This the seventh post in a series of posts commenting on the NAFTA renegotiation process. For Part I click here, for Part II click here, for Part III click here, for Part IV click here, for Part V click here, for Part VI click here.

  • The United States and Canada are continuing their meetings this week to try to hash out a bilateral deal in renegotiating NAFTA that would be not only satisfactory to both but also consistent with the bilateral deal arrived at by the United States and Mexico in late August. Reportedly, the three issues that stand in the way of a bilateral deal between the United States and Canada are: Chapter Nineteen (providing for review by binational panels of AD and CVD measures imposed by NAFTA signatories); Canada’s income and price support policy for its milk industry known as “supply management”;[1] and continuing exclusions from the scope of the treaty based upon “cultural” reasons. According to press reports, the United States wants Canada to agree to the elimination of Chapter Nineteen in NAFTA 2.0 while Canada regards the permanence of Chapter Nineteen as a “deal-breaker”. In this Post I explain why the merits that Chapter Nineteen might have today are grossly oversold.
  • Chapter Nineteen in the 1994 NAFTA is the direct descendent of Chapter Nineteen in the 1988 Canada-United States Free Trade Agreement (CUSFTA).[2] Reportedly, Canada would not have entered into CUSFTA had the United States not agreed to binational panel review, and the adoption of this mechanism required the personal intervention of Canadian Prime Minister Mulroney before President Reagan.
  • What binational panel review means is that decisions by national authorities to impose AD and CVD measures are reviewed by a binational panel; that is, an ad-hoc arbitration tribunal composed of individuals who are supposed to be trade remedy experts (not necessarily lawyers) from the two countries involved.[3] Crucially, this review is in terms of consistency with domestic law, and is available as an alternative to domestic judicial review. It is important to underscore in this connection that, while decisions by local courts can be appealed, in NAFTA there is no procedure for appealing the rulings of binational panels.[4] Unlike WTO dispute settlement (which is state-to-state), NAFTA panels are launched upon request by the affected companies themselves.
  • Canada’s rationale for seeking an alternative under CUSFTA to the review of AD and CVD measures by U.S. courts appears to have been twofold:
  • One, Canada was reportedly frustrated with the adoption by U.S. courts of the “deference doctrine” pursuant to the Chevron ruling in 1984.  In a nutshell, under the “deference doctrine” a court should defer to a legal interpretation by the relevant administrative agencies (which in AD and CVD investigations would include the U.S. Department of Commerce and the U.S. International Trade Commission) as long as such legal interpretation is based upon a permissible construction (that is, a permissible reading) of the provision involved.[5]
  • Two, the alternative of challenging AD and CVD determinations via GATT dispute settlement was of limited utility because GATT panel reports were only adopted if the losing party was in agreement that this be so.
  • In theory, bilateral panel review under CUSFTA should not have turned out all that different from judicial review because CUSFTA Chapter Nineteen, at Article 1904(3), required panels to apply the same standard of review and the same “general legal principles” that a local court would apply and, in the case of the United States, such legal framework would have included the “deference doctrine”.
  • Chapter Nineteen of CUSFTA was essentially transposed to Chapter Nineteen of NAFTA. NAFTA Chapter Nineteen, at Article 1904:(3), also requires panels to apply the same standard of review and the same “general legal principles” that a local court would apply. However, critics of NAFTA Chapter Nineteen have argued that the key difference between bilateral panel review and judicial review is precisely that panelists are inclined to misapply the proper standard of review.[6]
  • In my opinion, Chapter Nineteen of the NAFTA suffers from another four fundamental problems:
  • First, panelists often lack expertise on the subject matter (i.e., the application of AD and CVD law). It is not stretch to say that sometimes they even lack a basic understanding of the subject matter. Let’s take Mexico as an example. In Mexico the pool of individuals with extensive experience litigating AD cases comprises a dozen persons or so. In an AD investigation concerning a big-ticket import from the United States, all these individuals would be involved as counsel to the domestic producers, the exporters, the importers, and to the associations of domestic producers, exporters and importers. This means that, in the event of the imposition of measures and a subsequent Chapter Nineteen proceeding, Mexico’s true AD experts would be conflicted out from serving in the panel. In such circumstances, the individuals who typically would be appointed by Mexico as panelists would be experts on areas such as customs law, tax law, or investment arbitration, or academics who specialize on international trade law but have little or no background whatsoever on AD investigations. But things can get much worse: Individuals whose expertise is on financial law and even education have served as NAFTA panelists.
  • Second, even if a panel reviewing a measure is composed of AD experts, as a general rule this expertise is strictly circumscribed to local AD, and does not extend to the other country’s AD. This is only to be expected because there is no reason why AD experts should invest time and resources to gain expertise too on AD as practiced in a foreign jurisdiction. Of course, while AD in the NAFTA region has some common strands, since the legislation of WTO Members has to be consistent with WTO rules in general and WTO AD rules in particular, the similarities involved are not all-encompassing because the AD Agreement does not cover each and every aspect of a Member’s practice. By way of illustration, I once witnessed a Mexican panelist, with expertise on Mexican AD, chastise the U.S. Government at a hearing for not having considered applying the “lesser duty rule” completely oblivious of the fact that this approach is not provided for under U.S. law.
  • Third, since the NAFTA parties involved in a dispute get to appoint the panelists, they can “game” the system. For instance, the party whose action is being reviewed obviously would never appoint as panelist someone who is a known critic of its AD practice, because it fears pay-back time. Similarly, when the grounds for a panel are very poor, it makes sense for the exporting party to appoint panelists with no understanding of the subject matter because an utterly confused panelist is more likely to rule either way.
  • Fourth, the absence of an appeal mechanism makes poorly-reasoned panel decisions final.[7]
  • I am afraid the first three problems described above are irremediable because they stem from the nature of bilateral panels.
  • Crucially, if Chapter Nineteen were dropped from NAFTA 2.0, Canada would still have an alternative to judicial review via WTO litigation. Unlike how things functioned under GATT, the WTO system makes the adoption of panel and AB reports by the DSB virtually certain, because they can only be rejected by a consensus against adoption (which in practice does not happen since the winning party would block such consensus). Moreover, inaction by the losing party regarding implementing the recommendations arising from the panel and AB reports opens the door to retaliation by the winning party.
  • To say that the NAFTA track, unlike the WTO track, remains effective is not on point, because NAFTA panels can be derailed if one of the parties involved strategically fails to appoint panelists. This is not a theoretical possibility, as it has happened already.
  • If there are issues in the AD practice of the NAFTA parties that need be fixed and can only be addressed via Chapter Nineteen (because WTO litigation is not the right vehicle to do so), no one supporting Chapter Nineteen has taken the trouble to identify what such issues are.
  • For all the reasons above, one would hope that Canada abandons its position to regard the elimination of Chapter Nineteen as a “red line” that it will not cross in renegotiating NAFTA.

[1] Canada’s  “supply management”  system is similar to the European Union’s Common Agricultural Policy and involves controls on both domestic production and imports (via very high tariffs). It is also applied to a few agricultural products other than milk.

[2] The text of CUSFTA can be found at

[3] By an ad hoc arbitration tribunal I mean an arbitration tribunal that is not permanent. In the current NAFTA, binational panels are composed of five panelists. Each NAFTA party gets to appoint two panelists, “in consultation”  with the other party. The fifth panelist is selected jointly by the two parties. At least three of the five panelists should be lawyers. The system was the same under CUSFTA.

[4] NAFTA Article 1904(13) provides for an “extraordinary challenge procedure”  on a number of grounds, including the panelists having shown bias or having incurred in conflict of interest, as a result of which a panel decision can be vacated or remanded. This procedure, however, does not allow appealing the panel’s findings.

[5] See, for instance, ttps://

[6] See, for instance, J.F. Colares and John W. Bohn, “NAFTA’s Double Standards of Review”, Wake Forest Law Review, Vol. 42, Issue 1 (2007).

[7] For a critical assessment of the findings of NAFTA binational panels reviewing Mexican AD actions, see Jorge Miranda and Juan Carlos Partida, “Mexico: quasi-judicial review of trade remedy measures by NAFTA panels”, in Muslum Yilmaz, ed., Domestic Judicial Review of Trade Remedies, Cambridge University Press, 2013.

*The opinions presented in this Post are mine alone and do not represent in any way official views of King & Spalding or its clients

This page as PDF

Leave a Reply

Your email address will not be published. Required fields are marked *