Monday’s New York conference on “Arbitration with States and State Entities under the ICC Rules” got me thinking about the possibility of amicus submissions in investment cases before the ICC or other institutions beyond ICSID. A few musings:
Are amicus debates likely to arise in the ICC context? The answer is yes. Although most ICC cases involving States or State entities have arisen from contracts, a number already have alleged breach of investment treaties, and more presumably are to come. A recent ICC task force suggested that roughly 20% of BITs allow some possibility of using the ICC Rules, either by expressly providing that option or by allowing the parties to agree on an institution and rules beyond those specified in the treaty. Treaty cases frequently present issues of broader public interest. Indeed, such issues can arise even in contract cases involving States; the ICC Court reportedly twice has considered amicus issues in contract cases, one involving the European Commission and another a U.N. agency. The fact that ICC filings are not publicly posted, the way ICSID registrations customarily are, does not mean that civil society groups will not learn of their pendency and ask to be heard. The parties may quietly notify supportive groups and encourage their participation.
Would an ICC tribunal have authority to accept amicus submissions? The powers of the arbitrators fundamentally stem from the terms of consent. In investment cases, unless the claimant has a contract with the State providing for arbitration, the consent generally arises from treaty. But most treaties, with the exception of a few based on recent U.S. Model BITs, do not contain advance consent to nonparty submissions. Even in the NAFTA context, which has involved several amicus submissions, the consent stems not from the treaty but from a 2003 joint Statement of the Free Trade Commission.
If treaties are silent, the next potential source of consent are the procedural rules the State offered and the investor accepted. The ICSID Rules expressly address nonparty submissions, and place the issue squarely within the arbitrators’ discretion to decide. But the ICC Rules, like the current version of the UNCITRAL Rules, are silent. The UNCITRAL working group is currently considering a rules revision to parallel the ICSID approach. The task force behind the recent ICC Rules revision similarly considered adding a provision on amici but ultimately declined, reportedly worrying that it might complicate proceedings and discourage parties from choosing the ICC, perhaps over ICSID.
The ICC task force believed that if both parties agreed, a tribunal could admit amicus submissions under the general procedural discretion provided by Article 19(1). This surely is right: parties by consent generally may frame the arbitral process. But this general discretion probably could not be stretched to allow a tribunal to accept amicus submissions over the objections of a party, as may now occur in ICSID cases. The issue might be considered analogous to arbitrators’ relying on independent research, beyond the facts or law presented by the parties; doing so could well expose their award to challenge for exceeding their authority.
If a tribunal has discretion, what standards should it apply? ICSID Rule 37 provides useful guidance. It identifies as relevant the extent to which the nonparty submission “would assist the Tribunal in the determination of a factual or legal issue — by bringing a perspective, particular knowledge or insight that is different from that of the disputing parties,” “would address a matter within the scope of the dispute,” and would reflect “a significant interest in the proceeding” by the nonparty itself. The Tribunal should also consider whether the submission would “disrupt the proceeding or unduly burden or unfairly prejudice either party,” and ensure that the parties can present observations on the submission.
The track record of amicus applications in investment cases been mixed, depending on who the amicus is and whether it is considered to offer any special knowledge that could usefully guide the tribunal’s decision. The European Commission was granted permission in AES v. Hungary and Electrabel v. Hungary (full disclosure: I represented the respondent); the tribunals essentially accepted that it offered a distinct perspective. It was the enforcer in first instance of EU law, which Hungary invoked to defend its actions; the disputes involved claims by European companies against a European State; and the European Communities themselves were signatory to the Energy Charter Treaty, which governed the claims. By contrast, amicus applications by NGOs have had mixed success. Some have been granted, as in Biwater v. Tanzania, Suez v. Argentina and Glamis v. United States; others were rejected, as in Chevron v. Ecuador and in two parallel cases against Zimbabwe.
The Glamis case is interesting because it reminds us that amicus applications can be in favor of either side. Although environmental and labor groups arguably have been most vocal in arguing for transparency and nonparty submission rights, presumably to defend vigorous State regulation, business promotion groups also may have an interest in being heard on the side of aggrieved investors. In Glamis, the tribunal accepted briefs from both the Quechan Indian Nation, arguing for government’s duty to preserve sacred lands, and from the National Mining Association, arguing that regulation should not undermine protected interests of the mining sector.
What parameters should tribunals consider to maximize the utility of nonparty submissions while ensuring fairness to the parties? Here there are many factors to consider. To pose just three of the tougher ones:
1. Should a tribunal allow amici access to the parties’ submissions? The tribunals in AES and Electrabel came to opposite conclusions: the first denied the Commission’s request for access and the second granted it, subject to possible redactions. What is the argument? On one side, granting access allows an amicus to more meaningfully join issue; if the rationale for accepting its submission is that the tribunal believes the amicus has special knowledge or insight, wouldn’t the tribunal benefit from helping that amicus target its submissions to the precise questions presented for determination? That reasoning led the Foresti v. South Africa tribunal to allow several NGOs access to the parties’ submissions. But doing this raises questions about confidentiality. What if the amicus is a commercial actor in the same market sector as one of the parties, or an NGO with its own vested interest in the sector? The Biwater and Suez tribunals denied amici access to the parties’ filings, in part because of confidentiality concerns. That result seems even more logical outside the ICSID rules, which at least grant tribunals procedural discretion over the amicus process. ICC arbitrators, like arbitrators under most rules, are subject to obligations of confidentiality. Absent exceptions arguably flowing from discretion granted in the rules, how could arbitrators ethically release the parties’ submissions to a nonparty over their express objection?
2. Could a tribunal ever require the amicus to be available for cross-examination? Generally the parties oppose amicus participation at the hearing, and generally tribunals deny these requests. But to date the issue has been addressed mostly before the parties see the written submissions themselves. In principle a party may feel differently if the submission makes important assertions of fact. If the very reason for allowing the submission is that the non-party arguably offers some special factual knowledge or technical expertise, shouldn’t the parties have an opportunity to test its assertions through cross-examination? Why should all other presenters of fact or expertise be subject to questioning, so the tribunal can evaluate the persuasive value, but not the amici?
3. Finally, could a tribunal ever invite amicus contributions to costs? Nonparty submissions clearly add to the cost of a proceeding. The parties have to prepare observations in response, perhaps of significant length (some past cases have involved submissions from five different amici, each presenting its own perspective). This results in extra argument at the hearing and extra time by the arbitrators to consider the arguments and possible address them in their decision. Should the parties alone bear the attendant costs? Would it ever be proper for a tribunal to ask a nonparty to contribute to those extra costs? That could chill the involvement of civil society on issues of broader public interest. But if public interest is the appropriate consideration, why should the cost of subsidizing the broadening of the dispute, into a form of public notice-and-comment proceeding, fall entirely on the parties?
There are no easy answers. Fundamentally, these questions devolve into one basic one (“whose arbitration is this anyway?”), with a subsidiary one (“should the answer be different for investment disputes?). Are investment disputes still basically disputes between two distinct parties, or are they a forum for resolving issues of public interest?
The question leaves ample room for debate. As the ICC and other institutions try to move into the investment arbitration sphere largely dominated by ICSID, they too will have to grapple with these issues — with even less guidance in their procedural rules than the ICSID Rules now provide.