The Upshot of Up-Stream Losses in Mexico v Cargill: Judicial Deference to International Arbitration Tribunals
Cargill Incorporated (“Cargill”) is a US-based company that manufactures high fructose corn syrup (“HFCS”), a low-cost substitute for cane sugar that is used as a sweetener in soft drinks. Mexico, the world’s second largest per capita consumer of soft drinks, implemented various trade barriers to protect its domestic sugar cane industry from foreign-produced HFCS. In response to these measures, Cargill sought arbitration for breach of Chapter 11 of the North American Free Trade Agreement (“NAFTA”) on behalf of itself and its Mexican subsidiary distributor, Cargill de Mexico S.A. de C.V. (“CdM”).
The arbitration panel established to hear the dispute concluded that Mexico’s measures were in breach of its Chapter 11 obligations to afford a certain degree of protection to the investors (and investments) of the NAFTA parties and awarded damages to both the parent and subsidiary company in the amount of US$77,329,240. This arbitration award incorporated compensation for both “down-stream” and “up-stream” losses. The former includes the value of direct sale loss and associated costs suffered by CdM and is not contentious. The latter, however, represents the cost of lost sales to CdM of products manufactured by Cargill in its plant in the United States and was the subject of a recent case decided by the Ontario Court of Appeal (“OCA”).
Since the parties to the NAFTA dispute selected Toronto as the “place of arbitration,” Ontario courts were vested with the authority to review the panel’s award under the International Commercial Arbitration Act, RSO 1990, c 1.9, which adopted the UNCITRAL Model Law on International Commercial Arbitration (“Model Law”). In Mexico v Cargill, 2011 ONCA 622 [Cargill], the OCA affirmed the lower court’s dismissal of Mexico’s challenge to the jurisdiction of the panel to award up-stream damages.
In order to reach this conclusion, the OCA was tasked with identifying and framing the proper standard of review to be applied by domestic courts in reviewing the decisions of international arbitration panels. More specifically, the Court articulated when, and on what basis, an arbitration award is subject to being set aside by a domestic court on the ground that it lacks jurisdiction.
The Panel’s Decision – A disjuncture between NAFTA objectives and implementation?
In justifying the award for up-stream losses, the panel acknowledged that Chapter 11 applies only to “measures relating to investments that are in the territory of the State Party enacting the measures.” It also confronted a finding by the tribunal in Archer Daniels Midland v The United Mexican States that it lacked jurisdiction to award compensation for lost profits the claimants “would have produced in the United States and exported to Mexico ‘but for’ the Tax, as these losses were not suffered in their capacity as investors in Mexico.”
The panel in Cargill found that the business model of the claimant (ie. manufacturing HFCS in the United States, importing into Mexico through a border facility in Texas, and distributing in Mexico through CdM’s distribution centre in Tula) was sufficiently integrated so as to constitute a cohesive, indivisible investment for the purposes of Chapter 11. Since CdM’s selling of HFCS to the soft drink industry in Mexico was contingent on its being imported from the parent company in the US, the two operations “were so associated…as to be compensable under the NAFTA.”
Unsurprisingly, the panel’s “integrated” approach to the characterization of investment may have wide-ranging implications for governments seeking to ascertain the potential costs of implementing NAFTA inconsistent regulatory measures. It is difficult to tell exactly where the line of “association” between parent companies and their subsidiaries will be drawn given the intertwined and interdependent nature of their economic affairs.
The logical corollary of an “integrated” definition of investment is a corresponding consolidation of various economic roles into a single, broad conception of “investor.” In reaching its conclusion, the panel dismissed Mexico’s argument that the losses suffered by Cargill in the US were those suffered in the company’s capacity as producer and exporter and not as an investor in CdM, its Mexican investment.
The most interesting aspect of the decision—and perhaps the most troubling—is the way in which the panel distinguished the seemingly unequivocal (although not binding) findings of the panel in Archer Daniels. The Cargill panel found that up-stream losses were justified since CdM was not a producer of HFCS and thus depended on the HFCS sold to it by the parent company, unlike the investors in Archer Daniels whose investment took the form of a more extensive, joint venture production operation in Mexico.
If the oft-cited goals of attracting foreign direct investment are to increase jobs, technology and expertise transfer, and to capture more of the value-added production chain in host states, then why would more incentives be created for companies to produce and process within their home state and simply distribute to host countries? The panel’s understanding of how to implement NAFTA Chapter 11, in other words, is not consonant with the treaty’s underlying policy objectives. Or, at the very least, the treaty’s public justification.
The OCA’s Decision – The (in)correct framing of the reasonableness standard and an (un)reasonable application of the correctness standard?
Although the OCA distanced itself from importing domestic administrative law concepts into judicial review of international investment arbitration, it still found that correctness was the appropriate standard to review jurisdictional issues. What is unclear, however, is why the Court seemed to frame the correctness standard as being a lower one than reasonableness by suggesting, at para 51:
Any time the court reviews on the reasonableness standard, it undertakes an in-depth analysis of the reasoning and decision of the tribunal in order to decide whether the result is a reasonable one…Once a court enters into a reasonableness review, it is effectively considering the merits of the tribunal’s decision and deciding whether that decision is acceptable because it is reasonable, not because it was made within the jurisdiction of the tribunal.
However, adopting a reasonableness standard for the question of jurisdiction would simply require the court to ask: “was it reasonable for the tribunal to conclude that it had the authority to enter into this line of inquiry?” and not to conduct any substantive review of the merits of the case.
Nevertheless, having adopted a correctness standard, the Court was required to determine whether or not the tribunal correctly interpreted (and acted within) its jurisdiction. Article 34(1)(a)(iii) of the Model Law allows a Superior Court judge to set aside a decision of an international arbitral tribunal where “the award deals with a dispute not contemplated by or not falling within the terms of the submission to arbitration, or contains decisions on matters beyond the scope of the submission.”
Thus a tribunal’s jurisdiction is circumscribed by the scope of the parties’ submission, defined by the Court as consisting of: (1) the agreement of the parties; (2) the words of the relevant Articles from Chapter 11, and where relevant, from other chapters of NAFTA; and (3) any interpretation of those words subsequently agreed to by the NAFTA signatory parties.
In applying the standard, the Court found that the tribunal’s award of up-stream damages stemmed from its interpretation of damages arising from an investment and thus was within the scope of the submission and the NAFTA provisions. The Court gave two examples of scenarios where a tribunal would fail to meet this correctness standard: if it made a finding in relation to an investment in Brazil (i.e. a non-party to NAFTA) or in relation to a period of time which was not contemplated by the parties to the arbitration. The court’s posture, in other words, is a highly deferential one.
Both Canada and the United States appeared as intervenors in the appeal and supported Mexico’s position that the only compensable damages are those suffered in the territory of the Party where the investment is located and not losses suffered by the investor in its home business operation. A proper application of the correctness standard on the issue of jurisdiction would presumably give adequate consideration and weight to the statements of the parties to NAFTA as to the proper interpretation of the treaty (i.e. the third component of the scope of “submission” identified by the Court). The Court, however, simply found that these statements do not constitute “a clear, well-understood, agreed common position.” What would constitute a clearer common position, however, was not explained.
Overall, while the task of “navigat[ing] the tension between the discouragement of courts to intervene on the one hand, and on the other, the court’s statutory mandate to review for jurisdictional excess” is no doubt a complicated, politically sensitive one, the need to calibrate a sufficiently precise, conceptually sound standard for judicial review of international arbitration remains an outstanding project after Cargill.