16.1 Introduction
The present generation of dispute settlement mechanisms (DSMs) in trade agreements is not hugely varied in terms of institutional design, but that consistency is beginning to change. Today, a core feature of comprehensive trade agreements is a set of enforcement provisions designed to ensure that governments comply with the obligations they have undertaken. Unlike other areas of international law, state-to-state DSMs are well accepted and widely implemented in trade agreements. The fact that we speak of them as ‘mechanisms’ already suggests the advancement of their institutional form. They are no longer sole provisions or short references; in many instances, they create standalone bodies and occupy entire chapters of text. Moreover, trade agreement DSMs from around the world have converged in recent years towards an ad hoc adjudicatory process heard by a neutral panel of adjudicators. When the panel identifies a violation of the agreement, the agreement typically provides a further process for trade sanctions to induce compliance where needed.
Despite the proliferation of hundreds of trade agreements with dispute settlement provisions of some type, states rarely have activated those provisions and the cumulative number of formal disputes has remained low for many decades. As disputes have now appeared with greater frequency in the last five years, in part as a result of the World Trade Organization’s (WTO) own DSM decline, states are experimenting with new models of and approaches to dispute settlement, including some that empower other actors beyond the state in seeking to ensure compliance with trade agreement commitments. The moment is ripe to review the present evolution into new realms of disputes and to contemplate alternative designs to resolve conflict between states.
This chapter begins by briefly surveying DSMs that commonly appear in trade agreements. Many trade agreements include binding arbitration tools for the resolution of disputes among the parties, but not all commitments in the agreement may be subject to dispute settlement. Some trade agreements differ with respect to the legal result or outcome of the arbitration process. Agreements vary in the remedies they provide to the successful party. While a comprehensive study exceeds the bounds of this short chapter, reviewing the taxonomy is useful for understanding the landscape.Footnote 1 After mapping the terrain, the chapter then turns to the primary issues facing states and other stakeholders as they write new trade agreements and negotiate less orthodox mechanisms to address problems that arise in trade relationships. Finally, the chapter makes recommendations for future negotiations and for further research.
16.2 Mapping Trends
Much ink in international law scholarship has been spilled on questions of institutional design surrounding dispute settlement. In trade, commentators over the last forty years have praised the concept of third-party dispute settlement as a great achievement in our long-time sovereigntist discipline of international law. These are typically state-to-state mechanisms, although not exclusively so. We focus on state behaviour because one of the primary goals of international law is to shape that behaviour. That is true likewise in trade: commitments made in trade agreements seek to install reciprocal behavioural constraints. The primary mechanism for holding states accountable in trade is, like elsewhere, third-party dispute settlement – whether in the form of arbitration or through a standing judicial body. The culmination of that exercise is licence to one state to impose economic penalties (often in the form of suspension of concessions) on the party that has acted in breach of the agreement. States invoke the dispute settlement chapter as a means by which to ensure that other states are complying – in effect, to enforce the agreement. As this chapter shows, that story is changing, but state-to-state dispute settlement continues to be the norm, for now.
Scholars have identified different reasons as to why strong state-to-state DSMs have developed and why they have become more prevalent. The chapter mentions these in context below. One common rationale is that DSMs gained importance with the heightened demands of deeper agreements with wider membership (Allee and Elsig Reference Allee and Elsig2015). Thus, more elaborate DSMs emerged as trade agreements themselves became more elaborate along multiple dimensions.
16.2.1 Trends in Provisions
Nearly all (87 per cent according to the Design of Trade Agreements (DESTA) data set; Allee and Elsig Reference Allee and Elsig2015) comprehensive trade agreements negotiated in the last thirty years include some form of DSM.Footnote 2 Most modern trade agreements that contain binding and enforceable dispute settlement provisions, roughly half of the trade agreements overall, rely on ad hoc adjudication, similar to the WTO. But the institutional framework for dispute settlement is varied and the prevalence of arbitration is relatively new. This generation of trade agreements has been characterised by what I call ‘forum proliferation’: a shift in interpretive power from treaty parties to a plethora of ad hoc, independent panels unrestricted by precedent and unsupervised by any superior authority (Claussen Reference Claussen2018a). This section reviews the structural models across the cartography of trade agreements in brief and then turns to some recent highlights.
16.2.1.1 Common Taxonomies
There are several ways to categorise structural models of DSMs. Some scholars have undertaken to categorise them by region, while others have primarily organised them by type. Beginning in the early 2000s, political scientists and lawyers developed multiple taxonomies for this purpose. The WTO Secretariat reviewed the literature in 2013 (Chase et al. Reference Chase2013). For purposes of our institutional design review, it makes sense as others have done to divide DSMs into three broad groups: (1) political or diplomatic dispute settlement, (2) systems based on a standing panel, and (3) referral to an ad hoc arbitral panel (Porges Reference Porges, Chauffour and Maur2011). Of all the 727 trade agreements captured by the DESTA data set, 80 per cent have consultation mechanisms, 23 per cent have mediation mechanisms, 46 per cent have arbitration mechanisms, and 5 per cent have standing judicial bodies. But those total numbers are insufficient to understand the clear trend towards ad hoc arbitration processes in the last four decades of trade agreements.
Todd Allee and Manfred Elsig have captured not only this trend towards more ad hoc arbitration in trade agreement DSMs but also a tendency towards increased ‘legalisation’ through their textual analysis of DSMs (Allee and Elsig Reference Allee and Elsig2015). They consider stronger DSMs to exhibit greater legalisation which they measure along six dimensions. The first dimension tracks the Porges taxonomy and turns on the extent to which dispute settlement authority is delegated to a third party, legal body. On a three-tiered scale, the ‘weakest’ agreements are those lacking any provision for dispute settlement authority, the middle ground is those which rely primarily on ad hoc arbitration, and the ‘strongest’ are those that create a standing body. The other elements relied upon by Allee and Elsig include the presence of a choice of forum clause, the process for selection of the chair of the arbitral panel, time limits for dispute settlement, the availability of post-award sanctions, and the comprehensive scope of the dispute settlement provisions vis-à-vis the rest of the agreement. The following figures are drawn from the Allee and Elsig data as updated through 2023.Footnote 3
First, the DESTA data show more robust DSMs over time, and particularly beginning in the 2000s, led by the United States (US) but not limited to it. As Figure 16.1 reflects, up until the 1990s, there were in fact very few DSMs apart from judicial-styled arrangements or consultative mechanisms, with heavy reliance on joint committees – committees of the parties’ trade officials.
Figure 16.1 Strength of DSM over time.
Past research provides evidence that the most important predictor for DSM strength is depth of the agreement (see Allee and Elsig Reference Allee and Elsig2015). Put differently, the more ambitious or comprehensive the agreement, the stronger or more legalistic the DSM is on average. Over time, as trade agreements have grown deeper also DSM has increased in strength (Figure 16.2). And, as I discuss further below, the enforcement of trade agreement provisions on issues of social and environmental sustainability has catalysed this shift just as it has followed it. States have incorporated more sustainability commitments and gradually made them enforceable through the same DSM as the commercial provisions of the agreement.
Figure 16.2 Depths of preferential trade agreements (PTAs) and strength of DSMs.
Both these trends can be seen also when disaggregating the type of DSM. The following Figure 16.3 shows the diminution in agreements with no DSM at all, and likewise the small numbers in modern agreements of DSMs that have only consultative or mediation procedures. As can be seen, arbitration becomes popular as a form of dispute settlement beginning in the 1980s.
Figure 16.3 Types of DSM over time.
16.2.1.2 Regional Trends
Many trends in trade agreement design follow regional patterns as other contributions to this volume have shown. Dispute settlement is no different. The following Figure 16.4 shows the pathways towards arbitration across regions.
Figure 16.4 Types of DSM per region.
The DESTA data also show that on average intercontinental agreements have the highest degree of legalisation, followed by agreements in the Americas (Figure 16.5).
The US has led the way in making DSMs more legally elaborate. Nearly all US comprehensive trade agreementsFootnote 4 presently in force provide for ad hoc arbitration. In their most basic terms, these sorts of DSMs work in panels composed of three or five panel members with each side picking an equal number of arbitrators and agreeing on the chair. The panel sets a timetable in consultation with the parties that provides for the submission of written pleadings and a hearing, which is open to the public. The panel then provides a report to the parties with a decision as to whether the panel finds the responding party to have acted inconsistently with the agreement. The offending party is expected to come into compliance with panel decisions. If not, compensation, suspension of benefits, or fines are possible remedies.
Compared to the US, European Union (EU) trade agreements are less legalised as defined by the DESTA data because most of the older agreements were not subject to arbitration, but they have changed dramatically in recent years. With the exception of the European Economic Area (EEA), all EU trade agreements from before 2000 are based on a diplomatic approach to dispute settlement in which consultations are the key mechanism (Garcia Bercero Reference Bercero, Bartels and Ortino2006). These consultations occur within the framework of a joint committee. The EEA is the only agreement that provided for judicial enforcement procedures, modelled on the European Court of Justice (ECJ). The other agreements concluded by the EU before 2000 permit adjudication only upon agreement of the parties. Most of those intend arbitration, but the Association Agreement with Turkey is exceptional in that it allows the parties to choose the ECJ for disputes. Another exception is the Cotonou Agreement in that it foresees the possibility of multi-party dispute, and arbitration following consultations.Footnote 5
A fundamental shift in EU DSMs occurred in the conclusion of the agreements with Mexico (2000) and Chile (2001). This shift occurred in part at the urging of Mexico which had sought to adopt the North American Free Trade Agreement (NAFTA) model in its negotiations with the EU (Garcia Bercero Reference Bercero, Bartels and Ortino2006). These agreements use an arbitration model DSM, drawing heavily from the WTO Dispute Settlement Understanding (DSU). By this time, the EU was already a major player in the WTO DSM, and there was some growing dissatisfaction with the lack of effectiveness of the traditional diplomatic approach. Garcia Bercero explains that it was also during the negotiations with Mexico that leadership for the agreement within the Commission was allocated to the Directorate General for Trade, which was ‘institutionally more inclined toward procedures that ensure a rapid solution of commercial disputes’. All EU agreements concluded since that time have included similar dispute settlement procedures. Most recently, the trend has been to expand the topical coverage of EU trade agreement DSMs. The European Commission’s New Action Plan introduced a shift towards judicialisation of sustainability provisions in 2022. This model is being introduced for the first time in the EU–New Zealand free trade agreement (FTA).
A similar evolution can be seen among the European Free Trade Association (EFTA) countries: Norway, Iceland, Liechtenstein, and Switzerland. The original EFTA Convention (1961) employed a consultation approach to disputes, although it incorporated the possibility for an examining committee to make recommendations to the EFTA Council. By the 1990s, however, sentiments towards dispute settlement had changed and an arbitration procedure was introduced in 2001 as a backup to bilateral consultations and negotiations (Ziegler Reference Ziegler, Bartels and Ortino2006). Likewise, in the agreements concluded by the EFTA states with other countries, they sought to include arbitration clauses from the late 1990s forward, and even renegotiated some that did not include such clauses. Thus, many of the EFTA agreements today set out a path to consultation within a joint committee followed by mandatory arbitration.
The pattern continues with the story of Mercosur, the Southern Common Market. Mercosur has had a DSM since its inception, but introduced a panel system only in 1993. It included three stages: consultations, intervention by the Common Market Group, and arbitration before an ad hoc panel. In 2004, the parties introduced the Olivos Protocol, which added a permanent review panel to hear appeals from panels. Other trade agreements in the region follow a similar approach. For example, the Andean Community has a system which includes automatic third-party review and which grants treaty signatories and private actors the right to initiate disputes. Procedures to settle disputes among Member States over non-compliance with regional norms can be heard by a secretariat and ultimately by the Andean Panel of Justice.
Most Indian PTAs have followed the same trend from political DSMs to, by the early 2000s, arbitral models (Singh Reference Singh2022). The latter seek to emulate the WTO DSU, following consultations or direct negotiations. Some Indian agreements permit a complaining party to initiate arbitration directly after consultations, while others require the party to first refer the dispute to a committee. Further, in some agreements the committee referral step is optional whereas in others it is mandatory.
As the DESTA data show, African agreements exhibit the least legalisation, together with agreements from Oceania, but this may be changing. An important feature of the recent African Continental Free Trade Area (AfCFTA) is its Protocol on Dispute Settlement. Unlike the majority of the African regional economic community courts that are modelled after the ECJ, the AfCFTA DSM tracks the WTO model.
Asian agreements reflect a high degree of legalisation, even if there have been no known adjudicated disputes among them. The Association of Southeast Asian Nations (ASEAN) agreements followed the same trajectory seen elsewhere. Initially, the ASEAN agreements provided for relatively informal dispute settlement, but by the late 1990s they had instituted a panel system with the possibility of appealing a decision to the ASEAN ministers. The ASEAN DSM covers all ASEAN economic agreements and the reports from panels and the Appellate Body are binding (Luo Reference Luo, Bartels and Ortino2006). Even though Asian agreements include adjudicatory mechanisms, they often heavily emphasise cooperation and consultation may explain the lack of adjudication. For example, the Regional Comprehensive Economic Partnership (RCEP) chapter on dispute settlement promotes flexibility, consensus, and the avoidance of formal dispute. There is thus some initial evidence that for some Asian PTAs, legalisation has declined.
Not all Asian agreements have adopted legalistic mechanisms. Agreements from China are very much mixed – with some including complex dispute systems and some not. Indonesian agreements with other developing countries often have no formal DSM, which scholars have seen as reflective of a general reluctance among developing states to embrace legalistic institutions (Widiatedja Reference Widiatedja2020).
Several scholars have investigated the sources and consequences of the widespread move towards legalistic mechanisms for resolving disputes in international trade agreements. Gomez-Mera and Molinari (Reference Gomez-Mera and Molinari2014) sum up these reasons and note also some overarching trends. For example, they show that democracies are more likely than autocracies to prefer moderately strict DSMs. They also demonstrate that many countries have been motivated to formalise their systems given the perceived success of the WTO. Sometimes WTO disputes have prompted the inclusion of a formal DSM in a trade agreement, such as in the case of the trade agreement between Indonesia and Australia. In other circumstances, some legalised DSMs have been the result of colonial relationships. A further reason is the simple copy-paste, repetition effect that we have seen in the substantive commitments also (Claussen Reference Claussen2018b; Allee and Elsig Reference Allee and Elsig2019).
16.2.1.3 Additional Features
Formal disputes are ‘not just the tip, but the tip of the tip of the iceberg’ (Melillo Reference Melillo2019). Dispute settlement apparatuses can provide an extensive institutional architecture apart from the actual panel, or consultation, system. Trade agreement DSMs and other chapters typically establish multiple committees. Margherita Melillo has shown that trade agreements, especially those of the EU of the latest generation, have many subcommittees, and evidence suggests that parties are addressing potential dispute issues there (Melillo Reference Melillo2019).
Some agreements also exclude individual topics from their DSMs. Scholars have catalogued these exclusions and their evolution over time (e.g. Froese Reference Froese2016). By one count, more than half of trade agreements with arbitration mechanisms exclude at least one subject from their DSMs (Chase et al. Reference Chase2013). Most often excluded is competition, followed by trade in services, and then sanitary and phytosanitary (SPS) measures. Generally, however, states and civil society actors have pushed for bringing more topics into the main DSM rather than the other way around, at least until recently as discussed further below. While agreements have for many years provided special procedures for certain topics, and even for certain trading partners (Froese Reference Froese2016), the broad shift towards including trade-plus commitments such as labour and environment provisions among the set of binding and enforceable obligations to which the DSM applies has begun to fade. Recent trade agreements have also diminished some of the offerings of past agreements or otherwise departed from the strict procedures of the WTO DSU in what they permit such as the RCEP mentioned above.
A further and often-discussed matter is the relationship between trade agreement DSMs and the WTO DSM. Some agreements include ‘choice of forum’ clauses permitting parties to choose, while others rely on exclusive jurisdiction clauses. Twenty-seven per cent of agreements in the DESTA data set include choice of forum-type provisions. For example, the NAFTA included a preference for NAFTA resolution of disputes involving environmental, SPS, or standard-related measures and even provided that a responding party may insist that a WTO complaint be withdrawn. But this provision of the NAFTA is rather exceptional. Other disputes under the NAFTA were subject to a choice of forum clause where the complaining party could elect whether to constitute a WTO panel or a PTA panel. That practice is more common: of the agreements that discuss forum choice, 97 per cent restrict the use of the other forum when one has been selected by the claimant or jointly. Other disputes under the NAFTA were subject to a choice of forum clause where the complaining party could elect whether to constitute a WTO panel or a PTA panel. In some instances, parties have sought review of the same measure under different agreements which has led to concern about re-litigation and an abuse of forum shopping privileges (Hillman Reference Hillman2009). Panel members have queried states about the option of consolidation in litigation, but no such mergers have yet occurred. Some agreements, such as the EU–Chile Agreement, include clauses that express a preference for using the WTO where the breach is ‘equivalent in substance to a WTO obligation’.
16.2.2 Trends in Litigation
Scholars have queried why there have been so few disputes under trade agreements outside the WTO. Before 2020, fewer than twenty state-to-state disputes under trade agreements were known to have reached a final decision by an arbitral panel. Since 2020, however, more than a dozen cases have appeared on trade agreement dockets for a variety of reasons. One likely contributing factor is the WTO Appellate Body’s effective suspension, but not all disputes that have been litigated under FTAs in the last three years could have been brought to the WTO. Some parties to these recent disputes are not WTO Members. More importantly, most of the disputes that have materialised have arisen under obligations that are unique to the trade agreement in question. For example, a case between the EU and Korea arose under the sustainable development chapter of their trade agreement. The several United States–Mexico–Canada Agreement (USMCA) cases now ongoing or recently concluded were the result of heightened rules or special disciplines found in the USMCA.
A further rationale for the increase in litigation is that of demonstrating strength among trade agreement enforcement. This is especially true in the US. In the absence of support for the WTO dispute settlement system, the US has actively pursued enforcement under the USMCA to demonstrate the utility of that system in the face of scepticism. Even those economies that continue to use the WTO dispute settlement system regularly recognise the limitations of that system in light of the absence of the Appellate Body. They have developed new means for carrying out enforcement and new ways to invigorate their trade agreement engagement. A good illustration is the institution of the EU single entry point which provides a means for members of the public to bring issues to the attention of the European Commission. Some commentators have also suggested that litigating under trade agreements outside the WTO may be more politically desirable. They may receive less public attention – although that has varied – and they may appear to be less confrontational outside the presence of all other WTO Members. Further, remedies may be especially crafted for the context of a trade agreement in ways not possible at the WTO.
The increase in litigation has begun to give rise to new forms of challenges that future designers will need to consider. One is the interpretive difficulty facing panels where certain language is repeated in other agreements and parties may urge panels to take account of other decisions interpreting the same language. Agreements may benefit from ‘interpretation selection clauses’ to guide the contours of interpretational jurisdiction for decision-makers, particularly with respect to the appearance of the same terms elsewhere. The EU–Canada Comprehensive Economic and Trade Agreement, for one, directs panels as to how to consider WTO case law. Even WTO-accommodating provisions can be subject to debate, however. These provisions do not offer guidance with respect to the decisions of other trade agreement panels – guidance that may soon be needed. Under current conditions, a panel may or may not seek to reach consistent outcomes and states have no way to predict how an individual panel will proceed in this respect. One useful option may be for states to develop rules like in conflict of laws to govern the relationship between different systems (Claussen 2018). The New Zealand–Brunei Side Agreement to the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP) expressly provides for how arbitrators should treat other decisions (saying other agreements with the same language should be interpreted consistently).
Whereas the other chapters in this book speak to normative development in areas of commitments made by states, DSMs raise the stakes still more by outsourcing that development to panels of adjudicators. More attention ought to be paid to the argumentation, the sources, the outcomes, and the impacts on other areas of trade law that will soon evolve. Many recent disputes are in the public eye, which makes them both politically important and also legally relevant in ways that may be undercounted as yet. This is partly due to the fact that scholars tend to segregate trade and trade disputes from other areas of international law, but it need not be that way. There are lessons to be learned in both directions and more cross-pollination would be beneficial.
16.3 Novelties
This section turns to the recent specific advances in the development of trade agreement dispute settlement design. After some early experimentation, states have adopted, or considered, how to improve the dispute settlement experience – sometimes in response to popular pressure. Space does not permit a full review of novelties in individual agreements. Rather, I highlight a few movements with wide application. This section takes up six such areas of review: rules of procedure and evidence; transparency and accessibility; panel member requirements and selection; public participation; the role, if any, of a secretariat; and, means through which disputes end.
16.3.1 Rules of Procedure and Evidence
States in recent years have included rules of procedure in their dispute settlement chapter, or, more often, they have mandated the later development of such rules. These rules set out expectations regarding requirements for submission of pleadings, evidence and arguments, deadlines, conduct of hearings, language, translation and interpretation, among other topics. But the rules that parties have developed thus far sometimes have not covered all disputes, or all issues that have arisen in disputes. For example, arbitrators have on occasion noted that their disputes were underserviced by rules developed by the parties (Boisson de Chazournes and Lee Reference Chazournes and Lee2022).
Scholars likewise have called for rules of evidence to help manoeuvre new types of documents and unusual-to-trade forms of exhibits (Claussen and Wu forthcoming Reference Claussen and Wu2019). In a dispute between the US and Guatemala, the panel highlighted the ‘difficulty’ of the task before it, given the absence of rules to provide for the handling of evidence, the appearance of witnesses, and the testing of evidence through examination. In the absence of those additional tools, the panel decided to attribute little to no weight to several redacted statements, concluding that they were not credible because they did not have sufficient ‘indicia of reliability’ that the panel developed. Thus, although the panel admitted the statements, in some cases, the panel did not attribute any weight to certain of them because they were not accompanied by corroborating evidence or had insufficient detail or lacked precision. In direct response to the outcome of this dispute, the US insisted on the inclusion of rules of evidence in the USMCA.
Rules of evidence and procedure become more important as trade disputes evolve to cover new areas of regulatory cooperation and social policies. Our trade systems to date do not yet present a vision for how to resolve disputes that may not lend themselves to ordinary evidentiary principles or obvious presentation. A live question for DSM designers is how to develop a system for resolving disputes about trade problems that cannot be easily proven (Claussen Reference Claussen2019). These are not new problems, but they will become increasingly acute as these new DSMs face their first legitimacy tests.
A further procedural innovation that scholars have highlighted but which has not yet received sufficient attention by parties is that of the purpose and opportunity presented by interim review. Most trade agreements with arbitration DSMs, and 20 per cent overall, include an interim report stage, in which an initial draft of the panel’s report is presented to the parties prior to its publication. The parties comment on the initial report and the panel then pens a final report which is made public and serves as the ultimate decision and conclusion of the proceedings. But both in the WTO and in trade agreement disputes, states have given limited guidance to panels on the purpose of this intermediate stage, what changes the panel might make, and what the purpose of revision might be. Panellists and states have thus brought different approaches to bear on the interim process, and those divergences make the interim review ripe for further consideration and innovation in future designs.
16.3.2 Transparency and Accessibility
Trade agreements vary greatly in how much transparency is required by their DSMs. Since 1998, the US has pushed for much greater transparency in the WTO as well as in trade agreements, advocating that all panel submissions be available to the public and that panel hearings be open to public observation, except where there is a need to protect confidential business information. For example, the US has required all dispute documents to be made public within ten days of submission. In contrast, other economies like Canada or among ASEAN Members have authorised but not required their parties to make their own submissions public. US agreements also call for public hearings, though the degree of publicity has varied in the few disputes that have materialised. Some disputes have been webcasted via video livestream, whereas others offered a public hearing room in the building of the hearing, and still others provided an audio livestream. These provisions surrounding the public nature of the hearing and the location of the hearing, among others, are reflective of a general push towards greater transparency at least among some governments.
16.3.3 Arbitrators and Their Selection
Another area of recent innovation and of evolution is that of the arbitrator selection process in ad hoc arbitral dispute settlement. In nearly all instances, each of the two parties in a bilateral dispute appoints one arbitrator. Then agreements use at least four types of different appointment processes for the appointment of the chair: 5 per cent of trade agreements leave it to the parties to agree on the chair of the panel; 4 per cent leave it to the party-appointed arbitrators to select a chair; 18 per cent provide for selection of the chair by lot from a list; and 17 per cent provide for selection by an individual or organisation external to the dispute, usually a specific official of an intergovernmental organisation.Footnote 6
Some early ad hoc arbitration mechanisms did not appropriately safeguard from the possibility of a responding state blocking the constitution of an arbitral panel. For example, the NAFTA suffered from this flaw and allowed the parties to halt the dispute settlement process entirely. From the late 1990s until its termination in 2020, no NAFTA panels were formed (Lester et al. Reference Lester, Manak and Arpas2019). By contrast, in WTO disputes, the Director-General makes appointments where the parties cannot agree on the panellists. Only 17 per cent of trade agreements outside the WTO provide a mechanism for a third party to appoint panellists in the absence of agreement by the parties. That third party may be the WTO Director-General, or in other instances, another prominent leader of an adjudicatory body such as the Secretary-General of the Permanent Court of Arbitration or the President of the International Court of Justice. The latter approach is common among investor–state dispute settlement, as well.
Among the DSMs that feature ad hoc arbitration, some trade agreements now include rosters of potential arbitrators who the parties consider to be qualified and willing to serve so that the dispute settlement process can commence as soon as it is required. The practice of maintaining rosters also facilitates a ready list of individuals with specialised expertise. For example, the EU maintains a specialised roster of individuals with backgrounds in sustainable development for potential sustainable development disputes. The creation of specialised rosters coincides with parties’ insistence on specialised expertise for those types of disputes, a development that the US–Guatemala dispute precipitated (Claussen Reference Claussen2020). Generally, however, agreements that speak to qualifications of arbitrators typically limit those requirements to expertise in the relevant field and nationality, where they say anything at all.
These rosters can be either open, meaning that disputing parties may choose any panellists they wish including from the roster, or closed, meaning that the parties are constrained by the roster. The latter is rare – one example is the EU–Chile Agreement – but even among open lists, there is mixed practice among party choices. For example, in the USMCA disputes to date, all but one panel member was named from the roster. Additionally, most agreements prescribe rules to decide which panellists shall serve as an arbitrator. Apart from any requisite qualifications, several agreements constrain party choice on the basis of nationality (Chase et al. Reference Chase2013).
Agreements vary in the number of arbitrators that may serve on any given panel. All of the disputes except one that have resulted in a panel report since 2017 have involved three panel members. The dispute between Mexico, Canada, and the US concerning automotive parts under the USMCA involved five panel members, consistent with the requirements of the agreement which specifies that in case of three parties – two complainants, one respondent – five panellists are to be selected. In fact, under the USMCA, the default panel size for a bilateral dispute is also five unless the parties choose to have three panellists. Some PTAs, particularly Indian FTAs of recent years, provide for just one arbitrator, in case the second party refuses or is unable to nominate one (Singh Reference Singh2022).
Panel members have included retired trade officials, adjudicators, lawyers in private practice, and academics. Most panel members serve in a part-time, hourly capacity for a particular dispute apart from their full-time jobs elsewhere. With the rise in the number of disputes, a legal community is developing both with respect to counsel and with the arbitrator pool. Likewise, industry counsel are now increasingly engaged in trade agreement disputes. In the near term, scholars ought to study these sociological developments more closely.
Finally, more work is needed to understand the codes of conduct and ethical rules that apply in the trade agreement space. Most DSMs with active disputes require panellists to sign and commit to a code of conduct which relies heavily on disclosures and confidentiality. No publicly known challenges to arbitrators’ commitments have come to light, and this system appears to function with sufficient confidence from the parties. Most arbitral DSMs split panellist costs evenly among the parties to the dispute which also lends legitimacy to the arbitration arrangement just like nationality requirements do, although there are some notable exceptions (Chase et al. Reference Chase2013). Under the RCEP, for instance, each party pays costs of the panellist that it appointed.
16.3.4 Public Participation
Some of the recently concluded trade agreements include novelties that alter in small and large ways the nature of litigation that could arise. For example, both the USMCA and the CPTPP provide the opportunity for non-governmental entities (NGEs) to apply to the panel for permission to provide a written submission in the proceedings. To be granted such permission, the NGE based in the territory of one of the parties must identify specific issues of fact and law that the NGE will address in its submission and explain how the submission would assist the panel in the determination of the factual or legal issue related to the dispute by bringing a perspective or knowledge different from that of the participating parties. Non-governmental entities have made submissions in several of the disputes heard by panels in the last three years. Most of these NGEs have been industry associations supporting the complaining party’s position. Panels have granted leave liberally and permitted these NGEs to submit but they also have, in some instances, rejected an NGE’s request.
Likewise, US agreements and now EU agreements, among others, include public submission opportunities to trigger a review by the government into non-compliance by a trading partner. For example, under the NAFTA and now under the USMCA, the governments of North America provide contact points for members of the public to provide information about the practices of the other North American parties in the areas of labour and environment. The US Department of Labor has received numerous such submissions, though none has led to a state-to-state dispute settlement proceeding, apart from the one petition submitted under the auspices of the FTA among Central America, the US, and the Dominican Republic (Claussen Reference Claussen2020).
16.3.5 Secretariats
Few trade agreements concluded in the last forty years make reference to any permanent institution for dispute management. Secretariats are typical only where agreements provide for a standing court. Those courts require legal and administrative support. One example is the Andean Court of Justice which is to be given ‘all necessary facilities for proper fulfilment of its functions’, and is required to ‘appoint its necessary Secretary and the necessary personnel to perform its duties’. These agreements with standing bodies also will usually have a permanent budget process that involves substantial contributions by the parties.
Apart from standing bodies, only some trade agreements provide for the administration of state-to-state dispute settlement through a quasi-ad hoc secretariat with limited authority. These trade agreements require each party to maintain a government office, separate from the office representing the government in the dispute, to be ready to manage a case should one arise. In US trade agreements, the office of the responding party to the dispute picks up this role. The staff are charged with providing administrative assistance to the panel, serving as the panel’s contact point, arranging payment, organising and coordinating logistics, and performing ‘all other tasks’ as established under the agreement, the rules, or by the parties. These tasks may include those of management of document exchanges and hearings; coordination of any roster; secretarial, translation, and interpretation services; provision or rental of a place to hold hearings; research and drafting assistance to panellists; payment of panellist fees and expenses; information services; and capacity building. A similar model establishes a standing secretariat comprised of national sections across the parties. The national sections collaborate and may cooperatively provide administrative assistance to panels. The 2004 ASEAN Protocol on Enhanced Dispute Settlement Mechanism provides a clearly defined role in dispute settlement for a secretariat and is similar to the Mercosur Administrative Secretariat in that way.
Other approaches to dispute settlement management include those that permit panellists to hire assistants to provide research assistance as a complement to a separate office that provides logistical support. In recent practice, panellists have not often hired assistants. Still, other dispute chapters are silent entirely on assistance to panels.
States generally remain reluctant about setting up a robust secretariat given the potentially heavy financial burden they create, particularly for bilateral trade agreements where the costs are shared by only two parties. Questions have arisen as to who bears the cost of the secretariat; this was a debated question in the Trans-Pacific Partnership Agreement negotiations, for instance. Likewise, where a trade agreement relies on the parties to manage the administrative functions of a panel, parties have debated whether the complaining party or the responding party should bear that burden. In US trade agreements, for example, the burden falls on the responding party.
To address these issues, in future agreements, states could consider setting up specialised secretariats or elaborating guidance for ad hoc staff. Such changes could (1) better facilitate the management of the dispute; (2) clarify and delineate roles and responsibilities; and (3) likely yield financial savings to the extent those staff can assist with management issues presently imposed upon the panel members with their high hourly rates. Another alternative is for parties to add to the agreement an option for them to use an existing institution, such as the Permanent Court of Arbitration in The Hague, as a case manager for the dispute, much like what is done in trade agreement investor–state dispute settlement chapters. Although relying on a professionalised secretariat to manage a dispute comes at a cost, some secretariats may be willing to lower their usual rates to accommodate state needs.
16.3.6 The End of Disputes
Trade agreement disputes also provide an opportunity to consider alternative endings for disputes rather than simply a panel report with a pronouncement on breach. It is too early to tell just how well traditional outcomes serve parties’ interests, but we know that states are experimenting somewhat with their pathways to an outcome. After more disputes reach their conclusion, further research will be required to determine whether states are changing their measures in response or seeking alternative solutions.
Many agreements provide an option to the parties to agree to good offices, conciliation, and mediation anytime during the lifetime of the dispute. The DESTA data also tell us that as deeper agreements evolved, the parties also built in off-ramps. They made more references to mutually agreed solutions and made clear that they could stop litigation at any time they wished to solve the issue through negotiation (see Figure 16.6).
Figure 16.6 Strength of DSMs and references to MAS.
Another important feature of trade agreement DSMs is their short timeline. US agreements, for one, demand that a panel issue a final report within 150 days, or 120 days in the case of perishable goods, absent agreement by the parties. These timelines facilitate quick turnarounds on measures in question unseen in other fora.
I do not take up remedies and compliance with dispute settlement decisions here, but further research is urgently needed on that topic as trade agreements have begun to innovate in that area and many go beyond the WTO model. For example, some trade agreement DSMs provide for the possibility of provisional measures.Footnote 7 Some provide for financial compensation.Footnote 8 Approximately half of trade agreements in the DESTA data set provide for retaliation in case of non-implementation.
16.4 Future
As this chapter has described, the trend of avoiding trade agreement disputes has waxed and waned. More topics in trade agreements have been made subject to DSMs – having been brought into the ‘ordinary’ process – just as more economies adopted a more legalistic procedure; yet, few disputes have been brought. Only in the last three years have trade agreement disputes begun to emerge with any consistency. Those disputes have yielded some positive outcomes in that states engaged with the reports and have worked within their contours, so far as public information depicts. The growing body of jurisprudence will soon form the foundation for disputes of the future, as this section takes up. The reports issued by panels in these early years have the potential to shape what will follow as states treat them as sources of authority or otherwise rely on them for their policy choices.
Ironically, with trade agreement dispute settlement now beginning to normalise, states have turned their attention to the extent of dispute settlement’s reach, pushing the boundaries of what may be possible to ask a trade agreement to resolve. This section turns to the latest innovations in trade agreement dispute settlement: first, alternative styles of dealmaking, and sector-specific dispute designs; and second, disputes beyond the states, and specialised dispute tools.
16.4.1 Next-generation Deals and Their Disputes
The most important move by states in 2022–2023 has been the move away from comprehensive trade agreements entirely. In these US-led next-generation agreements, governments are unlikely to include binding commitments, and without binding commitments, enforcement mechanisms are unlikely. For example, more than a dozen countries are presently negotiating a new Indo-Pacific Economic Framework which will be ‘self-enforcing and incentive-based’ (Inside U.S. Trade 2022). This could mean ‘unilateral’ enforcement instead of traditional state-to-state dispute settlement, which, as some have noted, risks power plays and questions of durability (Lester Reference Lester2022). This is not necessarily an aberration, but it remains to be seen if large club-like systems will replace comprehensive trade agreements in other parts of the world, with actors other than the US. In such systems, enforcement is likely to take on a softer form.
The US has also led the charge on sector-specific deals, including product-specific deals such as in the area of critical minerals. Again, this is not a new trend, but it is one that could pick up speed in upcoming decades as narratives about free trade falter. Governments have for many years negotiated smaller trade-related deals. The US has more than 1,200 trade-related agreements with more than 100 countries that regulate everything from cross-border tomatoes to trousers to titanium. These are not insignificant. So far, few of these deals maintain any sort of DSM, but that too could depend on their upcoming direction as well as their linkage to existing agreements. If future trade-related agreements are not about market access, or are limited to individual issue areas, even if they do lead to disputes, these might look different from the traditional sorts of disputes and, if so, they may require distinct sorts of mechanisms. We may see less arbitration and more expert determination, for example. Indeed, we already have.
In these times of institutional diversity, states may wish to consider incorporating into future trade agreements graduated menus of dispute settlement options. Some states have sought to do so, building in systems that offer several routes to get to a result in a dispute, such as the environmental provisions of the USMCA. There, parties may invoke the DSM, go to a committee, hold Ministerial consultations, proceed to a panel, and include public submissions, among still other options. Diversification in DSMs is an appropriate way to manage diverse dealmaking. Governments ought to continue to think broadly about paths to resolution that are non-traditional and get the job done.
16.4.2 Disputes beyond the States
Another trend is now on the rise as states have turned attention to the potential for trade agreement DSMs to effectuate other goals beyond state-to-state compliance. Some of these innovative mechanisms for compliance have gained considerable public attention. They have garnered notice for their innovativeness as well as for their considerable reach and emphasis on ‘trade plus’ as seemingly more important than ‘ordinary trade’ matters, and they have altered the conversation on who or what is the target of our DSMs.
This trend is the increased attention to corporate accountability as subject to and part of trade agreement dispute settlement. The new tools in trade policing that most exemplify its corporate turn are those related to sustainability: both environmental and social sustainability. These are tools that US negotiators first developed and have recently implemented and deployed to ensure that companies are abiding by environmental and labour laws where they operate. Such instruments enable US bureaucrats to stop goods at the border that fail to meet these requirements.
A primary example is the USMCA Facility-Specific Rapid-Response Labor Mechanism (RRM, for short). The RRM is a bilateral annex to the dispute settlement chapter of the USMCA – between Canada and Mexico on the one hand, and separately between the US and Mexico on the other. It is designed to enable one of the governments to raise concerns about a denial of workers’ rights at a factory on the territory of the other and ultimately to impose tariff penalties on the goods from that facility if the problem is not resolved. More concretely, the RRM permits US trade agencies to detain goods at the border if those goods come from a factory that is denying workers their collective bargaining rights, such as by tampering with a union voting process. For example, Customs can stop cars manufactured in Mexico at the Texas border and hold them there until the problematic labour issue at the General Motors factory from which they came is resolved to US trade officials’ satisfaction. And it has done just that. General Motors was the first company to be subject to this tool. As of August 2023, the tool has been deployed a dozen times by the US against facilities in Mexico.
The point of the RRM is clear: it puts compliance responsibility not on the government but on companies. Rather than set up a panel to identify whether Mexico is at fault, the tool would allow a panel to make a determination as to whether there has been a violation of workers’ rights at a workplace – effectively, bringing action against a company rather than against the Mexican government as would have been the case under most ordinary DSMs.
And the RRM is not unique. The US has used a similar approach in a forestry annex to the environmental chapter of its trade agreement with Peru, and other governments are now contemplating similar moves in their trade agreements (Claussen Reference Claussen2024). These tools are more akin to corporate social responsibility and corporate due diligence tools than they are to ordinary trade agreement DSMs, but they coexist and are squarely situated among the DSMs of these agreements.
Thus, these recent mechanisms have implications for how we think about DSMs – what counts as a dispute and what constitutes enforcement? These questions align also with questions about standing and actors. The new dispute settlement tools do not rely on either state-to-state settlement or domestic adjudication or entities that use force as we can see in other areas of government and international law.
16.5 Conclusion
The future is bright for trade agreement dispute settlement design. This is a moment of unusual experimentation. After thirty years of convergence towards state-to-state arbitration, trading partners are now re-evaluating the potential for trade agreement enforcement tools to achieve other aims. Bespoke designs are emerging for bilateral problems that target not states but other actors, including private actors and including for trade-plus matters. Designs are beginning to advance beyond boilerplate towards innovative and tailored tools.
Building on this potential will be important to keep enforcement efforts at the supranational level rather than slide into unilateral enforcement as recent pressures on governments to demonstrate that trade agreements are meaningful tools for positive social change have led to some questioning and some faltering. But the new tools, even if reciprocal on paper and agreed among parties, may not necessarily apply equally or be deployed consistently. The next several years will be a proving ground for innovative DSMs and their parties.
Open access