Under Chapter 11 of the North American Free Trade Agreement (NAFTA), foreign investors received the right to directly challenge regulatory measures implemented by member states and to have their complaint adjudicated by procedures derived from international commercial arbitration. Much controversy has centred on investors' rights to challenge state actions directly, without the intervention of their own state. However, the method by which these disputes are settled may be just as problematic in altering the power relations between public actors (states) and private ones (investors). This article investigates mechanisms that might be used to protect the public interest should unwelcome decisions emerge from the arbitration process. It concludes that they are ineffective. Rather, states have sanctioned a significant transfer of authority from public to private control. Essentially, a portion of national sovereignty is surrendered, not just to international entities, but to private ones. Thus, the Chapter 11 system for adjudicating investor-state disputes plays an important but largely unobserved role—one that can illuminate ongoing debate about the role played by states in the era of globalization.
Early globalization theorists posited the demise or obsolescence of the nation-state; they claimed it was leaking authority and sovereignty upwards (to supranational organizations), downwards (to subnational units) and sideways to markets (and dominant players in them, like multinational corporations) and emerging players, such as international civil society.1
Ohmae (1990) provides the classic assertion of the obsolescence of the nation-state.
This is an important exercise, since the demolition of the straw man erected by early globalization theorists could lead to an equally flawed conclusion: that nothing much has changed. Clearly, the impact of globalization on state sovereignty, authority and capacity is contested.2
The following attempt to categorize positions is far from exhaustive and certainly the citations list represents only the tip of a very large iceberg. Nevertheless it may serve to highlight the role of definitional disputes in some of the debates on this subject and also to lead into the issue with which this article is concerned, namely the rise of private power and authority (see Nolke, 2004: 161). This is a topic that much international relations and international political economy theory finds hard to incorporate.
But this effect is not universally conceded. For example, Maria Gritsch employs a realist perspective to depict the advanced capitalist states as constructing globalization and using it “to acquire greater power over and autonomy from their national economies and societies” (2005: 2), as well as to “enhance their international politico-economic power at other states' expense” (2005: 13). Thus, state sovereignty is retained vis-à-vis other states, and the state also remains predominant as far as private interests are concerned, since it holds on to the capacity to monitor, regulate and exercise “direct control over [firms'] capacity to execute foreign investment, establish production networks, or complete sales. Governments can obstruct these behaviors if they conflict with the state's … interests” (Gritsch, 2005: 13–14). From a legal perspective, Kal Raustiala (2003) argues that sovereignty is little affected by agreements that are revocable, as is the case with most international agreements. What the state can revoke does not diminish its sovereignty (2003: 846). Indeed, Raustiala considers that structurally induced “interdependence” between states meant that sovereignty was already being lost. To restore or reclaim it, states have to collaborate in new international institutions (2003: 856–875).
To some extent, the dispute over ‘revocability’ is a definitional dispute, though it turns on how easily revocation might be implemented by states that have signed onto international agreements. Krasner, for example, broke down sovereignty into a number of components. One of these, “international legal sovereignty,” approximates Raustiala's abstract legal concept of undiminished (formal/legal) sovereignty. Krasner's other components—domestic sovereignty, interdependence sovereignty, Westphalian sovereignty3
Krasner (1999: 3–4) provides the following definitions:
International legal sovereignty refers to the practices associated with mutual recognition, usually between territorial entities that have formal juridical independence. Westphalian sovereignty refers to political organization based on the exclusion of external actors from authority structures within a given territory. Domestic sovereignty refers to the formal organization of political authority within the state and the ability of public authorities to exercise effective control within the borders of their own polity. Finally, interdependence sovereignty refers to the ability of public authorities to regulate the flow of information, ideas, goods, people, pollutants, or capital across the borders of their state.
For that matter, the accepted meaning of sovereignty may also change, especially when the state may be delegating some of its authority to private actors.4
According to Nolke (2004: 161), this is because established theories in international relations and international political economy remain too state-centric to easily accommodate transnational private authority.
The degree to which private governance is partially displacing public authority in the judicial sphere has been explored by Claire Cutler:
[F]undamental transformations in global power and authority are enhancing the significance of the private sphere in both the creation and enforcement of international commercial law. State-based, positivist international law and “public” notions of authority are being combined with or, in some cases, superseded by nonstate law, informal normative structures, and “private” economic power and authority as a new transnational legal order takes shape. (2003: 1)
This is accomplished with the full cooperation of governments, which are “providing a hospitable legal and regulatory framework for private, secretive, and closed arbitration proceedings” (Cutler, 2003: 27), increasingly serving to settle disputes in a wide variety of areas.5
As noted below there have been effective efforts to make the proceedings more transparent.
To partially test the thesis that authority is shifting from public into private hands, this article examines the procedures for settling disputes under NAFTA Chapter 11. It detects evidence of a delegation or transfer of public authority to the private sector. NAFTA Chapter 11's grant to investors of the right to make direct claims against signatory governments is dealt with only incidentally; rather, the focus is on the procedures for resolving such claims, and the means available to states to assert the public interest, given these procedures and the context of investor rights in which they are embedded.6
Giving investors the right to directly launch such claims was an innovation in itself, especially since the rules surrounding that transfer are such as to limit state regulatory capacity (Van Harten, 2005: 602).
Investor-state relations need not be covered by interstate treaties like NAFTA. They could be governed by contract, in which case the resolution of disputes would be as provided for in the contract and subject to the provisions of the domestic law of the host country. The relationship would be like a private contract between individuals or firms. But, as Gus Van Harten points out, the effect of covering such disputes in the provisions of a treaty is to render investor-state disputes “public” rather than private. However, in NAFTA and in many bilateral investment treaties (BITs), by the provisions of the same treaty, private authority is extended to the settling of disputes. This is achieved by recognizing the processes and institutions of the international commercial arbitration, as expressed in agreements like the New York and ICSID conventions, as the appropriate means of dealing with disputed issues (Van Harten, 2005: 604).
Van Harten traces the evolution of state recognition of the international commercial arbitration system through a number of stages, including the New York convention (1958), then ICSID 1965, and finally the BITs and NAFTA. At each stage except the last, according to Van Harten, the state retained final authority, since it did not grant a “general consent” to subject itself to compulsory binding commercial arbitration. But NAFTA and many BITs do grant an explicit general consent, with the result that investor-state arbitration is transformed “from a modified form of commercial arbitration into a system to control the exercise of [the state's] regulatory authority with respect to investors as a group” (Van Harten, 2005: 608).
NAFTA embeds foreign investors' rights equivalent to the rights of states in some areas and provides that investor-state disputes “shall be considered to arise out of a commercial relationship or transaction for the purposes of the New York Convention and Article I of the Inter-American Convention” (Article 1136.7). Essentially, this pre-commits the signatories to treat disputes between themselves and foreign investors as commercial rather than regulatory matters, and to have such disputes settled according to the private international commercial arbitration system. In line with international conventions, states have generally adopted legislation and rules that limit domestic review of international arbitral outcomes (Cutler, 2003: 225–6, 231–3). This produces a situation in which “the arbitrators act as private judges, holding hearings and issuing judgements. There are few grounds for appeal to courts, and the final decision of the arbitrators, under the terms of the widely adopted 1958 New York Convention, is more easily enforced among signatory countries than would be a court judgement” (Dezalay and Garth, 1996: 6). The New York Convention includes a commitment that signatories will enforce commercial arbitration awards made in other jurisdictions.
Even if a state were to alter its own domestic legislation, it would still be bound by international commitments to enforce arbitration decisions made in other countries. One of the chief effects therefore might simply be to lose attractiveness as a site of arbitrations since its own legislation might be viewed as less desirable. Of course, withdrawal from NAFTA itself would mean removal of the state's a priori general consent, opening space for new domestic legislation.
The Normative Debate
Both defenders and critics weigh in on the transferral of the adjudication process to the international level and the expansion of the role of private authority. Much of the discourse amongst international trade experts explicitly deals with methods of making the (liberal) rules of the international trading system more stringent in restraining governments operating under pressures of democracy. Jackson (1993: 572) observes that “making the rules more effective will tend to limit or constrain government discretion … in the face of domestic interest groups, or ‘rent seekers’ who have sought to reduce trade liberalization so as to enhance their own economic profits, rents, and positions.” Clearly, in this framework, the democratically elected legislative and executive branches of government are most prone to be captured by “rent seekers.” On a scale of increasing remoteness from state interference, placing decision-making authority in judicial institutions that are both international and private insulates them well from political interference.
There have been influential attempts to theorize the reconstitution of international economic relations on the basis of enhanced private rights (Petersmann 1991, 1993). Dypski (2002: 234) argues that “[a]llowing individuals to directly approach governments and seek redress is fundamental to the democratic expectations of the NAFTA countries, and indicates a greater evolution towards privatization of international law” (see also Byrne, 2000). The attraction of arbitration includes a capacity for parties to avoid being subject to the laws of another party's state in many respects. The process traditionally has also been more secretive than pursuing the matter through courts. Historically, recourse to international commercial arbitration rather than host-country courts has been particularly attractive to foreign investors in developing countries (Dezalay and Garth, 1996: 5, 86). For business, there is some assurance that selected arbitrators will have appropriate expertise (Dore, 1986: 5) and, perhaps more subtly, since arbitrators do not enjoy judicial independence but must establish a reputation in order to further their careers, a broadly sympathetic view of businesses' concerns can be anticipated (Dezalay and Garth, 1996: 117).
However, this system of private justice, with a rather closed cadre of arbitrators,7
See Dezalay and Garth (1996: 10, 23–4) on the personal attributes and class background of arbitrators.
Asserting the Public Interest: Methods of Control
Although there is debate about its desirability and implications, there is little doubt about the empirical reality of private authority in international investment dispute settlement. To what extent does this imply loss of sovereignty by the state? Using Krasner's typology, one can detect diminished domestic and Westphalian sovereignty, which in turn have an impact on interdependence sovereignty. As the system is state-sanctioned, international legal sovereignty remains intact. For Raustiala (2003), the key issue would be revocability. Having created this system, can states change it, if they so wish? If they can, there is no loss of sovereignty in Raustiala's terms. The remainder of this article probes the means available to states to “correct” unwanted decisions arrived at in the private arbitration system created by NAFTA Chapter 11. It concludes that only by setting the threshold for “revocability” at a highly abstract and legalistic level can the argument be sustained that there is no loss of sovereignty (in this case to private authorities). At a more reasonable, “real world” threshold, it is clear that states have created a system that they can only change or influence with great difficulty, if at all.
In NAFTA, five means of “correction” can be identified. The first is withdrawal from NAFTA, which can be accomplished by any one of the signatories on six months notice (NAFTA Article 2205). This establishes that revocation is legally possible and is a sufficient condition for Raustiala to conclude that sovereignty is unimpaired. However, as long as NAFTA is considered generally beneficial,8
Even if it ceased to be so considered, asymmetries of power might deter resort to this provision.
According to Sampliner (2003: 30), submissions under Article 1128 have been made frequently. Even if all three parties are in full agreement in a submission on a particular point, it is not binding on tribunals. The parties have argued that such agreements in submissions are technically “subsequent agreements,” which, under the Vienna Convention on the Law of Treaties, must be taken into account by tribunals. However, Sampliner (2003: 30–31) notes that the issue of whether tribunals are bound by agreements of the parties in such submissions under Article 1128 has not been conclusively tested, some tribunals having rendered decisions that appear consistent with agreed interpretations, but, in one instance at least, a tribunal may have disregarded an agreement in the Parties' submissions on an interpretation.
The Free Trade Commission's ability to bind tribunals seems more assured under Article 1131.2, since the language is explicit: “An interpretation by the Commission of a provision of this Agreement shall be binding on a Tribunal… .” As with amendment, however, it is necessary for unanimity to prevail amongst the parties, although it is an easier process than amendment, since legislative approval by the parties is not required. In the absence of such unanimity, tribunals enjoy considerable leeway. For example, a trilateral investment group has been meeting since 1998 to work towards a common understanding of Article 1110, which deals with expropriation. To date no agreement has emerged on expropriation, leaving it open, as Mann puts it, “to investors who wish to pursue broad readings of the expropriation provision, under which normal regulatory measures with an economic impact on foreign investors can be challenged under Chapter 11.” Panels have varied in their receptiveness to such challenges and until an authoritative interpretation is issued uncertainty will continue (Mann, 2003).9
Mann's assessment (2005) of the outcome of the Methanex case was much more positive. In Methanex the tribunal, amongst other rulings, defined the “like circumstances” issue more tightly than other tribunals, such as that in the S.D. Myers case; and ruled that regulations for a public purpose are not expropriations or measures tantamount to expropriation. However, as Mann notes (2005: 9), absence of binding precedent in arbitration hearings means “legal uncertainty remains the hallmark of this system of arbitration.”
The Free Trade Commission has issued Notes of Interpretation of Certain Chapter 11 Provisions promoting limited transparency and providing certain clarifications (Free Trade Commission, 2001). This interpretative note was to clarify Article 1105 (minimum standard of treatment) and procedural transparency provisions. On transparency, the interpretation held that “[n]othing in the NAFTA imposes a general duty of confidentiality on the disputing parties to a Chapter Eleven arbitration” and promised “to make available to the public in a timely manner all documents submitted to, or issued by, a Chapter Eleven tribunal,” subject to certain specified exceptions (Free Trade Commission, 2001). However, reaction has been mixed. Sampliner (2003: 31) considers that as a result of the note, many of the procedural criticisms of NAFTA arbitration have been addressed satisfactorily, such that “NAFTA arbitrations have become largely public proceedings, with open access on the Internet to awards, pleadings, and even in many cases, hearings, transcripts; acceptance of amicus curiae briefs from concerned public interest groups, and most recently, hearings open to the public.” And indeed, there is much more information available than before. However, VanDuzer (2002: 7) has highlighted the limited but significant exceptions to transparency that the FTC note permits, and which stem from the arbitral rules that govern Chapter 11. He emphasizes the veto power left to the investor in a Chapter 11 case. In his view, the interpretation amounts only to a commitment to “seek the consent of the investor to disclosure and a tribunal order permitting disclosure.”
Moreover, doubts have been expressed about whether the commission's power to make binding interpretations is as unlimited as it appears. Matiation (2003: 479) has pointed out that the commission has the right to interpret but not change or amend the agreement (see also Shapren, 2003: 349; Staff and Lewis, 2003: 328). One NAFTA panel did argue—though in dicta rather than in its decision, and without relying on the point—that it viewed the interpretation as an amendment and thus beyond the power of the commission (Sampliner, 2003: 32). The agreement, as a state-to-state treaty, is governed by principles of international law that may place limits on its interpretive capacity. Thus, the commission's authority to issue interpretations that address concerns arising about the agreement's impact on public policy and the common good is not “unfettered” (Matiation, 2003: 495). Similarly, Weiler argues that “it is not clear if the three trade ministers can actually use this power as a sort of ‘sore loser’ clause, ordering tribunals to ‘interpret’ NAFTA provisions in a way that even previous tribunals have already concluded should not be done…. [T]he NAFTA Parties are now issuing mere ‘recommendations’ which clearly could not be binding under NAFTA Article 1131(2)” (2006).
Judicial Review
The final public control mechanism is that of judicial review of arbitration decisions. To date, three judicial reviews of NAFTA Chapter 11 arbitration decisions have been completed: the Myers, Feldman and Metalclad cases. In the Myers and Feldman cases, the arbitration decisions were sustained; in the Metalclad case, portions of the arbitration decision were overruled, while others were sustained. Three cases is obviously not a large ‘n’—although it is the universe of cases that have proceeded under NAFTA Chapter 11 as far as judicial review. Consequently, a review of these cases should reveal much about the role that judicial review may play.
As we shall see, notwithstanding the partial overrule of the tribunal's decision in the Metalclad case, there are grounds for thinking that it is extremely difficult to use judicial review as a means of overturning damaging decisions by arbitrators. This is because the grounds for judicial review are quite narrow and because judges show great deference to the decisions of arbitrators and, consequently, exhibit considerable self-restraint in deciding whether to overturn their decisions.
Chapter 11 arbitrations can take place under one of three sets of rules—the ICSID, ICSID Additional Facility or UNCITRAL rules. Only the latter two are operative, since neither Canada nor Mexico has yet ratified the ICSID convention (VanDuzer, 2002: 2–3). NAFTA itself does not provide for judicial review of arbitrations, so the arbitration rules being followed determine this. Although the ICSID Additional Facilities and UNCITRAL rules provide that the results of arbitrations are final and binding, neither set of rules prevents arbitral awards from being subject to judicial review under national (or subnational) law. In practice, the applicable law on procedural matters is deemed to be that of the place of arbitration. VanDuzer (2002: 18) comments: “Under Canadian federal law and laws of each province the main grounds upon which a court may set aside an arbitral award are as follows: a party was not given proper notice of the appointment of an arbitrator, the award deals with a dispute not contemplated by or falling within the terms of the submission to arbitration, or the award is contrary to the public policy of the state in which the award is sought to be enforced.” The latter concept could, in the ordinary sense of the term, be construed as wide-ranging. In practice, a narrow interpretation of public policy renders it more of a theoretical than practical constraint on arbitrators (see Cutler, 2003: 225–36).
The cases
Metalclad was a US investor that initiated a claim against Mexico under NAFTA Chapter 11 (for details see The United Mexican States v. Metalclad Corporation 2001 BCSC 664: paras 2 through 18). Metalclad had acquired a hazardous waste transfer station and wished to construct a hazardous waste landfill on the site. Permission was refused by the local municipality, though the company received two federal permits and one state permit. Metalclad claimed, and the tribunal found, that Mexican federal officials had assured them that “they had all the authorities required to undertake the landfill project” (The United Mexican States v. Metalclad Corporation 2001 BCSC 664: para 8). Based on federal authority, construction of the site was completed in 1995, though the municipality issued a “stop-work” order and local protests prevented it from being opened and operated. In October 1996, Metalclad delivered a notice of intent to file a NAFTA Chapter 11 claim. On September 20, 1997, the state governor issued an ecological decree declaring the site an ecological preserve. On August 30, 2000, the NAFTA tribunal delivered its verdict.
The tribunal decision (see Metalclad v. United States of Mexico, ICSID Additional Facility Case No. ARB(AF)/97/1) found in favour of Metalclad and awarded damages of US$16.685 million. The tribunal's reasoning was that NAFTA aimed to increase cross-border investment opportunities, in part through enhanced “transparency.” This it construed to mean that “all relevant legal requirements … should be capable of being readily known to all affected investors of another Party” (para 76). The tribunal found that Mexican federal officials had assured Metalclad that it had all permits necessary, an assurance on which it was entitled to rely (paras 80, 89). The lack of clarity on whether a municipal permit was also required amounted to “a failure on the part of Mexico to ensure the transparency required by NAFTA” (para 88). These actions, coupled with the subsequent enactment by the state level of an ecological decree, meant that Metalclad was not treated “fairly or equitably” as required under NAFTA Article 1105. Such a breach of Article 1105 was considered a measure “tantamount to expropriation in violation of NAFTA Article 1101(1)” (paras 104 to 107). Further, the Ecological Decree, in the tribunal's view, had the effect of “barring forever the operation of the landfill” and provided further grounds for concluding that expropriation had occurred (para 109).
In the second of the cases, S.D. Myers, a US processor and disposer of hazardous waste, decided to acquire Canadian PCBs to sustain its business in the US. A Canadian company was established to arrange export of these materials to the US, and S.D. Myers persuaded the US Environmental Protection Agency (EPA) to relax its regulations against the import of PCBs. At that point, on November 20, 1995, Canada issued an interim order that had the effect of banning the export of PCBs.
Canada justified its decision on a number of grounds, including concerns about whether the EPA's administrative ruling complied with US law, and whether exports of PCBs would violate the Basel convention, to which Canada was a signatory. Further, allowing exports would violate Canada's own 1989 policy on domestic disposal. Then, too, Canadian disposal facilities needed to remain viable, since it was possible that the US would once more close the border to this traffic (see S.D. Myers v. Canada, Partial Award 13 November 2000: para 121).
The tribunal agreed with S.D. Myers' claims that Canada's actions constituted disguised discrimination against Myers and its investment in Canada, breaking the article's “national treatment” provisions and, in the process, failing to meet the minimum standard of treatment under international law.10
Myers' other claims—that Canada's actions violated Article 11.06 (Performance Requirements) and 1110 (Expropriation)—were rejected.
The Feldman case involved complex transactions pertaining to tax rebates available to exporters of Mexican cigarettes. The exporters took delivery in Mexico and hence paid a domestic tax. On exporting the product, however, the tax was rebated, subject to certain conditions. The tribunal itself found it difficult to ascertain the precise facts since some records had been destroyed and the parties to the dispute differed in their accounts of what had transpired (Feldman v. Mexico, paras 6–23). Feldman claimed that in certain periods in the 1990s, other (Mexican-owned) firms engaged in the same business had received rebates, whereas his firm had not. He alleged that this was discrimination based on nationality and thus violated the National Treatment provisions of NAFTA. The tribunal concluded that the favourable treatment of domestically owned firms did constitute discrimination against Feldman and that Mexico had violated NAFTA Article 1102 (paras 173–188). Damages were awarded against Mexico.
Outcomes of Judicial Review
In each of these cases the national government involved challenged the outcome of the arbitration and the cases proceeded to judicial review. Since all had an official “place of arbitration” listed as Canada, the applicable law to be followed was that of the jurisdiction listed as the place of arbitration.
Only in one case—Metalclad—was any part of an arbitration decision struck down as a result of judicial review. All Canadian international arbitration legislation, federal and provincial, closely adheres to UNCITRAL's Model Law on International Commercial Arbitration (Pepper, 1998: 809). The relevant sections of British Columbia legislation, closely patterned after the UNCITRAL Model Law, provide that an arbitration award may be set aside by the court if: it “deals with a dispute not contemplated by or falling within the terms of the submission to arbitration, or it contains decisions on matters beyond the scope of the submission to arbitration”; or “the composition of the arbitral tribunal or the arbitral procedure was not in accordance with the agreement of the parties”; or “the arbitral award is in conflict with the public policy in British Columbia” (United Mexican States v. Metalclad Corporation 2001: para 50).
The determination that the dispute was commercial was made on the basis that the dispute pertained to an investment and was thus commercial in nature (United Mexican States v. Metalclad Corporation 2001: paras 39–49). The terms of NAFTA Article 1136 (7) expressly state that disputes under Chapter 11 are commercial in nature. Similarly, other features of Chapter 11 confirm that the state signatories conceived the process as one of international commercial arbitration (Olasolo, 2002: 195). However, as Olasolo argues (195–6), “as NAFTA article 1136(7) implicitly points out, it is not clear, at all, that the underlying relationship between foreign private investors and a NAFTA state party is commercial in nature.” This is apparent from a case like Metalclad, where “a legitimate public interest goes far beyond the commercial private interests for which adjudication under the UNCITRAL Rules of Arbitration were designed” (Olasolo, 2002: 196). Yet in signing on to this provision, states assigned decision making on these broader matters of public interest to the private arbitration system that is embodied in Chapter 11.
The judicial review of the Metalclad decision concluded that the tribunal imported into NAFTA Article 1105 (minimum standards) a provision (transparency obligations) that was not there (United Mexican States v. Metalclad Corporation 2001: paras 67–76). In so doing, the tribunal “decided a matter beyond the scope of the submission to arbitration” (para 76). Since the tribunal relied on the transparency obligations it had imported in to NAFTA to determine that there had been a measure tantamount to expropriation, this determination, too, was struck down (para 79). However, other findings of the arbitration tribunal were sustained by the judicial review.11
But see Olasolo (2002: 208–9) for a critique of the BC Supreme Court's reasoning.
Critics (for example, sympathetic ones like Olasolo, 2002; and unsympathetic ones like Weiler, 2003) have argued that whilst the BC Supreme Court (Mr. Justice Tysoe) formally proclaimed a narrow scope for judicial review, it did, in practice, extend well beyond the anticipated limits (Olasolo, 2002: 190). For Olasolo (2002: 190), this stemmed from the unsuitability of the commercial arbitration system to deal with the issues arising in Chapter 11 cases, of “having tried to assimilate NAFTA Chapter 11 arbitrations into pure commercial international arbitrations…. important public interests are adjudicated in NAFTA Chapter 11 arbitrations, and, therefore, it is necessary to profoundly reform the current system of arbitration that mirrors the one created to adjudicate pure private interests.” In the Metalclad case the BC Supreme Court interpreted its role broadly and acted, in some respects, to protect the public interest. For this it attracted considerable criticism. The other cases, however, indicate much more self-restraint on the part of the courts and much greater deference to the arbitrators. Unsurprisingly, these decisions have been welcomed by the international arbitration community (see Herbert Smith, 2004; Weiler, 2003).
In the S.D. Myers case, Canada applied to set aside the results of the arbitration on two grounds. First, Canada alleged that the award exceeded the scope of the NAFTA agreement by dealing with matters not contemplated by NAFTA Chapter 11. Second, Canada alleged that the award contravened the public policy of Canada. Both these are among the very limited grounds on which judicial review may set aside arbitration awards (Attorney General of Canada, 2003: paras 6 and 7).
In pursuit of this claim, Canada raised a number of aspects of fact (such as whether S.D. Myers was an ‘investor’ within the meaning of NAFTA) and legal process (such as how much “deference” is due to arbitrators). The Federal Court (Mr. Justice Kelen) was unreceptive to any of the arguments advanced by Canada. His arguments highlight the very limited nature of judicial review that must be presumed normal, Metalclad notwithstanding, in such cases.
First, the judgement identifies the rules for interpreting international treaties and the applicable document. These included the Canadian Commercial Arbitration Act and Commercial Arbitration Code; NAFTA itself; UNCITRAL Arbitration Rules, and the Vienna Convention on the Law of Treaties. Reference to these authorities and judicial interpretation of them makes it clear, for example, that if a matter is within the scope of an arbitration tribunal, there is no allowance for “judicial review if the decision is based on an error of law or an erroneous finding of fact” (Federal Court of Canada, 2004: para 42). In arriving at this conclusion, the judge rejected Canada's argument that arbitration tribunals should be held to a standard of “correctness”—a standard which some commentators consider was followed by the judge in the Metalclad case (Macek 2003).
Canada argued that the arbitration decision violated “public policy,” construed by Canada in this case as respecting the US environmental statutes that prohibited the import of PCBs (notwithstanding the administrative waiving of statutory provisions described above) and respecting Canada's international obligations under the Basel Convention (Attorney General of Canada, 2003: paras 228–232). Justice Kelen, however, rejected this definition: “‘Public policy’ does not refer to the political position or an international position of Canada but refers to ‘fundamental notions and principles of justice’. To cast aside the tribunal decision it would be necessary to find the tribunal's decisions exceed its jurisdiction and be ‘patently unreasonable’, ‘clearly irrational’, ‘totally lacking in reality’ or a ‘flagrant denial of justice’.” These are high thresholds for an appellant against an arbitration tribunal to clear, and a hurdle, in the judge's opinion, not cleared by Canada (Federal Court of Canada, 2004: paras 55–56).
Running through the judgement, and citations from relevant jurisprudence, plus the terms of the various rules applied to judicial review of arbitration decisions, is a very high degree of judicial deference to the pronouncement of arbitrators. This includes the observation that the parties, having created the rules by which arbitrators are selected, must have confidence in the persons who will be adjudicating (Federal Court of Canada 2004: para 16); and the citation, with approval, of Mr. Justice Chilcott, who reviewed the Feldman case: “I accept the proposition that judicial deference should be accorded to arbitral awards generally and to international commercial arbitrations in particular” (Federal Court of Canada, 2004: para 37).
The Feldman judicial review was heard in the Ontario Superior Court, where the legislation to be applied is the UNCITRAL Model Law. In denying Mexico's attempt to have the ruling overturned, Mr. Justice Chilcote made a number of comments and citations of case law that reinforce the view that judicial review is a very limited instrument. He noted (United Mexican States v. Marvin Roy Feldman Karpa, Ontario Supreme Court of Justice Court File No. 03-CV-23500: para 77) that a “very high level of deference should be accorded to the Tribunal,” especially considering that Mexico was attempting to challenge the facts of the case: “The panel who has heard the evidence is best able to determine issues of credibility, reliability and onus of proof.” In any case, the grounds for review under the Model Law did not provide for a review of a finding on facts (para 81). Case law suggested to the judge that “[i]t is meet … as a matter of policy, to adopt a standard which seeks to preserve the autonomy of the forum selected by the parties and to minimize judicial intervention when reviewing international commercial arbitral awards…” (cited in para 78). The judge went on to praise the expertise of the panel with respect to international commercial arbitration and to conclude that there had been no breach of Ontario public policy. His remarks on the score serve to illustrate how limited a ground that is for a successful appeal to judicial review: “The courts of this province have consistently held that for an arbitral award to be interfered with as being against public policy, it ‘must fundamentally offend the most basic and explicit principles of justice and fairness in Ontario, or evidence intolerable ignorance or corruption on the part of the arbitral Tribunal.’… The Applicant must establish that the awards are contrary to the essential morality of Ontario” (para 87). Mexico's case was dismissed. This case was unsuccessfully appealed by Mexico to the Court of Appeal for Ontario. In it decision the Court of Appeal noted: “Notions of international comity and the reality of the global marketplace suggest that courts should use their authority to interfere with international commercial arbitration awards sparingly … the applicable standard of review in this case is at the very high end of the spectrum of judicial deference” (2005: paras 34, 43).
The reluctance to intervene in arbitral awards is thus firmly ensconced in the rules surrounding the arbitration process and in judicial attitudes. These considerations emanate from private international commercial law and are applied, in NAFTA Chapter 11, to situations that go far beyond the disputes that gave rise to them. This point was made, without effect, in Canada's case memorandum before the Federal Court:
137. NAFTA Chapter 11 arbitrations differ substantially from a private commercial arbitration in terms of the extent to which their decisions might affect interests beyond those of immediate parties to the dispute. Claims under NAFTA Chapter 11 are not contractual disputes but challenges to government ‘measures’, a term NAFTA Article 201(1) defines as including ‘any law, regulation, procedure, requirement or practice’. 138. The decisions of NAFTA Chapter Eleven Tribunals have important public policy implications that impact upon, and are of interest to Canadians generally and non-disputing NAFTA Parties. (Attorney General of Canada, 2003: paras 137–138)
Such arguments, of course, are undercut by NAFTA Article 1136.7.
Conclusions
NAFTA categorizes disputes between a foreign investor and a state as commercial rather than regulatory in nature. States grant general consent to have such disputes dealt with by procedures originally established for international commercial arbitration. However, the matters coming before arbitration tribunals raise issues of regulation that are removed from the traditional work of commercial arbitration. In agreeing to regard them as commercial, the state has assigned judicial power over noncommercial matters, in which states have a significant regulatory interest, to a system of private arbitration. The means for asserting a public interest into these proceedings are either cumbersome (withdrawal from NAFTA, amendment) or ineffective. In that few corrective devices to assert the paramountcy of the public interest are available, there is a lack of public control of a private arbitration process that resolves major disputes of public interest. While it might be argued that particular neoliberal governments find these mechanisms afford a desirable insulation from democratic pressures, it is difficult to escape the conclusion that this has been purchased by diminishing national state and public authority and enhancing that of the international corporate sector. The idea that states use internationalization to retain or restore control cannot be sustained. Through mechanisms like NAFTA Chapter 11, a transfer from public to private authority that is difficult to control or reverse has been engineered.
This example of enhanced private authority over public issues demonstrates the complexity of the state's role under globalization. The predictions of writers like Ohmae—that the state is obsolete—may be unconvincing, but it does not follow that states exercise as much authority as previously. In the field of investment protection particularly, states have chosen to confer public authority on private actors and, having done so, lack effective means to modify the results. Short of the drastic step of revoking their membership in international economic agreements like NAFTA, they have ceded some of their sovereignty to international, but more importantly, private bodies. As a result, their legal orders and public policy regimes are subject in important respects to the norms of international corporate law.
Acknowledgments
I would like to thank Greg Albo, David Schneiderman and Donald Swartz, who, together with a reader who wished to remain anonymous, provided valuable comments on earlier versions of this article. Thanks are also due to the anonymous referees of the Journal for their helpful comments.
