28 Aug The WTO Appellate Body and International Judicial Deference
One of the issues identified by Tomer Broude in his APSA virtual roundtable introduction is the allocation of rule-determination power between national bodies and international tribunals. He also notes the growing concern of judicial activism by international tribunals, with the WTO Appellate Body as the most notable example. He identifies an emerging international judiciary that represents an international level of governance that, while established by state consent, has the potential to exert normative power that is independent of the specific will of states. I would like to take those issues as the launching point for a discussion of how the WTO Appellate Body is acting in ways inconsistent with its mandate. While not prone to WTO bashing, I can’t help but conclude that in some respects the Appellate Body has displayed a capacity to overreach.
Let’s start with the premise that treaty negotiators wanted to draft a treaty that required international judges to confer broad discretion on the Member States’ reasonable interpretations of treaty obligations. Suppose further that the goal of an international dispute resolution process was to police extreme behavior by Member States, but to otherwise have an attitude of deference. In short, what if negotiators wanted to adopt something like Chevron deference? How would they do so? And would it work?
In effect, those are the questions that are raised with respect to one particular aspect of the WTO dispute settlement process. Unlike other WTO commitments, the Antidumping Agreement includes an interpretive mandate for WTO panels to defer to permissible interpretations of Member States. Specifically, Article 17.6(ii) of the Antidumping Agreement requires WTO panels
to interpret the relevant provisions of the Agreement in accordance with customary rules of interpretation of public international law. Where the panel finds that a relevant provision of the Agreement admits of more than one permissible interpretation, the panel shall find the authorities’ measure to be in conformity with the Agreement if it rests upon one of those permissible interpretations.
In essence, as the negotiating history makes clear this language incorporates administrative law principles analogous to Chevron deference, while also adopting traditional international law norms regarding treaty interpretation, i.e., the interpretive methodology reflected in the Vienna Convention on the Law of Treaties.
At least that was the intent. But something has happened in the dozen or so years that this provision has been applied. The WTO Appellate Body has highlighted the obligation to apply the Vienna Convention under the first sentence of Article 17.6, and concluded that the obligation to defer under the second sentence is only relevant if the WTO provision in question will admit of more than one interpretation. And, of course, the entire framework of Articles 31 and 32 of the Vienna Convention is structured to provide tools to reach the one best interpretation of an ambiguous treaty provision. The result is that WTO adjudicators have never concluded that the Antidumping Agreement will admit of more than one permissible interpretation. As Daniel Tarullo put in an article in 2002, “[i]t is difficult to identify any issue in any of the cases in which this special standard has produced an outcome different from that which would have prevailed had there been no Article 17.6.”
It didn’t have to be this way. The WTO could have adopted an approach similar to the European Court of Human Rights. Under the margin of appreciation doctrine adopted by the ECHR, national authorities are granted significant deference while preserving the authority of the European Court to give the final ruling on whether a Member State restriction is reconcilable with the treaty obligation. By incorporating the international law doctrine of the margin of appreciation, the requirements of deference in the second sentence of 17.6(ii) could have been viewed as fully consistent with the first sentence’s mandate to apply customary international law rules of interpretation. The WTO could have concluded that the first sentence of 17.6(ii) simply embraces an emerging interpretive approach in international law which requires some international tribunals to defer to national authorities in appropriate circumstances. In short, both the first and second sentences of Article 17.6(ii) could have supported an attitude of deference.
As I have written in a forthcoming article in the Columbia Journal of Transnational Law, (a draft of which is available for download here), one of the most notable instances of the WTO failing to apply Article 17.6(ii) came in a recent Appellate Body report involving the United States’ practice of zeroing. The subject is arcane, controversial and beyond the scope of this short post. But suffice it to say that it is a practice used to calculate the margin of dumping that has been in place for decades and that previous GATT panels have authorized. There is nothing in the drafting of the new Antidumping Agreement that suggests the practice is obviously prohibited.
Indeed, the WTO panel in US—Zeroing ruled that “we must conclude that [the Antidumping Agreement] does not consider zeroing to be unfair and thus prohibited in all circumstances, but rather recognizes that in some cases zeroing may be appropriate in order to accurately reflect the existence of dumping by an exporter.” But the WTO Appellate Body disagreed and concluded that the only permissible interpretation was one that prohibited the practice of zeroing. The Appellate Body’s interpretation may well be a permissible reading of the treaty obligation. And the result may well be the preferred policy outcome. But that is beside the point. What is critical in US—Zeroing is adherence to the appropriate standard of review. As I discuss in some detail in the article, if the WTO panel concluded that the text of the Antidumping Agreement compelled the result articulated by the United States, it is difficult to understand how the Appellate Body could reach the conclusion that there was not sufficient ambiguity in the text that the United States’ position was not at least one permissible interpretation of the treaty obligation. The Appellate Body appeared to focus on its understanding of the best reading of the treaty obligation, and ignored whether the Member State’s interpretation was a permissible one.
The WTO Appellate Body’s failure to adhere to the appropriate standard of review in cases such as US—Zeroing illustrates the perils of entrusting international tribunals with a mandate to defer to the reasonable policy choices of Member States. It suggests that it may be inappropriate for Member States to breezily assume that international tribunals will closely follow interpretive rules that seek to maintain Member State authority at the expense of enhanced international judicial power. It is, after all, more than a little odd that the European Court of Human Rights has developed an attitude of deference without textual support to do so, while the World Trade Organization has rejected an approach of deference despite the express obligation to do so.
One might go further and argue that the most effective international tribunals are ones that have rules that are capable of some flexibility when the political stakes are high. As one who is a strong supporter of the WTO in most instances, I might add that it does not profit the WTO for the Appellate Body to create brittle rules out of obligations that were intended to be supple.