Investor-State Dispute Settlement and the Future of the Precautionary Principle

Free trade agreements can be traced back to the ancient empires which existed before the Common Era.⁽¹⁾ However, the heyday of trade expansion came in the nineteenth century when the transport innovations of the industrial revolution, combined with Empire, and an adherence to laissez-faire economics - particularly in Great Britain - opened up enormous trade opportunities across the world.⁽²⁾ An early example of bilateralism, which featured the now-familiar most favoured nation status, was the Cobden-Chavalier agreement of 1860 between Great Britain and France. This led to a series of similar European agreements which created something of a golden era of trade in Europe. Unfortunately this was short-lived; an economic depression in the 1870s ushered in a period of increasing protectionism which was ultimately to lead to great instability and nationalism culminating in a half-century of economic and military conflict in the twentieth century.⁽³⁾ Following World War II, the General Agreement on Tariffs and Trade - the first major multilateral trade agreement and the forerunner of the WTO⁽⁴⁾ - was created in 1948, having been salvaged from the failed attempt to create an international trade organisation at a conference of states in the Cuban capital in 1947.⁽⁵⁾

GATT, and later, the WTO, facilitated the growth of the global economy in the later twentieth and twenty-first centuries, and offered an independent dispute settlement service in the form of the WTO’s dispute settlement body and appellate panel. These offered a means of settlement of trade disputes that avoided, initially at least, some of the political and diplomatic hurdles presented by state-state negotiation. However, the WTO seems to have become a victim of its own success. Viviane de Beaufort takes the view that despite its considerable success in attempting to broker agreements between 159 states, progress at the WTO has foundered on: “confrontations between states”; the increasingly complex nature of the negotiations which have moved from concerns with tariffs to complex technical matters such as phytosanitary protection and intellectual property; and the rise of “concerns on sovereignty.” ⁽⁶⁾ As a result multilateralism on the WTO model has suffered a decline in popularity originating in “a general lack of enthusiasm from states and … WTO governance issues.” ⁽⁷⁾ Consequently states have moved towards bilateral or multilateral free trade agreements, including the recent so-called Mega-Treaties, brokered without direct input from the WTO. A number of such agreements also contain provisions on investment and these tend to follow established patterns in bilateral investment treaties (BITs), and commonly define the terms for investments between states, make provision for most-favoured nation status, define what is meant by fair and equitable treatment, prohibit expropriation and may even make provision for dispute settlement, which may permit an investor to sue a host state directly.

In recent times, the United States has been particularly active in embracing this model of trade and, since its first free trade agreement with Israel in 1985, is now party to 20 free trade agreements,⁽⁸⁾ including the North American Free Trade Agreement (NAFTA),⁽⁹⁾ and the recently concluded Trans-Pacific Partnership (TPP).⁽¹⁰⁾ It is also a party to 42 bilateral investment treaties⁽¹¹⁾ as well as 52 trade and investment framework treaties.⁽¹²⁾ While the use of BITs by the United States may appear impressive it is dwarfed by Germany’s 135 (of which 32 are in force) and the United Kingdom’s 106 (96 in force). In fact, the proliferation of this type of agreement worldwide has been such that the United Nations Conference on Trade and Development puts the current number of BITs at 2948, of which 2317 are in force.⁽¹³⁾ The proliferation of BITs was near-exponential between 1970 (five agreements) and the peak in 1996 (221 agreements). Since then, the scope for new BITs has declined with only 11 BITs concluded in 2013.⁽¹⁴⁾

This increase of BITs and investment chapters in free trade agreements is coupled with a spectacular increase of case load in investor-state dispute settlement (ISDS) fora. The last summary of the case load statistics for the International Centre for the Settlement of Investment Disputes (ICSID) indicated that 52 applications were currently registered. Just a decade earlier, in 2005, the figure was approximately half that (at 27 applications), and in 1995 was a mere three applications.⁽¹⁵⁾ In the years 1972 to 1994, ICSID applications totalled only 32; by December 2015 this total had risen to 549. This has occurred against the background of a rise in preference for arbitration generally as evidenced by the growth in other arbitral fora. ⁽¹⁶⁾

The reasons for the rise in popularity in BITs and the introduction of investment chapters in free trade agreements, particularly the recent Mega-Treaties are not difficult to find. There are distinct advantages to states in brokering agreements directly and, as Caroline Foster has pointed out, such agreements “encourage and support international commercial private actors’ freedoms to invest, disinvest, repatriate capital, buy and sell goods and services, employ, navigate, exploit communal resources, and take business decisions.”⁽¹⁷⁾ As such, both developing and developed states favour investment agreements; developing states, because they facilitate foreign direct investment, and developed states, because they offer an attractive trading platform for private actors - invariably economically important multinational corporations - amongst others, with the introduction of an ISDS mechanism which avoids the diplomatic and political vagaries of state-state settlement‥

However, there are some distinct disadvantages to ISDS. In particular, it permits aggrieved investors to take host states directly to international arbitration; it is not necessarily the case that the investor need first exhaust domestic remedies,⁽¹⁸⁾ especially if they consider that seeking such remedies might be futile.⁽¹⁹⁾ Once at ICSID, or a similar forum, the nature of the ISDS that follows is profoundly different from normal “private” arbitration, in that one of the parties (the state) is constrained by public interest considerations that do not affect the other. This use of arbitration has been described as “internationalized public law”⁽²⁰⁾ and, in effect, takes the established principles of domestic public law and places them in an international arbitration forum which is not necessarily well suited for the purpose. Arbitration was developed for private investors as a relatively swift alternative to contentious litigation where the remedy was decided according to the dictates of commercial expediency and with a view to maintaining the commercial relationship. The rules of natural justice, due process, stare decisis and the requirements of transparency (i.e. justice being seen to be done) are secondary considerations. Perhaps most controversial is the fact that decisions of ICSID tribunals are not subject to appeal,⁽²¹⁾ apart from an annulment procedure which can overturn a decision on limited grounds.⁽²²⁾ This limitation does not apply to disputes involving non-parties who may take advantage of the ICSID Additional Facility rules; such decisions are subject to appeal.⁽²³⁾

The paper will examine just one aspect of ISDS, but one which has profound implications for regulatory sovereignty in nation states. At issue is the question of whether BITs and investment chapters in recent Mega-Treaties permit investors to challenge regulatory decisions on the basis, inter alia, that they amount to a lack of fair and equitable treatment for investors or are tantamount to expropriation of their assets. This is most likely to arise where regulation is based on an interpretation of scientific data (often related to public health or environmental protection) which is susceptible to different regulatory responses. In the European Union such decisions are often predicated on the precautionary principle - which is an important element of European primary law⁽²⁴⁾ - whereas in the United States and Canada, the precautionary principle is viewed with suspicion if not outright hostility.⁽²⁵⁾ Should an arbitral tribunal be tasked with having to decide whether the regulatory decision is a proportionate one, this could embroil it, not so much in an assessment of the scientific evidence per se, but in an assessment of the relative merits of the precautionary principle compared to the cost-benefit approach favoured elsewhere in the world. In other words an assessment of how uncertainty in scientific evidence should best be built into regulatory decision-making. This, in effect, places a tribunal in the position of adjudicating in the public interest, a role normally associated with a fully constituted judiciary. Moreover, its decision, if made according to normal arbitration principles, could have the effect of subjugating national sovereignty to private commercial interest. It might also permit one state to use its private sector investors to “adjust” or “chill” regulation in another. This is certainly a fear among some in the European Parliament as it continues to debate the negotiation of the Transatlantic Trade and Investment Partnership (TTIP) between the United States and the European Union.⁽²⁶⁾ This paper attempts to assess the reality of that fear.

However, rather than discuss this issue in purely abstract and theoretical terms, an attempt has been made to analyse the possibilities through a real-life and current issue, namely the European Commission’s restriction on the use of neonicotinoid insecticides. Hence this paper will assess what the possibilities might be were an investor to challenge this restriction under the terms of the proposed (though partly hypothetical)⁽²⁷⁾ investment chapter of TTIP and to reach some tentative conclusions on the reality of the threat to regulatory sovereignty in respect of the precautionary principle.

The European Commission has been at great pains to reassure concerned parties (including the “European Parliament, Member States, national parliaments and stakeholders”⁽²⁸⁾) that any ISDS arrangements in TTIP will not affect regulatory sovereignty. In its press release in September 2015 on new proposals for an “Investment Court System for TTIP and other EU trade and investment negotiations” it reassured readers that “governments’ right to regulate would be enshrined and guaranteed in the provisions of the trade and investment agreements.”⁽²⁹⁾

The Commission was perhaps well-advised to issue this reassurance since scrutiny of the operation of investment chapters in existing agreements, by analysis of the arbitral decisions that have emerged from them, could lead an analyst to a rather different conclusion. This Part will examine the general nature of the protections for environmental regulatory sovereignty in typical investment agreements, proposed and extant, before examining, in Part III, some of the arbitral decisions which have engaged with these types of provisions.

Modern model investment treaties such as the 2012 U.S. Model BIT⁽³⁰⁾ contain, at least on the face of it, built-in protection of the ability of states to regulate without interference from external investors. For example, Article 12 of the 2012 U.S. Model BIT concerns “Investment and Environment” and makes impermissible any attempts by states to:

This recognises the well-established fact that weakening environmental regulatory standards can have the effect of distorting trade and seeks to avoid a race to the bottom in terms of environmental protection.⁽³²⁾

The 2012 U.S. Model BIT also recognises the sovereignty of states in promulgating environmental regulation. Article 12(5) states that:

Thus treaties based on the 2012 U.S. Model BIT not only attempt to protect the sovereignty of states in maintaining, adopting and enforcing environmental regulation but also ensure that states do not weaken protection in order to attract investment.

The Comprehensive Economic and Trade Agreement (CETA) between Canada and the European Union, concluded in September 2014,⁽³⁴⁾ is viewed by many as presaging the likely final contents of the TTIP and has drawn on existing model BITs (including the 2012 U.S. Model BIT) for many of its provisions.⁽³⁵⁾ CETA makes provision for investment protection⁽³⁶⁾ which is similar to existing model BIT texts, though the investment chapter was significantly modified following legal analysis such that:

Hence the principal changes to the investment chapter in CETA were to create a bespoke Tribunal and Appellate Tribunal for the purposes of investor-state dispute settlement, including bespoke rules for the appointment of members of the arbitration and appellate tribunals. This is in contrast to the original text of CETA which drew on the existing services of ICSID. The new version makes use of certain of the rules of the ICSID Convention and of the ICSID Secretariat’s services but the arbitration hearings themselves will take place outside the ICSID system. This change was made to assuage some of the concerns, particularly in Europe, over the appointment and independence of ICSID arbitrators and the unavailability of appeal against ICSID arbitration decisions under Article 53(1) of the ICSID Convention.

In terms of CETA’s protection of environmental regulation in state parties’ territories, the investment chapter makes several concessions. Under Section B of CETA (Market Access),⁽³⁸⁾ the following do not constitute a denial of market access for the purposes of investment:

Similarly, for the purposes of Section D of CETA (Investment Protection),⁽⁴⁰⁾ which includes the provisions for fair and equitable treatment, compensation, expropriation, transfers and subrogation, ⁽⁴¹⁾ a clarification is included at the beginning of the section:

The recently agreed TPP uses similar language in its investment chapter in relation to performance requirements:

In relation to the TPP investment chapter as a whole, a general provision is included in Article 9.15:

The latest text of the investment chapter of the putative TTIP - though evidently incomplete⁽⁴⁵⁾- is very similar to that of CETA, and also makes provision for a bespoke “Tribunal of First Instance” and a “permanent appeal Tribunal” for investor-state dispute settlement.⁽⁴⁶⁾ It also contains articles, like CETA, relating to fair and equitable treatment, compensation, direct and indirect expropriation, transfers, subrogation and denial of benefits.⁽⁴⁷⁾

Very similar protections for important public interest matters are to be found in other, more established, trade agreements such as the North American Free Trade Agreement (NAFTA) and these, unlike the texts of CETA, TTP and TTIP, have been the subject of arbitral decisions. For example, the preamble to NAFTA, inter alia, exhorts the three state parties to “STRENGTHEN the development and enforcement of environmental laws and regulations”⁽⁴⁸⁾ and Article 1114 relates specifically to the investment chapter of the agreement and states that:

On the face of it such provisions should prevent interference with the environmental regulatory sovereignty which states must exercise in the public interest and prevent the dismantling of existing regulation as a means of attracting foreign direct investment. However, the reality has not always reflected these expectations. Some arbitral decisions have suggested that investor interests can, in certain circumstances, outweigh the right of a state to exercise sovereignty. It is to these decisions that we now turn.

Most ISDS proceedings have taken place under the auspices of the ICSID and a number of these have involved challenges based on attempts - at least on the face of it - to protect the environment or to preserve local sovereignty in respect of activities perceived to be environmentally unacceptable or contrary to local or national public health imperatives.

Perhaps the most controversial of these was the Metalclad arbitration, issued in August 2000 and heard under the Additional Facility rules of the ICSID Convention.⁽⁵⁰⁾ These rules were applied owing to the fact that Mexico, the state party, was not a party to the ICSID convention but had opted to use ICSID as a forum to determine a dispute relating to the construction and operation of a hazardous waste facility in the La Pedrera valley near Guadalcazar in the State of San Luis Potosi in southern Mexico.⁽⁵¹⁾ The arbitration was required to determine whether the Mexican municipal authorities had acted in such a way, in preventing the operation of the site (in response to strong local opposition), as to contravene Articles 1105 and 1110 of the NAFTA which state:

In short, the U.S. and Canadian investors in Metalclad were seeking to show that the lack of transparency in the operation of the Mexican regulatory system relating to hazardous waste sites, and the decision of the local municipality to create a nature reserve which included the entire site (thereby rendering it inoperable), amounted to a lack of fair and equitable treatment and to measures tantamount to expropriation.

The decision in relation to Article 1105(1) was based on the fact that there was considerable uncertainty on the part of the Mexican federal, state and municipal authorities as to which of them was in fact responsible for the permitting of hazardous waste facilities.⁽⁵²⁾ Metalclad had legitimately, and in good faith, relied on federal assurances about the legality of the construction and operation of the site. The absence of regulatory clarity, and uncertainty as to practice and procedure at the various levels of governance amounted to a lack of transparency. Given the strength of local opposition⁽⁵³⁾ it is perhaps understandable that the municipal authority so assiduously tried to prevent construction and operation of the plant. Nonetheless, the uncertainty unquestionably affected Metalclad’s ability to make decisions about, and to realise, their investment. The fact that ultimately this uncertainty resulted in an arbitral decision which compensated foreign investors at the expense of local autonomy may be seen as unfortunate, but Metalclad’s treatment was arguably unfair and inequitable treatment by the standards of international law, particularly as they had engaged extensively with domestic procedures in an attempt to resolve the issue before turning to ICSID.

However, aspects of the panel’s comments on the Article 1110 infringement were much more controversial. Much of the decision appeared to follow logically from the Article 1105 finding, on the basis that such a degree of opacity in the regulatory and legal process was not only unfair and inequitable treatment, but on such a serious scale as to qualify also as a measure tantamount to expropriation. However, the panel went further and considered the state governor’s decision to issue an Ecological Decree which encompassed the site as part of the Article 1110 analysis:

This was to attract considerable informed criticism and adverse judicial treatment in the aftermath. Many commentators considered that the Tribunal - which was led by no less a figure than Professor Sir Elihu Lauterpacht, CBE, QC - had construed the term “indirect measures tantamount to expropriation” unacceptably widely and in such a way as to trespass on the legitimate exercise of state sovereignty in environmental matters. Professor Phillipe Sands has characterised the Tribunal as taking “a very broad approach to the definition of expropriation”⁽⁵⁶⁾ and has written elsewhere that such decisions represent a worrying and retrograde development in the requirement to balance the interests of investment with those of environmental protection.⁽⁵⁷⁾

The decision received judicial scrutiny at the hands of Judge David F. Tysoe of the Supreme Court of British Columbia.⁽⁵⁸⁾ He reviewed the reasoning of the Tribunal and concluded that the Metalclad Tribunal had exceeded its jurisdiction when deciding whether the Mexican government had contravened NAFTA Article 1105 ostensibly by importing the concept of transparency into the text of the Article as one of the objectives of the NAFTA. In Tysoe J.’s view the international law concept of fair and equitable treatment does not incorporate the concept of transparency and the Tribunal exceeded its jurisdiction in importing it.⁽⁵⁹⁾ The question of transparency was thus beyond the scope of the submission to arbitration “because there are no transparency obligations contained in Chapter 11.”⁽⁶⁰⁾ In relation to Article 1110, Tysoe J. considered that the Tribunal’s Article 1105 analysis had “infected its analysis of Article 1110.”⁽⁶¹⁾ Thus since the decision on Article 1105 had improperly taken account of transparency and this was then used as the basis for deciding the Article 1110 issue, this too was beyond the scope of its jurisdiction for the same reason.

Turning to the question of whether the passing of the Eco-Decree amounted to a measure tantamount to expropriation, Tysoe, J. was not able to conclude that the Tribunal had been patently unreasonable in reaching its decision. However, he did conclude that the definition of expropriation used by the Tribunal was “sufficiently broad to include a legitimate rezoning of property by a municipality or other zoning authority.”⁽⁶²⁾ However, given that the definition used was a matter of law, the resulting conclusion was one with which Tysoe J. was “not entitled to interfere under the International CAA.”⁽⁶³⁾

The importance of the Metalclad decision lies partially in its demonstration of the power of arbitral decision-making vis-à-vis sovereign interests and in the potential of international trade law principles to influence the shape and form of local and national regulatory measures.⁽⁶⁴⁾ However, the subsequent use (or non-use) of the decision in other arbitral disputes serves also to illustrate the unpredictability and relative capriciousness of such decisions and the uncertainty that governments might face in designing and enforcing regulatory measures such that they do not fall foul of investor challenges or attempts to “chill” the regulatory process.

What then, is the legacy of the Metalclad decision on fair and equitable treatment and expropriation, and to what extent has it been followed subsequently by other tribunal panels? This is not such a straightforward question to answer for a number of reasons. Firstly, arbitral tribunals decide each case on the facts before them and the terms of the underlying agreement and hence the common law doctrine of stare decisis is not appropriate. Secondly, there is no requirement for ICSID arbitral awards to be published - and many are not, or are published as excerpts only.⁽⁶⁵⁾ Thus it is not possible in many cases to find out whether, or the extent to which, the Metalclad interpretation of “tantamount to expropriation” was relied upon. Thirdly, different investment agreements between states take different approaches to fundamental state-investor relations such as most-favoured nation treatment, fair and equitable treatment, expropriation and discrimination. Hence references to previous awards based on NAFTA will, at best, have persuasive authority, but may have none at all.

It is worth noting, however, that not long after the Metalclad decision was issued, the NAFTA rules were altered in respect of the meaning of fair and equitable treatment. The NAFTA Free Trade Commission adopted, on July 31, 2001, a Note of Interpretation entitled “Minimum Standard of Treatment in Accordance with International Law.”⁽⁶⁶⁾ It states, inter alia, that the interpretation of “ … the concepts of ‘fair and equitable treatment’ and ‘full protection and security’ do not require treatment in addition to or beyond that which is required by the customary international law minimum standard of treatment of aliens.” This change may or may not have been passed in response to the Metalclad decision but at any rate should have made such a broad interpretation less likely in the future, at least where NAFTA-based arbitrations are concerned.

And yet, barely three years after Metalclad arbitration, Mexico found itself back at ICSID facing similar claims of lack of fair and equitable treatment and expropriation relating to another hazardous waste landfill site - this time far to the north-west in the Municipality of Hermosillo, in the State of Sonora in Mexico.⁽⁶⁷⁾ This time the BIT between Spain and Mexico was the governing instrument and again it was dealt with under the ICSID Additional Facility rules. Notwithstanding the earlier criticism of Metalclad, the tribunal applied a similar analysis to that employed in Metalclad and reached essentially the same conclusions.⁽⁶⁸⁾

The Metalclad decision arose again in Waste Management Inc, yet another ICSID case involving Mexico.⁽⁶⁹⁾ This case also involved Article 1105(1) and Article 1110 claims against Mexico, again in the context of waste management, and again where local unrest played a part. The U.S. investor claimed that the failure of municipal authorities in Acapulco properly to cede a waste management concession amounted to “measures tantamount to expropriation.” This had occurred in circumstances where local people had taken great exception to the imposition of new waste collection charges and the dismantling of existing systems (poorly regulated though they were) and had refused to cooperate - or pay. This resulted in the entire scheme becoming economically unviable.

The Waste Management Inc Tribunal did refer to the Metalclad decision, and appeared to acquiesce in Justice Tysoe’s characterisation of the interpretation of tantamount to expropriation as “extremely broad.”⁽⁷⁰⁾ In the event, the Waste Management Inc Tribunal did not need to engage with this interpretation as it considered that the waste concession had not been a “regulatory taking” (as in Metalclad) but had failed because it was economically unviable from the start. Thus the Metalclad issue could be evaded.

Expropriation also arose in the case of Lucchetti where a conservation order (Decree 258) was placed on an area including a pasta factory owned by an investor based in Chile.⁽⁷¹⁾ A follow-up decree (Decree 259), aimed at implementing the first decree stated:

Evidently this entirely “closed and … removed” the factory - a clearer example of expropriation would be difficult to imagine. However, given that it was for the purposes of environmental protection (i.e. in the public interest) and appeared nondiscriminatory in the sense that all outlets in that area were similarly affected, it would have been interesting to see how the Tribunal would have approached it. In the event Lucchetti’s claim failed on technical grounds⁽⁷²⁾ and the expropriation issue was never dealt with so we can only speculate on whether the Metalclad approach would have been invoked or not.

In 2012, the Metalclad decision was invoked against Venezuela in Mobil Investment Canada Inc.⁽⁷³⁾ This is referred to not only because of citation of Metalclad but also because one of the arbitrators was Professor Philippe Sands QC, who had previously roundly criticized the Metalclad decision.⁽⁷⁴⁾ Perhaps unsurprisingly therefore, the Mobil tribunal refused to follow the Metalclad lead in relation to NAFTA Article 1105(1) and in a rather terse paragraph pointed out that the Metalclad decision had been partially set aside by the Supreme Court of British Columbia and further that the Tribunal “is not aware of any subsequent decisions which have followed the approach taken by the Metalclad Tribunal.”⁽⁷⁵⁾ It seems as if the fair and equitable treatment part of the Metalclad decision at least was by now considered unsound.

However, more recently still, the Metalclad decision was invoked in Gold Reserve Inc.⁽⁷⁶⁾This is one of many cases⁽⁷⁷⁾ brought against Venezuela in the wake of the Chávez administration’s endeavours to take back the country from foreign influence in the 2000s until his death in 2013.⁽⁷⁸⁾ The case involves the alleged failure of the Venezuelan authorities to honour gold mining concessions held by a Canadian parent company. It was not brought under the terms of NAFTA but instead under the 1998 BIT between the Government of Canada and the Government of Venezuela.

In a long and often difficult appraisal,⁽⁷⁹⁾ the Tribunal eventually turns to the question of fair and equitable treatment⁽⁸⁰⁾ and invokes the “transparency” analysis used in Metalclad.⁽⁸¹⁾ The Metalclad approach was approved without any reservation and certainly no mention of Justice Tysoe’s unfavourable review in the Supreme Court of British Columbia, nor the adjustment to the NAFTA interpretation of the fair and equitable treatment which occurred after the Metalclad decision, nor the comments of the panel members in Mobil & Murphy. This decision in particular demonstrates the uncertainty and volatility that can pervade arbitral decisions.

The Gold Reserve Tribunal also discussed the question of expropriation, and again invoked the Metalclad decision,⁽⁸²⁾ once again without any reservations expressed about the “broadness” of that interpretation. The tribunal went on to award damages to Gold Reserve Inc. in the amount of US$713,032,000.⁽⁸³⁾

Venezuela was also on the receiving end of another huge expropriation award at the end of 2014 in Venezuela Holdings B.V⁽⁸⁴⁾ Here a group of Dutch parent company investors were awarded in excess of US$1.6b as a result of a “lawful” expropriation by the Venezuelan government which was held not to have been adequately compensated in line with the terms of the Netherlands-Venezuela BIT of 1991.⁽⁸⁵⁾ This particular award makes no mention at all of the Metalcladdecision.

All these decisions were the result of disputes which went all the way to full assessment of the merits of the parties’ positions and at least each party’s position was fully pleaded. However, there are instances where the threat of ISDS proceedings appear, at least on the face of it, to have resulted in a “chilling” of regulatory measures without the benefit of detailed argument. The use of threatened proceedings to influence the exercise of regulatory sovereignty has caused even more disquiet, not least because it is difficult, if not impossible, to assess the extent of its influence.

One of the starkest examples of this “use” of ISDS is the 1998 Ethylcorpdecision.⁽⁸⁶⁾ Here Ethylcorp, a group of U.S. investors incorporated in Virginia, objected to the decision of the Canadian government to ban the import of methylcyclopentadienyl manganese tricarbonyl (MMT), a petroleum additive, on the grounds of public health protection.⁽⁸⁷⁾ This restriction was brought about by the passage of the Manganese-based Fuel Additives Act of 1997.⁽⁸⁸⁾ The Act proscribed the interstate trade in, or import of, MMT for commercial purposes unless in accordance with an authorisation. However section 5 of the Act specifically forbids the authorisation of MMT for the purposes of addition to petroleum. The effect of the Act, in the view of Ethylcorp, was to deprive it of its business, since their MMT was blended into more that 95% by volume of the petroleum sold in Canada.⁽⁸⁹⁾ Of course Ethylcorp were not alone in this in the sense that the Act banned the import and use of MMT as a petrol additive for every company operating in Canada. Hence there was no element of discriminatory practice. Nonetheless, Ethylcorp brought proceedings under the UNCITRAL rules alleging a breach of chapter 11 of the NAFTA, including an assertion that the passage of the Act contravened NAFTA rules on national treatment (Article 1102), performance requirements (Article 1106) and expropriation (Article 1110). In the decision on jurisdiction, delivered on June 24, 1998, the tribunal rejected the arguments of the Canadian government that (i) Ethylcorp had failed to observe procedural requirements related to timeliness; and, (ii) that the Act and statements relating to it, were not “measures”, nor were they measures related to “investments” or “investors.” ⁽⁹⁰⁾ The tribunal went on to claim jurisdiction.

In the event, however, the merits of the claim were never addressed as the Canadian Government instead reached an out of court settlement with Ethylcorp, doubtless owing to lack of confidence that it could win on the merits in another arbitral forum. Though this response cannot be thought of as a classic regulatory chilling in that the provision was promulgated and remained in place after the challenge rather than preventing its genesis in the first place, it does illustrate that even where a public health measure is put in place for entirely bona fide public health reasons supported by sound science, the investor can nevertheless use the threat of arbitration to protect its investment. Further illustration of this tactic, albeit in a different context, can be observed in the campaign by Phillip Morris against plain packaging regulations in Uruguay and Australia.⁽⁹¹⁾

Thus there is plenty of precedent for the use of threats of arbitration to control regulation deemed to be economically threatening to an overseas investor. This gives rise to the question of whether regulatory measures adopted to protect public health or the environment in the European Union might potentially be open to challenge under the putative investment chapter of the TTIP by investors based in the U.S. Self-evidently the answer is yes by virtue of the mere existence of the investment chapter. However, there is an added nuance to challenges to environmental measures in the European Union since so many of them are based on the precautionary principle. It is contended that the different perceptions of this principle on both sides of the Atlantic presents an avenue of attack on regulatory measures in the European Union. Challenges to regulatory measures may be undertaken on the basis that the precautionary principle may be a disproportionate and non-scientific response which results in regulation based on fear and irrationality not on sound science.⁽⁹²⁾ It is to this debate that we now turn.

The precautionary principle is, in essence, the idea that lack of scientific uncertainty about a potentially harmful phenomenon, product or process should not, of itself, be a barrier to the taking of precautionary measures. The origins of the principle are somewhat obscure but it is generally traced back to Swedish and German environmental policies (Vorsorgeprinzip) of the 1970s. It has also been said to have been highly influential at the London Dumping Conference of the International Maritime Organisation in 1972, which gave rise to the Convention on the Prevention of Marine Pollution by Dumping of Wastes and Other Matter (MARPOL) which marked the beginning of the precautionary approach to the disposal of waste at sea.⁽⁹³⁾ Certainly by 1992 the precautionary approach was a significant element of international environmental policy and was prominent in the Declaration of the United Nations Conference on the Environment and Development, 1992 (Rio Conference):

By 1998, the precautionary principle had become part of the legal order of the European Communities by virtue of the Treaty of Amsterdam 1997, which inserted a new Article 131(2) into the Treaty on European Union; essentially the same text is now incorporated in Article 191 of the Treaty on the Functioning of the European Union which states that:

Unfortunately the principle itself is not further defined in the Treaty. Some clarification is contained within the official “Communication from the Commission on the precautionary principle” of 2000⁽⁹⁶⁾ which indicates that the principle should be engaged:

However, the Communication also points out that

Thus the operation of the precautionary principle is envisaged as part of a wider risk assessment strategy where there is evidence that the risk exists, but the magnitude of that risk cannot be quantified with certainty. It is perhaps unfortunate that the term “sufficient certainty” in paragraph three of this extract receives no further elaboration particularly as it is in the response to the lack of sufficient certainty by policy and lawmakers that the controversy surrounding the precautionary principle inheres. The literature devoted to the general nature of the precautionary principle, the disagreement about its use, role and implementation across the Atlantic, and its role within trade law is voluminous⁽⁹⁹⁾ and the difference in transatlantic approaches to the principle is one of the sticking points in the negotiation of the TTIP itself.⁽¹⁰⁰⁾

On the whole law- and policy-makers in the United States take a negative view of the precautionary principle⁽¹⁰¹⁾ and this view has affected trade relations between the United States and the European Union in the past where fundamental differences have arisen in the response to unquantified risk. The United States (often joined by Canada, but also frequently supported by Australia, Brazil and New Zealand) has tended to prefer a cost-benefit approach to risk, taking the view that this is a more rational and scientific approach to risk management. The European Union, on the other hand takes a more precautionary approach based, as its foundational treaty requires, on the precautionary principle - a concept seen as irrational, post-enlightenment and risk averse by many in North America.⁽¹⁰²⁾ In crude terms these two approaches may be thought of as the innocent-until-proven-guilty and guilty-until-proven-innocent models respectively. This difference in views has underpinned the protracted trade disputes between North America and Europe over the licensing of genetically modified organisms⁽¹⁰³⁾ and the use of growth hormones in beef.⁽¹⁰⁴⁾ However, despite the commonly asserted difference in approaches between North America and Europe, which has to some extent been perpetuated above, it is necessary to bear in mind that the situation is far more complex than a simple divergence of opinion across the Atlantic. Bergkamp and Smith’s comprehensive analysis of the approach to precaution between the United States and the European Union is at pains to point out that both jurisdictions have embraced, and continue to embrace, precautionary regulation,⁽¹⁰⁵⁾ that the extent of the precaution taken is highly dependent on the issue under discussion,⁽¹⁰⁶⁾ and that each jurisdiction is at different stages of development in terms of its sophistication. However, they conclude that Europe, at least in 2013, remained the less sophisticated jurisdiction:

Of course a European may justifiably express some scepticism of this conclusion in the face of U.S. attitudes to the uncertainty relating, for example, to climate change among certain influential members of the U.S. polity who ignore the science altogether in favour of leftist conspiracy theory.⁽¹⁰⁸⁾

Whatever the realities of implementation, the difficulties with the precautionary principle as a concept are not scientific in origin. There is general consensus and mutual respect between U.S. and European scientists in relation to the scientific method itself, and hence in relation to the degree of uncertainty associated with a particular phenomenon, product or process. Disagreement emerges in relation to the most appropriate response to the uncertainty. A cost-benefit analysis approach places a higher premium on scientific rationality and regulation should only be imposed where a “significant risk” exists. The preference for this paradigm in the United States has been traced to Supreme Court precedent in the 1980s which “helped to ensure America’s economic and technological advancement and competitiveness during the past several decades”,⁽¹⁰⁹⁾ and to the subsequent influence of the Reagan administration (ably led by the Speaker of the House Newt Gingrich) whose chief concern was reducing regulatory burden as part of the “Contract with America” and hence embraced cost-benefit analysis because it tends to generate a higher threshold for the imposition of regulation.⁽¹¹⁰⁾ The precautionary paradigm embraced in Europe, on the other hand, is said to have its origins in postenlightenment European philosophies, particularly those of France and Germany⁽¹¹¹⁾and to embody ethical and equitable principles derived from Western notions of morality and deontology.⁽¹¹²⁾

Hence disagreement over approaches to precaution are disagreements over doctrinal interpretation rather than over science or law as such. It is this that makes regulations based on the precautionary principle peculiarly vulnerable to attack in arbitral proceedings, since arbitrators are asked to rule, not so much on the quality of the science underpinning a regulation or even the act of regulation itself (though this is sometimes challenged), but rather the rationality of the regulatory response to that science. An investor might invite a tribunal to find that a regulatory response is not based on sound science, or is too cautious to be justifiable given the lack of sufficient scientific certainty. As such, it may amount to a lack of fair and equitable treatment, or a measure tantamount to expropriation. Challenges on these grounds under the proposed TTIP are plausible, and the next section uses a hypothetical challenge based on a current EU restriction to analyse the possibilities of success.

An illustrative example of the operation of the precautionary principle in the European Union is to be found in regulations that implement the current partial restriction of the use of a class of chemicals known as the neonicotinoid insecticides (hereinafter neonics). Developed in the 1980s, principally by the chemical companies Bayer, Syngenta and Sumitomo Chemical, the neonics are a class of systemic insecticides (taken up by all parts of the plants to which they are applied), which attack the central nervous systems of pest insects which feed on the plants with fatal effect, but with none of the environmental persistence or high mammalian toxicity associated with the previous pyrethrin, pyrethroid, organophosphate and organochlorine alternatives.⁽¹¹³⁾ The background to the imposition of the current restrictions on the use of these chemicals has been fully described elsewhere⁽¹¹⁴⁾ and only a brief summary is necessary here.

Initially, Council Directive 91/444/EEC⁽¹¹⁵⁾ (concerning the placing of plant protection products on the market) made provision for substances to be added to Annex 1 of that directive which listed “Active Substances Authorized for Incorporation in Plant Protection Products.” Subsequent secondary legislation added a number of neonics to this Annex thereby permitting their use in plant products in the European Union.⁽¹¹⁶⁾ However, following accidental releases of these chemicals which resulted in the deaths of bee colonies, the use of the neonics clothianidin, thiamethoxam, imidacloprid and fipronil were subjected to additional risk assessment requirements.⁽¹¹⁷⁾ Subsequent regulatory amendments placed still further restrictions, specifically on the use of three of the more commonly used neonics, clothianidin, thiamethoxam and imidacloprid, as foliar treatments, as soil additives and in seed dressing.⁽¹¹⁸⁾

These restrictions were justified on the basis of a report, commissioned by the European Commission,⁽¹¹⁹⁾ by the European Food Safety Authority which reviewed the evidence on the effect of neonics on bee colonies.⁽¹²⁰⁾ The evidence suggested that although the normal use of these pesticides was not lethal to bees, there was some evidence that neonics residues have the effect of interfering with the bees’ navigation systems, thereby disorientating them and preventing them from returning to their colonies or from indicating the sources of food to the remainder of the colony even were they able to return. The European Food Safety Authority recommended that further evidence was required in order to establish the magnitude of the risk.⁽¹²¹⁾ On this basis the Commission, taking the precautionary approach required by Article 191 of the Treaty on the Functioning of the European Union, and explicitly referred to in EU Regulation (EC) No 1107/2009, recommended restrictions on the use of neonics.

These recommendations were viewed with some equivocation in the member states of the European Union⁽¹²²⁾ and the arable farming community has protested that the restrictions on the application of neonics are significantly adversely affecting their business.⁽¹²³⁾ In 2015 the National Farmers’ Union successfully applied to the UK Government for an emergency lifting of the ban in three eastern counties of England, a move which was unsuccessfully challenged by Friends of the Earth.⁽¹²⁴⁾Perhaps unsurprisingly the chemical industry has viewed these restrictions as being an overreaction based on flawed field studies.⁽¹²⁵⁾

The central regulation which introduced these restrictions, EU Regulation (EC) No. 1107/2009, is explicitly based on the precautionary principle.⁽¹²⁶⁾ Article 1(4) states:

The question is whether - should the TTIP come into being in anything like its current form - Bayer, Syngenta or Sumitomo could seek to challenge the restriction under the terms of the TTIP draft investment protection section⁽¹²⁷⁾ on the basis that the restrictions represent a failure to accord fair and equitable treatment or that the restrictions represent an expropriation of property.

As this paper was being completed, the TTIP negotiations entered their 15^th round⁽¹⁸²⁾against the backdrop of recent demonstrations in Germany against both TTIP and CETA.⁽¹⁸³⁾ The EU Commission will no doubt be even more conscious of the need to reassure European citizens that TTIP will not undermine national sovereignty - a message that has assumed even greater importance since the citizens of the United Kingdom voted in favour of “Brexit” in a referendum in July 2016. The “Leave” campaign in Britain made much of the need to “regain sovereignty” from the European Union,⁽¹⁸⁴⁾ so that questions of national sovereignty are now, more than ever at the forefront of politicians’ minds in the EU.

Where environmental sovereignty is concerned the text of the TTIP includes, at first sight, quite strongly worded provisions to permit nation states to maintain and enhance environmental protection. However, very similar provisions to these appear in other investment and trade agreements but they have not necessarily prevented investors from seeking compensation for the loss of their investment (or the value of it) as a result of local, regional or national measures designed to protect the environment or public health. Such claims have invariably been based on the fair and equitable treatment principle and/or on the basis that the measures represent measures tantamount to (indirect) exropriation. It is also the case that such claims do not necessarily need to reach the merits stage in order to bring about the desired effect as illustrated in the Ethyl Corp decision discussed above.

The current text of the TTIP does not suggest that it is likely to be significantly more resilient to such claims than many of its forbears. In fact, given that so much of the regulation in environmental matters and public health in the European Union is based on a version of the precautionary principle which is far more “precautionary” than envisaged in principle 15 of the Rio Declaration, investor claims which seek to attack its rationality are probably more likely.

Of course the final version of the TTIP may contain explicit protection for the precautionary principle, but, if this merely repeats the provision in CETA, then it is unlikely to prevent claims based on equivocal scientific evidence such as the restrictions currently in place for neonicotinoid pesticides. Moreover, the provision of a bespoke tribunal and appeal system within TTIP may not necessarily make claims founded on the irrationality of regulatory responses to the precautionary principle any less plausible or any more likely to fail.

Investment protection agreements have the potential to enhance environmental protection if they encourage greater transparency of regulation by the host state so that the investor is certain about the regulatory environment into which they are entering and know the risks. However, the approach of investment tribunal panels themselves needs to change in order to take a broader view of purposes of arbitration in an era of climate change, loss of biodiversity and the need to preserve ecosystem services. It can no longer be appropriate merely to consider the private property rights of investors as the primary consideration. The public interest in environmental quality and the need to adapt to environmental pressures by sovereign authorities must play a greater part. In the long run this is in the interests of investors also. However, it is the elected sovereign authorities who must be permitted the final word in how this is brought about rather than corporate entities who primary purpose is the generation of profit.