Over the past few years, investor–state arbitration has attracted a number of criticisms, such as the pro-investor tendency, the lack of procedural transparency, and the chilling effect on state regulatory power over public interests. In the context of such criticisms, the EU in late 2015 proposed an Investment Court System (ICS) to bring improvement to investor–state dispute settlement (ISDS) mechanisms that might be incorporated in all ongoing and future EU investment treaty negotiations, including the investment chapter of the Transatlantic Trade and Investment Partnership (TTIP) between the EU and the US. 1
In fact, TTIP negotiation is not the only occasion on which the EU brought forward the idea of ICS. Around the same time, in the investment chapter of the other two negotiated agreements, namely, the Comprehensive Economic and Trade Agreement between Canada and the EU (CETA), 2 and the Free Trade Agreement between the EU and Vietnam (EU–Vietnam FTA), 3 the ICS has also been incorporated. 4 Browsing through the EU’s TTIP proposal and the ISDS provisions of the CETA and EU–Vietnam FTA, several features can be found. First, the ICS is a two-tier system, consisting of a tribunal of first instance (TFI) and an appellate tribunal (AT). 5 A disputing party is allowed to appeal the decision of the TFI on the grounds of mistake of law or facts, in addition to procedural irregularity. 6 Second, the ICS is comprised of arbitrators appointed by special committee with fixed term of office. 7 Individual cases are heard by divisions consisting of arbitrators appointed by the head of the tribunals. 8 Third, procedural transparency is enhanced in the ICS proceedings with the mandatory application (with modification) of the UNCITRAL Transparency Rules. 9
There are also other provisions aiming at improving investor–state arbitration. For example, under CETA, a Contracting Party retains its right to regulate in order to achieve legitimate policy objectives and to protect public interests. 10 Consequently, the Contracting Party’s enactment or amendment of its legislation or regulation cannot be seen as a breach of a treaty obligation, even if a covered investment is negatively affected or an investor’s expectation is hindered. 11 Therefore, a host state’s right to regulate is excluded from the scope of substantive matters that can be challenged under CETA. From a procedural perspective, the introduction of early dismissal procedures and the ‘loser pays all costs’ principle might deter unfounded or frivolous claims. 12 These issues deserve more thorough discussion beyond the scope of this paper. Considering the integrity of arbitrators as an important factor affecting the legitimacy and viability of investor–state arbitration, the following sections will focus on the ICS’s new rules concerning the appointment of arbitrators. In addition, since under the three agreements the provisions concerning the appointment of arbitrators are generally identical, this paper will refer to the text of TTIP in the subsequent discussion.
Rules Regarding the Composition of the Tribunal and the Appointment of Arbitrators
The New Rules in a Snapshot
Like other ISDS procedures under most investment treaties, an investment dispute under the TTIP should preferably be resolved amicably by negotiation or mediation. 13 If amicable resolution is not possible, the investor may request consultations (Ref. 13, art. 4.1). If, after six months, the consultations have not produced a resolution of the dispute, the investor can file a claim with the TFI (Ref. 13, art. 6.1). However, when it comes to the phase of the tribunal being composed with appointed arbitrators, a clear departure from the common practice of parties appointing arbitrators is taken by the ICS.
As proposed by the EU, the TFI will comprise of 15 arbitrators, including five nationals of EU Member States, five nationals of the US, and five nationals of third countries. These arbitrators will be appointed by a special Committee upon the entry into force of the TTIP (Ref. 13, art. 9.2). The appointed arbitrators will hold their office for a fixed six-year term, renewable once, with a monthly paid retainer fee (Ref. 13, art. 9.5, 9.12). The TFI will have a President, appointed by the Committee from among the five arbitrators of third country nationality (Ref. 13, art. 9.8). The TFI will hear individual cases in divisions consisting of three arbitrators, one of whom shall be a national of an EU Member State, one a national of the US, and one from a third country; the arbitrators serving in the divisions will be appointed by the President of the TFI on a rotation basis; only the arbitrator from the third country can chair the division (Ref 13, art. 9.6, 9.7). The disputing parties may also agree to have their case heard by a sole arbitrator of third country nationality; the sole arbitrator shall also be appointed by the President (Ref. 13, art. 9.9). The composition and appointment of the arbitrators of the AT, the appellate branch of the ICS, is basically the same, except that the AT will consist of six arbitrators only (Ref. 13, art. 10).
Note that, under the Convention of the International Centre for Settlement of Investment Disputes (ICSID), the Contracting Parties and the Chairman of the Administrative Council are allowed to designate persons to form a Panel of Arbitrators, 14 but arbitrators can be appointed by the parties from outside the Panel to constitute a tribunal (Ref. 14, art. 40(1)). So, in ICSID proceedings, the disputing parties’ choice of arbitrators is not confined to the candidates designated to the Panel of Arbitrators, as long as the appointees possess the same qualification required by the ICSID Convention (Ref. 14, art. 40(2)).
Streamlined Process of the Appointment of Arbitrators
Under the proposed ICS the disputing parties are deprived of the right to appoint arbitrators. Instead, the President of the TFI/AT is invested with the power of appointment. For cases before the TFI, the division shall be constituted within 90 days of the submission of the claim (Ref. 13, art. 9.7). One possible advantage of this new approach is that the composition process under the ICS is more simplified and efficient, as compared with conventional investor–state arbitration, where the parties are entitled to elect the arbitrators. In ICSID arbitration, for example, if at the time of the registration of the request for arbitration the parties have not agreed upon the number of arbitrators and the method of their appointment, the requesting party may make relevant proposals to the other party, the other party may accept such proposals or make its own proposals; if the other party replies with its own proposals, then it is the requesting party’s turn to decide whether or not to accept such proposals. If no agreement has been reached after 60 days from the registration of the request for arbitration, the appointment procedure of Article 37.2(b) of the ICSID Convention becomes applicable at either party’s option. 15 In this case, each party appoints one arbitrator, and the third arbitrator, as the president of the tribunal, is appointed by agreement of the parties (Ref. 14, art. 37(2)(b)). To achieve this, either party shall nominate two persons, one as the arbitrator appointed, the other as the proposed third arbitrator to be agreed as the president; the other party shall in its reply appoint its arbitrator, and agree to the third arbitrator as president, or propose another third arbitrator, in which case it is the initiating party’s turn to decide whether to accept or not (Ref. 15, Rule 3). Although the term ‘promptly’ frequently appears in the text of the provisions, there is no time limit in this second phase of appointing arbitrators. It only leads to a third stage if the tribunal has not been constituted within 90 days after the dispatch by the ICSID of notice of registration of the request for arbitration; in this phase, either party may ask the Chairman of the Administrative Council of the ICSID to appoint arbitrators or to designate the president of the tribunal for the parties in order to complete the composition of the tribunal, which shall be done within another 30 days (Ref. 15, Rule 4). Accordingly, the constitution of an ICSID tribunal may be completed as late as 120 days after the registration of the request for arbitration. By contrast, the 90-day time limit of composition of the division under the ICS is indeed an improvement.
Problems with the New Rules
The Pro-investor Fallacy
One major criticism of the conventional investor–state arbitration is that, by allowing the investors to appoint arbitrators, a financial incentive is created for the arbitral tribunal to interpret the law and decide in favour of the investors. 16 In other words, it is argued that the asymmetrical claim structure (i.e. only the investor can file a claim for arbitration against the host state, not vice versa) could create incentives for arbitrators to favour the class of parties that is able to initiate the proceedings (i.e. the investors); in addition, due to their career interest, arbitrators could be influenced by the need to appease parties with power over appointment.Reference Van Harten 17 Under such presumption, the abolishment of the parties’ right to appoint arbitrators can correct such bias and eliminate the pro-investor tendency. This is probably the primary policy consideration under the ICS to have the arbitrators pre-installed by the Committee with fixed terms, and the appointment to hear individual cases to be made by the President of the TFI/AT, instead of by the disputing parties. 18
However, the pro-investor allegation is a fallacy that is logically unsound. In the case of a tribunal consisting of three arbitrators, if, owing to financial or career interest concerns, the appointment of an arbitrator by one party could result in the arbitrator’s bias in favour of that party, such an incentive should apply equally to the arbitrator appointed by the investor and the arbitrator appointed by the host state. In other words, if the arbitrator appointed by the investor were to have a financial incentive (for example, appointment in future cases) to please the investor, then the arbitrator selected by the host state might just as well have the same motive to appease the host state. As a result, allowing the disputing parties to appoint arbitrators could create pro-investor arbitrators as well as pro-state ones. However, the third arbitrator, who is to serve as the chair or president of the tribunal, is usually installed by mutual agreement of the disputing parties or the two arbitrators appointed earlier. Under the ‘appointment creating incentive leading to bias in favour of the appointing party’ hypothesis, the joint appointment provides no incentive for the third arbitrator to favour either party, as he or she is not appointed unilaterally by a single party. Consequently, if in a tribunal with three arbitrators there is a pro-investor arbitrator and a pro-state one, the bias of these appointees in favour of their respective appointing party would probably offset each other, and the jointly appointed third arbitrator serving as chair or president of the tribunal could act as a genuine neutral. As for cases heard by a sole arbitrator, the arbitrator is in principle jointly appointed by mutual agreement of the disputing parties, thus the pro-investor or pro-state hypothesis is not applicable.
Furthermore, under the ICS it is likely that the arbitrators might still have a pro-investor or pro-state tendency even though their appointment is not made by the disputing parties. If bias can be generated by financial incentives, it can equally be induced by national identification. A division of the TFI/AT hearing a case under the ICS consists of three arbitrators, one being an EU citizen, the other a US national, and the third a national of the third country. It could be presumed that, for the arbitrator from the EU, there might be a psychological connection, arising out of nationality or identification, with the investor-claimant from the EU, or the responding EU or EU Member State. Similarly, the arbitrator who is a US national might be under affective influence to rule in favour of the US claimant, or the responding US. That is why, under the ICS, the nationalities of the arbitrators are delicately balanced. In addition, the arbitrators are appointed by the President of the TFI, who is a national of a third country, and only the arbitrator from the third country can chair the division. These rules imply that nationality is also a potential threat to the impartiality and independence of the arbitrators. Therefore, to have the disputing parties to blame for the arbitrators’ pro-investor or pro-state tendency, simply because the latter are appointed by the former, is not a fair accusation.
In fact, evidence of empirical statistics suggests that there is no such thing as a pro-investor tendency. In order to avoid ambiguity, a case decided in favour of the investor is defined as the investor’s claim on merits being approved and payment/compensation granted in the final award; on the other hand, a decision in favour of the host state is construed as the investor’s claim being dismissed and/or the host state’s counterclaim being successful in the final award. 19 A preliminary award-finding jurisdiction is not considered as a pro-investor award, as it will not necessarily result in a ultimate decision ruling in favour of the investor’s substantive claim. 19 Under this definition, one can examine the relative percentages of cases decided in favour of the investor or the host state and find that by no means do the investors win more cases than the host states. For example, by the end of 2017 there have been cumulatively 650 cases filed with and administrated by the ICSID. 20 Among these cases, 66% were decided by a tribunal while the other 34% were settled or otherwise discontinued (Ref. 20, p. 13). Of the 66% decided cases, only 48% were ruled in favour of the investor (final awards upholding claims in part or in full); the remaining 52% were against the investor (including claims dismissed in full, jurisdiction denied, and claims found manifestly without legal merits) (Ref. 20, p. 15). According to the above statistics, the pro-investor argument is unfounded, otherwise the relevant statistics should have presented a result showing the investors winning more cases than the host states. In reality, however, the outcome is reversed. The hard evidence undoubtedly rebuts the false allegation of the pro-investor tendency arising from appointment by parties.
The design of the ICS is allegedly to reform the investor–state arbitration mechanism. As far as the instalment of arbitrators is concerned, the new rules under the ICS are supposedly intended to eliminate, by depriving the parties of the right to appoint, the potential bias an arbitrator might have with his/her appointing party. However, the pro-investor allegation has been proven as untrue and non-existent. On the contrary, the appointment of an arbitrator by a disputing party will not necessarily result in the appointee’s bias in favour of the appointing party. So there is nothing wrong with the current practice of allowing disputing parties to appoint their respective arbitrators, except that it could cause delay in arbitral proceedings. In this regard, although the approach of the pre-selection of the tribunal members and the appointment of arbitrators by a neutral of the ICS is indeed new and different from past experience, it should not be seen as an improvement as far as the prevention of an arbitrator’s potential bias is concerned. The justification and necessity of the new rules is therefore in doubt.
The Violation of the ‘Burden of Proof’ Principle
The deprivation of the disputing parties’ rights on grounds of bias resulting from appointment poses another problem. In civil proceedings, as well as international arbitration proceedings, there is the principle of ‘burden of proof’ deciding which party bears the responsibility for proving a particular allegation or contention; the generally accepted rule is that the party making a specific allegation carries the burden of proof of such an allegation. 21 This principle is recognised in many arbitration rules. For example, Article 27(1) of the United Nations Commission on International Trade Law Arbitration Rules (UNCITRAL Arbitration Rules) provides that ‘[e]ach party will have the burden of proving the facts relied on to support its claim or defence.’ 22 However, by depriving the investor of the right to appoint an arbitrator, the ICS seems to build its method concerning the composition of the tribunals/divisions on the presumption, without any proof, that arbitrators will have a bias in favour of the appointing investors. This is in conflict with the principle of burden of proof in civil and arbitration proceedings. The integrity of arbitrators should not be prejudged under the ICS; any lack thereof must be proved by the alleging party.
In conventional investor–state arbitration, where parties appoint the arbitrators, the allegation of a lack of impartiality and independence of arbitrators, in general based on the pro-investor argument, has been proven illogical and unsupported. However, that does not mean that, in a particular case, the parties cannot question an arbitrator’s integrity. The challenge mechanism is included in domestic arbitration legislation, institutional arbitral rules, and international instruments. Under the ICSID Convention, for example, high moral character and the ability to exercise independent judgement are, among other things, required qualifications of arbitrators (Ref. 14, art. 14 (1)). If an appointed arbitrator seems to fail to meet such requirements, a party is given the right to challenge him or her, asking the tribunal to disqualify such an arbitrator, on account of any fact indicating a manifest lack of the required qualifications (Ref. 14, art. 57). It indicates that an arbitrator is presumed impartial and independent unless the contrary is proven through the challenge process. Similar provisions are included in the UNCITRAL Arbitration Rules, which state that ‘[a]ny arbitrator may be challenged if circumstances exist that give rise to justifiable doubts as to the arbitrator’s impartiality or independence’ (Ref. 22, art. 12.1).
With regard to the criteria for the determination of lack of impartiality or independence, the International Bar Association’s Guidelines on Conflicts of Interest in International Arbitration (IBA Guidelines) serve to provide uniform standards. 23 In summary, the IBA guidelines consist of two parts. The first part introduces general standards of impartiality, independence, and duty of disclosure, with useful explanations. For example, ‘justifiable doubts’ to the impartiality or independence of an arbitrator is interpreted as a situation under which
a reasonable third person, having knowledge of the relevant facts and circumstances, would reach the conclusion that there is a likelihood that the arbitrator may be influenced by factors other than the merits of the case as presented by the parties in reaching his or her decision. (Ref. 23, Part I, (2)(c))
The second part is titled ‘Practical Application of the General Standards’. It provides users with three lists containing instances of potential conflicts of interest that may possibly undermine an arbitrator’s impartiality or independence. First, there are the Non-Waivable Red List (NWRL) and the Waivable Red List (WRL), which contain instances that give rise to justifiable doubts from the viewpoint of a reasonable third person as to the arbitrator’s impartiality or independence (Ref. 23, Part II, 2). The difference between the NWRL and the WRL is that situations described under the NWRL are more serious than those under the WRL, so the former cannot be waived even if accepted by the disputing parties, whilst the latter can be cured by the parties’ express willingness to have such a person act as arbitrator (Ref. 23, Part II, 2). Second, there is the Orange List (OL) containing situations that may, from the disputing parties’ perspective, give rise to doubts as to the arbitrator’s integrity, but the parties are deemed to have accepted the arbitrator if, after disclosure of such situations described by the OL, no timely objection is made (Ref. 23, Part II, 3). Finally, the Green List (GL) contains instances where no appearance and no actual conflicts of interest exist from an objective point of view (Ref. 23, Part II, 7).
The ICS, as proposed by the EU, also allows the parties to challenge the impartiality or independence of arbitrators (Ref. 13, art. 11.2), even though the arbitrators are installed by neutrals of the ICS. Although the ICS has not commenced its operation, the effectiveness of the challenge mechanism in previous arbitral proceedings can be observed. For instance, in recent ICSID cases, such as Caratube v. Kazakhstan, 24 Blue Bank v. Venezuela 25 and Burlington Resources, Inc. v. Ecuador, 26 arbitrators were challenged and successfully disqualified. These examples demonstrate that, if an arbitrator is found to be biased, the removal of such arbitrator can be done through the challenge procedure. The inclusion of the challenge mechanism in the ICS shows that, irrespective of how the arbitrators are appointed, there is still the potential risk of an arbitrator being biased; in other words, the deprivation of the parties’ right to appoint arbitrators cannot guarantee the integrity of the arbitrators; otherwise there would be no need for the challenge mechanism. It is the challenge procedure, not the deprivation of the parties’ right to appoint arbitrators, that serves to effectively safeguard the ICS proceedings from corrupt arbitrators. In short, the best policy to ensure the integrity of arbitrators is to include the challenge procedure in arbitral proceedings, rather than to presume all arbitrators are biased upon their appointment by the parties. The abolishment of the appointment of arbitrators by the parties is a breach of the presumed impartial and independent principle, and an overkill, in terms of the goal it attempts to achieve.
Conclusion
The rules regarding the composition of tribunals and the appointment of arbitrators under the ICS of the TTIP, as proposed by the EU, are different from past experiences. Under the new rules, the disputing parties are no longer allowed to appoint their respective arbitrators. Instead, such power is transferred to neutrals under the ICS. By doing this, the arbitral proceedings might be streamlined as far as the efficiency of the process of appointing arbitrators is concerned. This is indeed a needed reform. But apart from this procedural advantage, the new rules do not seem to improve investor–state arbitration, insofar as the rules are based on the pro-investor hypothesis. It is neither logical nor fair to blame the disputing parties’ appointment of arbitrators for the pro-investor tendency. In fact, according to empirical evidence the pro-investor hypothesis is merely a false assumption and does not even exist. Irrespective of how the arbitrators are appointed, they should be presumed impartial and independent, unless their bias is proven through the applicable challenge procedures. Accordingly, the new rules under the ICS, which allegedly serve to secure the integrity of arbitrators at the price of depriving the parties of their right to appoint arbitrators, is illogical, unfounded, unjustifiable and unnecessary, and an overkill to the improvement of investor–state arbitration.
Chi-Chung Kao is a Professor in the Department of Ocean and Border Governance, National Quemoy University, Kinmen, Taiwan, Republic of China, a Member of the Chartered Institute of Arbitrators, UK, and an Arbitrator at the Chinese Arbitration Association, Taipei, Republic of China. He holds several research awards. His areas of specialisation are International Investment Law, International Dispute Resolution, International Arbitration, and International Commercial and Business Law.
