Reform or withdrawal from the ECT: What does it mean for coal?

With negotiators preparing for the seventh round of ECT modernization talks, and no signs of tangible progress, the continued emergence of new investor-state disputes under the controversial agreement has highlighted the risks the treaty poses for states as they phase out fossil fuels. 

Of all fossil fuels, coal has the greatest climate impact, and UN Secretary-General António Guterres has called its phase-out “the single most important step to get in line with the 1.5-degree goal of the Paris Agreement.” However, sovereign decisions to phase out coal inevitably affect the investment of coal power plant operators, frequently leading to disputes concerning compensation for lost profits. 

With its 55 contracting states, the ECT grants protection to a particularly high number of such investments in coal. Specifically, it allows foreign investors to use ISDS to challenge governmental measures aimed at phasing out coal. In ISDS cases, arbitral tribunals take final and binding decisions on such disputes, without the intervention of national courts, often in full confidentiality, and frequently by awarding significant amounts of monetary compensation. 

Investors in coal are increasingly using this option under the ECT. In February, the treaty generated widespread media attention following the news that RWE had begun ISDS proceedings against the Netherlands, challenging the Dutch government’s decision to ban the burning of coal for electricity generation by 2030. 

More recently, the energy giant Uniper commenced parallel court and arbitration proceedings against the Netherlands on similar terms. With this legal action, Uniper is challenging a Dutch law that forces the shutdown of the 1070-megawatt coal power plant that it is currently operating in Maasvlakte, Netherlands, by 2029. While Uniper maintains the position that the law violates the investment protection provisions of the ECT, the Dutch government has stated, with the support of civil society organizations, that power plant operators could and should have anticipated the decision. The Dutch law was also necessary for the Netherlands to comply with their commitments under the Paris agreement.

Wary of this litigation risk, the ECT contracting parties began in 2018 to negotiate a “modernization” of the treaty. While at the time of writing no compromise for such a reform of the treaty has been reached, the two most prominent options for a way forward appear to be an amendment of the treaty as per the EU’s most recent amendment proposal (EU proposal), 0r a coordinated withdrawal of the EU and its member states (EU withdrawal). 

Given the vital importance of a swift coal phase-out for the energy transition, it is uncontroversial that the degree of protection that the ECT affords to coal investments should inform policymakers when considering these options. In other words, we can ask how many coal power plants would be protected if the treaty were amended as per the EU proposal,  and how does this number compare to the scenario of an EU withdrawal?

Attempting to answer these questions, this paper offers a brief overview of the two aforementioned options, i.e. the EU proposal and an EU withdrawal in the first section. In a second step, we attempt to discern the number of coal power plants that would generate ISDS risk under the ECT for each of these options  before drawing a conclusion.

Current options: the EU amendment proposal or a coordinated EU withdrawal

In February 2021, the European Commission released an updated proposal for amending the ECT, citing the need to bring the scope of the treaty in line with the  Paris Agreement. In this proposal, the Commission suggests distinguishing the treatment afforded to existing and future investments. 

The proposal suggests excluding future investments in fossil fuels from the ECT’s scope of protection while maintaining protection for future gas investments below a certain carbon threshold. For the purpose of the proposal, future investments are understood to be investments made after the proposed amendment enters into force. According to the proposal, future investors would therefore no longer have recourse to ECT-based ISDS.

Additionally, the EU suggests continuing to afford protection to fossil fuel investments that were made before the time of entry into force of the proposed amendment (existing investments) for a period of ten years after this date. This means that, if adopted, such existing fossil fuel investors could still sue ECT contracting states using ISDS for at least another decade.

To be adopted, an amendment such as that proposed by the EU would require the unanimity of all 55 contracting states of the treaty. However, some non-EU countries continue to resist the reform effort and, according to recently leaked diplomatic cables, support for the EU proposal is “close to non-existent.”

In light of the uncertainty of reaching a consensus on the EU proposal, the EU and some of its member states[1] no longer rule out the possibility of withdrawing from the treaty altogether. One of the perceived obstacles to this alternative is the ECT’s so-called sunset clause. According to this clause, the treaty’s provisions will continue to apply to existing investments for a period of 20 years post-withdrawal. 

However, recent legal analysis shows that the EU’s member states could neutralize the sunset clause among themselves before proceeding to a coordinated withdrawal from the treaty. Specifically, they could do so by modifying the treaty’s content in accordance with rules of public international law. In contrast to the EU proposal, this would allow the withdrawing states to bar access to ISDS for existing fossil fuel investments between these states immediately. Such an approach would also not require unanimity of all ECT contracting states. 

Assessing the impact of the EU proposal and a coordinated withdrawal on the protection of coal power plants

As stated above, states’ regulatory freedom to proceed with a swift phase-out of coal plays a decisive role in the ability to meet the objectives of the Paris Agreement. It is therefore vital to quantify the litigation risk the ECT is likely to generate with respect to existing investments in coal power plants. 

One way to quantify this litigation risk would be to discern the likely compensation arbitral tribunals would award to coal power plant investors under the ECT in case of a phase-out. However, such an attempt faces important difficulties due to various factors. These include, for instance, a plethora of different valuation techniques applied by arbitral tribunals, unclear future operating cycles, future fluctuations in the prices of electricity and coal, and the future development of emissions trading schemes.

As a more viable approach to quantifying the litigation risk, we suggest identifying the number of coal power plants with investors that could have recourse to ECT-based ISDS. Considering the two aforementioned options – an amendment as per the EU proposal and an EU withdrawal – we have analyzed data on foreign-owned coal power plants to identify this number for each option. 

More specifically, the meta-analysis was guided by the following two research questions: (1) How many existing coal power plants would be protected, if the EU’s amendment proposal were adopted today? And (2) how many of these plants would be protected if the EU and its member states, but no other state, neutralized the sunset clause and withdrew from the treaty today?

Methodology and results

To answer these questions, we created a dataset using data on global foreign-owned coal power plants compiled by the NGO End Coal and presented as part of the Global Coal Plant Tracker database (GCPT database). To compile this data, End Coal has gathered preliminary lists of coal power plants in each country from public and private data sources, including information on plant ownership. In a second step, specific categories (i.e. whether a coal power plant is still operating or has been retired) are assigned to the plants (for more information on the methodology, see here).

The data sample was then updated to include only those plants that are protected by the ECT as of 2021 and that are currently operating, as well as to exclude plants that were decommissioned in the meantime (i.e. between the moment when the data was compiled in 2020 and the moment of writing) or that have not yet been built. 

We assumed that the ECT “protects” a coal power plant if at least one of its owners would have standing under the current Article 26 of the ECT to bring an ISDS claim against the state in which the plant is located. Ownership information was also retrieved from the GCPT database, which contains some information as to direct owners and parent companies. More complex structures of corporate ownership that were not included in the GCPT database, such as certain minority stakes or sister companies with different nationalities, were not taken into account. 

As defined by Article 26 of the ECT, investors from an ECT contracting state (home state) can bring an ISDS claim, if their investment is located in another ECT contracting state (host state). Consequently, we only included those plants that are located within the area of an ECT contracting state and are at least partly owned by an investor based in another ECT contracting state. This step ensured the exclusion of coal power plants that are located within the area of an ECT contracting state[2] and owned exclusively by investors from the same state (domestically owned coal power plants). Coal power plants located within the area of an ECT contracting state and owned exclusively by investors from non-ECT contracting states were also excluded. Since the Russian Federation signed but neither ratified nor provisionally applied the treaty, coal power plants located in the area of the Russian Federation or in an ECT contracting state and at least partly owned by Russian investors were also excluded.

The resulting list contained all existing coal power plants that are currently protected under the ECT and therefore “generate” an ISDS risk. We counted 61 coal power plants that are currently generating an ISDS risk under the ECT.

If an amendment according to the EU proposal entered into force today, these 61 coal power plants would continue to be protected under the ECT for a period of ten years after the amendment takes effect. Assuming an optimistic scenario, in which negotiations came to a conclusion in one or two years, investment protection for fossil fuels would at the earliest end in 2032-33 or later. It is important to note that this is an unrealistically optimistic estimate. For such an amendment to enter into force, it would have to first be adopted unanimously by all 55 ECT contracting states. In a second step, it would then have to be ratified, accepted or approved by at least three-fourths of the contracting states, only taking legal effect 90 days after the last instrument of ratification, acceptance or approval has been deposited. 

In practice, this would allow foreign investors of these coal power plants based in any ECT contracting state to continue to sue ECT host states for a decision to phase out coal for well over a decade. For instance, if a state were to decide in 2028 to phase out coal by 2050, such investors would therefore still have three years to initiate ISDS proceedings to challenge the decision under the ECT. 

How would this compare to the scenario of a coordinated withdrawal of the EU and its member states? To determine the number of coal power plants potentially affected by such an EU withdrawal, we counted the following four categories of power plants: 

  1. 45 coal power plants located within the EU and owned by an investor from the EU;
  2. Five coal power plants located within the EU and owned by an investor from a non-EU ECT contracting state;
  3. Seven coal power plants located in the area of a non-EU ECT contracting state and owned by an investor from the EU; and 
  4. Four coal power plants located in the area of a non-EU ECT contracting state and owned by an investor from another non-EU ECT contracting state.

If the EU and all its member states were to withdraw from the treaty and assuming that these states and the EU would neutralize the ECT’s sunset clause among themselves (inter se neutralization), the 45 power plants in category 1 would no longer generate a litigation risk under the ECT. Indeed, a coordinated EU withdrawal would deprive investors based in the EU of access to ECT-based ISDS with regard to investments in power plants in the EU. More specifically, due to the inter se neutralization, these investors could no longer avail themselves of the sunset clause after the withdrawal. 

Further, the 16 remaining power plants in categories 2, 3, and 4 would continue to enjoy full protection under the ECT, since the inter se neutralization would only extinguish the sunset clauses effects between the EU and its member states. The clause would therefore remain intact between EU and non-EU ECT contracting states and investors in the latter three categories could continue to have recourse to ECT-based ISDS. 

Moreover, of these 16 power plants, four  are located outside the EU and owned exclusively by investors from non-EU ECT contracting states (category 4). These four power plants, therefore, do not generate litigation risk for the EU or its member states. The remaining 12 coal power plants are either based in the EU or owned by investors from the EU. Seven of these 12 coal power plants are located outside the EU and at least partially owned by investors from the EU (category 3). These investors could continue to sue the non-EU ECT host states even after a coordinated EU withdrawal. The remaining five coal power plants are based in the EU and are at least partially owned by investors from non-EU ECT contracting states (Switzerland and the United Kingdom). Only these five coal power plants would generate litigation risk for the EU and its member states after a coordinated EU withdrawal (category 2).

It is important to note certain limitations of this analysis. Firstly, it does not take into account post-withdrawal acquisitions of coal power plants or later changes in ownership that might generate an additional litigation risk. Furthermore, it also does not consider the post-withdrawal risk of forum shopping. As stated above, the analysis relies on the GCPT database and the ownership information indicated therein. In reality, it is likely that ownership structures are more complex, involving more intricate corporate structures, such as holdings, sister companies, more widespread minority stakes or shareholder nationalities. More detailed studies will therefore be required to take into account these aspects and, in particular, to quantify the post-withdrawal risk of forum shopping. Such studies could also complement data from the GCPT database with other databases on coal power plants, request more detailed information from public sources or imply more detailed research on ownership via commercial registers. We nevertheless argue that the results of the current analysis highlight the significant practical differences – in terms of concrete risk of ISDS – between the scenarios of an ECT amendment and an EU withdrawal. 

Conclusion

The results of the current analysis show a significant difference in the number of coal power plants that would be protected under the ECT in case of an amendment as per the EU proposal on the one hand and in case of an EU withdrawal on the other. The ECT currently protects 61 coal power plants that would continue to be protected for a period of ten years under the EU proposal. By contrast, if the EU and its member states were to withdraw from the treaty, only 16 coal power plants would continue to be protected by the ECT. Moreover, only five of these coal power plants are either based in the EU or owned by investors from the EU, and only three are based in the EU and have non-EU owners. Therefore, if the EU and its Member States were to withdraw from the ECT while neutralizing the sunset clause among themselves, they would only face an ISDS risk in respect of these three coal power plants.

The authors would like to thank Fabien Flues (PowerShift), Cornelia Maarfield (CAN Europe), Amandine Van Den Berghe (ClientEarth) and Nathalie Bernasconi-Osterwalder (IISD) for the valuable comments they provided.

Authors

Lukas Schaugg is an International Economic Law Fellow at IISD and a PhD-researcher in investment law at Osgoode Hall Law School, Toronto, Canada.

Lea Di Salvatore is a PhD researcher at the School of Law of the University of Nottingham and a visiting scholar at the Center for Environmental and Sustainability Research at the Nova University of Lisbon. 

Notes

[1] E.g., Spain; France; Poland
[2]Ownership data was retrieved from the Global Coal Plant Tracker Database that is compiled and updated by the NGO End Coal.