Eiser v. Spain: Reinforcing the Importance of Early Disclosure in Investment Arbitration

By Sumit Chatterjee (National Law School of India University, Bangalore)

An ICSID Committee, chaired by Ricardo Ramirez-Hernandez, recently annulled an arbitral award rendered in favour of a solar power investor in the case of Eiser Infrastructre Ltd. v Republic of Spain. [1] The primary ground on which the award was annulled was the undisclosed relationship between Stanimir Alexandrov, who was one of the arbitrators on the arbitral tribunal that rendered the award, and one of the experts appointed by the Claimants to make their case. The committee came to the conclusion that the undisclosed relationship created a “manifest appearance of bias”, which qualified the threshold of annulment on the grounds of improper constitution of the tribunal, and a serious departure from a fundamental rule of procedure under 52 of the ICSID Rules. [2]

After understanding the decision of the committee, and reconciling the same within the ICSID Rules framework, this post will explore two broader ramifications of this decision on investment arbitration; first, the importance of early disclosure of potential and existing conflicts by arbitrators, and second, the importance of this decision in understanding the double-hatting debate in international arbitration.

Decision of the committee

After the arbitral tribunal chaired by John Cook, and comprising of Stanimir Alexandrov and Campbell McLachlan, had decided the dispute between UK-based infrastructure firm Eiser Infrastructure Ltd. and the Republic of Spain in favour of the former, and ordered Spain to pay €128 million, Spain filed an application to annul the award, and to deliberate upon the same, a three-member committee comprising of Chairman Ricardo Ramírez-Hernández, Dominique Hascher and Teresa Cheng was constituted. After Teresa Chang stepped down from the committee, she was swiftly replaced on the committee by Makhdoom Ali Khan.

Spain had made their case for annulment of the award on two broad grounds. First, that the tribunal had been improperly constituted, under Art. 52(1)(a), as a result of the undisclosed relationship between the nominated arbitrator of the Claimant, Stanimir Alexandrov, and one of the experts of the Brattle group that was appointed by the Claimants, Carlos Lapuerta. The influence that Alexandrov exercised on being a part of the tribunal, and the failure to provide Spain with an opportunity to challenge his appointment on the ground of this relationship, was invoked by Spain to claim a serious departure from a fundamental rule of procedure under Art. 52(1)(d) of the ICSID Rules. Second, they claimed that the tribunal had failed to provide reasons, under Art. 52(1)(e), and had manifestly exceeded their powers, under Art. 52(1)(b), as a result of an improper award of damages.

The committee deliberated on the first ground, and analysed the relationship between Mr. Alexandrov and the expert retained by the Claimants in great detail. It was soon discovered that during the proceedings themselves, Mr. Alexandrov had been acting as counsel of a reputed law firm in other arbitration proceedings, and had employed the services of the Brattle group as experts. Furthermore, in four of these proceedings, the impugned expert, Mr. Lapuerta, had been the testifying expert on behalf of the Brattle group. Thus, these well-established past and present connections between the arbitrator and the expert retained by the Claimants suggested a manifest appearance of bias on the part of the arbitrator, and would thus qualify the threshold under Art. 52(1)(a) of the ICSID Rules to hold that the tribunal had been improperly constituted. The committee referred to the standard laid down in Blue Bank International v Bolivia, [3] by Chairman Kim, wherein it was stipulated that in order to determine whether an arbitrator had failed to comply with the standards of independence and impartiality, the standard should be one of whether “a third party would find an evident or obvious appearance of lack of impartiality on reasonable evaluation of the facts in this case”. [4]

The Committee also stated that the failure on the part of Mr. Alexandrov to disclose this conflict had severe effects on the proceedings themselves, as it hampered the constitution of an independent tribunal, and also adversely affected Spain’s right to a fair arbitration. It thus held that the failure to disclose had a “material effect” on the proceedings, and thus the tribunal had seriously departed from a fundamental rule of procedure under Art. 52(1)(d) of the ICSID Rules. Having considered the first ground sufficient to annul the award rendered by the tribunal, the committee did not delve into the intricacies of the second ground raised by Spain.

The next part of this post will explore the duty of disclosure in the context of investment arbitration, and analyse the decision of the committee from that perspective.

Tracing the contours of the duty to disclose in Investment Arbitration

One of the hallmarks of the arbitral process is having independent and impartial arbitrators on the tribunal to adjudicate the disputes between the parties. While independence and impartiality have often been used interchangeably in the context of understanding the duty of the arbitrators, it is well established that the former refers to a more objective standard of ensuring that the arbitrator does not have any personal, financial or professional ties with any of the parties, witnesses, counsel etc., while the latter is more of a subjective standard that is based on the conduct of the arbitrator during the proceedings. [5] With respect to investment treaty arbitration, the requirement of independence and impartiality assumes much accentuated significance, as a result of the public interest element, and the political and economic ramifications of the decision on the Respondent State. The duty to disclose is a corollary of the independence and impartiality requirement, as it places a positive duty upon the appointed arbitrators to disclose any and all potential and existing conflicts of interest with any of the parties, witnesses, counsel etc. involved in the arbitration. [6] The disclosure is also a safeguard to ensure that the arbitrator is secured from any future challenges on his/her independence or impartiality by one of the parties on the grounds which have been disclosed. [7] Under the ICSID Arbitration Rules of 2003, the disclosure duty is grounded in Article 6, the scope of which was expanded through the amendments made in 2006 to make it a continuing obligation on the part of the arbitrator. [8]

In the Eiser case, the committee assumed the role of a “guardian of the ICSID system [9] and held that the bar must be set high when it comes to disclosure requirements given the importance of early disclosure for a fair and just arbitral process. By holding the same, the committee reinforced the importance of prompt and early disclosure in investment arbitrations. Not only does a disclosure aid the parties in raising a timely challenge to the appointment of the arbitrator if it deems so necessary, it also waives the right of the party from raising such challenge at a later date, or post the rendering of the award, in cases where they fail to make such a challenge within the stipulated time period. The failure of Mr. Alexandrov to make a timely disclosure of the conflict resulted not only in declining Spain the opportunity to challenge his position on the tribunal, but also in influencing the decision of the other members of the tribunal with his continued presence on the tribunal, which in the eyes of the committee, would raise a reasonable suspicion of bias to any independent observer.

While the case certainly emphasised the importance of making prompt and early disclosure of conflicts of interest in Investment Arbitration, it also highlighted an issue that has garnered significant academic interest and debate for a long while: the issue of double-hatting.

Double-hatting in Investment Arbitration: A Necessary Evil?

Double hatting has gained significant traction in the academic discourse on investment arbitration ever since Prof. Phillipe Sands first alluded to the dilemma at the 2009 IBA Conference. [10] It essentially refers to the growing trend in investment arbitration, wherein lawyers who are appointed as arbitrators in particular cases continue to represent other parties as counsel in arbitration proceedings at the same time. Double hatting raises a number of poignant ethical and practical concerns, as a result of the unavoidable conflict of interests that arise in light of the interwoven nexus of relations which lawyers have, both in his/her role as a counsel, and as an arbitrator. One of the overarching concerns in this regard is of role confusion, which refers to the situation where arbitrators try to issue an award that would be favourable for them in a case where they are representing a different client as counsel. [11] Role confusion is also linked to another related concern with double hatting, which is the problem of issue conflict. An issue conflict arises when an arbitrator has to adjudicate on an issue that was in contention in an earlier or ongoing case where he/she served as a counsel, or as an arbitrator. [12] Double hatting thus increases the possibility of an occurrence of issue conflict for an arbitrator, as was evident in the case of Telekom Malaysia Berhad v Ghana, [13] and raise justifiable doubts as to the arbitrator’s impartiality and independence. [14]

The need for reform to combat the predicament of double hatting has been all the more pronounced as a result of the prevailing no-man’s land with respect to ethical standards that prevail in arbitration proceedings, not just for legal counsels, but also for arbitrators. This invigorated call for a reform in the prevailing paradigm has led to a number of recent developments, which also illustrate two extremely different approaches to tackling this predicament.

The Dutch Model BIT has employed a rather extreme approach, explicitly disallowing double hatting, and precluding arbitrators from acting as legal counsels. [15] It also mentions that no arbitrator should have acted as a counsel in any investment arbitration proceeding in the previous five years. [16] Another radical change that the Dutch Model BIT makes is to completely do away with party-appointed arbitrators, and instil the power to appoint arbitrators solely to a competent appointing authority.

This development has come in light of the increasing concern of politicisation of Investor-State arbitrations, and how the appointment of arbitrators to constitute the tribunal accentuate this concern more than any other factor. In fact, in the preliminary identifications of possible areas of reform in investor-state arbitration by the UNCITRAL Working Group III, [17] the concern that arises from completely shifting the burden of appointments from parties to an appointing authority is a re-politicisation of the investment arbitration paradigm. [18] It has been stated, for example, that the influence of States on appointments would continue to exist while the investor would lose out on having any say in the appointment process. While there have been suggestions as to limit the influence of States, and to include investors to be a part of the process, such as screening, consultations etc., there is no black and white position on this issue as of yet. [19]

However, the Working Group III was also responsible for the conceptualisation of the Draft Code of Conduct for Adjudicators in Investor-State Dispute Settlement, released by the ICSID and the UNCITRAL. The Draft Code flags the issue of double-hatting in Article 6, [20] which aims at a “limit on multiple roles” for adjudicators. The Draft Code came into being due to growing concerns about the numerous ethical and practical predicaments in investment arbitration proceedings, due to the different professional relations and roles of the appointed arbitrators. It illustrates the median approach, which seeks to remedy the concerns of double hatting within the prevailing framework in investment arbitration, while also maintaining the balance between the interests of both the investor and the Responding State party in the arbitral process, especially during appointments.

The public interest element, and the fact that proceedings in investment arbitration are indeed public, call for a greater scrutiny of the independence and impartiality of the arbitrators appointed. This tension, between party autonomy in the choice of arbitrators on one hand, and ensuring the right to a fair and independent arbitration on the other, has been the crux of the academic debate surrounding double hatting. And it is in this context that the Eiser case has taken a firm stand. The ICSID committee, by recognising the failure to disclose the conflict as a ground to annul the award, has illustrated the extremely serious implications of double-hatting in an investment arbitration, where it can cast a shadow over a successful award rendered in favour of a party, and ultimately lead to its annulment. Not only does the decision come at a crucial time, given the prevailing debate around double hatting, it also comes as a telling warning for lawyers who serve on tribunals in their roles as arbitrators to recognize and adhere to the duty of being independent and impartial throughout the proceedings. The Eiser case also demonstrated that the wide nexus of connections that a lawyer has, and people he/she engages with in order to represent their clients, leads to a number of potential conflicts, as was observed here with the expert of the Brattle group, retained by the Claimants. Arbitrators must be extremely wary of such conflicts, and comply with the best practice of early and prompt disclosure to the best of their abilities. The jury is still out on whether the setting up of an international investment adjudicatory body is in the best interests of resolving all the problems that exist in the investor-state dispute resolution settlement mechanism, but the Eiser decision has, by taking a firm stand against any minutiae of an appearance of bias, shown that the present system is also well equipped to provide parties what they wish for: a neutral, efficient and fair result.


[1] Eiser Infrastructre Ltd. v Republic of Spain (ICSID Case No. ARB/13/36).

[2] ICSID Rules, Article 52, “(1) Either party may request annulment of the award by an application in writing addressed to the Secretary-General on one or more of the following grounds: (a) that the Tribunal was not properly constituted; (b) that the Tribunal has manifestly exceeded its powers; (c) that there was corruption on the part of a member of the Tribunal; (d) that there has been a serious departure from a fundamental rule of procedure; or (e) that the award has failed to state the reasons on which it is based.”

[3] Blue Bank International v Bolivia (ICSID Case No. ARB/12/20).

[4] Ibid.

[5] The distinction was emphasized in Suez v. Argentina (ICSID Case no. ARB/03/19).

[6]Noah Rubins and Bernard Lauterberg, ‘Independence, Impartiality and Duty of Disclosure in Investment Arbitration’ in Christina Knahr, Chrishtian Koller et al., Investment and Commercial Arbitration – Similarities and Divergences (Eleven International Publshing, 2010).

[7] M.B. Feldman, ‘The annulment proceedings and the finality of ICSID arbitral awards’ [1987] 2(1) ICSID Rev.

[8] ICSID Arbitration Rules (2003), Article 6. Rubins and Lauterberg (n vi).

[9] Cosmo Anderson and Sebastian Perry, ‘Undisclosed expert ties prove fatal to ICSID award’ (Global Arbitration Review, 12 June 2020) < https://globalarbitrationreview.com/article/1227900/undisclosed-expert-ties-prove-fatal-to-icsid-award&gt; accessed 4 July 2020.

[10] Dennis H. Hranitzky and Eduardo Silva Romero, ‘The ‘Double Hat’ in International Arbitration’ (New York Law Journal, 14 July 2010) < https://www.law.com/newyorklawjournal/almID/1202462634101/The-Double-Hat-Debate-in-International-Arbitration/?slreturn=20200731133829&gt; accessed 4 July 2020.

[11] Frederick A Acomb and Nicholas J Jones, ‘Double-Hatting in International Arbitration’ (2017) 43 Litig 15.

[12] Ibid, at 16.

[13] Telekom Malaysia Berhad v Ghana (UNCITRAL Arbitration at the PCA, The Hague).

[14] In Telekom Malaysia Berhad v Ghana (UNCITRAL Arbitration at the PCA, The Hague), Ghana challenged the presence of the arbitrator nominated by the Claimants, Prof. Emmanuel Galliard, on the ground that he was acting as counsel for Morocco in the annulment proceedings of the award rendered in RFCC v Morocco (ICSID Case No. ARB/00/6), where he was challenging an argument that was being relied on by Ghana in the present case. The matter went up before the District Court in the Hague, which held that the dual roles would certainly hint at an appearance of being influenced by his role as counsel on his position on the tribunal. Subsequent to this decision, Prof. Galliard resigned from his role as the counsel of Morocco in the annulment proceedings and continued to serve as an arbitrator on the tribunal

[15] Netherlands Model Bilateral Investment Treaty, 2018.

[16] ‘The new draft Dutch BIT: what does it mean for investor mailbox companies?’ (HSF Notes, 30 May 2018) < https://hsfnotes.com/arbitration/2018/05/30/the-new-draft-dutch-bit-what-does-it-mean-for-investor-mailbox-companies/&gt; accessed 4 July 2020.

[17] UNCITRAL Working Group III: Investor-State Dispute Settlement Reform 2020.

[18] Fernando Dias Simoes, ‘UNCITRAL Working Group III: Would an Investment Court De-politicize ISDS’ (Kluwer Arbitration Blog, 25 March 2020) < http://arbitrationblog.kluwerarbitration.com/2020/03/25/uncitral-working-group-iii-would-an-investment-court-de-politicize-isds/?doing_wp_cron=1598784017.8404219150543212890625&gt; accessed 4 July 2020.

[19] Ibid.

[20] UNCITRAL and ICSID, Draft Code of Conduct for Adjudicators in Investor-State Dispute Settlement, Article 6 – “Adjudicators shall [refrain from acting]/[disclose that they act] as counsel, expert witness, judge, agent or in any other relevant role at the same time as they are [within X years of] acting on matters that involve the same parties, [the same facts] [and/ or] [the same treaty]”.

The Shortcomings of the Proposal for an “International Court System” (ICS)

by Dr. Nikos Lavranos LLM, Secretary General of EFILA*

During 2015 it became clear that the European Commission was under mounting pressure from the European Parliament (EP), Trade Ministers of several EU Member States, anti-ISDS NGOs and the media to propose more “reforms” of the investor-State dispute settlement (ISDS) system that is contained in CETA and envisaged to be included in TTIP as well.

EFILA decided to establish a Task Force – consisting also of non-EFILA members – to analyse the final proposal for a so-called “International Court System” (ICS), which the European Commission formally adopted on 12 November 2015 and transmitted it to the US as a basis for further negotiations within the context of the TTIP negotiations.

During the debate in the European Parliament and among several Trade Ministers of EU Member States one key issue pointing towards a “solution” and which was continuously repeated was the creation of a permanent investment court consisting of publicly appointed judges. It was argued that in contrast to the current system of ad hoc arbitration consisting of party-appointed arbitrators, which has been characterized as “private”, behind closed doors dispute resolution, which biased towards the investor, a permanent investment court with judges would ensure fairer and better adjudication of investment disputes. Another related key issue, which was considered important for a “solution” was the creation of an appeal mechanism. Again the rather simplified characterization that ISDS disputes have no appeal possibility and are completely beyond the control of national courts, was used as a justification for the need of an appeal mechanism.

The European Commission had to incorporate these points otherwise a ratification of TTIP by the EP and the Member States would seem rather illusory. Having had significant experience as a disputing party in the WTO, which happens to include the Appellate Body as a permanent (quasi)judicial body, it was a small step for the European Commission to copy and paste many of the WTO dispute settlement elements into its ICS proposal.

The structure of the 60-pages EFILA Task Force analysis is as follows:

Chapter 1 analyses not only the ICS proposal as such, but also the process that preceded the proposal. This is important in order to understand the political context in which this proposal is embedded. It critiques certain aspects of the ICS proposal and raises a number of issues which the Task Force considers should be addressed in developing the ICS proposal further.

Chapter 2 provides an extensive overview of the already existing forms of appeal and annulment of investment awards. It also highlights the reform efforts in this regard by the PCA and the ICSID Secretariat. This overview provides a detailed picture of the status quo (including both the mechanisms and methods of operation), from which the ICS proposal departs. This analysis also draws critical attention to features or elements of the current system of ISDS which could be addressed in developing the ICS proposal.

Chapter 3 turns towards the WTO dispute settlement system by first explaining the features of the appeal system and then by examining to what extent this system could successfully be transplanted into the ICS and the limitations in so-doing.

Finally, Chapter 4 wraps up this analysis by providing some general conclusions as to matters which require consideration by the Contracting Parties in developing the ICS proposal further. In particular, the issues highlighted concern the methods of selection of the judges (and the implications of a move towards a system whereby the Respondent maintains, but the Claimant is deprived of, a role), the size of the pool of candidates for the two-tiered system, the relationship between the ICS and the CJEU and how the ICS will operate in the wider context of resolution of investor-state disputes under other instruments.

The conclusions of the Task Force report can be summarized as follows:

  1. The paper concludes that the ICS proposal is, first and foremost, a bold move to appease the EP and the public opinion in many EU Member States, which are critical against TTIP generally, and in particular against including any type of ISDS. The ICS proposal attempts to make the inclusion of an investor-state dispute settlement mechanism in TTIP politically acceptable, while at the same time trying to address the perceived shortcomings of the existing ISDS.
  1. The paper notes that – in contrast to the public perception – mechanisms for limited review of investment arbitration awards are already in place, such as the ICSID annulment mechanism and the setting aside procedure for non-ICSID awards by national courts. These mechanisms – while not perfect – provide useful corrective tools.
  1. The analysis of the WTO dispute settlement mechanism illustrates that caution should be exercised in simply transplanting it to investor-state disputes. The reason is that WTO law is structurally different from investment law, serves different purposes and has different users.
  1. Generally, it can be concluded that the ICS proposal clearly breaks with the current party-appointed, ad-hoc ISDS as provided for in practically all BITs and FTAs. The main result is that it deprives claimants of any role in the appointment of the judges, while giving the respondent States the exclusive authority to do so, albeit in advance of a particular case. The appointment of the judges by the Contracting Parties raises several problems, which the ICS proposal does not sufficiently address.
  1. The pre-selection of the TFI and AT judges by the Contracting Parties carries the inherent risk of selecting “pro-State” individuals, in particular since they are paid by the States (or rather their tax payers) alone. Apart from this danger, it remains doubtful whether a sufficient number of appropriately qualified individuals with the necessary expertise can be found. This is particularly true since many professionals currently working in arbitration may be excluded on the basis that they could be considered to be biased. The pool of TFI and AT judges would seem to be limited to academics, (former) judges and (former) Governmental officials. That might not be sufficient to guarantee the practical experience and expertise needed and/or independence from the State.
  1. The standard of impartiality and independence of the judges is highly subjective, and their independence on a practical level is not assured by the proposed text. Also, the system of challenging TFI judges and AT members can be further criticised for envisaging that the presiding judge will decide the challenge against one of his own colleagues on the bench, rather a decision being made by an independent outside authority.
  1. The system of determination of Respondent (in the case of the EU or Member States), in particular the binding nature of that determination, which is done by the EU and its Member States alone, creates significant disadvantages for the claimant and does not allow the ICS tribunals to correct any wrong determinations. This could result in cases being effectively thrown out because of a wrong determination of the Respondent.
  1. Since the ICSID Convention is not applicable to the EU, the recognition and enforcement of

ICS decisions remains limited to the EU and the US. The proposal also fails to clarify the difficulties elated to the New York Convention 1958.

  1. The ICS proposal does not address the difficult legal situation between the CJEU and other international courts and tribunals. There is no reason to believe that the CJEU would be more positive towards the ICS as compared to its outright rejection of the European Court of Human Rights when it comes to the potential interpretation or application of EU law. Also, the CJEU’s consistent rejection of any direct effect of WTO AB panel reports – even those that have been approved by the DSB and after the implementation deadline has lapsed – raises doubts as to the legal effects of ICS decisions within the European legal order.
  1. In sum, the suggested creation of a two-tier (semi)permanent court system would give the Contracting Parties a significantly stronger role in the whole dispute settlement process – potentially at the expense of both the investor/claimant and the authority of the ICS. In particular, the appeal possibility carries the risk of burdening small and medium investors by increasing the potential length of the proceedings and costs.
  1. While the US position towards the ICS proposal remains unclear for the time being, it also remains unclear how the ICS proposal could be multilateralized. Indeed, the perceived shortcomings of the current ISDS system is based on the fact that more than 3,000 BITs/FTAs are in place, which have been concluded by practically all countries in the world. The ICS proposal – limited to TTIP and perhaps extended to CETA – does not change that. The way the UNCITRAL Transparency Rules of 2014 are incrementally applied by way of an opt-in system established by a separate international treaty could be a possible way forward.
  1. As the TTIP negotiations between the US and the EU are now focusing on the ICS proposal, this is a perfect moment to further improve the proposal by addressing the matters identified in this analysis.
  1. Finally, the US and the EU should also consider whether it would not be more preferable to modify and improve existing systems, such as turning the ICSID annulment procedure into a full appeal mechanism.

This in-depth analysis is very timely and arguably one of the first ones following the formal adoption of the ICS proposal by the European Commission last November.

The EFILA Task Force paper raises many issues and provides some answers, but certainly leaves many problems untouched. At the EFILA Annual Conference which will take place on 5 February in Paris, the last panel will specifically discuss this report. All members of the investment arbitration community are welcome to (still) register for the conference or to submit their constructive comments to Dr. Nikos Lavranos, LLM, Secretary General of EFILA, at: n.lavranos@efila.org