Investor Protection and Issuer Confidence after Kolassa
By Matteo Gargantini, Senior Research Fellow MPI Luxembourg
The decision rendered by the ECJ in Kolassa (Case C-375/13) offers a good opportunity to assess the European rules on jurisdiction from the point of view of investor protection and issuer confidence. A first comment on Kolassa has already been published on this Blog by Professor Matthias Lehmann. In his post, Professor Lehmann mainly focuses on the application of Art. 5(3) Brussels I Regulation to prospectus liability and on the evidence a court needs to consider when the disputed facts are relevant both for establishing jurisdiction and for deciding on the merit (these topics are addressed respectively in the third and the fourth questions referred to the ECJ). Full reference can therefore be made to Professor Lehmann’s accurate analysis both for such points and for the description of the relevant facts. This post will instead sketch some general remarks from the perspective of financial markets law (for a more detailed analysis based on the Opinion of the Advocate General in Kolassa see Gargantini, Jurisdictional Issues in the Circulation and Holding of (Intermediated) Securities: The Advocate General’s Opinion in Kolassa V. Barclays, Rivista di diritto internazionale privato e processuale (2014), 1095).
To better understand the issues raised by Kolassa, it is worth considering in more detail the first two questions referred by the Austrian court, namely whether for the purpose of Art. 15 Brussels I Regulation Barclays, the issuing company, and Mr Kolassa, the final investor, are part of a contract, or whether for the purpose of Art. 5(1) Brussels I Regulation the relationship between them can at least be considered contractual. As opposed to the claim considered by the third question – which only refers to prospectus liability and to “breach of obligations to protect and advise” – the claims dealt with by the first two questions were also based on “the bonds terms and conditions”. Hence, it appears that Mr Kolassa was relying not only on prospectus liability, but also on a direct violation of the bond terms, that being the missing payments. Therefore, the clarifications provided by the ECJ on prospectus liability are not the full story. First, nothing prevents investors from filing claims exclusively – or, as Mr Kolassa did, also – on the basis of violation of the bond terms and conditions. Second, it might well be the case that a security offering is carried out with no prospectus being published at all, for example because one of the exemptions set forth by Art. 4 Directive 2003/71/EC (on the prospectus to be published when securities are offered to the public or admitted to trading) applies.
The first two questions referred to the ECJ raise difficult problems because, in Kolassa, not only are the securities bought on the secondary market, with no direct contact between issuer and investor, but they are also held by Mr Kolassa’s bank (direktanlage) rather than by Mr Kolassa himself. In such a scheme, Mr Kolassa only has a claim against his bank and cannot be regarded as the holder of the securities. The distinction between the problems raised by security circulation, on the one hand, and security holding, on the other, is clearly drawn in the questions referred by the Austrian courts. Both the Opinion of the Advocate General and the ECJ decision deny that Art. 5(1) and Art. 15 apply, but they are unfortunately not as clear as the referring court in discerning the two aspects. Para. 26 of the decision seemingly links the absence of a contract to the fact that Mr Kolassa is not the bearer of the bond. Hence, it could be inferred that the “chain of contracts through which certain rights and obligations of the professional […] are transferred to the consumer” (para. 30) refers to the contracts that compose the holding chain of the securities. However, para. 35 is more elliptical and might also include security circulation when it refers to “an applicant who, as a consumer, has acquired a bearer bond from a third party professional, without a contract having been concluded between that consumer and the issuer of the bond”. Likewise, the applicability of Art. 5(1) is excluded on the basis that “a legal obligation freely consented to by Barclays Bank with respect to Mr Kolassa is lacking”, it being unclear whether this is linked to the fact that the bonds were purchased on the secondary market or to the fact that direktanlage, rather than Mr Kolassa, should be regarded as the bearer of the certificate (para. 40).
Whether the inapplicability of Arts. 5(1) and 15 Brussels I derives from the fact that the bonds are bought from previous purchasers rather than underwritten directly from the issuer or, instead, from the fact that Mr Kolassa is not the holder of the securities is however key to understanding the implications of the decision. If the first explanation prevailed, the consumer protection regime of Art. 15 would not easily apply in securities offerings whenever – as is often the case – a bank syndicate first underwrote the securities and then resold them to investors at large (so-called “firm commitment syndicate”). At the same time, ruling out a contractual obligation pursuant to Art. 5(1) on similar grounds would imply that issuers might be held liable for violation of the bonds’ terms and conditions in any jurisdiction where their investors suffered economic loss according to Art. 5(3). Such a system would exclude retail investor protection with no economic rationale and would paradoxically expose the offering companies to the risk of being sued by professional investors in jurisdictions where they published no prospectus and, consequently, addressed no investor.
Therefore, although the distinction between circulation and holding of securities may not be decisive in Kolassa, its implications remain whenever the investor/accountholder is the bearer of the relevant securities. Since Kolassa does not provide a conclusive answer to these questions, it might be appropriate to give a narrow reading to the decision, hence considering the intermediated and indirect holding of the securities through direktanlage as the reason why Arts. 5(1) and 15 do not apply.
To be sure, even a restrictive reading of Kolassa, although preferable, is no panacea. First, it would leave open the question whether the circulation of the securities might still prevent the identification of a contract or even a contractual obligation between issuers and investors pursuant to Arts. 15 and 5 respectively. This would seem to be the case for Art. 15, because ECJ case law usually requires a direct contact between the two parties (see Von Hein, Verstärkung des Kapitalanlegerschutzes: Das Europäische Zivilprozessrecht auf dem Prüfstand, in Eur. Zeitschrift für Wirtschaftsrecht, 2011, 370). A different result may perhaps be reached for Art. 5(1), considering that it might apply in the absence of a direct contact and that the ECJ has stated that conditions incorporated in a security may be transferred along with the security when this is handed over (see e.g. Coreck, Case C-387/98), which is exactly the purpose of incorporating a restitution obligation into a bond. Second, linking the applicability of Arts. 5(1) and 15 to the formal qualification of the investor as security holder might easily create a differential treatment of investors that are regarded as mere beneficial owners in countries such as the United Kingdom, where security holding is mainly based on trusts. In this context, the strict interpretation of Art. 15 and the raison d’être of the autonomous interpretation of jurisdictional rules come into conflict.
To what extent a different reading of the applicable rules could ensure a better regulatory framework remains to be seen. The Brussels I Regulation does not always seem to leave room for different interpretations, at least in the light of consolidated case law. Art. 15 and its traditional understanding is a clear example. What is sure, from the point of view of securities law, is that the drawbacks of the current system reduce both issuer confidence and investor protection.
It is worth adding that the parallel claims based on the bond itself is a rather peculiar feature of the present case and is not typical for prospectus liability litigation. You are of course absolutely right that the Kolassa judgment is also quite unclear on how to deal with them under the Brussels I Regulation. In my opinion, one should not look at the technical way in which these instruments are held, acquired or distributed. For instance, it cannot matter for jurisdictional purposes whether the issuance of the bond was based on a “firm commitment” or on a “best efforts underwriting”, or how the indirect holding system is organised. From the investors’ perspective, these points are purely coincidental. As you rightly suggest, it is necessary to find a forum that can be predicted by both the investor and the issuer. Article 15 Brussels I should be discarded, as it would lead to a fragmentation of bond disputes. Given that the provision requires a direct contract between a business and a consumer, it can only be applied in the rare cases in which the investors themselves underwrite the bond. Generally, it seems more appropriate to apply Article 5(1). Lit. a requires to determine the place of performance of the bond according to the national law applicable to the investor’s claim (see ECJ cases Tessili and De Bloos). Since the conflict-of-laws rules on contractual obligations have been harmonised in the EU by the Rome I Regulation, every Member State should apply the same law to these questions. The result is that one and the same court will be competent for all bond claims of the entire issue, which is consistent with international practice. The actions based on statutory prospectus liability would still have to be dealt with under Article 5(3) Brussels I.
Kolassa is a difficult decision to interpret. On one view it may be a rather clumsy attempt to extend the consumer protection provisions of the Brussels Regulation into an area they do not expressly cover.
Had Mr Kolassa had a contract with Barclays Bank, then he could have relied on Article 15 of the Regulation, and brought proceedings in Austria as the place of his domicile (Regulation 44/2001 Article 16(1)). That seems to have been the result the CJEU were striving for, and perhaps that is not surprising given the strong policy in favour or protecting consumer rights reflected in a number of European private international law instruments.
But on the facts, Mr Kolassa did not have a contract with his proposed Defendant, Barclays Bank. He had invested in securities indirectly, through an intermediary bank in Austria, where he was domiciled. His contract was with that bank, and not with the Barclays, the ultimate issuer. Thus, there was no consumer contract (for the purposes of Article 15 Regulation 44/2001), and for that matter neither was there a contract within the extended meaning of that term in Article 5(1).
What then to do? There is nowhere else to go than to the tort/delict head in Article 5(3). However, the Court’s justification for allowing Mr Kolassa to rely on Article 5(3) is somewhat strained.
Advocate-General Szpunar, in his Opinion of 3 September 2014, seems to have proceeded on the basis that Mr Kolassa could sue in Austria, because his complaint was about deficiencies in Barclays’ Prospectus, and the Prospectus had been published in Austria. One can understand the logic of that approach. If the gist of the complaint made by a Claimant is that he was induced to enter into an investment because of incomplete or misleading information in a Prospectus published in his home jurisdiction, then one can see the justification for concluding that that is “the place where the harmful event occurred”: it is where the decision to invest is made.
But the CJEU does not appear to adopt this this analysis. Perhaps there were insufficient facts in the record to enable it to do so (see Judgment at para. , which suggests there was no information in the case-file to show that the Prospectus was in fact distributed “anywhere other than the Member State in which Barclays Bank has its seat”).
Instead, obscurely, at para. , a rather more general proposition seems to be advanced, namely that: “The courts where the applicant is domiciled have jurisdiction”.
Of course under the Brussels Regulation regime, it is not the applicant’s place of domicile, but that of the Defendant, which is relevant. So this is a rather meaningless statement looked at in isolation. Outside the consumer protection context, the courts for the place of the Claimant’s domicile cannot have jurisdiction simply because they are the courts for the place of the Claimant’s domicile. There has to be some other connecting factor linking the case to that jurisdiction, which justifies displacement of the usual rule that the Claimant must sue in the place of the Defendant’s domicile. What that reason is in the Kolassa case, if it is not reliance on the Prospectus published there, is difficult to discern.
The Judgment at para.  hints that Austria was the justified as an available jurisdiction because the Claimant’s loss “occurred itself directly in [the Claimant’s] bank account held with a bank established within the area of jurisdiction of those courts”. But again, this seems a rather surprising statement, given the Court’s earlier pronouncement on the meaning of loss in Kronhofer: in a sense, a Claimant will almost always sustain financial loss in the place of his domicile, and so an approach to jurisdiction which looks at damage in those terms will subvert the overall approach of the Regulation which gives prominence to the domicile of the Defendant.
It is suggested this is all rather unsatisfactory. It is difficult to know what really underlies the CJEU’s reasoning; but as noted, one plausible explanation is that it was a desire to afford to the Claimant, who was undoubtedly a consumer, the same option of suing in the place of his domicile that he would have had if he had had a contract with the Defendant. The problem is, he didn’t have one. In that sense, the case exposes a gap or lacuna in the consumer protection regime under the Regulation. That may be so, and the gap is no doubt one which deserves to be plugged. The trouble is that the way the Court has gone about it involves a distortion of the Article 5(3) test and reasoning which is obscure and difficult to understand.
Perhaps legislative intervention is required to deal with those in the position of Mr Kolassa. In the meantime, we should remember that outside the consumer context, Issuers should still be able to include provisions on jurisdiction (including exclusive jurisdiction) and on choice of law (including in relation to non-contractual claims) which will be effective as against wholesale or institutional investors in their securities, at any rate if they are contractual counterparties.