Long-Term Relationship over Litigation: Mediation in WTO Dispute Settlement Proceedings

Abstract

The Dispute Settlement Mechanism (DSM), once regarded as the jewel in the crown of the World Trade Organization (WTO), has been facing a variety of serious criticisms for its inherent limitations and problems while its appellate review function has been paralyzed. Discussions on the reform of the WTO DSM have been under way for several years now. Many key items are on the reform agenda, one of which is to introduce Alternative Dispute Resolution (ADR) proceedings to the WTO DSM. Among several options of ADR, ‘mediation’ can offer an important set of tools for the WTO and its Members to resolve disputes in a more efficient and prompt manner. If properly structured, mediation can complement the existing binding proceedings of panels and the Appellate Body. At the same time, introduction of mediation to the WTO DSM may also cause additional legal and practical problems. It may cause further delays, confidentiality traps, due process myriads, and enforcement loopholes. It is vital to introduce mediation provisions to address those critical problems. Systematized and structuralized mediation in the WTO DSM will be able to offer a viable alternative path to resolve certain complex and sensitive disputes.

1. Introduction

The Dispute Settlement Mechanism (DSM) of the World Trade Organization (WTO) has been in a dire situation since 2016 as the reappointment of retiring Appellate Body (AB) members has been repeatedly blocked. Subsequently, the AB ceased to function in December 2019, and the final AB members' terms ultimately expired in January 2021.

In an effort to find common ground and initiate a breakthrough, the WTO Members began a WTO DSM ‘reform’ discussion. In October 2019, Ambassador David Walker of New Zealand drafted a decision on the functioning of the AB, following a year of consultations with Members through an informal process.¹ The draft decision, however, failed to secure consensus. After the United States had delivered a statement at the General Council on the AB reform in October 2019² and published its own report on the topic in February 2020,³ a US-led informal discussion was launched in March 2022 to build consensus before the then upcoming 12th Ministerial Conference (MC-12) in June 2022.⁴ After the MC-12, Members' efforts continued. An informal process on dispute settlement was convened by Marco Molina of Guatemala in February 2023⁵ and a new consolidated text was produced in February 2024 just before the 13th Ministerial Conference (MC-13) later in the month.⁶ Consensus was not reached in the MC-13.

Despite the continued efforts of the Members to find consensus and reach agreement on a reform package of the WTO DSM and the attendant amendment of the WTO Understanding on Rules and Procedures Governing the Settlement of Disputes (DSU), a breakthrough is not in sight yet, nor is there a positive signal that the gap is being reduced by key Members on core topics on the table. Such being the case, restoring the WTO DSM to full operation, with or without an appellate review function, as a result of a final package to be adopted is arguably one of the most important tasks for the WTO and its Members at the moment. In the course of the reform discussions, various proposals, suggestions, and ideas have been presented and examined. Whatever results from the current reform discussions, it seems likely that a new scheme will constitute a WTO DSM 2.0 with significant modifications and changes from the current one.

One critical topic among the proposals and suggestions is whether Alternative Dispute Resolution (ADR) will be useful and effective in settling disputes among WTO Members, and thus whether it will have to secure a firmer place in the WTO DSM.⁷ If the panel and AB proceedings represent a binding DSM, various ADR options embody a non-binding DSM. Among diverse ADR options, mediation stands out with its advanced and sophisticated features compared to others and is currently receiving renewed attention in various fora as a new means to settle international disputes. Arbitration is also regarded as a form of ADR, but as a binding proceeding, it is excluded from ADR as the term is used for this article. While all ADR methods may contribute to resolving disputes one way or another, a more practical and focused question in this regard would be whether mediation can play a meaningful role in a future WTO DSM. This article argues that mediation will offer a useful tool in resolving certain types of disputes, and that a future WTO DSM should contemplate including mediation as one of the major avenues of settling disputes.

The WTO DSM with its fully operating panel and AB proceedings should be restored as soon as possible. Although the WTO DSM demands an innovative and forward-looking reform, its core ‘rule of law’ elements anchored in the current DSU should be retained. This article is not intended to criticize the current WTO DSM or certain provisions of the DSU. Rather it aims to suggest a new approach to supplement and complement the current DSM. Thus, proposed mediation is not to replace the current system but instead to offer a diverse set of dispute settlement options for the Members.

With this in mind, this article is structured as follows. Section 2 examines the renewed attention to mediation in the global community at present as a context in which to import the recent phenomenon into the WTO DSM. Section 3 then looks into mediation provisions in current trade agreements, analyzing their similarities and differences together with reasons for the rare utilization of mediation. Section 4 explains why mediation will be useful in a future WTO DSM, and what needs to be done to achieve the objectives through the current WTO DSM reform discussion. Finally, Section 5 presents a conclusion.

2. Mediation as an Alternative Means to Settle International Disputes

ADR is defined as a means of out-of-court dispute resolution.⁸ Thus, any method other than litigation can be categorized as ADR, including arbitration, negotiation, good offices, mediation, relevant committee proceedings, and contracting parties' joint interpretation. The WTO Agreements also stipulate ADR that Members can select to settle their trade disputes in the place of panel and AB proceedings: DSU Article 25 sets forth arbitration; DSU Article 5 offers negotiation, good offices, mediation and conciliation; DSU Article 4 mandates parties' consultation before moving on to a panel proceeding; DSU Article 3.12 permits Director-General-brokered good offices for certain disputes under BISD 14S/18; Marrakesh Agreement Article IX provides for interpretation by Members; and Marrakesh Agreement Annex 3 contains the Trade Policy Review Mechanism (TPRM) where Members exchange their views on specific trade measures of the Member under review. Among these, as noted above, arbitration under DSU Article 25 is excluded from the scope of ADR for the purpose of this article as a binding proceeding.

Notably, the WTO has also seen the emergence of a practice where Members raise their concerns over particular measures through an array of official proceedings. The Committee on the Technical Barriers to Trade (TBT Committee)⁹ and the Committee on Sanitary and Phytosanitary Measures (SPS Committee)¹⁰ have developed Specific Trade Concern (STC) proceedings where Members can raise their concerns regarding specific TBT and SPS measures of other Members to resolve differences amicably through discussions in the committees. The broad mandates of the committees have been interpreted to accommodate the STC proceedings.¹¹ Over time, the committees have administered the STC proceedings akin to multilateral mediation.¹² Thus, the STC proceedings can also be said to provide another example of the WTO's ADR schemes.

This menu of ADR arguably stands for the proposition that the WTO Agreements are also receptive to ADR handling trade disputes. As a matter of fact, in the WTO, it is the Dispute Settlement Body (DSB), a collective entity comprising all Members, that makes a final decision after a panel or AB proceeding. A panel and the AB merely assist the DSB in discharging its authority as such to settle disputes. In other words, even binding dispute settlement proceedings of the WTO are structured in a way to place the Members at the apex, not panels or the AB, apparently an arrangement recalling ADR.

Among these ADR methods, mediation is receiving increased attention at the moment with the introduction of the United Nations Convention on International Settlement Agreements Resulting from Mediation (Singapore Convention).¹³ The United Nations Commission on International Trade Law (UNCITRAL) initiated negotiations in September 2015 for a new convention that guarantees the international enforcement of the results of mediation. After three years of negotiation, the text of the convention was finalized in February 2018, and was then adopted by the United Nations General Assembly in December 2018. The new convention was signed in August 2019 in Singapore and went into effect in September 2020. As of July 2024, there are 57 signatories, 14 of which have ratified it. The Singapore Convention aims to facilitate mediation to settle international commercial disputes by agreeing to enforce the outcome of mediation in multiple jurisdictions. As long as the disputes are commercial in nature, it may also apply to those involving states including inter-state ones.¹⁴

Similarly, Investor-State Dispute Settlement (ISDS) proceedings also share the potential of mediation as a means to settle international investment disputes arising from International Investment Agreements (IIAs) in addition to conventional arbitration.¹⁵ The main driving force behind the momentum is the recognition of mediation's unique traits to preserve long-term relationship among stakeholders.¹⁶ Against this backdrop, the International Centre for Settlement of International Investment Disputes (ICSID) adopted mediation rules in 2022,¹⁷ and the UNCITRAL completed mediation provisions and guidelines in April 2023, which were then adopted in the Commission session held in July 2023.¹⁸

At the same time, some recent free trade agreements (FTAs) have adopted mediation proceedings with various different titles and modes as an additional means to settle disputes between contracting parties. These recent developments indicate that mediation can be a useful tool for state-to-state dispute settlement proceedings.¹⁹ In fact, mediation has long been an important tool in resolving sensitive political and diplomatic disputes among states.²⁰ Of course, mediation has its own shortcomings and limitations as well.²¹

From the recent discussion at various fora, mediation is generally defined as any proceeding, regardless of its specific description, involving a third person who facilitates disputing parties' negotiation but lacks the authority to impose a decision upon them.²² It may, therefore, include diverse proceedings satisfying the definition. Notably, the term is also used to cover conciliation; as a matter of fact, mediation and conciliation differ only as regards the extent of an intermediary third party's role, and if the depth of the involvement is set aside, the difference between the two is minimal, if any.²³ Mindful of this, the majority of states seem to prefer the term ‘mediation’ which now covers conventional conciliation as well.²⁴

Mediation involves a third person (mediator) who hears from both parties, assesses the parties' arguments, and offers his or her own solution package.²⁵ With a reliable third person at the center, it may ensure objective evaluation of the matter at issue, unlike negotiation or good offices where only parties meet directly. Objectivity may lead to a reasonable assessment of the strength and weakness of the parties' arguments, which in turn may lead to finding a reasonable solution package. While a third person is engaged, that person is not able to impose a solution; in other words, a mediator merely suggests a possible solution, and it is the parties that decide to accept (or not accept) the solution. Thus, party autonomy is preserved throughout. The combination of objectivity and party autonomy has apparently made mediation attractive in recent discussions of the global community. Once a dispute is resolved through mediation, the disputing parties are believed to be able to maintain their long-term relationship.²⁶

The increasing global attention to mediation and its perceived long-term benefit arguably shed an important light on the WTO DSM. In the current discussion on the WTO DSM reform, Members are well advised to contemplate mediation carefully as an additional option for settling disputes arising from WTO Agreements. Needless to say, mediation has its own shortcomings and limitations, and inserting a new proceeding to the already thinly stretched current DSM would invite a new risk. If a proper balance is struck between the merits and demerits of mediation in the WTO DSM, however, and if carefully drafted provisions are introduced so as to facilitate mediation both in terms of legal and practical perspectives, mediation arguably has the potential to deal with, at least, certain types of disputes in an efficient manner.

3. Mediation in Trade Agreements

Mediation is no stranger to trade agreements. It appears in various places in trade agreements. One could argue that it has been on the radar of trade agreements for some time now. If mediation is regarded as a useful tool to settle certain international disputes, a presumption which seems to be broadly accepted considering recent developments on mediation debates on multiple fronts, it may also apply, mutatis mutandis, to settling disputes arising from trade agreements.

3.1 Mediation in the WTO Agreements

Article 5 of the DSU sets forth mediation along with other ADRs.²⁷ It provides that Members can resort to mediation at any time. Any statement made during mediation cannot be referred to in subsequent panel or AB proceedings. Other than this generic, open-ended provision, the DSU or any other covered agreement is silent on details of mediation. In fact, similar provisions of a generic nature also appear in FTAs. Since the inception of the WTO in 1995, there has been just one mediation under Article 5 so far.²⁸ This case relates to a dispute between the then European Communities and Thailand and the Philippines concerning the former's preferential tariff rates for canned tuna imported from African, Caribbean, and Pacific states.²⁹ Through a mediation conducted between September 2002 and December 2002, the dispute was amicably resolved between the parties.³⁰ The WTO has also seen two examples of good offices since 1995. One concerns the long-lasting banana dispute between the European Union on the one hand, and Colombia, Costa Rica, Ecuador, Guatemala, and Honduras on the other hand. This dispute was resolved through the good offices of the WTO Director General Pascal Lamy conducted during the period of November 2007 to December 2009.³¹ It should be noted that the good offices in this dispute were under Article 3.12 of the DSU, instead of Article 5, where the Director-General is authorized to conduct good offices in a dispute brought by a developing Member against a developed Member.³² The other concerns the tuna dispute mentioned in the mediation example above. Before initiating mediation, the disputing parties resorted to the Director-General for the appointment of a mediator.³³ During September 2002 to October 2002, then Director-General Supachai Panitchpakdi apparently engaged with the parties in preparation for the mediation.³⁴ To the extent that the Director-General facilitated dialogue and provided assistance during this period, his role could be considered as an informal good office.³⁵

There is no knowing what has caused the almost complete non-utilization of mediation, but the absence of detailed provisions on mediation seems to be one of the important reasons in this regard. In other words, Article 5 merely offers a possibility to pursue mediation, but fails to offer details. The design of the mechanism as envisaged in Article 5 lacks detailed procedural rules that could be readily utilized by the WTO Members.³⁶ In addition, the language of Article 5 is also confusing; the first and second sentences of Article 5.3 stipulate that mediation can be utilized at any time, whereas the third sentence conditions the request for the establishment of a panel on the termination of mediation, suggesting mediation is only a pre-panel option.³⁷ An exact timeframe of mediation and its status within the DSU thus seem to be rather elusive.³⁸

Under these circumstances, government officials from disputing parties could hardly possess authority or exercise an initiative in offering and accepting a mediation proposal. The extraordinary nature of mediation would loom large in the absence of a textual anchor, and an official would be incentivized to not pursue that irregular path. Similarly, even if mediation is conducted and a compromise package is somehow structured, which may look mutually beneficial in the long term, an official would still be reluctant to take the deal because of the political and legal consequences from domestic constituents. The negotiating official would have to take all responsibilities, a consequence hard to accept for many government officials of Members. A defeat in a panel or AB proceeding is a national matter, but acceptance of the package deal would be his or her personal matter. This may explain the relatively rare usage of mediation in the WTO DSM.

The same mediation-averse tendency is also observed in ISDS proceedings. In the ICSID's 967 investment disputes as of the end of 2023, only 14 disputes have been resolved through mediation – a very low utilization rate (1.4%) indeed. Observers also tend to point to the lack of a system and sufficient provisions for mediation. Quite possibly, government officials participating in ISDS proceedings are simply forced to look the other way when it comes to mediation because of the same political and administrative burden. The ICSID's new rules on mediation now aim to overcome this traditional obstacle by introducing detailed and structured provisions on mediation. The UNCITRAL is also joining forces in this regard by adopting detailed provisions and guidelines that states can refer to when contemplating mediation in their investment disputes. These recent developments on the ISDS front arguably further highlight the WTO DSM's falling behind when it comes to mediation as a new form of settling disputes.

When robust panel and AB proceedings were in place, the situation before the paralysis of AB in December 2019, few Members would have in fact been incentivized to examine the possibility of mediation as requisite details were simply absent from the text of WTO Agreements. The appropriateness and viability of the current panel and AB proceedings are now going through a rigorous and critical assessment, and once the current reform debates are completed, a future WTO DSM is likely to have a different shape and structure. Such being the case, unlike in the past, Members would now be required to look beyond the existing DSM or otherwise look to other directions than the conventional dispute settlement paths. If so, the remaining ‘push’ would be the elaboration and systematization of mediation clauses in the WTO Agreements to incorporate and facilitate mediation as a meaningful alternative in a WTO DSM.

3.2 Mediation in FTAs

Notably, FTAs take a more active stance toward mediation than the WTO Agreements. There are three different dimensions where mediation appears in FTAs. Firstly, there is a general provision offering mediation as one of the menus of ADRs, as seen in Article 5 of the DSU.³⁹ A general provision of this sort uses the same or similar title with Article 5 of the DSU (Good Offices, Conciliation, and Mediation).⁴⁰ It should be noted, however, that this general provision does not seem to be a universal feature: some FTAs do not contain such a general ADR provision including mediation.⁴¹

Secondly, committees in charge of respective chapters of FTAs possess mediation as one of the options to explore in settling disputes between or among the contracting parties.⁴² In these instances, sometimes mediation is simply optional.⁴³ At other times, mediation is procedurally mandatory. A mandatory feature emerges in two different schemes: one is where a party (or parties) should proceed to mediation before it moves to a binding dispute settlement proceeding,⁴⁴ while the other is where a party (or parties) should go through procedures containing mediation as one of the menu of options before it moves forward to a binding dispute settlement proceeding.⁴⁵ In the latter category of mandatory features, referral to regular DSM of the FTA is only permitted upon the completion of respective, specific proceedings.⁴⁶ In other words, even if mediation suggested by the committee at issue is not selected because of a party's or parties' refusal, or mediation suggested by a party is not selected because of the other party's refusal or non-cooperation, at least the entire process in the respective chapters should be followed and completed. Stated differently, even if mediation in these circumstances is obviously not ‘mandatory’, the consideration of mediation as a possible option can be said to be still mandatory.⁴⁷ Similarly, joint committees of FTAs are usually in charge of the entire issues arising under the respective FTAs, including settling disputes between parties, which in turn may include mediation as one of the available options for the joint committees to pursue to settle disputes between the contracting parties.⁴⁸

Thirdly, specific types of dispute arising from particular chapters of FTAs are designed to proceed to mediation.⁴⁹ In other words, in this category, mediation is the only option to resolve the disputes arising from the specific chapters. To the extent that parties' voluntary acceptance of mediation as the only solution tool is provided in a blanket fashion in advance, it could be called an ‘enhanced’ version of mediation. It should be noted that sector-specific dispute settlement proceedings in FTAs are not always mediation; sometimes, they exist in the form of a binding proceeding.⁵⁰ Mediation, however, is selected to resolve certain types of disputes. Disputes regarding labor, environment, and non-tariff barriers (NTBs) are such examples. As regards specific descriptions and formats of mediation, there are diverse variations in different FTAs. Therefore, it should be noted that in determining whether there exists mediation as the term is now defined in various fora of the global community, one should examine the essence of a proceeding (i.e., a voluntary, non-binding proceeding involving a third person facilitator).⁵¹ Consequently, as long as a certain procedure contains core elements of mediation as such, it may (or should) be defined as mediation for the purpose of this discussion.

As a matter of fact, regarding the appellation, diverse titles are used. The non-binding proceedings may be referred to as mediation,⁵² panel of expert proceedings,⁵³ or review panel proceedings,⁵⁴ or simply no specific designation is given.⁵⁵ And yet, they could all be categorized as mediation as the term is currently defined.

Post-mediation schemes also vary in various FTAs. Sometimes mediation proceedings are final, and there is no further proceeding to pursue, at least for that particular dispute.⁵⁶ Other times, upon the completion of mediation, parties can still advance to a further proceeding, i.e., a binding dispute settlement proceeding.⁵⁷ In this case, mediation is a way station to the binding DSM. The latter category can be further broken down into (i) where mediation is voluntary (that is, merely an option) and (ii) where mediation is mandatory (that is, a prerequisite). In any event, the fact that the outcomes from all these mediation proceedings are non-binding does not change. Nor does the fact that a third person (mediator) is engaged to facilitate the party-to-party negotiation without an authority to impose an outcome.

Of the three mediation categories summarized above (i.e., general reference provisions, option offering provisions, and chapter-tied unitary option provisions), the first and second forms of mediation have barely been utilized, probably because of the same mediation-averse atmosphere among government decision makers mindful of political and legal repercussions from domestic interest groups. On the other hand, the third form has begun to make inroads into dispute settlements in FTAs. The reason that the third one has increasingly become a viable path seems to be that ‘detailed provisions’ are stipulated in advance for these types of mediation. Relatively concrete contents and elaborate procedures are set forth in the text of the FTAs in question, so that disputing parties are aware of what to do and how to proceed as regards the mediation at issue. Such being the case, the phenomenon in FTAs also supports the notion that elaboration and systematization should lie at the core of projects to facilitate mediation in the WTO DSM in the long run.

One example falling under the third category is mediation for NTBs. The ‘Free Trade Agreement between the European Union, of the one part, and the Republic of Korea, of the other part’ (EU–Korea FTA) and the ‘Free Trade Agreement between the Government of the People's Republic of China and the Government of the Republic of Korea’ (Korea–China FTA) adopted mediation clauses for NTB disputes.⁵⁸ NTB disputes are considered complex and controversial as they are frequently related to national regulatory policy space. The parties of the FTAs were of the view that some NTB disputes are better suited to non-binding dispute settlement proceedings with a third person leading – i.e., mediation – and they thus introduced special proceedings for this purpose. These proceedings are not intended to replace the DSMs of respective FTAs: they are merely added to the DSM as another option if parties so choose. Even between the EU–Korea FTA and the Korea–China FTA, there is a wide variance. While the former contains detailed provisions on mediation in an independent annex, the latter includes just one provision referring to mediation as an additional option for NTB disputes.⁵⁹ As discussed below, mediation utilization is arguably dependent on the existence of detailed provisions so as to overcome the otherwise prevailing mediation-averse atmosphere from government officials’ perspectives. In that respect, the EU–Korea FTA's approach seems more practical in terms of facilitating mediation in actual disputes.

The sustainable trade chapter of the EU–Korea FTA also offers mediation to deal with disputes relating to labor and environmental issues.⁶⁰ Not all disputes arising from these two areas are subject to ordinary, binding DSM of the FTA. Instead, they can only be referred to the special DSM in the form of mediation. One thing to note is that the title of the dispute settlement proceeding here is termed as ‘panel of expert’ proceedings. Irrespective of the use of the term ‘panel’, the panel proceedings here are non-binding and the third person(s) lacks the authority to impose the decision. The parties also possess the authority to determine the future course of action after the mediation, so the procedure still carries the core traits of mediation. The same scheme is also adopted in the ‘EU–Canada Comprehensive Economic and Trade Agreement’ (CETA), the ‘Free Trade Agreement between the European Union and the Republic of Singapore’ (EU–Singapore FTA), and the ‘Free Trade Agreement between the European Union and the Socialist Republic of Vietnam’ (EU–Vietnam FTA).⁶¹

To date, there are two cases that merit careful scrutiny in this regard. While they both dealt with links between labor and trade, the type of labor provisions in the two proceedings were not the same. Moreover, the FTAs at stake do not deal with sustainable development in the same way either.⁶² The first case concerns the dispute between the EU and the Republic of Korea, which took place under the EU–Korea FTA's trade and sustainability chapter.⁶³ This proceeding arguably presents a variety of novel issues in this respect.⁶⁴

The first Panel of Experts under Article 13.15 of the EU–Korea FTA (EU–Korea proceeding) was formally established on 30 December 2019 and rendered its report on 25 January 2021. The EU–Korea proceeding was initiated by the EU on 17 December 2018 by submitting a request to Korea pursuant to Article 13.14.1 of the EU–Korea FTA concerning certain measures, including provisions of Korea's Trade Union and Labour Relations Adjustment Act (TULRAA) and Korea's non-ratification of certain core International Labour Organization (ILO) Conventions.⁶⁵ Following governmental consultations, on 4 July 2019, the EU requested a Panel of Experts to be convened in accordance with Article 13.15 of the EU–Korea FTA. Relying on the first and last sentences of Article 13.4.3 of the FTA, the EU alleged that the TULRAA did not comply with Korea's commitment to respect, promote, and realize freedom of association and the effective recognition of collective bargaining, and that Korea had made insufficient efforts to ratify fundamental ILO Conventions given that four of those conventions remained unratified by Korea at the time.⁶⁶ In its decision, the panel found Korea to be in violation and provided their suggestions for the resolution of the dispute. It then requested that the EU and Korea continue to discuss to reach a final settlement of the matter. Given subsequent interactions between the two parties after the proceeding, it appears that the dispute is now resolved.

The other case involves a dispute arising from a United States' FTA. On 14 June 2017, the arbitral panel established under the Dominican Republic–Central America–United States Free Trade Agreement (CAFTA-DR) rendered its decision for a dispute between the United States and Guatemala.⁶⁷ It should be noted that this proceeding is a binding DSM under the CAFTA-DR. That said, the dispute also contains mediation components in the entire proceeding and illustrates its possible usefulness. First of all, the labor chapter and environment chapter of the CAFTA-DR also include mediation as an option to settle disputes relating to these subjects.⁶⁸ As the parties are obligated to complete the chapter-specific dispute settlement, which includes mediation as one of the available avenues to reach the binding DSM, these chapters can be said to have a hybrid format when it comes to dispute settlement. Secondly, the proceeding itself contains mediation elements. In the dispute, the US panel request was submitted on 9 August 2011.⁶⁹ Negotiations between the US and Guatemala ensued and the panel was composed on 30 November 2012, almost 19 months later.⁷⁰ Once the panel was composed, the panel proceeding was again immediately suspended.⁷¹ The two sides then continued their remaining negotiation for the next five months or so, and reached agreement on 11 April 2013 with a mutual action plan.⁷² However, when the action plan was not implemented to the satisfaction of the US, the panel proceeding then resumed on 19 September 2014, almost 17 months later.⁷³ Thus, until 19 September 2014 when the panel proceeding effectively began, the two sides had already engaged in negotiations for three and a half years along with an agreed (although subsequently aborted) settlement package. To the extent that the three and a half years of negotiations were carried out with the panel in the background, ready to step in, this procedure arguably carries a close semblance to and similarities with mediation as well.⁷⁴

4. WTO DSM Reform and Mediation

Mindful of the increasing attention to and appetite for mediation in various international dispute settlement proceedings, and novel experiments in FTAs to apply mediation for various purposes, WTO Members are advised to explore mediation in the WTO DSM in a more serious manner. As the current discussions on the WTO DSM reform undertake to examine a wide spectrum of issues, the Members would have to make a decision on prioritization at some point. In this process, a strong argument can be made that mediation should be included in the priority topic list. This section first provides an analytical framework to assist the current WTO DSM reform negotiations in this regard and then specific questions to be asked in the discussion and negotiation process.

4.1 Analytical Framework

In carrying out discussions on mediation, more than anything else the Members should put mediation into the right perspective as a new means to resolve trade disputes. Both the merits and demerits of mediation are currently being debated robustly. These dialogues should enlighten the WTO Members about the benefits mediation can bring together with the new burden it can entail, which will then pave the way for a more balanced and practical approach to mediation in the WTO DSM. While the Members are encouraged to explore the mediation avenue actively in the current WTO DSM reform negotiations, they should remain keen to assess how specifically it can contribute to resolving disputes between Members. A simple expectation that mediation may lend a helping hand one way or another is not preferable; prying open the already procedure-saturated DSU to find a place for mediation may end up further complicating the WTO DSM and cause further pretexts for delay.

Mediation's benefits are widely recognized. All things being equal, it ensures relatively prompt resolution of a dispute at low cost.⁷⁵ As it is a party-autonomous proceeding with a mediator assisting, parties can come up with flexible solutions and diverse packages, tailored to the dispute and disputing parties in question.⁷⁶ These benefits of mediation may well apply to WTO disputes, though to varying degrees. The experiments being carried out by recent FTAs arguably imply the potential that the same or similar formula may also apply to the WTO DSM.⁷⁷ As a matter of fact, quite a few FTAs have specific provisions to encourage and facilitate ADRs as a means to resolve disputes between parties in the free trade area.⁷⁸ In particular, mediation is being fine-tuned and elaborated for new possibilities to handle disputes arising from FTAs.

More importantly, mediation is valued as a means to settle disputes while preserving long-term relationships between disputing parties.⁷⁹ This ‘relationship-maintaining’ feature is believed to be a core value of mediation.⁸⁰ As such, mediation is being explored and applied in international commercial disputes, investment disputes and state-to-state disputes when a long-term relationship is more important than a short-term gain.⁸¹ The same feature is arguably in great demand in none other than the WTO regime where the level of conflict and tension has been on the sharp rise for the past several years.⁸² If the ultimate question boils down to long-term relationship versus short-term trade gain, mediation may offer a reliable vehicle to strike a right balance and to resolve a dispute.

Mediation's downside should not be overlooked either. Enforceability raises a critical concern.⁸³ Financial and human resources might be wasted if an agreed deal is not carried out. It may be useful only for some type of disputes.⁸⁴ It may expose government agencies and officials to domestic political criticism and legal consequences concerning a decision to settle, a situation that they would not encounter in an all-out effort in litigation.⁸⁵ The relative lack of familiarity and experience with mediation also raises a practical hurdle.⁸⁶ Flexibility in form and outcome also means uncertainty.⁸⁷ Mediation may also lead to a longer and costlier proceeding if it turns out to be yet another layer added to a binding proceeding.

Thus, in a future mediation discussion on the WTO DSM, a balance point should be struck between these merits and demerits. In particular, it is imperative to explore ways to allay the concerns stemming from the demerits. In the context of the WTO DSM, this would mean how to ensure (i) an elaborated textual basis to systematize and structuralize mediation; (ii) a more reliable and available option for Members to turn to; (iii) a harmonized insertion to the existing panel and AB proceedings; and (iv) avoidance of unreasonable and unnecessary prolongation of the overall dispute settlement proceedings.

With these framework issues in mind, and in order to put the mediation discussion at the current WTO DSM reform debates into perspective, the following four specific questions should be asked in future debates and negotiations. Focused efforts need to be made to find answers to these four specific questions.

4.2 Specific Questions to Be Posed

The fundamental questions when it comes to mediation at this juncture are the following: (i) Why has mediation almost never been used by WTO Members?; and (ii) Why would Members use it more today? As regards the first question, the discussion in Section 1 above arguably points to the lack of an institutionalized system, which has made mediation an impractical choice for government. The answer to the second question is that once the institutionalization hurdle is overcome, governments are likely to turn to mediation more actively given the increasing awareness of the benefit of mediation. In response to the two fundamental questions, the four specific questions to be tackled in upcoming discussions and negotiations in the particular context of the WTO DSM are as follows:

1. - Will mediation help resolve disputes among Members?

2. - What needs to be done to facilitate mediation?

3. - How do we align mediation with panel/AB proceedings?

4. - How do we manage novel challenges and obstacles? These four questions are closely intertwined. Each of these four questions is examined in turn below.

5. (i) Will Mediation Help Resolve Disputes among Members?

If mediation is added to the mix of the WTO DSM, as an additional option that disputing parties can turn to, it may help them settle their disputes out of panel or AB proceedings. In particular, mediation may be useful in dealing with disputes with high political or diplomatic implications. It may also be mobilized for disputes involving provisions in the WTO Agreements, which are outdated because of fast technological development and societal changes. In the absence of an amendment, panels and the AB are forced to be confined to the textual bounds because of the general rules of treaty interpretation set forth in Article 31 (1) of the Vienna Convention on the Law of Treaties (VCLT). Adjudicators would have to apply existing provisions even if distinctly outdated. Mediation can offer flexibility wherein the disputing parties can pursue a reasonable compromise in these circumstances. Thus, at least some, if not all, disputes at the WTO may be susceptible to mediation.

In the WTO, the Members tend to present wide differences and diverging views when it comes to sensitive matters falling under their regulatory space. Regular panel and AB proceedings may as well resolve disputes arising from these matters successfully. But recent experiences tell us that panels and the AB may simply fail to settle such disputes to the satisfaction of the parties. Mediation may play a critical role at this juncture. In particular, mediation to resolve these politically or diplomatically loaded disputes may help parties preserve their long-term relationships by prodding them to find a mutually acceptable middle ground.⁸⁸ As a matter of fact, more and more disputes lodged at the WTO panel and AB proceedings require a ‘value judgment’ or juxtaposition of different perspectives and views.⁸⁹ For instance, NTBs, local barriers to trade (LBTs), subsidies (particularly indirect subsidies relating to industrial policies), domestic regulation in the services market, regulating newly emerging businesses and transactions (as in the digital economy), general exceptions (GATT Article XX), and national security exceptions (GATT Article XXI) have triggered sharp divisions and intense controversies among the WTO Members recently. Disputes involving such sensitive topics as these may be candidates for mediation.⁹⁰ Of course, these issues can and should be referred to binding DSM as necessary because often times binding DSM offers the only viable solution. That said, however, some of these issues may well be suitable to more candid, flexible discussions through mediation.⁹¹

Tellingly, it should be noted that sometimes a decision from a panel or the AB does not necessarily resolve the underlying dispute despite the official rendering of the decision.⁹² In other words, disputes are officially resolved but practically persist. Mediation may be able to lead to a more comprehensive and longer-lasting resolution. Consider, for example, the decades-long large civil aircraft disputes. Multiple panel and AB proceedings have failed to offer finality to the dispute between the United States and the EU.⁹³ Mediation could have offered a workable solution. As a matter of fact, Brazil and Canada reached a mutually agreed solution in their own aircraft disputes after several panel and AB proceedings spanning over almost 14 years.⁹⁴ Mediation could have helped expedite the resolution of the dispute.

Consider also the long-lasting ‘zeroing’ disputes from this angle. In the past two decades, the AB has held more than a dozen times that zeroing is inconsistent with the WTO's Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade 1994 (Antidumping Agreement),⁹⁵ and yet the core issue still remains unresolved so that a losing Member refuses to implement the ruling and recommendations of the AB,⁹⁶ and a prevailing Member chooses not to take a DSB-authorized retaliatory measure.⁹⁷ In the course of the two decade-long multiple proceedings, mediation could have helped the Members involved find a mutually satisfactory solution.

Subsidy disputes offer another example of mediation's possible usefulness. Amid the growing tension on the trade front between the United States and China, subsidy issues take the center stage of the bilateral standoff. Some of high-profile subsidy disputes have proceeded to consultation under the DSU, only to be stalled at that stage.⁹⁸ The completion of the binding proceedings under the DSU is not expected any time soon. Furthermore, the jurisprudence on key subsidy issues arguably shows some internal conflict. For instance, as regards ‘public body’ under Article 1.1(a)(1) of the Agreement on Subsidies and Countervailing Measures (SCM Agreement), the AB took a different approach in US–Anti-Dumping and Countervailing Duties ⁹⁹ and US–Countervailing Measures.¹⁰⁰ In the earlier case, the United States Department of Commerce (USDOC)'s determination on ‘public bodies’ was found to be inconsistent with the SCM Agreement's language of ‘any entity controlled by a government’.¹⁰¹ The AB determined that the USDOC's decision to treat certain State-Owned Enterprises (SOEs) of China as public bodies was in violation of Article 1.1(a)(1) of the SCM Agreement.¹⁰² On the contrary, in the latter case the AB took a different approach to the same question.¹⁰³ The AB found that a close relationship between an entity exercising government function and transaction at issue is not required, broadening the scope of public bodies enough to cover many Chinese SOEs and thus upholding the standards adopted by USDOC.¹⁰⁴ The different approaches (or evolution of jurisprudence) to the WTO DSM might further intensify already controversial subsidy disputes between the United States and China, and between other Members. As subsidy norms have long been scrutinzed at the WTO for possible changes and amendments with the changing role of a government in the background in the course of a pandemic, economic crises, and rampant industrial policies, present and future disputes involving subsidies may not be amenable to easy resolution through the mechanical application of the SCM Agreement. Unless and until new subsidy deals are formulated in the future, mediation could offer an interim alternative route to resolve disputes between Members.

1. (ii) What Needs to be Done to Facilitate Mediation?

In order to facilitate mediation in the WTO regime, the first and foremost task is elaboration, structuralization, and systematization of mediation-related provisions in WTO Agreements. As discussed in the sections above, in trade agreements mediation can be offered in many different ways and forms. The preferable course of action for the WTO DSM would be to introduce mediation clauses, either in the DSU or in other guidance documents, which are elaborated, systematized, and structuralized so that they could help address the concerns identified above.

Firstly, as for ‘elaboration’, if possible, mediation should be offered as a readily available option together with clear and step-by-step explanations as regards how to initiate, administer, and manage it. The WTO Members should be informed in advance of the ins and outs of mediation proceedings, instead of merely being told to form the proceedings from scratch themselves. In this regard, perhaps making mediation mandatory for certain disputes (either categorically or by the decision of the DSB) can be contemplated. Even if mediation is left as a voluntary procedure to be chosen by the parties on an ad hoc basis, a stronger suggestion and tone in the provisions of the DSU or other relevant texts will help foster and facilitate mediation. For example, the Members might want to include a provision (even though hortatory in nature) to the effect that mediation is a viable mechanism to resolve disputes, and disputing parties should consider mediation actively before they decide to move on to panel proceedings. The purpose of such a provision is to tell the government officials and domestic constituents of disputing parties that mediation is readily available and sometimes preferable.

In fact, concrete actions are now being contemplated for this purpose, although at an early stage. For instance, in a recent General Council report containing a consolidated text summarizing informal discussion on the WTO DSM reform, the tentative general principle on mediation purports to mandate the Director-General or the Chairperson of the DSB to offer mediation with a view to assisting the disputing parties in settling their disputes.¹⁰⁵ The same report also includes a more detailed appointment procedure for the mediator, where the parties can turn to a pre-established list of mediators, and the selected mediators should comply with codes of conduct.¹⁰⁶ It also stipulates procedural rules for mediation including specific deadlines.¹⁰⁷ In light of this, overall this report can be said to be an important stride in the right direction. In addition, these efforts reflected in the report largely keep up with recent developments in other international fora.

There are several provisions in the DSU to be amended to import mediation. Of course, it is also possible to consider adopting a stand-alone instrument, including mediation-related provisions without amending the DSU.¹⁰⁸ Some elements can be introduced through a new, stand-alone instrument. But some elements need to be introduced through the DSU. For instance, those touching upon structural aspects would arguably demand adjustment of the DSU. Consider timeframes of the current DSU in this regard. In principle, mediation may take place at any stage of the WTO DSM – pre-panel, inter-panel, post-panel, pre-AB, inter-AB, and post-AB proceedings. When arbitration under Article 21.3(c) or Article 22.6 of the DSU is pursued, mediation may also take place prior, during, or after the proceeding. Conduct of mediation inevitably stands to affect these existing proceedings of the DSU, and the associated timelines stipulated in the DSU. It is not clear whether and how mediation-triggered delay can be accounted for; the parties may agree to extend the deadlines in some instances, but they cannot do so when the deadlines are fixed, such as the 90-deadline for an AB review under Article 17.5. Questions may also arise when mediation is delayed despite parties’ desire, because of the reasons on the part of the mediator or institution. To safeguard against legal uncertainties, provisions need to be inserted to adjust, coordinate, or exempt various timelines in case of mediation.

Similarly, definitions of key terms should preferably be included in the DSU. It would be odd to have the term ‘mediation’ included in the current DSU (Article 5) and have it defined in a lower legal instrument.¹⁰⁹ Adoption of a panel report under Article 6 and an AB report under Article 17 provides legal finality under the current DSU. Should the parties be permitted to deviate from the contents of a DSB-adopted report, the possibility should be clearly stipulated in the relevant provisions of the DSU. Likewise, surveillance of implementation by the DSB is only confined to adopted panel and AB reports under Article 21. It is questionable whether an instrument outside the DSU can expand the surveillance to something else. General principles contained in Article 3 may also need new provisions specifically tailored to mediation and ADRs. These prospective changes would arguably demand an amendment of the DSU. Only core issues can be reflected in the text of the DSU, and details can be compiled in a new appendix (Appendix 5) to be attached to the DSU.

On a different note, from the policy perspective a bifurcated approach – i.e., the current DSU remains intact while another WTO instrument is adopted for specific rules for mediation – might/would further weaken the stature of the DSU by outsourcing a procedure to another instrument. Note that Article 23.1 stipulates that ‘([w]hen Members seek the redress of a violation … [they] shall have recourse to, and abide by, the rules and procedures of [the DSU])’. This provision was introduced for the purpose of ‘strengthening of the multilateral system’. An introduction of a competing instrument may run the risk of chipping away at the central status of the DSU. In short, both from the legal and policy perspectives, it would be more prudent to amend the DSU to incorporate key elements related to mediation together with an appendix to set forth detailed rules and guidelines.

In light of this, more specifically the following provisions of the DSU can be contemplated for a possible amendment. Article 2.1 of the DSU would have to be expanded to include mediation proceedings within the DSB scope of authorities even though a particular settlement decision will be made by disputing parties. Similarly, one more paragraph would have to be inserted into Article 3 as it currently includes only panel and AB proceedings (paragraphs 2, 3, and 4), arbitration (paragraph 4), and a mutually agreed solution (paragraph 6). The term ‘conciliation’ in paragraph 6 should be changed as well for consistency purposes. Most importantly, Article 5 should be further elaborated to provide detailed rules on mediation.¹¹⁰ Recommendation on mediation referring to its usefulness, efficiency, and flexibility should be stipulated as separate paragraphs in Article 5. General principles on mediation should also be stipulated in Article 5 as well, such as procedural due process, neutrality of mediators, confidentiality, timing of initiation, and ‘without prejudice’ affirmation. Further details such as rules of procedure for mediation, selection of mediators, and codes of conduct for mediators can be included as an annex (or multiple annexes as the case may be) to the DSU. One might also consider adopting a separate plurilateral agreement to regulate mediation.¹¹¹ It may not be a preferable option because a separate agreement would run the risk of creating a dispute settlement proceeding outside the DSU, thus detached from the existing DSM, which would further complicate the already complex WTO DSM.

Article 23 of the DSU only addresses, in paragraph 2, panel and AB proceedings, 21.3 arbitration on the reasonable period of time, and 22.6 arbitration on the level of suspension. A stand-alone paragraph would have to be inserted to cover mediation. Article 24, paragraph 2 can also be amended to stipulate broader mediation proceedings for disputes involving least developed country (LDC) Members. At present, the paragraph only addresses ADRs between consultation and request for the establishment of a panel. The provision should be expanded to set forth mediation in the entire dispute settlement proceedings in order to assist LDCs in WTO DSM.

Outside the legal text, the WTO Secretariat may also consider education and information sharing programs for the spread of mediation among Members.¹¹² In order to secure reliable mediators, each WTO committee may prepare its own list of experts in the respective expertise area, who can serve as mediators for the disputes arising in the specific area.¹¹³

Secondly, regarding ‘structuralization’, logistical and administrative issues relating to mediation should be prepared thoroughly as well. For example, for mediation to begin, proceed, and succeed, a roster of mediators and their code of ethics/conduct should be prepared in advance. Their compensation scale and conflict of interest assessment are also necessary. Supporting staff and liaison officers should be either composed or secured as well. Otherwise, disputing parties will have to deal with all these logistical issues in the early stage of the dispute, which is obviously a drag on the prospect of mediation in the eyes of the decision makers in the respective governments. While mediation is confidential in principle, some information on mediation may have to be made public to meet the minimum transparency requirement under the WTO regime and under domestic laws and regulations. Finding the right balance is also required and someone should be in charge of these matters on behalf of the disputing parties and mediator – yet another reason to secure administrative and logistical arrangements in advance.

With respect to structuralization, enforcement is an important issue as well in this respect.¹¹⁴ If the result of mediation is not enforced, the trust in the option would evaporate rather quickly. In fact, one of the major concerns for mediation is that it is non-binding. However, it is important to note that although mediation is non-binding, the outcome of the mediation can be made binding if the parties so choose. In other words, while mediation itself is not legally binding, the instrument reflecting the outcome of mediation may evolve into an agreement between the parties if they vow to keep to the terms in the agreement.¹¹⁵ It depends on the will of the disputing parties. This is the scheme states and governments are currently trying to sharpen in other international fora discussing mediation.

For instance, this is the very objective of the Singapore Convention. The disputing parties’ agreement can be recorded in an instrument (called settlement agreement) and the parties would have to abide by the contents of the instrument.¹¹⁶ As a contract, it is now enforceable in the place of mediation as well in other jurisdictions of the signatories of the Singapore Convention. A similar feature may be contemplated for mediation in the WTO DSM. An instrument agreed upon by the disputing WTO Members via mediation can sometimes be a non-binding one, but other times it can also be a binding one if it is formulated to constitute a treaty under international law once the parties record their willingness to abide by it in accordance with Article 2 of the VCLT.¹¹⁷ In other words, the questions whether mediation is binding and whether the outcome of the mediation is binding are separate and distinct, and the disputing WTO Members may (in fact should be encouraged to) choose to have a final instrument evolve into an international agreement between them in which they vow to keep to the terms of the settlement agreement.

As a matter of fact, mediation clauses in some of the FTAs also rely upon this avenue of enforcement.¹¹⁸ This would also support the proposition that a similar feature is likely to work for the WTO DSM. At times, there are instances where a more effective enforcement mechanism for the outcome of mediation is adopted in advance.¹¹⁹ For example, a third person(s) is engaged to hear claims from the disputing parties, and his or her decision is rendered but is not binding in and of itself. The implementation is left to the mutual agreement between the parties but the third person also presents his or her plan for enforcement. Only when the parties' voluntary agreement for action is not carried out as agreed upon does a pre-determined enforcement procedure come into play. To the extent that the third person cannot impose a decision with a final decision reserved to the disputing parties, these proceedings may still be categorized as mediation. When it comes to enforcement, the WTO DSM will be able to explore various ‘creative’ ways for enforcement.

In this regard, the inclusion of the enforcement issue in the recent General Council report is a welcome move. The report stipulates that any mutually agreed solution reached between the disputing parties as a result of mediation should be notified to the DSB and the relevant Councils and Committees.¹²⁰ Once undertaken, such notification may help lay the groundwork for future monitoring of the implementation of the agreed settlement package. According to the said report, any Member can raise the issue of implementation of the settlement package at the DSB.¹²¹ This could ensure regular monitoring of the implementation.

Article 21 of the DSU stipulates surveillance of the implementation of recommendations and rulings of the DSB as a result of the adoption of the panel and the AB reports. The article can be expanded to cover surveillance of settlement packages agreed on between disputing parties as a result of mediation. In the alternative, a mirror article can be newly created in the DSU to set forth surveillance proceedings for settlement agreements of mediation. In practice, after mediation, the two parties should preferably agree on a settlement package. The package is preferably reduced in writing so as to become a treaty. As a treaty, it is now enforceable on its own accord. At the same time, the settlement agreement should be reported to the DSB, so that it can be placed on the regular surveillance list, placing more pressure on implementation.

In sum, enforcement is a critical issue for the success of mediation. Mediation's flexibility should be preserved as much as possible throughout the proceeding, but the outcome of such flexible dialogue should preferably turn into a written instrument that the parties vow to implement, which then becomes enforceable subsequently.

Thirdly, as for ‘systematization’, mediation should be explored, designed, and applied in tandem with the existing WTO DSM scheme. What is good for mediation generally may not be so for mediation for WTO disputes because of the unique characteristics embedded in the WTO DSM. Mediation cannot be clinically separated or administered in a vacuum. It can only operate properly when it is well integrated and aligned with the overall architecture of the WTO DSM. This is discussed in detail in the following sub-subsection.

1. (iii) How to Align Mediation with Panel/AB Proceedings?

In practice, one of the critical challenges in introducing mediation to the WTO DSM is how to align mediation with the existing panel/AB proceedings. As mentioned in sub-subsection ii above, ‘systematization’ should be kept in mind when introducing mediation to the WTO DSM, at the core of which lies the possible alignment between new mediation and existing panel/AB proceedings. To repeat a critical point, mediation in the WTO context cannot and should not replace the existing panel/AB proceedings; it should be a supplementary proceeding and parties are free to move on to panel/AB proceedings when they do not agree to conduct mediation or, even if they do, when they fail to reach agreement through mediation. Therefore, any mediation should always contemplate a subsequent or concurrent panel/AB proceeding. Prospective mediation provisions in the DSU or other WTO instruments should be carefully examined to fathom the possible impact on panel/AB proceedings and vice versa. The success of a new mediation scheme hinges on how to align and liaise it with the existing panel/AB proceedings.

In this regard, mediation can be voluntary or mandatory. Mandatory mediation may help enhance the stature of mediation in the overall architecture of the WTO DSM. This, however, seems to place too much weight on mediation. It is feared that such an ‘enhanced’ approach might further sap the vitality of the already suffering panel/AB proceedings. It might be prudent, at least for now, for mediation to be offered as a voluntary choice of disputing parties. Disputing parties have more options that are available both practically and legally. In the meantime, the burden on panels and the AB can be lightened. In view of this, mediation can be an option either in pre-panel/AB, ongoing panel/AB, and post-panel/AB proceedings as the case may be.

Assuming the panel and AB proceedings continue to take the core element of the WTO DSM, it would be critical to introduce mediation in line with the panel/AB proceedings as opposed to an independent or stand-alone proceeding. It would be more appropriate and practical to design mediation in connection with the panel/AB proceedings as much as possible. Mediation should always be contemplated and implemented with upcoming, pending, or completed panel/AB proceedings in mind.

Once mediation is understood as a subsidiary to panel/AB proceedings, the time assigned for mediation can be reduced to a minimum. During the pendency of mediation, the time clock for a panel or AB proceeding should be suspended. If a mediation proceeding is subject to a short timeframe, with a narrow possibility of extension with good cause, it would help ensure to avoid inserting another lengthy proceeding into the already delayed and saturated WTO DSM, which poses a real concern for mediation in the WTO DSM.

In this context, if mediation is initiated during a panel or AB proceeding, a mediator can be appointed from the three panelists or three AB members hearing the case. Or the panel or the AB may recommend a mediator. In the alternative, the WTO Secretariat may maintain a roster of mediators. If a standing roster for panelists is somehow prepared in future WTO DSM's, an expert from the roster may also be a good candidate to serve as mediator. In any event it is the decision of the disputing parties whom to choose as mediator.

Often times, mediation clauses include a non-reference obligation in subsequent panel/AB proceedings. Parties are not allowed to refer to the contents of mediation at the panel or AB proceedings. At times, what is prohibited from being referred to is enumerated. By way of example, in the mediation of NTB disputes in the EU–Korea FTA, the following three issues are prohibited:

1. – positions taken by the other Party in the course of the mediation procedure;

2. – the fact that the other Party has indicated its willingness to accept a solution to the non-tariff measure subject to mediation;

3. – proposals made by the mediator.

In other words, only positions, willingness to compromise, and proposals are prohibited. Other statements and documents are therefore open to subsequent utilization and referral. Broad prohibition of subsequent referral may lead to controversies and disputes as to what is permitted and what is not in subsequent panel/AB proceedings. It is preferable to set forth in the DSU what is prohibited in clear terms in advance.

Furthermore, should mediation be adopted in the middle of the panel or AB proceedings, the non-reference provision (refraining from using statements made in mediation in panel and AB proceedings) may not work in practice, and even if they can work, a complex choreography would be needed. It is possible that statements made in a panel or AB proceeding and those from mediation are commingled and blended at this stage. It may be difficult to tell which is which. Furthermore, there may be instances where parties and the panel may want to consult with the mediators or entities overseeing mediation to better understand the dispute. If that is the case, this avenue may have to be kept open in some instances.¹²² This reality shows the non-referral obligation would warrant further discussion if it could work in practice.

The recent General Council report on the WTO DSM reform also touches upon the importance of the alignment. Members' participation in mediation should not harm their rights and obligations under the DSU or WTO Agreements.¹²³ Similarly, unless the parties agree otherwise, the mediator should not advise or comment on the consistency of the measure at issue with the covered agreements.¹²⁴ Procedures undertaken, advice offered, and solutions proposed should all remain confidential too.¹²⁵ In the same vein, the report states that the outcome of mediation ‘shall be consistent with the covered agreements and shall not nullify or impair the benefits accruing to any member … nor impede the attainment of any objective.’¹²⁶ As regards the non-reference obligation, the report clarifies that a party in a mediation should not rely on, or introduce as evidence, in other dispute settlement procedures under the DSU, any advice or proposed solutions from a mediator or any confidential information obtained from mediation proceedings.¹²⁷ Notably, the same obligation also extends to any other international dispute settlement proceeding beyond the WTO DSM.¹²⁸ It may help enhance the effectiveness of a non-reference promise. In addition, unless the parties agree otherwise, a mediator ‘shall not serve’ as a member of a panel or an arbitrator in other proceedings of the WTO DSM involving the same measure subject to the said mediation.¹²⁹ Mediator's non-participation obligation also extends to other international dispute settlement proceedings to ensure its effectiveness.¹³⁰

To reflect these ideas, the DSU would have to be amended. Article 5 of the DSU needs to be expanded or, if needed, a stand-alone article needs to be introduced so as to stipulate that mediation can be adopted in all proceedings under the DSU; i.e., pre-panel, inter-panel, post-panel, inter-AB, post-AB, implementation and retaliation proceedings. As regards evidence and statements submitted during mediation, a new article would have to be inserted to set forth what is permitted to use in future panel/AB proceedings and what is not. Considering the need for efficiency and streamlining the timeframe, the parties should be permitted to choose their mediator as a panelist if they so wish. Likewise, a panelist should be permitted to serve as a mediator in post-panel mediation, again if the parties so desire. Clear textual language may be inserted in the prospective DSU for these issues. The overall timeframe of the DSB decisions in Article 20 would require further adjustment in case of mediation. A clear stipulation of an overall deadline would help guard against utilization of mediation as a delay tactic by a respondent.

1. (iv) How to Manage Novel Challenges and Obstacles?

Mediation is not problem-free. It brings novel challenges and obstacles to the WTO DSM as well. Benefits should be weighed against these challenges and obstacles. Future discussions should focus on reducing these challenges and overcoming obstacles.

Firstly, mediation should not merely add another layer of procedural components. If so, a mediation proceeding might simply further delay the whole proceeding of the WTO DSM. It might also increase the total cost for utilizing the WTO DSM. Therefore, introducing a reliable procedural guardrail to preempt an unreasonable increase of cost and duration of settling a dispute should receive careful attention in future discussions.

Any introduction of mediation basically means an addition of a new proceeding to the existing panel/AB proceedings. Thus, it naturally tends to lengthen the total time spent for the whole proceedings as much as the days spent for mediation. Assuming mediation is usually scheduled to take 60 days, this change would then mean an extension of as many days in a WTO dispute. Mindful of the effort of Members to reduce the current timeframe of the DSU by eliminating various small segments here and there (such as dispensing with a consultation period of 60 days under Article 4 of the DSU), the new addition of 60 days is by no means trivial. If managed improperly, it may run the risk of exacerbating the current situation. Under these circumstances, if mediation is merely inserted into existing dispute settlement proceedings as another procedural layer, then the whole cost and duration might also end up increasing. In particular, a real concern in this regard is a situation where a responding party suggests or agrees to mediation only to buy time. Such a delaying tactic would undermine the integrity of mediation and the whole WTO DSM.

The addition of another layer would only find a rationale if the system is so designed as to induce robust mediation capable of resolving underlying disputes early on and avoid the possible abuse of process effectively. Some ideas that can be explored in this respect would be (i) to prepare a standing roster of mediators with respected backgrounds and proven expertise, (ii) to introduce a mediation proceeding with a short timeframe and a strict deadline which can only be extended by both parties' consent, (iii) to enable a party to terminate mediation as soon as deemed necessary at any time of the proceeding, (iv) to permit a party to request skipping consultation before a panel is established when mediation has been selected, and (v) to allow a party to disclose basic facts about mediation such as time, location, participants and key arguments (except the contents of mediation) to a subsequent panel proceeding. These are merely exemplary and other creative ideas are needed to expedite the entire proceeding and block any attempt to abuse the process. Prospective mediation in the WTO DSM with these elements would help address the ‘additional procedural layer’ problem.

Secondly, the WTO DSM reform discussions and recent FTAs' DSM experiments underscore transparency. This is a reflection of the emerging trend of trade agreements to cater to the demand of various stakeholders for more information and updates on disputes of national concern. Increasingly, documents and hearings are open to the general public, and non-governmental entities' views are solicited during proceedings. Mediation, however, is not susceptible to transparency; it is generally based on the confidentiality obligation of the participants involved.¹³¹ As mediation finds its new place in the WTO DSM, this ideological fissure will rise to the surface.

As both transparency and confidentiality serve respective legitimate objectives, the question is how to find a balance between the two in designing future mediation proceedings in the WTO DSM. One solution would be to disclose at least the basic facts of mediation to the general public. For instance, the fact that mediation is or has been conducted, names of mediators and parties' representatives, duration and location of mediation, and success or failure to reach an agreement may be disclosed. Substantive matters of mediation should not be disclosed in the absence of the consent of the parties. A bifurcated approach such as this one would offer a workable compromise to satisfy both transparency and confidentiality. In order for a compromise to take root, a specific provision needs to be included in a future DSU or relevant instruments to inform parties prior to mediation what types of information will be disclosed.

Thirdly, due process concerns raise yet another challenge for a meaningful discussion of mediation. As noted above, mediation's benefit lies in the flexibility of the proceeding.¹³² Mediators are able to communicate with the parties freely, both unilaterally or simultaneously. Such flexibility, however, may invite criticism that a particular mediation proceeding has violated the due process principle. One example would be a mediator's engagement with one party alone, which is not permitted in litigation or arbitration.¹³³ In mediation, however, ex parte communication is frequently utilized in the course of finding common ground.¹³⁴ Similarly, there is no obligation on the part of a mediator to disclose the contents of the discussion with one party to the other party. He or she can spend more time with one of the parties discussing the subject matter. A mediator can also reach out to other experts or sources of expertise to get information as long as the confidentiality obligation is not breached. As a matter of fact, mediators do not have to be lawyers in the first place. Useful though it may be, the inherent flexibility associated with mediation raises a due process concern. Ensuring due process principles while also preserving flexibility is another critical issue to facilitate mediation in the WTO DSM.

Assuming that we are unable to change the inherent traits of mediation, the Members could instead turn to regulation of mediators. Selection of reliable mediators for WTO disputes, strong ethics guidelines and rules, the possibility for parties' peremptory challenges, and possible sanctions by the DSB if good cause is shown may be contemplated in this regard. Provisions to that effect need to be stipulated in a new DSU or WTO instruments. Mindful of this, the wisdom of adopting the same ethics standards for both arbitrators and mediators in some recent FTAs can be questioned.¹³⁵ Given the unique function performed by mediators, ethics standards for mediators and those for arbitrators may overlap in some areas but may not in other areas. Separate standards for mediators are arguably more appropriate and mediation in the WTO DSM needs them as well.

Fourthly, as for mediation, enforcement should be guaranteed as much as possible. Even if mediation has resulted in a settlement, if one party does not abide by the deal the dispute remains unresolved and the other party will have to proceed to a panel proceeding anyway. Such being the case, an enforcement scheme is important for any prospective mediation. Ensuring enforcement of the settlement agreement between disputing Members reached by mediation is important for the utilization of mediation in the WTO DSM. Granted, mediation is a non-binding proceeding. As explained above, however, it is still possible and even common that parties agree to a settlement package in which they commit to performing specific actions. Depending on the contents, this settlement agreement can be enforceable.

For this to happen, disputing parties' agreement should preferably be reduced to a text, confirmed by a mediator, and established as a treaty satisfying the requirements of Article 2 of the VCLT. These elements can be included in the future formulation of mediation proceedings in the WTO DSM. If not in the text of the DSU, a recommendation together with consequences can be included in relevant WTO instruments to alert the Members to the importance of enforcement.

Another avenue to explore would be to utilize a panel proceeding for enforcement purposes. A settlement agreement resulting from mediation can be submitted to a panel and the panel, after a brief confirmation procedure, can record it in a short panel report, which is then adopted by the DSB to proceed for regular implementation under the DSU. In the alternative, the settlement agreement may be directly reported to the DSB which will then christen the instrument to be implemented within the specific timeframe. The DSB discussions and conclusions should be recorded in the DSB documents. In short, the general notion that mediation is not capable of being enforced is not necessarily accurate, and there seem to be diverse ways to address the issue, as evidenced in the Singapore Convention and ISDS reform discussions at UNCITRAL.

As mediation has not yet been fully explored or digested on the trade dispute front, there will be many novel questions and untested problems going forward. The WTO Members will have to adjust mediation proceedings using their tools in the DSB and Ministerial/General Council as necessary. As with anything else, a learning curve and ensuing adjustment will be necessary here as well. Perhaps, it is against this backdrop that mediation provisions of recent FTAs have tended to contain review provisions.¹³⁶ The WTO DSM should also contain a review provision to check the status and adjust accordingly.

5. Conclusion – Mediation as a Viable Component in a Future WTO DSM

The DSM has been regarded as the jewel in the crown of the WTO regime for the past 28 years. It represents the ‘rule of law’ in the multilateral trading regime. The core elements of the DSM should be preserved for the future of the WTO and multilateralism. Most notably, its AB should be restored to resume the appellate review function as soon as possible. At the same time, the Members have also observed recently that the WTO DSM has its inherent limitations and shortcomings in handling and resolving some complex and sensitive disputes among the Members. Over the years, panels and the AB have reviewed these disputes and rendered their decisions. Official closure of proceedings notwithstanding, sometimes underlying disputes remain unresolved and new ones in the same vein have kept arising.

In light of this experience, it is time to contemplate whether ADRs will help address these disputes. In addition, the WTO DSM reform discussion has been currently underway, providing an opportunity to explore various ADRs in the WTO DSM, on top of the current panel/AB proceedings. This is not to substitute for or replace the binding panel/AB proceedings but instead to provide an additional option to Members to resolve certain disputes through different avenues, i.e. ADRs.

Among various ADRs, mediation deserves careful attention. As an ADR with a more structuralized format than other ADRs, mediation is regarded as the most advanced version of a non-binding proceeding. It involves a neutral third person, procedural guardrails and, notably, an enforcement mechanism. Accordingly, mediation is gaining increased attention at various fora in the global community, spanning commercial, investment, and state-to-state disputes. The rationale for mediation in these disputes can arguably apply to WTO disputes with equal force. The benefits of mediation will help the WTO Members manage and resolve their disputes in a more efficient manner. The Members will have an additional option in addition to binding proceedings of panels and the AB.

At the same time, mediation in the WTO DSM also raises various concerns. Further delays, confidentiality traps, due process myriads, and enforcement loopholes are such examples. While they pose legitimate concerns which may test the integrity of any mediation proceeding if left unbridled, proper formulation of provisions will be able to avoid or mitigate them. More specifically, mediation provisions should be elaborated so that parties know in advance how mediation will proceed. Also, in order for mediation to be employed in real disputes, it should shed its current ad hoc feature so as to become systematized with proper personnel, procedures and facilities put in place beforehand or readily available on short notice. Finally, mediation should be structuralized within the overall architecture of the WTO DSM to exist and operate in tandem with the current panel/AB proceedings, ensuring a two-way interaction throughout. If the current discussion on the WTO DSM reform manages to attain the required elaboration, structuralization, and systematization, mediation will have a good chance of taking root in the WTO DSM and offering an alternative path to resolve certain complex and sensitive disputes.

As the WTO's authority is weakening and multilateralism faltering, it is imperative to reinvest the DSM with its previous trust and confidence. At the core of this objective lies establishing a WTO DSM capable of resolving disputes among the Members satisfactorily and efficiently. A carefully crafted form of mediation will provide a viable option in this regard.