Smart consumer goods and digital content: international jurisdiction and law applicable to the claims of consumers

1. Introduction

2. International jurisdiction in consumer contracts over smart consumer goods and digital content or digital services

3. Law applicable to consumer contracts for smart goods and digital content or digital services 3.1. General principles

3.2. Directive-based domestic contract law norms as overriding mandatory norms under Rome I?

3.3. Incorrect transposition of directive-based conflict of law rules: the example of Estonia

4. International jurisdiction for product liability claims due to defective consumer goods and digital content

5. Law applicable to product liability claims due to defective consumer goods and digital content 5.1. Applicability of Article 5 of Rome II Regulation

5.2. The “marketed” and “foreseeability” criteria in Article 5 of Rome II Regulation in case of smart consumer goods

6. Conclusions

* * *

The article concentrates on international jurisdiction and applicable law in cases where consumers wish to assert their claims due to defective smart goods and digital content. It shows that the new digital content and consumer sales directives do not create many new problems concerning international consumer jurisdiction or applicable law rules. However, in cases where a sales contract of a smart good is interconnected with a digital services contract, the dichotomy of contracts is reflected in the jurisdiction and applicable law: it may happen that the jurisdiction and applicable law of a sales contract of a smart good deviates from the jurisdiction and applicable law of the interconnected contract for digital services. It is further argued that domestic contract law provisions goldplating the rules of digital content and consumer sales directives should, as a rule, not be considered overriding mandatory norms within the meaning of Article 9 of Rome I Regulation. The article also analyses the European jurisdiction and conflict of law rules on product liability and concludes that in most cases the jurisprudence of the Court of Justice affords consumers who have suffered health or property damage due to a smart good a de facto plaintiff jurisdiction. For privacy infringements - also an important risk in case of smart goods and digital content - the plaintiff jurisdiction also derives from Article 79, 2. of GDPR.

1.

Introduction

1.The adoption of the new digital content directive [2] (DCD) and consumer sales directive [3] (CSD) in late spring 2019 have brought about intense academic discussion on the new rules on “digital contracts” and consumer protection. As contracts concerning smart consumer goods and digital content often have cross-border nature, it is logical to ask how private international law deals with such contracts. For instance, do the rules on “goods with digital elements” in the new directives raise new legal questions concerning jurisdiction and applicable law, and if yes, can they be reasonably solved under the existing regimes of the Brussels Ibis Regulation [4] and the Rome I Regulation? [5] Are the provisions of the new directives to be considered mandatory norms within the meaning of Article 9 of Rome I and would the status of mandatory norms even be necessary for protecting European consumers?

2.The effective protection of consumers who have suffered bodily harm or property damage due to defective smart goods or digital content depends not only on their contractual claims but also upon whether, where and how they can assert claims against producers under product liability rules. Therefore, the second part of the article looks at the jurisdiction and applicable law rules in the area of product liability under the existing EU law and the jurisprudence of the Court of Justice of the European Union (CJEU), mapping the possibilities of consumers to claim damages at the courts of their habitual residence as well as to rely upon their “home law” while doing so.

2.

International jurisdiction in consumer contracts over smart consumer goods and digital content or digital services

3.As is well known, Brussels Ibis Regulation provides consumers with a privileged jurisdiction regime: in cases defined in Article 17, 1. of Brussels Ibis, a consumer may be sued only in the courts of the Member State of his domicile and he is also entitled to file a claim against the trader there. [6] Pursuant to Article 19 of Brussels Ibis, jurisdiction agreements to the detriment of consumers are invalid, as a rule. Therefore, in these cases the trader bears the risk of ending up in litigation abroad. In addition to credit-financed consumer sales contracts, consumer contract jurisdiction rules are also applicable if “the contract has been concluded with a person who pursues commercial or professional activities in the Member State of the consumer's domicile or, by any means, directs such activities to that Member State or to several States including that Member State, and the contract falls within the scope of such activities”. [7] The CJEU has given guidance on the interpretation of what constitutes “direction of trader's activities to consumer's Member State” in several judgments, starting with the famous Alpenhof / Pammer decision [8] followed by Mühlleitner [9] and Emrek [10] judgments and has interpreted the concept of “direction of activities” remarkably widely. [11] The new digital content and consumer sales directives do not create many new problems concerning these international consumer jurisdiction rules of Brussels Ibis Regulation. Yet, one specific issue does arise, and this concerns buying goods with interconnected digital services, eg a smart car with a digital navigation system or a smart home system which consists of both hard- and software.

4.Namely, as the starting point, the directives are based on the assumption that the sales contract of such a smart good also covers the ancillary/interconnected digital services. This is so if two cumulative conditions are met: a) the digital service is inter-connected with the good in such a way that the absence of that digital service would prevent the good from performing its functions [12]; and b) the digital service is provided with the good under the sales contract. [13] However, the recitals of the directives reveal that there is a great deal of room left for party autonomy with the consequence that in many cases the consumer will have one contract with the hardware seller and the other with the software/digital services producer. [14]

5.To start with, recital (15) CSD stresses that whether or not inter-connected digital service forms part of the sales contract depends upon the sales contract. And if the consumer concludes a contract for the supply of digital content or a digital service which does not form part of the sales contract, that contract should be considered to be separate from the sales contract, even if the seller acts as an intermediary of the digital services contract with the third-party supplier. [15] Recital (16) CSD expressly allows for a situation where the consumer buys a smart phone without a specific operating system and the consumer subsequently concludes a contract for the supply of an operating system from a third party. We can also use the example brought by Carvalho: if the consumer buys a car with no GPS application installed and then “buys” the latest version of a GPS application over the Internet there are two contracts, one of which is a sales contract of the car and the other a digital services contract for the GPS application. [16]

6.This dichotomy of contracts is reflected also in the jurisdiction and applicable law: in specific cases it may happen that the sales contract of a smart good (eg a connected car) underlies jurisdiction and applicable law other than the inter-connected contract for digital services (eg a navigation and entertainment system of that car). For example, an Estonian consumer goes to Poland and decides to buy a new Mercedes from the local trader: here the consumer cannot rely upon his home jurisdiction as the contract does not fall under the scope of Article 17 of Brussels Ibis. [17] Then he goes back to Estonia and concludes a contract for Mercedes MeConnect digital services package: as the trader (Mercedes-Benz) clearly directs its digital services to Estonia, the specific consumer contract jurisdiction (consumer's habitual residence) under Articles 17, 18 of Brussels Ibis is available to the consumer. [18] The same is true if a consumer buys a smart home system or a smart watch from a Latvian department store but then installs it at home in Estonia and successfully concludes a “license agreement” there.

7.It can therefore happen that the jurisdiction for asserting claims over the defects of the good (the car or the smart home system) and the jurisdiction for defects in digital services deviate from each other. In many cases, however, it is hard to determine whether a particular problem is due to the defects of the car (or the watch) or that of the inter-connected digital service. The possibility of Article 8, 1. of Brussels Ibis to pursue connected claims against multiple co-defendants before the court of the state where any one of the co-defendants is domiciled does not offer real protection as it operates only upon general jurisdiction and not special jurisdiction such as Article 18, 1. of Brussels Ibis. [19] Consequently the consumer faces an additional burden of filing two separate claims in two separate jurisdictions even though the claims are very much connected to each other. The risk of contradicting judgments is considerable as well.

8.In order to reduce these negative consequences, one could argue that the interpretation developed by the Court of Justice in the Maletic [20] case should be applied to such inter-connected contracts of consumer sales and digital services. In Maletic - still decided under the “old” Brussels I Regulation - the Court of Justice allowed consumers who had booked a package holiday via a website intermediary to sue both the intermediary as well as the tour operator before the court of their domicile. The Court considered these two contracts to be “inseparably linked” [21] - an argument that could be valid also for sales contracts with inter-connected digital services contracts. The CJEU also took notice of the objectives of protecting the consumer as the weaker party and of minimising the possibility of concurrent proceedings and therefore irreconcilable judgments and found that these objectives preclude a solution which would force the consumer to pursue parallel proceedings before different courts. [22] However, it would be too far-reaching to apply it also to the sales of smart goods with digital content/services and to allow the consumer to sue both contract parties at his domicile. This would put the seller into a considerably worse jurisdictional position than he otherwise would have been. Moreover, the consumer's home jurisdiction would be totally unforeseeable to the seller. In order to avoid contradicting judgments, I would therefore argue that these contracts are inseparably linked but I would allow the consumer to sue both parties - the seller as well as the digital service provider - only in the courts of the country where he had bought the smart item and not in the country of his domicile where he had afterwards concluded the license agreement. True, this result cannot be based upon Articles 17, 18 of Brussels I Regulation as these provisions only regulate the jurisdiction of consumer's domicile. Therefore I would suggest to apply the concept of “inseparably linked contracts” developed in the Maletic case also in the context of Article 7, 1., (b) of Brussels I Regulation and allow both disputes to be settled at the place of performance of the “main” contract, ie the sales contract of a smart good. This solution would give the consumer a choice whether to go a) against different contracting parties in different countries (in Poland against the seller and in Estonia against the digital service provider) or b) to sue both parties in Poland.

3.

Law applicable to consumer contracts for smart goods and digital content or digital services

3.1.

General principles

9.The most important instrument determining the law applicable to international contracts in the European Union is the Rome I Regulation. Under Rome I Regulation, the applicable law is first and foremost determined through the principle of party autonomy. This principle also applies to consumer contracts under Article 6, 1. of Rome I. However, Rome I expressly states that the presumably weaker parties should be protected by conflict-of-law rules that are more favourable to their interests than the general rules. [23] With regard to consumers, Article 6, 2. of Rome I stipulates that even if the parties have agreed that a particular system of law is to be applied to the contract, such choice may not deprive consumers of the protection afforded to them by the mandatory provisions of their state of habitual residence. Similarly to Article 17, 1. of the Brussels Ibis Regulation, this rule comes into play if the trader in a way or another directs his commercial activities to the Member State where the consumer has his habitual residence. [24]

3.2.

Directive-based domestic contract law norms as overriding mandatory norms under Rome I?

10.Previously, various consumer law directives obliged Member States to ensure that the consumer does not lose the protection granted by that consumer law directive by choosing the law of a non-EU-member country as applicable to the contract, if the contract has a close connection with the territory of the Member States. [25] By contrast, newer consumer contract law directives such as the new Digital Content Directive, Consumer Rights Directive [26] and the new Consumer Goods Directive no longer contain such provisions and refer all questions of determining whether the consumer retains the protection granted by the directive to Rome I Regulation. [27] Under Article 3, 4. of Rome I choice of the law of a third country cannot not prejudice the application of mandatory provisions of Community law, if all elements relevant to the situation at the time of the choice are located in one or more Member States. The standard of Article 3, 4. of Rome I is somewhat higher than the one in the private international law provisions of the older consumer law directives as the directives only required a “close link with the territory of one or more Member States”.

11.As indicated above, Rome I protects consumers as weaker contractual parties by virtue of Article 6. In addition to Article 6, Rome I Regulation contains another correction mechanism: the doctrine of overriding mandatory rules in Article 9. Article 9, 1. of Rome I defines overriding mandatory provisions as provisions the respect for which is regarded as crucial by a country for safeguarding its public interests, such as its political, social or economic organisation, to such an extent that they are applicable to any situation falling within their scope, irrespective of the law otherwise applicable to the contract under the Regulation. One can therefore wonder whether national implementation norms of the new consumer contract law directives not only constitute “provisions that cannot be derogated from by agreement” within the meaning of Article 6, 2. of Rome I but are also to be seen as rules crucial for safeguarding the forum's public interests within the meaning of Article 9 of Rome I. Could, for instance, the rules on trader's or seller's liability of the new consumer contract law directives be considered not only mandatory but also as crucial for a Member State for safeguarding its public interests?

12.One must note, of course, that the question whether the norms of the new consumer contract law package should be considered overriding mandatory norms is a more general one: can directive-based consumer contract law provisions generally be considered overriding mandatory norms? It obviously needs very good reasons to consider some consumer contract law directives loi de police while rejecting such nature for others.

13.The question whether Article 6 should be considered lex specialis over Article 9, 2. of Rome I Regulation is controversial and not yet decided by the Court of Justice. Several prominent authors in the German legal literature see Article 6 as lex specialis over Article 9 for contracts which fall within the scope of Article 6 of Rome I, as in these cases there is usually no need for additional consumer protection under Article 9. [28] Before the adoption of Rome I Regulation the German High Federal Court took the position that German contractual consumer protection provisions can, in principle, be applied via the provision on internationally mandatory rules but only in cases where the contract has a close connection with Germany. [29] In a later judgment - still decided under the national conflict of law provisions and not under Rome Convention or Rome I Regulation - the High Federal Court expressly stated that the conflict of laws provision on consumer contracts has no lex specialis character at least in cases where the contract does not fall within its scope. [30]

14.In practice, the question does not become relevant too often as usually the forum of the dispute coincides with the habitual residence of the consumer under Article 18, 1. of the Brussels I Regulation and also the benchmark of Article 6, 2. of Rome I against which the legality of choice of law is tested. The problem could arise when we are outside the scope of Article 6 of Rome I, eg when the trader has not targeted its activities to the consumer's country of residence. [31] For example, an Estonian consumer visits Poland and decides to buy a car from the local trader: if there is no choice of law then Polish law, including its unfair terms rules will be applicable under Article 4, 1., (a) of Rome I. [32] If a defect shows up later, then Estonian consumer protection norms will not come into play as the trader had not targeted its sales activities to Estonia. It is also not likely that Estonian consumer protection norms could be applied as overriding mandatory norms as the forum of the dispute will most likely be Poland as well. [33] It could be different, however, if the trader had undertaken an obligation to deliver the car to Estonia which would open the jurisdiction of Article 7, (1), (b) of the Brussels I Regulation. In this case the Estonian consumer could sue the Polish trader in Estonia and the question of the applicability of Estonian implementation norms as overriding mandatory rules under Article 9, 2. of Rome I could become important.

15.A similar situation arises if an EU consumer buys a good in a third country, for instance when a Lithuanian consumer buys a consumer good on an excursion to neighbouring Belarus - provided that the Lithuanian domestic jurisdiction rules [34] allow him to file a claim in Lithuania. Or, if a Lithuanian consumer makes the same transaction in Poland but for whatever reason Belarusian or Californian law is chosen. In the latter case the issue is solved by Article 3, 4. of Rome I [35] that would enable the application of Lithuanian consumer sales, digital content as well as unfair terms rules. Finally, the issue becomes relevant also in cases where the jurisdiction and applicable law do not coincide due to the exceptions in Article 6, 4. of Rome I - which would rarely be the case for consumer goods and digital services. Consequently, one may agree with Berndt von Hoffmann that the question of overriding mandatory norms in contract law is of great interest to scholars but of little practical importance. [36]

16.While it is clear that the overriding mandatory norms have to be applied despite any choice of law clause in the contract and supersede the law applicable under the statutory conflict-of-law provisions, the relationship of consumer protection rules in Article 6 of Rome I to overriding mandatory provisions in its Article 9 is not uniformly solved neither in legal literature nor in the judicial practice of the Member States. The underlying question is whether the doctrine of overriding mandatory norms in Article 9 is meant to protect also weaker parties of the contract or whether its main function is to protect the more general public interests and values. And, of course, whether protection of weaker contractual partners could be seen as such general public interest. In short: while “non-mobile” consumers are protected by virtue of Article 6, 2. of Rome I Regulation, “mobile” consumers could be protected by its Article 9 and thereby reach the same level of protection, depending on the view of the national judge.

17.The Court of Justice has this far not given clear instructions on the issue. True, the Court has stressed that the notion of overriding mandatory norms must be interpreted strictly [37] and when determining them the judge has to take account not only of the exact terms of the law, but also of its general structure and of all the circumstances in which that law was adopted. [38] The Court has also ruled when interpreting the overriding mandatory norms provision of the Rome II Regulation [39] that a national rule which provides a three-year limitation period for actions seeking compensation for damage resulting from a car accident cannot be considered an overriding mandatory provision unless the national court finds a special justification for it. [40] Whether one can generally consider consumer contract law rules as overriding mandatory norms within the meaning of Article 9 of Rome I has, however, not yet been answered by the Court of Justice as the agency law cases Ingmar [41] and Unamar are not about consumer contract law.

18.The views of national courts and doctrines differ on the issue. For instance, German doctrine [42] and the High Federal Court [43] do not consider consumer protection rules overriding mandatory provisions if the protection of public interest is only reflected by the primary purpose of protecting private interests of the contract parties. By contrast, French [44] and Dutch [45] doctrine have taken a wider approach and consider also rules meant to protect the weaker party as overriding mandatory provisions, because avoiding the abuse of the weaker party as such is also of general public interest.

19.When asking whether the consumer protection norms of the new directives constitute overriding mandatory norms, one must first note that at least in principle, the new directives are maximum harmonizing [46] which clearly reduces the practical relevance of the question. In the Digital Content Directive only the question of time limits of the trader's liability remains subject to minimum harmonization. [47] The new Consumer Sales Directive, by contrast, gives more discretion to Member States, allowing higher consumer protection standards not only concerning time limits of the seller's liability [48] but also concerning longer reversal of the burden of proof. [49] In certain questions it also allows Member States to reduce consumer protection standards, eg to opt for notification obligation of the consumer. [50] Finally, both directives also entitle Member States to extend the directives' regime to contracts [51] and issues [52] outside of their scope.

20.Therefore, as the legal regime concerning consumer contracts on digital content and consumer goods is still not totally harmonized across the EU, the question of so-called “goldplating” [53] arises. For example, could a French court dealing with a consumer sales contract between a mobile French consumer and Estonian trader apply French law (which does not foresee a notification obligation) instead of the otherwise applicable Estonian law containing a notification obligation? Or, can the French court apply the French rules on vice caché which are unknown to the Estonian law? The rulings of the Court of Justice do not give a direct answer as the Unamar case concerned agency directive which is outside of the scope of Article 6 of Rome I [54] and Silva Martins case concerned domestic rules on time limits of non-contractual claims that are not harmonized at all. The issue is also controversial in legal literature as eg van Bochove allows a national court to qualify a domestic rule based on a minimum harmonisation Directive but transgressing the level of protection as an overriding mandatory rule [55] whereas Lüttringhaus and van Calster reject the idea. [56]

21.Yet, Silva Martins case may give a hint also concerning the limitation periods that are minimum harmonised under DCD and CSD: I would argue that national rules providing for longer limitation periods in consumer sales are not to be considered overriding mandatory norms under Rome I Regulation. Otherwise a state could impose their own law to traders who are not reckoning with it as they trust being outside the scope of Article 6, 2. of Rome I. Member States could also do so without the hard work of convincing other Member States during the EU legislative process. Therefore, in my view a domestic contract law provision goldplating the rules of digital content and consumer sales directives could be considered an overriding mandatory norm only in exceptional cases with compelling reasons - which, frankly, is very hard to imagine. The issues regulated by the Digital Content Directive and Consumer Sales Directive - consumer's remedies for lack of conformity, guarantees and regress claim of the trader - are not likely to protect the public interest of the forum within the meaning of Article 9, 1. of Rome I. This position would be also in line with Recital (37) of Rome I which underlines that the concept of “overriding mandatory provisions” should be distinguished from the expression “provisions which cannot be derogated from by agreement” and be construed more restrictively.

22.The same applies if the consumer sales contract is subject to the law of a third country: either the consumer is protected via Article 6 of Rome I or via Article 3, 4. of Rome I Regulation. In the majority of cases additional protection under Article 9 of Rome I is not necessary - at least if one takes the position that the third-country producer of a consumer good does not constitute an element connecting it to the third country and thereby excluding the protection of Article 3, 4. of Rome I. [57] The situation is, however, somewhat more complicated in case of digital content or digital services: in many cases the contract party of an EU consumer is a trader outside the EU, often situated eg in California or some other US state. True, in the majority of cases the consumer is protected by Article 6, 2. of Rome I as he concludes the contract online and the US trader directs its activity to the state of consumer's residence within the meaning of Article 6, 1., (b) of Rome I - leading to the applicability of the mandatory norms of the habitual residence of the consumer. Yet, there can be cases where eg an Estonian consumer buys a videogame in Finland sold by a US trader under US law and the game is not available in Estonia: here the consumer is not protected by Article 6, 2. nor by Article 3, 4. of Rome I. In these exceptional cases the applicability of at least some of the provisions of the Digital Content Directive via Article 9, 2. of Rome I cannot be ruled out.

3.3.

Incorrect transposition of directive-based conflict of law rules: the example of Estonia

23.In Estonia an additional issue arises from the fact that the directive-based conflict of law rules have been incorrectly transposed in the Estonian Law of Obligations Act (LOA). [58] Namely, Sections 36(2), 53(1), 237(2) and 403(6) of the LOA stipulate that the provisions determining the rights and obligations of the consumer and of the trader apply to contracts with consumers residing in Estonia or the EU, if the contract is entered into as a result of a public tender, advertising or other similar economic activities taking place in Estonia or if the contract is essentially linked to the territory of Estonia for any other reason, regardless of which state's law applies to the contract. This raises a specific question of whether these norms are of an overriding mandatory nature, which their original language would suggest.

24.The incorrect transposition is evident from the fact that the scope of application of the Estonian consumer contract law provisions under the above-mentioned provisions is wider than foreseen by the consumer directives, since the directives only require them to apply if the law of a third country - and not of a Member State - has been chosen. The Estonian provisions as they stand now would also require precedence over a choice of law of another Member State, eg Belgian law. This would mean that consumers would be left without protection under Belgian law, if Belgian law offers a higher protection standard in a particular issue than the Estonian law does. [59] This would clearly contradict Article 6, 2. of Rome I Regulation as it allows the consumer to rely upon the provisions that are most favourable to him. Therefore, it is convincingly argued by Piir that Estonian national implementation provisions of the consumer-related directives should be considered subordinate to Article 6 of Rome I and should apply only in cases where the prerequisites for applying Article 6 of Rome I are not met. [60] Moreover, the aforementioned conflict rules of the Estonian Law of Obligations Act are not to be considered overriding mandatory rules within the meaning of Article 9 of Rome I but only domestically mandatory. [61] In the end, the Estonian legislator would in my view be well advised to abolish § 237(2) LOA when transposing the new consumer sales directive [62] which does not foresee a special conflict of law norm as did its predecessor. [63]

4.

International jurisdiction for product liability claims due to defective consumer goods and digital content

25.After long reflexions over jurisdiction and applicable law for contractual claims let us now briefly touch upon international jurisdiction and applicable law concerning non-contractual claims of consumers who have suffered health or property damages due to defective smart goods or digital content. Whereas Rome II Regulation contains a specific conflict of law provision for product liability, jurisdiction for these claims is determined under the general tort jurisdiction of Article 7, 2. of Brussels Ibis Regulation. Apart from this provision, Brussels Ibis Regulation offers alternative jurisdictions, namely the defendant's domicile (Art. 4), the domicile of one of the defendants in case of joint tortfeasors (Art. 8) and the place chosen by the parties (Art. 25). In this article I only concentrate on the head of jurisdiction under Article 7, 2. of Brussels Ibis, the underlying rationale of which is not the protection of consumers but rather facilitating a predictable and close-to-court jurisdiction rule for sound administration of justice. [64]

26.As is known, Article 7, 2. of Brussels Ibis provides for special jurisdiction “in matters relating to tort, delict or quasi-delict, in the courts for the place where the harmful event occurred or may occur”. As is long established in the CJEU case law, the “place where the harmful event occurred” covers both the place where the damage occurred (locus actus) and the place of the event giving rise to it (locus damni), with the result that the defendant may be sued in the courts for either of those places. [65] In product liability cases this ubiquity rule leads to the following consequences. In Kainz the CJEU held that locus actus is the place where the product was manufactured. [66] Apart from domestic situations, this is usually not a favourable option for the consumer who has suffered damage due to a defective consumer good. However, the alternative jurisdiction of locus damni may offer a possibility to sue in his habitual residence. Namely, in Zuid Chemie, the CJEU clarified that the locus damni “is the place where the event which gave rise to the damage produces its harmful effects, that is to say, the place where the damage caused by the defective product actually manifests itself”. [67] As in most product liability cases the damage manifests itself in the consumer's habitual residence, the consumer has a de facto plaintiff jurisdiction - exactly as in contractual claims under Article 18, 1. of Brussels Ibis.

27.One must stress, however, that under Dumez [68] and Marinari [69] case law the plaintiff's jurisdiction under Article 7, 2. of Brussels Ibis is only available to the consumer if he has suffered initial damage (and not only indirect or consequential damage) at his place of habitual residence. This turned out to be one of the crucial issues in the recently decided Dieselgate case [70] where the referring Austrian court wanted the Court of Justice to say whether mere purchase of the manipulated cars from car dealers established in Austria, and the delivery of those vehicles in Austria, is sufficient to establish its jurisdiction under Article 7, 2. of Brussels Ibis as the “place where the damage occurred”. The Austrian court was also unsure whether the damage in the form of a reduction in the value of those vehicles constituted initial or consequential damage resulting from the defect of the cars - as only the former would establish the plaintiff's jurisdiction under the above-mentioned case-law. Finally, it was also disputable whether such damage is not a pure financial damage which again under the established case law of the CJEU would not give rise to the plaintiff's jurisdiction. [71]

28.The CJEU confirmed that the Austrian court had jurisdiction over Volkswagen AG under Article 7, 2. of Brussels Ibis as the place where the purchase took place is to be considered the place where the damage occurred. The Court reasoned that the damage in the form of reduction in value of the car constitutes initial damage and not only an indirect consequence of the harm suffered by others. [72] Moreover, it also does not constitute purely financial damage because contrary to the situations in previous case law where the reduction in the financial assets of the persons concerned had no connection to a tangible asset, here the case concerns a defect that affects vehicles, which are tangible assets. [73]

29.The reasoning of the Court which uses the tangibility of the asset as the main criterion for differentiating between purely financial damage and direct damage creates an interesting question concerning potential damages claims due to “defective” digital content. Could consumers who have bought manipulated (eg knowingly made non-secure) software use their home jurisdiction for claiming damages - that is, the reduction of the value of the software - from the software producer? If one only stresses the tangibility aspect as the Court of Justice did then the answer could turn out to be negative. True, it is very much disputable whether software is a product within the meaning of the Product Liability Directive. [74] However, as far as is known, the claim of the VKI in the Dieselgate case was not a product liability one, either. I would argue that producers of manipulated software should anticipate to be sued at the plaintiff's jurisdiction, ie at the place where the defective asset was bought, exactly as the producers of manipulated cars should. [75] Moreover, we should remember the competition law case Hydrogen Peroxide where the damages consisted of the additional cost that the plaintiff had to pay because of a cartel and where the CJEU accepted the plaintiff's jurisdiction. [76] One can argue that in case of defective software the damages also consist of additional payment of the buyer as they have paid too much for the reduced value of the software and therefore the plaintiff's jurisdiction should be available here as well.

30.Finally, it should of course be admitted that in many cases the problem described would be mitigated by the fact that in cases where the manipulation of the software has led to data breaches the affected data subject (the buyer) may use an alternative plaintiff's jurisdiction under Article 79, 2. of the GDPR. [77] As the damages claim under Article 82 of GDPR encompasses not only immaterial but also material damages including financial loss [78], reduction of value of digital content due to violations against data protection standards can be claimed at the court where the data subject has his habitual residence.

5.

Law applicable to product liability claims due to defective consumer goods and digital content

5.1.

Applicability of Article 5 of Rome II Regulation

31.Contrary to the jurisdiction rules in Brussels Ibis Regulation, Rome II Regulation determining the law applicable to non-contractual claims contains a specific - but very complex - conflict of law provision on producer's liability in Article 5. Therefore, when a consumer has suffered damage due to digital content or (inter-connected or other) digital services, the much-debated issue of whether software or other digital content can be qualified as a product within the meaning of that provision becomes relevant. For example, a research lab has developed a digital contact lens for persons with diabetes which is able to measure blood sugar level in tear fluid and send the data to a smartphone or a tablet. [79] Say, the consumer uses such a smart medical device (bought by another person so that no contractual claims are available) which has interconnected digital services and a defect in that digital service causes him health damage. Should the applicable law be determined according to Article 5 of Rome II or can only the general conflict of law norm of Article 4 be relied upon? The issue is obviously of most importance if the producer is situated outside the European Union.

32.Article 5, 1. of Rome II provides for a cascade of connecting factors which are to be applied in successive order: (a) the country of the consumer's habitual residence; (b) the country in which the product was acquired; and (c) the country in which the injury occurred. They all are subject to the condition that the product was marketed in that country. However, these connecting factors are not applicable if the defendant could not reasonably have foreseen the marketing of the product in the respective country. In such cases the law of the country of the habitual residence of the producer applies. [80] Finally, Article 5, 2. provides for a closer-connection-exception. [81] And of course, all these connecting factors may be set aside in case of a choice of law under Article 14 [82] or a common place of residence of the parties (Art. 5, 1. and 4, 2. of Rome II).

33.When speaking about the claims of consumers, we should also recall that although the Product Liability Directive is based on a consumer protection rationale, Article 5 of Rome II Regulation as a specific conflict of law provision for product liability is not specifically aimed at the protection of weaker parties. [83] The European legislator regarded the general applicable law principles of Article 4 of Rome II inadequate for product liability and sought to establish a conflict-of-law rule which meets “the objectives of fairly spreading the risks inherent in a modern high-technology society, protecting consumers' health, stimulating innovation, securing undistorted competition and facilitating trade”. [84] The creation of a cascade system of connecting factors, together with the foreseeability clause in Article 5, 1., was considered to be a balanced solution in regard to all these competing objectives.

34.When starting to apply the cascade provision of Article 5 of Rome II Regulation, we should first determine whether the concepts of “producer”, “product” and “damages” used there must be interpreted in line with the ones of the Product Liability Directive. At least as far as the notion of “product” is concerned, the majority tend to answer in the affirmative [85], not least due to the referral to the Product Liability Directive in the original Commission's proposal for Rome II. [86]

35.This, however, does not make it any easier as interpreting the concept of “product” under the Directive involves the unsolved debate of whether and under what conditions software can be considered a “product”. First, it is clear - at least in principle, that services as such are outside the scope of the Directive: the Court of Justice denied the applicability of the Product Liability Directive to service providers in the Besançon case. [87] This should, at least in principle, exclude digital services from the scope of Article 5 of Rome II but when taking a closer look the issue becomes even more blurry. At least in German literature the discussion concentrates on the intangible nature of the software and considers software a product if it is on a tangible carrier. [88] However, one also tries to overcome the intangibility issue by saying that when a user downloads software from the Internet to the hard disk then it becomes tangible and, hence, a product. [89] Only software used online (eg software-as-a-service) would not be qualified as a product and therefore not be covered by Article 5 of Rome II. [90]

36.At least software on or connected to a medical device delivering data to consumer's smartphone and leading to health damage should still be considered a product [91] allowing us to determine the law applicable to the non-contractual liability under the cascade system of Article 5, 1. of Rome II Regulation. This would most probably be different if health damage is caused by software-as-service or a medical app such as a blood pressure tracking and analysis tool which is unrelated to a specific tangible product. These cases may fall out of the scope of Article 5 of Rome II as a specific product liability conflict of law provision and hence the law applicable to the claim of a harmed person must be determined under the general rule of Article 4 of Rome II Regulation - allowing us, as we soon will see, to come to an even more consumer-friendly result.

37.As the ubiquity principle does not come into play under Article 4 of Rome II, the law applicable to the non-contractual claim of the consumer is the law of the country in which the damage occurs (lex loci damni) irrespective of the country in which the event giving rise to the damage occurred and irrespective of the country or countries in which the indirect consequences of that event occur. In cases of personal injury, the country in which the damage occurs is the country where the injury was sustained. [92] This often coincides with the habitual residence of the injured person and hence his interests are more protected under the general conflict of law rule than under Article 5 of Rome I.

38.True, the producer could still try to invoke the closer-connection rule in Article 4, 2. of Rome II and escape the “home law” of the consumer. This should, however, not be successful as such software is normally available on a contractual basis leading to the applicability of the mandatory consumer protection norms of the consumer's habitual residence under Article 6, 2. of Rome I Regulation. And even if the consumer using the software is not the contractual partner of the software producer, it does not seem unfair to subject the damages claim to the law of the consumer's habitual residence because the producer was aware of its software being used in that country and hence the applicable law is foreseeable to him.

5.2.

The “marketed” and “foreseeability” criteria in Article 5 of Rome II Regulation in case of smart consumer goods

39.A closer look at Article 5, 1. of Rome II reveals that every connecting factor of this cascade provision is only to be applied if the product was marketed in the given country. The concept of “marketing of the product” is to be interpreted in line with the corresponding concept of “putting into circulation” in Article 7, (a) of the Product Liability Directive. [93] The Court of Justice has not yet given clear guidance on the interpretation of this issue, defining in the O'Byrne case mostly the time when (and not necessarily the place where) the product was put into circulation. [94] Advertising a specific product on the Internet targeting the particular country is deemed sufficient for “marketing” within the meaning of Article 5 of Rome II, provided that at least one product has also been sold there. [95]

40.As described above, in case of smart consumer goods the defect leading to health damage may lay within the interconnected digital service or within embedded software. An obvious case would also be a connected car with an autonomous parking system: if a self-parking car causes bodily injury to a person due to a defect in the digital parking service, the relevant question when determining the law applicable to his damages claim should be whether the car (and not the parking software, eg the Mercedes MeConnect digital service) was marketed in the country of the injured person's habitual residence (Art. 5, 1., (a) of Rome II). The same result could be reached via the closer connection rule of Article 5, 2. of Rome II. A contradicting view, ie to ask whether the parking software was marketed in the injured person's habitual residence, would lead to problems of whether the software was put into circulation at the place of the relevant server (or cloud?) or maybe even anywhere in the world where a consumer may access it. [96]

41.In order to protect the interests of the producer as the potential tortfeasor, Article 5, 2. of Rome II declares that the applicable law cannot be determined under the cascade system of Article 5, 1. of Rome II if the person claimed to be liable could not reasonably foresee the marketing of the product in the respective country. Instead, the law of the country where the person claimed to be liable is habitually resident should be applied. Concerning the foreseeability element it is pointed out by van Calster that it requires certain activity of the producer towards the territory where the product ended up, thereby resembling the “targeting” concept for consumer contracts in Rome I and Brussels Ibis Regulations. [97] Surely, these concepts cannot be considered identical as directing its activities towards the country of consumer's habitual residence requires somewhat stronger input by the trader than mere foreseeability that its products are marketed there. It is rightly stressed by Bach that the foreseeability requirement becomes problematic and should be interpreted restrictively especially in case of secondary markets on the Internet, for instance in the case of selling second-hand products via online-auctions or digital platforms: these secondary distribution channels are often not foreseeable nor controllable for producers. [98] Thus, product liability claims for second-hand smart goods which were originally meant only for eg US markets but which have ended up in Europe via eBay are subject to the law of the residence of the producer. This should not, however, apply to cases where the producer allows a non-US consumer to activate the smart product over the Internet also in Europe: in this case the producer cannot rely upon the unforeseeability defence.

6.

Conclusions

42.The new digital content and consumer sales directives do not create substantial new problems concerning international consumer jurisdiction or applicable law rules. Yet, in cases where a sales contract of a smart good is interconnected with another contract concluded with a digital services provider, the dichotomy of contracts is reflected in the jurisdiction and applicable law: it may happen that the jurisdiction and applicable law of a sales contract of a smart good (eg a connected car) deviates from the jurisdiction and applicable law of the interconnected contract for digital services (eg a navigation and entertainment system of that car). In many cases these contracts can be considered “inseparably linked” within the meaning of the Maletic case as the consumer would not be able to use the smart good without the digital service. However, in most cases a “mobile” consumer who has acquired the smart item from abroad but concluded a digital services contract later at his domicile should not be able to use his home jurisdiction for asserting claims both against the seller as well as the service provider.

43.Domestic contract law provisions goldplating the rules of digital content and consumer sales directives should, as a rule, not be considered overriding mandatory norms within the meaning of Article 9 of Rome I Regulation. The issues regulated by the Digital Content Directive and Consumer Sales Directive - consumer's remedies for lack of conformity, guarantees and regress claim of the trader - are not likely to protect the public interest of the forum within the meaning of Article 9, 1. of Rome I. Moreover, due to the interplay with jurisdiction rules, the question will only arise in specific circumstances and the discussion therefore remains mostly of theoretical nature.

44.Consumers who have suffered health or property damage due to a smart good can assert their product liability claims against producers at the place where the damage caused by the defective product actually manifests itself - which in most cases coincides with the habitual residence of the consumer. Hence, the jurisprudence of the Court of Justice has afforded them a de facto plaintiff jurisdiction. This plaintiff jurisdiction is also available for claiming the reduction of value of a defective smart good, (as recently clarified in the Dieselgate case) as well as for damage caused by privacy infringements under Article 79, 2. of the GDPR. The law applicable to the product liability claims due to defective smart goods is usually to be determined under the cascade system of Article 5 of Rome II Regulation. However, in cases where the harm was caused by a digital service such as a health application, we are back at the old dispute of whether software as such can be considered a product leading to product liability. Even if the answer were negative, in most cases the consumer could still rely upon the law of his habitual residence under Article 4 of Rome II.