Margins of Wrongfulness in Arbitration: Measuring the Tinker in Eiser v. Spain

Apr 11, 2024 | HILJ-HIALSA International Arbitration Collaboration, Online Scholarship

Galo Márquez*

The saga of Spanish investment arbitrations born out of Spain’s decision to disapply incentives in the renewable sector in the face of a financial crisis continues to generate awards on quantum and liability. The recent decision in Eiser v. Spain is one of many that has contributed to a pool of inconsistent arbitral awards. Given the increasing number of decisions, it is not surprising that arbitral tribunals have reached divergent decisions as to the wrongfulness of Spain’s measures. In his Dissent on Liability and Quantum, Prof. Philippe Sands affirms that “[n]o doubt, reasonable folk will question the wisdom of creating a system that allowed so many competing and contradictory awards to flower, and introduce the changes that seem so necessary” (¶ 1). This article addresses some of the inconsistencies arising from how arbitral tribunals have recognized that a state may make regulatory changes. Still, this recognition has not meaningfully impacted the damages awarded to investors.

Nor have the competing views of arbitral tribunals resulted in a uniform application of the law of damages in international arbitration. Typically, when tribunals determine that a state’s measure violates an investment treaty, they award damages for injury attributable to any part of the measure. By reviewing the decision in Eiser v. Spain, this article suggests a different approach. It argues that states should only be liable for the portion of damages attributable to the difference between the injury caused by the measure actually imposed, and the hypothetical injury that would have been caused by a lawful alternative measure. I will refer to this concept as the “Margin of Wrongfulness.”

It must be mentioned that this article builds upon the work of the Academic Forum on Investor-State Dispute Settlement (“ISDS”) before the United Nations Commission on International Trade Law Working Group III (“UNCITRAL”) on matters related to damages. The author is an appointed Member of the Academic Forum on ISDS and a representative observer before Working Group II on Arbitration and Dispute Resolution. However, the views expressed here are mine alone. The benefit of this article is twofold. First, it demonstrates an inconsistency in the way that arbitral tribunals have awarded damages, which the ISDS community has sparingly addressed. Second, by reviewing over a few dozen awards, the article identifies that the concept of ‘legitimate expectations’ in international investment law cannot continue developing as it has in the last few decades without a shift in how damages are awarded.

I. Introduction to the Spanish financial crisis

The Eiser v. Spain case is an old tale with a new twist. Like many other countries, Spain introduced at the end of the 90s a package of regulatory amendments and financial incentives to promote renewable energy in the country. Spain sought to promote a specific type of solar power through state subsidies. The new regulatory regime was an attempt to realize the spirit of the 1992 Framework Convention on Climate Change, the 1994 Energy Charter Treaty, and the 1997 Kyoto Protocol, as well as the 2001 European Union policy for reducing greenhouse gasses by means of developing renewable energy in the region.

Starting in 1998, Spain issued a series of decrees to promote renewable energy by means of a guarantee for long-term supply of energy (Eiser, ¶¶ 105-107). In the long term, however, these incentives were not sustainable. Spain was not left alone in the crusade to advance renewable energy throughout the 90s. In the first weeks of 1991, the Committee on Science and Technology of the European Parliament made an aggressive call to move towards renewable energy on the basis of the results of the Brundtland report—document recounting the world’s critical environmental problems. In the Committee’s view, “[e]very effort should be made to develop the potential for renewable energy which could form the foundation of the global energy structure during the 21^st century” (p. 2, ¶ 5).

Answering this call, Spain issued Royal Decree 661/2007. The International Energy Agency viewed the Decree as seeking “[t]o contribute to Spain’s efforts to achieve its 2010 national target for the promotion of electricity from renewable energy under EC Directive 2001/77/CE.” More than an effort, several investors—including Eiser (¶¶ 357-35)—argued that Decree 661/2007 was a public policy creating an attractive environment for investment (Charanne v. Spain, ¶ 515).

This Decree was later retracted by Spain, causing significant harm to energy projects. The rationale behind a state’s incapacity to unreasonably and unfairly backtrack from a commitment is quite logical. In the same way that a person would not purchase a car for a price other than the one advertised, investors expect to receive the benefit of protective guarantees or commitments that the government has made to them. UNCTAD estimates that the Spanish Decree has been at the heart of over 40 investment arbitrations. This  number keeps growing as investors submit new ISDS cases arising from Spain’s retreat from its regulatory commitments to the energy sector (See, WOC v. Spain).

A few years into the application of Decree 661/2007, the Tribunal in Eiser recounts that Spain became concerned due to a “tariff deficit.” The tariff deficit is generally understood as “[t]he financial gap between the costs of subsidies paid to renewable energy producers and revenues derived from energy sales to consumers” (¶ 124). To combat this deficit, in December 2012, the Spanish Parliament imposed a 7% tax on the total value of the energy fed into the national grid. This tax eliminated the subsidies for renewable energy. This mechanism was followed by several other decrees that blunted the financial incentives and subsidies for foreign investors. In June 2014, the government dealt the final blow to Decree RD 661/2007 through a ‘ministerial order’ setting up a whole new regime for existing power plants (the “Disputed Measure”).

The Eiser case arose from a so-called ‘failed’ investment in the solar power sector in Spain (¶¶ 94-95). During the arbitration, Eiser valued its investment through the Discounted Cash Flow (“DCF”) method at €124.3 million (¶ 136). The relevance of damages in investment arbitration is a growing concern in the field. Certain circles claim that arbitral tribunals are awarding increasingly larger claims in favor of the investor. On this, the International Institute for Sustainable Development mentions that its “[r]esearch has found over 50 known cases in which an investor–state tribunal has awarded a foreign investor over USD 100 million in compensation. In at least eight claims, the award reached over USD 1 billion” (¶ 14).

The energy generation mechanism promoted under Decree 661/2007 was allegedly categorized as a renewable source. The new policies materialized in Decree RD 661/2007. The Arbitral Tribunal considered that the Decree contained a myriad of key elements, including: (i) a guaranteed “priority of dispatch” into Spain’s grid subject to certain conditions; and (ii) it allowed energy producers to adopt different tariffs for production in accordance with the specific characteristics of each project (¶ 112).

Eiser brought an investment arbitration under the Energy Charter Treaty as the general partner of a limited partnership, claiming that it invested in reliance on Royal Decree 661/2007 and its stabilization clause¾which restricted Spain from amending its regulatory framework (¶¶ 357-358). Eiser’s business model depends on identifying low-risk investments in public infrastructure.

The State’s decision to retract Decree 661/2007 had a devastating effect on several companies. Although many foreign investors had access to bilateral investment treaties or the Energy Charter Treaty, locals recount that Spain’s decision to retroactively backtrack from the Decree 661/2007 (the Disputed Measures) “[l]ed to […] the total ruin of 62 000 families […] and that their opportunities to file claims before the Spanish Government have been restricted, given that they did not have access to more impartial, international courts.” This was not the case for Eiser, which enjoyed its position as a United Kingdom incorporated company with the capacity to bring an investment arbitration against Spain.

II. The Tribunal’s Award

Against this backdrop, the Eiser Tribunal was tasked with reviewing if the Disputed Measures breached the fair and equitable treatment (“FET”) standard. Embarking on this task, the Arbitral Tribunal had to address the State’s capacity to regulate¾a heavily disputed point in international investment law (Isolux v. Spain, ¶¶ 409-426; Novenergia II v. Spain, ¶ 542-697). Although the nuances to identify a breach of legitimate expectations may be contested, arbitral tribunals have generally considered the following points (¶¶ 369-372):

• Whether the investor’s expectations were legitimate, reasonable or fair, and not based on subjective considerations.
• The reliance of the investor on such expectations when making its an investment.
• The State’s unilateral conduct to the detriment of the legitimate expectation.
• The existence of a damage to the investor.

Following the steps of Parkerings v. Lithuania, the Tribunal in Eiser departed from the premise that “[a]bsent explicit undertakings directly extended to investors and guaranteeing that States will not change their laws or regulations, investment treaties do not eliminate States’ right to modify their regulatory regimes to meet evolving circumstances and public needs” (¶ 369). Accordingly, FET itself does not paralyze the regulatory powers of the governments.

III. Overlapping principles for determining the margin of wrongfulness

An award on quantum must flow from a prior determination of liability, granted through the same award or in a prior stage. Arbitrators sometimes split their decisions on liability and quantum into different awards, but these components may very well be included in the same award. Several breaches have been claimed in the Spanish arbitrations, but a notable one is the myriad of claims concerning the FET provision. Tribunals have considered that FET may spin off in several protections, including a violation of legitimate expectations, lack of transparency, lack of due process, and arbitrariness (Electrabel v. Hungary I, ¶ 7.74.; PV Investors v. Spain II, ¶ 565; RREEF v. Spain II, ¶ 260; Operafund v. Spain, ¶ 524).

EDF v. Romania, an important case on the analysis of a claimant’s legitimate expectations, explained that a mere regulatory instability does not amount to a breach of a FET provision (¶ 217):

The idea that legitimate expectations, and therefore FET, imply the stability of the legal and business framework, may not be correct if stated in an overly-broad and unqualified formulation. The FET might then mean the virtual freezing of the legal regulation of economic activities, in contrast with the State’s normal regulatory power and the evolutionary character of economic life.

The Eiser Tribunal considered that Spain’s withdrawal of Decree 661/2007 amounted to a breach of the FET standard and the investor’s legitimate expectations. The Tribunal notably awarded full compensation for all losses to the project attributable to Spain’s retraction of Decree 661/2007. By applying a Margin of Wrongfulness, the Tribunal should have discounted from the award losses attributable to the portion of Spain’s measure that would not have amounted to an arbitrary abrogation of Decree 661/2007. In other words, if not all amendments to the regulatory framework are abusive, then the financial gap between the measure that Spain could have taken without breaching FET and the measure that, if applied, would be illegal constitutes the Margin of Wrongfulness.

Reduction of damages in findings where a government disregards an investor’s legitimate expectations is not uncommon. In MTD v. Chile, the Arbitral Tribunal found a breach to the FET standard by considering that Chile had induced a legitimate expectation that a real estate project would be feasible, even when the acquired land could not be developed for commercial purposes (¶ 217). In what has been called a Solomonic decision (Potestà, pp. 38-39), the Tribunal reduced by 50% the damages awarded to the investor due to its failure to conduct an independent assessment of the land.

As such, not all damages need to be compensated. The Margin of Wrongfulness would imply the existence of an ‘unjustified damage’ standard (Muhammad, p. 108) and not a mere reduction in the value of the investor’s investment (Wöss, ¶ 9-10). The Eiser Tribunal unfortunately missed this aspect. Incoherently, it considered that states are not confined to a straitjacket when amending prior legislation, without analyzing the resulting effect on damage quantum. The Eiser Tribunal is not alone in this omission. In the Renergy case, the Tribunal, by majority, invoked “[w]ell-established arbitral case-law” holding “that even in the absence of any specific commitment, Article 10(1) ECT does protect investors against legislative changes that exceed a (wide) acceptable margin” (¶ 642). The recognition of an exceeding acceptable margin might justify the existence of a Margin of Wrongfulness.

The Eiser case is also enlightening as to the impact that a fact witness might have on damages. In most awards, the decision on quantum is driven by the technical viewpoints of experts. In Eiser, the Tribunal reflected that during the arbitration hearing, representatives of the investor opined on the weight that the Disputed Measure had on the investment:

In response to the Tribunal’s question at the Hearing, Mr. Meissner, a founding partner of Eiser, drew a distinction between the changes in Spain’s regulatory regime and other regulatory situations where regulators might “tinker a little bit with the returns.” In contrast, he deposed that “here we had a complete value destruction. We lost all value in this particular project.”

The factual witness’ statement suggests that “tinker[ing] a little bit with the returns”—i.e., the Margin of Wrongfulness—would be acceptable. This implies an additional layer of analysis that exceeds the scope of this article, where the Margin of Wrongfulness might also be justified since an investor is expected to value its investment considering the risks of the project. This is relevant, because under several methods to quantify damages in arbitration a discount rate needs to be applied to the future cash flows that an investor might expect. The ICCA Task Force on Damages considers than an investor’s “[u]nbiased cash flows are expected (average) cash flows—not the cash flows that investors may hope for if everything goes well”.

The Margin of Wrongfulness does put into question some of the premises on which the law of state reparation has been built. Through state responsibility doctrine, investors expect to receive compensation after a violation of international law, which “[s]hould reflect all financially assessable damages” (Marboe, ¶ 3.289). This principle is derived from the acclaimed Factory at Chorzów case, where the Permanent Court of International Justice settled that “[r]eparation must, as far as possible, wipe out all the consequences of the illegal act and reestablish the situation which would, in all probability, have existed if that act had not been committed” (Chorzów, PCIJ 1927, p. 47). The decision in Chorzów has become a cornerstone of the law on damages and is widely cited by tribunals, including in the Eiser case (Eiser, ¶ 421). The Margin of Wrongfulness brings to light some potential deficiencies in the development of damages in ISDS since the Chorzów case.

While the Margin of Wrongfulness questions the legality of the Disputed Measure, it could also be applied by questioning the legitimate expectation of the investor to receive damages. Reparation as understood in Chorzów is built on two premises: (i) the existence of an illegal act, and (ii) the reestablishment of the situation had the act never been committed (Amoco v. Iran, ¶ 191-195). The first premise is compatible with the Margin of Wrongfulness since it implicitly recognizes that damages cannot be awarded for a lawful act, subject to certain nuances on the type of breach committed (e.g., lawful expropriation still requires compensation). The second premise, however, is more questionable. Compensation should restore the injured party not to their pre-breach position, but to the position they would have arrived at had a ‘lawful’ measure been enacted. Restoring the injured party to the pre-breach position presumes that the State’s measure is unlawful in its entirety—but this may not necessarily be true. Arguably, the Tribunal’s recognition of the Margin of Wrongfulness and the Chorzów case invites discussion on how the premises of international damages should be reconsidered or disregarded by future arbitrators in cases such as Eiser v. Spain.

If the Margin of Wrongfulness is accepted as a legal premise then the arbitrators would need to identify the lawful-scenario of the State’s measure. This would require tribunals to heavily engage in hypothetical factual and financial situations, which ISDS tribunals are familiar with. Most investment arbitrations necessitate that arbitrators identify a “but-for” scenario, derived from the principle of causation in international damages. According to the Brattle Group, one of the most active quantum experts in ISDS cases, “[c]ausation requires the careful construction of a counterfactual or ‘but-for’ world that eliminates only the conduct at issue but retains all relevant features of the actual world. The construction of the but-for world is necessarily hypothetical and must remain internally consistent”.

Assessing the Margin of Wrongfulness by considering a potential hypothetical scenario is then inherent to the ordinary course of damages identification conducted by tribunals. This alternative factual situations are also grounded as a matter of law. In the 1987 Amoco decision before the Iran-US Claims Tribunal, Judge Bower separately opined that once liability is found with a degree of certainty, then quantification may be performed with “[t]he best available evidence, even though this process be inherently speculative” (n. 142, ¶ 26). An alternative evaluation of the Margin of Wrongfulness would then be compatible with the perspective taken by some stakeholders.

IV. Conclusion

Eiser v. Spain underscores the intricate nature of investment disputes and the importance of balancing a state’s regulatory authority with investor protections. Recognizing the Margin of Wrongfulness and considering how this concept impacts damages may contribute to a more nuanced approach in future arbitration cases, fostering a fair and equitable resolution for all parties involved. It would also incentivize uniformity in cases where a state, such as Spain, is faced with a wave of investment claims. Moreover, achieving certainty as to the standard on which an investment tribunal should award damages might also function as a prevent mechanism of control for States when enacting amendments to their international commitments.

*Galo Márquez is an Associate in the International Arbitration practice at Creel, García-Cuellar, Aiza y Enríquez, and a Member of the Academic Forum on ISDS before UNCITRAL. He is also a Business Law Professor at Tec de Monterrey.

Cover image credit 

The Civil Liability of Arbitrators: A Transition from Absolute to Qualified Immunity in the United States

Apr 11, 2024 | HILJ-HIALSA International Arbitration Collaboration, Online Scholarship

José Ramón Villarreal Martínez*

I. Introduction

Several jurisdictions have recorded a rise in lawsuits against international arbitrators and arbitral institutions in national courts (p.13).^[1] These cases are occasionally unfounded and may be initiated by disgruntled parties who are dissatisfied with the outcome of an award. They may attempt to file their claims as breaches of public policy or acts of bad faith. This trend has caused a reputational crisis for international arbitration.

The issue is significant because the UNCITRAL Model Law does not specifically cover the matter of arbitrator liability. As a result, each jurisdiction has taken a different approach addressing this issue. In common law countries,^[2] arbitrators are granted the same immunity as judges.

In contrast, in jurisdictions that follow a civil law tradition, the role of the arbitrator is considered sui generis. This is because the arbitrator is seen as both a professional service provider and as someone who performs a jurisdictional function akin to a judge. As a result, the arbitrator has certain contractual rights and obligations towards the parties involved. Additionally, the arbitrators are protected by a system of qualified immunity, meaning that they can only be held liable in serious cases, such as when there is gross negligence, fraud, or bad faith.

Nevertheless, even within common law jurisdictions, there is a lack of consistency in the degree of immunity granted. For instance, in the U.S., the arbitrator enjoys absolute immunity, thereby shielding them from civil responsibility claims, even in instances involving fraud, carelessness, or bad faith. In contrast, in England, the arbitrator’s immunity is not absolute: it does not apply if the arbitrator acts in bad faith.

The problem gained significance after a recent news event where the Paris Court of Appeals annulled an award after it was discovered that an arbitrator had publicly acknowledged a personal acquaintance with one of the lawyers involved in the arbitration process, in a eulogy he published in the prestigious French publication Dalloz. The annulment occurred because of a breach of disclosure, which led to concerns about the arbitrator’s impartiality and independence. In qualified immunity jurisdictions,^[3] the violation of contractual obligations (failure to deliver a timely award, breaches of the duty of disclosure, and excluding an arbitrator from the deliberations) has been ground to impose liability on the arbitrator.^[4]

In contrast, in Grupo Unidos por el Canal S.A. et al. v. Autoridad del Canal de Panama, the U.S. Court of Appeals for the Eleventh Circuit, while analyzing the possibility to vacate an award due to an alleged breach of the duty of disclosure by the arbitrators, stated that “[it] is little wonder, and of little concern, that elite members of the small international arbitration community cross paths in their work . . .  [w]e refuse to grant vacatur simply because these people worked together elsewhere.”

An argument can be made that absolute immunity is a potential solution to avoid frivolous claims from being brought against arbitrators. However, this legal doctrine fails to provide arbitrators with the motivation to perform their duties diligently and cautiously. This has resulted in arbitrators avoiding liability, even in cases where they have acted negligently, to the detriment of the parties and the reputation of arbitration.

From my perspective, the United States must shift from absolute to qualified immunity. This move is crucial because it safeguards the arbitrator’s position in making decisions and offers a recourse for parties involved in situations when the arbitrator acts dishonestly, engages in fraudulent behavior, or displays gross negligence. This transition would establish a liability framework that protects the arbitrator’s judicial function and, at the same time, protects parties against arbitrators’ wrongdoing.

The experience in jurisdictions that have implemented qualified immunity shows that only a few successful cases where arbitrators have been held accountable have been reported. For this reason, an acknowledgment of some level of liability of arbitrators contributes to the high standards of quality that are expected from an arbitration procedure.

II. Arbitrator Liability Regimes

The doctrine of judicial immunity, which originated in England in the 17th century in the cases of Floyd v. Barker and The Marshalsea, has been adopted by common law jurisdictions. However, the degree of immunity adopted by each of them is different.

The doctrine of judicial immunity states that judges are not legally liable for any potential harm resulting from their judgments. The purpose of this doctrine is to uphold the reliability of the judicial system by allowing judges to render decisions without undue pressure by the parties. In the U.S., arbitrators are granted absolute immunity; in England, this immunity allows an exception in cases of bad faith.

In contrast, civil law countries acknowledge that arbitrators carry out their role through a contract, functioning as professional service providers. To safeguard the arbitrator, these jurisdictions have established a sui generis approach^[5] that acknowledges both the judicial role of the arbitrator and his contractual duties.

The sui generis approach has established a form of qualified immunity, whereby the arbitrator is protected from liability for his jurisdictional role, while also safeguarding the parties involved from any unjustified infringements that arbitrators may commit, which could be considered as contractual violations.

III. The Liability of Arbitrators in the U.S.

Within legal systems based on common law, the principle of judicial immunity extends to arbitrators and other individuals who carry out adjudicatory duties. In the U.S., the doctrine of absolute immunity for arbitrators was initially acknowledged in Jones v Brown and further affirmed by the Supreme Court in Butz v Economou. This doctrine provides arbitrators with absolute immunity from legal claims, even in situations of extreme carelessness, and intentional dishonesty.

While the rationale for granting absolute immunity to arbitrators is based on their adjudicatory role, it is indisputable that there are more disparities than similarities between judges and arbitrators, as Pierre Lalive suggests:

One should hesitate to assimilate the position of the arbitrator to that of a judge. In any case, the reasons seem obvious to exclude an assimilation, and even an analogy, between a judge and arbitrator. First a judge is in no way chosen by the agreement of the parties (…) Secondly, when exercising their judicial function, State judges exercise power authority conferred by the State and in its own name.

To sum up, the differences between the function, activity, position and status of a judge on the one hand, and those of an arbitrator on the other, are so great that no sufficient analogy can be drawn between the two which can possibly justify the immunity of the arbitrator.

Consequently, arbitrators bear a closer resemblance to professional service providers rather than judges. For this reason, an absolute degree of immunity fails to motivate arbitrators to adhere to the utmost standards of care and thoroughness.

The absolute immunity approach is problematic for its failure to acknowledge the contractual nature of the arbitrator’s role. Also, it disregards that professionals from analogous fields may be held accountable for civil liability if they incur in a breach of contract.

As Lorena Malintoppi declares:

There is no question that arbitrators and arbitral institutions should be held liable if they commit gross negligence, or act in bad faith. Needless to say, arbitrators are bound to act fairly, to respect due process and the integrity of the proceedings, to ensure the efficiency of the process, and avoid delays. It is also universally accepted that arbitrators have the duty to be impartial and independent and to disclose for the duration of the arbitral proceedings any facts or circumstances that may put into question their capacity to decide a dispute independently and impartially.

Furthermore, “[n]ot only does absolute immunity yield bad results as a matter of policy, but the doctrine also rests on shaky legal foundations . . . [d]espite its dubiousness, the doctrine or arbitral immunity has gone largely unquestioned.” Notwithstanding these concerns, most courts in the U.S. have blindly adhered to it.

From my perspective, it is necessary for the U.S. to transition from the absolute immunity doctrine to qualified immunity. Under this approach, the arbitrator would be allowed absolute immunity for his adjudicatory role but would also be considered a professional service provider who may be held accountable for negligence, bad faith, or misconduct.

IV. The Transition to Qualified Immunity

Although the idea of absolute immunity of arbitrators has been widely accepted in the U.S., the California Court of Appeal in Baar v Tigerman deviated from this doctrine and refused to apply it to an arbitrator, who had failed to deliver a timely award.

While the absolute immunity doctrine is unquestioningly implemented in the U.S., this was the first instance where a court specifically highlighted that arbitrators are immune from liability in their jurisdictional role, and that the refusal to issue an award is separate from the decision-making process. While American case law has firmly established the application of the absolute immunity doctrine for arbitrators, this precedent has sparked research and debate over the possibility to transition from absolute to qualified immunity, a discussion that has been abandoned in recent years.

In addition, debate over arbitrators’ civil liability is crucial for effective arbitration, since absolute immunity may hinder diligence and good faith. It is also a bad policy to shield people who willfully engage in bad faith or gross negligence, such as not delivering a timely award or disclosing conflicts of interest. As Susan Frank suggests, “[o]verly broad immunity fails to create an incentive for arbitrators to be responsible for their actions, to the parties who are paying fees, or to the integrity of the international arbitration system.”

For that reason, other common law countries such as England have transitioned from the absolute immunity doctrine towards a qualified approach, by recognizing in Section 29 (1) of the Arbitration Act of 1996 that “[a]n arbitrator is not liable for anything done or omitted in the discharge or purported discharge of his functions as arbitrator unless the act or omission is shown to have been in bad faith.” Also, in England, an arbitrator is immune in all the activities that involve a judicial function; however, “this immunity . . . [does not] have anything like the same force when applied to professional men when they are not fulfilling a judicial function.”

Although absolute immunity is conceded to arbitrators as a public policy measure, absolute immunity should not protect arbitrators when they voluntarily engage in bad faith or other misconduct. The reason for concern lies not only in its impact on the arbitration’s outcome, but also in its detrimental effect on the reputation of arbitration as a viable alternative to court litigation.

V. Reasons to Advocate for Qualified Immunity

Even though other common law jurisdictions have transitioned to qualified immunity, I view Baar v Tigerman as a precedent that has the potential to ignite the discussion on qualified immunity. This is because qualified immunity acknowledges the contractual nature of an arbitrator’s appointment, wherein the arbitrator assumes rights and responsibilities by consenting to deliver a fair and enforceable award. Additionally, qualified immunity also safeguards arbitrators when they carry out judicial duties, while ensuring that they are held responsible for their lack of care in fulfilling their contractual obligations.

There is a concern that if qualified immunity is adopted as a matter of public policy, because arbitrators, unlike judges, are vulnerable to “(1) unhappy parties [that] might threaten arbitrators, or (2) arbitrators might not make principled decisions if they are concerned about being sued,” nevertheless, these policy justifications are insufficient to support absolute immunity; however, there is evidence in that qualified immunity will not threat the impartiality or independence of arbitrators, since “the number of successful cases brought against arbitrators and institutions is limited. A French study (…) identified five cases since 1804 where arbitrators were found liable by the French courts: one for untimely resignation, twice for lack of independence and impartiality, and twice for excessive delays.”

In England, since the Arbitration Act of 1996 “there have been no reported English cases which have interpreted the bad faith requirement in Section 29 . . . Bad faith is a deliberately high threshold, and this carve-out seeks to strike the balance between immunity and permissible recourse for parties in respect of egregious arbitrator behaviour.”

Thus, based on the experience of France and England, it can be inferred that qualified immunity does not pose a risk to the impartiality and independence of arbitrators; rather, it supports these qualities because it imposes a “balance that addresses the courts’ dual concerns: protecting the public from possible arbitrator abuse and providing arbitrators with immunity to ensure independent decision making . . . [q]ualified immunity would hold arbitrators accountable when arbitrator.”

VI. Conclusion

The advantage of transitioning from absolute to qualified immunity is that the latter recognizes the sui generis legal relationship that exists between the arbitrator and the parties, where the arbitrator is chosen by a contract to perform an adjudicatory function. In addition, qualified immunity safeguards arbitrators when they carry out judicial duties, while also ensuring that they are held responsible for any wrongdoing committed in bad faith or negligence.

A flaw of the absolute immunity approach is that it ignores that the parties, when they chose an arbitrator, have a reasonable expectation that the arbitrator will remain impartial, independent and that he will fulfil his duties (contractual and adjudicatory) in good faith. Moreover, “their acceptance of the arbitral risk did not cover the case of fraud, of corruption, nor (it would seem), cases of gross and inexcusable negligence. But it did cover, or include the possibility of mistakes in law and legal procedure . . . there is no justification for the immunity of the arbitrators, especially when it is based on the misconceived assimilation to the status of judges.”

Another critique of absolute immunity is that it grants protection to arbitrators based on a misunderstood policy argument to protect them from undue pressure from the parties, and to provide them legal certainty that they will be immune from civil liability claims. Even when the reason behind this argument is true, and that it is undeniable that without some degree of immunity fewer professionals would accept to be appointed as an arbitrator, there is no justification to shield arbitrators in cases of bad faith or gross negligence.

As Dario Alessi suggests:

No law of contract would allow the gross unfairness of exempting a party to a contract from liability because of some policy argument. In particular, the independence of arbitrators may not be obtained at the expense of fairness, producing impunity for breach of obligations that the arbitrators have freely assumed. (…) The unpleasant effects of liability of arbitrators which may occur, such as a risk of vexatious litigation, collateral disputes or harassing lawsuits, cannot as such justify the denial of the right to enforce the promises of arbitrators.

In addition, it is also a good policy to provide the parties of an arbitration a remedy in case that the arbitrator acts in bad faith (such as deliberatively not disclosing conflicts of interest) or in cases of fraud or gross negligence. At the international stage the trend is to improve the current standards about impartiality and independence applicable to arbitrators, evidence of this tendency is the recent publication of the IBA Guidelines of Conflicts of Interest in International Arbitration 2024, that was updated to incorporate “the best current international practice . . . [t]he General Standards and the Application Lists are based upon statutes, practices, and case law and other decisions in a cross-section of jurisdictions, and upon the judgment and experience of the main participants in international arbitration.”

While international arbitration is imposing stricter standards to arbitrators, the U.S. lags behind the international trend by granting explicitly immunity to arbitrators that fail to disclose conflicts of interests, as it is recognized in Section 14 of the Uniform Arbitration Act. There current approach in the U.S. towards the liability of the arbitrator fails to  “balance the various interests of parties, counsel, arbitrators, and arbitration institutions, all of whom have a responsibility for ensuring the integrity, reputation, and efficiency of international arbitration.”

To promote the adoption of qualified immunity in the U.S., I consider it necessary to adopt qualified immunity through legislation at state level, and to implement it in the Uniform Arbitration Act, since it does not seem likely^[6] that the U.S. Supreme Court will hear a case that modifies the existing status quo.^[7] As M. Rasmussen says, “because large sophisticated parties would not forego their access to national courts without carefully exploring the advantages and disadvantages of the process, they must perceive that the advantages of arbitration outweigh the disadvantages.”

In addition, the transition to “contractual liability, would increase transparency, accountability, independence, impartiality, and integrity in the arbitral process. Henceforth, international arbitration would enjoy greater recognition and public confidence,” a transition to qualified immunity will help arbitration to overcome its actual reputational crisis.

*Facultad Libre de Derecho de Monterrey (LLB), Escuela Libre de Derecho (LLM), University of Southern California, Gould School of Law (LLM), Harvard Extension School (ALM Government candidate). The author specializes in civil, commercial litigation, and commercial arbitration in Mexico. I’m grateful to Roberto Cuchí Olabuenaga for his comments in early drafts, and to the members of the editorial board of the HILJ-HIALSA for their work in preparing this piece for publication. All errors are my own.

Cover image credit 

In Defence of the Impartiality of Barristers and Door Tenants in ISDS: A Call for Departure from Hrvatska Elektroprivreda d.d. v. Republic of Slovenia

Mar 29, 2024 | HILJ-HIALSA International Arbitration Collaboration, Online Scholarship

*Batuhan Betin

To venture so brazenly as to critique an order promulgated by the most distinguished of tribunals, comprised of Jan Paulsson, Charles N. Brower, and David A. R. Williams QC (as he was then), presiding, may be comparable to an act of sacrilege. Alas, heresy is precisely what the tribunal’s Order, dated May 6^th, 2008, in Hrvatska Elektroprivreda d.d. v. Republic of Slovenia provokes. In classrooms, misinformed articulations of the case garner polarizing responses from students. It prompts those educated in the institutions of the British Isles (and equally, the Commonwealth) to react viscerally, raising their hands in hopes of being called upon to clarify the particularities of the seemingly alien system of barristers’ chambers—much to the amusement of their contemporaries hailing from “enlightened” civil law jurisdictions who stubbornly refuse to distinguish barristers from the lawyers of their native lands.

Despite its notoriety, the case has attracted limited academic commentary.

At the crux of Hrvatska Elektroprivreda d.d. v. Republic of Slovenia was a list communicated on April 25^th, 2008, tabulating the members of the Respondent’s representatives who would be attending the  inaugural arbitral hearing scheduled a fortnight from then (¶ 3). Amongst the constituents of the list was Mr. David Mildon QC (as he was then) (¶ 3). Mr. Mildon KC (as he is now) was, and still is, a barrister operating out of Essex Court Chambers, in London (¶ 3). Mr. Williams KC (as he is now) was, at the time, a “door tenant” operating out of the very same set of chambers (¶ 3). The Claimant’s representatives declared that they had not been aware of Mr. Mildon KC’s retainer before the above-mentioned letter (¶¶ 4-5). They promptly requested disclosure of Mr. Mildon KC’s affiliations with the presiding arbitrator (¶ 5), and subsequently petitioned for Mr. Mildon KC to be removed on grounds that Mr. Williams KC could not ‘be relied upon to “judge fairly”’ (¶ 15).

The tribunal evaluated the interrelationship between Mr. Mildon KC and Mr. Williams KC and found that there was justifiable “apprehension of partiality” (¶ 31).  It ordered that “Mr. David Mildon QC may not participate further as counsel in this case” (¶ Ruling). The Tribunal’s decision to order the removal of Mr. Mildon KC was inspired by a medley of considerations. The seemingly most decisive and foundational consideration was the Tribunal’s finding that there was a justifiable “apprehension of partiality” because “the London Chambers system is wholly foreign to” the Claimant (¶ 31). Admittedly, this was bolstered by the tardy timing of the disclosure of Mr. Mildon KC’s retainer (only two weeks prior to the first hearing). Without reserving a firm position as to whether this case was wrongly decided, this article seeks to challenge the precedential value of this Order vis-à-vis future cases where members of the same chambers are appointed as arbitrators and counsel. Even the most authoritative of arbitral treatises, much (presumably) to their editors’ chagrin, cautiously cite this Order as a persuasive authority that the presence of both members serves as a basis for a serious risk of impropriety (Redfern and Hunter on International Arbitration (7th Edition), ¶ 4.138).

This article seeks to deconstruct two critical features of the Order. First, it shall seek to prove that, in hindsight, the Tribunal applied an erroneous subjective standard in discerning whether there was a risk of an appearance of partiality in contradiction to Article 14(1) of the Convention on the Settlement of Investment Disputes between States and Nationals of Other States. Secondly, it advocates that the Tribunal could have more precisely characterized and investigated the interrelationship between Mr. Mildon KC and Mr. Williams KC. In particular, it seeks to demonstrate that the position of “door tenant” may be distinguished from that of a traditionally tenanted member of chambers. Consequently, the particularities of the former may mandate that in future disputes, tribunals should be cautious when applying principles generally applicable to “inter-barrister” relationships mutatis mutandis to “barrister-door tenant” relationships. To this extent, it seeks to discourage blanket recourse to the present Order.

The Tribunal applied erroneous standards in assessing the appearance of partiality. As previously mentioned, it articulated firstly, and seemingly decisively, that the Claimant had a justifiable “apprehension of partiality” because “the London Chambers system is wholly foreign to” them (¶ 31) (emphasis added). This assessment is problematic in that it endorses a subjective evaluation of the arbitrator’s partiality. It essentially adopts the position articulated by the Claimant’s representatives, that whereas the interrelationship “may not be cause for concern in London . . . [v]iewed from the Claimant’s cultural perspective, such concerns are justified” (¶ 10) (emphasis added). However, the prevailing view amongst ICSID tribunals today is that the independence of an arbitrator must be assessed, objectively, on the basis of “whether a reasonable third party, with knowledge of all the facts, would consider that there were reasonable grounds for doubting that an arbitrator possessed the requisite qualities of independence and impartiality” (Schreuer’s Commentary to the ICSID Convention p.1570 (emphasis added); see also Suez and Vivendi v Argentina ¶ 78; EDF v Argentina ¶¶ 109-111). A negative formulation of the test can also be articulated as follows: “a reasonable and informed third party would find it highly likely that the arbitrator’ lacked independence and impartiality” (p.1591; see also Caratube and Hourani v Kazakhstan ¶ 90). The Tribunal deviates from objectivity in that it assesses partiality exclusively from the subjective eyes of the particular Claimant to the dispute, as opposed to the objective lenses of the reasonable third party.

The Tribunal also failed to consider whether the Claimant could have become familiar with this purportedly ‘wholly foreign’ system through knowledge of all the relevant facts. Indeed, this is perplexing given that the Tribunal unequivocally acknowledged earlier in the Order that “[b]arristers are sole practitioners . . . [t]heir Chambers are not law firms” (¶ 17) (emphasis added). Whereas the Tribunal did caveat this statement by cautioning that “th[e] practice [of barristers] is not universally understood,” (¶ 18) it nonetheless tacitly acknowledged how readily accessible this information was. For instance, it pointed to the website of Essex Court Chambers in which unambiguous wording declared, “Chambers is not a firm, nor are members partners or employees. Rather, Chambers contains the separate, self-contained offices of individual barristers each self-employed and working separately” (¶ 17) (emphasis added).

Nowadays there is even greater transparency regarding this phenomenon. The Bar Council of England and Wales’ 2014 Information Note, and its recent 2022 Brochure, underscore that “barristers practicing from traditional sets of chambers are self-employed and are not [emphasis added] in partnership,” and “[b]arristers in the same chambers are fully independent of each other,” respectively. Similar disclaimers are customarily promulgated on the websites of Chambers. In the increasingly globalized world of arbitration, where the participation of barristers is ever-increasing, there is no sound reason precluding the representatives of opposing parties from explaining the realities of this system to their clients.

The tribunal also sought to caveat its earlier statement on the basis that the practice was not “universally agreed” (¶ 18). This proposition is hardly palatable. The independence of barristers is one of the Ten Core Duties promulgated by the Bar Standards Board (BSB) (the body responsible for the ethical regulation of the profession). It is imbued in various rules codified in the readily and publicly accessible BSB Code of Conduct, which, inter alia, underscores that “[m]embers of chambers are not in partnership but are independent of one another and are not responsible for the conduct of other members” (BSB Handbook v. 4.7, gC131). Even the failure to explain to “unsophisticated lay clients [emphasis added]” that “members of the chamber are, in fact, self-employed individuals who are not responsible for one another’s work,” is itself considered a breach of this Code of Conduct (gC56) (emphasis added). Consequently, any arbitrary attempt to doubt the legitimacy of this inviolable and fundamental principle must be readily rejected by future tribunals.

The Tribunal ought to have more precisely characterized the interrelationship between “door tenants” and barristers.

Even the most widely cited of treatises characterize the present matter as a conflict of interest between two barristers who are members of the same chambers. However, Mr. Williams KC was not a traditional member of Essex Court Chambers, nor did he practice as a litigator therein. In relation to Essex Court Chambers, his capacity was that of a “door tenant” (¶ 3). He appears to have retained his practicing certificate from the Bar of New Zealand where he was a traditional tenant of Bankside Chambers. In the absence of knowledge as to the extent and nature of his relationship with the English set, this article does not, per se, reserve a position as to whether the Tribunal erred in treating Mr. Williams KC akin to a traditional member of chambers. Indeed, as elaborated in detail below, there may be circumstances where such treatment may be justified.

The purpose of this Part is to acknowledge that, and identify why, the Tribunal conflated the roles/capacities of traditional members of chambers and that of their door tenants. It seeks then to dissuade other tribunals from blindly conflating the two, in future disputes, in prospective reliance of the current Order.

The Tribunal’s reluctance to distinguish between traditional members of chambers and door tenants should be attributed to the Background Information section of the 2004 IBA Guidelines on Conflicts of Interest in International Arbitration.  Therein, it is asserted (without much elaboration) that for all relevant intents and purposes, the term “members of chambers” shall, because it is “proper,” include door tenants (pp.456-457). However, the Background Information fails to provide any reason as to why the conflation is necessary or proper.

In practice, the relationship of a door tenant vis-à-vis a traditional member operating out of a set of chambers is not, per se, similar to the latter’s relationship with their contemporaries. One set of construction law barristers distinguishes the former as follows: “A number of individuals who are classed as ‘door tenants’ of Chambers also use the clerking and administrative services of ACL. Door tenants are not members, but the courtesy of displaying their name at the entrance to Chambers has been extended to them” (emphasis added).

Door tenants will likely have other permanent occupations. They may be career academics; they may be “affiliated to, and practice out of, another chamber”; they may even have vocations as lawyers employed full-time in traditional law firms operating out of “foreign” jurisdictions. The roles of a door tenant may also be vastly different vis-à-vis traditional members. Some may perform purely consultative functions. Whereas, admittedly, others will take on casework. One regional set of chambers describes the function of their door tenants as follows: “door tenants . . . are available . . . as advisers and consultants and they take on casework which is particular to their field of expertise.” Evidently, one must not blindly conflate the two. Assessment must be made on a case-by-case basis, having regard to the specific nature and extent of the door tenant’s affiliation. Otherwise, one risks arbitrarily mischaracterizing the probability of partiality.

In finding a justifiable appearance of partiality, the Tribunal placed great emphasis on the basis that the promotional materials disseminated by English chambers created a real sense of partnership and collective association amongst their traditional members (¶¶ 17-18). To this end, the Tribunal opined, it was appropriate to treat barristers, in respect of their perception by litigants, akin to solicitors employed in traditional law firms (¶ 19). However, surely, this litmus test can have a countervailing effect. The proactive efforts by chambers to distinguish between permanent members and door tenants (as exemplified for instance by the abovementioned quote from Atkin Chambers) creates a line of delimitation between the two types of tenants. It distances the latter from the former and conveys an unambiguous message that the character of the latter’s affiliation with chambers is likely to be fundamentally different from that of the former.

This is not to say that a “barrister-door tenant” relationship may never warrant disclosure. For instance, where a career academic (Professor A), who is also a door tenant, receives extensive appointments via chambers, their relationship vis-à-vis an ordinary tenanted member may be more akin to a traditional “barrister-barrister” relationship. There, mutatis mutandis application of principles governing the latter may be entirely warranted. However, it is submitted, that due regard must also be had to the secondary nature of a door tenant’s relationship to chambers. Traditionally, a barrister’s principal, if not exclusive, line of business is his self-employed vocation as a litigator, who operates out of the relevant set of chambers. On the contrary, a door tenant has a principal employer or vocation (which may be their primary source of material income). If the same hypothetical Professor A was instead a “hobby arbitrator,” instructed in less than half a handful of cases per annum, it would surely be less persuasive to treat them as if they were any other tenanted barrister.

The failure to distinguish door tenants from traditional barristers also results in certain perplexing outcomes under the 2014 IBA Guidelines on Conflict of Interests in International Arbitration. Professor A who is a door tenant at Chambers B, but a tenured member of the faculty at University C, would be expected to disclose their relationship with Counsel Y who is a junior tenant at Chambers B (¶ 3.3.2). Nonetheless, they would be exempt from disclosing their relationship with Counsel X who is also a tenured member of the faculty at University C (¶ 4.3.3). In light of Professor A’s financial independence from Counsel Y, it is unclear why the risk of bias in favor of Counsel Y is deemed inherently higher than that vis-à-vis Counsel X. Surely, in this hypothetical, the risk of bias should be greater vis-à-vis Counsel X, as a favorable award could bolster the reputation of the faculty, which may in turn indirectly benefit Professor A.

In the modern age, where information is publicly and readily accessible, no reasonable observer armed with the requisite information pertaining to the ethical regulation of barristers should have a good reason for believing that a door tenant is likely to be partial towards counsel hailing from a shared set of chambers by that connection alone. Hrvatska Elektroprivreda d.d. v. Republic of Slovenia must not be interpreted as promulgating a presumption applicable indiscriminately, and capable of producing dispositive outcomes without consideration of the specificities of each prospective conflict. This case also illustrates the desirability of expanding the 2014 IBA Guidelines on Conflict of Interests in International Arbitration such as to provide greater elucidation as to the susceptibility of door tenants (specifically) towards certain biases, with special attention being held to the secondary nature of their vocational relationship with the set. Future tribunals should and must not rely on the present Order as legitimizing the indiscriminate conflation of the two, potentially vastly different, roles.

Skepticism vis-à-vis the Order has started to emerge outside of the investment arbitration context. In 2012, the London Court of International Arbitration (“LCIA”) Court engaged in perhaps the most “sober” interpretation of the Order. In its Challenge Decision No. LCIA81116 the Court unequivocally found that the mere fact that counsel and arbitrator operate out of the same set of chambers cannot serve as “a [decisive] basis to impose upon him an obligation to disclose the activities of other barristers in his chambers; therefore, non-disclosure of such activities does not give rise to ‘justifiable doubts’” (¶48). It rejected any contention that the collective marketing or promotion of members of chambers could ever usurp this presumption (¶47).  Nonetheless, it cautioned, that the presumption of impartiality amongst barristers is neither inviolable nor irrefutable. On each occasion, it is paramount that one conducts a “fact-based enquiry” into whether that particular relationship, between those two members of the same set of chambers, meets the requisite threshold for an appearance of partiality (¶48). This is a most welcome development. Admittedly, this skepticism may stem from the reality that the LCIA is an inherently English institution. The LCIA Court is constituted predominately of English and common law qualified practitioners (or of those intimately familiar with the English legal system). Nonetheless, the author hopes to see the LCIA Court’s sobering approach proliferate and prevail amongst its more transnational investment arbitration contemporaries.

*Batuhan Betin holds an LL.M from Queen Mary University of London where he graduated first overall in his cohort. He extends his sincerest gratitude to his good friend Ms. María Rosario Tejada for bringing the HILJ-HIALSA Collaboration on International Arbitration to his attention.

Cover image credit