauhan and Aryan Yashpal, Change to Improve, Not to Unhinge—A Critique of the Indian Approach to International Arbitration, Indian Journal of Arbitration Law, Volume X, Issue 2 (2021), pp. 1-11.

[xxvi] See Helen Shi, Have Chinese Courts Adopted an Arbitration- Friendly Approach Towards International Arbitration?, in Neil Kaplan, Michael Pryles, et al. (eds), International Arbitration: When East Meets West: Liber Amicorum Michael Moser(Netherlands: Kluwer Law International, 2020), pp. 235-244.

[xxvii] Luck Treat Limited v. Zhongyuancheng Co, Ltd, 2019 Zui Gao Fa Min Te No.1(Supreme People’s Court of China, decided on 18 September 2019).

[xxviii] Terence Wong et al, China: Draft Revised Arbitration Law of PRC Published for Comments, available at https://www.mondaq.com/china/arbitration-dispute-resolution/1104356/draft-revised-arbitration-law-of-prc-published-for-comments- last visited on 4 December 2022. See also Weina Ye et al, Key Changes under Revised Draft of PRC Arbitration Law, available at  https://hsfnotes.com/arbitration/2021/08/11/key-changes-under-revised-draft-of-pcr-arbitration-law/, last visited on 4 December 2022.

Written by Eduardo Silva de Freitas (Erasmus University Rotterdam) & Xandra Kramer (Erasmus University Rotterdam/Utrecht University), members of the Vici project Affordable Access to Justice, financed by the Dutch Research Council (NWO), www.euciviljustice.eu.  

Introduction

On 9 November 2022 the District Court Amsterdam accepted international jurisdiction in an interim judgment in a collective action brought against TikTok (DC Amsterdam, 9 November 2022, ECLI:NL:RBAMS:2022:6488; in Dutch). The claim is brought by three Dutch-based representative organisations; the Foundation for Market Information Research (Stichting Onderzoek Marktinformatie, SOMI), the Foundation Take Back Your Privacy (TBYP) and the Stichting Massaschade en Consument (Foundation on Mass Damage and Consumers). It concerns a collective action brought under the Dutch collective action act (WAMCA) for the infringement of privacy rights of children (all foundations) and adults and children (Foundation on Mass Damage and Consumers). In total, seven TikTok entities are sued, located in Ireland, the United Kingdom, California, Singapore, the Cayman Islands and China. The claims are for the court to order that an effective system is implemented for age registration, parental permission and control, and measures to ensure that commercial communication can be identified and that TikTok complies with the Code of Conduct of the Dutch Media Act and the GDPR.

After an overview of the application of the WAMCA, which has been introduced in a different context on this blog earlier, we will discuss how the Court assessed the question of international jurisdiction.

The class action under the Dutch WAMCA

 Following case law of the Dutch Supreme Court in the 1980s concerning legal standing of representative organisations, the possibility to start a collective action was laid down in Article 3:305a of the Dutch Civil Code (DCC) in 1994. However, this was limited to declaratory and injunctive relief. Redress for compensation in mass damage cases was only introduced in 2005 with the enactment of the Collective Settlement of Mass Claims Act (Wet collectieve afwikkeling massaschade, WCAM). This collective settlement scheme enables parties to jointly request the Amsterdam Court of Appeal to declare a settlement agreement binding on an opt-out basis. The legislative gap remained as a collective action for compensation was not possible and such mass settlement agreement relies on the willingness of an allegedly liable party to settle.

This gap was closed when in 2019, after a lengthy legislative process, the Act on Redress of Mass Damages in a Collective Action (Wet afwikkeling massaschade in collectieve actie, WAMCA) was adopted. The WAMCA entered into force on 1 January 2020 and applies to mass events that occurred on or after 15 November 2016. The WAMCA expanded the collective action contained in Article 3:305a DCC to include actions for compensation of damage (Tillema, 2022; Tzankova and Kramer, 2021). While the WAMCA Act generally operates on an opt-out basis for beneficiaries represented by the representative organisation(s), there are exemptions, including for parties domiciled or habitually resident outside the Netherlands. In addition, the standing and admissibility requirements are relatively strict, and also include a scope rule requiring a close connection to the Netherlands. Collective actions are registered in a central register (the WAMCA register) and from the time of registration a three-months period starts to run (to be extended to maximum six months), enabling other claim organisations to bring a claim, as only one representative action can be brought for the same event(s). If no settlement is reached, an exclusive representative will be appointed by the court. Since its applicability as of 1 January 2020, 61 collective actions have been registered out of which 8 cases have been concluded to date; only a very few cases have been successful so far. These collective actions involve different cases, including consumer cases, privacy violations, environmental and human rights cases, intellectual property rights, and cases against the government. Over one-third of the cases are cross-border cases and thus raise questions of international jurisdiction and the applicable law.

As mentioned above, in the TikTok case eventually three Dutch representative foundations initiated a collective action against, in total, seven TikTok entities, including parent company Bytedance Ltd. (in the first action, the claim is only brought against the Irish entity; in the other two actions, respectively, six and seven entities are defendants). These are TikTok Technology Limited (Ireland), TikTok Information Technology Limited (UK), TikTok Inc. (California), TikTok PTE Limited (Singapore), Bytedance Ltd. (Cayman Islands), Beijng Bytedance Technology Co. Ltd. (China) and TikTok Ltd. (also Cayman Islands). The claim is, in essence, that these entities are responsible for the violation of fundamental rights of children and adults. The way in which the personal data of TikTok users is processed and shared with third parties violates the GDPR as well as the Dutch Telecommunications Act and Media Act. It is also claimed that TikTok’s terms and conditions violate the Unfair Contract Terms Directive (UCTD – 93/13/EEC) and the relevant provisions of the Dutch Civil Code.

International jurisdiction of the Amsterdam District Court

 The first stage of the proceedings, leading up to this interim judgment, deals with the international jurisdiction of the District Court of Amsterdam, as the TikTok entities challenge its international jurisdiction. TikTok requested the Court to refer preliminary questions to the CJEU but the Court refused this request, stating that the questions on (a) how the GDPR and Brussels I-bis Regulation regimes interact and (b) the applicability of Article 79(2) GDPR were deemed resolved.

Relevant jurisdiction rules

Considering the domicile of the defendant(s) and the alleged violation of the GDPR, both EU and Dutch domestic jurisdiction rules come into the picture. TikTok alleges that the Dutch courts do not have jurisdiction over this case under Article 79(2) GDPR. Moreover, TikTok alleges that, since Article 79(2) GDPR is a lex specialis in relation to the Brussels I-bis Regulation, the latter cannot be applied to override the jurisdictional rules set out in the GDPR. The three representative organisations argue that the Dutch courts have jurisdiction under both EU private international law rules and the Dutch Code of Civil Procedure (DCCP). Before delving into how the District Court of Amsterdam construed the interaction between the legislations concerned, we will describe the applicable rules on international jurisdiction for privacy violations. The alleged violations occurred, or the claims relate to violations occurring, after 25 May 2018, that is, after the entry into force of the GDPR. TikTok Ireland is a data controller subject to the GDPR. Under Article 79(2) GDPR the “data subjects” (those whose rights are protected by the GDPR) shall bring an action for the violation of their rights in either the courts of the Member State in which the data controller or processor is established or of the Member State in which the data subject has its habitual residence. Furthermore, Article 80(1) GDPR provides for the possibility of data subjects to mandate a representative body which has been properly constituted under the law of that Member State, has statutory objectives which are in the public interest, and is active in the field of the protection of data subjects’ rights and freedoms to file actions on their behalf under Article 79 GDPR.

The case also deals with non-GDPR-related claims, which triggers the application of the Brussels I-bis Regulation, at least as far as the entities domiciled in the EU are concerned. Article 7(1)(a) Brussels I-bis states that, for contractual matters, jurisdiction is vested in the Member State in which the contract is to be performed. More importantly for this case, with regards to torts, Article 7(2) provides jurisdiction for the courts of the place where the harmful event occurred or may occur. Finally, in relation to the TikTok entities that are not domiciled in the EU, the international jurisdiction rules of the Dutch Code of Civil Procedure (Articles 1-14 DCCP) apply. This is the case regarding both GDPR and non-GDPR-related claims. These Dutch rules are largely based on those of the Brussels I-bis Regulation and also include a rule on multiple defendants in Article 7 DCCP.

The claims against TikTok Ireland

The Amsterdam District Court starts its reasoning by addressing whether it has jurisdiction over TikTok Technology Limited, domiciled in Ireland, the entity that is sued by all three representative organisations. The Court states that Article 80(1) GDPR does not distinguish between substantive and procedural rights in granting the possibility for data subjects to mandate a representative body to file actions on their behalf under Article 79 GDPR. Therefore, actions brought under Article 80(1) GDPR can rely on the jurisdictional rule set out in Article 79(2) GDPR which allows for the bringing of actions before the courts of the Member State in which the data subject has its habitual residence. The Court further reasons that the word ‘choice’ enshrined in Recital 145 GDPR, when mentioning actions for redress, allows for the interpretation that it is up to the data subject to decide where she prefers to file her claim.  In the case at hand, since the data subjects concerned reside in the Netherlands, they can mandate a representative body to file claims before the Dutch courts.

As to the non-GDPR-related claims and GDPR violations that also qualify as tortious conduct, the District Court considered first whether the case concerned contractual matters, to decide whether Article 7(1) or Article 7(2) Brussels I-bis Regulation applies. For this purpose, the District Court relied on the rule established by the CJEU in Wikingerhof v. Booking.com (Case C-59/19, ECLI:EU:C:2020:95), according to which a claim comes under Article 7(2) when contractual terms as such and their interpretation are not at stake, but rather the application of legal rules triggered by the commercial practices concerned – or, in other words, contractual “interpretation being necessary, at most, in order to establish that those practices actually occur”. Given that, in this case, the question is whether TikTok’s terms and conditions are abusive under both the UCTD and the DCC, the claim was deemed to fall under Article 7(2) Brussels I-bis Regulation.

Next, the District Court assesses whether the criteria for establishing jurisdiction under Article 7(2) are met. For this purpose it refers to the CJEU ruling in eDate Advertising and Others (Case C-509/09, ECLI:EU:C:2011:685). In this case the CJEU ruled that, when it comes to “publication of information on the internet” that triggers an “adverse effect on personality rights”, the habitual residence of the victim being his centre of interests can be regarded as the place in which the damage occurred. The District Court rightfully ruled that since the rights of TikTok users that have their habitual residence in the Netherlands had been violated through online means, the Netherlands can be regarded as the place in which the damage occurred.

The Court confronts TikTok’s argument that, since Article 79(2) GDPR is a lex specialis in relation to the Brussels I-bis Regulation, the latter cannot be applied to override the jurisdictional rules set out in the GDPR. As per the Court, the rules on conflict of jurisdiction established by the Brussels I-bis Regulation are general in nature and, as such, cannot be derogated from other than by explicit rules. Hence, the Court interprets Recital 147 GDPR – which states that the application of the Brussels I-bis Regulation should be without prejudice to the application of the GDPR – as being unable to strip away the applicability of the Brussels I-bis Regulation. In the Court’s understanding, Recital 147 GDPR points to the complementarity of the GDPR in relation to the Brussels I-bis Regulation, and both regimes coexist without hierarchy. Therefore, according to the Court, the GDPR is not a lex specialis in relation to the Brussels I-bis Regulation. Furthermore, the Court notes that, under Article 67 Brussels I-bis Regulation, its regime is without prejudice to specific jurisdictional rules contained in EU legislation on specific matters. While the relationship between the jurisdiction rules of the GDPR and the Brussels I-bis Regulation is not wholly undisputed, in the present case the provisions do not contradict each other, while at the same time in this case also non-GDPR issues are at stake.

The claims against non-EU based TikTok entities

Having established international jurisdiction in the case against TikTok Ireland, the Amsterdam District Court rules on its international jurisdiction in relation to the other TikTok entities sued by two of the foundations. As no EU rules or international convention applies, the Dutch jurisdiction rules laid down in Articles 1-14 DCCP apply. Article 7(1) DCCP contains a rule for multiple defendants and connected claims similar to that in Article 8(1) Brussels I-bis. The Court considers that both legal and factual aspects are closely intertwined in this case. The claims concern several different services, not only the processing of data, and all defendants are involved in the provision of these services. The claims are therefore so closely connected that it is expedient that they are dealt with in the same proceedings.

Outlook

TikTok attempted to appeal this interim judgment on international jurisdiction. Under Article 337(2) DCCP, it is at the court’s discretion to grant leave to appeal interim decisions when the appeal is not filed against the final judgment at the same time. In this case, the Court did not find sufficient reasons to allow for such appeal. The case will now proceed on other preliminary matters, including the admissibility of the claim under the WAMCA, and (if admissible) the appointment of the exclusive representative. For this purpose, at the end of its judgment the Court orders parties to provide security as to the financing of the case, which requires submitting to the Court a finance agreement with the third-party financer. After that, assuming that no settlement will be reached, the case will proceed on the merits. It may well be that either of the parties will appeal the final judgment, and that on that occasion TikTok will raise the jurisdictional question again.

To be continued.

Dr Benjamin Hayward*

The United Nations Convention on Contracts for the International Sale of Goods (‘CISG’), currently adopted by 95 States, is a treaty intended to harmonise the laws governing cross-border goods trade: and thereby promote trade itself.  So much is made clear in its Preamble:

The States Parties to this Convention, …

Being of the opinion that the adoption of uniform rules which govern contracts for the international sale of goods and take into account the different social, economic and legal systems would contribute to the removal of legal barriers in international trade and promote the development of international trade,

Have agreed as follows: …

Art. 7(1) CISG’s instruction for interpreters to have regard ‘to its international character and to the need to promote uniformity in its application and the observance of good faith in international trade’ establishes a requirement of autonomous interpretation.  This, in turn, facilitates the CISG’s global jurisconsultorium: whereby courts, arbitrators, lawyers, academics, and other interested stakeholders can influence and receive influence in relation to the CISG’s uniform interpretation.  A recent publication edited by Peng Guo, Haicong Zuo and Shu Zhang, titled Selected Chinese Cases on the UN Sales Convention (CISG) Vol 1, makes an important contribution to this interpretative framework: presenting abstracts and commentaries addressing 48 Chinese CISG cases spanning 1993 to 2005, that may previously have been less accessible to wider international audiences.

A review of this case law collection discloses an interesting phenomenon affecting the CISG’s Chinese application: at least, until very recently.  Pursuant to Art. 142(2) General Principles of the Civil Law (which was effective in the People’s Republic of China until repealed as of 1 January 2021):

[I]f any international treaty concluded or acceded to by the People’s Republic of China contains provisions differing from those in the civil laws of the People’s Republic of China, the provisions of the international treaty shall apply, unless the provisions are ones on which the People’s Republic of China has announced reservations.

(Translation via Jie Luo.)

Numerous contributions to Guo, Zuo and Zhang’s volume – including by Wang, Guo and Zhang; Luo; Luo again; Wang; and Xu and Li – observe that some Chinese courts have interpreted this provision to require the CISG’s application only where it is inconsistent with non-harmonised Chinese law.  Whilst this approach to the CISG’s application is noteworthy for its inconsistency with international understandings of the treaty, it is arguably more noteworthy for highlighting that national law itself is often ‘where the relationship between the convention and national law is regulated’.[1]  Scholarship has given much attention to the success (or otherwise) of Art. 7(1) CISG in securing the treaty’s autonomous interpretation.  However, machinery provisions giving the CISG local effect in any given legal system (themselves being matters of ‘local legislative judgment’) have an apparently-underappreciated role to play, too.

Wang’s contribution quotes Han as writing that the Chinese inconsistency concept’s effective implementation of a reverse burden of proof in establishing the CISG’s application is a situation that ‘I am afraid … is unique in the world’.  On the contrary, and not unlike China’s former Art. 142(2) General Principles of the Civil Law, Australia’s CISG implementing Acts still ostensibly frame the treaty’s local application in terms of inconsistency.  The Sale of Goods (Vienna Convention) Act 1986 (NSW) s 6 is representative of provisions found across the Australian state and territory jurisdictions: ‘[t]he provisions of the Convention prevail over any other law in force in New South Wales to the extent of any inconsistency’.  Case law from Victoria and from Western Australia has read those jurisdictions’ equivalent inconsistency provisions as implying the CISG’s piecemeal application, only where particular provisions are inconsistent with local law.  Looking even further afield, Australia’s own use of the inconsistency device is far from unique.  Singaporean and Canadian legislation make use of the inconsistency concept, as does Hong Kong’s recently-promulgated CISG Ordinance.  In the latter case, the statutory interpretation risks associated with the adoption of an inconsistency provision were drawn to the Hong Kong Department of Justice’s attention.  However, Australia’s statutory model prevailed, perhaps in part because it has previously been put forward as a model for Commonwealth jurisdictions looking to implement the CISG.

At the risk of being slightly controversial, at least some scholarship addressing the failings of national CISG interpretations may have been asking the wrong question: or at least, missing an important additional question.  Instead of asking why any given court has failed to apply and respect Art. 7(1) CISG’s interpretative directive, we might instead (or also) usefully ask whether that given State’s CISG implementation legislation has been drafted so as to invite the local law comparisons that have then been made.  Some responsibility for problematic CISG interpretations might lie with the legislature, in addition to the judiciary.

In Australia, the Playcorp decision – Victoria’s inconsistency case referred to above – has been taken by subsequent cases in both the Federal Court and in the Full Federal Court of Australia as authority for the proposition that Art. 35 CISG’s conformity requirements equate to the implied terms contained in the non-harmonised Goods Act 1958 (Vic) s 19.  The Federal Court’s first-instance decision was itself then cited in New South Wales for that same proposition: leading to a problematic CISG interpretation that is now entrenched under multiple layers of precedent.  Whilst the equation being made here is rightly criticised in itself, it has Australia’s inconsistency provisions – in addition to our courts’ failures to apply Art. 7(1) CISG – resting at its core.

Guo, Zuo and Zhang’s Selected Chinese Cases on the UN Sales Convention (CISG) Vol 1 thereby makes a valuable contribution to the Convention’s jurisconsultorium: first, by virtue of its very existence, but secondly, by its additional disclosure of China’s former inconsistency struggles to the wider scholarly community.

[1] Bruno Zeller, ‘The CISG in Australasia: An Overview’ in Franco Ferrari (ed), Quo Vadis CISG?  Celebrating the 25^th Anniversary of the United Nations Convention on Contracts for the International Sale of Goods (Bruylant, 2005) 293, 299.

 

* Senior Lecturer, Department of Business Law and Taxation, Monash Business School

Twitter: @LawGuyPI

International Trade and International Commercial Law research group: @MonashITICL

Disclosure: The author is a confirmed contributor to the forthcoming Selected Chinese Cases on the UN Sales Convention (CISG) Vol 3.