The Danger of the Interpretation of Facts: Legal Uncertainty in the Spanish Saga Cases

Abstract

Enhancing legal certainty is one of the main values that are sought in the investor–state dispute settlement system. The importance of legal certainty is strengthened in the case of renewable energy investments, which are in the global public interest, long-term and capital-intensive up-front. The first part of the paper presents the importance of legal certainty in investment arbitration in general, its limits and its importance in the context of the green energy transition. In addition, it addresses the special features of renewable energy investments. The second part of the paper analyses from the perspective of legal certainty the Spanish renewable energy cases initiated under the Energy Charter Treaty (ECT), which deal with similar factual and legal issues. In this respect, the paper presents the varying weight tribunals gave to the important facts that led them further to conclude whether Spain breached the fair and equitable treatment standard, and if so, whether the investor was entitled to full compensation or a reasonable rate of return. In addition, it presents different approaches to perceiving the stability provision of Article 10 (1) of the ECT. The paper concludes that it remains uncertain to what extent RE investors will be protected under the ECT’s stability condition in the case of fundamental or small-scale changes. Although one group of arbitrators may argue that the fundamental change triggers per se a breach of a stability condition, others may argue that for the breach to be established, the host state’s measures must be arbitrary, unreasonable or discriminatory. Moreover, the threat to legal certainty might not only be the vague provisions of the ECT but also the significant discretion tribunals have towards the interpretation of facts, leading to different outcomes. Indeed, it is at the discretion of arbitrators to consider whether the timing of investment, presence of evidence indicating possible regulatory changes, and the reasonable rate of return prescribed in Spain’s domestic law will be relevant or irrelevant.

1. Introduction

Legal certainty should secure predictability among the relations between people (Mitrović 1996). It requires that those that are subject to the law can predict reliably how the rules that govern their conduct will be interpreted and applied (Tamanaha 2004). It is asserted that the application phase in international law is of pivotal importance in order to know what rights and obligations apply in any given case as it is significantly less systematic and coherent than domestic legal systems (Casanovas 2001).

Different actors express that predictability and consistency are one of the key concerns of investment arbitration, which are needed for promoting legal certainty and, thus, contributing to the system’s legitimacy and credibility (UNCITRAL 2018a; OECD 2012; European Commission 2015). Investment arbitration is the main forum for the resolution of renewable energy (RE) disputes between foreign investors and states (The World Bank and International Energy Charter 2023).1 Foreign direct investment plays an integral role in funding renewable projects globally (ibid., p. 7). RE investment projects are long-term, capital-intensive upfront and reliant on support schemes, which makes them particularly vulnerable to regulatory risks (Balcerzak 2023; Busch et al. 2023; Boute 2009). It is argued that investment arbitration has the potential to protect low-carbon investments against the risks of regulatory changes (Boute 2012). The extent to which investment arbitration contributes to regulatory certainty of low-carbon investments depends on the consistency and certainty of the arbitral process (ibid.; Selivanova 2018). In the past decade, numerous cases were initiated against Spain under the Energy Charter Treaty (ECT) arising out of a series of reforms affecting the renewables sector undertaken by the Spanish Government for the sake of tackling the growing tariff deficit of the electricity system (Spanish saga cases). Due to legal and factual similarities and the growing arbitral practice, Spanish saga cases provide an opportunity to analyse to what extent investment arbitration secures predictability in investor–state relations and as such provides legal certainty.

Against this backdrop, this paper will analyse Spanish saga cases from the perspective of legal certainty. It will analyse to what extent investors and states can predict reliably how the rules that govern their conduct will be interpreted and applied. Accordingly, the paper will not examine the relevance of legal certainty in the context of the protection of legitimate expectations as the element of the fair and equitable treatment (FET) standard (see Henckels 2023a, 2023b). The main focus will be on presenting the varying weight given to distinct facts that ultimately led to different outcomes related to the breach of the FET standard. In addition, different perceptions of the stability condition under Article 10 (1) of the ECT will be presented to demonstrate inconsistencies in the arbitral practice from a broader perspective.

The first part of this paper will present the arguments in favour of legal certainty in investment arbitration as well as its limits more generally. It will then focus on the importance of legal certainty in investment arbitration in the context of the green energy transition, bearing in mind the special features of RE investments. The second part of the paper will analyse Spanish saga cases. After presenting briefly the developments in the Spanish legal framework due to which disputes were initiated, the paper will, first, present inconsistencies related to the weight given to the facts important for determining the breach of the FET standard. In this respect, it will focus on the moment of making the investment, the presence of the Spanish Supreme Court’s judgments as (potential) indicators of possible changes to the legal framework, and the reasonable rate of return prescribed in the Spanish domestic law as a (potential) reason for triggering the reasonable rate of return threshold. Afterwards, it will present inconsistencies related to the weight given to the stability condition under Article 10 (1) of the ECT, discussed normally within the framework of investors’ legitimate expectations, compared to the host state’s right to regulate. The paper will end with a conclusion that, even though tribunals refer to previous decisions dealing with similar factual and legal issues, Spanish saga cases do not provide investors in RE with regulatory certainty on which to base their investment.

2. Legal Certainty in Investment Arbitration

The issue of consistency among and predictability of arbitral decisions and, thus, legal certainty in investment arbitration has been extensively discussed in the literature (Rivkin et al. 2015; Castellarin 2018; Weidemaier 2010; Verburg 2019; Kaufmann-Kohler 2008; Dietz et al. 2019; Subedi 2016; Guillaume 2011). Different actors put forward the inconsistency among arbitral decisions as a concern. UNCITRAL Working Group III has identified concerns that are commonly expressed about the investor–state dispute settlement (ISDS) regime (UNCITRAL 2017a, 2018b, 2018c, 2019a, 2019b, 2019c). Concerns that are related to legal certainty were categorised as ‘arbitral outcomes’. Arbitral outcomes concern issues related to consistency, coherence and predictability as well as limited mechanisms for ensuring the correctness of arbitral decisions. Public consultations conducted in the EU show that all categories of respondents2 agreed that the main argument in favour of an appellate mechanism is that it contributes to more consistency and coherence, and as such to legal certainty (European Commission 2015). Governmental discussions organised under the auspices of the OECD also put forward inconsistency as one of the key issues of the ISDS system (OECD 2012). In addition, in certain arbitral decisions, arbitrators themselves claim they have a duty to seek consistency and contribute to the harmonious development of international investment law (Saipem v. Bangladesh; Burlington v. Ecuador; The PV Investors v. Spain; ADC v. Hungary).3

One of the typical examples used to demonstrate the issue of inconsistency in investment arbitration is represented in the cases against Argentina during its 2001 economic crisis, where disagreements among the arbitrators focused on whether Argentina was not liable for the breach of several obligations under the Argentina–US BIT based on the necessity defence under customary international law and the BIT’s emergency clauses (Guillaume 2011; Ten Cate 2013; Subedi 2016). Some tribunals found that Argentina successfully established emergency defence under the BIT, while others concluded that it did not meet the standards for either defence (UNCITRAL 2017b; Enron v. Argentina; Continental v. Argentina).4 Another example seems to be the Spanish saga cases that will be discussed below.

It is not surprising that one of the major issues concerning the current ISDS mechanism is inconsistency in its jurisprudence, considering the decentralised nature of investment arbitration, the vague principles they are tasked to interpret, and the absence of the formal doctrine of precedent and appellate mechanism (Verburg 2019; UNCTAD 2012; UNCITRAL 2018a). The doctrine of precedent does not exist in international law (Guillaume 2011; Prosecutor v. Zlatko Aleksovski; SGS v. Philippines; El Paso v. Argentina).5 Strictly speaking, arbitral tribunals are at full liberty to deviate from previous awards (separate opinion of Thomas W Wälde on Thunderbird v. Mexico). Nevertheless, it is argued that decisionmakers have an obligation to strive for consistency and predictability and, hence, follow precedents, and the extent to which they are obliged to do so depends on the stage of development of the law (Fuller 1969; Kaufmann-Kohler 2007, 2008). Judicial consistency as far as possible in the area of international investment law is regarded as desirable (Total v. Argentina).6 Since most investment-related disputes are resolved by arbitration, arbitrators bear the primary responsibility for granting certainty and predictability to investments (Weidemaier 2010).

The credibility of the entire dispute settlement system depends on consistency because the system that produces unpredictable results will lose the confidence of the users (Kaufmann-Kohler 2007; Knahr and Reinisch 2007; UNCITRAL 2018d). Furthermore, tribunals departing from prior decisions without justifiable reasons undermine the legitimacy of a dispute settlement system (Bhala 2001). Inconsistency is regarded as one of the main legitimacy gaps of the ISDS system (Dietz et al. 2019). In particular, it is discussed ‘that the existing system of ICSID precedent may have resulted from arbitrators’ efforts to attain legitimacy in the eyes of a wide range of external actors’ (Weidemaier 2010, p. 1954). Indeed, conflicting decisions are regarded as a legitimatory problem (Bogdandy and Venzke 2013).

2.1. The Limits of Legal Certainty

Nonetheless, one should not ignore that a degree of inconsistency is inherent in any legal system and is not intolerable (Kaufmann-Kohler 2008; Mitrović 1996). Even if it were possible, striving for absolute certainty would impoverish international law of its argumentative, procedural and rational side (Papić 2021). Courts can depart from previous decisions when it appears right to do so even in the case of the formal doctrine of precedent (Prosecutor v. Zlatko Aleksovski). During UNCITRAL Working Group III discussions, it was agreed that consistency should not be achieved to the detriment of decisions’ correctness, i.e., predictability and correctness should be the objective rather than uniformity (UNCITRAL 2018a). Further, Schultz calls for caution when pursuing consistency in investment arbitration, as consistency has relative value, meaning that it can do both harm and good, depending on what is being made consistent (Schultz 2014). He claims that consistency should be sought only when it contributes to a desirable regime (ibid., p. 316). Thus, consistency in applying rules is not per se desirable (ibid., p. 316). For example, it was stressed that to improve the stability of the investment climate for low-carbon projects, it would suffice to clarify certain arbitral tribunals’ interpretation of the FET standard already providing significant investment protection (Boute 2012). From this perspective, the desirable regime would be the one that offers greater protection to foreign investors investing in the sector of public interest, and as such would be the one worthy of consistency.

2.2. Legal Certainty and the Green Energy Transition

International investment law could support the objectives of the international community, such as environmental protection and the related protection of human rights, if it was harmonised and greater consistency was achieved (Subedi 2016). On the one hand, the lack of consistency in arbitral practice makes states uncertain about what (climate-related) measures they may adopt without breaching investors’ rights. Such uncertainty may lead to the adoption of measures that breach investment protection provisions or result in states’ regulatory chilling (UNCITRAL 2018d; dissenting opinion of Philippe Sands on RENERGY v. Spain; Tienhaara and Downie 2018). On the other hand, it is argued that the lack of predictability is an issue for (RE) investors as it could constitute a risk factor and may as such hinder investments (UNCITRAL 2018d). Indeed, given the costs involved, (RE) investors value a dispute settlement system that provides predictability (ibid., para. 37) Thus, it is asserted that interpretations providing adequate protection to low-carbon investments should be officially endorsed (Boute 2012).

It could be useful to emphasise the particular features of RE investments before proceeding to the analysis of the arbitral practice concerning the sector in order to understand the importance of providing greater predictability through arbitral decisions to such investments. International investment agreements, including the ECT, are technology-neutral, meaning that investment protection provisions apply to both fossil fuel and renewables investors.7 Nevertheless, the particular significance of providing adequate RE investments’ protection could be approached by assessing, inter alia, the environmental benefits such investments provide to the host state and international community. One of the common features of RE investments is that they have a significant positive impact on the environment (The PV Investors v. Spain; Balcerzak 2023). It is emphasised that potential claims related to low-carbon investments or directed against domestic climate policies should be assessed in the context of the political priority at the international level for drastically reducing GHG emissions (Boute 2009). Preservation of the environment and the related fight against climate change are in the global public interest, as they are essential not only to states but to all individuals of present and future generations (Casanovas 2015; Rodrigo Hernández 2018; Juste 2015). Moreover, environmental protection and adverse effects of climate change have an impact on the enjoyment of human rights, whose protection is by itself in the global public interest (Casanovas 2015). Bodansky stressed that the reduction in GHG emissions can be considered a global public good (Bodansky 2012, 2016). Accordingly, RE plants, in addition to fulfilling the private purpose of electricity supply, provide global public benefit by taking the place of coal-fired or other conventional power plants that would have emitted GHG emissions (Bodansky 2012). In this respect, the tribunal in The PV Investors v. Spain found the need to find the right balance between the protection of investors who invested substantial capital in a sector that provides environmental benefits to the host state, on the one hand, and the host state’s right to regulate in a proportionate, reasonable and non-arbitrary manner, on the other.

Other particular features of RE investments should also be taken into account when discussing the adequate protection required for such investments. These investment projects are long-term, capital-intensive upfront and have low working capital (SolES v. Spain; Green Power Partners v. Spain; The World Bank and International Energy Charter 2023).8 They require huge amounts of capital before beginning operations, producing electricity and generating income (Balcerzak 2023; Hirth and Steckel 2016). Due to high upfront investment costs, in 2019, foreign companies sponsored almost 40 per cent of all RE generation projects (The World Bank and International Energy Charter 2023). In the case of developing countries and transition economies, foreign companies sponsored more than 70 per cent of the respective projects (ibid.). This remains the case even though, between 2000 and 2021, the cost of green technology has decreased sharply due to technological developments (ibid., pp. 10, 16–18). RE investments are dependent on governmental support schemes introduced during the first decade of this century9 for the sake of allowing ‘the technologies to be developed in the hope that over time the costs associated therewith will decline, thus making RE technologies more competitive’ (Antin v. Spain, para. 540; UNCTAD 2010; European Commission n.d.; Directive 2001/77/EC).10 The purpose of such incentives is to reduce financial and regulatory risks by guaranteeing RE producers benefits over a long fixed period of time. Regulatory risks concern the probability that the rules existing at the time of investing will be changed (Boute 2009, 2012). Regulatory risks may be more relevant for certain types of foreign direct investments, such as renewables, than others (The World Bank and International Energy Charter 2023). Indeed, in over 100 identified RE investor–state disputes, the most common political risk raised in the proceedings is adverse regulatory changes (ibid., p. 12).

3. Legal (un)Certainty in Action: Spanish Saga Cases

The cases that will be discussed in this section concern measures adopted by Spain ‘in an attempt to tackle the overrun cost derived from the excessive RES support incentives (mostly related to solar PV promotion), and therefore the growing tariff deficit of the electricity system’ (Coronas et al. 2022, p. 3). These measures were adopted in the public interest of securing sustainability and reliability of the electricity system and protecting the state budget, in the context of the global financial crisis faced after 2008 (dissenting opinion of Philippe Sands on RENERGY v. Spain). Due to the impact of the undertaken measures on investments, RE investors claimed breaches under, inter alia, Article 10 (1) of the ECT, which prescribes the FET standard. The most important function of the FET standard is the protection of investors’ reasonable and legitimate expectations (Electrabel v. Hungary).11

The purpose of this section is to analyse the extent to which legal certainty is respected, i.e., to what extent investment arbitration secures predictability in investor–state relations. Accordingly, it will first present relevant developments in the Spanish legal framework due to which disputes were initiated. Afterwards, it will present inconsistencies in the Spanish saga cases by addressing the interpretation of facts that had a varying impact on the determination of investors’ legitimate expectations. In addition, the paper will present different approaches towards the interpretation of the stability condition prescribed in Article 10 (1) of the ECT.

3.1. Developments in the Spanish Legal Framework12

Countries have adopted incentive schemes during the 2000s to stimulate investments in the renewables sector in order to meet their UN climate regime commitments and, in the case of EU member states, EU regulations encouraging renewables (Antin v. Spain; Silver Ridge v. Italy).13 Indeed, climate change commitments require multi-level actions (Casanovas and Rodrigo 2022). In this respect, Spain has adopted the General Electricity Law 54/1997 (Electricity Law)14 that distinguished the ‘Ordinary Regime’, which was applicable to conventional power plants, from the ‘Special Regime’, applicable to electricity production facilities of less than 50MW that generated electricity from non-consumable RE sources (Antin v. Spain). Under the Ordinary Regime, remuneration depended on the wholesale market price of electricity, while under the Special Regime, investors were to benefit from the premium that was paid over the wholesale market price (Masdar v. Spain; Balcerzak 2023).15 Article 30 (4) of the law guaranteed RE investors a premium complementing the normal market price in order for them to receive a reasonable rate of return (Green Power Partners v. Spain). Following the adoption of the Electricity Law’s first Royal Decree16, the RE promotion period began in Spain that lasted until 2008 (Coronas et al. 2022).

The Electricity Law was implemented through the Royal Decree (RD) 2818/1998, the RD 436/2004, and later the RD 661/2007.17 The RD 661/2007 is ‘the bedrock on which the Claimants [of Spanish Saga cases] anchor their claims’ as it was the legislation based on which investors decided to invest (Balcerzak 2023, p. 33; The PV Investors v. Spain para. 189). Article 24 (1) of the RD 661/2007 prescribed that Special Regime installations could benefit from one of the two remuneration schemes. The first option was to receive for the sale of electricity a regulated tariff expressed in EUR cents per kilowatt hour (the FiT), while the second was to sell electricity directly to the market for the market price or a price freely negotiated, supplemented by a premium expressed in EUR cents per kilowatt hour (feed-in premium). The decree also provided for the priority of the dispatch of electricity generated through RE. Of importance for further discussions, its Article 44 (3) prescribed that the tariffs, premiums and incentives and upper and lower limits defined in the decree will be reviewed in 2010 and every four years thereafter based on the specific criteria, always guaranteeing reasonable rates of return with reference to the cost of money in the capital market. Further, the following was provided by Article 44 (3): ‘the revisions to the regulated tariff and the upper and lower limits indicated in this paragraph shall not affect facilities for which the deed of commissioning shall have been granted prior to 1 January of the second year following the year in which the revision shall have been performed’.

Moreover, in regard to solar PV, RD 1578/200818 has replaced RD 661/2007 by establishing lower FiTs (by approximately 30 per cent) and annual capacity quotas for PV facilities registered after 29 September 2008 (Green Power Partners v. Spain; Foresight and others v. Spain)19. With RD 1578/2008 started the containment period of the Spanish solar electricity sector, which further extended to other RE power sectors (Coronas et al. 2022).

Disputed measures in the Spanish saga cases can be divided into those that modified the regime (the 2010 measures, Law 15/2012, Royal Decree Law (RDL) 2/2013), and those that completely repealed the initial regime (‘New Regulatory Regime’) (Foresight and others v. Spain; SolEs v. Spain). Some of the measures that modified the RD 661/2007 (in the case of PV installations also RD 1578/2008) are as follows:

• RD 1565/2010 limited the period during which the FiT prescribed in RD 661/2007 was payable to 25 years (later was extended to 28 and 30 years);20
• RDL 14/2010 imposed an annual cap on the number of hours during which PV installations could sell electricity under the FiT;
• Law 15/201221 imposed a seven per cent tax on electric energy production; as well as a new levy on hydropower producers; eliminated the FiT scheme under the Special Regime for electricity produced using non-renewable back-up fuel (Hydro Energy v. Spain; InfraRed and others v. Spain)22;
• The RDL 2/201323 modified the inflation index used to update FITs.

Some of the measures that have repealed the existing legal framework, i.e., RD 661/2007 (in the case of PV installations also RD 1578/2008) are listed here:

• RDL 9/201324 was adopted on an urgent basis due to consumer protection in the context of the economic crisis and the guarantee of economic sustainability of the electricity system. It amended Law 54/1997 and repealed RD 661/2007 and RD 1578/2008. It eliminated the regime of fixed tariffs and premiums both for new and existing installations by substituting it with ‘a system providing for ‘specific remuneration’ based on ‘standard’ costs per unit of installed power, plus standard amounts for operating costs’ (Hydro Energy v. Spain, para. 145; InfraRed and others v. Spain; Balcerzak 2023; dissenting opinion of Kaj Hobér on Stadtwerke v. Spain);
• Law 24/201325 confirmed and enshrined the provisions of the new remuneration regime (‘Specific Regime’) enacted by RDL 9/2013 (InfraRed and others v. Spain);
• RD 413/2014 and Ministerial Order IET/1045/2014 implemented the new Law 24/2013 by, inter alia, prescribing the methodology for calculating the specific remuneration scheme available to RE producers, fixing the lifetime for which plants could receive specific remunerations, and the maximum number of operating hours for which specific remuneration would be received (InfraRed and others v. Spain).

3.2. Inconsistencies Regarding the Weight Given to the Facts When Determining the Breach of Investors’ Legitimate Expectations

Tribunals regard facts as central in resolving disputes (Perrone 2021). Emphasis on the facts, whose interpretation is not that different from the interpretation of the law, ‘opens up a space of arbitral subjectivity, allowing arbitrators to portray the relationship between foreign investors and states in different ways’ (ibid., p. 44). Disputed measures in the Spanish saga cases are almost identical (Balcerzak 2023). Nevertheless, some arbitrators in the Spanish saga cases have outlined the importance of particular facts when deciding on a breach of investors’ legitimate expectations. For example, the moment of making the investment, the presence of Spanish Supreme Court judgments as (potential) indicators of possible changes to the legal framework, and the reasonable rate of return prescribed in the Spanish domestic law were facts that some tribunals used to justify the reaching of different conclusions in particular cases. Nevertheless, it seems that these same facts can be disregarded. Accordingly, the extent to which these facts will have an impact on the outcome of a case (if any) is uncertain, as although in some cases these facts had an important impact on the outcome of the case, in others, this particularity had a rather irrelevant effect on tribunal’s decision.

In the Isolux case, which concerned PV investments, the moment of making investments was decisive for determining that there was no breach of the FET standard, as the investor made its investments after Spain had already modified the regulatory framework and was subject to several studies that made its revision inevitable.26 The investor claimed that Spain violated its legitimate expectations by modifying the regulatory framework through Law 15/2012 and RDL 2/2013, and abolishing it through RDL 9/2013.27 The tribunal in Isolux found 29 October 2012 to be the relevant date for assessing whether the existing regulatory framework could have generated the investor’s legitimate expectations, as it was the date when the claimant acquired 65,434,220 registered shares (equivalent to 58.8632 per cent of the share capital) from T-Solar.28 The tribunal based its reasoning that the existing regulatory framework could not have generated a legitimate expectation for the claimant that it would not be modified on the fact that the regulatory framework had already been modified29, and the Spanish Supreme Court judgments verified the legality of such amendments.30 All these facts were seen to demonstrate the unstable nature of a regulatory framework that the government had the power and duty to adapt to changing circumstances, within the framework of the Electricity Law.31 Moreover, the tribunal found that the existence of the Special Regime throughout the life of the plants could not be an expectation per se, regardless of its content.32 Accordingly, the finding was that at the time of making the investments (October 2012), claimants knew or should have known that the system was going to be modified (see footnote 32 above). In this respect, the tribunal also referred to the report of the Spanish Energy Commission from March 2012 that stressed the economic imbalance of the electricity system and the necessary revision of the efficiency incentives of the existing regulation (see footnote 32 above). Similarly, the tribunals in certain other cases emphasised the importance of the date of investment when assessing investors’ legitimate expectations, such as Novenergia II and Cube.33 In Novenergia II34, the tribunal confirmed that the regulatory changes that occurred prior to October 2012 ‘certainly must have been an indication to the investor in Isolux that significant changes were being made to the Special Regime as set out in RD 661/2007.’ (Novenergia II v. Spain, para. 686).

However, although it is confirmed in the literature and arbitral practice that ‘the timing of the respective investments has been a critical distinguishing factor in ascertaining the legitimate expectations of investors’, in the case of Watkins35, the exact timing of the investment was not even determined (concurring and dissenting opinion of Charles N Brower on The PV Investors v. Spain, para. 14; Selivanova 2018; Noihac 2020). The disputed measures in the Watkins case are the same as those in the Isolux case, but concerned the wind sector. Although the tribunal did state that the claimant’s expectations must be assessed at the time the investment was made, it did not determine whether it was made in August 2011, as claimed by the investor, or in May 2012, as claimed by Spain, nor did it explain how the date of investment could change the intensity of legitimate expectations (dissenting opinion of Helene Ruiz Fabri on Watkins v. Spain).36 The Annulment Committee on the Watkins case concluded that the tribunal’s omission of a specific finding on the date of investment does not justify the annulment of the award, as it would not have any impact on the outcome, including the extent of the economic impact of the breaches.37 One can agree that the tribunal would have reached the same conclusion if it had chosen between August 2011 and May 2012 as the date of investment, since the tribunal did not regard changes in the regulations affecting the RE sector prior to the former date as indicators that the system was going to be modified, and, thus, put investors who invested in 2007 and those who invested in 2011 or 2012 on equal footing. Royal Decree 1614/2010 could have been viewed as the ‘temporary precursor’ to the 2012–2014 disputed measures as it imposed production limits on and temporarily reduced incentives for, amongst others, wind power facilities (Eurus Energy v. Spain, para. 132–33).38 The preamble of the mentioned decree prescribes that it was adopted to guarantee the economic sustainability of the Spanish electricity system and resolve the inefficiencies regarding wind and other technologies.39 Accordingly, it may not be excluded that if equal weight had been given to the date of investment in the Watkins case as it was in Isolux, the outcome of the case would have been different. Indeed, while in the Isolux case, the tribunal ruled in favour of the respondent, in the Watkins case, the claimant was awarded full compensation.

It was also highlighted that tribunals had inconsistent approaches related to the weight given to other facts. For example, it was indicated that in the cases OperaFund40 and Cube, the tribunal completely rejected the relevance of the Spanish Supreme Court’s judgments, while in the case of Charanne41, Isolux and Stadtwerke München and others42, these judgments were considered to be relevant sources of information on the basis of which an investor could have predicted possible changes to a regulatory framework (Levashova 2020). For instance, the tribunal in the Cube case did not find Supreme Court decisions to be indicators to the claimants of the potential reduction and withdrawal of support schemes established under RD 661/2007. However, the tribunal in Isolux highlighted that it must take into account Spanish Supreme Court decisions in order to determine whether the investor was aware there were no obstacles for Spain to amend its regulatory framework. It was further concluded from this practice that the tribunals that took Spanish Supreme Court judgments as relevant in this respect ‘set more elaborate requirements for proper due diligence’, while those that regarded them as irrelevant did not consider due diligence as a requirement for the protection of legitimate expectations (Levashova 2020, p. 251). In the OperaFund and Cube cases, claimants were ultimately entitled to full reparation. In contrast, in the cases Charanne, Isolux and Stadtwerke München and others, the tribunals found Spain not liable for the breach of the FET standard and, hence, ruled in its favour.

Arguably, the most significant inconsistencies related to facts in the Spanish saga cases concern the interpretation of RD 661/2007 and the Electricity Law, which further led tribunals to conclude whether the investor was entitled to full compensation or limited compensation, i.e., a reasonable rate of return, for a breach of the FET standard. The compensation awarded to the investor significantly varies depending on whether a tribunal found that a full violation of the FET standard occurred or if it was only to the extent that the new regulatory framework did not secure a reasonable rate of return (Balcerzak 2023). As Brower pointed out in his dissenting opinion regarding The PV Investors, if the tribunal in The PV Investors were to have followed the approach of tribunals in preceding cases dealing with the same issues (offering full reparation), the investors would have been awarded approximately €632 million, contrary to the €91.1 they were entitled to due to the tribunal’s adoption of a different approach (offering reasonable profitability) (concurring and dissenting opinion of Charles N Brower on The PV Investors v. Spain, para. 17). For example, in the case of The PV Investors, the tribunal concluded ‘that the cardinal principle emerging from Article 30.4 of the Electricity Law and the implementing decrees up to RD 661/2007 is reasonable profitability’ (The PV Investors v. Spain, para. 596).43 In contrast, in the Novenergia II case, the tribunal found Spain’s arguments on the reasonable rate of return as unconvincing because the principle was vague and undefined at the time of investment (Novenergia II v. Spain, para. 673). In the OperaFund case, by relying on, inter alia, the interpretation reached in Novenergia II, the tribunal rejected the ‘reasonable return’ argument (OperaFund v. Spain, para. 489).44 Consequently, it concluded that ‘it is hard to imagine a more explicit stabilization assurance than the one mentioned in Article 44(3): ‘revisions […] shall not affect facilities for which the functioning certificate had been granted’. (ibid., para. 485). Hence, it concluded that the claimant was entitled to receive full compensation instead of reasonable profitability.

3.3. Going beyond Inconsistency in the Interpretation of Facts: The Stability Condition in the ECT

Article 10 (1) of the ECT provides that ‘Each Contracting Party shall, in accordance with the provisions of this Treaty, encourage and create stable, equitable, favourable and transparent conditions for Investors of other Contracting Parties to make Investments in its Area. Such conditions shall include a commitment to accord at all times to Investments of Investors of other Contracting Parties fair and equitable treatment.’45 The reason why the ECT is most-invoked treaty in the renewables sector may be founded on the recognition that its FET provision is not a traditional FET provision as it makes explicit reference to stability, and as such it appears to place a greater emphasis on ‘stable’ conditions for investments than other treaties (The PV Investors v. Spain; Watkins v. Spain; dissenting opinion of Philippe Sands on RENERGY v. Spain; dissenting opinion of Peter Cameroon on Infracapital v. Spain). Stability is particularly relevant in the case of the energy sector due to the above-mentioned characteristics of investments (dissenting opinion of Peter Cameroon on Infracapital v. Spain). Overall, there seems to be a consensus among tribunals that the stability condition in the ECT’s Article 10 (1) is not absolute (The PV Investors v. Spain; Infracapital v. Spain). Accordingly, the stability condition under Article 10 (1) does not imply freezing of the business conditions as they were when the investment was made, since it should not be seen as a form of stabilisation clause found in energy contracts or as an insurance policy (dissenting opinion of Philippe Sands on RENERGY v. Spain; Infracapital v. Spain; dissenting opinion of Peter Cameroon on Infracapital v. Spain; Hydro Energy v. Spain). This is so because the FET standard, including its stability obligation, needs to be balanced with the host state’s inherent right to regulate in the public interest and adapt its regulatory framework to changing circumstances when there is an economic or social justification for such action (The PV Investors v. Spain; Infracapital v. Spain; Eurus Energy v. Spain; Electrabel v. Hungary). The stability condition has been found to be related to the principle of reasonable and legitimate expectations (Charanne v. Spain; OperaFund v. Spain; Eurus Energy v. Spain; Krzykowski et al. 2021). Claims related to the breach of legitimate expectations arise when an investor is suffering losses due to the host state’s measures that violated the legitimate expectations the investor had when making the investment (UNCTAD 2012). Tribunals normally take for granted the idea that the principle of legitimate expectations is relevant when deciding on the breach of the FET standard and, thus, do not question its origins (Potesta 2013). This was found to be particularly an issue in the case of the interpretation of the stability condition under Article 10 (1) of the ECT, which was analysed by the arbitrators under the framework of investors’ legitimate expectations (dissenting opinion of Philippe Sands on RENERGY v. Spain). The idea of legitimate expectations is perceived as an extremely flexible tool that allows arbitrators to balance investors’ interests and host states’ right to regulate (Potesta 2013). Indeed, in the Spanish saga cases, different approaches of arbitrators can be identified depending on the weight given to the stability condition, discussed normally within the framework of investors’ legitimate expectations, compared to the host state’s right to regulate.46

The OperaFund and Watkins cases seem to pertain to the group of cases that provide greater emphasis on the stability condition. In the OperaFund case, the tribunal found separate breaches of the FET standard and the stability obligation (OperaFund v. Spain, para. 490, 508–13). Correspondingly, the stability condition, which is prescribed in the first sentence of Article 10 (1), was regarded as an autonomous requirement that provides for a distinct cause of action (dissenting opinion of Peter Cameroon on Infracapital v. Spain; Verburg 2019). In this case, breaches of the stability condition and the FET standard were discussed within the framework of investors’ legitimate expectations. Regarding the breach of the stability condition, the tribunal found that with RD 661/2007, the respondent ‘assumed an obligation of regulatory stability, which […] gave rise to legitimate expectations of stability under the ECT’ (OperaFund v. Spain, para. 512). Furthermore, the tribunal concluded that stable conditions were breached by the adoption of disputed measures as they fundamentally changed the expectations initially made under RD 661/2007 (ibid., para. 513). In a similar way, the tribunal in the Watkins case found that Spain cannot substantially alter the regulatory framework the investor relied upon or subject it to periods of legal uncertainty without violating the stability provision (Watkins v. Spain, para. 543). This is the case since Spain entered voluntarily into the ECT, by way of which it accepted limitations when it comes to modifying its regulatory framework (ibid.). Accordingly, it seems that, based on these decisions, a fundamental change in the legal framework per se triggers a breach of the stability condition.

Furthermore, could the breach of the stability condition be expected when the modifications are not fundamental but represent a small-scale change? Disputed measures resulting in a small-scale change could be identified in the Charanne case. The dissenting arbitrator in the case argued that the host state must compensate for the damage it caused by its valid exercise of regulatory power, even though in this case the change was not fundamental (dissenting opinion of Guido Santiago Tawil on Charanne v. Spain). The different conclusion of the dissenting arbitrator was based on his perception regarding the basis of the investor’s legitimate expectations. Contrary to the majority opinion in the Charanne case, the arbitrator argued that the creation of legitimate expectations is not limited to the existence of specific commitments but may also be based on the legal order in force when the investment was made. Such an approach, which regards the incentive scheme prescribed in general legislation as capable of constituting a specific commitment of regulatory stability, can be seen also, inter alia, in the Opera Fund and Watkins cases (Ipp et al. 2022). Correspondingly, depending on what arbitrators may perceive as a basis of investors’ legitimate expectations, RE investors could also be protected from small-scale changes such as those that occurred with the adoption of RD 1565/2010 and RDL 14/2010 mentioned above.

However, different conclusions should not be excluded, regarding both fundamental and small-scale changes. In the cases Isolux and The PV Investors, the tribunals rejected the existence of stability as an autonomous standard by finding that the stability condition cannot per se generate rights in favour of investors (Isolux v. Spain, para. 764–65; The PV Investors v. Spain, para. 567). The tribunals supported the view that the stability condition forms part of the FET standard. Similarly, the tribunal in RWE Innogy interpreted the first sentence of Article 10(1) as the one that is concerned ‘only with the conditions in which the Investment is made, as opposed to establishing any ongoing obligation of stability.’ (RWE Innogy v. Spain, para. 426). Accordingly, it concluded that the stability obligation exists only to the extent that it forms part of the FET commitment (ibid., para. 429). Moreover, the tribunal in the Infracapital case provided that the word ‘shall’ in the first sentence of Article 10 (1) of the ECT indicates an obligation but leaves open the question of whether it is autonomous from the obligation to accord FET (Infracapital v. Spain, para. 517). It, nevertheless, accepted the approach that stability forms part of the broader FET context (ibid., para. 518–20).

Moreover, the majority in Infracapital, contrary to the OperaFund and Watkins cases, rejected the general proposition that ‘a fundamental or continual change in regulations constitutes a breach of the FET standard.’ (ibid., para. 527) It went on to conclude that a significant, constant or complex regulatory change would not per se lead to the breach of the FET (ibid., para. 528). This is so because states have the right to regulate in response to changing circumstances to address a public concern (ibid., para. 527–28). Hence, absent specific assurances, measures must be arbitrary, unreasonable or discriminatory in order for the breach of the FET standard to be found (ibid., para. 531, 566). The approach of the tribunal in Infracapital was endorsed by Professor Philippe Sands in his dissenting opinion on Renergy (dissenting opinion of Philippe Sands on RENERGY v. Spain, para. 27). Professor Sands argues that the stability obligation should not be analysed under the framework of the doctrine of legitimate expectations as it does not depend on any commitment or assurance on the part of the state or consequential expectation on the part of an investor (ibid., para. 31–32). Correspondingly, he claims that the central issue is not whether the regulatory changes were foreseeable but whether Spain, when modifying the legal framework, acted in the public interest and in a proportionate, reasonable, consistent and non-discriminatory manner (ibid., para. 32).

The majority in Charanne came to conclude that, in the absence of a specific commitment towards stability, an investor cannot have a legitimate expectation that a regulatory framework will not be changed during the lifetime of the installation due to market needs or in the public interest (Charanne v. Spain, para. 510–11). Hence, the conclusion was that, absent specific commitments, investors’ legitimate expectations could be frustrated only if regulations were amended unreasonably, against the public interest, or in a disproportionate fashion (ibid., para. 512–40). The majority rejected considering RD 661/2007 and RD 1578/2008 as specific commitments (ibid., para. 492–93). Accordingly, it may be concluded that if a state adopts measures in the public interest that are not arbitrary, unreasonable or discriminatory the breach of the stability condition might not be established. This may be the case both concerning a fundamental change and a small-scale change of the regulatory framework.

4. Conclusions

Legal certainty is identified as one of the key concerns of the current ISDS system. As Raz has argued, vague laws can mislead or confuse those willing to be guided by them (Raz 1979). The FET standard under Article 10 (1) of the ECT is a vague standard. As was presented above, tribunals differently approach its stability condition and the interconnected state’s right to regulate. Accordingly, it remains uncertain to what extent RE investors will be protected under the ECT’s stability condition in the case of fundamental or small-scale changes. Although one group of arbitrators may argue that the fundamental change triggers per se a breach of a stability condition, others may argue that for the breach to be established, host states’ measures must be arbitrary, unreasonable or discriminatory. Nevertheless, it seems that it is not only the vague FET provision that has led to inconsistent decisions in the Spanish saga cases. The threat to legal certainty, at least in the similar line of cases such as Spanish saga cases, seems to be the varying weight given to the relevant facts, based on which the tribunal found a breach of the FET standard or decided to award a higher or lower amount of compensation to the investor. It is at the discretion of arbitrators to consider whether the timing of the investment, the presence of evidence indicating possible regulatory changes, and the reasonable rate of return prescribed in Spain’s domestic law will be relevant or irrelevant.

This paper agrees with the findings previously reached that the insurance role of the ISDS system in respect to RE cases is doubted as, although investors may prevail in the cases, there is certainly no guarantee of a positive outcome or clarity about the amount of damages they may receive in the case that a breach of an international investment agreement is established (Tienhaara and Downie 2018; Selivanova 2018; Boute 2012; concurring and dissenting opinion of Charles N Brower on The PV Investors v. Spain).