Hostname: page-component-848d4c4894-75dct Total loading time: 0 Render date: 2024-06-07T20:47:06.643Z Has data issue: false hasContentIssue false

AN INTERNATIONAL LAW PRINCIPLE OF NON-REGRESSION FROM ENVIRONMENTAL PROTECTIONS

Published online by Cambridge University Press:  11 January 2023

Andrew D Mitchell
Affiliation:
Professor, Associate Dean (Research), Faculty of Law, Monash University, andrew.mitchell@monash.edu
James Munro
Affiliation:
Principal Legal Counsel, Anti-Dumping Commission (Australia), j.munro2@gmail.com.
Rights & Permissions [Opens in a new window]

Abstract

A principle that prohibits States from weakening their domestic levels of environmental protection continues to emerge at varying speeds within international trade, investment and environmental law. This article explores the principle's diverse history, rationale and legal expression in each of these domains and finds that its various articulations in different international treaties suffer the same shortfalls and deficiencies. Non-regression clauses may leave the complexities and nuances of implementing environmental protections unaddressed, including identifying and measuring when a regression has occurred and balancing these environmental protections with other legitimate policy and environmental measures. As these clauses are increasingly subject to investor–State and State–State dispute procedures, States expose themselves to heightened liability for changes to their environmental laws, even where those changes might be legitimate and reasonable. The particular emergence of this principle in environmental law offers treaty-makers an opportunity to clarify the rights of States to derogate from otherwise narrowly drafted clauses that require them to maintain their level of environmental protection strictly.

Type
Articles
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
Copyright © The Author(s), 2023. Published by Cambridge University Press on behalf of the British Institute of International and Comparative Law

I. Introduction

A so-called principle of ‘non-regression’ from environmental protections is emerging in public international law. According to this principle, States are prohibited from weakening their domestic levels of environmental protection. The emergence of this principle is timely considering significant rollbacks of domestic environmental protections worldwide.Footnote 1 However, its emergence is curious. This principle's rationale, history, and level of development vary substantially between different international law domains.

In international investment law—the fragmented collection of over 2,800 treaties comprising bilateral investment treaties (BITs), plurilateral investment treaties, and preferential trade agreements (PTAs) with investment chapters—this principle first emerged in the North America Free Trade Agreement (NAFTA) signed in 1992Footnote 2 and has since become ubiquitous and often subject to binding dispute settlement. Its original rationale in international investment law was to prevent industrial relocation resulting from weakened environmental protections, with the underlying concerns flowing from the potential for so-called ‘pollution havens’.

In international trade law—comprising the Marrakesh Agreement Establishing the World Trade Organization (WTO Agreement)Footnote 3 and over 300 PTAsFootnote 4—the principle is absent from the WTO Agreement signed in 1994, but first emerged in the United States–Jordan PTA signed in 2000. More recently, it has become a feature of PTAs involving the United States (US), the European Union (EU), and China as State parties. The principle's original rationale in international trade law was preventing the promotion of exports through the weakening of environmental protections, with the underlying concern grounded in the potential use of environmentally harmful processing and production methods (PPMs) as an element of competitive advantage in trade relations.

In international environmental law—comprising multilateral environmental agreements (MEAs), regional treaties on various environment-related matters, and certain elements of customary international lawFootnote 5—the principle has only begun to emerge in the past few years out of growing concerns about backsliding and lack of progress over major environmental challenges like climate change and biodiversity. Advocates of this principle have drawn on the contemporaneous emergence of a human right to a clean, healthy and sustainable environment—from which no derogation would ordinarily be permitted under international human rights law—to justify a general principle of non-regression from existing levels of domestic environmental protections. The principle's first iteration in international environmental law appositely appeared in a recent Latin American treaty on access to justice and procedural rights in environmental matters.Footnote 6

This article examines the varying normative bases for the non-regression principle and how the same basic concept has arisen in parallel across these three fields of international law, from different sources, rationales, and contexts (sections II, III, and IV below). The implications for the principle's emergence in these domains, especially its potential impact on investor–State and State–State dispute settlement, is also examined.

II. Non-Regression from Domestic Environmental Protections in International Investment Law

A. Background and Rationale

Clauses enshrining a principle of non-regression from domestic environmental protections have become ubiquitous in international investment agreements (IIAs), particularly in the past decade. Over 150 States have subscribed to a non-regression clause in at least one IIA.Footnote 7

The underlying concern to prevent industrial relocation resulting from weakened environmental protections flowed from the potential for ‘pollution havens’.Footnote 8 The original non-regression clause in NAFTA was a response to concerns that, whilst NAFTA would promote investment in Mexico, the US would be unfairly harmed by the loss of the production facilities (and jobs) that relocated to obtain the benefit of lower environmental compliance costs,Footnote 9 and that such relocation could undermine the stronger environmental protections maintained in the US.Footnote 10 In such circumstances, maintaining stronger environmental protections in the US could perversely stimulate more pollution-intensive methods of production and less environmentally-friendly products in the other party to the IIA (eg Mexico)—the very outcomes the stronger protections were intended to ameliorate.

The US's underlying motivation to prevent industrial relocation has persisted in its pursuit of non-regression clauses in subsequent IIAs, including most recently in the Agreement between the United States, the United Mexican States, and Canada (USMCA),Footnote 11 ie the ‘new’ NAFTA.Footnote 12

In the late 2000s, the EU began including NAFTA-like non-regression clauses in its IIAs.Footnote 13 However, the EU cited a wholly different rationale for their inclusion. The EU's pursuit of non-regression clauses was, and continues to be, a function of issue linkage: using its negotiating leverage in the international economic sphere to further sustainable development goals internationally, thereby strengthening global environmental governance.Footnote 14 That said, in the particular context of the Brexit negotiations, the EU's pursuit of a non-regression clause appeared to be motivated more by pragmatic competitiveness-related concerns (akin to the US's advocacy of these clauses) rather than by the normative goal of promoting sustainable development globally.Footnote 15

While the US and EU have been major advocates of non-regression clauses in IIAs, these clauses are now a part of the standard practice of major economies like Japan, China, Korea, Canada, Brazil and Turkey,Footnote 16 as well as regional groupings like the Eurasian Economic Union,Footnote 17 the European Free Trade Area,Footnote 18 and the Southern African Development Community.Footnote 19 They have also become part of the ‘model’ IIAs of economies like Morocco,Footnote 20 Colombia,Footnote 21 and Slovakia,Footnote 22 and have featured in a diverse array of bilateral IIAs between, for instance, Nigeria and Singapore,Footnote 23 the United Arab Emirates and Rwanda,Footnote 24 Argentina and Chile,Footnote 25 and Guatemala and Trinidad and Tobago.Footnote 26

Thus, the proliferation of non-regression clauses in IIAs is not solely a function of asymmetrical power relations in IIA negotiations, whereby larger economies such as the US and the EU are the demandeurs and smaller economies are the unwilling recipients of this principle. While asymmetrical power-relations have certainly played a part, the recent proliferation of non-regression clauses has also taken shape through mechanisms like acculturation and socialisation.Footnote 27 This has generated a critical mass whereby non-regression clauses have become standard in international investment law.Footnote 28

The international investment regime's prototype non-regression clause in the original NAFTA continues to be influential in the drafting and structure of such clauses in more recent IIAs. It provides:Footnote 29

The Parties recognize that it is inappropriate to encourage investment by relaxing domestic … environmental measures. Accordingly, a Party should not waive or otherwise derogate from, or offer to waive or otherwise derogate from, such measures as an encouragement for the establishment, acquisition, expansion or retention in its territory of an investment of an investor ….

In the US's more recent IIAs, the operative obligation has evolved progressively from ‘should’ to ‘shall’, together with the application of formal dispute settlement procedures administered by independent adjudicators.Footnote 30 This shift is epitomised by the revisions made to the original non-regression clause in the USMCA. In contrast to the original NAFTA's use of ‘should’ (signalling its exemption from dispute settlement procedures)Footnote 31, the equivalent clause in the USMCA is framed as a stricter ‘shall’-based obligation. It is subject to binding State–State dispute settlement procedures that permit economic countermeasures in the event of violation.Footnote 32 A similar process of ‘legalisation’ of non-regression clauses over time can be observed in the practice of the EU, China, India, Japan, Canada, and Korea.Footnote 33

A diversity of terminology, definitions, and textual clarifications or carve-outs have led to variations in the scope and application of non-regression clauses across different IIAs. They all nonetheless share the same two-part structure pioneered in the original NAFTA clause above. The first part concerns the existence of a law or other instrument that protects the environment and has been subsequently modified in a way that reduces its level of environmental protection. In that regard, the IIAs of major jurisdictions like the US, the EU, Japan, China, Korea, and Canada replicate the language from the original NAFTA requiring a ‘waiver’ or ‘derogation’ from an environmental lawFootnote 34 in a manner that ‘relaxes’, ‘weakens’ or ‘reduces’ its environmental protections.Footnote 35

The second aspect involves delimiting the circumstances in which the non-regression clause proscribes these modifications to domestic environmental laws. Again, most IIAs replicate the language from the original NAFTA, whereby regressions from such laws are impugned only if they ‘encourage’ investment.Footnote 36 Specifically, the NAFTA clause proscribing regressions where they act ‘as an encouragement for the establishment, acquisition, expansion or retention in its territory of an investment of an investor’ has been replicated frequently,Footnote 37 with ‘to encourage investment’ appearing as a variation in some IIAs,Footnote 38 and with the formulation ‘encourage investment by relaxing’ appearing in the opening statements of principle in many non-regression clauses.Footnote 39

The ordinary meaning of ‘encourage’ is to ‘incite, induce, instigate; in weaker sense, to recommend, advise’, and to ‘stimulate (persons or personal efforts) by assistance, reward, or expressions of favour or approval’.Footnote 40 A tribunal construing the similar concept of ‘promot[ing] investment’ understood it as referring to a ‘duty to create the conditions for the flowing of investments by nationals of one State into the territory of the other State’.Footnote 41 The prepositions ‘as’, ‘to’, and ‘by’ describe the relationship between the ‘encouragement’ on one hand, and the ‘derogation’, ‘waiver’, or other regression, on the other. In particular, these prepositions suggest that, to fall within the scope of the non-regression clause, the regression must be the mechanism through which the ‘encouragement’ is given effect.Footnote 42

However, the use of ‘an encouragement’ instead of ‘the encouragement’ in many iterations of non-regression clauses foreshadows that regression could be part of several influences that ultimately lead to a stimulation of investment. Likewise, the textual features that pertain to ‘investment’ generally as opposed to identifiable investments—such as the clarification in NAFTA that its clause applies to ‘all investments in the territory of the Party’Footnote 43—indicate that such clauses are directed more at the economic conditions that affect capital flows as opposed to individual investment projects.

Accordingly, the second limb of most non-regression clauses regarding the ‘encouragement’ of investment calls for an analysis of whether the regression at issue has changed the conditions of competition for capital between jurisdictions such that inward flows of capital are apt to increase into the jurisdiction regressing from its environmental law.Footnote 44 That said, a minority of non-regression clauses include textual features that refer to specific investments or a more prescriptive role for the ‘encouragement’ as the deliberate and exclusive vehicle for conferring a competitive advantage in capital flows.Footnote 45 Such features may warrant a more restrictive interpretation of the ‘encouragement’ limb. Accordingly, the specific terminology and formulation for the ‘encouragement’ limb would likely be determinative to its scope and meaning. In general terms, however, the key question at this second stage is whether the regression at issue stimulates investment through the putative host State becoming a more competitive destination for capital.

A minority of IIAs take a laxer approach by omitting any need for an increase in capital flows and instead impugn regressions that ‘affect’ flows of capital between jurisdictions in some way.Footnote 46

B. Implications of a Principle of Non-Regression in International Investment Law

As a discipline of public international law, international investment law is somewhat unique in that most IIAs permit private investors of one State party to bring claims of IIA violation directly against the other State party in whose territory its investment is located (referred to as ‘investor–State dispute settlement’) (ISDS). The emergence of non-regression clauses in international investment law is significant because States could be exposed to additional liability in ISDS proceedings. This additional exposure could be manifested in one of two ways. First, some IIAs permit the non-regression clause to itself form the basis of a claim in ISDS proceedings (section II.B1). Secondly, non-regression clauses could play a decisive role as interpretive context in an arbitral tribunal's examination of claims under the fair and equitable treatment (FET) standard (section II.B2).

The potential for non-regression clauses to play these roles in ISDS proceedings is far from abstract. Rather, in recent years many States have sought private investment in certain sectors and activities to fulfil environment-related public objectives, often establishing regulatory frameworks to incentivise and induce such investment.Footnote 47 Subsequent State actions that have weakened or undermined such regulatory frameworks have formed the basis of numerous recent ISDS claims. For instance, investors have initiated ISDS claims against Canada over the repeal of Ontario's emissions trading scheme,Footnote 48 against Panama over the repeal of a rule requiring gas to be blended with bioethanol,Footnote 49 and against Serbia and the Dominican Republic respectively over actions by their authorities that allegedly undermined waste management schemes.Footnote 50 Moreover, dozens of ISDS claims have been initiated to contest regressions by various States from schemes to incentivise renewable energy.Footnote 51 Whilst many of these renewable energy-related claims—such as those against Peru, Romania, Ukraine and ArgentinaFootnote 52—are ongoing, over two dozen awards have now been rendered in claims against Spain, Italy, and the Czech Republic regarding regressions from their renewable energy schemes.Footnote 53 These awards offer an insight into how non-regression clauses could tangibly affect the outcome of ISDS proceedings, and they are drawn upon in the analysis below.

1. Non-regression clauses as the direct basis for a claim under IIAs

Various bilateral IIAs amongst an assortment of State parties contain non-regression clauses subject to ISDS, including with the provision for retrospective compensation.Footnote 54 These non-regression clauses thus provide a stand-alone basis for ISDS claims against measures by the host State that regress from domestic environmental laws.Footnote 55 This section focuses on how such claims could unfold. At the same time, it is important to note that other IIAs—including under the predominant practice of some major jurisdictions—exempt their non-regression clauses from dispute settlement entirely and thus foreclose any direct ISDS claims,Footnote 56 or limit claims under those clauses exclusively to State–State dispute settlement procedures.Footnote 57 In such instances, non-regression clauses could nonetheless play a role in ISDS proceedings involving other obligations in these IIAs (eg the FET standard) that impugn a host State's regressive measures (see further section II.B2 below).

As mentioned above, dozens of recent ISDS proceedings have involved challenges against measures that weaken the effectiveness of domestic environmental laws. However, non-regression clauses have yet to be invoked as the direct basis of a claim. This is probably due to the absence of non-regression clauses in the older IIAs, under which these regressive measures have been litigated thus far (eg the Energy Charter Treaty and the Czech Republic's IIAs with Germany and the United Kingdom). At a time when there is a concurrent proliferation of non-regression clauses in IIAs and ISDS proceedings involving weakened environmental laws, it is by no means far-fetched to anticipate that non-regression clauses could be invoked as the basis of a direct ISDS claim. Indeed, the regressions from domestic environmental protections currently being litigated in ISDS proceedings under other IIA provisions enliven the kinds of measures and fact patterns that could form the basis of such a challenge.

Despite a degree of variation in terminology and scope, as discussed in section II(A) above, non-regression clauses across IIAs share the same two-part structure and basic legal elements. A complainant would first need to demonstrate that there has been a ‘waiver’ or ‘derogation’ from an environmental law (or some other instrument) in a way that weakens its effectiveness.

On its face, this first element is apparent in the renewable energy-related ISDS disputes against Italy, Spain, and the Czech Republic insofar as they involved laws intending to reduce greenhouse gas emissions that were subsequently repealed or amended in a way that reduced incentives to transition to renewable energy sources. Likewise, the dispute against Canada concerns the repeal of an Ontarian law setting up an emissions trading scheme whose objective was to reduce greenhouse gas emissions. Through such repeals or amendments, these States ‘derogate[d]’Footnote 58 from their environmental laws by effectively reducing—or, indeed, eliminating—the incentives or requirements to achieve the environmental objectives in these laws, and hence their stringency or effectiveness.

As a second element, having established a ‘waiver’ or ‘derogation’ from an environmental law (or other instrument), a complainant alleging a breach of a non-regression clause would need to show that the derogation was undertaken to ‘encourage’ investment. In most IIAs, the concept of ‘encouraging’ investment would be sufficiently broad to encompass creating economic conditions that are attractive to investors and investment generally.Footnote 59 As discussed above, this somewhat capacious reading arises from two textual features of most non-regression clauses. First, most clauses refer to encouraging ‘investment’ generally without any caveats suggesting that the encouragement must, for instance, be targeted at attracting specific investment projects or investment in certain sectors or by particular investors.Footnote 60 Second, most clauses use the phrase ‘as an encouragement’ for investment, rather than stipulating that the regression must be the sole factor encouraging investment in a given instance.Footnote 61

Notably, the repeal of the Ontarian law described above was expressly intended to ‘create jobs’ and ‘allow [businesses] to grow’.Footnote 62 There was thus a clear link between the regression and encouraging investment in that instance. The repeal or amendment of renewable energy schemes by Spain, Italy, and the Czech Republic involved additional intermediary steps between the regression from environmental laws and a putative encouragement of investment. For instance, Spain reduced its feed-in tariffs to ameliorate a broader economic crisis and avoid a default on public debt.Footnote 63 The regression was intended to lead to an improvement in public accounts to in turn stabilise the economy, which would in turn create more favourable conditions for economic growth and thereby stimulate investment. Likewise, the Czech Republic and Italy sought to mitigate dramatic increases in electricity prices in the context of the broader economic crisis,Footnote 64 thus using the regression to improve economic conditions and thereby encourage economic activity.

In such instances, a non-regression clause's applicability would turn on whether the complainant could show that the regression is the mechanism through which an ‘encouragement’ of investment is effectuated. In that regard, most clauses use terms such as ‘by’, ‘to,’ and ‘as’ to denote the relationship between the ‘encouraging’ investment on the one hand, and the regression at issue on the other.Footnote 65 These terms are sufficiently broad to encompass fact-patterns involving intermediary steps between the initial regression and a subsequent ‘encouragement’ of investment, particularly where the underlying objective is to incentivise and stimulate economic activity ultimately. That said, the more intermediary steps between the regression and the encouragement, the more difficult it may be to demonstrate that the regression is the mechanism through which an ‘encouragement’ is given effect.

In terms of remedies, a host State found to have violated a non-regression clause in ISDS proceedings would be liable to pay monetary compensation for the harm caused. This differentiates ISDS claims under non-regression clauses from ISDS claims under the FET standard in relation to regressions from environmental laws of general application. In particular, some ISDS tribunals have found that investors can be compensated under the FET standard only to the extent that a violation exceeds what is proportionate or rational in a given case.Footnote 66 By contrast, there is no equivalent standard of proportionality or reasonableness in non-regression clauses.Footnote 67 Thus, not only do non-regression clauses offer a direct legal avenue for a claim under an IIA, but they also expose host States to greater liability by requiring compensation for all harm suffered by the investor concerning that regression.

2. Non-regression clauses as interpretive context in fair and equitable treatment claims

The exclusion of non-regression clauses from forming the basis of an ISDS claim in several IIAs means that such clauses may be more likely to play a role as interpretative context in ISDS claims under another provision. The FET standard is the most obvious provision in this regard, because regressions from domestic environmental protections that harm investors usually take the form of a State changing its existing regulatory framework. The FET standard comprises several elements, and its precise scope and content is contested.Footnote 68 For present purposes the protection of legitimate expectations held by investors regarding the durability of the host State's regulatory framework is focused upon.Footnote 69

The outcomes of ISDS awards involving Italy, Spain, and the Czech Republic's regressions from their renewable energy laws illustrate how claims against regressions can be pursued under the FET standard. Tribunals adjudicating these claims have usually balanced two considerations in determining whether the rollbacks from the renewable energy schemes infringed an investor's legitimate expectations and thereby breached the FET standard.Footnote 70

On the one hand, tribunals have assessed what an investor can legitimately claim to have expected by reference to the degree of specificity of any assurance given to the investor that the framework would not change.Footnote 71 For instance, a clear and direct promise by governmental authorities that there would be no change to the applicable framework has typically been protected by tribunals as a legitimate expectation under the FET standard.Footnote 72 If a host State subsequently reneged on this promise by changing its regulatory framework, the host State would be required to compensate the affected investor under the FET standard.Footnote 73

On the other hand, absent a specific promise to the contrary, tribunals have considered that investors cannot legitimately expect there would be no change to a framework simply because that framework is enshrined in laws and regulations.Footnote 74 Rather, States have a right to regulate in the public interest. In such circumstances, a host State's change in its regulatory framework would give rise to a breach of the FET standard only if it represents an irrational, arbitrary, or disproportionate exercise of the right to regulate, or the total subversion or recission of the regulatory regime for certain investments.Footnote 75 Against that background, there would be at least two ways that the presence of a non-regression clause in an IIA could provide interpretive context under the FET standard.Footnote 76

First, a non-regression clause could be used as evidence that an investor legitimately did not expect the host State to regress from its domestic environmental laws. In some cases, the absence of a provision in an environmental law indicating that it would not be amended has led the tribunal to conclude that the investor had no legitimate expectation there would not be amendments or modifications in the future.Footnote 77 A non-regression clause could constitute evidence to the contrary. As a legally-binding obligation at the international level, a non-regression clause could be evidence of a host State's commitment to refrain from regressing from its domestic environmental laws in certain circumstances.Footnote 78 It is unlikely that a non-regression clause would alone suffice as evidence of a promise to investors that a given regulatory framework would not change, particularly since such clauses lack the degree of specificity usually attaching to the kinds of promises successfully protected in ISDS. This is because such clauses typically apply to all domestic environmental laws and investments rather than specific schemes and particular investors.Footnote 79 However, the presence of such a clause could augment other evidence, such as public statements and policy documents issued by officials, the registration of investments as qualifying for the regulatory regime, or contractual arrangements between the host State and the investor, which could collectively suffice to evince a legitimate expectation.Footnote 80

Secondly, a non-regression clause could shed light on whether a given regression reflects a reasonable and proportionate exercise of its regulatory power under the FET standard on the one hand, or whether it is irrational and disproportionate—or an unreasonable or total subversion of the regulatory framework—in violation of the FET standard, on the other hand. This is because the very function of a non-regression clause is to constrain a host State's regulatory power. By committing to refrain from lowering environmental standards to obtain a competitive advantage in attracting capital, a State party is effectively excluding such actions from the domain of its legitimate right to regulate under an IIA.Footnote 81 Indeed, Canada and the EU clarified explicitly in a joint interpretative statement that the ‘sovereign right’ to regulate under their IIA did not override their ‘agree[ment] not to lower levels of environmental protection in order to encourage trade or investment’ as enshrined in its non-regression clause.Footnote 82

This is significant because a series of tribunals have rejected ISDS claims under the FET standard against regressions, finding instead that they amounted to a reasonable and proportionate exercise of the right to regulate in the public interest. For instance, tribunals hearing the disputes against Spain, Italy, and the Czech Republic regarding their respective renewable energy schemes accepted that these States acted to protect other public interest objectives like managing crises in public finances, mitigating dramatic rises in consumer electricity prices, and correcting for inaccurate modelling in the design of the laws.Footnote 83 These tribunals accepted the legitimacy of measures whose objective was to safeguard other public interests despite this resulting in a weakening of environmental laws.

By contrast, as discussed further in section IV.(B) below, non-regression clauses are agnostic as to the underlying rationale for a given regression. They make no distinction between, on the one hand, a regression that is solely intended to reduce cost pressures on business and thereby obtain a competitive advantage in attracting capital, such as Ontario's repeal of its climate law, and on the other hand, regressions that are primarily intended to protect other public interest objectives, such as the Spanish, Italian, and Czech renewable energy laws.Footnote 84

Rather, non-regression clauses essentially repudiate the legitimacy of measures that regress from environmental laws and result in an encouragement of investment irrespective of the underlying rationale for such measures. Therefore, the effect of a non-regression clause is to remove such measures from the scope of the range of regulatory actions permissible under that IIA. It is thus difficult to envisage a basis on which a tribunal would accept a measure as a legitimate exercise of regulatory power under the FET standard in cases where a complainant has demonstrated that the measure would be prohibited under the non-regression clause. That approach would be contrary to the basic premise that the provisions of a treaty are cumulative and complementary, and are to be interpreted harmoniously.Footnote 85

In short, the FET standard in a given IIA could be interpreted in light of its non-regression clause such that measures prohibited by the clause are excluded from the host State's legitimate exercise of regulatory power. Such a role for non-regression clauses as interpretative context in construing the FET standard could reflect another way in which non-regression clauses expose host States to additional liability in ISDS proceedings. In particular, the presence of a non-regression clause could be pivotal to whether there is a finding of violation in instances where the regression at issue would otherwise fall within the bounds of proportionality or reasonableness under the FET standard. Moreover, even in instances where a regression would, in any case, violate the FET standard as disproportionate or irrational, the presence of a non-regression clause could inflate the level of compensation awarded. This is because, if the presence of a non-regression clause results in an interpretation whereby no regression is justifiable under the FET standard, a host State would be liable for all harm caused by the regression, as opposed to only harm resulting from aspects of the regression that exceed a proportionate or reasonable exercise of regulatory authority.Footnote 86

C. Summary: Non-Regression Clauses under IIAs

Non-regression clauses have become a ubiquitous feature of IIAs at a time when regressions from domestic environmental laws have led to a significant stream of ISDS disputes. These clauses expose host States to additional liability in ISDS disputes because they offer a direct basis for an IIA claim under some IIAs, and because they could strengthen claims under the FET standard. Moreover, by removing such regressions from the scope of a host State's right to regulate under an IIA, non-regression clauses could also inflate the level of compensation that would otherwise be available under an FET claim. Against that background, the current approach to non-regression clause drafting is problematic. It appears to impugn conduct by a host State that may otherwise seem reasonable, such as taking steps to protect another public interest matter or to respond to a change in circumstances. In that regard, the current preponderance of non-regression clauses is reminiscent of the similarly imprecise and overly-broad drafting of early clauses enshrining the FET standard, which led to unanticipated outcomes in ISDS disputes and subsequent rounds of clarifications and redrafting in more recent IIAs.Footnote 87

III. Non-Regression from Domestic Environmental Protections in International Trade Law

A. Background and Rationale

Clauses in the international trade regime that enshrine a principle of non-regression from domestic environmental protection are generally less prevalent than in the international investment regime. In a dataset on file with the authors, 158 States have subscribed to such a clause concerning investment in an IIA, while 109 States have subscribed to such a clause concerning trade in a PTA.Footnote 88

Such clauses are not necessarily mutually exclusive. In some PTAs, both international trade and international investment are addressed in a single non-regression clause.Footnote 89 Further, a PTA that includes investment obligations in an investment chapter can itself comprise an IIA that incorporates a non-regression clause that is limited to investment.Footnote 90

Despite these ways in which IIAs and PTAs can overlap, non-regression clauses in the international trade regime and those in the international investment regime can be differentiated.Footnote 91 These non-regression clauses have distinct origins and rationales. As described earlier, the first non-regression clause in the international investment regime was included in the original NAFTA. Although the original NAFTA generally contained obligations relating to international trade and international investment, the policy concern underlying its non-regression clause was to prevent industrial relocation resulting from weakened environmental protections,Footnote 92 and was thus limited to flows of investment. It did not extend to international trade,Footnote 93 and was thus not designed to regulate a State party's ability to weaken environmental laws to promote exports.Footnote 94

That notwithstanding, there was a contemporaneous concern of environmentalists around the time of NAFTA negotiations about a perception that international trade law permitted States to use environmentally-harmful PPMs as an element of competitive advantage.Footnote 95 Certain rulings by adjudicatory panels convened under the GATT (the forerunner to the WTO) led to fears that State parties to the GATT would be deterred from differentiating between products based on the environmental impact of their PPMs. Non-regression clauses in the international trade regime were thus seen as a tool to ensure that PPM-related environment protections could not be weakened to enhance the competitiveness of a State party's products in export markets.Footnote 96 International trade non-regression clauses were thus designed to prevent the promotion of exports through weakening environmental protections, with the underlying concern being the use of undesirable PPMs as an element of competitive advantage.

Against that background,Footnote 97 the first non-regression clause covering international trade appeared in the PTA between the US and Jordan signed in 2000:Footnote 98

The Parties recognize that it is inappropriate to encourage trade by relaxing domestic environmental laws. Accordingly, each Party shall strive to ensure that it does not waive or otherwise derogate from, or offer to waive or otherwise derogate from, such laws as an encouragement for trade with the other Party.

Contrary to the original iteration in NAFTA, this non-regression clause was limited to bilateral trade—excluding investment flowsFootnote 99—and was viewed as precedent-setting.Footnote 100 Indeed, the US's legislative history concerning this and subsequent PTA negotiations confirms the distinction between using weakened environment protections to ‘promote exports’ (ie trade) on the one hand, and to ‘attract investment’ on the other.Footnote 101 Likewise, a distinction was drawn between the measures precluded by non-regression clauses that ‘affect [] exports to the United States’ vis-à-vis those that ‘affect [] investment by US persons’.Footnote 102 There was thus a clear delineation in the original development of non-regression clauses between international trade and international investment.

Though distinct, the rationales in the respective spheres of international investment and international trade overlap in certain ways. For instance, the weakening of environmental regulation of PPMs could, in a given circumstance, attract foreign investment in that sector through lower compliance costs. Such a scenario would engage both the policy concern regarding the attraction of investments through regressions from environmental standards and the policy concern regarding the promotion of exports through regressions from the environmental regulation of PPMs. However, there is nothing inherent in the weakening of environmental regulation of PPMs that attracts foreign investment; nor does anything inherent in the attraction of investment due to reduced environmental compliance costs necessarily promote exports. Accordingly, there is only partial overlap between these respective rationales. Both stem from the potential use of environmental protections as an element of competitive advantage, but their origins are distinct and they respond to different policy concerns.

Thus, some PTAs cover international trade and international investment in a single non-regression clause,Footnote 103 but others contain non-regression clauses that differentiate between international trade and international investment.Footnote 104 Some PTAs include non-regression clauses that only address international investment.Footnote 105 Conversely, others include clauses that only address international trade.Footnote 106

While the distinction between the underlying rationales for non-regression clauses in the international trade and investment regimes is apparent, it is less clear why the uptake of non-regression clauses has been muted in the international trade regime. This slower uptake may, at least in part, reflect fewer opportunities in the international trade regime; IIAs are far more common than PTAs.Footnote 107 However, though non-regression clauses covering international trade have recently been included in several large plurilateral PTAs,Footnote 108 they have also been absent from others like the Regional Comprehensive Economic Partnership (signed in 2020 by China, Korea, Japan, Australia, New Zealand, and the members of ASEAN) and the African Continental Free Trade Agreement (signed in 2019 by the members of the African Union).

Accordingly, some States may be hesitant to adopt commitments regulating environmental standards as an element of competitive advantage in international trade. This could represent a vestige of the debates of the late 1990s and early 2000s, in which the regulation of PPMs through environmental standards was posited as a tool of ‘eco-imperialism’ imposed by wealthier countries on developing countries.Footnote 109

Nonetheless, beyond PTAs involving major economies, trade-related non-regression clauses are increasingly emerging in PTAs amongst diverse pairings of States, such as between Korea and Turkey,Footnote 110 Canada and Ukraine,Footnote 111 Australia and Peru,Footnote 112 New Zealand and Malaysia,Footnote 113 Indonesia and the EFTA Member States,Footnote 114 and Chile and Argentina,Footnote 115 to cite a few. Despite the slower uptake of trade-related non-regression clauses vis-à-vis those in IIAs, their proliferation seems to be following a similar trajectory, where the concept is diffusing from its initial advocates amongst a wider variety of States. Non-regression clauses covering international trade seem to be evolving incrementally into a standardised aspect of modern PTAs.

As one of the demandeurs for non-regression clauses in PTAs, it is noteworthy that the US's practice has evolved since its PTA with Jordan in a similar way to its investment-related non-regression clauses, namely through a stringent mandatory obligation and the application of State–State dispute settlement procedures linked to economic countermeasures for violations.Footnote 116 During this evolution, the US's motivation has persisted as being ‘to prevent environmental abuse as a means to gain an advantage in international trade’.Footnote 117 The EU also looks set to strengthen its approach to the drafting and implementation of non-regression clauses covering trade, whilst continuing to cite normative goals and issue linkage as the main driver, ie ‘leverag[ing] sustainable development’ by using trade concessions as a means of achieving environmental objectives.Footnote 118 Although the inclusion of trade-related non-regression clauses has become standard in China's PTAs, China's approach opts for State–State consultations over adjudicatory procedures to resolve disputes.Footnote 119

Analysis will now turn to the implications of the emergence of non-regression clauses in the international trade regime, particularly in relation to State–State dispute settlement.

B. Implications of a Principle of Non-Regression in International Trade Law

As with the international investment regime, the implications of non-regression clauses emerging in the international trade regime manifest in two main ways. In some instances, they can give rise to a stand-alone cause of action in State–State dispute settlement under PTAs containing trade-related non-regression clauses (section III.B1). Secondly, they can form the interpretive context to other obligations being construed in a State–State dispute (section III.B2).

1. Non-regression clauses as the direct basis for a claim under PTAs

Some trade-related non-regression clauses in PTAs are exempt from State–State dispute settlement, particularly the earlier iterations.Footnote 120 More recently, non-regression clauses in the international trade regime are increasingly subject to State–State dispute settlement by independent adjudicators.Footnote 121 Nonetheless, there continues to be a divergence in the remedies available upon a finding that a State has violated the non-regression clause. As mentioned earlier, the US's practice has moved towards permitting economic countermeasures in response to findings of violation.Footnote 122 By contrast, under the practice of the EU, a panel of independent adjudicators produces findings of whether the respondent State has violated the non-regression clause at issue, but these findings have no binding or tangible force.Footnote 123 Rather, the findings are intended to contribute to a broader dialogue between the disputing States concerning the contested measure.Footnote 124 That said, recent developments indicate that the EU's practice is set to become more sanction-oriented.Footnote 125

As with ISDS in the international investment regime, non-regression clauses concerning international trade have yet to be tested in State–State dispute settlement under a PTA (although, recent reports suggest the EU is considering bringing a claim under the Brexit Agreement regarding the UK's weakening of sewage rules).Footnote 126 However, the absence of such disputes thus far should not be taken as an indication that they are unlikely to arise. Rather, recent years have seen a distinct movement towards States invoking dispute settlement procedures under PTAs in non-traditional areas such as labour and the environment. For instance, the US has on several recent occasions invoked environment-related dispute mechanisms under its PTA with Peru concerning illegal loggingFootnote 127 and under its PTA with Korea concerning illegal fishing.Footnote 128 In the analogous labour context, an adjudicatory panel convened under the Dominican Republic–Central America–United States Free Trade Agreement became, in 2017, the first to issue findings in a labour-related dispute under a PTA.Footnote 129 This was followed by an adjudicatory panel reaching findings in a labour-related dispute under the PTA between the EU and KoreaFootnote 130 and by the US twice triggering labour-related dispute mechanisms under USMCA.Footnote 131

Indeed, one of the key concessions to Congressional Democrats for their support for the USMCA concerned amendments to make it easier for complainant States to succeed in environment- and labour-related cases.Footnote 132 Similarly, several EU Member States have advocated strengthening the State–State dispute settlement procedures for labour and environment provisions in the EU's PTAs by subjecting those provisions to economic countermeasures.Footnote 133 In the context of the Brexit negotiations, including stringent dispute settlement procedures for non-regression clauses was one of the EU's highest priorities in concluding its Brexit agreement.Footnote 134

Accordingly, against a background of major jurisdictions expressing concerns about the potential use of environmental standards as an element of competitive advantage,Footnote 135 the proliferation of non-regression clauses subject to State–State dispute settlement procedures is a significant development. Moreover, one need not look far to identify the kinds of fact patterns that could well be viewed as within the purview of trade-related non-regression clauses. In numerous recent instances, States have explained changes to environmental standards in terms of a trade-off between the prior more stringent standards and alleviating cost pressures on sectors like manufacturing and agriculture to improve those sectors’ international competitiveness. High-profile examples include the US's repeal (under the Trump Administration) of the Clean Power Plan to save ‘$33 billion in avoided compliance costs’, Australia's repeal of its emissions trading scheme to ‘boost Australia's economic growth’ and ‘enhance Australia's international competitiveness’, Brazil's weakening of laws requiring reforestation to promote the ‘monetisation’ of rainforests, and recent moves by India and Indonesia to roll back rules on environmental impact assessments for economic development.Footnote 136

Thus, as States increasingly bind their domestic environmental laws in PTAs, they may find themselves either constrained in their ability to relax those laws or otherwise facing liability for doing so. Some examples suggest that even proponents of trade-related non-regression clauses could find themselves at odds with these clauses.

2. Non-regression clauses as interpretive context in disputes over other obligations in PTAs

Aside from offering a direct cause of action under some PTAs, non-regression clauses could serve as interpretive context in State–State disputes over other PTA rights and obligations. For example, in a dispute under the PTA between Ukraine and the EU over a Ukrainian export ban on timber, Ukraine relied on the PTA's non-regression clause as interpretive context to rebut the EU's argument that the ban was not ‘necessary’ under the PTA's environment-related exceptions.Footnote 137 The EU had argued that alternative measures were available to Ukraine that were less restrictive than an export ban.Footnote 138 For Ukraine, however, adopting such alternatives would necessarily entail a regression from the export ban to encourage trade, thus creating an inconsistency with the non-regression clause. While the arbitral panel accepted that the non-regression clause could play a role as interpretive context in this regard, it ultimately rejected the EU's case on other grounds.Footnote 139

As this example illustrates, non-regression clauses could be used by respondent States to argue against a finding of violation of another obligation in a PTA. Ukraine was effectively contending that the arbitral panel could not reach a finding of violation since any remedial action to comply with such a finding would necessarily violate the PTA's non-regression clause.

One could also envisage fact patterns in which the complainant State uses a non-regression clause under a PTA as interpretive context to support its case. If the measure at issue arose from a regression by the respondent State from its domestic environmental laws, the complainant State could argue in light of the PTA's non-regression clause that the measure falls outside the scope of what may otherwise constitute a reasonable or justified exercise of the right to regulate under the PTA. This could undermine the respondent State's ability to claim that the contested regression was necessary to protect another public interest matter, even another environmental value. If the PTA's non-regression clause prohibited the regression, it would be inconsonant to find that it was nonetheless somehow permitted under the host State's right to regulate under the PTA (eg under general exceptions or provision-specific flexibilities).

The ability for non-regression clauses in PTAs to function as interpretive context that delimits a State's right to regulate is distinguishable from the multilateral context in which there is no generally-applicable non-regression clause. For instance, WTO Members are typically understood to have a sovereign right to set their own levels of protection in public interest and regulatory matters without conditions on whether those levels reflect increases or decreases.Footnote 140 The significance of this to PTAs lies in increasing attempts in PTAs to harmonise outcomes in disputes and interpretive issues vis-à-vis the multilateral context.Footnote 141 It is unclear how these kinds of provisions in PTAs would accommodate outcomes in disputes under other agreements that, based on the respondent's chosen means of compliance, result in a regression from its domestic environmental standards to better facilitate trade, akin to Ukraine's argument mentioned above.Footnote 142 Moreover, some have also speculated that discrepancies between respective obligations could invite forum shopping by the complainant State.Footnote 143

Most PTAs do not address this potential tension, even in instances where the PTA explicitly enshrines a State's sovereign right to set its own levels of protection in a manner akin to the WTO Agreement.Footnote 144 A small number of PTAs do, however, include clarifications that the non-regression clause is ‘without prejudice to’ or ‘subject to’ a State's right to regulate.Footnote 145 Such clarifications are counterintuitive because they appear to subordinate the non-regression clause to the State's right to regulate. If a State maintained an overriding right to decrease its levels of environmental protection, a non-regression clause would be rendered inutile. It is doubtful that adjudicators under a PTA would interpret these clarifications in a way that deprives non-regression clauses of any practical meaning, unless no other interpretation is reasonably available.

C. Summary: Non-Regression Clauses under PTAs

Notwithstanding that non-regression clauses covering international trade are not yet as ubiquitous in PTAs as they are in IIAs, they appear to be following a similar trajectory. The emergence of trade-related non-regression clauses is significant in light of increasing concerns amongst major economies that environmental standards could be misused as a source of competitive advantage in international markets. Non-regression clauses could be a source of new claims in the international trade regime, as reflected in reports that the EU is considering such a claim against the UK and could, in any event, operate as interpretive context in trade disputes over other rights and obligations in PTAs. It is noteworthy, in that regard, that the concept at the heart of non-regression clauses—namely, to preclude downward movements in a State's domestic levels of environmental protection—could be difficult to reconcile with other provisions in PTAs that, similar to the WTO Agreement, enshrine a sovereign right to set their own levels of protection irrespective of whether they are higher or lower.

IV. Non-Regression from Domestic Environmental Protections in International Environmental Law

Some have postulated that an established principle of non-regression from domestic environmental protections exists in international environmental law that finds expression in the Paris Agreement and other multilateral and environmental treaties.Footnote 146 This review of the origins and rationale of such a principle suggests the contrary. As will be shown, this principle is largely an unrealised concept that appears in theoretical and advocacy literature but is not readily observable in the primary sources of international environmental law.

Recently, however, there have been signs that this principle could transform from a theoretical concept to a legal reality. It was mentioned in a 2018 treaty on environmental rights in Latin America. It was also identified for inclusion in a potential new multilateral environmental treaty discussed at the United Nations General Assembly (UNGA). Analysis begins with an overview of the origins and emerging rationales for this principle in the international environmental regime (section IV.A) before considering its potential implications (section IV.B).

A. Background and Rationale

A principle of non-regression in the international environment regime first emerged in the lead-up to the Rio + 20 Conference in 2012 and the negotiation of its outcome document ‘The Future We Want’.Footnote 147 The European Parliament ‘call[ed] for the recognition of the principle of non-regression in the context of environmental protection as well as fundamental rights’ in its resolution on its preferred EU position.Footnote 148 The concept was promoted actively by the International Centre for Comparative Environmental Law, which explained that ‘[a]s the right to a healthy environment is increasingly recognized, it should be protected—like all UN human rights—by preventing any type of backsliding or regression of existing levels of environmental protection’.Footnote 149 The concept was included in an early draft of the Rio + 20 outcome document as a preference of the G77/China negotiating group, but did not achieve consensus,Footnote 150 and was reportedly opposed by the US, Japan, and Canada.Footnote 151 Thus, the final document does not refer to a principle of non-regression. While proponents of establishing a principle framed their advocacy in terms of a potential human right to the environment, the primary impetus seems to have been more a pragmatic reaction to environmental backsliding in the face of major challenges like climate change than a normative reaction.Footnote 152

Subsequently, proposals for a principle of non-regression in international environmental law resurfaced in a process—ultimately taken up by UNGA—to establish a new multilateral environmental agreement that would codify universal principles of international environmental law. A 2018 study commissioned by the United Nations Secretary-General to inform this process described the principle of non-regression as ‘relatively new to the field of environmental law’ and ‘not yet fully developed’.Footnote 153 According to this study, ‘once a human right is recognized, it cannot be restrained, destroyed or repealed’ under international law.Footnote 154 This, in turn, provided the rationale for inferring that States should be precluded from regressing from their domestic levels of environmental protection in international law to the extent of overlap between human rights and environmental protection.

However, several States participating in this UNGA process remained unconvinced. Some delegations such as China and Colombia called for further clarity on the substance of a non-regression clause in international environmental law.Footnote 155 Others such as New Zealand and Canada expressed reservations that the clause could perversely lead to lower environmental standards in domestic laws to preserve flexibility for future changes.Footnote 156 The UNGA discussions on this new agreement appear to have since been abandoned.Footnote 157

A separate stream of thought has sought to posit a principle of non-regression as somehow pre-existing and perhaps even forming a part of customary international law.Footnote 158 The authors find this unpersuasive. Proponents of this perspective cite the Paris Agreement—particularly Article 4.3—as a prominent manifestation of an existing principle of non-regression.Footnote 159 The relevant aspect of Article 4.3 that forms the basis of this contention provides: ‘[e]ach Party's successive nationally determined contribution will represent a progression beyond the Party's then current nationally determined contribution’.Footnote 160 The logical corollary of continuous progressions is that regressions are precluded under Article 4.3.

However, nothing in the provision's negotiating history supports the view that it is derived from a principle of non-regression that exists more generally in the international environmental regime. Rather, that history reveals that the provision is better understood as an innovation to address the so-called ‘ambition gap’ in the Paris Agreement. The first appearance of the concept that was ultimately reflected in Article 4.3 is found in the 2014 Lima Call for Climate Action and its attendant draft negotiating text for what became the Paris Agreement. The negotiating parties had, at that stage, ‘agree[d] that each Party's intended nationally determined contribution … will represent a progression beyond the current undertaking of that Party’,Footnote 161 and the negotiating text included various references to ‘tak[ing] action at the highest level of ambition and to progressively increase that level of ambition’ and to ‘progressively enhance the level of ambition of mitigation commitments’.Footnote 162 As these textual aspects suggest, the underlying concern was that the aggregate ambition of the negotiating parties was insufficient to meet the objective of avoiding dangerous climate change.Footnote 163 The discussions were thus focused on how to enhance this level of ambition over time. The negotiating parties explored mechanisms for ‘scaling up’ ambition.

Mexico, for instance, stressed for need for ‘a mechanism that allows for the evolution of its provisions and adapts to changing conditions … as a way to gradually increase the collective and individual level of ambition oriented to reach the above-mentioned objective of the Convention’,Footnote 164 and South Africa submitted that the ‘2015 agreement would include an adjustment procedure to ensure that the long-term aspirational goals are met’.Footnote 165 The EU called for ‘a mechanism within the 2015 Agreement to review periodically, and if necessary scale up, the level of ambition of mitigation commitments to stay on track to achieve the below 2°C objective’,Footnote 166 and the Rainforest Coalition of 18 developing countries submitted that a ‘review process should be considered to allow increasing ambition over time’.Footnote 167 New Zealand likewise argued for ‘an obligation on each Party to increase the ambition of its commitment over time’.Footnote 168 Other groups of negotiating countries made similar proposals for a mechanism to increase the ambition of mitigation commitments as a means of bridging the gap in ambition that would be necessary to keep the increase in global temperature to below 2°C or 1.5°C.Footnote 169

Thus, the inclusion of Article 4.3 was not indicative of a pre-existing principle of non-regression in international environmental law. Rather, it is better understood as part of a suite of such tools in the Paris Agreement referred to as the ‘ambition cycle’,Footnote 170 and was understood by negotiating States as a pragmatic mechanism to ‘scale up’ ambition.Footnote 171 There is no evidence that the non-regressive elements of this provision have a normative character or are connected to a pre-existing principle. The same is true for the other MEAs sometimes cited as incorporating a non-regression clause.Footnote 172

It is also noteworthy that this principle is largely absent from domestic legal systems, despite recent movements in jurisdictions like Australia and Brazil that have drawn on Article 4.3 to ‘lock in’ existing climate standards.Footnote 173 In particular, although Boyd's landmark study on the coverage of environmental matters in 193 national constitutions is sometimes cited in support of the existence of a principle of non-regression domestically,Footnote 174 his study indicates only a handful of domestic legal systems exhibit a principle of non-regression in that way: France, Belgium, Hungary, and several Latin American countries.Footnote 175 Further, Boyd recognises that dozens of national constitutions make no mention of environmental rights, including such jurisdictions as the US, China, Japan, and Canada.Footnote 176 A separate study into the existence of non-regression in US environmental law found that it does not exist as a general principle in that system. Relatedly, measures to avoid regression in US environmental law did not have justifications rooted in a commitment to protecting individual rights.Footnote 177

Thus, a principle of non-regression cannot be said to be pre-existing or already-established in the international environment regime, nor a well-established and ubiquitous feature of domestic legal systems. Rather it is a nascent concept whose primary normative basis would be derived from, if anywhere, a human rights framework. Recent developments in the United Nations Human Rights Council (HRC) have strengthened the viability of this normative basis. In late 2021 the HRC explicitly recognised, for the first time, the existence of ‘the right to a safe, clean, healthy and sustainable environment as a human right that is important for the enjoyment of human rights’.Footnote 178 Through this recognition, the HRC went beyond acknowledging that environmental degradation can interfere with the enjoyment of other human rights by declaring that environmental protection can itself more directly manifest as a human right.

As part of the preparatory work leading up to this recognition by the HRC, the Special Rapporteur to the HRC produced numerous studies (subsequently cited by the HRC)Footnote 179 on the subject, including one finding that ‘there is a strong presumption against retrogressive measures’ from environmental protections falling within the purview of human rights.Footnote 180 In other words, a principle of non-regression applies to a putative human right to environmental protection. Within the human rights framework, this principle finds its origins in Article 30 of the Universal Declaration of Human Rights, which has since been replicated in several major human rights treaties,Footnote 181 and provides that ‘[n]othing in this Declaration may be interpreted as implying for any State, group or person any right to engage in any activity or to perform any act aimed at the destruction of any of the rights and freedoms set forth herein’.Footnote 182 This principle also derives from the concept of the ‘progressive realisation’ of human rights set out in certain human rights treaties, which presupposes that the realisation of human rights will continuously improve and, accordingly, will not regress.Footnote 183

The link between environmental protection, human rights, and non-regression is gaining broader currency. A recent UN report on domestic environmental laws called for a ‘human rights approach’ to ‘strengthen [the] environmental rule of law through application of the non-regression principle’.Footnote 184 Relatedly, as mentioned earlier, a 2018 treaty in Latin America on procedural rights and access to justice in environmental matters incorporated the first explicit reference to a principle of non-regression in international environmental law, providing that ‘Each Party shall be guided by the following principles in implementing the present Agreement: … (c) Principle of non-regression’.Footnote 185

The extent to which environmental protection and human rights are linked—and, accordingly, the breadth of a non-regression principle in this domain—remains somewhat unclear. There is not a complete alignment between human rights protections and environmental protections.Footnote 186 The Human Rights Council's Special Rapporteur on this subject reported that ‘[t]he obligation to protect human rights from environmental harm does not require States to prohibit all activities that may cause any environmental degradation’.Footnote 187 If there is no nexus between a particular area of environmental protection and the enjoyment of a human right, there may not be a basis for subsuming that area of environmental protection within the rubric of human rights protections.Footnote 188 Not every regression from domestic environmental protections will necessarily be associated with a derogation from human rights.

Nonetheless, the concept of a principle of non-regression in the international environment regime seems to be approaching an inflection point, where the theoretical proposition is transforming into a legal reality underpinned by a human rights framework.

B. Implications of a Principle of Non-Regression in International Environmental Law

Advocates of a multilaterally-adopted non-regression clause in the international environment regime have postulated that it could influence litigious proceedings over environmental matters on both the domestic and international planes.Footnote 189 However, unlike in the international trade and investment regimes, non-regression clauses have yet to emerge in international environmental law in a way that gives rise to tangible implications. As mentioned earlier, a non-regression clause has only appeared in an environmental treaty as an expression of principle rather than an operative provision subject to dispute settlement. Nonetheless a brief overview of two aspects that would be particularly relevant to any future operationalisation of a principle of non-regression in the international environment regime will be provided.

One shortcoming of non-regression clauses in the international trade and investment regimes relates to the absence of any flexibilities or exceptions to derogate from environmental laws to fulfil another public interest matter (see above sections II.B2 and III.B2). By contrast, the normative basis of a non-regression principle in the international environment regime has a human rights framework at its source, which introduces the possibility for certain limited derogations.Footnote 190 This possibility was recognised by the Special Rapporteur to the HRC on human rights and the environment:Footnote 191

… as with all other rights in the Covenant, there is a strong presumption that retrogressive measures taken in relation to the right to health are not permissible. If States do take deliberately retrogressive measures, then they have the burden of proving that they first carefully considered all alternatives, and that the measures are duly justified by reference to the totality of the rights provided for in the Covenant in the context of the full use of the State party's maximum available resources.

Therefore, it seems that a non-regression principle derived from a human rights framework would not be absolute. Rather, whilst emphasising that ‘there is a strong presumption against retrogressive measures’, the Special Rapporteur considered that ‘States have discretion to strike a balance between environmental protection and other legitimate societal interests … [b]ut the balance cannot be unreasonable, or result in unjustified, foreseeable infringements of human rights’.Footnote 192 In practical terms, instances where States regress from domestic environmental laws to ameliorate another urgent public interest, such as the renewable energy schemes discussed in section II.B above, may not infringe the principle of non-regression in the international environment regime. However, instances where States regress from their environmental laws despite those laws operating as envisaged, and despite the absence of any external event or change in circumstances that warrant a reconsideration, could be impugned by this principle.

For example, the Trump Administration's repeal of the Clean Power Plan to save ‘$33 billion in avoided compliance costs’,Footnote 193 Australia's repeal of its emissions trading scheme to ‘boost Australia's economic growth’ and ‘enhance Australia's international competitiveness’,Footnote 194 and Ontario's repeal of its climate law to ‘remove a costly burden from Ontario businesses, allowing them to grow, create jobs and compete around the world’Footnote 195 could, on one view, be indicative of the kinds of regressions that would fall foul of an international environmental law principle of non-regression. Future iterations of non-regression clauses in the international environment regime would do well to clarify this distinction explicitly and provide guidance delimiting the circumstances under which regressive measures could be permitted.

A second consideration pertinent to the practical application of non-regression clauses in the international environment regime concerns how, in practice, to identify a benchmark for determining whether a regression has actually occurred. This is because a single environmental protection objective can be articulated in various ways that offer varying benchmarks. For instance, Australia adopted a renewable energy law whose 2020 target was simultaneously defined as both 20 per cent of electricity from renewables by 2020, as well as 41,000GWh generated from renewables by 2020.Footnote 196 The legislation was subsequently weakened to a level of 33,000GWh by 2020.Footnote 197 Despite this tacit weakening of the law, the 20 per cent target remained unchanged since the legislated level of 41,000GWh had represented an overestimated future electricity demand.Footnote 198 A benchmark based on the legislated 41,000GWh target would result in an impermissible regression, whereas a benchmark based on the 20 per cent target would not.

Indeed, the kinds of modelling errors underlying this divergence are not uncommon in environmental policymaking; the complexity of variables can make it difficult to accurately forecast the chosen mechanism's costs, effectiveness, and efficiency.Footnote 199 Drawing on another example discussed earlier, the Czech Republic designed a feed-in tariff scheme in 2005 on the assumption that 15 GWh production of electricity would be produced from solar power by 2010. However, actual production of electricity from solar power was 616 GWh in 2010, growing to 2,182 GWh in 2011, due to a series of unexpected developments (most significantly, the collapse in solar panel prices).Footnote 200 The feed-in tariffs uptake was correspondingly far higher than forecast, with the consequence that consumer electricity prices increased dramatically. To account for the measure's unexpected operation, the Czech Government amended its renewable energy legislation to reduce the price support for solar-generated electricity.

Again, whether this amendment qualifies as a regression depends on the benchmark used. The expected level of electricity generated from solar by 2010—which provided the basis for designing the legislation—represents one benchmark (ie 15 GWh); the degree of price support and tax breaks to stimulate investment in solar—which were actually enshrined in legislation as the instruments for improving environmental outcomes—represent another. The existing iterations of non-regression clauses in the international trade and investment regimes fail to accommodate instances where errors in assumptions and modelling underpinning legislated targets necessitate modifications to that legislation to realign it with the original policy intent. Instead, the legislation (or other instrument) is itself the benchmark, and a weakening of that benchmark constitutes a derogation irrespective of the rationale.

Thus, as a principle of non-regression begins to take shape in the international environment context, there would be strong grounds for addressing these considerations explicitly when translating the principle into operative non-regression clauses.

C. Summary: A Principle of Non-Regression in International Environmental Law

Contrary to some views, the authors do not consider that a principle of non-regression is reflected in major environmental treaties like the Paris Agreement. Clauses in those treaties, which preclude backsliding from environmental protections, are better understood as pragmatic responses to particular policy problems rather than reflecting a general, principled prohibition on regressing from domestic environmental protections.

Nonetheless, this principle debuted recently in a Latin American treaty on environmental rights. It has also found its way into various debates and discourses in international environmental law. As the concept of environmental protection as a human right that cannot be diminished or repealed gains wider recognition, it seems likely that the principle of non-regression in the international environmental regime will continue to emerge and evolve.

V. Conclusion

This review of non-regression from domestic environmental protections—as a principle emerging across the respective international law domains of trade, investment, and the environment—presents a confusing picture. One might expect such a principle to be most firmly grounded in the international environmental law domain given its environmental focus, and yet this is where it is least developed. Counterintuitively, it is the international trade and investment regimes—not always known for environmental sensibilities—in which this principle has seen the widest uptake and progression.

This dynamic could be due to the nature of the obligations typically incurred in these respective domains. Obligations to protect and attract investment, and to liberalise market access for goods, services and capital, could—in theory—risk fuelling a race to the bottom in terms of compliance costs from environmental regulation. Non-regression clauses could thus be viewed as a kind of safeguard or risk-management tool to mitigate against the negative externalities that arise when States liberalise their economic relations. While the focus of non-regression clauses may be on domestic environmental protections, their function in assisting to manage more open economic relations between States is inherently international.

These considerations are less pertinent to the international environmental regime due to the very nature of obligations in environmental treaties. In particular, it is not obvious why States would need the kind of safeguard offered by non-regression clauses in an environmental treaty when that treaty itself enshrines obligations to protect environmental matters. Rather, a regression from the environmental protections enshrined in an environment treaty would be a breach of the treaty itself. The practical function of non-regression clauses in that context would be to extend the treaty's reach to a State's domestic environmental protections, which are not otherwise prescribed by the treaty itself. Such a function seems to lack the kind of international character that one might expect of a principle in international law.

That said, non-regression clauses could play a niche role in relatively non-prescriptive environmental treaties that leave it up to State parties to determine the form and stringency of their commitments. Non-regression clauses in such instances would place parameters around State parties’ discretion to set their own commitments, thereby facilitating mutual confidence in the durability and trajectory of each other's commitments. As a tool to assist the management of States’ relations in these types of environmental treaties, non-regression clauses would have a more obvious international character.

The challenges in linking regressions from domestic environmental protections with international relations in the particular context of the international environmental regime may explain why its proponents have reached for a human rights framework. A human rights framework elevates the domestic to the international, providing a legal rationale for treating domestic environmental protections as a matter to be regulated in international relations.

However, these challenges might also explain why some proponents of non-regression clauses in the international trade and investment regimes have been less enthusiastic in the international environment framework. The need to assuage domestic interest groups concerned about potential negative externalities from economic liberalisation could explain the incentive for countries like the US to enshrine non-regression clauses in its IIAs and PTAs. By contrast, the very nature of environmental treaties as prescribing certain environmental protections eliminates this kind of incentive structure. Rather, such an incentive structure may only likely arise where domestic interest groups harbour concerns about the potential for backsliding in ways not explicitly provided for in the environmental treaty itself.

Finally, non-regression clauses present a puzzle regarding how principles form in international law. Thus far, they represent a case of convergent evolution. The same basic concept has emerged in different domains for different reasons at varying speeds. Concerns regarding industrial relocation from weakened environmental rules, the erosion of environmental PPMs, and a human right to environmental protection, have respectively elicited the same response: a principle of non-regression from domestic environmental protections. The future trajectory of non-regression clauses will perhaps showcase a unique way that principles can form in international law and provide a ripe area for further research, particularly if they start to merge across domains, as seems to now be the case with trade and investment in some PTAs.

Footnotes

Nothing in this article is reflective of the views or position of any past or present employer. The authors would like to thank Theodore Samlidis for his research assistance.

References

1 As potential examples in relation to Brazil, Indonesia, India, Canada, the United States and Australia, see eg, N Popvich et al, ‘76 Environmental Rules on the Way Out Under Trump’ (New York Times, 6 July 2018); US EPA, ‘EPA Takes Another Step To Advance President Trump's America First Strategy, Proposes Repeal Of ‘‘Clean Power Plan’’’ (News Release, 10 October 2017); Soares-Filho, B, Rajão, R, Macedo, MN and Carneiro, A, ‘Cracking Brazil's Forest Code’ (2014) 344 Science 363, 363–4CrossRefGoogle ScholarPubMed; Vale, M et al. , ‘The COVID-19 pandemic as an opportunity to weaken environmental protection in Brazil’ (2021) 255 Biological Conservation 23CrossRefGoogle ScholarPubMed; A Schipani and B Harris, ‘Brazil minister calls for the Amazon to be monetised’ (Financial Times, 23 August 2019); Fitch Ratings, ‘Indonesia's Reform Package Boosts Growth Prospects’ (FitchWire, 14 October 2020); Sembiring, R, ‘Indonesia's Omnibus Bill on Job Creation: a Setback for Environmental Law?’ (2020) 4 Chinese Journal of Environmental Law 97, 100–1CrossRefGoogle Scholar; Office of the Premier, ‘A Government for the People: Speech from the Throne’ (12 July 2018); Second Reading Speech, Clean Energy Legislation (Carbon Tax Repeal) Bill 2014 (Cth) (Australia); J Mazoomdaar, ‘Explained: Reading the draft Environment Impact Assessment norms, and finding the red flags’ (Indian Express, 10 August 2020); J Nandi, ‘‘‘Nothing disturbing in the clauses of draft EIA 2020’’, says RP Gupta’ (Hindustan Times, 17 August 2020).

2 North American Free Trade Agreement (signed 17 December 1992, entered into force 1 January 1994) (NAFTA).

3 Marrakesh Agreement Establishing the World Trade Organization (opened for signature 15 April 1994, entered into force 1 January 1995) 1867 UNTS 3 (WTO Agreement). As such, nothing in this article expresses a view on obligations under the WTO Agreement or the meaning of associated case law, nor on any interaction between the WTO Agreement and PTAs.

4 The Regional Trade Agreement Database of the WTO records 304 PTAs as being notified and in force (as of 15 February 2020): <https://rtais.wto.org/UI/PublicMaintainRTAHome.aspx>.

5 See, eg, Sands, P and Peel, J with Fabra, A and Mackenzie, R, Principles of International Environmental Law (4th edn, CUP 2018) Ch 6CrossRefGoogle Scholar; Dupuy, PM and Viñuales, JE, International Environmental Law (2nd edn, CUP 2018) Ch 3CrossRefGoogle Scholar.

6 Regional Agreement on Access to Information, Public Participation and Justice in Environmental Matters in Latin America and the Caribbean (opened for signature 9 April 2018, entered into force 22 April 2021) 56654 UNTS, art 3(c).

7 Appendix 1.

8 SR Fletcher, Trade and Environment: Treatment in Recent Agreements – GATT and NAFTA (Report for Congress, Congressional Research Service, 4 April 1994) 1, 9–10; Feeley, MS and Knier, E, ‘Environmental Considerations of the Emerging United States-Mexico Free Trade Agreement’ (1992) 2 Duke Journal of Comparative & International Law 259Google Scholar; US Congress (Office of Technology Assessment), Trade and Environment: Conflicts and Opportunities (Report No. OTA-BP-ITE-94, US Government Printing Office, May 1992) 17–18; A Chapman, North American Free Trade Agreement: rationale and issues (Report No. BP-327E, Canadian Parliament Research Branch, January 1993) section ‘Environment’.

9 DC Esty, Greening the GATT Trade, Environment and the Future (Institute for International Economics, 1994) 155–6; R Carbaugh and D Wassink, ‘Environmental Standards and International Competitiveness’ (1992) 16 World Competition 81, 82–3; S Charnovitz, ‘The North American Free Trade Agreement: green law or green spin?’ (1994) 26 Law&PolIntBus 1, 38; G Grossman and A Krueger, Environmental Impact of a North American Free Trade Agreement (Working Paper No 3914, National Bureau of Economic Research, 1991) 2; DC Esty and D Geradin, ‘Market Access, Competitiveness, and Harmonization: Environmental Protection in Regional Trade Agreements’ (1997) 21 HarvEnvtlLRev 265, 320 and 333.

10 Grossman and Krueger (n 9) 1–2, 4; Esty and Geradin (n 9) 313–14.

11 Agreement between the United States, the United Mexican States, and Canada (signed 30 November 2018, entered into force 1 July 2020) (USMCA).

12 USTR, The United States Canada Mexico Agreement Fact Sheet: Environment (December 2019) 1.

13 Debuting in: CARIFORUM-European Community Economic Partnership Agreement, signed 15 October 2008 (entered into force 29 December 2008) art 188.1 (CARIFORUM–EU FTA).

14 See, eg, Council of the European Union, Review of the EU Sustainable Development Strategy (No. 10117/06, 9 June 2006) 21; European Commission, Trade for All. Towards a More Responsible Trade and Investment Policy (European Union 2015) 23.

15 European Commission, Task Force for Relations with the United Kingdom, Internal EU27 preparatory discussions on the future relationship: ‘Level playing field’ (UKTF (2020) 4 – Commission to EU 27) 4, 5, 13.

16 See eg, CPTPP, art 20.3.6; Free Trade Agreement between the Government of the People's Republic of China and the Government of Georgia (signed 13 May 2017, entered into force 1 January 2018) art 9.2 (China–Georgia FTA); Canada–Ukraine Free Trade Agreement (signed 11 July 2016, entered into force 1 August 2017) art 12.5 (Canada–Ukraine FTA); Korea–New Zealand Free Trade Agreement (signed 23 March 2015, entered into force 15 December 2015) art 16.2.2 (Korea–New Zealand FTA); Agreement between Japan and Georgia for the Liberalisation, Promotion, and Protection of Investment (signed 29 January 2021, not yet in force) art 20 (Japan–Georgia IIA); Investment Cooperation and Facilitation Treaty between Brazil and India (signed 25 January 2020, not yet in force) art 22.2 (Brazil–India IIA); EU–Japan Economic Partnership Agreement (signed 17 July 2018, entered into force 1 February 2019) art 16.2.2 (EU–Japan FTA); Framework Agreement between Korea and Turkey (signed 2 August 2012, entered into force 1 May 2013) art 5.7.2 (Korea–Turkey FTA).

17 See, eg, Free Trade Agreement between the Eurasian Economic Union and Viet Nam (signed 29 May 2015, entered into force 5 October 2016) art 12.4.3 (EEU–Vietnam FTA).

18 See, eg, Comprehensive Economic Partnership Agreement between Indonesia and the EFTA States (signed 18 December 2018, entered into force 1 November 2021) arts 4.8.2 and 8.3.4 (Indonesia–EFTA FTA); Free Trade Agreement between the EFTA States and Albania (signed 17 December 2009, entered into force 1 October 2011) art 34.2 (Albania–EFTA FTA).

19 Economic Partnership Agreement between the European Union and the SADC EPA States (signed 10 June 2016, not yet in force, provisionally applied in part) art 9.3 (EU–SADC FTA); Southern African Development Community, SADC Model Bilateral Investment Treaty Template with Commentary (2012) 41.

20 Morocco Model Investment Treaty (2019) art 17.

21 Colombia Model Investment Treaty (2017) art XI.

22 Slovakia Model Investment Treaty (2019) art 3.2.

23 Investment Promotion and Protection Agreement between Nigeria and Singapore (signed 4 November 2016, not yet in force) art 10 (Nigeria–Singapore IIA).

24 Agreement between the United Arab Emirates and Rwanda on the Reciprocal Promotion and Protection of Investments (signed 1 November 2017, not yet in force) art 9.2.

25 Free Trade Agreement between Argentina and Chile (signed 2 November 2017, entered into force 2 May 2019) arts 8.14 and 13.4.14 (Argentina–Chile FTA).

26 Agreement between Guatemala and Trinidad and Tobago on the Reciprocal Promotion and Protection of Investments (signed 13 August 2013, entered into force 23 June 2016) art 16.

27 See generally AD Mitchell and J Munro, ‘No Retreat: An Emerging Principle of Non-Regression from Environmental Protections in International Investment Law’ (2019) 50 Georgetown Journal of International Law 625, 655–8.

28 Appendix 1.

29 NAFTA, art 1114(2).

30 See Mitchell and Munro (n 27) 649–50 and 662–7.

31 Kinnear, M et al. , ‘Article 1114 – Environmental Measures’ in Investment Disputes under NAFTA: An Annotated Guide to NAFTA Chapter 11, Supplement No. 1 (Kluwer Law International 2006) 1114–13Google Scholar.

32 USMCA, art 24.32, 31.19.1.

33 Mitchell and Munro (n 27) 663.

34 The more recent IIAs of the US and EU contain an explicit definition of ‘environmental law’ that captures both legislation and regulations (eg CETA art 24.1; CPTPP, art 20.1). Others use the broad and undefined terminology of ‘environmental measures’ (eg IIAs between China and Canada, Mexico and Switzerland, Japan and Kenya), and yet others still list the types of instruments covered without attendant definition such as ‘environmental laws, regulations, policies and practices’ (eg IIAs of China).

35 This is reflected in the recent drafting practice across major jurisdictions including the US, China, the EU, Japan, Korea, Brazil, and Canada. See, eg, USMCA, art 24.4.3; CPTPP, art 20.3.6; China–Georgia FTA, art 9.2; Canada–Ukraine FTA, art 12.5; Korea–New Zealand FTA, art 16.2.2; Japan–Georgia IIA, art 20; EU–Japan FTA, art 16.2.2; Korea–Turkey FTA, art 5.7.2; Protocol to Upgrade the New Zealand–China Free Trade Agreement (signed 4 November 2019, not yet in force) Ch 22 art 3.2 (China–New Zealand Protocol). Brazil and India used the equivalent terminology of ‘amend’ and ‘repeal’ in their recent IIA: Brazil–India IIA, art 22.2.

36 See ibid.

37 See, eg, Switzerland–Mexico IIA, ad art 3; Japan–Iraq IIA, art 22; Slovakia–Iran IIA, art 10.1; Japan–India FTA, art 99; Japan–Korea–China IIA, art 23.

38 See, eg, Vietnam–EEA FTA, art 12.4.3; EFTA–Central American States FTA, art 12.4.3; EU–Korea FTA, art 13.7.2.

39 See, eg, NAFTA, art 1114(2); Switzerland–Mexico IIA, art 3; Japan–Iraq IIA, art 22; Slovakia–Iran IIA, art 10.1; Japan–India FTA, art 99; Japan–Korea–China IIA, art 23.

40 Oxford English Dictionary Online (Oxford University Press), ‘encourage’, v.

41 Philip Morris v Uruguay, ICSID Case No. ARB/10/7, Jurisdiction (2 July 2013) [168].

42 See, eg, Appellate Body Report, Australia – Apples, [172]: ‘The word “to” in adverbial relation with the infinitive verb “protect” indicates a purpose or intention. Thus, it establishes a required link between the measure and the protected interest.’

43 NAFTA, art 1101(1)(c) (emphasis added).

44 See section II.B1.

45 Mitchell and Munro (n 27) 663–5.

46 See, eg, Trade and Cooperation Agreement between the European Union and the United Kingdom (signed 30 December 2020, entered into force 21 April 2021) art 391.2 (‘Brexit Agreement’); Korea–New Zealand FTA, art 16.2.6; Colombia–United States Trade Promotion Agreement (signed 22 November 2006, entered into force 15 May 2012) art 18.3.3 (Colombia–US FTA). See, by analogy, Panel Report, Guatemala – Article 16.2.1(a) of the CAFTA-DR (4 June 2017) [165], [175], [177], [190].

47 See, eg, OECD, Database on Policy Instruments for the Environment: Tradable Permits Systems – Annual Information All Countries <https://pinedatabase.oecd.org/QueryResult_8.aspx?Key=4765c3a1-f053-4c04-bec9-e3d3df693f34&QryCtx=2)>.

48 L Bohmer, ‘Koch conglomerate launches NAFTA legacy claim against Canada over cancellation of emissions trading program’ (18 December 2020) Investment Arbitration Reporter.

49 Campos de Pese v Panama, ICSID Case No. ARB/20/19.

50 Lee-Chin v Dominican Republic, ICSID Case No. UNCT/18/3; Zelena v Serbia, ICSID Case No. ARB/14/27.

51 See YS Selivanova, ‘Changes in Renewables Support Policy and Investment Protection under the Energy Charter Treaty: Analysis of Jurisprudence and Outlook for the Current Arbitration Cases’ (ICSID Review – Foreign Investment Law Journal, 30 August 2018).

52 Latam Hydro and Mamacocha v Peru, ICSID Case No. ARB/19/28, Claimant's Memorial (14 September 2020) [29], [35]; LSG Building Solutions GmbH and others v Romania, ICSID Case No. ARB/18/19; Orazul International v Argentina, ICSID Case No. ARB/19/25; V Djanic, ‘Wind farm investor lodges treaty-based claim against Ukraine’ 29 October 2021 Investment Arbitration Reporter.

53 See, eg, Eskosol v Italy, ICSID Case No. ARB/15/50, Award (4 September 2020); Greentech v Italy, SCC Arbitration V (2015/095), Award (23 December 2018); 9Ren v Spain, ICSID Case No. ARB/15/15, Award (31 May 2019); Antin v Spain, ICSID Case No. ARB/13/31, Award (15 June 2018); RWE Innogy v Spain, ICSID Case No. ARB/14/34 (30 December 2019); JSW Solar v Czech Republic, Ad Hoc Arbitration, Permanent Court of Arbitration Case No. 2014-03, Award (11 October 2017); Charanne v Spain, Stockholm Chamber of Commerce Arbitration No. 2012/062, Award (21 January 2016); Blusun v Italy, ICSID Case No. ARB 14/3, Award (27 December 2016); Stadtwerke v Spain, ICSID Case No. ARB/15/1, Award (Majority of Jeswald Salacuse and Zachary Douglas) (2 December 2019); ESPF Beteiligungs v Italy, ICSID Case No. ARB/16/5, Award (Majority of Henri C. Alvarez and Michael C. Pryles) (14 September 2020); Novenergia v Spain, SCC Arbitration 2015/063, Award (15 February 2018); SunReserve Luxco Holdings v Italy, SCC Arbitration V 2016/32 (25 March 2020); Belenergia v Italy, ICSID Case No. ARB/15/40 (6 August 2019); InfraRed v Spain, ICSID Case No. ARB/14/12, Award (2 August 2019); OperaFund v Spain, ICSID Case No. ARB/15/36, Award (Majority of August Reinsch and Karl-Heinz Böckstiegel) (6 September 2019); RREEF v Spain, ICSID Case No. ARB/Í3/30, Award (Majority of Alain Pellet and Pedro Nikken) (30 November 2018); Hydro Energy v Spain, ICSID Case No. ARB/15/42, Award (9 March 2020); Antaris v Czech Republic, Ad Hoc Tribunal, Permanent Court of Arbitration Case No. 2014-01, Award (2 May 2018).

54 See, eg, Nigeria–Singapore IIA, arts 10 and 12; Agreement between Slovakia and Iran for the Promotion and Reciprocal Protection of Investments (signed 19 January 2016, entered into force 30 August 2017) arts 10.1, 17.1 (Slovakia–Iran IIA); Agreement between the Belgium–Luxembourg Economic Union and Montenegro on the Reciprocal Promotion and Protection of Investments (signed 16 February 2010, not yet in force) arts 5.2, 12, and 13; Agreement among Japan, Korea and China for the Promotion, Facilitation and Protection of Investment, (signed 13 May 2012, entered into force 17 May 2014) arts 23 and 15.2 (Japan–Korea–China IIA); Agreement between Colombia and Turkey concerning the Reciprocal Promotion and Protection of Investments (signed 28 July 2014, not yet in force) arts 11.2, 12.2 and 14.1; Agreement between Japan and Kenya for the Promotion and Protection of Investment (signed 28 August 2016, entered into force 14 September 2017) arts 15 and 22; Comprehensive Economic Partnership Agreement between India and Korea (signed 7 August 2009, entered into force 1 January 2010) arts 10.16.2, 10.21.1 and 14.2.1.

55 There is no basis to interpret clauses containing the term ‘should’ instead of ‘shall’ as being excluded from dispute settlement, particularly where there is no indication or negotiating history to the contrary and they have been explicitly subject to dispute settlement. The term ‘should’ is capable of connoting not only an aspiration, but also a binding duty or responsibility (albeit less prescriptive than ‘shall’): see, eg, El Paso v Argentina, ICSID Case No. ARB/03/I5, Jurisdiction (27 April 2006) [106], [110]; Grand River v USA, UNCITRAL Arbitration, Jurisdiction (20 July 2006) [58]; Occidental v Ecuador, LCIA Case No UN3467, Award (1 July 2004) [70]; Enron v Argentina, ICSID Case No ARB/01/3, Jurisdiction (14 January 2004) [65]; Appellate Body Report, Canada – Aircraft, [187]; Panel Reports, US – Clove Cigarettes, [7.575]; Korea – Radionuclides (Japan), [7.429]; Guatemala – Cement II, n 854.

56 See, eg, Agreement between Canada and Hong Kong for the Promotion and Protection of Investments (signed 10 February 2016, entered into force 6 September 2016) art 20.1; China–Georgia FTA art 9.6; Brazil–Ethiopia IIA art 24.3; Free Trade Agreement between Colombia and Korea (signed 21 February 2013, entered into force 15 July 2016) art 20.2 (Colombia–Korea FTA).

57 See, eg, USMCA art 24.32; CPTPP art 20.23; Canada–Ukraine FTA, art 12.21; Colombia–US FTA, art 18.12.

58 See ‘derogate’, Black's Law Dictionary (10th edn, BA Gardner ed, 2014): ‘to detract’ or ‘[t]he partial repeal or abrogation of a law by a later act that limits its scope or impairs its utility and force’. See also ‘derogate’, Oxford English Dictionary (2nd edn, OUP 1989): ‘[t]o repeal or abrogate in part (a law, sentence, etc.); to destroy or impair the force and effect of; to lessen the extent or authority of’ or ‘[to] detract from; to lessen, abate, disparage, depreciate’.

59 See Philip Morris v Uruguay, ICSID Case No. ARB/10/7, Jurisdiction (2 July 2013) [168].

60 See above (n 35).

61 This is the formulation in the original NAFTA (see NAFTA art 1114(2)) and has since been replicated widely, eg: Japan–Korea–China IIA, art 23; Agreement between Switzerland and Mexico on the Promotion and Reciprocal Protection of Investments (signed 10 July 1995, entered into force 14 March 1996), Ad art 3; Agreement between Japan and Iraq for the Promotion and Protection of Investment (signed 7 June 2012, entered into force 25 February 2014) art 22; Slovakia–Iran IIA, art 10.1; Comprehensive Economic Partnership Agreement between Japan and India (signed 16 February 2011, entered into force 30 June 2011) art 99.

62 Ontario Government, ‘Relief on the Way: Ontario Passes Legislation to End Cap and Trade Carbon Tax’ (Press Release, 31 October 2018).

63 See, eg, Stadtwerke (n 53) [258]–[259]; 9Ren (n 53) fn 4. See also IMF, Spain: 2012 Article IV Consultation (IMF Country Report No. 12/202) 5; European Commission, Spain: Memorandum of Understanding on Financial-Sector Policy Conditionality (20 July 2012) [31].

64 Antaris (n 53) [120]–[127] and [444]; JSW Solar (n 53) [383]–[391]; Eskosol (n 53) [388]–[389], [400], [402] and [410]; SunReserve Luxco Holdings (n 53) [854]–[855].

65 See above (n 35).

66 See, eg, RREEF (n 53) [545], [547]; RWE (n 53) [599]–[600], [732]. These claims related to changes in the regulatory framework, as opposed to specific promises.

67 See further below section II.B2.

68 ESPF Beteiligungs (n 53) [443]–[444].

69 Novenergia (n 53) [646], [648]; Blusun (n 53) [315(c)]; SunReserve (n 53) [684]; Belenergia (n 53) [570]–[571]; InfraRed (n 53) [350], [365]; OperaFund (n 53) [426]; RREEF (n 53) [260]; Hydro Energy (n 53) [548].

70 SunReserve (n 53) [685]–[687]; RREEF (n 53) [262]; Novenergia (n 53) [655]–[658], [694]; Hydro Energy (n 53) [583]; Antin (n 53) [530]–[531].

71 OperaFund (n 53) [481]; ESPF (n 53) [512]; Belenergia (n 53) [579]–[580]; InfraRed (n 53) [366]–[367], [418]; Hydro Energy (n 53) [673]; Charanne (n 53) [492]–[497]; RWE (n 53) [452]; 9Ren (n 53) [253]–[257]; Antin (n 53) [538], [552]; Antaris (n 53) [360].

72 ESPF (n 53) [512]; 9Ren (n 53) [653]–[657]; Antaris (n 53) [360].

73 Some tribunals have diverged as to whether the right to regulate could nonetheless justify reneging on such promises. Some appear to have considered that a right to regulate can override such promises: SunReserve (n 53) [703]; Belenergia (n 53) [572]. Others have found that reneging on promises – even if in pursuit of legitimate objectives – must be compensated: CEF Energia v Italy, SCC Arbitration V (2015/158), Award (16 January 2019), [240]–[243], ESPF (n 53) [418], [421], [520]; Blusun (n 53) [366]; OperaFund (n 53) [485]; 9Ren (n 53) [258]–[259]

74 Blusun (n 53) [319(4)–(5)], [367]–[371]; SunReserve (n 53) [702]; Belenergia (n 53) [579]–[582]; RWE (n 53) [448], [457]–[458]; RREEF (n 53) [318]–[321]; Hydro Energy (n 53) [584]–[586], [594]; Stadtwerke (n 53) [264]; Charanne (n 53) [493], [499]. That said, some tribunals have considered that laws themselves can create legitimate expectations: Novenergia (n 53) [650]–[652]; Antaris (n 53) [360]. For an overview of the two schools of thought on whether laws and regulations of general application can amount to a ‘legitimate expectation’ protected under the FET standard, see RWE (n 53) [453]–[458]; Masdar v Spain, ICSID Case No. ARB/14/1, Award (16 May 2018), [489]–[503] and [521].

75 SunReserve (n 53) [692]; ESPF (n 53) [577]; Blusun (n 53) [319(5)], [363], [372]; InfraRed (n 53) [368]; RWE (n 53) [551], [553]; OperaFund (n 53) [509]–[510]; Novenergia (n 53) [654]; Hydro Energy (n 53) [568], [590]; Antin (n 53) [532]; Charanne (n 53) [514], [517]; Antaris (n 53) [360]; Eiser v Spain, ICSID Case No. ARB/13/36, Award (4 May 2017) [363]; Novenergia (n 53) [656].

76 SunReserve (n 53) [678], [680], [684].

77 SunReserve (n 53) [800], [804], [813]; Blusun (n 53) [374]; RWE (n 53) [542], [548]; RREEF (n 53) [321]; Isolux Infrastructure v Spain (Award) (Arbitration SCC V2013/153, 6 July 2016), [793]; Hydro Energy (n 53) [592]–[594]; Stadtwerke (n 53) [261]; ICW (n 53) [544].

78 See, by analogy, ESPF (n 53) [418]; RREEF (n 53) [243].

79 SunReserve (n 53) [703], [817]–[818]; ESPF (n 53) [518]; RWE (n 53) [461]; Antin (n 53) [538]. Some tribunals have taken the broader view that an assurance to an investor can indeed be implied from generally-applicable legal instruments: see eg Novenergia (n 53) [650].

80 Representations or promises made through the ‘international law obligations’ and ‘treaties’ of the host State were explicitly referenced in this regard in: ESPF (n 53) [513]; and Frontier Petroleum. v Czechia (Award) (UNCITRAL Arbitral Rules, 12 November 2010) [285]. For the general proposition that legitimate expectations can arise from various sources and actions, see: SunReserve (n 53) [699]–[700], [770] and [788]. InfraRed (n 53) [409]–[410]; OperaFund (n 53) [483]; 9Ren (n 53) [265]–[266]; Antin (n 53) [548] and [552]–[553]; Antaris (n 53) [366]–[367]. Some tribunals have taken a narrower view, considering that commitments specifically addressed to a particular investor can give rise to legitimate expectations: Belenergia (n 53) [580]; Charanne (n 53) [490]–[493].

81 See, eg, ESPF (n 53) [418]; CEF Energia v Italy, SCC Arbitration V (2015/158), Award (16 January 2019) [240]–[242].

82 Council of the European Union, Joint Interpretative Instrument on the Comprehensive Economic and Trade Agreement (CETA) between Canada and the European Union and its Member States, [8] (No. 13541/16, Oct. 27, 2016).

83 See, eg, Eskosol (n 53) [388]–[389], [400], [402], [410]; Charanne (n 53) [535]–[536]; JSW (n 53) [391], [406]; Antaris (n 53) [444]; ICW European Investments (n 53) [535], [638]–[639]; Stadtwerke (n 53) [260]–[261], [320]–[321] and [354]; Isolux (n 53) [823]; Novenergia (n 53) [688]–[689]; Belenergia (n 53) [604]–[605]; SunReserve (n 53) [855].

84 For limited exceptions to this general feature of non-regression clauses in IIAs, see IIAs involving Canada as a State party applying a treaty-wide ‘general exceptions’ clause to the non-regression clause; see also IIAs with EFTA as a party, in which the non-regression clause requires the ‘encouragement of investment’ or ‘enhancement of a competitive trade advantage’ to be the ‘sole intention’ of the Party in engaging in the regression at issue, thus permitting a consideration of the underlying rationale (see, eg, Indonesia–EFTA FTA, art 8.3.4(a); Comprehensive Economic Partnership Agreement between Ecuador and the EFTA Member States (signed 25 June 2018, entered into force 1 November 2020) art 8.3.4(a) (EFTA–Ecuador FTA).

85 Weeramantry, JR, Treaty Interpretation in Investment Arbitration (OUP 2012) 6670Google Scholar. See also Bear Creek Mining v Peru, ICSID Case No. ARB/14/21, Award (30 November 2017) [473]–[474]; SunReserve (n 53) [678], [680], [684].

86 See, eg, RREEF (n 53) [545], [547]; RWE (n 53) [599]–[600], [732].

87 UNCTAD, Fair and Equitable Treatment (UNCTAD Series on Issues in IIAs II, 2012) 103–4.

88 Given that the WTO Agreement does not contain a non-regression clause, this analysis of trade-related non-regression clauses is limited to PTAs.

89 This is typical of the recent practice of the United States, the EU, China, Japan, Korea and Canada, see above (n 35).

90 See, eg, Agreement between Japan and Chile for a Strategic Economic Partnership (signed 27 March 2007, entered into force 3 September 2007) art 87 (Japan–Chile FTA); Free Trade Agreement between Israel and Colombia (signed 30 September 2013, entered into force 11 August 2020) art 10.14 (Colombia–Israel FTA).

91 For a discussion on the relationship between the international trade regime and the international investment regime, see: AD Mitchell and E Sheargold, Principles of International Trade and Investment Law (2021); J Kurtz, The WTO and International Investment Law: Converging Systems (2016).

92 See above section II.A.

93 Thomas, C and Tereposky, GA, ‘The NAFTA and the Side Agreement on Environmental Co-operation: Addressing Environmental Concerns in a North American Free Trade Regime’ (1993) 27 JWT 6Google Scholar, 13; Esty and Geradin (n 9) 313–14.

94 The North American Agreement on Environmental Cooperation—which was a side agreement to the original NAFTA—covers only failures to enforce environmental laws and does not encompass a trade-related non-regression clause.

95 Fletcher (n 8) 1–2; Feeley and Knier (n 8) 269–71; Office of Technology Assessment (n 8) 15–16; Chapman (n 8), section ‘Environment’; S Charnovitz, ‘Free Trade, Fair Trade, Green Trade: Defogging the Debate’ (1994) 27 CornellIntLJ 459, 468–70.

96 Senate Committee on Finance, ‘Hearing before the Committee on Finance: Jordan Free Trade Agreement’ (107th Session of Congress, 20 March 2001) 4; Report from the Committee on Finance on the Bipartisan Trade Promotion Authority Act of 2002 (Report 107-139, 28 February 2002) 30.

97 J Ruebner, US–Jordan Free Trade Agreement (Congressional Research Service Report for Congress, May 2001) 6, 8; Senate Committee on Finance (n 96) 43–4; E Harwood, ‘The Jordan Free Trade Agreement: Free Trade and the Environment’ (2002) 27 William and Mary Environmental Law and Policy Review 509, 520–3.

98 Agreement between the United States of America and Jordan on the Establishment of a Free Trade Area (signed 24 October 2000, entered into force 17 December 2001) art 5.1 (emphasis added) (US–Jordan FTA).

99 USTR, Final Environmental Review of the Agreement on the Establishment of a Free Trade Area Between the Government of the United States and Jordan (2000) 6, 8, 23, and annex III, 2; Office of the Press Secretary, Overview of U.S.–Jordan Free Trade Agreement (White House Press Release, 28 September 2001) 1–2.

100 Ruebner (n 97) 6, 8; Senate Committee on Finance (n 96) 4, 8, 10, 13, 29.

101 See also Senate Committee on Finance (n 96) 4.

102 Committee Report (n 96) 107–39 (emphasis added).

103 See above (n 89).

104 Some PTAs contain separate non-regression clauses with varying legal standards for international trade on the one hand, and international investment on the other: see, eg, Argentina–Chile FTA, arts 8.14 and 13.4.14. Some PTAs contain different clauses with varying legal standards that deal with international investment alone in one clause, and international trade and investment collectively in another: see, eg, Free Trade Agreement between China and Korea (signed 2 June 2015, entered into force 20 December 2015) arts 12.16, 16.5.2; CARIFORUM–EU FTA, arts 73 and 188.1; Indonesia–EFTA FTA, arts 4.8.2 and 8.3.4. Other PTAs set out differentiated legal standards for international trade and international investment in a single non-regression clause: see, eg, European Union–Central America Association Agreement (signed 29 June 2012, not yet in force, provisionally applied in part) art 291.2; Albania–EFTA FTA, art 34.2.

105 See, eg, Japan–Chile FTA, art 87; Colombia–Israel FTA, art 10.14.

106 US–Jordan FTA, art 5.1

107 The UNCTAD dataset records over 2850 IIAs whereas the WTO RTA records 304 PTAs.

108 See, eg, USMCA, art 24.4.3; CPTPP art 20.3.6; EEU–Vietnam FTA, art 12.4.3; EU–SADC FTA, art 9.3; Free Trade Agreement between MERCOSUR and the European Union (agreement in principle reached on 28 June 2019, not yet signed) Trade and Sustainable Development Chapter art 2.4 (EU–MERCOSUR FTA).

109 See Charnovitz, S, ‘The Law of Environmental ‘‘PPMs’’ in the WTO: Debunking the Myth of Illegality’ (2002) 27 YaleJIntlL 59, 62–5Google Scholar.

110 Korea–Turkey FTA, art 5.7.2.

111 Canada–Ukraine FTA, art 12.5.

112 Free Trade Agreement between Australia and Peru (signed 12 February 2018, entered into force 11 February 2020) art 19.3.3.

113 New Zealand–Malaysia Agreement on Environmental Cooperation (signed 26 October 2009, entered into force 1 August 2010) art 2.6.

114 Indonesia–EFTA FTA, art 8.3.4.

115 Argentina–Chile FTA, art 13.4.14.

116 USTR, Bipartisan Agreement on Trade Policy (May 2007) 2–3; USTR, The United States–Canada–Mexico Agreement Fact Sheet: Environment (December 2019) 1; CB Rangel, ‘Moving Forward: A New, Bipartisan Trade Policy That Reflects American Values’ (2008) 45 HarvJonLegis 377, 395; MA Villarreal and IF Ferguson, USMCA: Amendment and Key Changes (Congressional Research Service, 20 January 2020) 1–3.

117 Rangel (n 116) 396, 397 and 399.

118 See Non-paper from the Netherlands and France on trade, social economic effects and sustainable development (July 2020). See also M Bronckers and G Gruni, ‘Retooling the Sustainability Standards in EU Free Trade Agreements’ (2021) 24 JIEL 25, 37–9 and 49–50.

119 eg China–Georgia FTA, arts 9.2 and 9.6; China–New Zealand Protocol, Ch 22 arts 3.2 and 9.2.

120 See, eg, Australia–United States Free Trade Agreement (signed 18 May 2004, entered into force 1 January 2005) [2005] ATS 1, art 19.7.5 (AUSFTA); Colombia–Korea FTA, art 20.2; Free Trade Agreement between Canada and Korea (signed 22 September 2014, entered into force 1 January 2014) arts 8.10, 8.18, 17.5.3 and 17.15; China–New Zealand Protocol, Ch 22 arts 3.2 and 9.1; EFTA–Ecuador FTA, art 8.13.

121 See, eg, USMCA art 24.32; CPTPP art 20.23; CETA art 24.14.10; EU–Japan FTA, art 16.18.5; EU-MERCOSUR FTA, art 17.9; Canada–Ukraine FTA, art 12.21; EU–Korea FTA, art 13.15; Colombia–US FTA, art 18.12.

122 See above (n 116).

123 CETA art 24.16.1; EU–Japan FTA, art 16.17.1; EU–MERCOSUR FTA, arts 15.5 and 17.11; EU–Korea FTA, art 13.16. The Brexit Agreement (arts 410.2, 410.3, 749 and 750) represents a noteworthy exception.

124 CETA art 24.15.11; EU–Japan FTA, art 16.18.6; EU–MERCOSUR FTA, art 17.11; EU–Korea FTA, art 13.15.2.

125 See, eg, R Francis, ‘EU Ministers want FTA labour and environment chapters to be upgraded’ (Borderlex, 17 October 2022).

126 B Waterfield, ‘Channel sewage ‘‘could violate Brexit deal’’’ (The Times, 2 September 2022).

127 USTR, ‘USTR Requests First-Ever Environment Consultations Under the U.S.-Peru Trade Promotion Agreement (PTPA)’ (Press Release, 1 April 2019); USTR, ‘USTR Announces Enforcement Action to Block Illegal Timber Imports from Peru’ (Press Release, 19 October 2020); USTR, ‘USTR Announces Unprecedented Action to Block Illegal Timber Imports from Peru’ (Press Release, 19 October 2017).

128 USTR, ‘USTR to Request First-Ever Environment Consultations Under the U.S.-Korea Free Trade Agreement (KORUS) in Effort to Combat Illegal Fishing’ (Press Release, 19 September 2019).

129 Panel Report, Guatemala –Article 16.2.1(a) of the CAFTA-DR (4 June 2017).

130 Report of Panel of Experts Constituted under Article 13.15 of the EU–Korea Free Trade Agreement (20 January 2021).

131 USTR, ‘United States Seeks Mexico's Review of Alleged Worker's Rights Denial at Auto Manufacturing Facility’ (Press Release, 5 May 2021); AFL-CIO, ‘Filing of First USMCA ‘‘Rapid Response Mechanism’’ Labor Case to Fight for Mexican Workers Denied Independent Union Representation’ (Press Release, 10 May 2021).

132 Villarreal and Ferguson (n 116) 1–3.

133 See above (n 118).

134 European Parliamentary Research Service, The level playing-field for labour and environment in EU-UK relations (European Parliament, April 2021) 1, 7–10.

135 See eg proposals and measures by the US and EU to counteract weak environmental standards as a market distortion: USTR, Advancing Sustainability Goals through Trade Rules to Level the Playing Field: Draft Ministerial Decision (WTO Doc WT/GC/W/814, 17 December 2020); European Commission, The EU's new trade defence rules and first country report (Memo, 20 December 2017) 3.

136 See above (n 1).

137 Final Report of the Arbitration Panel under the Association Agreement between the EU and Ukraine, Ukraine – Wood Products, [227], [246]–[252].

138 ibid [271]–[273], [342].

139 ibid [339], [341]–[342].

140 Appellate Body Reports, US – Gambling [308]; Brazil – Retreaded Tyres [156]; Colombia – Textiles [5.115]; Korea – Various Measures on Beef [176], [178]; China – Publications and Audiovisual Products [318]; India – Solar Cells [5.59], n 214; EC – Seal Products [5.200]; Australia – Salmon [199]; US – COOL (Article 21.5 – Canada and Mexico) [5.266], n 761. As stated above, nothing in this article expresses a view on obligations under the WTO Agreement or the meaning of associated case law.

141 See, eg, USMCA, arts 31.3 and 32.4; CPTPP, art 28.12.3; Australia–China FTA, art 15.9.2; RCEP, 19.4.2; EU–Korea FTA, art 14.16; EU–Central America FTA, art 322.2.

142 For analogous examples where the means of pursuing compliance may involve lowering the actual degree of contribution made by the challenged measure in line with the intended level of protection, see Panel Report, Japan – Apples (Article 21.5) [8.193]; Appellate Body Report, Korea – Beef [178]–[180].

143 See, eg, J Pauwelyn and LE Salles, ‘Forum Shopping Before International Tribunals: (Real) Concerns, (Im)Possible Solutions’ (2009) 42 CornellIntLJ 77, 81–3.

144 See, eg, CETA, art 24.5.1; EU–Japan FTA, art 16.2.2; China–Georgia FTA, art 9.2; Canada–Ukraine FTA, art 12.5; Korea–New Zealand FTA, art 16.2.2; EU–Mexico FTA, arts 2.3 and 2.4; Colombia–US FTA, art 18.3.3.

145 See, eg, USMCA, art 24.4.3; CPTPP, art 20.3.6; Indonesia–EFTA FTA, art 8.3.4; EFTA–Ecuador FTA, art 8.13.

146 L Boisson de Chazournes, ‘One Swallow Does Not a Summer Make, but Might the Paris Agreement on Climate Change a Better Future Create?’ (2016) 27 EJIL 253, 254–5; M Prieur, ‘The Principle of Non-Regression’ in M Faure, Elgar Encyclopedia of Environmental Law (Edward Elgar 2016) 251, 257; M Vordermayer-Riemer, Non-Regression in International Environmental Law (Intersentia 2020) 332 and 429 (albeit recognising that this is not necessarily reflective of a ‘principle’ at 435). See also Report of the Secretary-General, Gaps in international environmental law and environment-related instruments: towards a global pact for the environment (UNGA, A/73/419, 30 November 2018) [22].

147 M Prieur and G Garver, ‘Non-regression in environmental protection: a new tool for implementing the Rio Principles’ (2012) Future Perfect: Rio+20 30, 30–1; M Prieur, ‘Non-Regression in Environmental Law’ in G Mainguy (ed), Surveys and Perspectives Integrating Environment and Society (vol 5, No 2, 2012) 53, 54; Prieur (n 146) 251, 254 and 256; E Rehbinder, ‘Contribution to the Development of Environmental Law’ (2012) 42 Environmental Law and Policy 210, 213; Vordermayer-Riemer (n 146) 2–3. See also NA Robinson, ‘Reflecting on Measured Deliberations’ (2012) 42 Environmental Policy and Law 219, 223.

148 European Parliament resolution of 29 September 2011 on developing a common EU position ahead of the United Nations Conference on Sustainable Development (Rio+20) (P7_TA(2011)0430) [97].

149 International Centre for Comparative Environmental Law, ‘Non-regression in environmental policy and law’ (2012) 1. See also Rehbinder (n 147) 213.

150 IISD, Earth Negotiations Bulletin (5 July 2015) 5.

151 Prieur (n 147) 55.

152 Prieur and Garver (n 147) 30; Prieur (n 146) 251, 253–4; Prieur (n 147) 54.

153 Secretary-General Report (n 146) [22], [102].

154 Secretary-General Report (n 146) [22].

155 Colombia, Documento de Discusión – Posición de la Republica de Colombia – Tercera Sesión Pacto Global, 4; China, Comments of China to the Ad Hoc Open-Ended Working Group Established Pursuant to General Assembly Resolution 72/277 (19 February 2019) 6.

156 IISD, Earth Negotiations Bulletin (21 January 2019) 5–6.

157 UN Ad Hoc Open-ended Working Group, Report of the ad hoc open-ended working group established pursuant to UNGA resolution 72/277 (UN Doc A/AC.289/6/Rev.2, 13 June 2019) [54].

158 See, eg, Prieur (n 146) 251, 253–4 and 258.

159 See, eg, Boisson de Chazournes (n 146) 254–5; Vordermayer-Riemer (n 146) 332 and 429.

160 Paris Agreement (opened for signature 22 April 2016, entered into force 4 November 2016) art 4.3 (emphasis added). See also art 3.

161 Conference of the Parties, Decision 1/CP.20: Lima Call for Climate Action (Report of the Conference of the Parties on its twentieth session, 1–14 December 2014, FCCC/CP/2014/10/Add.1) [10].

162 Ibid, annex: ‘Elements for a draft negotiating text’.

163 Summary of the round tables under workstream 1 on the 2015 agreement ADP 2, part 1 Bonn, Germany, 29 April–3 May 2013 Note by the Co-Chairs, [13]–[14] and [44] <http://unfccc.int/resource/docs/2013/adp2/eng/5infsum.pdf>.

164 Submission by Mexico to the Ad-Hoc Working Group on the Durban Platform for Enhanced Action (ADP), pp. 1 and 3 <http://unfccc.int/files/bodies/application/pdf/submission_by_mexico,_indicative_elements_for_a_lbi.pdf>.

165 Submission by South Africa on the Determination and Communication of Parties’ Intended Nationally Determined Contributions May 2014, p. 7 <http://unfccc.int/files/bodies/application/pdf/adp_indc_southafrica.pdf>.

166 Submission by Ireland and the European Commission on Behalf of the European Union and its Member States (Dublin, 27 May 2013): Process for Ensuring Ambitious Mitigation Commitments in the 2015 Agreement, para (v), <https://unfccc.int/files/documentation/submissions_from_parties/adp/application/pdf/adp_eu_workstream_1_20130527.pdf>.

167 Submission of Views on the Work of the ADP Coalition for Rainforest Nations, para 16, <http://unfccc.int/files/documentation/submissions_from_parties/adp/application/pdf/adp2-5_submission_by_cfrn_20140604.pdf>.

168 New Zealand Submission to the Ad Hoc Working Group on the Durban Platform for Enhanced Action, para 8(c), accessed 29 October 2018 <http://unfccc.int/files/documentation/submissions_from_parties/adp/application/pdf/adp2-4_submission_by_new_zealand_submission_20140312.pdf>.

169 See, eg, Independent Association of Latin America and the Caribbean (AILAC) Submission on the Ad Hoc Working Group on the Durban Platform (ADP), p. 4 <http://unfccc.int/files/documentation/submissions_from_parties/adp/application/pdf/adp2.4_submission_by_ailac_20140310.pdf>; Submission by Sudan on behalf of the Africa Group: Elements of the draft negotiation text under the ADP, para 3, <http://unfccc.int/files/bodies/application/pdf/adp_w1_elements_africangroup.pdf>.

170 See D Bodansky, J Brunnée and L Rajamani, International Climate Change Law (OUP, 2017) 235–6.

171 Summary of the round tables under workstream 1 on the 2015 agreement ADP 2, part 1 Bonn, Germany, 29 April–3 May 2013 Note by the Co-Chairs, [13]–[14] and [44]; Submission by Mexico to the Ad-Hoc Working Group on the Durban Platform for Enhanced Action (ADP), 1, 3; Submission by South Africa on the Determination and Communication of Parties’ Intended Nationally Determined Contributions May 2014, 7; Independent Association of Latin America and the Caribbean (AILAC) Submission on the Ad Hoc Working Group on the Durban Platform (ADP), 4; Submission by Sudan on behalf of the Africa Group: Elements of the draft negotiation text under the ADP, [3]; Submission by Ireland and the European Commission on Behalf of the European Union and its Member States (Dublin, 27 May 2013): Process for Ensuring Ambitious Mitigation Commitments in the 2015 Agreement, (v); Submission of Views on the Work of the ADP Coalition for Rainforest Nations, [16]; New Zealand Submission to the Ad Hoc Working Group on the Durban Platform for Enhanced Action, [8(c)],

172 Prieur (n 146) 253–4; Vordermayer-Riemer (n 146) 332–5 and 518.

173 See AH Benjamin, ‘We, the Judges and the Environment’ (2012) 29 PaceEnvtlLRev 582, 589–90; Explanatory Memorandum for Climate Change Bill 2022 (Cth), para 23.

174 DR Boyd, Constitutions, Human Rights, and the Environment: National Approaches’ in A Grear and LJ Kotzé (eds), Research Handbook on Human rights and the Environment (Edward Elgar 2016) 170, 172.

175 ibid 170, 189.

176 ibid 197–198.

177 RL Glicksman, ‘The Justifications for Nondegradation Programs in U.S. Environmental Law’ (Legal Studies Research Paper No. 2012-46, George Washington University Law School, 2012) 1–2.

178 Resolution 48/13, ‘The human right to a safe, clean, healthy and sustainable environment’ (Human Rights Council, A/HRC/48/L.23/Rev.1, 8 October 2021), [1] (‘HRC Res 48/13’).

179 HRC Res 48/13, preambular recital 15 and fn 3.

180 JH Knox, Mapping Report (A/HRC/25/53, 30 December 2013) [80]; Report of the Special Rapporteur on the issue of human rights obligations relating to the enjoyment of a safe, clean, healthy and sustainable environment (A/HRC/31/52, 1 February 2016) [67], [68], [75].

181 See, eg, WA Schabas, The European Convention on Human Rights: A Commentary (OUP 2015) 611–15; S Joseph and M Castan, The International Covenant on Civil and Political Rights: Cases, Materials, and Commentary (3rd edn, OUP 2013) 922.

182 Adopted by UNGA Resolution 217 A(III) of 10 December 1948. See Secretary-General Report (n 146) [22], fn 76.

183 B Saul, D Kinley and J Mowbray, The International Covenant on Economic, Social and Cultural Rights: Commentary, Cases, and Materials (OUP 2013) 152; D Shelton and A Gould, ‘Positive and Negative Obligations’ in D Shelton (ed), The Oxford Handbook of International Human Rights Law (OUP 2013) 569; M Scheinin, ‘Core Rights and Obligations’ in D Shelton (ed), The Oxford Handbook of International Human Rights Law (OUP 2013) 537.

184 C Bruch et al, Environmental Rule of Law: First Global Report (UNEP 2019) 151–2.

185 See (n 6) art 3(c).

186 PM Dupuy and JE Viñuales, International Environmental Law (2nd edn, CUP 2018) 386; N Bryner, ‘A constitutional right to a healthy environment’ in D Fisher (ed), Research Handbook on Fundamental Concepts of Environmental Law (Edward Elgar 2016) 168, 171–3; P Sands and J Peel with A Fabra and R Mackenzie, Principles of International Environmental Law (4th edn, CUP 2018) 815.

187 Knox (n 180) [80].

188 See A Boyle, ‘Environment and Human Rights’ in R Wolfrum (ed), Max Planck Encyclopedia of Public International Law (article updated April 2019) [5], [33].

189 Group of Experts for the Global Pact for the Environment, White Paper: Toward a Global Pact for the Environment (September 2017) 32–4.

190 See further LR Helfer, ‘Rethinking Derogations from Human Rights Treaties’ (2021) 115 AJIL 20, 22–5.

191 Knox (n 180) [55] (citing Committee on Economic, Social and Cultural Rights in its general comment No. 3, [32], and Committee's general comment No. 15, [19]).

192 Knox (n 180) [67].

193 EPA (n 1).

194 Revised Explanatory Memorandum, Clean Energy Legislation (Carbon Tax Repeal) Bill 2014

195 See (n 62).

196 Climate Change Authority, 2014 Renewable Energy Target Review – Report, 17.

197 See Explanatory Memorandum, Renewable Energy (Electricity) Amendment Bill 2015; The Hon. G Hunt MP (Minister for the Environment) and The Hon. I Macfarlane MP (Minister for Industry and Science), ‘Certainty and growth for renewable energy Joint media release’ (Press Release, 23 June 2015).

198 Wharburton review (Expert Panel), Renewable Energy Target Scheme – Report of the Expert Panel (2014) 17.

199 See, eg, S Bell and D McGillivray, Environmental Law (7th edn, 2008 OUP) 48; J Black, ‘The Role of Risk in Regulatory Processes’ in R Baldwin, M Cave and M Lodge (eds), The Oxford Handbook of Regulation (OUP 2010) 301, 317–23; BR Copeland and M Scott Taylor, ‘Trade, Tragedy, and the Commons’ (2009) 99 AmEconRev 725, 725.

200 JWS Solar v Czech Republic, PCA Case No. 2014-03, Final Award (11 October 2017) [375], [377], [382].