The Precedential Value Case Law of the WTO Dispute Settlement Mechanism

by Zihan Liu*



The Dispute Settlement Mechanism (DSM) of World Trade Organization (WTO), which is one of the important pillars of the WTO, improves the security and predictability of multilateral trade. However, the US has raised the issue of the status of precedent within the WTO DSM. WTO dispute settlement reports carry high precedential weight in judicial practice, and subsequent practice can be departed from only if there is a cogent reason;[1] the US regards such departures as precedent. Therefore, they constitute a systemic issue in WTO reform. This paper contends that the nature of WTO dispute settlement reports should be regarded as jurisprudence constante, in lieu of precedent, providing strong guidance to the Appellate Body (AB) and the panel compared to case law, but have no binding force for the subsequent cases; this is from the French legal system and is quite different from common law system. At the same time, this paper contends that the incorporation of jurisprudence consistante helps to address the concern of the US while continuing to preserve the judicial independence of the AB.



The Statement by the US on the Precedential Value of Panel or AB Reports was submitted at the meeting of WTO Dispute Settlement Body Geneva on December 18th in 2018. It asserted that the AB broke the rules of the Dispute Settlement Understanding and the WTO Agreement, and gave the authority in precedent to its dispute settlement reports, especially the interpretation of the treaties involved.[2] The US criticised the practice of the AB of considering the interpretation of the treaties as a precedent that is binding on subsequent cases. It used as an example the case the US – Stainless Steel (Mexico) (2008), one of twelve cases from 2004 to 2016. In this case, the panel disagreed with the interpretation of the provision in the well-established case law of the AB, but it agreed with the interpretation set out by the US. In contrast, the AB believed that the panel should take into account the legal interpretation embodied in an adopted panel and the AB reports, and is not free to disregard the legal interpretation.

Therefore, the US contends that the practical effect of this case is to attach significance to the role of precedent. It mentioned that the AB should not treat its adjudication as a precedent, which would go beyond the authority granted to it by WTO members.



The panel in the case the US – Stainless Steel (Mexico) (2008) said that it had “no option but to respectfully disagree with the line of reasoning developed by the AB regarding the WTO consistency of simple zeroing in periodic reviews”[3]. The AB mentioned in its report that it was ‘deeply concerned’ about the decision of the panel, which differed from the well-established case law of the AB, noting that the panel’s decision had “serious implications for the proper functioning of the WTO DSM”.[4] The AB stated:

AB reports are not binding…. Ensuring ‘security and predictability’ in the DSM implies that, absent ‘cogent reasons’, an adjudicatory body will resolve the same legal question in the same way in a subsequent case.[5]

According to the report of the AB, the adopted reports control subsequent practice[6], unless there are cogent reasons to depart from this assumption. Thus, if there is no cogent reason to do so, the panel should not depart from the legal interpretation embodied in the adopted reports of the panels and the AB, and should have consistency when resolving the same legal question.

This prompts the question: Is there any way to flesh out the precise elements of “cogent reason”?

The AB has proposed the condition of “cogent reason” in US – Stainless Steel (Mexico) (2008), but has not specified what constitutes a “cogent reason”. The panel in the China – Rare Earths (2014) stated, however, that it “may be understood as referring generally to a high threshold”.[7] Later that year, in the US – Countervailing and Anti-Dumping Measures (China) (2014), the panel defined the concept and the scope of “cogent reason” in the following terms:

To our minds, ‘cogent’ reasons, i.e. reasons that could in appropriate cases justify a panel in adopting a different interpretation, would encompass, inter alia: (i) a multilateral interpretation of a provision of the covered agreements under Article IX:2 of the WTO Agreement that departs from a prior AB interpretation; (ii) a demonstration that a prior AB interpretation proved to be unworkable in a particular set of circumstances falling within the scope of the relevant obligation at issue; (iii) a demonstration that the AB’s prior interpretation leads to a conflict with another provision of a covered agreement that was not raised before the AB; or (iv) a demonstration that the AB’s interpretation was based on a factually incorrect premise.[8]

Although the definition of “cogent reasons” provides the possibility of not following a prior AB’s decision, the panel did not provide other useful information or discussion except these reasons; therefore, the scope of cogent reason remains unclear. Moreover, the AB has not previously defined  what constitutes a ‘cogent reasons’ to depart from prior interpretations , [9] and it remains unclear whether future panels will do so.



In my opinion, the statement of the US in which it criticised the WTO’s AB’s regarding its rulings as precedent lacks merit. The arguments of the US are refuted as follows.

First, some WTO members such as China and Brazil disagree with the US on this issue. After the US presented its critique, many WTO members took the floor to comment that there was no allowance under the DSU or the WTO Agreement giving the AB authority to issue rulings that set a binding precedent, and questioned whether the AB was asserting such an authority.[10] They also said that the “cogent reasons” pronouncement that the US claimed the AB was asserting in Stainless Steel was not correct, and that in fact, the AB had said in the same ruling that it “is well settled that the AB reports are not binding, except with respect to resolving the particular dispute between the parties”.[11] There is a difference between previous rulings setting binding precedent and previous rulings serving as guidance for the AB and panels.

Second, the US selectively criticised the precedent role of the AB.[12] If the US thinks that the AB carries precedential weight in US – Stainless Steel (Mexico) (2008), then it carries precedential weight at the beginning of the WTO’s dispute settlement process. For example, in US – Gasoline (Venezuela) (1996), the AB sided with the US claim and overturned the panel’s interpretation of treaties in which the panel found that the measures involved a violation of of GATT Article 20(g) by the US; it then created a new “two-step” analysis method in interpreting the Article.[13] This interpretation was affirmed and applied in US – Shrimp (1998),[14] which subsequent panels and the AB continued to cite frequently over the next two decades, and about which the US said nothing. Thus, it is fair to say that the critique of the US is selective.

Finally, there are many de facto precedents in WTO dispute settlement procedures when a judicial decision provides authority or prescriptive principles for subsequent cases with the same or very relevant legal issues. Almost every panel or AB report in the DSB published reports cites the prior reports of the panels and the AB. For instance, Japan – Taxes on Alcoholic Beverages (1996), US – Gasoline (1996), EC – Bananas III (1997), and US – Softwood Lumber (2001), among others, were cited in more than 40 reports concerning the WTO DSM.



(1) Benefits in the Role of Precedent

First, carrying precedential weight in a WTO dispute settlement report can fill in gaps left by WTO rules and ensure the continuity, stability, and predictability [15]of WTO rules, as an adjudicatory body is much more likely to resolve the same legal question in the same way in a subsequent case.[16] The rules of the WTO are more abstract, general, and even vague, as when a deal is badly needed for political reasons, but a substantive agreement cannot be reached, negotiators sometimes opt for a vague arrangement that permits conflicting interpretations. The implicit decision in such instances is that the matter can be addressed further in future disputes while providing the immediate appearance of the accord that is required for political reasons.[17] But it still needs specific interpretation when applying the rules of the WTO in a specific dispute, so that it can be consistent and exercisable. Regarding reports of the panels and the AB as precedents is to apply and make a reasonable interpretation of WTO rules, which can help the panels and the AB find guidance from prior cases.[18] This then allows them to correctly apply abstract and generalised WTO rules to the settlement of specific disputes and ensures the consistent interpretation and application of WTO rules.

Second, the role of precedent can strengthen the persuasiveness of the WTO dispute settlement reports’ interpretation and reasoning, making them more convincing to the parties and thus enhancing the enforcement of the WTO dispute settlement reports. As Edgar Bodenheimer writes in Jurisprudence: The Philosophy and Method of the Law, “The authority and power of the jurisprudence does not come from the jurisprudence itself, but what is hidden behind the jurisprudence.[19] The power that makes the jurisprudence effective is not the judge’s will or order, but the inherent advantages of the principle or the reality of a habit embodied in a jurisprudence.[20]

Either the principle or the reality of a habit can enhance members’ confidence in the process of solving disputes under the WTO DSM. WTO members can predict legal effects before taking actions and confirm their rights, obligations, and duties, which enhances the enforcement of the WTO dispute settlement reports.

(2) Disadvantages in the Role of Precedent

Although the use of precedent has some benefits, it also has a number of disadvantages that could erode the legitimacy of the WTO DSM.

First, regarding the WTO dispute settlement reports as binding precedents is unfair to developing countries.[21] Developed countries, especially the US, are actively participating in the WTO dispute settlement, and can thus apply the rules of WTO through precedents in a way that serves their interests. This mechanism has the potential to put smaller and poorer countries, which do not have the legal capacity to participate intensively in the system, at a disadvantage in the legal process. [22] Their participation is limited because numerous interpretations already made by the panels and the AB are treated as binding precedents that benefit the interests of developed countries. [23]

Second, giving AB reports strong precedential weight would propagate any judicial mistake that might occur in one case over time. This has been portrayed by reference to by Maggi and Staiger’s (2011) cynical views on the DSB’s mandate to maximise the joint welfare of the disputing parties. [24] Professors Beshkar and Chilton also agree with it. [25] If there are mistakes in precedents, it is possible for subsequent cases to follow them, which hinders the development of the DSM.

Third, regarding the report of the AB as a binding precedent will increase both the frequency and the cost of disputes. Although an emphasis on precedent could reduce the uncertainty of disputes, reducing their frequency, it could, paradoxically, also lead more-frequent disputes, as to make the precedent established in the dispute settlement report more favorable to a given party, that party will repeatedly interact. As disputes emerge that conflict with the precedent established in dispute settlement reports, the likelihood of litigation and cost of lawsuits, as well as of instituting any changes to prior agreements, may increase. Because judges have the power to create binding precedents for the development of law, it is a matter of course for them to change the relevant law to adapt to changing circumstances. In fact, judges can invite legislators to pass new legislation if they disagree with a certain decision in some cases.[26] In general, however, overturning the agreements they made before requires consensus. The result is that the cost of treating WTO dispute settlement reports as precedents may be higher than that of similar actions in domestic settings.[27]

To sum up, the use of precedent in the WTO dispute settlement mechanism contains more disadvantages. Thus, this paper proposes that the AB’s decisions could be regarded as jurisprudence constante, as explained below.



The notion of iurisprudence constante derives from civil legal systems, while the notion of ‘precedent’ connotes the case law formed by common law systems. [28]In the view of this paper, WTO dispute settlement reports by their nature are better considered as jurisprudence constante. The book Legal Theory, Sources of Law and the Semantic Web explains jurisprudence constante as “basically a weakened formulation of stare decisis: the longer a series of identical decisions in identical cases based on the same legal rule by the courts becomes, the more determinative the decision found in the series becomes for future cases”[29].

The WTO Analytical Index: Guide to WTO Law and Practices, which is regularly updated by the WTO secretariat, uses the title of jurisprudence and treats the interpretation involved as jurisprudence constante to guide subsequent cases.[30] In addition, the panel of Brazil – Desiccated Coconut (1995) said that panel reports are useful and persuasive guidance in relevant cases. The panel of Japan –Taxes on Alcoholic Beverages (1995) also said that although unadopted panel reports have no validity in GATT or WTO, panels can find useful references in the analysis of unadopted panel reports. Therefore, my opinion is that it is better to consider the nature of WTO dispute settlement reports as jurisprudence constante than as precedent.

Treating the interpretation of treaties as jurisprudence constante is not unique to WTO dispute settlement. In fact, some international adjudication bodies, such as ICJ and ICSID, do so. For example, ICJ emphasised in the judgment of Temple of Preah Vihear (Cambodia v. Thailand) in 1961 that “[i]n so doing, the Court must apply its normal canons of interpretation, the first of which, according to the established jurisprudence of the Court, is that words are to be interpreted according to their natural and ordinary meaning in the context in which they occur”[31]. ICJ used the word “jurisprudence” as the basic principle of treaty interpretation. Another example is the ICSID case Daimler Financial Services AG v. Argentine. ICSID mentioned in this case that

[w]hile the Tribunal is not bound to follow the example of prior tribunals, it can find no justification either in the text of the German-Argentine BIT or in general international law to depart from the overwhelming jurisprudence constante that has emerged around this particular legal question[32].

Ultimately, WTO dispute settlement reports are not binding precedents, nor can they increase or decrease the rights and obligations under agreements. Although panels and the AB can cite prior reports, they can do so only for guidance. Although prior reports are not precedents and the panels and the AB have no requirement to follow them in subsequent cases, they should be taken into full account in subsequent practice, and the panels and the AB should follow the consistency principle, which is the key point of jurisprudence constante. Although jurisprudence constante may have a precedent effect, it is it is, in fact, merely strong guidance for subsequent judgments.



Reasons for the US’ objection about the nature of WTO dispute settlement reports are insufficient, which could confirmed from the following three grounds: the comments from other members, the historic case in the US, and cases in the practice of the WTO. It is unnecessary to treat the AB’s decisions as binding precedents, as the use of precedent in the WTO dispute settlement mechanism contains more disadvantages.. Considering it as jurisprudence constante is a better course.


*Attorney-at-law (PRC Bar). LL.M, University of Edinburgh. ORCID: 0000-0002-2845-6013

[1] Appellate Body Report, US – Stainless Steel (Mexico) (WT/DS344/R, 2008) para.160.

[2] Statement by the United States on the Precedential Value of Panel or Appellate Body Reports (Geneva 2018) para 3.

[3] Panel Report, US – Stainless Steel (Mexico) (WT/DS344/R, 2008) para 7.106.

[4] Ibid. para. 162.

[5] Ibid., para. 158-160.

[6] D Palmeter and P C Mavroidis, “The WTO Legal System: Sources of Law” (1998) 92 AM.J. INT’LL 398.

[7] Panel Report, China – Rare Earths (United States) (WT/DS433/R, 2014) paras 7.59–7.61.

[8] Panel Report, US [FIGURE DASH] Countervailing and Anti-Dumping Measures (China) (WT/DS449/R, 2014) paras 7.317

[9] B Mostafa and C. Adam, “Revisiting Procedure and Precedent in the WTO: An Analysis of US – Countervailing and Anti-Dumping Measures (China)” (2015) 728 Coase-Sandor Working Paper Series in Law and Economics 11.

[10] Supra note 3.

[11] Ibid.

[12] J Bacchus and S Lester, “Of Precedent and Persuasion: The Crucial Role of an Appeals Court in WTO Disputes” (2019)74 CATO Institute 2.

[13] US v Venezuela 1996 DSR 22.

[14] US v India 1998 DSR 118-110.

[15] G Gilbert, “The Use of Precedent by International Judges ad Arbitrators” (2012) 2 JIDS 6.

[16] Supra note 9, at 15.

[17] B M Hoekman and M M Kostecki, The Political Economy of The World Trading System


[18] L Simon, “What Is the Precedential Value of Appellate Body Reports?” (2018) 3 International Economic Law and Policy Blog, available at

[19] E Bodenheimer, Jurisprudence: The Philosophy and Method of the Law (1962) 390.

[20] Ibid.

[21] Supra note 9, at 17.

[22] Ibid.

[23] Ibid.

[24] Panel Report, US-CVD and AD Measures (China) (WT/DS379/12/Add.7) para. 7.317.

[25] Supra note 9, at 17.

[26] D Palmeter and P C Mavroidis, “The WTO Legal System: Sources of Law” (1998) 92 AM.J. INT’LL 398.

[27] Ibid.

[28] Z Naigen, “The Precedential Value or Treaty Interpretation of Appellate Body” (2019) 2 International Economic Review 45.

[29] A Roer, Legal Theory, Sources of Law and the Semantic Web (2009), 246.

[30] WTO Analytical, WTO Analytical Index (Guide to WTO Law and Practice), 1995-2011 (June 2012).

[31] International Court of Justice Reports, Temple of Preah Vihear (Cambodia v Thailand): Preliminary Objections para 32.

[32] Daimler Financial Services AG v. Argentine, ICSID Case No. ARB/05/1, Award, 22 August 2012, para 34.