Confidentiality vs. Transparency In Commercial Arbitration: A False Contradiction To Overcome

1. Introduction: the general context of confidentiality in commercial arbitration

As it has been stated by the French newspaper “Le Figaro” on September 9, 2008, “the custom is not to say who arbitrated what”. Confidentiality is considered to be one of the crucial features of commercial arbitration. Historically, arbitration proceedings – as well as arbitral awards – have been considered fully confidential.

During the revision of the Uncitral Arbitration Rules, the need of an higher level of transparency emerged, more specifically for treaty-based investor-State arbitration[1]. Since then, the debate has been also focused, among scholars, to commercial arbitration.

We know that arbitration is an expression of party autonomy. This autonomy has to be expressed in a contract: the arbitration agreement. Most of the times, parties agree upon the features of the future proceedings in such agreement. Among those features – or “rules of the game”, such as the applicable law, the seat of arbitration, the language of the proceedings – parties may also include a provision to govern confidentiality issues. For this reason, an arbitration clause may contain, among others, a specific provision on confidentiality. No question, then, that an arbitration clause may contain provisions like the following:

“…awards should be treated as confidential and not be communicated to third parties unless all parties [and the arbitrator] consent; or they fall into the public domain as a result of enforcement actions before national courts [or other authorities]; or they must be disclosed in order to comply with a legal requirement imposed on an arbitrating party or to establish or protect such a party’s legal rights against a third party”; or, even with a broader wording, “…no information concerning an arbitration, beyond the names of the parties and the relief requested, may be unilaterally disclosed to third party by any participating party unless it is required to do so by law or by a competent regulatory body, and then only: (i) by disclosing no more than what is legally required, and (ii) furnishing to the arbitrator details of the disclosure and an explanation of the reason for it[2]”.

The IBA Guidelines for Drafting International Arbitration Clauses[3] – adopted in 2010 – at paragraphs 60-65 expressly suggest that the parties, if concerned about confidentiality, should address this issue in their arbitration clause.

But how often do we see arbitration clauses dealing with confidentiality issues? I would say, very rarely. Nevertheless, confidentiality is considered to be a feature of all arbitration proceedings, as part of the original arbitration clause or as a “natural” consequence of such agreement, no matter whether a confidentiality clause has been included or not in the contract.

The survey conducted by Queen’s Mary College[4] expressly shows that 50% of the corporations interviewed considers arbitration confidential even where there is no specific clause to that effect in the arbitration rules adopted or in the arbitration agreement.

But the same survey reports also that 30% of those interviewed believe that, in the absence of an express agreement of the parties, arbitration is not confidential.

Accordingly, legislation, arbitration rules, Court decisions, international treaties either did not address the issue or did not precisely define confidentiality[5].

Some countries and arbitral institutions incorporate confidentiality provisions into their national laws and sets of rules, some others chose not to.

As far as arbitration rules of the major institutions are concerned, not all of them incorporate confidentiality provisions, although the majority provides for some degree of confidentiality. However, even in the case that these rules expressly lay down confidentiality obligations, often they relate only to some issues to which the duty of confidentiality can pertain.

Case law itself offers different solutions, to the point that a coherent doctrine is still a mirage. It was only in the 1990s, when some Courts decisions rejected the idea that arbitration is per se confidential[6], that the above mentioned assumption was questioned.

Also scholars are divided between an inherent nature of confidentiality on one side, and a legal concept which is relatively limited[7] on the other side. For the latter, arbitration does not have a confidential nature per definition but it may be confidential if the parties so wish and expressly agree (directly, in an ad hoc proceedings, or by reference to a set of Rules, containing a provision on confidentiality, in administered arbitration).

For all these reasons, we cannot reasonably affirm that confidentiality is a duty strictly and automatically related to the parties’ choice to arbitrate. The different sources just mentioned show that there is no general and absolute recognition of the duty of confidentiality in arbitration.

Consequently, in the present scenario, confidentiality obligations vary substantially depending on the arbitration agreement, in which the parties may have addressed the issue, the lex loci arbitrii and the applicable rules of arbitration. Furthermore, even when these sources provide to some extent a confidentiality obligation, questions on the scope, limits and enforceability are still far from being settled.

Exceptions include the possibility to disclose confidential information in specific circumstances, such as in proceedings to enforce or to set aside an arbitral award, to establish a party’s legal right, to comply with a compulsory order or request of a governmental or regulatory body or with law requirements.

Personally, I believe that this debate – and the related “dualistic” view – has very limited significance. Rather than investigate the legal background (if any) of confidentiality in commercial arbitration, we should pay much attention to understand, on one side, whether the interest of the parties to confidentiality is a real interest and, on the other side, whether transparency is a real need. Furthermore, we should wonder how and to which extent other players in arbitration, such as arbitral institutions, may contribute to find a balance between the two (I believe, only apparently) opposing interests: confidentiality on one side, transparency on the other.

As transparency comes across confidentiality, “the conflict between transparency and confidentiality cannot permit the victory of one on the other, and their settlement turns out necessary[8]. Let’s see how such balance can be achieved.

The present contribution – focused on commercial arbitration only, investment arbitration excluded for obvious reasons – does not mean to offer an exhaustive analysis of confidentiality versus transparency. The topics are very complicated and not easy to be covered in few pages. My aim is to raise some ideas on which, hopefully, a more extensive debate will follow.

2. The interests of the parties: confidentiality vs. transparency

2.1. Confidentiality: is it a real interest?

A vast majority of scholars, when listing the main advantages of arbitration for the parties, includes confidentiality. 

It is worth just to be mentioned that privacy and confidentiality – important and interrelated features of international commercial arbitration – are different concepts. While the privacy of the proceedings precludes any stranger from attending it, confidentiality is concerned with the obligation not to disclose information relating to the content of the arbitration. Here we will deal with the latter only. 

But is confidentiality really one of the main reasons for a company to opt for arbitration? No question that, in some very specific situations, confidentiality may play a crucial role in the parties’ choice. For example, in intellectual property agreements or when business information and trade secrets are involved[9].

In other circumstances, parties may not wish to make public allegations of bad faith, incompetence or lack of adequate financial resources.

The above mentioned Queen’s Mary College survey shows that 62% of corporate counsel interviewed considers confidentiality not the essential reason for recourse to arbitration, although “very important”. Several reasons could explain this.

First of all, corporations are often obliged to report to shareholders, and to disclose their annual accounts which might include information that cuts across confidentiality. Indeed, the duty of confidentiality disappears once there is an obligation to reveal information.

Furthermore, many commercial arbitration matters do not involve sensitive commercial information, making confidentiality not such a serious concern[10].

Concretely speaking, it is hard to find a real interest of the parties – i.e. companies and/or individuals involved, not their counsel or the arbitrators – not to disclose information about the existence of a dispute and of an arbitration proceedings, as well as about the way such a dispute has been decided; not even for the losing party, whose alleged strong interest to confidentiality vanishes quickly when it decides to challenge, on a serious or specious ground, the award before a State Court (or, before the award, it challenges an arbitrator or seeks an interim measure of protection by a State Court).

It is rarely the case that a company’s image may be damaged because of its involvement in a dispute (nowadays, more and more a physiological – rather than pathological – event of any complex business relation). In the vast majority of situations, such involvement would not cause any serious business loss to the parties.

In conclusion, it seems to me that there is no real basis in stating that most of the parties choose arbitration because of its confidential nature. Whenever such interest is essential for the parties, they can expressly state the confidentiality requirement in the arbitration agreement, exactly as they do for the other crucial “rules of the game”.

If parties do not do it, the broader interest of the business and legal community in transparency may prevail.

2.2. Transparency: is it a real need?

But which are the reasons calling for transparency in commercial arbitration? In which cases and why should confidentiality give way to transparency? Is transparency a real need?

The reasons for transparency – or the downsides of confidentiality – are several.

First of all, although arbitration is binding only for the parties who are signatories, “quality” awards can have persuasive value for future parties and arbitrators. The access to an highly competent and specific “case law” may lead to many advantages for arbitration practitioners and, in general, for arbitration.

There are also reasons of predictability and consistency. Business people do not like uncertainty and unpredictability. More visible proceedings and transparent awards would guarantee an higher level of consistency and predictability, which, in turn, would enhance the legitimacy of the process itself, with the parties having a greater understanding of it. If they are more satisfied, and because of transparency, have the perception that the process is fair, they are likely to use arbitration again.

Obviously, the publication of awards would make people learn of mistakes and misbehaviors of others, avoiding future disputes.

Furthermore, confidentiality of arbitral decisions may lead to inconsistent resolution of disputes arising out of the same business transaction but decided by different arbitral tribunals. This carries the risk of conflicting awards. In these circumstances, more transparency is desirable, especially for the stakeholders in order to benefit of documents and information relevant to each of the disputes.

The access to arbitral awards may also contribute to the education and training of future and less experienced arbitrators. The non-publication of arbitral awards leaves parties, arbitrators and judges without guidelines in legally and factually similar cases. Again, the efficiency of the proceedings would benefit from the public availability of arbitral awards. 

Moreover, transparency may help users to control and evaluate the quality of service provided by individual arbitrators and arbitral institutions. Reading arbitral awards, future arbitrating parties will be able to assess how a particular arbitrator has handled past similar proceedings, whether that individual would be appropriate to be appointed in a similar future arbitration, his/her level of specific technical skills, how an arbitral center has fulfilled its duties etc. 

Not rare are claims for greater transparency by the users of arbitration. The situation leads to a paradox: indeed, parties want confidentiality but at the same time, they search for predictability while choosing the right person who will have the power to render a decision on their dispute.

There is a lack of information which transparency can easily help to correct, without going against confidentiality. The present contribution is not devoted to support transparency against confidentiality, but it intends to “simply” show how arbitration may be more accessible and more transparent without any harm to the parties’ interests.

3. The interests of the other players

3.1 The arbitrators

As I just underlined, transparency would allow practitioners to check the quality of arbitrators, as to the conduct of the proceedings and their awards. Such information may play a crucial role in future cases for the selection of the “ideal” arbitrator.

The same survey of Queen’s Mary College points out that arbitrators’ skills and experience, knowledge of the applicable law and reputation are the major features taken into consideration by the parties in the selection process.

Therefore, transparency may contribute to an higher level of awareness in the appointment of arbitrators, and for the arbitrators, a sort of “publicity” of their decision making abilities.

It is understandable that parties want to have information, as detailed as possible, before appointing arbitrators. This includes not only the information contained in the candidate’s curriculum vitae but also information about his/her performance as arbitrator.

High visibility of their performance should presumably suggest that arbitrators would take greater care in conducting the arbitration proceedings and in drafting the final award. The quality of their work, under the control of the public opinion, should be higher.

Therefore, (good…) arbitrators should also be in favor of transparency in commercial arbitration.

3.2 The scholars

Arbitration has always been presented as a little community in which everyone knows each other. Nowadays, with the development of arbitration throughout the world, with more than 140 States part of the New York Convention, it is undoubtedly true that practitioners are more and more numerous and the related practice more sophisticated.

Through transparency, scholars can have easier access to information related to the arbitral process, contributing to the study of arbitration and to its development.

3.3 The Institutions

Parties using arbitration can, on one hand, opt for an ad hoc proceedings in which they are free to determine the rules of the process. On the other one, they can refer to an institution, its Rules and its administration activity.  

Arbitral awards rendered under the auspices of a center can contribute to the promotion of arbitration itself and to the creation of a wider arbitration culture.

But the way institutions deal with the issue of confidentiality shows an heterogeneous context of the different regimes adopted. For instance, the Rules of the London Court of International Arbitration (art. 30.1) provide that “unless the parties expressly agree in writing to the contrary, the parties undertake as a general principle to keep confidential all awards in their arbitration…”.

The Milan Chamber of Arbitration-CAM in its 2010 Arbitration Rules, expressly provides the existence of a duty of confidentiality in arbitration, including the parties among the entities bound by the scope[11]. But it also considers, for purposes of research, the possibility to publish the content of arbitral awards.

Despite the different approach followed by the major arbitral centers, I strongly believe that the first player involved in surmounting the false debate of confidentiality vs. transparency is the Institution.

As most of the cases are administered, it is an obligation for the Institution to point out the conditions of a good balance between them. Let’s see how and to what extent.

4. Which role can an Institution play among those different interests?

Transparency does not mean that all information related to a specific proceedings should be disclosed to everyone.

The information may be related to the proceedings (to the hearings and to the documents produced), to the status of the arbitrators and to the final arbitral decision.

In administered arbitration, the institution can play – because of its neutrality – a very important and unique role in balancing the right of the parties to keep some information confidential, on one side, and the interests of the various players (users, practitioners, scholars, arbitrators and institutions) to have access to information related to arbitration proceedings, on the other. Thanks to the institution, those interests can be equally taken into consideration and the integrity of the proceedings preserved. 

4.1 Decisions on the arbitral proceedings

How confidential are arbitral proceedings? This is a highly controversial and difficult issue.

On the one hand, and mostly in intellectual property disputes, confidentiality of certain documents is sometimes of crucial importance. The situation does arise (and, e.g., has arisen in the IBM/Fujitsu arbitration[12]) that a party may wish to rely on documents which should not be seen by the other party (which may be its main competitor in this specific field).

Unless there is a specific provision to this effect within the arbitration clause – or in the Arbitration Rules applicable, according to the arbitration agreement – it will be almost impossible, once a dispute has arisen, to agree on a procedure whereby such confidential documents (or further confidential information) can be validly introduced and considered by the arbitrators without making such documents or information directly available to the other party (since the other party will say that its right to be heard will not be satisfied unless it has been able to have full and unrestricted access to any such documents or information and to comment thereon accordingly).

However, within the arbitration clause it would be possible and appropriate to contemplate and specifically agree that such confidential documents and information (which will have to be defined carefully) will only be made known and fully disclosed to the arbitrators (but not to the other party), or will be disclosed and made available to a neutral third party (such as an expert, or auditing firm), which would then issue a certification, or a report, or an assessment for the arbitral tribunal.

The following various wordings[13] – as well as the ones suggested at paragraph 1 – may be proposed for inclusion in the arbitration clause or in an arbitration agreement:

1. Any documentary or other evidence given by a party or a witness in the arbitration shall be treated as confidential, and shall not be disclosed, by any party whose access to such evidence arises exclusively as a result of its participation in the arbitration, to any third party for any purpose without the consent of all parties or order of a court [or arbitral tribunal] having jurisdiction. (For the purpose of this rule, a witness called by a party shall not be considered a third party. To the extent that a witness is given access to evidence obtained in the arbitration in order to prepare his testimony, the party calling such a witness shall be responsible for his maintaining the same degree of confidentiality as that required of the party).

2. [To the extent that they describe or refer to evidence] written pleadings shall not be disclosed to third parties for any purpose save as stated in 1 above.

3. An arbitrator, when issuing an order for the production of documentary or other evidence, may in his discretion make such order conditional upon the other party or parties’ specific written undertaking not to disclose any of the evidence (or details of it) to third parties”.

Any person serving as arbitrator, or expert appointed by an arbitral tribunal, or appearing as the representative of a party in an arbitration, thereby undertakes on his own behalf mutatis mutandis to respect the rules of confidentiality defined in Articles, etc“.

But, again, very rarely do we see such detailed agreements on confidentiality. And the majority of  Arbitration Rules do not cover – so precisely – all the above mentioned aspects related to the confidentiality of the proceedings.     

4.2 Decisions on arbitrators’ independence and impartiality

Something we are certain about is that challenges to arbitrators are numerous. The more arbitration is used, the more conflicts of interest exist. The situations in which arbitrators, counsel or parties, have had relationships between them, have increased in the last few years.

The IBA Guidelines on Conflict of Interests are useful reference for parties, counsel and arbitrators in order to identify specific recurrent cases in which these actors can be implied. The goal of these Guidelines is to propose common standards in order to identify possible circumstances of partiality and/or dependence of those relationships.

Arbitral institutions pay much attention to this issue. Many arbitral centers promote various initiatives in order to better regulate conflicts of interest. Some of the major arbitral institutions (like ICC, LCIA, CAM etc) provide that each arbitrator must disclose to the Institution all the information concerning past or present relationship with the proceedings and/or the players involved in the case. In those cases, the institution – through either a specific committee or the technical body in charge with the supervision of the proceedings – decides whether the appointment is confirmed or not and whether a challenge is grounded or not.

Having access to the decisions on arbitrators’ independence can be of fundamental importance for the parties and for the arbitrators themselves, reducing the risk of “unsuccessful” appointments and, therefore, of additional costs and waste of time.

This is the reason why the LCIA has published, through a Special Edition of Arbitration International, a Digest of its decisions on arbitrators’ challenges, and the Milan Chamber of Arbitration, although with a different approach, has decided to launch in 2013 a similar project.

4.3 Arbitral awards

Usually, transparency in commercial arbitration is mostly focused on the publication of sanitized arbitral awards. For instance, the ICC Secretariat publicizes synthesis of awards in the ICC International Court of Arbitration Bulletin for educational purposes. In this publication reference is made only to the docket number and the award is sanitized by removing the names of the parties, geographical and industrial facts that would risk to render the case and its participants identifiable.

More recently, the Milan Chamber of Arbitration has adopted a set of guidelines for the anonymous publication of arbitral awards. Their purpose is clearly embodied in paragraph (1) of the Preamble, which reads: “The Guidelines aim to provide a set of common and uniformly applicable standards in order to publish arbitral awards and provisions anonymously and confidentially…”, especially when the parties have not expressly and directly agreed on confidentiality issues”[14].

Generally speaking, as we said an higher transparency – and, consequently, wider predictability – would also represent a crucial step forward in the promotion of arbitration in the business community. Arbitration, although having a contractual nature, is a system for rendering justice. Arbitration plays a sort of “social” role, having a social impact. For this reason, we must render commercial arbitration – starting from the awards – more accessible, more transparent. We cannot see arbitration as a purely private phenomenon. But how can we reach all these advantages, combining the general interest to transparency with the parties’ interest to confidentiality?

First of all, a good “sanitization” of the arbitral award (an intervention to make it entirely anonymous, being impossible for anyone to understand the identity of the parties involved in the case) would render such interest (if any) real.  

The “sanitization” of arbitral awards can be better guaranteed in an administered arbitration, where the Institution – thanks to its Rules – can take into equal consideration, on one hand, parties’ interest to confidentiality and, on the other hand, the wider interest of potential users to access information about arbitration practice and arbitral decisions.

In order to ensure that parties would not be easily recognizable, an efficient treatment – that the Institution has the duty to guarantee – is essential.

For this purpose, many Arbitral Institutions specifically provide for a discipline of such treatment in their Rules. It must be pointed out that those provisions are extremely useful: they give the parties the certainty that during all the proceedings, their needs would be protected. The Institution is the first actor in arbitration to have the obligation to insure the maximum privacy of the whole proceedings and its integrity.

As regards to the Milan Arbitration Rules, they provides also the publicity of awards for purposes of research (see above, note 11) and, of course, any additional publicity the parties may wish.

Among all the information related to an arbitral proceedings, awards are surely the most important ones. But the general interest for transparency is definitely not to show the whole world which parties were involved in the arbitration and why. The goal of transparency is not to disclose everything but mostly to promote research and at the same time to improve the quality of arbitrations in general.

In institutional arbitration, the center has a general duty of constantly building case law, and in so doing, it also provides information on performances of arbitrators.

But, of course, such case law has to be carefully built. If publicity is made properly – that is to say, without any evidence for recognizing what has to be hidden – there would not be any problem with transparency. Transparency is not criticized per definition, as a principle, but for what it could lead to if publicity has been made incorrectly.

Therefore, quality has to be met not only by the arbitrators in the award but also by the arbitral institution in publishing such awards. Publication that has not to harm parties’ rights.

As we said, the solution to prevent such risks is a good “sanitization” of the arbitral award. This technique consists of cleaning the entire text by selecting only the elements which have a general interest for arbitration users and scholars, avoiding the disclosure of any aspects irrelevant for those purposes and able to identify the identity of the parties. CAM’s Guidelines are very detailed on these techniques.  

5. Conclusion: a balance between confidentiality and transparency in arbitration

As we have seen, apart the common understanding that the arbitral process is inherently private, there is no general consensus as to its confidential nature. The rules not only differ significantly amongst jurisdictions and arbitral institutions, but very often they are also unclear and not exhaustive on their scope and extent.

The present situation leads to uncertainties even on some fundamental issues. Accordingly, it is difficult to identify a solid legal entitlement to preserve confidentiality, as well as to delimitate its subjective and substantive reach. And the absence of a coherent judicial support contributes to such uncertainties.

Very often, even the identification of the relevant applicable provisions is not easy, due to the different principles of conflicts of laws and jurisdiction.

Furthermore, the applicable law of many countries may be inconsistent with each other. Consequently, an individual bound by an obligation of confidentiality may also be subject to an opposing obligation to disclose the very same information.

As a result, it is impossible to draw any general conclusion to establish the existence or non-existence of a confidentiality obligation in international commercial arbitration.

Given the different approaches to confidentiality and the absence of universally recognized standards, the best way to safely guarantee confidentiality is to sign (either prior to or during the proceedings) specific clauses in which parties should specify the scope, the extent, the duration of the confidentiality obligation, its exceptions and how it may be enforced. Agreements that most legal systems do recognize and enforce.

But they are clauses that, as we know, are very rarely included in the contract. Should we then, in the absence of such agreements, give up any possible use of arbitration to meet the above mentioned expectations and interests? Because of the uncertainties surrounding the issue of confidentiality, should we abandon the possibility to create a specialized case law, to educate future arbitrators, to contribute to an higher level of predictability, to provide useful information about arbitrators and institutions’ performances? I would say no.

No doubt the interests of the parties must be protected, as well as the integrity of the process. But I strongly believe that such interests (whether specific or more general) are not frustrated by an higher level of transparency, if properly governed by an arbitral institution, under its control and guidance, thanks to its neutrality, its competence, its professionalism. 

Let’s briefly return to the Queen’s Mary College survey, where it emerged that corporations select arbitral centers because of their internationalism, their neutrality and their reputation in the market.

If the reputation of a given arbitral center is strong, parties may be sure that, by selecting its rules and relying on its administration activity, the proceedings will be confidential and the use (if any) of the award will not harm their interests and rights. This applies even more whether the institution has defined – as the Milan Center has – a set of guidelines to be followed in the “sanitization” of its awards.

It is not whether confidentiality is better than transparency or transparency is more important than confidentiality: it is just a question of balance between two different (but not necessarily, opposing) interests.

Balance that can be found through the important role of an arbitral institution.

Stefano Azzali
The author, presently Fellow at the NYU Centre for Transnational Litigation and Commercial Law, is the Secretary General of the Milan Chamber of Arbitration and acts as Secretary Treasurer of the International Federation of Commercial Arbitration Institutions (IFCAI). Since 2005, he is Visiting Professor of Arbitration Law at Bocconi University School of Law in Milan and, from 2001 to 2007, he chaired the Disciplinary Commission of the Italian Football Federation (FIGC), where he is now member of its Federal Court of Justice.


[1] Official records of the General Assembly, 63rd Session, Supplement No. 17 (A/63/17)

[2] Paulsson, J. and Rawding, N., “The Trouble with Confidentiality”, Arbitration International, 1995, 3, 315-319

[3] http://www.ibanet.org/Document/Default.aspx?DocumentUid=D94438EB-2ED5-4CEA-9722-7A0C9281F2F2

[4] “2010 International Arbitration Survey: Choices in International Arbitration”, http://www.arbitrationonline.org/research/2010/index.html

[5] For a comparative study, see Noussia K., “Confidentiality in International Commercial Arbitration. A Comparative Analysis of the Position under English, US, German and French Law”, Springer, Heidelberg, 2010. See also ILO Report on “Confidentiality in International Commercial Arbitration”, October 2010, 5-10

[6] Australian High Court, Esso Australia Resources Limited v Plowman (1995) 183 CLR 10 and Swedish Supreme Court, Bulgarian Foreign Trade Bank Ltd v Al Trade Finance Inc N T 1881-99, judgment 27 October 2000 (“Bulbank”)

[7] “Expert Report of Stewart Boyd QC” (in Esso/BHP v. Plowman), Arbitration International, Kluwer Law International, 1995, 3, 266-268

[8] Fages F., “La confidentialité de l’arbitrage à l’épreuve de la transparence financière”, Revue de l’Arbitrage, Comité Français de l’Arbitrage, 2003, 1, 5-39 [The translation is not official]

[9] Dessemontet F., “Arbitration and Confidentiality”, The American Review of International Arbitration, 2001, 7:1, 299

[10] Queen’s Mary College Report, 29

[11]Art.8: 1. The Chamber of Arbitration, the parties, the Arbitral Tribunal and the expert witnesses shall keep the proceedings and the arbitral award confidential, except in the case it has to be used to protect one’s rights. 2. For purposes of research, the Chamber of Arbitration may publish the arbitral award in anonymous format, unless, during the proceedings, any of the parties objects to publication

[12] Smit H., “Case-note on Esso/BHP v. Plowman (Supreme Court of Victoria)”, Arbitration International, Kluwer Law International, 1995, 3, 299 – 302

[13] Paulsson, J. and Rawding, N., “The Trouble with Confidentiality”, Arbitration International, 1995, 3, 315 and 318

[14] The Guidelines are available at the CAM website (http://www.camera-arbtrale.it/Documenti/guidelines_anonym_aw.pdf). A commented edition will be soon published by Juris Publication.

The Systemic Integration of International Investment Treaties and the New York Convention

A.            Introduction

There have recently been various cases where investors successfully asserted a violation of an international investment treaty on the grounds that the host State failed to recognize and enforce a commercial arbitral award as foreseen in the New York Convention on the Recognition and Enforcement of Foreign Arbitral Awards of 1958 (the “New York Convention”).

Effectively, investors have thus been able to recover damages for the non‑compliance with obligations under the New York Convention within the framework of investment arbitration. In the following, it will be set out that this is due to a shift towards more systemic integration: instead of applying international investment law in clinical isolation, arbitral tribunals are increasingly willing to take into consideration other sources of international law such as the New York Convention.

Importantly, however, this development has not led to a defragmentation[1] of international law: international investment law remains a separate regime, which cannot be used as a mere vehicle to enforce obligations beyond investment law. For this reason, the possibility to obtain damages for non-compliance with obligations under the New York Convention within the framework of investment arbitration only exists under limited circumstances.

B.            Procedural challenge

Investors who want to assert their rights before an ICSID tribunal face the procedural challenge that ICSID is not a forum where States can generally be held responsible for the non-compliance with their obligations under the New York Convention. Instead, its jurisdiction is limited to legal disputes arising “directly out of an investment”.[2]

This jurisdictional requirement can, however, be met, if the transaction underlying the commercial award qualifies as an investment. To put it differently: an investor may initiate ICSID arbitration, if the host State fails to recognize and enforce a commercial arbitral award, which resulted from a transaction that qualifies as an investment.

The main precedent for this view was established by the arbitral tribunal in Saipem S.p.A. v. The People’s Republic of Bangladesh.[3] In this case, the arbitral tribunal had to decide whether a dispute arising from the non-enforcement of an ICC award fell within its jurisdiction. The arbitral tribunal confirmed this. It held that the “entire operation” would have to be considered in order to determine whether there is an investment under Article 25 of the ICSID Convention.[4] Given that the ICC award crystallized rights, which had arisen under a pipeline construction contract, an investment under Article 25 ICSID Convention would be given. The arbitral tribunal left open whether the ICC award as such qualified as an investment.

A similar approach was taken by the arbitral tribunal in ATA Construction, Industrial and Trading Company v. the Hashemite Kingdom of Jordan.[5] In this case, the arbitral tribunal was confronted with the question whether an arbitral award, which had resulted from a dispute concerning a dike construction contract, qualified as an investment. The arbitral tribunal confirmed this on the grounds that the “entire operation” of which the arbitral award formed part, i.e., the construction of the dike, qualified as an investment.[6]

The only case, where an ICSID tribunal refused to consider whether the underlying transaction, from which the arbitral award had arisen, qualified as an investment, was GEA Aktiengesellschaft v. Ukraine.[7] Here, the arbitral tribunal argued that a sharp analytical distinction would have to be maintained between the commercial arbitral award and the transaction from which it arose.[8] Given that the award itself involved none of the elements of an investment, such as a contribution to or relevant economic activity within Ukraine, the arbitral tribunal denied its jurisdiction.

While the arbitral tribunal was correct in its characterization that the commercial award did not – in and of itself –constitute an investment, there is little support for its proposition that a sharp analytical distinction has to be maintained between the commercial award and the underlying transaction. The practical consequence of this approach would be that parts of an investment (here,: the commercial award) could be denied protection simply by assessing them out of context (here, the underlying transaction). This is hardly desirable.

The jurisprudence of non-ICSID tribunals also confirms that arbitral awards cannot be assessed without taking into account the underlying transaction. As an example, one might consider the decision in White Industries Australia Limited v. The Republic of India.[9] In this case, the question arose whether an ICC award, which had been rendered following a dispute under a contract for the supply of equipment and development of a coalmine, constituted an investment within the meaning of the Australia-India BIT. The arbitral tribunal confirmed this and endorsed the claimant’s view that the award, while not being an investment in itself, was part of the original investment.[10] Even if one takes for granted that the notion of investment under Article 25 ICSID Convention cannot be equated with the notion of investment pursuant to the respective BIT, this decision lends support for the proposition that tribunals have to consider all relevant circumstances in taking their decisions, instead of considering parts of an investment in isolation.

The decisions of the arbitral tribunals in Frontier Petroleum Services vs. the Czech Republic[11] and in Romak S.A. vs. The Republic of Uzbekistan allow for the same conclusion.[12] While the arbitral tribunal in Romak S.A. vs. The Republic of Uzbekistan denied its jurisdiction to hear a dispute arising from the non-enforcement of a GATFA Award, it did so on the grounds that the underlying transaction was a wheat supply transaction and thus not an investment within the BIT. Again, this confirms that arbitral tribunal may have to look at the contractual relationship from which the commercial arbitral award arose in ruling upon its jurisdiction.

C.            Substantive challenges

Even if ICSID tribunals have jurisdiction, they are not entitled to award damages based on a mere finding that the New York Convention has been violated. Instead, they have to assess, whether the non-enforcement of the commercial award triggers responsibility under the respective investment treaty. This is not to suggest that the New York Convention would be irrelevant. To the contrary, it is part of the normative environment, which will have to be taken into consideration by arbitral tribunals.

1.             The New York Convention is relevant for the assessment of whether a treaty standard has been violated

Above all, the New York Convention provides interpretative guidance for the assessment of whether one of the treaty standards has been violated. In Saipem S.p.A. v. The People’s Republic of Bangladesh, for example, the arbitral tribunal had to assess whether Bangladesh had violated the protection against unlawful expropriation by depriving Saipem of the benefits under an ICC award. The arbitral tribunal confirmed this. It underlined that Bangladesh had acted unlawfully by abusing its rights and violating its obligations under the New York Convention.[13] The arbitral tribunal thus had recourse to the New York Convention in order to assess the lawfulness of the deprivation of benefits under the commercial award. Conceptually, one might designate this as a form of systemic integration as foreseen by Article 31 (3) (c) Vienna Convention.[14]

In ATA Construction, Industrial and Trading Company v. The Hashemite Kingdom of Jordan took a similar approach. The arbitral tribunal based its finding that Jordan had “violated both the letter and the spirit of the Turkey-Jordan BIT”[15] on the fact that Jordan had unlawfully extinguished Claimant’s right to arbitration contrary to Article II New York Convention.[16] While the arbitral tribunal refrained from specifying in greater detail which guarantee of the Turkey-Jordan BIT was violated, it mentioned that Jordan had assumed the obligation to accord Respondent’s investment fair and equitable treatment as well as treatment no less favorable than that required by international law.[17]

Likewise, the arbitral tribunal in Frontier Petroleum Services vs. the Czech Republic took into consideration the New York Convention in assessing whether the host State had violated the fair and equitable treatment standard. The arbitral tribunal explicitly confirmed that it had the power to review the decision of a national court’s conception of the public policy exception under the New York Convention and rejected Respondent’s allegation to the contrary.[18] However, it only examined whether the Czech courts had applied a plausible interpretation of the New York Convention, i.e., an interpretation that was tenable and made in good faith.[19] This deference to the decision of the State courts seems reasonable in view of the fact that the fair and equitable treatment standard is only a yardstick for certain minimum treatment. Not every form of illegality triggers responsibility under this standard. In the case at hand, the arbitral tribunal considered that the Czech courts’ interpretation of the New York Convention was not unreasonable or impossible.[20] Accordingly, the Czech courts had not acted arbitrarily, discriminatorily, or in bad faith so that no breach of the fair and equitable treatment standard was given.[21]

2.             The New York Convention is relevant for the assessment of the damages

The New York Convention is not only relevant when determining whether a BIT has been breached. It also has to be considered when assessing the quantum of damages flowing from such breach. Arbitral tribunals even have to enter into a more in-depth examination of the New York Convention in order to assess the damages.

The decision in White Industries v. The Republic of India is highly instructive in this regard. In that decision, the arbitral tribunal found that India had violated its obligation to provide for effective means of asserting claims and enforcing rights by delaying the decision on enforceability of an arbitral award over a period of nine years.[22] In order to determine the damages flowing from this violation, the arbitral tribunal determined whether the arbitral award was enforceable in India. In doing so, it carefully examined whether there was a ground for refusing the recognition and enforcement of the award under the New York Convention. It concluded that the award was enforceable, since no such ground was given.[23] In the view of the arbitral tribunal, White Industries was therefore entitled to full compensation for the loss it had suffered.

Interestingly, the arbitral tribunal seems to have felt a certain unease to enter into such full-fledged examination of grounds for refusing a declaration of enforceability under the New York Convention. It therefore explicitly asked the parties for a mandate to do so. Both parties agreed that the tribunal had been provided with sufficient material and that it should engage in a determination of the enforceability of the award in India.[24]

One might wonder whether such mandate was necessary in the case at hand? While it is true that the New York Convention leaves it up to the domestic State courts to assess the enforceability of commercial arbitral awards, it is important to note the decision of the arbitral tribunal only resulted in an award to pay damages. The ICSID tribunal did not order the execution of the arbitral award as such. Against this background, it seems reasonable to conclude that the arbitral tribunal only acted out of precaution and in order to respect to the parties’ right to be heard.

D.           Conclusion

International investment law stands in systemic relation with other sources of international law. As can be concluded from the above-mentioned jurisprudence, arbitral tribunals are increasingly willing to apply sources beyond international investment law such as the New York Convention. Importantly, however, such systemic integration only occurs, provided that the jurisdiction of the respective tribunal is given. Investors who seek damages for the non-enforcement of a commercial arbitral award before an ICSID tribunal therefore have to demonstrate that the underlying transaction, from which this award resulted, constitutes an investment.

Dr. Friedrich Rosenfeld, Hanefeld Rechtsanwälte, Hamburg, Germany.


[1] On fragmentation see Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law, Report of the Study Group of the International Law Commission, Finalized by Martti Koskenniemi, UN Doc. A/CN.4/L.682, 13 April 2006.

[2] Art. 25 ICSID Convention.

[3] Saipem S.p.A. (Claimant) v. The People’s Republic of Bangladesh, ICSID Case No. ARB/05/7, Decision on Jurisdiction and Recommendation on Provisional Measures, 21 March 2007.

[4] Ibid., para. 110.

[5] ATA Construction, Industrial and Trading Company (Claimant) and The Hashemite Kingdom of Jordan (Respondent), ICSID Case No. ARB/08/2, Award, 18 May 2010.

[6] Ibid., paras. 115, 120.

[7] GEA Group Aktiengesellschaft v. Ukraine, ICSID Case No. ARB/08/16, 31 March 2011.

[8] Ibid., para. 162.

[9] White Industries Australia Limited and The Republic of India, UNCITRAL Arbitration in Singapore under the Agreement between the Government of Australia and the Government of the Republic of India on the Promotion and Protection of Investments, Final Award, 30 November 2011.

[10] Ibid., para. 7.6.4.

[11] Frontier Petroleum Services v. The Czech Republic, Final Award, PCA – UNCITRAL Arbitration Rules, 12 November 2010, para. 233.

[12] Romak S.A. (Switzerland) and The Republic of Uzbekistan, PCA Case No. AA280, Award, 26 November 2009.

[13] Saipem S.p.A. (Claimant) v. The People’s Republic of Bangladesh, ICSID Case No. ARB/05/7, Award, 30 June 2009, para. 170.

[14] On systemic integration see C. McLachlan, The Principle of Systemic Integration and Article 31 (3) (c) of the Vienna Convention, 54 ICLQ (2005) 279 (279 ff.). See also Fragmentation of International Law: Difficulties Arising from the Diversification and Expansion of International Law, Report of the Study Group of the International Law Commission, Finalized by Martti Koskenniemi, UN Doc. A/CN.4/L.682, 13 April 2006, paras. 410 ff. and M. Waibel, International Investment Law and Treaty Interpretation, in: R. Hofmann and C. Tams (eds.), International Investment Law and General International Law (Baden Baden, 2011), p. 29 ff.

[15] ATA Construction, Industrial and Trading Company (Claimant) and The Hashemite Kingdom of Jordan (Respondent), ICSID Case No. ARB/08/2, Award, 18 May 2010, para. 125.

[16] Ibid., para. 128 f.

[17] Ibid., para. 125 and footnote 16.

[18] Frontier Petroleum Services v. The Czech Republic, Final Award, PCA – UNCITRAL Arbitration Rules, 12 November 2010, para. 525.

[19] Ibid., para. 527.

[20] Ibid., para. 530.

[21] Ibid., para. 529.

[22] White Industries Australia Limited and The Republic of India, UNCITRAL Arbitration in Singapore under the Agreement between the Government of Australia and the Government of the Republic of India on the Promotion and Protection of Investments, Final Award, 30 November 2011, para. 10.4.19. By contrast, it held that the obligation to provide fair and equitable standard was not violated. In the view of the arbitral tribunal, White Industries could not have the legitimate expectation that India would apply the New York Convention properly (para. 10.3.13). While the delay of the Indian courts would be unsatisfactory, it would not yet have reached the stage of constituting a denial of justice, either (para. 10.4.22). Besides, the arbitral tribunal held that the delay in declaring the award enforceable would not constitute a form of expropriation (12.3.6).

[23] Ibid., para. 14.2.66.

[24] Ibid., para. 14.2.2.