Posts Tagged CJEU
‘Like Dassonville on steroids’. Bobek AG in Rheinland on personality v territoriality, the nature of EU harmonisation, and its links with (as well as historic roots of) conflict of laws and regulatory competition.
In advising on a territorial restriction in an insurance clause earlier this month, I studied the CJEU judgment in C-581/18 Rheinland, important for the (limitations to the) reach of Article 18 TFEU, the general non-discrimination requirement on the basis of nationality. Bobek AG had earlier opined, and the Court followed, that in the absence of harmonisation and in a scenario with no EU links, Article 18 TFEU is not engaged. I had missed the AG’s earlier opinion – forgive me if I am late to this party.
It is important to sketch the context: Bobek AG had summarised the facts as
A German patient received, in Germany, defective breast implants manufactured by Poly Implant Prothèse SA (‘PIP’), a French undertaking that is now insolvent. The patient seeks compensation before the German courts from Allianz IARD SA, the French insurer of PIP. In France, manufacturers of medical devices are under a statutory obligation to be insured against civil liability for harm suffered by third parties arising from their activities. That obligation led PIP to conclude an insurance contract with Allianz, which contained a territorial clause limiting the cover to damage caused on French territory only. Thus, PIP medical devices that were exported to another Member State and used there were not covered by the insurance contract.
In this context, the Oberlandesgericht Frankfurt am Main (Higher Regional Court, Frankfurt am Main, Germany) enquires whether the fact that PIP was insured by Allianz for damage caused by its medical devices on French territory only, to the exclusion of that potentially caused in other Member States, is compatible with Article 18 TFEU and the principle of non-discrimination on grounds of nationality contained therein.
This post is not on Article 18 TFEU. Rather, consider the excellent (and eloquent) discussion by Bobek AG at 109 ff. Does the imperative of equal protection of all European citizen-consumers, in the absence of EU harmonising law on the issue, preclude a national rule that, in effect, limits insurance cover to persons who undergo surgery on the territory of the Member State, thus indirectly limiting the cover to citizens of that Member State? Bobek AG emphatically and despite moral sympathy for the victims, says no. The alternative would be ‘like Dassonville on steroids’ (at 111), it would ‘turn regulatory competence within the internal market on its head’ (at 109).
Consider his link with conflict of laws at 114-115:
In other words, the fact that goods once came from another Member State is not a sufficient reason to suggest that any matter later concerning those goods is covered by EU law. If that logic were to be embraced, by a questionable interpretation of Article 18 TFEU, the movement of goods in Europe would become (once again) reminiscent of medieval legal particularism, [at footnote 78 he refers to the excellent work by my legal history colleague Randall Lesaffer] whereby each product would, like a person, carry its own laws with it. Goods would be like snails, carrying their homes with them in the form of the legislation of their country of origin, to be applicable to them from their production to their destruction.
Such a consequence would not only displace any (normal) territoriality in the application of laws, but would also generate conflicts of regulatory regimes between the Member States. Indeed, such an expansionist interpretation of Article 18 TFEU could make the legislation of any of the Member States potentially applicable on the same territory without any clear and objective criteria as to which legislation should prevail in a given dispute, with the victim being able to choose the most favourable legislation.’
Most delightful analysis.
CJEU in Novo Banco: confirms mere presence of a natural person’s core immovable asset (the ‘family home’) does not in itself determine COMI (in insolvency).
When I reviewed Szpunar AG’s opinion, I pointed out that the crux of this case is the determination of ‘centre of main interests’ in the context of natural persons not exercising an independent business or professional activity, who benefit from free movement. The CJEU has now held.
With respect to natural persons outside of a profession, the Insolvency Regulation 2015/848 (‘EIR 2015’) determines ‘(i)n the case of any other individual, the centre of main interests shall be presumed to be the place of the individual’s habitual residence in the absence of proof to the contrary. This presumption shall only apply if the habitual residence has not been moved to another Member State within the 6-month period prior to the request for the opening of insolvency proceedings.’
‘Habitual residence’ is not defined by the EIR 2015. The CJEU runs along the usual themes: need for predictability and autonomous interpretation; emphasis on the Regulation generally defining COMI as ‘the place where the debtor conducts the administration of his interests on a regular basis and is therefore ascertainable by third parties’ (at 19 and referring to recital 13 of the previous Regulation); among those third parties, the important position of (potential) creditors and whether they may ascertain said centre (at 21); to agree with the AG at 24 that
relevant criteria for determining the centre of the main interests of individuals not exercising an independent business or professional activity are those connected with their financial and economic situation which corresponds to the place where they conduct the administration of their economic interests or the majority of their revenue is earned and spent, or the place where the greater part of their assets is located.
Like the AG, the CJEU holds that the mere presence of a natural person’s one immovable asset (the ‘family home’, GAVC) in another Member State than that of habitual residence, in and of itself does not suffice to rebut COMI (at 28).
At 30, the Court specifically flags that COMI in effect represents the place of the ’cause’ of the insolvency, i.e. the place from where one’s assets are managed in a way which led the insolvent into the financial pickle:
In that regard, although the cause of the insolvency is not, as such, a relevant factor for determining the centre of the main interests of an individual not exercising an independent business or professional activity, it nevertheless falls to the referring court to take into consideration all objective factors, ascertainable by third parties, which are connected with that person’s financial and economic situation. In a case such as the one in the main proceedings, as was observed in paragraph 24 above, that insolvency situation is located in the place where the applicants in the main proceedings conduct the administration of their economic interests on a regular basis or the majority of their revenue is earned and spent, or the place where the greater part of their assets is located.
As in all other scenarios of rebuttal, the ascertainability in particular by (potential) creditors is key and is a factual consideration which the national courts have to make.
(Handbook of) EU private international law, 2nd ed. 2016, Chapter 5, Heading 5.6.1.
The CJEU in Movic on enforcement of unfair trading practices and the less than abstract determination of ‘civil and commercial’.
I reviewed Szpunar AG’s Opinion in C-73/19 Belgische Staat v Movic BV et al here. The CJEU held this morning. At the time of posting an English version of the judgment was not yet available. The case at issue concerns enforcement of Belgium’s unfair trading act by the public authorities of the Member State. Movic BV of The Netherlands and the others defendants practices ticket touting: resale of tickets for leisure events.
The Court is more succinct than the AG in its analysis yet refers repeatedly to points made by Szpunar AG without itself therefore having to refer to so extensive an analysis.
The fact that a power was introduced by a law is not, in itself, decisive in order to conclude that the State acted in the exercise of State authority (at 52). Neither does the pursuit of the general interest automatically involve the exercise of public powers (at 53). With respect to the authorities’ powers of investigation, it would seem that the Court like the AG reads (at 57) C‑49/12 Sunico as meaning that to exclude proceedings from the scope of ‘civil and commercial matters’, it must be determined, in concreto, whether the public authority uses evidence which it has in its possession as a result of its public powers of investigation, hence putting it in a different position as a person governed by private law in analogous proceedings. Collecting evidence in the same way as a private person or a consumer association could, does not fall within that category (at 58).
Neither the request for penalty payments nor an application for an injunction makes the proceedings drop out off Brussels Ia: both instruments are available to private parties, too. That is not however the case for the observation of continued infringement by mere civil servant oath as opposed to bailiff certification. This, the Court holds like the AG, does amount to exercise of public authority (at 62) however (at 63) that element alone escapes BIA, it does not so taint the other part of the proceedings.
As I noted in my review of first Advocate General Szpunar’s Opinion, the need for highly factual considerations sits uneasily with the Regulation’s expressed DNA of predictability. However this squares with the CJEU case-law on ‘civil and commercial’.
(Handbook of) EU Private International Law, 2nd ed. 2016, Heading 2, Heading 2.2.
The CJEU’s locus damni determination in Volkswagen dismisses a US style minimum contacts rule. Like the passat, it risks picking up suits and landing them almost anywhere.
Update 10 July 2020 a few hours after posting: I revisited the pending, distinct reference by the Austrian Supreme Court (see Rouzbeh Moradi’s flag here) on type approval issues (which the High Court has actually dealt with as acte clair in  EWHC 783 (QB), referred to here). I was hoping there might be scope in those questions for the CJEU to fill in the blanks signalled below. I fear there is not.
I earlier reviewed Sánchez-Bordona AG’ opinion in C‑343/19 Verein für Konsumenteninformation v Volkswagen. I noted then that despite attempts at seeing system in the Opinion, the ever unclearer distinction between direct and indirect aka ‘ricochet’ damage under Article 7(2) Brussels Ia is a Valhalla for reverse engineering.
The AG did not suggest a wild west of connecting factors for indirect damage (please refer to my full post for overview), instead suggesting a Universal Music style requirement of extra factors (over and above the location of damage) to establish jurisdiction. In particular he put forward a minimum contacts rule such as in US conflict of laws: at 75: ‘the defendant’s intention to sell its vehicles in the Member State whose jurisdiction is in issue (and, as far as possible, in certain districts within that State).’
The CJEU’s judgment yesterday was received as giving ‘consumers’ the right to sue Volkswagen in their state of domicile. This however is not quite correct. Firstly, the parties at issue are not ‘consumers’ at least within the meaning of European conflicts law: the suit is one in tort, not contract, let alone one that concerns a consumer contract. Further, the AG was clear and the CJEU arguably held along the same lines, that it is only if the car was purchased by a downstream (third party) buyer and the Volkswagen Dieselgate story broke after that purchase, that the damage may be considered to only then have come into existence, thus creating jurisdiction. See the CJEU at 29 ff:
29. That said, in the main proceedings, it is apparent from the documents before the Court, subject to the assessment of the facts which it is for the referring court to make, that the damage alleged by the VKI takes the form of a loss in value of the vehicles in question stemming from the difference between the price paid by the purchaser for such a vehicle and its actual value owing to the installation of software that manipulates data relating to exhaust gas emissions.
30 Consequently, while those vehicles became defective as soon as that software had been installed, the view must be taken that the damage asserted occurred only when those vehicles were purchased, as they were acquired for a price higher than their actual value.
31 Such damage, which did not exist before the purchase of the vehicle by the final purchaser who considers himself adversely affected, constitutes initial damage within the meaning of the case-law recalled in paragraph 26 of the present judgment, and not an indirect consequence of the harm initially suffered by other persons within the meaning of the case-law cited in paragraph 27 of the present judgment.
That ‘case-law cited’ is the classic lines of cases on locus damni per A7(2) BIa, with Trans Tibor as its latest expression.
The CJEU does not qualify the damage as purely financial: at 33, citing the EC’s court opinion: ‘the fact that the claim for damages is expressed in euros does not mean that the damage is purely financial.’: the car, a tangible asset, actually suffers a defect, over and above the impact on its value as an asset. That is a statement which cuts many a corner and which has relevance beyond the EU regime for all ‘money judgments’ (think of e.g. the Hague Judgments Convention). Update 10 July 2020 after initial posting: thank you Gordon Nardell QC for pointing out that the CJEU view here is at odds with the English conflicts rules (and with many other, I reckon) on characterisation of loss as pecuniary.
Predictability, which is firmly part of the Brussels Ia Regulation’s DNA, the Court holds, is secured seeing as a car manufacturer which ‘engages in unlawful tampering with vehicles sold in other Member States may reasonably expect to be sued in the courts of those States (at 36).
Finally, the Court throws consistency with Rome II in the mix, by holding at 39
Lastly, that interpretation satisfies the requirement of consistency laid down in recital 7 of the Rome II Regulation, in so far as, in accordance with Article 6(1) thereof, the place where the damage occurs in a case involving an act of unfair competition is the place where ‘competitive relations or the collective interests of consumers are, or are likely to be, affected’. An act, such as that at issue in the main proceedings, which, by being likely to affect the collective interests of consumers as a group, constitutes an act of unfair competition (judgment of 28 July 2016, Verein für Konsumenteninformation, C‑191/15, EU:C:2016:612, paragraph 42), may affect those interests in any Member State within the territory of which the defective product is purchased by consumers. Thus, under the Rome II Regulation, the place where the damage occurs is the place in which such a product is purchased (see, by analogy, judgment of 29 July 2019, Tibor-Trans, C‑451/18, EU:C:2019:635, paragraph 35).
The extent to which A6 Rome II applies to acts of unfair competition being litigated by ‘consumers’ (in the non-technical sense of the word), is however not quite clear and in my view certainly not settled by this para in the Court’s judgment.
Finally, on locus delicti commissi as I noted at the time, the AG had not in my view given a complete analysis. The CJEU is silent on it.
Not many will feel much sympathy for Volkswagen facing cluster litigation across the EU given its intention to cheat. However the rejection of a minimum contacts approach under A7(2) will have implications reaching small corporations, too. The Volkswagen ruling has many loose ends and will need distinguishing, with intention to defraud the consumer arguably a relevant criterion for distinction given the Court’s finding in para 36.
It is to be feared that many national judges will fail to see the need for distinguishing, adding to the ever expanding ripple effect of locus damni following the Court’s epic Bier judgment.
Ps reference to the Passat in the title is of course to the VW Passat, named after the Germanic name for one of the Trade winds.
(Handbook of) EU Private International Law, 2nd ed. 2016, Chapter 2, Heading 184.108.40.206.7
Szpunar AG in Ellmes Property Services. Again, on rights in rem and, more challenging, on forum contractus and the spirit of CJEU De Bloos.
Acte clair is in the eyes of the beholder, I assume. However a confident judge would have sufficient CJEU authority to help them hold on the A24(1) BIa issues in C‑433/19 Ellmes Property Services in which Szpunar AG opined last week. (No EN version available at the time of publication of this post).
Do actions brought by a co-owner seeking to prohibit another co-owner from carrying out changes to his property subject to co-ownership, in particular to its designated use, arbitrarily and without the consent of the other co-owners, concern the assertion of a right in rem? In the negative, is the forum contractus per A7(1)(a) Brussels Ia the location of the property? The less clear issue in my view is the forum contractus element.
The location is Zell am Zee, contested use is, not surprisingly, tourist accomodation. Applicant in the national proceedings is an individual who lives in the apartment building. Defendant is a UK corporation who uses it for short-term lets despite the residential designation assigned to the building as a whole in the co-ownership agreement.
From CJEU authority including C-438/12 Weber v Weber it should be clear that other than the hardcore cases of ownership of real estate, the erga omnes v in personam character of rights in real estate depends on national law. The Advocate General in this respect points out that for the rights of co-owners in the case at issue to be rights in rem, Austrian law would have to be enable them to exercise these rights not just vis-a-vis the other co-owners, but also vis-a-vis third parties such as tenants. Whether this is the case in Austrian law has not been sufficiently explained in the reference, it seems.
For the impact of entry in the land register (where third parties can consult the co-ownership agreement), Szpunar AG reviews and contrasts C‑417/15 Schmidt v Schmidt, and C-630/17 Milivojević v Raiffeisenbank. Mere registration does not always entail erga omnes impact.
The Advocate General reminds us of the overall interpretation of Article 24, including the need for restrictive interpretation, and flags (with reference inter alia to the Handbook, p.73, for which I am, as always, sincerely humbled) that it is not just, or not even so much sound administration of justice which underlies A24. At least partially, Member States’ strategic interests are served by the issues listed in the Article.
Ellmes Property Services does not seem to raise additional issues such as we saw in C-25/18 Kerr. The Austrian courts could have dealt with this on their own, and seeing as the referring judge did not provide the kind of detail for the CJEU to judge, the AG’s suggestion is to leave it up to them to verify the erga omnes character.
That leaves (whether it will be needed depends on what the eventual insight will be on the erga omnes element), the forum contractus under A7(1). Parties differ as to the qualification of the contractual duty: is it a positive one (do!) or a negative one (must not!). The AG opts for the latter, with reference to CJEU 14/76 De Bloos: A7(1) refers to the contractual obligation forming the basis of the legal proceedings. I find the precedent value of De Bloos problematic in light of the many changes that have been made to Article 7 since, and in light of the engineering possibilities it hands to parties.
The AG advises that forum contractus will have to be determined by the Italian judge following the conflicts method per CJEU 12/76 Tessili v Dunlop, with little help from European harmonisation seeing i.a. as the initial co-ownership agreement dates back to 1978.
I am curious to see how far the Court will go in entertaining the issues at stake.
(Handbook of) EU Private International Law, Chapter 2, Heading 220.127.116.11 (cited by the AG) and Heading 18.104.22.168.
Alexander bros v Alstom. A reminder of the relevance of EU law for New York Convention refusal of recognition of arbitral awards on ordre public grounds.
In Alexander Brothers Ltd (Hong Kong SAR) v Alstom Transport SA & Anor  EWHC 1584 (Comm) Cockerill J discussed inter alia (at 177 ff) the impact of EU law on the ordre public assessment for potential refusal of recognition of an arbitral award under section 103 of the 1980 New York Convention.
CJEU authority are C-126/ 97 Eco Swiss (concerning EU competition law) and C-168/ 05 Claro (unfair terms in consumer contracts). At 183 Cockerill J does not suggest the CJEU authority should no longer stand. Indeed she suggests obiter that there is no reason to suggest the CJEU’s line of reasoning should not apply to wider issues than just competition law or consumer law. However, the burden of proof of showing that particular parts of EU law are of a nature to justify the ordre public exception, lies upon the party objecting to recognition. In casu Alstom have fallen short of that duty. Yes, there is scant reference to anti-corruption in the private sector; and yes there is EU money laundering law. However (at 186) ‘the EU has, in general terms, set its face against corruption. But aside from the area of money laundering it has not put in place mandatory laws or rules. In the context of international corruption of the kind in focus here it has left it to the individual member states to adopt what measures seem good to them. There is, in short, no applicable mandatory rule or public policy.’
An interesting discussion.
In T‑574/18 Agrochem-Maks the General Court at the end of May upheld the Commission Regulation not extending market authorisation for the active substance oxasulfuron, a pesticide. The EC Regulation noted that EFSA, the European Food Safety Authority, had identified a large number of data gaps resulting in the inability to finalise the risk assessment in several areas and that ‘in particular, the available information on oxasulfuron and its metabolites did not allow finalising the assessment of the overall consumer exposure, the groundwater exposure, the risk to aquatic organisms, earthworms, soil macro and microorganisms and non-target terrestrial plants’. Since ‘it has not been established with respect to one or more representative uses of at least one plant protection product that the approval criteria provided for in Article 4 of Regulation … No 1107/2009 [on plant protection products; see here, GAVC] [were] satisfied’, authorisation was not renewed.
The case at issue is brought by a small Croatian, family-owned company. That is a change from the classic pattern in this kind of cases, with large bio-agricultural industry routinely taking cases to the CJEU in laser-shoot fashion, hoping they might hit the target once or twice.
The General Court extensively outlines the procedure foreseen in the relevant EU laws, thereby identifying the core issue in near all of these cases held under the precautionary principle: the EU courts do not carry out a merits review; rather, they assess whether holes have emerged in the preparation of a decision, which could mean that the Institutions could not reasonably have come to the decision they came to.
That is no different here: at 62: ‘the EU Courts must verify that the relevant procedural rules have been complied with, that the facts admitted by the Commission have been accurately stated and that there has been no manifest error of appraisal or misuse of powers’. At 65, per CJEU T-13/99 Pfizer: ‘a scientific risk assessment carried out as thoroughly as possible on the basis of scientific advice founded on the principles of excellence, transparency and independence is an important procedural guarantee whose purpose is to ensure the scientific objectivity of the measures adopted and preclude any arbitrary measures.’
Specifically for current Regulation: at 66: ‘the burden of proving that the conditions for approval or renewal under Article 4 of Regulation No 1107/2009 are met lies, in principle, with the notifier.’ At 67 per CJEU T-584/13 BASF Agro: ‘it is the person seeking approval who must prove that the conditions of such approval are met in order to obtain it, and not the Commission which must prove that the conditions of approval are not met in order to be able to refuse it’.
The General Court then at length considers the procedure followed, including the reasons for the identified gaps, and then assesses the application of the precautionary principle to same: at 109 ff with reference to the 2000 Communication on the Precautionary Principle, COM(2000)1. Crucially, at 121, as noted ‘(u)nder Regulation 1107/2009 when the applicant words its renewal application, it bears the burden of proving the efficacy and safety of the substance in question.’ ‘Since it did not discharge that burden, the approval of the active substance could not be renewed.’
The case highlights once again the crucial nature of administrative compliance with the rulebooks under EU regulatory law. Many of us will have sat through presentations by EFSA or EC officials outlining the rules in excruciating and yes, not very sexy detail. Yet to follow procedure to a tee is crucial to ensure defence against corporations taking issue with the findings at the CJEU.
The case also emphasises the importance of burden of proof, as specified in the secondary law at issue and, preferably, the ‘no data, no market’ rule in EU regulatory law.
There might of course still be an appeal with the Court.
EU environmental law (with Leonie Reins), Edward Elgar, 2018, p.28 ff.
Must Article 107 TFEU be interpreted as meaning that a system whereby a private, non-profit eco-body, approved by the public authorities, receives contributions from those who place on the market a particular category of product and who enter into a contract with it to that effect, in return for a service consisting in the organisation on their behalf of the treatment of the waste from those products, and redistributes to operators responsible for the sorting and recovery of that waste, subsidies the amount of which is set out in the approval, in the light of environmental and social targets, is to be regarded as State aid within the meaning of that provision?
That is the question as phrased in C‑556/19 Société Eco TLC and on which Pitruzzella AG Opined on 28 May. TLC stands for Textiles, Lignes de maisons, and chaussures (textiles, household linen and shoes). Producers or as the case may be first importers pay a fee to the collective body in lieu of their personal commitments under extended producers responsibility per Waste Framework Directive 2008/98.
The AG of course revisits the definition of ‘State Aid’ under CJEU C-379/98 Preussen Elektra, on which more here and here. Preussen Elektra remains controversial for it would seem to give Member States quite a bit of room for manoeuvre to reach the same result as direct State Aid more or less simply by inserting a private operator who receivs funds directly from private operators however in line with direct State instructions on level and modalities of payment. The AG opines that in the case at issue there is no State Aid however he directs further factual lines of enquiry (ia re the State control over payments by the collective body to recyclers.
Handbook of EU Waste law, 2nd ed. 2015 OUP, para 4.116 ff.
C-500/18 AU v Reliantco was held by the CJEU on 2 April, in the early fog of the current pandemic. Reliantco is a company incorporated in Cyprus offering financial products and services through an online trading platform under the ‘UFX’ trade name – readers will recognise this from  EWHC 879 (Comm) Ang v Reliantco. Claimant AU is an individual. The litigation concerns limit orders speculating on a fall in the price of petrol, placed by AU on an online platform owned by the defendants in the main proceedings, following which AU lost the entire sum being held in the frozen trading account, that is, 1 919 720 US dollars (USD) (around EUR 1 804 345).
Choice of court and law was made pro Cyprus.
The case brings to the fore the more or less dense relationship between secondary EU consumer law such as in particular the unfair terms Directive 93/13 and, here, Directive 2004/39 on markets in financial instruments (particularly viz the notion of ‘retail client’ and ‘consumer’).
First up is the consumer title under Brussels Ia: Must A17(1) BIa be interpreted as meaning that a natural person who under a contract concluded with a financial company, carries out financial transactions through that company may be classified as a ‘consumer’ in particular whether it is appropriate, for the purposes of that classification, to take into consideration factors such as the fact that that person carried out a high volume of transactions within a relatively short period or that he or she invested significant sums in those transactions, or that that person is a ‘retail client’ within the meaning of A4(1) point 12 Directive 2004/39?
The Court had the benefit of course of C-208/18 Petruchová – which Baker J did not have in Ang v Reliantco. It is probably for that reason that the case went ahead without an Opinion of the AG. In Petruchová the Court had already held that factors such as
- the value of transactions carried out under contracts such as CFDs,
- the extent of the risks of financial loss associated with the conclusion of such contracts,
- any knowledge or expertise that person has in the field of financial instruments or his or her active conduct in the context of such transactions
- the fact that a person is classified as a ‘retail client’ within the meaning of Directive 2004/39 is, as such, in principle irrelevant for the purposes of classifying him or her as a ‘consumer’ within the meaning of BIa,
are, as such, in principle irrelevant to determine the qualification as a ‘consumer’. In Reliantco it now adds at 54 that ‘(t)he same is true of a situation in which the consumer carried out a high volume of transactions within a relatively short period or invested significant sums in those transactions.’
Next however comes the peculiarity that although AU claim jurisdiction for the Romanian courts against Reliantco Investments per the consumer title (which requires a ‘contract’ to be concluded), it bases its action on non-contractual liability, with applicable law to be determined by Rome II. (The action against the Cypriot subsidiary, with whom no contract has been concluded, must be one in tort. The Court does not go into analysis of the jurisdictional basis against that subsidiary, whose branch or independent basis or domicile is not entirely clear; anyone ready to clarify, please do).
At 68 the CJEU holds that the culpa in contrahendo action is indissociably linked to the contract concluded between the consumer and the seller or supplier, and at 71 that this conclusion is reinforced by A12(1) Rome II which makes the putative lex contractus, the lex causae for culpa in contrahendo. At 72 it emphasises the need for consistency between Rome II and Brussels IA in that both the law applicable to a non-contractual obligation arising out of dealings prior to the conclusion of a contract and the court having jurisdiction to hear an action concerning such an obligation, are determined by taking into consideration the proposed contract the conclusion of which is envisaged.
(Handbook of) EU private International Law, 2nd ed. 2016, Chapter 2, Heading 22.214.171.124.