Appeal Without Remand: A Design Flaw in WTO Dispute Settlement and How to Fix It
Under current dispute settlement rules, a favourable panel ruling may become inoperable if the Appellate Body lacks the factual basis to complete its analysis. Such an outcome is particularly detrimental to developing countries.
Imagine you are a poor developing country that mustered the political courage and financial resources to file a WTO complaint. Turns out, the WTO panel finds in your favour. Sadly, however, on appeal, the Appellate Body reverses the panel and, worse still, decides that it cannot come to any conclusion because of holes in the panel’s factual record. So, after (on average) one and a half years of litigation, you end up with empty hands. Does this sound like a nightmare for policy-makers from developing countries? Yes, it does. Is it possible under current WTO dispute settlement rules? Yes, perfectly so.
A Design Flaw Hurting Developing Countries
This nightmare scenario can, and has, happened because of a design flaw in WTO dispute settlement: the Appellate Body does not have the mandate to decide on factual questions, which sometimes means it cannot complete a case. Yet, at the same time, it cannot either remand a case back to the original panel, which sometimes forces a complainant to re-file a case from scratch. In a growing number of disputes the Appellate Body has thus left entire cases, or parts of cases, unresolved: in EC – LAN Equipment, it left the United States without an answer; the same happened in US – DRAMS, this time sending Korea home without any finding on whether or not there was a subsidy. Similarly, in EC –Sugar, the Appellate Body could simply not come to any conclusion under the Subsidies Agreement, thereby potentially withholding stronger remedies from the complainants Australia, Brazil and Thailand.
What does this mean for developing countries? The absence of a remand procedure can force developing country complainants to go through two full proceedings before they achieve any result. Given time and resource constraints, such re-filing of a case subsequent to the Appellate Body finding that it ‘cannot complete the analysis’ may simply be excluded. The money and/or human resources may just not be available for another round of consultations, panel proceedings and Appellate Body hearings.
Moreover, another one to two years of litigation may simply be too long to wait for relief: not to have access to crucial export markets (such as the EU or the US) during another one to two years may force developing country businesses to close shop.
So what should be done?
Three strategies are available: (i) enable the Appellate Body to more frequently ‘complete the analysis’ in the first round of proceedings; (ii) expedite the re-filing of disputes in a second round of proceedings within the current system, or; (iii) amend the Dispute Settlement Understanding (DSU) and introduce a remand or referral system. These strategies are not mutually exclusive and should, ideally, complement each other.
Early Resolution by the Appellate Body
First, panels and disputing parties should do their best to enable the Appellate Body (AB) to ‘complete the analysis’. If the AB can resolve the case, there is no need for a remand in the first place. Panels can contribute by making alternative findings, especially factual findings, thereby laying the groundwork for consequential legal findings by the Appellate Body. Panels should also refrain from exercising too much ‘judicial economy’ and at a minimum decide all claims that are necessary to give an effective resolution to the dispute (the so-called Salmon test on false judicial economy). Disputing parties, in turn, should be careful enough to request a ‘completion of the analysis’ before the Appellate Body, lest they be surprised by a reversal and subsequent silence. The AB itself can also facilitate early resolution of disputes. So that more disputes can be resolved before the Appellate Body itself, the latter should only refuse to complete the analysis in case the panel’s factual record is incomplete. If the AB carefully questions the parties at the hearing and potentially even thereafter (it could also make preliminary rulings), concerns of ‘novelty’ of the legal claims or due process should not arise, nor prevent the Appellate Body from deciding the case (jura novit curiae).
Finally, with a DSU amendment, the limits of WTO appellate review could be loosened by permitting the Appellate Body itself to make decisions on factual disagreements. Although this would take away the parties’ right to appeal, such right of appeal is not absolute and must be weighed against the benefits of a timely resolution of the dispute. No doubt, to have the Appellate Body itself decide the case would be much faster and simpler than having to go through a remand process, let alone a complete refiling of the case.
Expedite Re-filing
A second strategy to avoid the nightmare scenario of spending one and a half years on WTO litigation without any substantive outcome is to creatively use the current dispute settlement system to obtain an expedited second proceeding. One avenue is to resort to Article 21.5 implementation proceedings to obtain a completion of the analysis. Such proceedings should be completed in 90 days. This is exactly what the United States and New Zealand did in Canada – Dairy (Article 21.5 – II): subsequent to the Appellate Body’s finding in the first Article 21.5 proceeding where it could not complete the analysis, the complainants simply asked for a second Article 21.5 panel. One could imagine that a first Article 21.5 proceeding can also be used to complete elements of the analysis that could not be completed in the original proceeding. Yet, in those cases where the original proceeding did not lead to any finding of violation (as in EC – LAN Equipment), a complainant cannot invoke implementation proceedings under Article 21.5, as there is nothing to implement in the first place.
Another way to expedite a second round of panel proceedings, without an actual remand process, is for the parties to agree to shorten and/or simplify the re-filed proceedings: they could appoint the same panelists, agree to one round of submissions and one hearing only, and even do away with the interim review stage. Where the parties cannot agree to an expedited refiling, panels themselves, after mere consultation with the parties, can change standard working proceedings (not DSU provisions themselves).
A Remand or Referral System
A third strategy to avoid a complete refiling in cases where the Appellate Body cannot complete the analysis is, of course, to amend the DSU and to provide for a remand process. The original drafters of the DSU did not include a remand process mainly based on fears that it would extend the time limits (US argument) or make the proceedings too complicated (the view of many developing countries, even in respect of adding an appellate procedure in the first place). Other negotiators may have considered that a remand process was not necessary given that the Appellate Body itself can ‘modify’ panel findings. Because GATT cases were not usually factually complicated, such modifications could have been expected without requiring new factual findings. With more than ten years of DSU experience, it has become clear, however, that a remand would actually save time as opposed to a complete re-filing (rebutting the US time-related argument against a remand), that the added complexity of a remand is outweighed by the waste of resources and time linked to a complete re-filing (rebutting developing country concerns in respect of a remand) and, finally, that the increasing factual complexity of WTO disputes often makes it impossible for the Appellate Body to ‘modify’ panel findings without making new factual findings.
As a result, most people agree that the DSU needs a remand process. Yet, the devil is in the details: what should such a process look like? No less than four proposals are currently on the table of the DSU review negotiations (by Jordan, the EU, Korea and, jointly, by Argentina, Brazil, Canada, India, New Zealand and Norway). These proposals raise three core questions.
Who triggers the remand, when and what happens next?
First, who can trigger a remand? All four proposals put the right to request a remand in the hands of the disputing parties. The Appellate Body would not be able to remand at its own initiative. Some party control may, indeed, be wise so as to avoid an unnecessary extension of proceedings (if the complainant is happy with, for example, one finding of violation why waste the time needed for a remand in respect of a second claim of violation?). More specifically, since not completing the analysis on either a claim or a defense means that no conclusion either way can be reached on a particular claim, remand can only benefit complainants. Thus, in this author’s view, the right to request a remand ought to be reserved exclusively to complainants (not defendants).
A second thorny question is whether remand should occur before or after adoption of the first Appellate Body report where it was decided that the analysis could not be completed. The EU and the six country group proposals favour early adoption of the first Appellate Body report and would trigger remand only after such adoption. Korea, in contrast, would remand during the first Appellate Body proceeding and move to DSB adoption, of all reports involved, only after the remand is over. If the remand can be expeditious, Korea’s approach is preferable as it avoids the complications of two parallel implementation proceedings (one for the original Appellate Body report; a second for the findings on remand).
In respect of the third remand question, however, it is Korea’s proposal that is flawed: Korea would remand the case back to the original panel for it to make new factual findings only; after that, the dispute must in all cases be sent back to the Appellate Body for a completion of the analysis. This would unnecessarily extend the proceedings and take away the parties’ right to appeal over legal questions (as only the Appellate Body on remand, not the remand panel, would make legal findings). One of the core reasons for a remand process is, however, to preserve the parties’ right to appeal. Korea’s proposal would not serve that purpose. Instead, in this author’s view, a remand panel should decide on both the facts and the law, after which the panel report on remand must be adopted unless the parties appeal. That is also how the proposals by the EU, the six country group and Jordan envisage a remand process.
A carefully tailored and sufficiently expedited remand process – that respects the panel’s factfinding role and the parties’ right to appeal, and does not lend itself to delay tactics by defending parties – would clearly be simpler and less time consuming than a full-fledged second proceeding. It is in the interest of both developed and developing countries to amend the DSU accordingly. Failure to do so will inevitably lead to a waste of resources and time and, with sufficient cases left unresolved by the Appellate Body, may eventually undermine the credibility of the entire WTO dispute settlement system.
Joost Pauwelyn is Professor of Law at Duke University in Durham, North Carolina. This article is a summary and preview of his longer ICSTD-comissioned study on the subject.