Not all investments are created equal. In some instances foreign direct investment (FDI) leads to substantial job creation, positive spillover effects, and the introduction of novel and useful technologies. In other cases, FDI crowds out local investment, creates pollution and contributes little to nothing to the economy. This has always been the case, but in a climate-constrained world it is ever more apparent that FDI is not an unmitigated good.
In 2015, 187 governments took a step closer to reaching the conclusion that investment in fossil fuels must end by signing the Paris AgreementFootnote 1 under the United Nations Framework Convention on Climate Change (UNFCCC).Footnote 2 In this agreement, countries committed to keeping average global temperature change below 2 degrees Celsius (°C) of warming above pre-industrial levels. McGlade and Ekins have demonstrated that in order to have a reasonable chance of meeting this commitment globally, a third of oil reserves, half of gas reserves and over 80% of current coal reserves should remain unused from 2010 to 2050.Footnote 3 This means that a substantial proportion of known fossil fuel deposits would have to be left in the ground and further exploration activities could not be undertaken. While it is true that most governments have yet to accept this reality, barring an unlikely and dramatic technological breakthrough (for example, the development of a viable and scalable method of carbon capture and storage), they will have to begin curtailing fossil fuel investments in the very near future.
In this context, it is important to recognize that over 3,000 bilateral investment treaties (BITs) include provisions to provide legal protection for all forms of foreign investment, including investment in sectors that must eventually be rendered obsolete to prevent warming beyond the 2°C limit set by the Paris Agreement. Additionally, regional trade agreements such as the North American Free Trade Agreement (NAFTA),Footnote 4 as well as a new wave of agreements (not yet in force) – namely the Trans-Pacific Partnership (TPP),Footnote 5 the Transatlantic Trade and Investment Partnership (TTIP),Footnote 6 the Regional Comprehensive Economic Partnership (RCEP),Footnote 7 and the Comprehensive Economic and Trade Agreement (CETA)Footnote 8 – will have chapters on investment protection. Specifically, all of these agreements (except the CETA) will provide corporations with access to a process known as investor-state dispute settlement (ISDS).Footnote 9 This process allows foreign investors to bypass local courts and bring claims for monetary compensation to an international tribunal. Investment tribunals are composed of three private parties (arbitrators): one chosen by the state involved in the dispute; one chosen by the investor; and a third that is mutually agreed upon or, failing that, appointed by an arbitral body such as the International Centre for the Settlement of Investment Disputes (ICSID).
In 2015, the year in which the Paris Agreement and the TPP were signed (although not yet in force), a record number of 70 ISDS cases were launched. The United Nations Conference on Trade and Development (UNCTAD) has documented 767 known disputes in total (a lack of transparency in the system means that this could be a substantial underestimation).Footnote 10 The ISDS cases brought by tobacco giant Philip Morris against Uruguay and Australia over the introduction of strict cigarette packaging laws are perhaps the most notorious.Footnote 11 However, disputes related to tobacco or health more broadly are quite rare. Of all the disputes registered with ICSID up to 2015, 26% concerned oil, gas and mining (more than any other sector) and another 15% concerned electric power and other energy resources.Footnote 12 Thus, 41% of all ICSID cases relate to the energy sector or extractive industries.
Recently, there has been a surge of ISDS cases launched by renewable energy companies (primarily against Spain).Footnote 13 Nonetheless, long-established industry players bring the majority of claims. The ISDS system works best for incumbents because it was designed to ensure stability in the investment environment. States are expected to maintain the conditions that were conducive to attracting incumbent investors in the first place, and may be punished if they change their policy position. Such a framework provides cause for concern given the significant policy developments required to transition to a low-carbon economy. At present, there are no known cases where an investor has taken the step of formally launching ISDS over the introduction of a carbon tax, emissions trading scheme or renewable energy incentive scheme. The latter have been challenged only when modified with retroactive effect.Footnote 14 However, claims have certainly been contemplated, including against the Australian government’s proposed carbon pollution reduction scheme in 2009.Footnote 15
ISDS cases have arisen in response to:
∙ a moratorium on oil and gas operations within 12 nautical miles of the Italian coastline;Footnote 16
∙ restrictions placed on a coal-fired power station by the German city of Hamburg;Footnote 17
∙ a ban by the Canadian Province of Quebec on hydraulic fracturing (‘fracking’, a process with potentially significant climate impacts through methane releases as well as implications for groundwater);Footnote 18 and
∙ the United States (US) Obama administration’s rejection of a proposal by TransCanada Corporation to build the Keystone XL pipeline to transport oil produced from Alberta’s tar sands to various refineries in the US.Footnote 19
The German case was settled and, at the time of writing, the case in Quebec had not concluded, whereas the Italian case was only in the very early stages. TransCanada withdrew its claim against the US following President Trump’s executive order allowing construction of the Keystone XL pipeline to move ahead.Footnote 20
Obama’s rejection of Keystone XL was probably influenced by his desire to demonstrate leadership on climate change in the run-up to the Paris Climate Conference.Footnote 21 While it is unlikely that the existence of the ISDS case factored into Trump’s decision to allow the pipeline to proceed, it is worth noting that when the case was launched, a number of legal experts suggested that TransCanada (seeking US$15 billion in damages) had a good chance of winning.Footnote 22 In a different jurisdictional context, a case like this would have had the potential to create regulatory chill, which is the main focus of this article.
The regulatory chill hypothesis suggests that governments will fail to regulate in the public interest in a timely and effective manner because of concerns about ISDS.Footnote 23 This article outlines three distinct varieties of regulatory chill: internalization chill, threat chill, and cross-border chill. It also details why regulatory chill is of particular concern in the climate change policy space, drawing parallels between the fossil fuel and tobacco industries. Finally, the article addresses how some governments are dealing with regulatory chill: by introducing ‘safeguards’ and, in the case of the TPP, specifically carving out tobacco policies from ISDS provisions. Trump’s withdrawal from the TPP has rendered it defunct (although there is still the possibility that a modified ‘TPP minus one’ version will be negotiatedFootnote 24 ), but its text continues to have an influence on other agreements, such as the RCEP. This article argues that, in these agreements, governments should not limit themselves to a narrow tobacco carve-out and that Big Oil actually poses a greater threat to public policy than Big Tobacco. One option would be for negotiators to broaden the scope of the carve-out to cover all government measures taken in pursuit of international obligations, including those found in climate change treaties.
2. VARIETIES OF REGULATORY CHILL
There is a common misconception that governments can be forced by ISDS tribunals to roll back regulations that have been put in place; this is not the case.Footnote 25 As such, ISDS has no direct impact on health and environmental regulation. However, there is significant potential for indirect impacts to occur as a result of the effect ISDS has ‘on the way in which host states exercise their regulatory powers’.Footnote 26 ISDS is expected to have an effect on states because of the substantial financial risk involved (governments have been found liable for hundreds of millions and, in some cases, even billions of dollars) as well as the difficulty in predicting outcomes.Footnote 27 The debate on this topic in the literature and in policy circles has generally been framed around the phenomenon known as regulatory chill.Footnote 28 As indicated earlier, the regulatory chill hypothesis suggests that in some instances (not all, or there would be no ISDS cases) governments will fail to enact or enforce bona fide regulatory measures (or modify measures to such an extent that their original intent is undermined or their effectiveness is severely diminished) as a result of concerns about ISDS.Footnote 29
Two distinct types of regulatory chill have been discussed in the literature. This article proposes a third variant.
2.1. Internalization Chill
Internalization chill is a process whereby regulatory progress is dampened across all areas that affect foreign investors because government officials are concerned about the risk of an investor-state dispute arising.Footnote 30 Policy makers take into account potential disputes with foreign investors before they even begin to draft a policy, thereby prioritizing the avoidance of such disputes over the development of efficient regulation in the public interest. Internalization chill is extremely difficult to measure. It requires ‘counterfactual evidence about the regulations that would have existed in the absence of the purported chilling’, in addition to untangling the regulatory decision-making process and identifying the key factors that resulted in the failure to adopt a measure.Footnote 31
Evidence of internalization chill in the literature is limited and mixed. Côté conducted interviews with regulators in health, safety and environment agencies in the Canadian federal government and concluded that overall awareness of ISDS tended to be low across the board and, as such, that the potential for internalization chill was low.Footnote 32 However, she did find that awareness was greater among the highest-ranking officials and in those agencies that had been directly involved in disputes. Van Harten and Scott also conducted interviews with regulators in Canada, but at the provincial level (a significant percentage of ISDS cases in Canada have arisen over subnational government measures).Footnote 33 They found that government ministries have changed their decision making to account for concerns about ISDS and that government legal advisers play a key role in assessing ISDS risks.
The findings of Van Harten and Scott suggest that, while the everyday regulator may have no awareness of ISDS (as suggested by Côté), practices developed within government to ensure that policies are vetted by lawyers with expertise in investment law prior to adoption could lead to regulatory chill.Footnote 34 This is also supported by the comments of one of the world’s leading arbitration lawyers, Toby Landau, who stated on an Australian radio programme in 2014:
Without a doubt. Regulatory chill, so-called, in my opinion definitely exists, and there’s palpable evidence of it. There are those who deny it, but I can say that, in my role as counsel, on a number of occasions now, I’ve actually been instructed by governments to advise on possible adverse implications or consequences of a particular policy in terms of investor-state cases.Footnote 35
A number of countries now have mechanisms in place to screen health and environmental policies for potential trade and investment impacts prior to their implementation.Footnote 36 For example, Regulatory Impact Analysis Statements in Canada necessitate the consideration of trade and investment implications of any new regulation.Footnote 37 Several governments in the developing world have also recently embarked on comprehensive ISDS implementation policies, some of which include the development of ‘dispute prevention’ mechanisms.Footnote 38 The UNCTAD Secretariat has assisted the governments of Peru, Guatemala, Panama and the Dominican Republic in setting up such policies and mechanisms. The focus of UNCTAD’s work is to help developing countries in preventing costly disputes from arising. However, the organization does not appear to have considered the impact that these types of arrangement will have on regulatory authority. Will they increase the likelihood of settlements in which regulations are changed or not implemented and thus, in effect, institutionalize regulatory chill? Alternatively, will they conversely increase regulatory confidence in the ability of a measure to survive a challenge and thereby reduce any chilling effect? Although these types of arrangement have been mentioned in discussions of regulatory chill,Footnote 39 they have not been adequately assessed from a regulatory perspective.
2.2. Threat Chill
Threat chill is the most familiar form of regulatory chill. It concerns the chilling of specific regulatory measures that have been proposed by governments following an investor’s threat to arbitrate. A regulatory lack of awareness about ISDS is irrelevant to this type of chill. In fact, a lack of knowledge about the specifics of investment law makes the threat of arbitration all the more potent, because regulators will be less likely to recognize when an investor is bluffing.Footnote 40
The fact that investors brandish ISDS as a threat is indisputable; law firms actively advertise ISDS as a useful tool ‘to assist lobbying efforts to prevent wrongful regulatory change’.Footnote 41 The empirical question is, therefore, whether governments take these threats seriously and change their behaviour as a result.
Defenders of ISDS argue that if measures are bona fide public policies, governments do not have to worry about threats of arbitration.Footnote 42 However, such assertions are overly simplistic. First of all, in practice regulation is a messy business and governments frequently have multiple motivations for adopting regulatory measures. The extant case law demonstrates that the various actors involved in ISDS cases have held widely diverging opinions on where the line between bona fide and illegitimate measures should be drawn. Furthermore, regulators are subject to bounded rationality: they experience time and resource constraints and face serious limitations on their ability to predict the outcomes of legal cases.Footnote 43 This is particularly true in ISDS because the ambiguous nature of the provisions found in investment treaties leaves significant scope for arbitrators to interpret the law (for example, the requirement to provide fair and equitable treatment), and because awards rendered in investment arbitration are binding only on the parties involved in the dispute (there is no system of precedent). In this environment of uncertainty, governments may have a distorted perception of their chances of success in arbitration, and may also be reluctant to risk a negative outcome because the stakes involved in ISDS are very high. Even if a government wins a case, it may be burdened with a significant legal bill.Footnote 44 Regulators in developing countries are likely to be especially cautious about inciting an ISDS case given the limited capacity of their governments to cover this level of costs.
As Miles notes: ‘In operation, [regulatory chill] is generally a subtle creature and is difficult to detect. And yet, it has the potential to frustrate initiatives designed to implement national and global environmental objectives’.Footnote 45 One example of threat chill that has been detailed in the literature,Footnote 46 and more recently in the media,Footnote 47 is Indonesia’s decision to allow open-pit mining in protected forests following a threat of arbitration. In-depth case studies of this nature are often dismissed as anecdotal. To some extent, this may reflect a bias in the trade and investment field against qualitative social science methodologies. However, it is true that results have sometimes been inconclusive because researchers are unable to find a smoking gun. Government officials are wary of disclosing information about ISDS threats for a variety of reasons and would be understandably reluctant in many cases to admit that they abandoned a bona fide measure that was in the public interest out of fear of ISDS, especially if they have signed or continue to sign treaties providing access to arbitration.
Furthermore, an arbitration threat is likely to be just one in a suite of tools employed by an investor faced with a potentially negative regulatory change. Investors will also employ more traditional lobbying tactics and threats to exit the jurisdiction,Footnote 48 with consequences in terms of job losses and lost government revenue. Distinguishing the relative importance of these various tactics in influencing a government is very difficult. Additionally, they may act to reinforce one another. In Australia, for example, the tobacco industry launched a multimillion dollar public relations campaign to oppose legislation mandating the plain packaging of cigarettes. Included in some of the advertisements in this campaign was the claim that the legislation would waste taxpayers’ money because it would mire the government in costly litigation, including ISDS.Footnote 49 While this strategy ultimately proved unsuccessful for Big Tobacco, it points to the fact that the threat of ISDS can be used in multiple ways.
2.3. Cross-Border Chill
Recent experience with tobacco control measures suggests that there is a need to distinguish a third form of regulatory chill. Tobacco corporations have certainly directly threatened countries with legal action – for example, in Namibia, Togo and UgandaFootnote 50 – but a broader implicit threat has also been created by Philip Morris actually launching ISDS cases in Uruguay and Australia.Footnote 51
Defenders of ISDS have continued to misunderstand how regulatory chill operates in this context and cite the recent victories of Uruguay and Australia as proof that a chilling effect does not exist.Footnote 52 There are a number of flaws with this reasoning. In the case of Uruguay, the government has acknowledged that it would not have been able to defend itself in ISDS without the financial support of a foundation set up by Michael Bloomberg.Footnote 53 The country’s success in the case, therefore, does not disprove regulatory chill. Instead, it reinforces the notion that the high cost of ISDS can be sufficient to dissuade a government from defending a policy that would ultimately be determined to be compliant with international investment law. As for Australia, the case was decided on jurisdiction; consequently, it has no bearing on the regulatory chill debate.
More importantly, the real issue with these cases is not who won and who lost (most experts predicted that Philip Morris would lose),Footnote 54 although it is concerning that there was a dissenting opinion in the Uruguay case. Rather, the issue is how tobacco corporations exploited the existence of these cases over the many years during which they dragged on and how governments have responded. Margaret Chan, Director General of the World Health Organization (WHO), has stated that the ISDS cases were ‘deliberately designed to instill fear’ in countries trying to reduce smoking.Footnote 55 As Gruszczynski points out, the aggressive strategy taken by Big Tobacco against Australia, which is a relatively small market for cigarettes, can be explained by ‘the precedential nature of the new law and a fear that the packaging requirements will be copied in other jurisdictions’, and further notes that ‘their approach sends a strong signal to other countries contemplating the introduction of plain packaging laws that any such initiative may be challenged’.Footnote 56 Both Gruszczynski and Bonnitcha argue that New Zealand’s decision to delay plain packaging until the dispute against Australia had been resolved is a clear-cut case of regulatory chill.Footnote 57
Based on this example, this article proposes a third distinct variety of regulatory chill: cross-border chill. Investors pursue this type of chilling effect when a government adopts a policy that affects a form of investment common to many jurisdictions, is easily transferable, and is highly likely to be emulated by other governments. The most effective way to induce this type of chill is to actually launch an ISDS case in one or more jurisdictions (preferably the first jurisdiction that adopts a policy if an investment treaty is accessible). The aim from the investor’s perspective in launching a case is not necessarily to prevent or delay the take-up of a policy in the jurisdiction in which the claim is brought – this jurisdiction might not even represent a particularly important market or investment location – but in all other jurisdictions in which the investor is active. It is significant to note that even if a delay in the adoption of a policy is the best that an investor can hope for (for example, if it has very little chance of actually winning the ISDS case), the lengthy duration of ISDS cases can mean that the investor reaps significant economic rewards from this strategy, which far outweigh the costs of the ISDS case itself.
Dow AgroSciences LLC v. Canada is another case in which cross-border chill, this time of the subnational rather than national variety, was the likely aim of the investor.Footnote 58 In this case, Dow challenged Québec’s 2003 Pesticides Management Code,Footnote 59 which banned the use of the pesticide’s active ingredient 2,4-D for cosmetic purposes (i.e. lawn care) in the province. Québec’s regulators relied on the precautionary principle as justification for their ban of 2,4-D in the absence of conclusive scientific evidence of its environmental and health impacts. At the time, numerous other municipal and provincial governments were in the process of considering similar cosmetic pesticide bans.Footnote 60 Dow sought CAD$2 million plus costs. This paltry sum, which was lower than the average cost to participate in an ISDS case, led several observers to hypothesize that Dow initiated the dispute in the hope of having a chilling effect on other jurisdictions in Canada that were considering or developing similar bans.Footnote 61 The case was ultimately settled and does not appear to have had a detrimental impact on the regulation of pesticides in Canada. Nevertheless, Dow clearly thought that it was worth trying and, in a different context, it might have been a more viable strategy.
3. FOSSIL FUELS AND TOBACCO: FIGHTING FOR SURVIVAL
As noted above, a number of ISDS cases relate to climate change. There may be many other cases that have not (yet) come to light in which the threat of arbitration has been used to stall action on climate change. However, the focus of this section is not on what might have already occurred but instead on what is likely to occur in the future. It is argued here that the strong parallels between the tactics used by the tobacco industry and the fossil fuel industry in response to proposed regulation suggest that when countries begin to ramp up action on climate change, fossil fuel corporations will employ ISDS in the same way that Philip Morris has.
In recent years, a number of authors have charted the collaborations between the tobacco and fossil fuel industriesFootnote 62 and commented on the use by both industries of ‘merchants of doubt’: scientists or faux experts willing to argue against scientific consensus for a fee.Footnote 63 The Center for International Environmental Law (CIEL) has documented how the ‘tobacco playbook’ was actually invented by the fossil fuel industry, and how the two industries have, over the years, shared advertising agencies and public relations firms, as well as ‘experts’.Footnote 64 In the context of this article, what is most important is that all of these efforts for the past 50 to 60 years have been focused primarily on delaying regulatory action.
A number of parallels can be drawn between the tobacco and fossil fuel industries that explain their focus on stalling policy development. They also show why ISDS is a viable tactic for achieving this in addition to the tried-and-true methods of misinformation and lobbying. The first parallel is that both industries face ‘existential threats’ and are ‘fighting for survival’.Footnote 65 This is not the case for all industries faced with increased regulation, which often only translates into increased costs and possibly lower profit margins. Tobacco cannot be made healthy, nor can burning fossil fuels be made environmentally friendly (unless one believes the fantasy of ‘clean coal’). If a government or the public decides that putting an end to lung cancer or climate change is a priority, then this means that the tobacco or fossil fuel industries must be regulated out of existence. While corporations in such a situation may consider selling different products, they are likely to fight tooth-and-nail to market their existing products for as long as possible because of their substantial sunk investments in the sector (such as infrastructure and control of commodity chains). Corporations in both the tobacco and fossil fuel industries have experimented with alternative products (e-cigarettes and renewable energy, respectively), but neither has been willing to give up on the more profitable and well-established products. The continued desire of corporations in these industries to hold on for as long as possible means that they will use every strategy at their disposal, including ISDS. This is true even if they know that they are highly unlikely to win a case, because the years during which proceedings are ongoing may deliver them considerable profits in other jurisdictions.
The second parallel is that neither industry has a particularly good reputation with the public. Appearing ‘nice’ is largely irrelevant to their brand value. For some corporations, bringing a massive claim for compensation against a sovereign state might have a negative impact in terms of public relations. For example, Bechtel dropped its ISDS case against Bolivia following a concerted international civil society campaign, presumably because the harm being done to the firm’s reputation outweighed the potential benefits of winning the case.Footnote 66 There is no indication that suing Australia and Uruguay has had any impact on Philip Morris cigarette sales.
Thirdly, both industries are being targeted for regulation at the international level. In the case of tobacco, there is the Framework Convention on Tobacco ControlFootnote 67 and, in the case of fossil fuels, there are the UNFCCC and its Paris Agreement, as well as other initiatives such as the G20 pledges to reduce fossil fuel subsidies.Footnote 68 While framework agreements are largely voluntary and do not specify the particular actions that countries should take, they certainly play a role in pressuring states to act and also amplify the potential for a policy developed in one state to be mimicked in others. This is what happened with plain packaging legislation: first introduced in Australia, it has had a domino effect around the world. However, Philip Morris successfully delayed the falling of some of the dominoes by using ISDS. Fossil fuel corporations will hope to do the same when governments begin to adopt stricter climate policies.
Finally, both industries are dominated by a small group of extremely wealthy and truly multinational corporations. This means that they have the means and opportunities to engage in ISDS practically anywhere in the world. If one corporation does not have direct access to a treaty, the investment can be restructured through a subsidiary in a country where a treaty relationship exists. Philip Morris lost its case against Australia because it restructured its investment through Hong Kong, but other tribunals have been willing to accept jurisdiction when investors have engaged in ‘treaty shopping’.Footnote 69 Tobacco and fossil fuel corporations can also collaborate across their respective industries to bring cases in different parts of the world or to pile on in one jurisdiction to have maximum impact. The costs of ISDS, while high for many governments, are insignificant for corporations of this size.
Both industries are clearly aware of the power of ISDS and have therefore lobbied for its inclusion in agreements like the TPP. Philip Morris made a public submission to the US Trade Representative on the TPP, describing ISDS as ‘vital’, and effectively mapping the countries that the company wanted to sue.Footnote 70 Similarly, in a submission to the US Trade Representative in 2013, Chevron argued that the existence of ISDS panels ‘increases the likelihood’ of disputes being settled outside them.Footnote 71 Furthermore, the minutes of a meeting in April 2014 between unnamed Chevron executives and European Commission officials (obtained by The Guardian under access to information laws) note that Chevron believes that ‘the mere existence of ISDS is important as it acts as a deterrent’.Footnote 72 However, while tobacco and fossil fuel corporations both recognize the power of ISDS, the fossil fuel industry has far more experience with the process in practice than the tobacco industry does. Fossil fuel corporations have been launching contractual and treaty-based investment disputes over direct expropriations and other issues, such as tax increases, for decades. Thus, they are already comfortable with ISDS and understand how and when to use it. Applying this tactic to the new area of climate change policy will be fairly straightforward.
In summary, fossil fuel corporations are faced with an existential threat to their business model; they have the resources to engage in ISDS and no good reason not to. Critically, they do not need to believe that they will win the cases that they launch, simply that dragging a policy through ISDS in one jurisdiction might delay its adoption in another. In light of this, one could reasonably question why there have not to date been any ISDS cases directly focused on climate policy. The obvious answer is that, thus far, the other tactics employed by fossil fuel corporations (such as lobbying and misinformation campaigns) have been very successful in preventing the adoption of any policies that would be worth challenging in ISDS.
4. RESPONDING TO REGULATORY CHILL: SAFEGUARDS AND THE TPP CARVE-OUT
In order to respond to concerns both about the impact that ISDS cases have on the public purse and the potential for regulatory chill, governments have made attempts to tighten up the language in regional trade agreements as well as in new BITs. The aim has been to restrict the scope for creative interpretation of provisions by arbitrators, but experts are divided about whether any of the ‘safeguards’ that have been introduced will prove effective in protecting public policies from challenge.Footnote 73 In some cases, it is clear that the provisions that are lauded by governments are merely window dressing. For example, Article 9.16 of the TPP states:
Nothing in this Chapter shall be construed to prevent a Party from adopting, maintaining or enforcing any measure otherwise consistent with this Chapter that it considers appropriate to ensure that investment activity in its territory is undertaken in a manner sensitive to environmental, health or other regulatory objectives.Footnote 74
This provision provides no real protection for public policies because the term ‘otherwise consistent’ renders it tautological. However, it is worth addressing in greater detail some of the other, more significant, safeguards. This section will focus primarily on those included in the TPP, as this is the most significant recent trade agreement to complete the negotiation stage. Although the TPP is now effectively defunct – at least in any form that will include the US – it is considered by many actors (such as the governing Liberal/National Coalition in Australia and the former National Party-led government in New Zealand) to be the ‘gold standard’ of trade agreements and it is likely that its provisions will crop up elsewhere, particularly in the RCEP, which has a significant degree of overlap in membership.
4.1. National Treatment and Most Favoured Nation Treatment
One would think that the clauses in investment treaties that prevent governments from discriminating between investors on the basis of nationality would be fairly uncontroversial. Unfortunately, arbitral tribunals have read ‘national treatment’ and ‘most favoured nation (MFN) treatment’ clauses broadly as covering not only de jure discrimination but also de facto discrimination. In other words, there does not have to be evidence of intent to discriminate on the part of the state.
Some treaties, including the TPP, restrict the definitions of national treatment and MFN treatment by including the qualifier that these provisions apply only to investors in ‘like circumstances’. In theory, this approach offers a narrower scope for comparison than treaties with no qualifying language, but in practice a large ambit for interpretation remains. Tribunals are free to determine the criteria on which to assess which investors/investments are in like circumstances; there is no requirement that a tribunal must consider issues such as the environmental impact of an investment activity, which from a state’s perspective could provide a justifiable reason for differentiation.Footnote 75 If a tribunal is willing to consider the regulatory context (for instance, taking into account environmental and health policy objectives), it is likely to place the burden on the state to show that the discrimination was ‘reasonable’.Footnote 76 Furthermore, even where policy objectives are deemed to be reasonable, a tribunal may still find that the regulatory measures themselves are not if the same objectives could have been achieved in a less restrictive way.Footnote 77
Miles suggests that states that differentiate between investments on the basis of sustainability goals are at risk of facing investor claims for breach of the national treatment standard.Footnote 78 In the decision in Bilcon v. Canada the tribunal determined that the decision of an independent environmental review panel constituted a breach of the national treatment standard because a higher standard of environmental review was applied than was the case for other investments in like circumstances.Footnote 79 Canada’s arguments that the other projects reviewed were different in scope, in different locations, and involved different concerns (and therefore were not in like circumstances) were rejected by the tribunal.Footnote 80
The TPP contains a footnote that is intended to deal with this issue:
For greater certainty, whether treatment is accorded in ‘like circumstances’ under Article 9.4 (National Treatment) or Article 9.5 (Most-Favoured-Nation Treatment) depends on the totality of the circumstances, including whether the relevant treatment distinguishes between investors or investments on the basis of legitimate public welfare objectives.Footnote 81
However, this language will not be effective in preventing cases like Bilcon from arising under the TPP, because investors have proved their willingness to question before tribunals the legitimacy of government actions taken in the public interest. As Johnson and Sachs point out, a far more effective safeguard would have been to ‘clearly specify that a foreign investor seeking to recover on a non-discrimination claim must establish that the government discriminated against it on account of its nationality’.Footnote 82
An additional problem with the MFN provision in the TPP is its potential use by investors to access provisions in other investment treaties. In Maffezini v. Spain the tribunal decided that the MFN clause in the Spain–Argentina BIT could be applied to allow the claimant to have access to dispute resolution provisions in a BIT between Spain and Chile.Footnote 83 The Maffezini decision prompted states to explicitly carve out ISDS provisions from the application of the MFN clause in more recent treaties. The TPP follows this trend.
However, investors have also successfully imported the substantive provisions of other treaties through the application of MFN.Footnote 84 The CETA attempts to deal with this issue, stating:
Substantive obligations in other international investment treaties and other trade agreements do not in themselves constitute ‘treatment’, and thus cannot give rise to a breach of this article, absent measures adopted by a Party pursuant to such obligations.Footnote 85
The 2015 Indian Model BITFootnote 86 deals with the problem even more effectively by simply excluding any mention of MFN in the text (the country having been stung by the use of MFN to import substantive obligations from another treaty in a case brought by an Australian investor).Footnote 87
The TPP does not contain an effective safeguard against the ‘MFN loophole’ and, as such, it is possible that investors will try to exploit it, thus rendering the other safeguards in the TPP irrelevant. George Kahale III, chairman of Curtis, Mallet-Prevost, Colt & Mosle LLP (a leading arbitration firm), argues that including an MFN clause in the investment chapter of the TPP was a ‘major mistake’ and it is ‘a dangerous provision to be avoided by treaty drafters whenever possible’ because it can turn one bad treaty into protections ‘never imagined for virtually an entire world of investors’.Footnote 88
4.2. The International Minimum Standard and Fair and Equitable Treatment
The international minimum standard treatment (MST) and fair and equitable treatment (FET) (which could be two distinct standards or one and the same, depending on the treaty in question and the person interpreting it) are the most problematic standards in investment treaties from the perspective of the right to regulate.
The difficulty stems primarily from the fact that ‘the concepts “fair” and “equitable” are by themselves inherently subjective and therefore lacking in precision’.Footnote 89 Under one interpretation of the standard, states are required to meet the ‘legitimate expectations’ of the investor.
The most common safeguard is to clarify, as the TPP does, that FET does not require treatment in addition to or beyond that which is required under customary international law. Unfortunately, this safeguard has proved to be ineffective in practice because some arbitrators view the decisions of past tribunals (who have interpreted the standard broadly) as now forming part of customary international law. The same safeguard was introduced in the NAFTAFootnote 90 in 2001 through a joint note of interpretation by the parties.Footnote 91 This did not prevent the majority of the NAFTA tribunal in Bilcon v. Canada from determining that an alleged violation of Canadian domestic law amounted to a breach of customary international law. The US and Mexico have criticized the Bilcon award in submissions filed in another arbitration for failing ‘to properly assess customary international law’.Footnote 92
A further safeguard in the TPP’s investment chapter is the qualifier to MST/FET that ‘[t]he mere fact that an action is taken that is inconsistent with an investor’s expectations does not constitute a breach … even if there is loss or damage to their investment as a result of that action’.Footnote 93 Unfortunately, this language not only fails to act as a safeguard but could actually expose governments to greater risk. Firstly, no finding of an MST/FET violation has ever rested solely on the fact that an investor’s expectations were not met. Todd Weiler, a Canadian lawyer/arbitrator and proponent of ISDS, has noted that he ‘can’t recall any tribunal that, if you put this provision in that agreement, the result would be different either way’.Footnote 94
Secondly, in the past, governments (including the US) have clearly asserted that an investor’s ‘expectations’ are not to be considered at all in the determination of a breach of MST/FET. The TPP, by stating that a breach of an investor’s expectations does not alone give rise to a claim, ‘implicitly recognizes that “expectations” may in fact be relevant to establishing a violation of the FET standard’.Footnote 95 Thus, the TPP ‘codifies – rather than corrects – problematic decisions such as the Bilcon award’.Footnote 96
The preferable option to safeguard against expansive interpretations of the MST would be to exclude any mention of ‘fair and equitable treatment’. This is the approach taken by India in its 2015 Model BIT.
4.3. Indirect Expropriation
Historically, the direct taking of foreign property has been one of the most significant risks to foreign investment. Instances of outright taking are now considered rare in most parts of the world. For the last 20 years, the key debate in academic and policy circles has been on the coverage in investment treaties of so-called indirect expropriation. Indirect expropriation falls short of actual physical taking of property but results in the effective loss of management, use or control, or a significant depreciation in the value of the assets of a foreign investor.Footnote 97
In establishing whether or not an indirect expropriation has occurred, tribunals have tended to adopt one of two basic approaches. Under the first approach, the tribunal focuses solely on the effect of the regulation on the investor.Footnote 98 To evaluate the effect of a measure, tribunals are likely to examine both its economic impact and its duration. Tribunals which adopt the second approach will also examine the effect of a measure on an investor, but will additionally address its purpose. The tribunal will assess whether a measure was taken for a legitimate public purpose and may also evaluate whether the need to fulfil that purpose is proportionate to the negative effect felt by the investor. Given the difficulty of drawing a bright line between bona fide non-compensable regulation and a taking, many commentators and arbitrators suggest that such a determination can be achieved only on a case-by-case basis.Footnote 99
Safeguards to prevent tribunals from adopting overly broad, investor-friendly interpretations of indirect expropriation that might interfere with the right to regulate have typically followed the US model, which is based on domestic case law on regulatory takings.Footnote 100 This model relies on an annex with two distinct elements. The first element is a three-part test which addresses the effect of a measure, its character, and the extent to which it interferes with the investor’s ‘reasonable’ or ‘legitimate’ expectations. The second element is a statement that ‘except in rare circumstances’ non-discriminatory regulatory measures are not considered to be expropriations.
In some treaties, a provision (usually stated before the three-part test) is included which stipulates the level of deprivation that must have occurred for a measure to be considered a possible expropriation. For example, the Taiwan–New Zealand BIT (2013) states that ‘[i]n order to constitute indirect expropriation, the Party’s deprivation of the investor’s property must be: (a) either severe or for an indefinite period’.Footnote 101 The Canada–Korea Free Trade Agreement (2014) states: ‘An action or series of actions by a Party cannot constitute an expropriation unless it interferes with a tangible or intangible property right in an investment and eliminates all or nearly all of its value’.Footnote 102
In the treaty practice of countries such as Canada and Colombia, an additional note is included in the expropriation annex clarifying what ‘rare circumstances’ might entail: ‘such as when a measure or series of measures is so severe in the light of its purpose that it cannot be reasonably viewed as having been adopted and applied in good faith’. Some countries (Turkey, for example) opt to exclude the ‘rare circumstances’ language altogether.Footnote 103
The TPP adopted the US model annex without any of these additional qualifications. The only amendment to the standard model is a footnote with the following explanation:
Whether an investor’s investment-backed expectations are reasonable depends, to the extent relevant, on factors such as whether the government provided the investor with binding written assurances and the nature and extent of governmental regulation or the potential for government regulation in the relevant sector.Footnote 104
In some other treaties, this issue is dealt with by using the language of ‘specific commitments’ rather than ‘investment-backed expectations’. The term ‘specific commitments’ is narrower than ‘investment-backed expectations’, although some investment law experts maintain that neither belongs in the analysis of whether a regulatory measure is an expropriation.Footnote 105
What the TPP safeguard achieves is a rejection of the sole-effects approach to determining whether an indirect expropriation has occurred. However, it leaves open the door to a proportionality analysis, which Sornarajah argues has been imported into investment law from the trade context to ‘keep alive the idea that compensation is possible even in the case of regulatory expropriations’.Footnote 106
It is unclear what impact Article 3b (the ‘except in rare circumstances’ clause) might have in protecting the right to regulate. As Edsall notes in relation to a similar clause in the Central American Free Trade Agreement (CAFTA):Footnote 107 ‘Given the vague nature of the term “rare circumstances”, it could be a comparatively broad exception to the presumption in favor of “nondiscriminatory regulatory actions”’.Footnote 108 It is also unclear how this clause relates to those that precede it. In an intervention into a case brought under CAFTA against Costa Rica, the US government made it clear that it does not view the relationship between the clauses in the Annex on indirect expropriation as hierarchical, noting that the paragraph stating that regulatory measures are not expropriatory except in rare circumstances ‘is not an exception, but rather is intended to provide tribunals with additional guidance in determining whether an indirect expropriation has occurred’.Footnote 109 This case, which arose over Costa Rica’s move to establish a park to protect the nesting sites of endangered leatherback sea turtles, also demonstrates the willingness of investors to argue that their case is a ‘rare circumstance’.
4.4. The Tobacco Carve-Out
What the above demonstrates is that, despite the introduction of a number of safeguards, substantial ambiguity exists in the text of the TPP. Ambiguity equals regulatory uncertainty and thus regulatory chill remains a concern. However, there is one safeguard included in the TPP that was widely expected to be effective in countering regulatory chill: the so-called tobacco carve-out. Article 29.5 of the TPP on Tobacco Control Measures allows governments to ‘elect to deny the benefits of Section B of Chapter 9’ (i.e. ISDS) ‘with respect to claims challenging a tobacco control’.Footnote 110 Unless a government wishes to leave itself open to litigation over a policy like requiring plain packaging of tobacco, it would not be, at least not under the TPP.
Although some questions have been raised about the wording of the carve-out (particularly the fact that invoking it is optional) and what it fails to cover (tobacco leaf and possibly e-cigarettes), its inclusion in the agreement is generally considered a substantial win for health advocates.Footnote 111 However, somewhat ironically, the mere existence of the carve-out tends to create at least the perception that the other safeguards in the TPP’s investment chapter are inadequate to shield health policies from investor challenges.Footnote 112
A cynic might suggest that the carve-out was included simply to quell opposition to the TPP from civil society, particularly in Australia. However, a recent development casts some doubt on that theory. In October 2016, Australia and Singapore quietly adopted an amendment to their existing free trade agreement by inserting a tobacco carve-out.Footnote 113 This move, made after the Philip Morris claims against Australia and Uruguay had been dismissed, and with no fanfare, suggests that one or both of these countries continues to view tobacco ISDS claims as a threat. The carve-out serves to rule out the possibility of an ISDS case (although only from corporations based or structured through Australia or Singapore) and thereby any cross-border chilling effect from cases brought in other jurisdictions.
While widely viewed as effective,Footnote 114 there are critics of the tobacco carve-out at both ends of the political spectrum. Under the influence of the tobacco lobby, Republicans in the US Senate held up the ratification of the TPP in 2016, in part because they objected to the carve-out.Footnote 115 They expressed concern about a ‘slippery slope’ where tobacco would just be the first in a long line of industries to be excluded from protection. This concern is shared by many ISDS proponents. For example, Mercurio argues: ‘After tobacco, other disfavoured industries could subsequently be targeted – perhaps alcohol, sugar or even mining’.Footnote 116 At the other end of the spectrum, opponents of ISDS would welcome any or indeed all of these industries being excluded from the investment protection provisions in trade agreements. For these critics the problem with the TPP carve-out is not that it could set a dangerous precedent, but that it is too narrow in scope. They argue that tobacco control is not the only policy area worthy of special protection.Footnote 117
This article sides with the latter group of critics, based on the evidence that far more ISDS cases have been brought in relation to environmental regulations than to tobacco control measures, and the concerns outlined earlier in this article about the specific threat posed by fossil fuel companies challenging measures taken to address climate change. A carve-out for climate change measures in the TPP would have been far more significant than the tobacco carve out. The European Parliament has, in effect, called for such a carve-out in the TTIP (an agreement with an uncertain future under the Trump administration).Footnote 118 In the lead-up to the December 2015 Paris conference, the European Parliament adopted a resolution calling on the European Commission and European Union (EU) Member States to:
ensure that any measure adopted by a Party to the Paris Agreement relating to the objective of stabilising greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system, or relating to any of the principles or commitments contained in Articles 3 and 4 of the UNFCCC, will not be subject to any existing or future treaty of a Party to the extent that it allows for investor-state dispute settlement.Footnote 119
Finally, a preferable alternative to carving out either tobacco or climate change measures is to carve them both out, as well as all other measures adopted by governments in pursuit of their international obligations. The Framework Convention on Tobacco ControlFootnote 120 could be explicitly listed alongside the climate treaties as well as a myriad of other environmental, human rights and indigenous rights agreements, leaving the option for further agreements to be added to the list as they are signed.
Investment agreements have been designed primarily to protect the status quo. Conversely, compliance with the objectives of the Paris Agreement will require radical change: a future in which governments have met the collective goal of keeping below the 2°C guardrail is a future without fossil fuels. Civil society and governments at all levels will have to fight for this future, regardless of whether any of the recently negotiated regional trade agreements ever actually come into force. However, providing fossil fuel corporations with ISDS under these agreements is akin to handing your opponent extra weapons and ammunition before stepping onto the battlefield. Fossil fuel corporations will always have sufficient incentive to bring ISDS cases because they are fighting for their survival. For as long as there is any ambiguity in the substantive provisions of investment agreements – allowing cases to play out over several years, cost millions, and leave governments uncertain about outcomes – there will be policy delays. In a rapidly warming world, we simply cannot afford these delays.
The only provision that has been introduced into recent negotiations that would be remotely effective at protecting public policy from investor challenges is the tobacco carve-out found first in the (now defunct) TPP and more recently inserted into the Australia–Singapore Free Trade Agreement.Footnote 121 However, in the future, negotiators should not confine themselves to carving out tobacco control measures. Instead, it is logical to carve out all government measures taken in furtherance of obligations under international agreements. Of course, an even simpler option is to leave investment protection out of regional trade agreements altogether (and cease to sign BITs), given the complete absence of any evidence that providing foreign investors with greater legal rights than those of domestic investors has any public benefit.