ABOUT TDM
|
Comment
Secrecy has never been a concept used in international commercial arbitration. And even the concept of confidentiality is not assumed anymore to be automatically applicable as it has come under judicial attack in a number of countries, such as in Esso/BHP v.P lowman (1995) 128 A.L.R. 391, High Court of Australia, US v. Panhandle Eastern Corp. (D.Delaware1988) 118 F.R.D. 346, or Bulgarian Foreign Trade Bank Ltd. v. A.L. Trade finance Inc., a October 27, 2000 decision by the Swedish Supreme Court. One should aslo note that the ICC rules do not provide for any general duty of confidentiality. Art. 21 only establishes the confidentiality of ICC hearings.AAA and Vienna only require of arbitrators but not of the parties to observe confidentiality. Confidentiality is therefore from the outset a relative concept in international commercial arbitration.
There is of course something to be said in favour of confidentiality as well as transparency. This also goes for investment disputes. While the potential for amicable solution is one of the elements speaking for confidentiality there are others of a more general nature such as the wish to not make a dispute public at all or to protect business secrets that are also valid. On the other hand it iso bvious that not only do institutions publish more than ever about arbitration cases, law firms and the parties they represent speak more about it, but that the entire arbitration "community" is far better connected, resulting in more informal "sanitized" exchange on cases of interest. Obviously, the balanceb etween confidentiality and transparency seems to tilt slightly more in favour of transparency and the need for determining the right balance appears to be different from case to case. I do not see, however, that in the case of investment disputes the balance should automatically and generally be more in favour of transparency than in other commercial cases.What is true , though, is that that the state party may be subject to its own rules of transparency with regard to parliamentary or ministerial control which impose as such ad egree of transparency not to be adhered to by a private party.
In the end, the issue can be very much one of style. Confidentiality/Transparency is not so much a question of what and how much becomes public but how information generated in arbitration proceedings is being dealt with.
Comment 2:
There are of course degrees of secrecy and confidentiality, depending on the function in question. At the one extreme, the INTERNAL DELIBERATIONS of the tribunal are and should be secret: see e.g. ICSID Arbitration Rule 15. At the other extreme, it is difficult or even inappropriate in some circumstances to keep the arbitral award itself confidential. E.g. if there is to be the possibility of enforcing the award in domestic courts, then it obviously cannot be kept secret. The ICSID Convention (Art. 48(5) and Arbitration Rules prohibit the CENTRE from publishing an award without the consent of the parties; but (a) this has not prevented the Centre from publishing information about the fact of proceedings, and the stage they are at, in outline form; and (b) the Convention and Rules do not expressly deal with the possibility of unilateral publication by one of the parties or a third party. For instance, my friend and erstwhile opponent in the Holiday Inns case, Pierre Lalive, published an account of the case in the British Year Book of IL, without, so far as I am aware, the consent of Morocco.
The non-respect for the confidentiality of awards has certainly increased increased over the years. I recall that when I was given a copy of the Dubai/Sharjah award by one of the arbitrators, it was on the strict condition that, not only was I not to refer publicly to its content, but even to the fact that there been an arbitration. Clearly this was the basis on which the two states had agreed to go to arbitration. Perhaps quaintly, I stuck firmly to my undertaking, but this did not prevent others citing it to the ICJ in the Qatar/Bahrain case nor, indeed, the Court from citing it itself. (The award had in the meantime been published in ILR, presumably by means of unauthorized disclosure.)
It seems to me self-evident that, if the rules of the tribunal or the agreement of the parties provide for secrecy or confidentiality, it is not for a member of the tribunal or its support staff, a party, or a lawyer, to violate that confidentiality or secrecy, be it for reasons of disgruntlement, self-advertisement, desire to enlighten, or whatever. If an outside party comes by this information without giving an undertaking, he/she is not necessarily under any corresponding obligation. Whether domestic law or the rules of the institution concerned SHOULD allow for greater transparency is a separate question, and one which has already been discussed interestingly within this group.
Comment 3:
It is absolutely true. The proper description is "confidentiality". The trend towards maintaining a balance between confidentiality and transparency is also understandable. In weighing the concepts of confidentiality against transparency one may bear in view that much of the reporting done on arbitral proceedings are not primarily made by the general media and with the intention of informing the public (as partly the case may be with major commercial litigation) but rather with the aim of keeping the professional circles informed of the developments. Further, the content of what is reported of arbitral proceedings is of some relevance. One gets the feeling that the intention behind reporting of the arbitral developments is to inform the professional circles, mostly, of the legal developments rather than of detailed facts. That is, the reporting made by the arbitration journals cited should be seen in that light, without, however, denying the transparency effect of their publications.
Comment 4:
I do not think one can address the legitimacy concerns of NGOs, some of which are very well informed indeed, by playing around with words like this.
The point you make that within the existing legal framework, NAFTA ch. 11 parties and investors and tribunals have already gone some distance towards addressing those concerns is an important one, and certainly ch. 11 practice is going beyond the WTO towards transparency.
I have not heard any NGO suggest that elements of confidentiality to protect govt or commercial secrets are illegitimate, where justified by the substance. The question is whether one moves towards transparency as a default rule or more precisely a presumption, rebuttable in appropriate cases. While I am a big supporter of transparency, I think one also has to bear in mind the costs of arbitrations, and the extent to which private parties end up having to pay substantially in money and also in delay for this progress towards transparency, i.e. for what is essentially a "public good". this is even more the case for NGO participation, which--when not limited to amicus briefs submitted in a timely fashion and which are concise--can obviously create burdens on parties to the proceedings and even delay them. While burdens as such are not a knock-down argument against going farther in the participatory direction, it is clear that at present there is not an adequate "due process" code to make sure those burdens are distributed fairly among the relevant actors.
So the issue is one of balancing different values and interests in a legitimate way, and the extent one can do that by moving forward within the existing legal framework, as opposed to creating new or different rules.
Comment 5:
refer to several very interesting comments today.
A recent significant addition to the jurisprudence is given in a UK Privy Council decision No. 93/2001 in the Bermudan case Associated Electric & Gas Insurance Services Limited v European Reinsurance Company of Zurich; judgement delivered 29th January 2003.
The case concerned two arbitrations between the same parties and whether materials from one could be used in another to create issue estoppel, substantially the same specifics as arose in the early-1990s leading English case Dolling-Baker v Merrett; further, in full consideration of Esso v Plowman, the English Court of Appeal reaffirmed, in Ali Shipping v Shipyard Trogir [1999] 1 WLR 314, that there IS an implied duty of confidentiality in an English arbitration, subject to certain exceptions.
That case was extensively relied on before their Lordships in AEGIS but they chose not to refer to it in their judgement since the present case involved the construction of an express confidentiality agreement so hat general statements concerning the privacy of arbitration proceedings and the duty of one party to respect the confidentiality of the other were of less assistance and relevance. However, their Lordships noted that different types of confidentiality attach to different types of document or to documents which have been obtained in different ways. In particular, an award may have to be referred to for accounting purposes or for the purpose of legal proceedings (as AEGIS referred to it for the purposes of the present injunction proceedings) or for the purposes of enforcing the rights which the award confers (as European Re sought to do in the 2nd AEGIS/European Re arbitration).
The Privy Council concluded (inter alia): "Generalisations and the formulation of detailed implied terms are not appropriate."
I will shortly publish (on my website) a full commentary on this important case.
Permit me a postscript regarding Esso v Plowman, having revisited it in detail recently: (i) the case evidently turns on its very particular facts eg the significant "freedom of information" driver (ii) Sir Patrick Neil QC comprehensively demolished the leading judgements in a lecture delivered (and published in "ARBITRATION") shortly after the decision; (iii) as I understand the case, it binds only the State of Victoria and I am advised that, at least in New South Wales, it would not be followed.
Comment 6:
It may also be worthwhile to recall the interests of the various parties in confidentiality. It is not, in my experience, the question of confidentiality that leads international investors to accept arbitration instead of court proceedings. As you are undoubtedly aware, many (if not most) Western investors would be very happy to have their disputes with State parties heard in courts such as the London and New York courts. Those very same investors routinely submit to English or New York court jurisdiction in their financing and M&A documents, as well as many other significant corporate transactions. Those courts, of course, permit free attendance at hearings and public access to all filings of record and evidence presented to the court, except where the court has approved a joint confidentiality stipulation or ordered information to be sealed for good cause shown. In my experience, Western investors agree to arbitration of investor-State disputes principally because they cannot get the host State to agree to the jurisdiction of a Western court and they are themselves unprepared to accept the jurisdiction of the local courts. Consequently, the pressure for confidentiality of arbitration proceedings in an investor-State dispute ordinarily emanates from the host State, not the investor [I acknowledge of course that there are certainly exceptions to this general observation]. With this in mind, it may be sensible to focus on the policies of host States in blocking non-party access to investor-State arbitrations in which they are parties. With respect to US agencies involved in international investment activities, it is noteworthy that US Eximbank has an internal policy against agreeing to confidentiality clauses in documents to which it is a party. While the focus of such clauses is the confidentiality of information delivered under the terms of the relevant contract, such clauses, of course, are often broad enough to block disclosure to third-parties of information filed of record in related arbitration proceedings except as required by law. US Eximbank's policy is based in part on the US Privacy Act and US Freedom of Information Act, which set paramters as to public availability of information held by a US agency. However, US Eximbank's policy extends beyond the strict parameters of these statutory requirements out of concern that an agency's conduct should be open to public scrutiny except to the extent limited by law. Note, however, that US Eximbank is certainly prepared to finance a transaction in which the underlying parties have included a confidentiality clause in their own documents. To my (admittedly limited) knowledge, no other OECD or other export credit or investment insurance agency has a similar objection to confidentiality clauses. I would be interested to hear the experience with other countries from list participants.
Comment 7:
An interesting point that arises, and from the points made on the policy issues surrounding questions of confidentiality in investment agreements, is the origin of the obligation of confidentiality in arbitration in English law. The Ali Shipping -v- Shipyard Trogir case to which Hew refers confirmed that the obligation is to be found, in English law, as an implied term in the arbitration agreement. As such, it arises a function of English law being the law applicable to an arbitration agreement and not as a function of English law as the lex arbitri - as I read it, Ali Shipping would not be relevant to an arbitration with its seat in England but pursuant to an agreement to arbitrate not governed by English law. Potter LJ said in Ali Shipping
"It seems to me that, in holding as a matter of principle that the obligation of confidentiality (whatever its precise limits) arises as an essential corollary of the privacy of arbitration proceedings, the Court is propounding a term which arises 'as the nature of the contract itself implicitly requires ..."
"A clear distinction is to be drawn 'between the search for an implied term necessary to give business efficacy to a particular contract and the search, based on wider considerations, for a term which the law will necessarily imply as a necessary incident of a definable category of contractual relationship".
In international commercial arbitration, I think interesting issues would arise if the English Courts were to be presented with similar issues to Ali Shipping, where the seat was in England but the governing law of the arbitration agreement was not English law (particularly if the arbitration agreement was governed by the law of Victoria). Some have observed that the English Courts would apply Potter LJ's "wider considerations" by way of the exercise of an inherent power in relation to matters of public policy but it may well turn on very specific facts, as many of these cases appear to. I think those problems would be compounded in an investment arbitration, where in interpreting the relevant treaty the "implied term" route taken before now by the English Courts would not be available. As highlighted by Mark and as today's exchange illustrates, the question of whether confidentiality was a "necessary incident of a definable category of ... relationship" would, be much more open to question. It might indeed follow that confidentiality was not a "necessary incident" of the investor/State relationship, including the investor/State arbitration.
One observation (made to me by a very distinguished British arbitrator) comes to mind following Professor Walde's original text: very often the publication or circulation of Awards is by those individuals or parties who have an interest in certain views or philosophies being seen to be accepted in international arbitration. Sometimes papers or articles are written based on a small number of Awards that have come into the public domain - it may be that there are other Awards that have rejected or at least eschewed a particular view or philosophy but have not been circulated because their circulation serves no particular interest or view. Although many significant Awards are published and are the basis for much of what we do, we should not treat the available corpus of published Awards as if they were the equivalent of a fully reported body of case law from a state court system where all judgments are available.
Comment 8:
People often ask if there is a difference among arbitration rules and arbitral institutions. The various rules are uniform in imposing privacy on the hearings, so far as I know. The ICC Rules do not impose confidentiality duties on the parties with respect to the award. However, the UNCITRAL Rules prohibit publication of awards without the consent of the parties (see UNCITRAL Arbitration Rule 32(5)), the LCIA Rules impose a duty of confidentiality on the parties generally unless they expressly agree to the contrary in writing (see LCIA Rules Art. 30) and the AAA International Rules also prohibit making the award public unless the parties have consented or as required by law (see AAA International Rule 27.4). In contrast, the ICSID Rules prohibit the Centre from publishing awards without the consent of the parties (see ICSID Arbitration Rule 48(4) and ICSID Financial and Administrative Regulation 22(2)), but at least one knowledgeable author believes that the parties are free to publish ICSID awards unless they agree otherwise. See Schreuer, The ICSID Convention: A Commentary (2001) at 822. I hope this minutiae is of use (or at least amusing).
Comment 9:
It may be appreciated that the issue of confidentiality quite recently has been tested by the Supreme Court of Sweden (2000). Having the Supreme Courts judgement in mind it may - for those who wish for confidentiality rather than transparency - be good advise to be very specific on confidentiality when drafting arbitration clauses which will be applicable in proceedings governed by Swedish law, including when arbitration may take place under the rules of the Arbitration Institute of Stockholm Chamber of Commerce (the "SCC Rules").
To my mind its a good thing that those who wish confidentiality specifically have to accept the burden of having to agree on that, especially since transparency and the risk of public display generally seems to work as a drivers for settlement (and perhaps mediation subject to complete secrecy) before even starting arbitration proceedings. This does surely not apply to all governements or companies, but certainly for some "sensitive" governments (thinking of themselves as adhering to good governance etc) and most of all listed companies. Therefore, the recent clarification by the Swedish Supreme Court is quite sound.
An arbitration clause contained in any contract has in Sweden for a long time been considered to be an agreement within the agreement. The law governing the material part of the agreement does therefore not include the arbitration clause. Given the principles of separability in many jurisdictions this may not come as a surprise.
The Swedish Arbitration Act provides no rules with regard to confidentiality for parties and arbitrators, and neither do the ICC Rules or the SCC Rules. Notably, the Arbitration Institute of the Stockholm Chamber of Commerce does not itself disclose any information regarding proceedings being under their supervision. Yet, the general perception in Sweden - as well as other jurisdictions judged from the discussion on OGEMID for the last couple of days - is nevertheless that some kind of obligation of confidentiality exist. Especially since confidentiality is one of the common arguments for using arbitration as opposed to general and public court proceedings in the first place.
In essence the Supreme Court in the case of 2000 ruled that a party in arbitration proceedings governed by Swedish law is not bound by confidentiality, unless the parties have entered into a specific agreement to that effect. The dispute occurred after a debt had been transferred to Company X from an Austrian creditor. The agreement in question was entered into between the Austrian creditor and a Bulgarian Bank. The Bulgarian Bank never consented to the transfer of the debt. Based on the lack of consent the Bulgarian Bank refused to pay the outstanding and transferred debt. Company X initiated arbitration proceedings, as provided in the agreement, which also stipulated that the ECE-rules were to apply to the proceedings, and that the proceedings were to be held in Stockholm and that Austrian law was to apply.
A decision of the Tribunal was subsequently sent to Mealey's by a representative of Company X. The Chairman of the Tribunal was informed of the disclosure while the arbitration proceedings continued. Later on, the Chairman of the Tribunal attended a dinner party where he disclosed information about the decision to a Justice of the Supreme Court who was working on a case which also involved the transfer of rights in an agreement to a third party. The Supreme Court Judge later used the decision and in the subsequent judgement reference was made to the published article in Mealey's.
In the arbitration proceedings, the Bulgarian Bank claimed revocation of the arbitration agreement and disqualification of the Chairman of the Tribunal. The Bulgarian Bank claimed that the disclosure of the decision by Company X and by the Chairman, rendered the arbitration agreement null and void. The Tribunal dismissed this claim and later issued its award.
The Svea Court of Appeals held that the ECE-rules (I think Article 29) applied to oral proceedings before a tribunal, but not to written documents or to awards or decisions given by a tribunal. Given the nature of the information that had been disclosed through the publication of the decision, the Svea Court of Appeal held that it did not constitute a material breach of the parties' mutual duty of loyalty argued by the Bulgarian Bank, which would have given the Bulgarian Bank the right to revoke the agreement.
On appeal the Supreme Court later tested only the issue of whether Company X's disclosure had given the Bulgarian Bank the right to revoke the agreement, rendering the award invalid. Firstly, the Supreme Court established that a judgment in favour of the Bulgarian Bank would only be possible if Company X had been bound by an obligation in the agreement to adhere to confidentiality. It was established that there was no such explicit obligation in this respect in the agreement and that Swedish law did not provide a basis for any general and legally sanctioned undertaking to this effect.
It was further established that the ECE-rules only applied to oral proceedings before a tribunal, and not to the dispute as a whole. The Supreme Court also emphasised that in an arbitration dispute, the nature of the information dealt with varies to a great extent, and therefore that a disclosure by a party may lead to different considerations depending on the nature of the information in question (not a very surprising conclusion). The Supreme Court therefore concluded that a disclosure may constitute a breach under the specific rules of the Swedish Trade Secrets Act, i.e. disclosure of specific information being kept secret by an entrepreneur.
The Supreme Court went on and stressed that the private character of arbitration proceedings implies that third parties do not have the right to attend hearings or obtain submissions or petitions of the parties. Notably, this does not hinder disclosure of information about the arbitration proceedings. The Supreme Court seems to have given great wight to the fact that the recent Swedish Arbitration Act of 1999, does not contain any rules on confidentiality. The Supreme Court also embarked on a comparative search for a common foreign legal position on this subject. For example, it was noted that the general opinion under the laws of England, obliged the parties to adhere to confidentiality. Specific reference was also made to the mentioned Australian case of 1995, where the High Court ruled in favour of the opposite position. In conclusion, the Supreme Court found no international consensus that could enlighten the position of the Swedish law on the subject.
The Supreme Court unanimously ruled that a party in arbitration proceedings governed by Swedish law could not be bound by confidentiality absent a specific agreement to this effect. Company X's disclosure in Mealey's therefore did not costitute a material breach of the agreement giving the Bulgarian Bank the right to revoke it. The award was deemed valid and the appeal was dismissed.
Comment 10:
I can appreciate the distinction below between what is termed as "commercial arbitration" and "arbitration of investment disputes", though there is a good deal of overlap as the latter still has to deal with many issues of commercial character. The distinguishing feature between the two lies in the presence of "State" as a party in its sovereign capacity and, in many respects, for determination of its obligations and rights under international law, i.e., the bilateral or multilateral treaty. It might be useful to recall here while in private law there could be private and confidential agreements, international law does not recognize a concept of secret treaty. Thus, a dispute arising from an investment treaty could hardly be regarded as a confidential matter, at least as far as the State is concerned.
In arbitrations between a State and a private party, be it arbitration of investment disputes or traditional mixed claims commissions created by a treaty there has always been a good deal of publicity and transparency. ICSID is a good example. Publication of the Iran-US Claims Tribunal awards can also be cited.
However, in commercial arbitration privacy of the proceedings is still a major attraction for the parties. For instance, the parties assume that content of the arguments they make or the documents they submit will not be made public, unless at some stage it is publicized through court proceedings, e.g., for enforcement. It may be that the confidentiality of the proceedings also places certain obligations on those involved in the proceedings, in addition to their professional obligations, such as the arbitrators, secretaries (or the arbitral institution), stenographers, etc. How do we deal with these matters if a wider transparency is advocated?
Comment 11:
comment was thought provoking. Even if one accepts the proposition that there can be no such thing under public international law as a secret treaty, does that mean that all disputes about their interpretation (and about disputes generally under public international law) must be public? That would be an interesting proposition to put to States. Even the UN Security Council has in-camera sessions. Still, I quite like the idea that I have a right to sit in with Tony and Jacques the next time they meet to resolve their disputes. And that reminds me: I will ask George W to put another steak on the barbie for me at his ranch the next time a foreign dignitary is visiting him, so that I can have my say.
The analogy of BIT dispute to administrative review is interesting. However, it may be proposed from a failure to reflect on the basic nature of what a BIT is. A BIT dispute is a dispute that relates to the allegation that one State has violated its obligations under public international law to another State. A BIT is a treaty. The parties are two States. The dispute is inter-State.
The States-party to a BIT have agreed to certain terms in the BIT. Generally, one term is that disputes between the States over the application of the treaty can be resolved by negotiation and, failing that, binding arbitration. Should such negotiations be public? Must they, under public international law? Should such arbitrations be? Perhaps.
Another general term of BITs is that the States-party agree a method whereby their nationals who invest in the other State can oblige that other State to enter into negotiations and arbitration in the event that that other State violates its obligations under the treaty (i.e., the obligations the State owes under public international law to the first State: treatment of aliens; State responsibility; etc.). The rights at issue are still those pertaining to the State of the investor's nationality. The parties to the BIT have merely agreed a way by which nationals (not traditionally subjects of international law) can enforce the rights of their State directly against the host State, without the traditional intermediary of diplomatic protection. The same questions thus pertain.
I have seen nothing in the terms of any BIT that that requires the States-party to allow non-party interventions in the dispute resolution process. The correctness of the NAFTA decisions on the point has been challenged elsewhere as the view of individuals not well-versed in public international law and the law of treaties. The contrast in terms of the new US-Singapore FTA underscores this.
If it can be taken that there is nothing compelling the parties to a BIT dispute to include non-parties, is there a sound policy or ordre public reason for tribunals to do so of their own initiative? There appears to be no enormous call from States or investors for such a development. So who is agitating for it?
A number of NGOs appear to be keen on it. They appear to dislike what they interpret to be a lack of transparency in "the process". Of course, anyone who attempts to discover the funding sources and decision-making matrix of most NGOs will consider the processes of BIT arbitration to be a public circus by comparison. It relates to one of those famous verb declentions, like "I am a freedom-fighter, you are a national liberator, he is a terrorist": I am discrete and protect my innocent members who speak for the world; you are well-intentioned but have been co-opted; he is illegitimate and trying to hide his tyrranical prepensities.
But leaving that conjugation to one side, what the NGOs appear to be agitating for generally is transparency. Personally, I like the idea of transparency. It is good for citizens to know what their government is proposing to undertake as obligations on the intenrational stage. But is that transparency best effected at the national stage, when a State is considering entering into a BIT or a BIT is put before a parliament for ratification (i.e., "democracy")? That would mean addressing issues, such as permissible or legitimate activities by a government re the environment or labour laws or whatever, before the inter-State obligations have been undertaken.
The alternative would be to have transparency at the stage where there is an allegation that the obligations already undertaken have been violated. What I have read of NGO objections to the lack of transparency in BIT arbitrations is that they want to be able to address a tribunal on subjects such as whether environmental safeguards should be exempt from the provisions in BITs prohibiting certain acts by governments. Surely that debate is best engaged at the natinoal stage during negotiatiion of the BIT rather than after the obligations are already in place.
Perhaps it is easier for NGOs to convince a three-person tribunal than a national government (or voters countries where there is "democracy"). On the other hand, perhaps there should properly be more transparency in international relations between States.
Comment 12:
There are certainly very good arguments for lifting the veil of secrecy in investment arbitration. But one must also consider that open, public arbitration proceedings have a drawback: there seems to be a lack of control for investment arbitration awards. Awards cannot be set aside or modified in higher courts if the arbitrators misinterpreted the law or if their reasoning runs counter to good policy reasons. The Metalclad-Vancouver decision and the Aguas-annullment seem to be the exception to the rule.
This hardly seems to have been a problem as long as investment arbitration awards have been scarce, and even more scarcely published. But with now 15 new cases for ICSID in 2003, this lack of control seems apparent. As Jan Paulsson said in Cologne: "Let's face it: Sometimes tribunals get it wrong". But when ? There seldom is a critical discussion in literature or subsequent awards. From the viewpoint as a practitioner, this might be understandable.The tribunals are filled with eminent lawyers and professors, and is not easy to claim their decision is wrong. Also, one always meets twice.
To enhance the credibility of investment arbitration, further review mechanisms which also deal with the substantive content of an award might be necessary.
