Shoehorning Substance into a Procedural Mandate? The Right to Regulate and UNCITRAL Working Group III

This post addresses Draft Provision 12, on ‘Right to Regulate’, in the recent ‘Draft Provisions on Procedural and Cross-Cutting Issues’ that were published by the UNCITRAL Secretariat in late July and will be discussed at UNCITRAL Working Group III’s next session in October 2023 (A/CN.9/WG.III/WP.229 [11]). The Draft Provisions have been ‘prepared for inclusion in existing and future’ International Investment Agreements (IIAs) (A/CN.9/WG.III/WP.231, [3]). While Draft Provision 12(1) contains a very standard right to regulate clause, similar to those found in many newer IIAs, the next two paragraphs are far more novel. These provide:

2. When assessing the alleged breach by a Contracting Party of its obligation under the Agreement [the relevant IIA], the Tribunal shall give a high level of deference that international law accords to Contracting Parties with regard to the development and implementation of domestic policies, the right to regulate in the public interest and the right to adopt, maintain and enforce measures sensitive to the protection of public health, public safety or the environment, the promotion and protection of cultural diversity, or […].
3. No claim may be submitted for resolution pursuant to Draft Provisions 3 or 4, if the measure alleged to constitute a breach of the Agreement was adopted by the Contracting State to protect public health, public safety or the environment (including compliance with the Paris Agreement or any principle or commitment contained in articles 3 and 4 of the United Nations Framework Convention on Climate Change), the promotion and protection of cultural diversity, or […].

In a nutshell, Draft Provision 12(2) is an attempt to codify certain ideas around deference that have been endorsed, in various formulations, in the case law of many tribunals constituted under IIAs (see eg Shirlow (2021), Fahner (2020) pp. 71–84, Paine (2019) and (2022), Henckels (2015)). The reference to the ‘high level of deference that international law accords to Contracting Parties with regard to the development and implementation of domestic policies’ bears important similarities, even if it is not identical, to some of the most commonly cited arbitral formulations (eg SD Myers v Canada (2000) [263]). While most existing IIAs do not attempt to codify a generally applicable notion of deference, several of India’s recent IIAs include a provision similar to Draft Provision 12(2). These Indian IIAs, in a provision on ‘Burden of Proof and Governing Law’, provide that ‘[t]his Treaty shall be interpreted in the context of the high level of deference that international law accords to States with regard to their development and implementation of domestic policies’ (see eg India–Belarus BIT (2018) art 23.1, India–Kyrgz Republic BIT (2019) art 23.1, India–Taiwan BIT (2018) art 22.1, all reproducing 2015 India Model BIT art 23.1).

Draft Provision 12(3) is a carve-out from any dispute settlement mechanism contained in the relevant IIA, and from the State–State dispute settlement mechanism established in Draft Provision 4 (see below), for measures that pursue a broad range of public purposes. In the second part of this post, we comment in more detail on this carve-out proposal. However, before doing that, we retrace how UNCITRAL Working Group III, a process deliberately focused on procedural reforms to investor–State dispute settlement (ISDS), ended up addressing the above issues.

Regulatory Chill within ‘Other Issues’ in UNCITRAL Working Group III

Draft Provision 12 responds to concerns about regulatory chill that have been raised at several points in the Working Group III process (A/CN.9/WG.III/WP.232, [30]). For example, in Working Group III’s September 2022 meeting, in a discussion of ‘other cross-cutting issues’, ‘regulatory chill’ was identified as one of several issues ‘requiring further work’ (A/CN.9/1124, [101]–[104]). In April 2019, when Working Group III concluded work on the first two stages of its mandate – regarding identifying concerns over ISDS and potential areas where reform was desirable – and shifted to the present stage of developing reforms, there was also a discussion of ‘other concerns’ where the issue of regulatory chill was raised. At that point it was agreed regulatory chill ‘would not be addressed at this stage as a separate concern by the Working Group, while the potential impact of ISDS on the regulatory policy of States should guide the work on ISDS reform’ (A/CN.9/970, [36]–[37]), see also Langford and Roberts’ account of the 2019 discussion of ‘other concerns’. An intersessional meeting of the Working Group III in South Korea in September 2021 also addressed the various ‘cross-cutting issues’ with presentations on regulatory chill and the right to regulate (A/CN.9/WG.III/WP.214, [50]–[51]).

It is clear from reports of the Working Group’s sessions that in discussion of ‘other issues’ some interventions have emphasised the Working Group’s mandate extends only to procedural reform of ISDS, not reform of the substantive standards contained in IIAs (see eg A/CN.9/970 [27]). Not all States participating in Working Group III agree with this interpretation of the mandate. For example, South Africa’s submission to the Working Group challenged this position, arguing that ‘the Working Group would not be fully discharging its mandate if discussions on the substantive concerns were excluded’, and, among other proposed reforms, suggested carve-outs or exclusions from ISDS to prevent challenges to ‘legitimate public interest regulations’ (A/CN.9/WG.III/WP.176, [19]–[20], [56]–[57], [63]). Indonesia’s submission also criticised the Working Group’s procedural focus, noting ‘procedural law is inherently substantive and vice versa. Substantive and procedural provisions in the international investment agreements (IIAs) are intertwined in nature’. Indonesia highlighted the issue of regulatory chill and outlined numerous reform options arising from its review of its IIAs including ‘exclusion of claims’ (A/CN.9/WG.III/WP.156, [1]–[2], [15]–[16]). Thailand’s submission also urged the Working Group to remain open to reforms on substantive issues, suggesting it might develop model substantive provisions that could be used by States (A/CN.9/WG.III/WP.162, [28]–[29]).

If we rewind to the decision to grant the Working Group a mandate on ISDS reform at UNCITRAL’s annual meeting in July 2017, even at the stage there were a diversity of views, ranging from interventions that stressed that work on ISDS reform should not be limited to procedural issues, to those noting ‘work on substantive standards was deemed less feasible than work on the procedural aspects’, and some interventions that were sceptical of UNCITRAL undertaking work on ISDS reform at all, highlighting prior failed attempts to negotiate multilateral investment rules (A/72/17, [244]–[245], [257]). At UNCITRAL Working Group III’s first session in December 2017 there was also a move to highlight, at the start of the session, ‘that the mandate given to the Working Group focused on the procedural aspects of dispute settlement rather than on the substantive provisions’ in IIAs (A/CN.9/930/Rev.1, [20]).

We retrace the above chronology for two reasons. First, in our view, it highlights a significant point of ambiguity and tension regarding to what extent Working Group III’s agenda can extend beyond procedural aspects of ISDS reform. Second, it provides important background on the concerns that the new Draft Provision 12 on Right to Regulate may be responding to. In this light, it appears Draft Provision 12 may serve an important political function, simply by being part of the secretariat draft on cross-cutting issues, even if it is not viewed as acceptable in its current form by most states who participate in the UNCITRAL process.

We are not the first to highlight this tension or ambiguity in Working Group III’s mandate. For example, Bonnitcha et al (2023), in discussing how damages came to be added to the Working Group III agenda, describe something very similar, noting ‘there has been a practical acknowledgment that the distinction between procedural and substantive matters can be illusory’ (at 214–216). Roberts and St John described South Africa as a ‘widener’ given its efforts to ‘widen the terms of the UNCITRAL mandate’ (here and here). Langford et al. (2020) also observed that ‘this mandate dispute has been resolved pragmatically and with reliance on the strategic ambiguity in wording’ (at 173–174). See also the discussions of this issue by Arucri and Violi (2019: 9–15), Kelsey, Schneiderman and Van Harten (2019: 2–3), and Polonskaya (2020: 965–69).

Draft Provision 12(3): A Sweeping Carve-Out for Public Interest Measures

The carve-out proposed in Draft Provision 12(3) is an original proposal that is far wider than the carve-outs seen in IIAs concluded to date. Perhaps because Working Group III’s mandate focuses on procedural reforms to ISDS, the provision is phrased as a carve-out from dispute settlement, rather than a carve-out from the scope of coverage of the relevant IIA. The provision would prevent claims challenging a range of public interest measures under any dispute resolution mechanism(s) provided in relevant IIA (Draft Provision 3 is a placeholder for the dispute resolution mechanism(s) provided for in the relevant IIA), and under Draft Provision 4, which enables a State to bring a State–State claim on behalf of one of its investors, alleging that the other Party to the IIA has breached the Agreement and caused loss or damage to the investor (A/CN.9/WG.III/WP.232, [5]–[8]; A/CN.9/WG.III/WP.231 pp. 4–5). As this carve-out would prevent covered measures from being challenged in both ISDS and State–State dispute settlement, there may not be much practical difference from a carve-out that would exclude such measures from the coverage of the treaty.

While carve-outs for measures with specific public purposes are far from unheard of in IIAs, they are usually very narrow (see generally Paine and Sheargold 2023, 291–93). A well-known example is the denial of benefits clause in the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP) which allows treaty parties to block ISDS claims relating to tobacco control measures (art 29.5). Similar tobacco-focused carve-outs have been used in several subsequent Australian and Singaporean IIAs (see Paine and Sheargold 2023, 292), and two Australian IIAs include a broader carve-out from ISDS for all public health measures (Indonesia–Australia CEPA art 14.21(1)(b)), Peru–Australia FTA ch 8 sec B fn 17). Even these relatively narrow carve-outs have been controversial and subject to criticism from some commentators for the risk that they create a loophole which may allow protectionism or unfair treatment of investors (eg Lester and Mercurio, 2017, 6–8).

In light of how narrowly carve-outs have been used in IIAs to date, the breadth of public purposes covered by Draft Provision 12(3) is striking. The proposed carve-out covers measures adopted ‘to protect public health, public safety or the environment … [and] the promotion and protection of cultural diversity.’ It will be interesting to see how States react to this Draft Provision at Working Group III’s next session, but if the history of carve-outs in IIAs is any indication, it seems likely that many States will object to the breadth of the carve-out. A particular emphasis in the Draft Provision is given to the issue of climate change, as the text specifies that protecting the environment includes compliance with the Paris Agreement and Articles 3 and 4 of the United Nations Framework Convention on Climate Change (UNFCCC). The urgent need for IIA reform to protect climate policy space and align the investment treaty regime with the Paris Agreement has been highlighted in recent years by relevant international organisations (see eg ‘Track 1’ of the OECD’s work programme on the Future of Investment Treaties; OECD 2021, pp. 16–19; UNCTAD 2022). A risk of Draft Provision 12(3) is that covering such a broad range of public interests will make it unlikely that many states will support the adoption of the provision. Carve-outs that are more narrowly targeted are more likely to receive the political support necessary for urgent action.

Draft Provision 12(3) starts with the words ‘[n]o claim may be submitted for resolution’, which is similar phrasing to several of the ISDS carve-outs for tobacco control or public health measures noted above. For example, the Singapore–Australia FTA, as amended in 2016, provides that ‘[n]o claim may be brought’ through the ISDS mechanism in respect of a tobacco control measure (Ch 8, art 22). Carve-outs phrased in this way typically do not specify what should happen if an investor brings a claim that the respondent State believes is covered by the carve-out (although see Indonesia–Australia CEPA art 14.21(2), which permits the respondent to raise the application of various exclusions from ISDS, including the public health carve-out, as a preliminary objection). In the absence of a clear alternative procedure being set out in the treaty itself, the relevant investor–State or State–State tribunal would end up adjudicating over whether the claim falls within the scope of the carve-out. As we have argued elsewhere, States who are considering public interest carve-outs for their IIAs should also incorporate procedural mechanisms that allow the application of the carve-out to specific investor claims to be determined by the authorities of the treaty parties, or if they cannot agree, through State–State dispute settlement before a tribunal with relevant subject-matter expertise (Paine and Sheargold 2023, esp. 300–304). Some IIAs already establish procedures like these for determining whether a measure is covered by an exception for prudential or monetary policies (see eg CPTPP, art 11.22(2); Canada–Hong Kong BIT art 22(3)–(4)).

These procedural mechanisms are important because they remove questions that require subject-matter expertise or are politically sensitive from the authority of ISDS tribunals (van Aaken 2015, esp. 42–44). The public interests mentioned in Draft Provision 12(3) – including public health, public safety, the environment and cultural diversity – are all policy areas which are outside the expertise of most ISDS adjudicators, and a strong case can therefore be made that the applicability of a carve-out to these measures should not be determined by an ISDS tribunal (Paine and Sheargold 2023, 302–3). An additional concern arises in relation to climate change measures. Draft Provision 12(3) specifically mentions measures taken to comply with the Paris Agreement or Articles 3 and 4 of the UNFCCC, which reflects the approach taken in some earlier proposals for a climate carve-out (see eg Van Harten 2015; European Parliament Resolution 2015/2112(INI), [80]; Brauch et al, art 5.1(3)(j); CIEL, ClientEarth and IISD 2022, [44]). If a carve-out along these lines is not accompanied by a specific procedural mechanism to govern its application, then an ISDS tribunal would have to make findings about whether a measure has been taken to comply with the Paris Agreement or the UNFCCC. Most ISDS tribunals lack climate-specific expertise, and having them interpret and apply international environmental treaties raises systemic concerns.

Even if a carve-out is accompanied by a procedural mechanism that would avoid these issues being adjudicated by an ISDS tribunal, defining the scope of a carve-out by reference to other international treaties creates other potential problems. The UNFCCC regime is not static, as shown by the evolution from the Kyoto Protocol to the Paris Agreement, and if an IIA carve-out is defined by reference to current treaties then its utility may be undermined when the next climate treaty is negotiated. Moreover, the Paris Agreement and the UNFCCC both contain complex and contested principles, such as common but differentiated responsibilities (UNFCCC art 3.1 and 4.1, Paris Agreement art 2.2 and art 4.3), which may complicate the interpretation of the IIA carve-out. A simpler way to frame a carve-out to protect policy space for climate change mitigation measures would be to cover measures that have the purpose of reducing greenhouse gas emissions (Paine and Sheargold 2023, 299).

Conclusion

While we have argued in this post that States who are considering including public interest carve-outs in their IIAs should consider changing or adding to certain aspects of Draft Provision 12(3), the UNCITRAL Secretariat has put forward a significant and novel proposal to address concerns around regulatory chill. Even if in its current form Draft Provision 12 may not be adopted by many States, the debate it sparks, including in discussion at Working Group III’s next session in October, may be important for wider ongoing discussions of IIA reform. Additionally, the inclusion of such a broad carve-out in these Draft Provisions (along with Draft Provision 12(2) on deference) may serve an important political purpose by showing that the Working Group III process is addressing the concerns of those States who believe that the Working Group’s mandate should extend beyond strictly procedural issues.