The lack of a binding effect of a State’s right to interpret treaties has been raised as one of the reasons to reform the current Investor-State Dispute Settlement (“ISDS”) regime. The movement to reform the current ISDS regime led to the UNCITRAL Working Group III discussion (“WG III discussion”), which has been addressed in this blog multiple times. The WG III discussion considers both incremental and structural ISDS reform. The incremental reform approach focusses on improving some features of the ISDS system without structurally changing the current ISDS regime. On the other hand, structural reform, which is mainly supported by the EU, suggests to introduce a new ISDS mechanism (i.e., creation of a multilateral investment court).

Among various incremental reform approaches submitted by member States, WG III has focused on discussing the creation of an advisory centre, treaty parties’ involvement and control mechanism on treaty interpretation, the creation of a dispute prevention and mitigation mechanism, and arbitrators’ appointment methods and ethics. On 4 June 2020, the WG III held a webinar to discuss treaty parties’ involvement and control mechanism on treaty interpretation. This post further examines issues discussed in the WG III webinar and focuses on the binding effect of a joint or multilateral interpretation on a tribunal in a pending case.

 

Current Framework on States’ Right to Interpret a Treaty

Article 31(3) of the Vienna Convention on the Law of Treaties (“VCLT”) provides a general rule of interpretation, which requires to consider any subsequent agreements on interpretation (joint or multilateral interpretation) between the parties. However, a joint or multilateral submission on treaty interpretation under the VCLT will not bind a tribunal. It merely asks a tribunal to consider States’ submissions on treaty interpretation.

On the other hand, the Contracting States could amend the treaty pursuant to Articles 39 and 40 of the VCLT. When the treaty is amended it will be binding on a tribunal. However, amendment of the treaty will not retroactively bind a tribunal in a pending case. The process to amend a treaty could be complicated and time consuming.

To better address States’ concern on its right to interpret a treaty, some modern Investment Agreements (“IAs”) include specific provisions on a States’ right to treaty interpretation. Contracting States can submit unilateral, joint, or multilateral submissions on treaty interpretation. These submissions can be either binding or advisory depending on the specific treaty. For example, the new United States-Mexico-Canada Agreement (“USMCA”), the Free Trade Commission could submit its interpretation on the treaty provision, which will be binding on the tribunal. Additionally, the USMCA also allows a non-disputing party to submit its interpretation on the treaty provision.

Some other modern IAs provide both options. For example, the United States – Korea Free Trade Agreement (“FTA”) provides both options. Similarly, the India Model BIT provides both options.

In general, a unilateral State interpretation would not bind the tribunal. A non-disputing State’s interpretation could be used as guidance by the tribunal. For example, in the USMCA and other modern IAs such as the Korea – Vietnam FTA, the United States – Korea FTA, and the India – Kyrgyzstan Bilateral Investment Treaties (“BIT”). However, in practice, non-disputing State submissions are not utilized often.1)See Lise Johnson, The Role of States in Treaty Interpretation (Columbia Centre for Sustainable Investment (CCSI)) on Webinar on Treaty Parties’ Involvement and Control Mechanisms in Treaty Interpretation, UNCITRAL WG III, 4 June 2020 (out of 480 concluded IIA ISDS cases, only 10 cases had non-disputing State submission on treaty interpretation). A study by the Columbia Centre for Sustainable Investment (“CCSI”) identified that the reasons for the lack of non-disputing State submission were:

(1) cost-benefit calculations;

(2) lack of knowledge of the pending cases;

(3) political considerations;

(4) the treaty does not allow non-disputing State submissions.

In other circumstances, a non-disputing State may be reluctant to issue an interpretation of the treaty when its position conflicts with the investor’s position.

A joint or multilateral interpretation could become binding on a tribunal if it is stated as such in the applicable treaty.2)It is noted that 126 out of 2,573 (4.9 percent) analyzed treaties contain provision including binding effect of the treaty interpretation by Contracting States’ or interpretative committees. See WG III working document A/CN.9/WG.III/WP.191, para 38 (referring to UNCTAD’s data). For instance, Article 11.22 (3) of the United States – Korea FTA provides:

“A decision of the Joint Committee declaring its interpretation of a provision of this Agreement under Article 22.2.3(d) (Joint Committee) shall be binding on a tribunal, and any decision or award issued by a tribunal must be consistent with that decision.”

However, when should a joint or multilateral submission on treaty interpretation become binding? Would it bind the tribunal in pending cases? Can the State submit a treaty interpretation after the dispute has arisen?

Whether a Joint or Multilateral Interpretations Bind a Tribunal in a Pending Case3)For Contracting States’ submission on treaty interpretation on various time lines such as during treaty negotiations, after treaty negotiations see Catherine Titi, The Timing of Treaty Party Interpretations A Treaty-Design Perspective on Webinar on Treaty Parties’ Involvement and Control Mechanisms in Treaty Interpretation, UNCITRAL WG III, 4 June 2020.

Both unilateral and joint or multilateral interpretations can be released even after a dispute arises. Given that a unilateral interpretation will typically only be utilized as guidance by a tribunal, closer consideration is needed on the timing of the binding effect of a joint or multilateral interpretation on a tribunal. This could be particularly controversial if a joint or multilateral interpretation binds a tribunal in a pending case.

Whether a tribunal in a pending case will be bound by a joint or multilateral interpretation depends on the specific treaty. For instance, The Dutch Model BIT expressly prevents the binding effect on a pending case:

“A joint interpretative declaration adopted as result of consultations by the Contracting Parties shall be binding on a Tribunal established under Section 5 of this Agreement. Such joint interpretative declaration is not applicable in cases where a Tribunal was already established.” (Article 24.2)

Some IAs are silent on the binding effect on a pending case (such as the United States – Korea FTA and the India – Kyrgyzstan BIT). However, a joint interpretation mechanism under these treaties suggests that the interpretation may be binding on a tribunal in a pending case.

Regarding the EU – Canada Comprehensive Economic and Trade Agreement (“CETA”), Article 8.31 provides that the joint committee’s interpretation “shall be binding on the Tribunal established under this Section. The CETA Joint Committee may decide that an interpretation shall have binding effect from a specific date.” On 30 April 2019, the Court of Justice of the European Union (“CJEU”) held that the CETA Joint Committee’s interpretation does not retroactively bind the tribunal. The Court reasoned as follows:

It is important, moreover, in the light of the requirement of independence of the CETA Tribunal and Appellate Tribunal, that interpretations determined by the CETA Joint Committee have no effect on the handling of disputes that have been resolved or brought prior to those interpretations. If it were otherwise, the CETA Joint Committee could have an influence on the handling of specific disputes and therefore participate in the ISDS mechanism. (para 236)

[A]ccordingly, Article 8.31.3 of the CETA cannot be interpreted, having regard to Article 47, as permitting the Union to consent to decisions on interpretation of the CETA Joint Committee that would produce effects on the handling of disputes that have been dealt with or are pending. (para 237)

Even if the CETA Joint Committee’s interpretation will not be binding on a tribunal in a pending case, a tribunal can still consider the interpretation as a reference according to Article 31(3) of the VCLT.

 

Should a Joint or Multilateral Interpretations Bind a Tribunal in a Pending Case?

During the WG III discussion, States submitted to seek a binding effect of joint interpretations on a tribunal. However, whether a joint or multilateral interpretation should bind a tribunal in a pending case was not specifically addressed in the WG III working documents.4)This issue has been discussed by the panels in the WG III webinar, but working documents do not address this issue.

Giving binding effect of joint interpretations could contribute to greater consistency and predictability in treaty interpretation. Contracting States could also clarify their intention on specific provisions.

On the other hand, it is important not to compromise investors’ rights under the treaty. The retroactive application of a joint interpretation in a pending case could compromise foreign investor’s rights. This is because the joint interpretation did not exist when an investor made its investment, nor when the dispute was filed. Thus, requiring a joint interpretation to be binding on a tribunal in a pending case could compromise substantially the investor’s reasonable anticipation of its rights under the treaty.

Moreover, as pointed out by the CJEU, the independence of a tribunal is an important factor to consider. If a joint interpretation could bind a tribunal in a pending case, this could compromise the tribunal’s ability to adjudicate a dispute between the parties. By agreeing to ISDS in IAs, Contracting States confer to a tribunal the power to resolve disputes between foreign investors and States. ISDS tribunals have a duty to determine a case, interpret a treaty and apply it. Upon making this determination, a tribunal should independently and impartially make its decision. If a joint interpretation becomes binding on a tribunal in a pending case, the tribunal arguably will not be able to independently decide on the proper interpretation of a treaty and apply it to a pending case.

Thus, the WG III in a subsequent discussion on the States’ right to interpret a treaty should consider these concerns carefully.

 

Final Remarks

The purpose of addressing the issues of a joint or multilateral interpretation mechanism during the WG III discussion is to assess whether a joint or multilateral interpretation mechanism is necessary and should be introduced into the current ISDS regime, or multilateral court system in the future.

When answering this question, the WG III discussion should carefully consider whether a joint or multilateral interpretation should be binding on a tribunal in a pending case. In doing so, the WG III should especially give a serious concern on finding the right balance between the States and foreign investors’ rights. Additionally, the concern on compromising a tribunal’s ability to independently interpret a treaty should also be addressed.

If the WG III concludes on a necessity to provide a mechanism on joint or multilateral interpretation, it should further consider an effective way to incorporate this reform to existing IAs. The Mauritius Convention on Transparency, which implemented the UNCITRAL Rules on Transparency in Treaty-based Investor-State Arbitrations on existing IAs, could provide an illustrative way on how to introduce a joint or multilateral interpretation mechanism for the existing ISDS regime.

Finally, allowing Contracting States to submit treaty interpretations during ISDS proceedings adds an additional procedural step which could lead to increased costs and time. Additionally, in practice, Contracting States may not reach an agreement on treaty interpretation within a reasonable time when the dispute is pending. Thus, it is important to safeguard the procedural efficiency of the proceedings. Effective tools could include time limits. For instance, the United States – Korea FTA provides a 60 days’ limit to submit joint interpretations. This kind of time limit would minimize procedural delays caused by the treaty interpretation.


________________________

To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here. To submit a proposal for a blog post, please consult our Editorial Guidelines.

References

References
1 See Lise Johnson, The Role of States in Treaty Interpretation (Columbia Centre for Sustainable Investment (CCSI)) on Webinar on Treaty Parties’ Involvement and Control Mechanisms in Treaty Interpretation, UNCITRAL WG III, 4 June 2020 (out of 480 concluded IIA ISDS cases, only 10 cases had non-disputing State submission on treaty interpretation).
2 It is noted that 126 out of 2,573 (4.9 percent) analyzed treaties contain provision including binding effect of the treaty interpretation by Contracting States’ or interpretative committees. See WG III working document A/CN.9/WG.III/WP.191, para 38 (referring to UNCTAD’s data).
3 For Contracting States’ submission on treaty interpretation on various time lines such as during treaty negotiations, after treaty negotiations see Catherine Titi, The Timing of Treaty Party Interpretations A Treaty-Design Perspective on Webinar on Treaty Parties’ Involvement and Control Mechanisms in Treaty Interpretation, UNCITRAL WG III, 4 June 2020.
4 This issue has been discussed by the panels in the WG III webinar, but working documents do not address this issue.
This page as PDF

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.