Jump to Content Jump to Main Navigation
The Oxford Handbook of International Investment Law edited by Muchlinski, Peter T; Ortino, Federico; Schreuer, Christoph

Part III Procedural Issues, Ch.27 Review of Awards

Vladimír Balaš

From: The Oxford Handbook of International Investment Law

Edited By: Peter T Muchlinski, Federico Ortino, Christoph Schreuer

From: Investment Claims (http://oxia.ouplaw.com). (c) Oxford University Press, 2015. All Rights Reserved.  Subscriber: null; date: 23 October 2019

Subject(s):
Awards — Annulment — Failure to apply applicable law — Failure to state reasons — Review of arbitral awards — Revision of award — Applicable law — International investment law

(p. 1125) Chapter 27  Review of Awards

  1. (1)  Revision of International Arbitral Decisions: The Background 1127

    1. (a)  The Notion of International (Investment) Arbitration 1128

    2. (b)  Different Forms of Review in Investment Arbitration 1130

    3. (c)  A Historical Excursus 1131

    4. (d)  Judicial Review—Continental Traditions 1134

  2. (2)  Review of Arbitral Awards in Investment Disputes 1136

    1. (a)  Annulment Procedure under the ICSID Convention and the Issue of the Applicable Law 1137

    2. (b)  Experience with Award Revisions in ICSID 1146

    3. (c)  Review of Arbitral Awards in (International) Commercial Arbitration—Mixed Views 1150

  3. Concluding Remarks 1151

(p. 1126) Rudyard Kipling once said that nothing was settled until it was settled right.1 Could or should this idea be used as the motto for the settlement of international investment disputes?2 In my short contribution, I should like to deal with the fundamental question of whether and under what circumstances this principle can be applied also to the settlement of disputes in international law, with particular regard to international (investment) arbitration.

I have tried to find an answer to a similar question in an article for the Czech professional public.3 In that article I dealt with the issue of review of decisions generally in the settlement of international economic disputes, affording only a fleeting mention of the possibilities of review within the ICSID framework. The main purpose of the quoted Czech article was, in fact, to draw the reader's attention principally to one of the aspects of the relatively newly developing settlement of disputes in the World Trade Organization. As these issues are becoming, once again, an integral part of legal discussions on the issue of investment disputes, I believe that a brief review of the past would not go amiss.

The present chapter will first sum up briefly the development of opinions on the possibility of review of the decisions of international tribunals in general. It is obvious that international investment arbitrations fall under the notion of dispute settlement in public international law. It is equally obvious, however, that the drafters of the relevant international treaty and, in the end, also the arbitrators themselves have also drawn—and are still drawing to a considerable extent—from private law, in the first place from international commercial arbitration. Subsequently, the chapter will attempt to analyse briefly several key decisions of ad hoc committees established in accordance with the provisions of Article. 52 of the ICSID Convention.4 In the conclusion, I shall briefly assess whether it is necessary to broaden the scope of the review of arbitral awards in investment disputes and particularly whether it would (p. 1127) be convenient to introduce into investment disputes an appellate system allowing for the review of the merits of investment disputes.

Marginally, I should like to mention that the review of a legal act concerns, naturally, an allegedly or actually defective act. Similarly, as court decisions, arbitral awards can suffer from various defects. With respect to the review procedure, it is important whether the defect is such that it could influence the decision and cause its factual wrongness.5 Generally, a decision can be considered defective because of errors in appreciation of the facts or it can just suffer from various legal defects, such as procedural errors or errors in interpretation and/or application of law. In arbitration, the view of defectiveness is rather narrow and a defective act is deemed generally to be an act that shows only legal defects of some kind. In most cases of the review of an arbitral award, regardless of whether the award was rendered in commercial arbitration or in international investment disputes,6 such an award is considered defective if it does not comply with the relevant legal rules of procedure.

(1)  Revision of International Arbitral Decisions: The Background

Before dealing with the possibility of revision of awards rendered within the framework of investment disputes, we shall discuss the definition of the notion of international arbitration as well as the different forms of review of investment awards. Moreover, this section will provide a brief historical excursus on revision in international law as well as an examination of the concept of judicial review in the continental legal tradition.

(p. 1128) (a)  The Notion of International (Investment) Arbitration

The notion of arbitration may cover various types of dispute settlement, in particular the settlement of disputes between different subjects, the object of which involves various claims arising from various legal relations. Although it is possible to find numerous identical features in different types of arbitration proceedings, it is also true that many differences also exist, particularly in light of the character of the parties to the dispute and the object of the dispute. That is probably also the reason why, from a historical perspective, views have changed over some procedural aspects of arbitration proceedings. Among the disputed issues, one finds the question of whether it is appropriate to permit revision of the arbitral award, and if so, whether it is appropriate to limit such revision to procedural aspects only or to allow review of merits as well.7

According to Schlochauer, arbitration is the process of dispute settlement between states by means of arbitration tribunals appointed by the parties, where the tribunal may be established before or after a dispute has arisen between the parties.8 Article 37 of the Hague Convention on the Peaceful Settlement of International Disputes of 1907 (cf Hague Peace Conferences of 1899 and 1907) provides that the aim of international arbitration is the settlement of disputes between states by the judges appointed by them on the basis of respect for law. Recourse to arbitration involves also the obligation to abide by the award in good faith.9 Even though arbitration proceedings may arise between states also in the framework of investment disputes (eg in connection with the interpretation of investment treaties), we shall not deal much with this type of arbitration.

According to the same author, the notion of international arbitration does not include mixed arbitration tribunals, which deal with the claims of private natural or legal persons from foreign states, as such disputes have been generally litigated by the state of nationality of the private person concerned through diplomatic protection.10 The institution of mixed arbitration tribunals has its origin in the Mixed Claims Commissions established in the 19th century between the USA and the (p. 1129) states of South and Central America, although most of these commissions actually satisfied the definition of international arbitration tribunals with reference to the de facto restriction of their competence to settle the disputes between states. Mixed arbitration tribunals were established in great numbers on the basis of peace treaties after World War I. As a result of the broad spectrum of provisions dealing with the consequences of the war, the number of cases decided by mixed arbitration tribunals exceeded by a large margin the number of all international arbitration awards ever rendered. The number of mixed arbitration tribunals established after World War II includes in particular two tribunals established according to Articles 29 and 32 of the London Agreement on German External Debts 1953 (Annexes I and IV), the Arbitral Commission on Property, Rights and Interests in Germany 1954, and the Austro-German Property Treaty Arbitral Tribunal, 1957, established according to Article 108 of this treaty.11

The classification of the settlement of investment disputes by arbitration according to this criterion is not entirely clear. On the other hand, it is clear that with reference to the parties to the dispute, it can be considered a category approaching mixed disputes, as it involves disputes between states and investors (juridical or natural persons) of another state. However, the current system of investment dispute settlement developed from the espousal by states of claims of their nationals through diplomatic protection. This is attested to by the decisions of the International Court of Justice in such notorious cases as Nottebohm, Elettronica Sicula, and others.12 However, as this system proved cumbersome and started hindering the development of cross-border economic cooperation,13 states started looking for another model.14 The result was a number of bilateral investment treaties (BITs) (supplemented also with several multilateral treaties) that have gradually replaced FCN treaties negotiated in the 19th century and well into the 1930s, which were based on the old model of espousal of claims.15 The earliest BIT dates back to 1959 (between Germany and Pakistan) and the Washington Convention (establishing ICSID) came a few years later in 1965. The treaties on the protection of investments concerned, in particular, two types of disputes: (a) disputes between the parties to the treaty concerning the interpretation of the treaty itself and (b) disputes between a state and an (p. 1130) investor of the other party to the treaty concerning inter alia an alleged breach of the treaty itself.

(b)  Different Forms of Review in Investment Arbitration

Different types of review mechanism are quite routinely used without raising any controversy, while others are heavily debated. Just a glance at the ICSID Convention shows that it contemplates several remedies, spanning from Article 49 to Article 52.

‘Supplementation’ and ‘rectification’ are provided for in Article 49(2) and they serve as a remedy for certain types of omissions or errors in the award. They can be made by the tribunal that rendered the award and are meant to deal only with minor technical or clerical mistakes in the award itself.

Some problems might require ‘interpretation’ as provided for in Article 50 and some parties to the dispute may be tempted to use it as a tool enabling them to avoid an adverse award. Interpretation is meant to clarify the meaning of the award when parties disagree on its construction.

‘Revision’ provided for in Article 51 takes account of new facts that were unknown when the award was rendered. Revision is a quite traditional16 post-award or post-judgment remedy and (on the international plane) we can track it also to Article 61 of the Statutes of the International Court of Justice and of its predecessor the Permanent Court of International Justice.17 Revision can bring substantive alteration of the original award on the basis of newly discovered facts. It is not necessarily the kind of revision that has evolved within the theory of civil procedure as a third system of review alongside ‘appellation’ and ‘cassation’.

The most controversial issue by far is whether the ‘supreme’ post-award remedy in an investment dispute consisting of vacating the award18 should be limited to the annulment (or setting aside) of the award in narrowly defined circumstances or whether a broader remedy should be allowed. The crucial point is what the subject of the review should be. As we have already mentioned above, there is a broad range of possibilities, some typical of appellate systems, others of cassation or even of revision.19

(p. 1131) While some national systems allow for the review of all aspects of the decision (errors of law as well as errors of facts), it seems that the debate in the international arena revolves around the acceptance of the mechanism implemented within the WTO dispute settlement system, and for those who favour it, whether it should deal just with errors of law or allow for a limited review of the facts in exceptional circumstances. An effective change of the system would probably require agreement on an international scale that would include not only change to the ICSID Convention, but would require accord among states to harmonize relevant domestic legislation since national courts in ad hoc (mostly UNCITRAL) investment arbitration play an important role in setting aside the investment arbitration awards. A third alternative, would be the creation of a kind of appellate body with the mandate to review all investment disputes in ad hoc as well as institutional arbitration. The present chapter is limited to the last-mentioned post-award remedy, that is, the vacation of the award.

(c)  A Historical Excursus

The revision of the decisions of international tribunals is by no means a new subject. We could even call it a traditional subject—a classification attested to by the fact that it has been treated by whole generations of international jurists, from Hugo Grotius20 to Emmerich de Vattel,21 the latter considered today the author of the modern theory of international law dealing with the possibility of revision of an arbitral award in the field of international public law.

The revision of arbitral awards as such may be considered without doubt a subject of the history of law and can be traced back to the works of classical Roman jurists. It goes without saying that the founding fathers of international law were influenced considerably by the Roman law tradition.22 When speaking of the Roman law tradition, we do not mean international public law and the dispute settlement in relation thereto. In the field of public law, arbitral awards in Roman times were issued during the exercise of Roman sovereignty or as hegemonic measures for the (p. 1132) maintenance of pax romana. Thus, it is no wonder that Rome itself never submitted to such international arbitration.

Under Roman private law, the arbitral award was not subject to review, as noted, for instance, by Ulpianus.23 It was impossible to appeal an arbitral award or request its annulment. As soon as the arbitrator issued his award, his role as arbitrator became functus officio (its functions terminated). Paulus noted that an arbitration award could not be changed24 and Gaius that the arbitrator could not correct his award even if he had committed an error.25

The period of the Middle Ages is somewhat nebulous. This follows also from the citations of Goldschmidt26 and Novakovich27 in the quoted work by Ku?era.28 According to Goldschmidt, in the Middle Ages arbitral awards, both judicially and by legislation, were subject to the same system of remedies of judicial decisions. An appeal was permitted as if the award was a judgment issued by a court of first instance, unless there had been an express agreement to exclude an appeal. Annulment could also be sought in case of obvious unlawfulness. Even in cases where an appeal was forbidden, it was possible to resist arbitral awards due to gross injustice, which mixed the office of decision-maker with that of the arbitrator.

Novakovich, on the other hand, asserts that in arbitration between states it was possible to (judicially) revise arbitral awards in the Middle Ages. After an analysis of the cases of arbitration from the 12th to the 15th centuries Novakovich states that arbitrators, having issued their award, could decide on the issues resulting from the interpretation or execution of their decision, but the award itself was final. If we compare this approach with the relevant provisions of the ICSID Convention, we can easily find certain parallels between the medieval and the contemporary approaches.

If we compare the statements of both above-mentioned authors, a certain discrepancy may appear at first sight. A more painstaking analysis, however, will lead to the conclusion that this discrepancy is only apparent. While Goldschmidt deals with arbitration proceedings in general or analyses the issue of arbitration proceedings (p. 1133) on a national level, Novakovich deals with the settlement of disputes on the international plane. After all, decisions of international tribunals in traditional international disputes between states have not been subject to review. If the losing party is not satisfied with the results of the proceedings, it has no remedial instance (provided the parties to the disputes have not agreed on such a possibility) to which it could address its request for review of the judicial decision or the arbitral award. There was only the possibility of refusing to accept such a decision and to initiate a new proceeding concerning the issue in question which, naturally, entailed a number of risks.

Let us return, however, to the classics of international law. Grotius who, like a number of other jurists of his time, was inspired by Roman law to which he added the new concept of natural law, was of the opinion that an appeal has no place between the king and the nation.29 As this categorical opinion of Grotius had certain limits, a change did not take long to come. From the 18th century onwards, some authors started admitting the possibility of not respecting ‘unjust’ arbitral awards. As mentioned above, the writings maintain that the author of such a modern approach was the author of the theory of exces de pouvoir (excess of power) in the modern judicial system, Emerich de Vattel. According to de Vattel:

When sovereigns cannot agree on their claims and yet crave for the maintenance or restoration of peace, they entrust often the decision of their disputes to arbitrators; they have undertaken to do so and should observe their commitments. However, if the arbitrators themselves by their obviously unjust and unreasonable arbitration award have betrayed their commitments, their award does not deserve attention and the parties have respected it only in doubtful issues. Let us assume that the arbitrators, to remedy the suffered wrong, condemn a sovereign state to becoming the subject of the offended state: will a reasonable man say that this state would respect the award? If the injustice is small, it should be suffered for the good of peace and if it is not entirely obvious, it should be suffered as a wrong which the state intended to suffer. Because if the conviction of the justice of the arbitral award should be required for the state to abide by it, it would be unnecessary to apply to arbitrators.

It is not necessary to fear that by granting the parties the freedom not to abide by the obviously unjust and unreasonable arbitral award the arbitration judiciary will become superfluous; moreover, such decision is not at variance with the character of submission or compromise. A difficulty may arise only in case of submission to arbitration judiciary if it is vague and unlimited, namely in the case in which the object of the dispute has not been stated accurately, or the limits of disputed claims have not been defined. It may happen, like in the case which will follow, that the arbitrators will exceed their power and will pass a sentence on what has not been submitted to them. Having been asked to decide about the satisfaction which a State should grant for an offence, they will condemn it to become the subject of the offended State. Obviously that State has never granted them such great power and their nonsensical award is not binding for it. To prevent such difficulties, to eliminate every pretext for dishonesty, the compromise should define accurately the object of (p. 1134) the dispute, mutual claims and the content of the requests of both parties. That is what is submitted to arbitrators and in which the parties promise to abide by their award. And if the arbitral award remains within these accurate limits, it should be observed. It is impossible to maintain that it is unjust, because it gives the opinion on the issue which the disagreement of the parties has rendered questionable and which was submitted to arbitrators as such. To disregard such an arbitral award it would be necessary to prove by undisputable facts that the award is the result of corruption or obvious bias.30

As stated, for example by E Nys, Vattel's theory has struck firm roots in international law and was further developed by legal positivists. In the 19th century, we can find attempts to enumerate expressly the reasons for which states may refuse to abide by the arbitral award. They all agree that there may be cases in which the decision of the arbitration tribunal is not binding upon the parties to the dispute. In this way, the 19th century authors follow de Vattel. The reason, in their opinion, is particularly the fact that ‘there is no real decision’. The parties are confronted by a decision which is incorrectly qualified as an arbitration award.31 For instance, Bluntschli maintains that a party to the dispute may claim invalidity of the arbitration court award for one of the following reasons:

  1. (a)  that the arbitration court in its decision has exceeded its power,

  2. (b)  that the arbitrators' approach was dishonest,

  3. (c)  that the arbitration court has denied the parties a hearing or has broken obviously the basic principles of all proceedings in some other way,

  4. (d)  that the content of the arbitral award is incompatible with the principles of international law and human rights (international public order reservation).

On the other hand, according to Bluntschli, the fact that the award is incorrect or unjust to one party does not constitute a reason for its annulment.32

(d)  Judicial Review—Continental Traditions

To complete the picture, it is useful to shortly describe the kind of review that has been available in continental Europe since the French Revolution.33 Civil procedure distinguishes two fundamental remedial systems by which it understands the definition of certain remedy-guiding principles. They are ‘appellation’ and ‘cassation’ supplemented later by the so-called ‘revision’. While appellation and cassation follow (p. 1135) the two-stage system, revision was originally thought as part of the three-stage system (third instance).

The broadest remedial system is appellation, since it allows review of decision without any limitation. Both errors of law and errors of fact can be subject to appellation. A challenged decision can be overruled, altered, set aside, and/or upheld. Within appellation, a distinction can be made between a system of total appellation and a system of partial appellation. Total appellation allows a second-stage court to take into consideration new facts and evidence that had not been submitted and known to the court at the first stage.

A system of cassation allows a second-stage court to review issues of law but not issues of fact. Courts of cassation review whether the decision is in compliance with substantive and/or procedural laws. New facts and evidence are not admitted and courts of cassation are not allowed to review facts. Challenged decisions of the lower court can be either set aside or upheld, but they cannot be altered by the higher court.

A revision system allows the higher court to review the legal aspects of the challenged decision. The difference with the cassation system is that, unlike the courts of cassation, revision enables alteration of challenged decision, but only in cases where the lower court had correctly ascertained the facts of the case but its evaluation of the merits of the case was legally erroneous (incorrect). The result may be to uphold or alter the challenged decision but this will be based on the findings of fact made by the lower court. As we shall see, this kind of revision is not identical with the revision in international law, as described above.

Current review of investment awards ranges from a limited cassation system (a tribunal is mostly allowed to review only procedural errors of law)34 to a full cassation system (allowing the tribunal to deal with errors of law generally). It is questionable where the ICSID system now sits and where it should go in the future. It is also interesting to point to the WTO Appellate Body, whose competence is ‘limited to issues of law covered in the panel report and legal interpretations developed by the panel’.35 It is clear that the WTO appellate system is based on revision and cannot be considered as a true ‘appellate’ system.

It is quite obvious that this context gives rise to a number of further questions, for example, whether to introduce the principle of appeal, revision, or whether to adhere to the principle of cassation and if so, whether to remand the award to the (p. 1136) same arbitration tribunal, and so on.36 In international practice it is possible to find a number of the most varied approaches. As mentioned earlier, we can submit that what is suitable for international commercial arbitration need not be suitable for the settlement of investment disputes or the settlement of disputes between states; what may be suitable for quasi-judicial dispute settlement at the WTO need not be so for investment disputes either. In this chapter, I should like to concentrate on investment disputes; this does not mean, however, that the issues in other fields of arbitration should be neglected.

(2)  Review of Arbitral Awards in Investment Disputes

The current rules concerning review of arbitral awards in investment disputes are not uniform either at the national or the international level. The reason for this lack of uniformity is particularly due to the fact that the bilateral or plurilateral treaties on investment protection usually comprise the settlement of disputes under ICSID, ICSID Additional Facility Rules, or UNCITRAL rules. Depending on the rules governing the arbitration, the review of awards may be left, somewhat unsystematically,37 to international tribunals (for ICSID arbitration) or national courts (for ICSID AF and UNCITRAL arbitration).

The following analysis deals mostly with review of arbitral awards under the ICSID Convention rules. Judicial review of arbitral awards issued in ad hoc arbitration under UNCITRAL rules or under ICSID Additional Facility Rules is mentioned just marginally, as national laws adopt different approaches even if there are some common principles, for instance, in countries that have adopted the UNCITRAL Model Law on Arbitration. In order to map this field correctly, one would have to deal with some 200 national legislations, to say nothing about the fact that legislation or practice is not uniform even within certain states (as differences in interpretation (p. 1137) may appear even where the relevant legislation identically worded). Another problematic issue might be the applicability of rules whose primary purpose is to serve (international) commercial arbitration in review of investment arbitration awards in particular states.38

In practical terms, judicial review of international investment awards by national courts is relatively limited, in particular for two sets of reasons: (1) most of the cases fall under the ICSID Convention, which precludes judicial review by national courts and (2) the handful of non-ICSID cases left would only affect certain states.39

(a)  Annulment Procedure under the ICSID Convention and the Issue of the Applicable Law

Under Article 52 of the Convention on the Settlement of Investment Disputes between States and Nationals of Other States (ICSID Convention) parties may request the annulment of the award. The grounds for annulment of the award are that:

  1. (a)  the tribunal was not property constituted,

  2. (b)  the tribunal has manifestly exceeded its power,

  3. (c)  there was corruption on the part of a member of the tribunal,

  4. (d)  there has been a serious departure from a fundamental rule of procedure,

  5. (e)  the award has failed to state the reasons on which it is based.

The application must be submitted to the ICSID Secretary-General within 120 days from the date on which the award was rendered, except for cases of corruption where there is an objective three-year term of limitation, which starts from the day that the award was rendered, and a subjective term of 120 days from the day that corruption was discovered. Having received the request, the Chairman of the ICSID Board of Directors (ex officio the Chairman of IBRD) appoints a three-person committee from the Panel of Arbitrators. The members of an ad hoc committee may not be the same arbitrators who decided the underlying dispute previously. Every member of the ad hoc committee must be of a different nationality and none of them may be a citizen of the state party to the dispute or of the state of the disputing investor. The ICSID annulment procedure proceeds in accordance with a principle reminiscent of ‘limited cassation’: the ad hoc Committee may only confirm or annul the award and (p. 1138) if the award is annulled, the dispute may be decided by a new arbitration tribunal constituted in accordance with section 2 of Chapter IV of the Treaty.40

Of five grounds, two seem to be interesting from the point of view of understanding the nature, and the future development, of the review mechanism in ICSID practice. These are ‘manifest excess of powers’ and ‘failure to state reasons’. Manifest excess of powers is not just limited to the most obvious example of it, that is, decision on the merits by a tribunal that lacks jurisdiction.41 As is known from Vivendi, manifest excess of powers is also constituted by failure to exercise jurisdiction in cases where jurisdiction does, in fact, exist. Open-ended examples of possible or alleged manifest excess of powers might come up in regard to issues already decided or pending in a different proceeding, that is, res judicata and lis pendens issues.42 While these cases are more or less clear, a third example of manifest excess of powers, the failure to apply the proper law, might be considered more problematic and jurisprudence on this issue is far from clear. The remainder of this section addresses this issue in more detail.

Article 42(1) of the ICSID Convention deals with the applicable law to the dispute. This provision is quite crucial and is one of the essential elements of the parties' agreement to arbitrate. There is still a distinction being made between failure to apply proper law and incorrect application of that law. However, from the point of view of the final outcome, both seem to have the same impact. While the application of a law other than that agreed to by the parties constitutes an excess of powers and is a valid ground for annulment at the ICSID level, erroneous application of the proper law is not a ground for annulment.43 This was emphasized by ad hoc committees in several awards.44 The ad hoc Committee in Amco v Indonesia is somewhat illustrative on this point:

The law applied by the Tribunal will be examined by the ad hoc Committee, not for the purpose of scrutinizing whether the Tribunal committed errors in the interpretation of the requirements of applicable law or in the ascertainment or evaluation of the relevant facts to which such law has been applied. Such scrutiny is properly the task of a court of appeals, which the ad hoc Committee is not. The ad hoc Committee will limit itself to determining (p. 1139) whether the Tribunal did in fact apply the law it was bound to apply to the dispute. Failure to apply such law, as distinguished from mere misconstruction of that law, would constitute a manifest excess of powers on the part of the Tribunal and a ground for nullity under Article 52(1)(b) of the Convention.45

Such a clear message is in contradiction with the Swedish Court of Appeal in the CME case. In its partial award, the tribunal arbitrarily, as it appears from that award, omitted to apply part of the applicable law and only subsequently and under pressure tried to improve its image in the quantum decision. Although one could argue that this could have led to annulment at the ICSID level, the Swedish Court of Appeal avoided having to deal with the whole issue in a very formalistic way.46 The whole issue rests on the legal certainty and predictability of the system. Even annulment decisions should be reasonable and as just as possible. It is not a problem of differing decisions at the ICSID level and at the level of 200 national jurisdictions. There are just two possible decisions available in any given case: to confirm or to set aside the award. What is fundamental is that the procedures are standardized and predictable.

The Czech Republic alleged that the arbitral tribunal failed to apply the law that the arbitrators were obligated to apply in accordance with the Treaty, namely Czech law and international law. The arbitral tribunal, instead, based the award on general assessments of reasonableness. According to the Republic, the arbitral tribunal thereby exceeded its mandate and, in any event, committed gross procedural errors, each of which individually, and in any event taken together, affected the outcome of the arbitration proceedings.

The Swedish Court in its decision dealt with the differences between UNCITRAL Model Law and the Swedish Arbitration Act, which does not contain any rules as to the legal premises on which arbitrators should determine a dispute. In the view of the Swedish Court, on the basis of the legislative history of the Arbitration Act, such a rule may be dispensed with. It also stated as follows:

As it was added by the Court, in Sweden, there is probably a unanimous view that arbitrators should base their awards primarily on governing law, unless the parties may be deemed to have decided differently. In light of the desire to restrict the possibilities of challenge, in favour of the finality of an arbitration award, there exist predominant reasons against the implementation of any rule as to the legal premises on which a dispute shall be determined. However, the aforesaid does not prevent the parties from entering into an agreement that the dispute shall be determined in accordance with the law of a particular country or that the arbitrators shall determine the dispute based on reasonableness. Where it is evident that the arbitrators have applied the law of a different country in violation of such an (p. 1140) agreement, upon application to the Court, the award may be set aside on the ground that the arbitrators have exceeded their mandate. In doing so, the Court should not, of course, engage in a determination of whether the arbitrators erroneously applied the law agreed upon by the parties, as that should not lead to the arbitration award being set aside (Government Bill 1998/99:35 p. 123).47

The Swedish Court also stuck to the generally shared view on the issue in another part of its decision where it stated:

A general conclusion which may be drawn from that which is stated in the legislative history is that the legislature has sought to reduce the possibilities to challenge an arbitration award on the ground that the arbitrators have applied the wrong law. The arbitrators may be deemed to have exceeded their mandate only where they have applied the law of a different country in violation of an express provision that the law of a particular country shall govern the dispute; in the opinion of the Court of Appeal, an almost deliberate disregard of the designated law must be involved. There is no excess of mandate where the arbitrators have applied the designated law incorrectly. Nor can there hardly be any question of excess of mandate where the arbitrators have been required to interpret the parties' designation of applicable law and, in so doing, have interpreted the designation incorrectly.48

The same conclusion of the Swedish Court can also be found in the section of the decision that dealt with Articles 8.2 and 8.5 of the BIT entered into between the Czech Republic and the Netherlands:

The arbitration proceedings have taken place within the scope of the UNCITRAL rules. Articles 33.1 and 33.2 of the UNCITRAL rules provide, inter alia, that the arbitral tribunal shall apply the law which the parties stated to be applicable to the dispute and that the tribunal shall base its decision on assessments of reasonableness (shall decide as amicable compositeur or aequo et bono) only where the parties have expressly authorized the tribunal to do so.49

The relevant choice of law clause was correctly found by the Court in Article 8.6 of the Treaty. It is worded as follows:

The arbitral tribunal shall decide on the basis of the law, taking into account in particular though not exclusively:

•  the law in force of the Contracting Party concerned;

•  the provisions of this Agreement, and other relevant Agreements between the Contracting Parties;

•  the provisions of special agreements relating to the investment;

•  the general principles of international law.

(p. 1141) The following reasoning of the Swedish Court of Appeal was quite surprising, and not only for the Czech Republic:

The first part of the provision whereby the arbitrators ‘shall decide on the basis of the law’ is clear, while the following parts leave room for interpretation, as shown by the arbitrators' differing opinions regarding the correct legal purport of the provision. The arbitrators were obliged to comply with the choice of law provision and, accordingly, their mandate has also included interpretation of the provision insofar as it is not clear or not unambiguous. In the opinion of the Court of Appeal, an excess of mandate may be involved only where the arbitrators' interpretation of the choice of law clause proves to be baseless such that their assessment may be equated with the arbitrators almost having ignored a provision regarding applicable law.50

The wording in Article 8.6 of the Treaty that the arbitral tribunal shall ‘take into account in particular although not exclusively’ must be interpreted such that the arbitrators may apply rules other than those listed. The four references to the applicable law are not numbered, nor is there any indication that one should take precedence over another. The un-numbered list almost gives the impression that the contracting states have left to the arbitrators the determination, on a case by case basis, as to what specific law should be applied … .51

The parties to the dispute had agreed upon the interpretation of the choice of law clause in the BIT: the arbitral tribunal must ‘take into account as far as they are relevant to the dispute the law in force of the Contracting Party concerned and the other sources of law set out in Article 8.6’. The interpretation which can be given to the wording of the clause is thus hereby confirmed, namely that the clause leaves the arbitral tribunal to take into account Czech law and other sources of law in so far as they are relevant in the dispute.

The Court of Appeal did not believe that the various sections in the arbitral award were to be reviewed in order to ascertain which law was applied by the arbitral tribunal from the list contained in Article 8.6 of the Treaty. In the Court of Appeal's opinion, when assessing whether the arbitrators had exceeded their mandate, it was sufficient to clarify whether the arbitral tribunal applied any of the rules listed in the choice of law clause or whether the tribunal had not based its decision on any law at all but, rather, judged in accordance with general reasonableness. The various sections in the arbitration award under review, which the parties had invoked to support their respective arguments with regard to which source of law was not applied, led to no conclusion other than that the arbitral tribunal had complied with the provisions of the choice of law clause (ie it had applied the relevant sources of law, primarily international law). Accordingly, the Swedish Court concluded that the tribunal had violated the Treaty by basing its decision on a general assessment of reasonableness devoid of any basis in law. The fact that each legal statement in the award was not directly and expressly connected with a rule of law cannot be deemed to mean that (p. 1142) the tribunal conducted a general assessment of reasonableness. Accordingly, taking into consideration the aforesaid, the Court of Appeal found that the arbitral tribunal did not exceed its mandate by failing to apply the applicable law.

Schreuer's comments,52 following the opinion he and Reinisch provided to the Czech Republic as expert witnesses, expressed a different view. It reads as follows:

The key passage on applicable law in the Final Award is contained in para. 402. The Tribunal analyzes Article 8(6) of the Bilateral Investment Treaty between the Czech Republic and the Netherlands (‘the BIT’) containing the choice of law clause and comes to the conclusion that this provision is broad and grants a discretion to the Tribunal. The Tribunal adds that there is no exclusivity in the application of these laws. The Tribunal points out that the provision instructs the Arbitral Tribunal to take into account (not apply) the sources of law, in particular though not exclusively. …

The Tribunal's chief argument is that Article 8(6) gives discretion. But under the provision's wording, the discretion refers to the application of additional sources not to disregarding the sources listed in Article 8(6) but to disregard others. If that were true the Tribunal would also have discretion to disregard the BIT itself—an obvious absurd result.53

In their opinion, Professors Schreuer and Reinisch summarized the consistent practice of ICSID tribunals on the issue of applicable law, and under their opinion, the ICSID practice could also be accepted in UNCITRAL arbitration. Not only did the arbitral tribunal refuse to recognize its failure to apply proper law, its attitude was upheld by the Swedish Court of Appeal. The ratio decidendi of the Swedish Court of Appeal is far from persuasive. If such an experienced court can deliver such a decision, what can one expect from the national courts of a country with minimal or no arbitration tradition? One can just speculate how the revision would have been dealt with, for example, by the Czech Courts. This was yet another example of how the current investment dispute settlement does not work properly, nor does it work properly at the level of review of awards. The reason is obvious, and it is submitted that it is lack of predictability.

Commercial arbitrations are based on the general principle that the national court at the venue of arbitration proceedings may set aside the award for reasons provided in the respective national legal system, in the law governing the arbitration process, and in recognition and enforcement of arbitral awards. At first sight, it may appear that there is no substantial difference between legal regulation on the possibility of annulment of an award under the ICSID Convention and setting aside (p. 1143) arbitral awards under most national laws,54 for example Czech law,55 Swiss law,56 or the Federal Arbitration Act (FAA) of the USA.57 As noted above, there may not be a substantial difference in the wording as such, but there may be such a difference in the application and interpretation of the relevant legal documents.

A problematic situation is admitted even at the UNCITRAL level. It should be noted that the national arbitral legislation in many states is based on the UNCITRAL (p. 1144) Model Law on International Commercial Arbitration.58 Reasons for setting aside the arbitral award in Russia,59 India,60 or elsewhere are copied from Chapter VII (p. 1145) dealing with ‘Recourse Against Award’.61 An explanatory note by the UNCITRAL secretariat on the Model Law on International Commercial Arbitration62 supports our view:

Although the grounds for setting aside are almost identical to those for refusing recognition or enforcement, two practical differences should be noted. Firstly, the grounds relating to public policy, including non-arbitrability, may be different in substance, depending on the State in question (i.e. State of setting aside or State of enforcement). Secondly, and more importantly, the grounds for refusal of recognition or enforcement are valid and effective only in the State (or States) where the winning party seeks recognition and enforcement, while the grounds for setting aside have a different impact.63

These rules do not apply to arbitral awards issued pursuant to the ICSID Convention. As opposed to awards issued under the UNCITRAL rules, the ICSID Convention (p. 1146) awards are protected against challenges before national courts (particularly those of the host state), which may not always be impartial to the investor.64 This feature is considered one of the principal advantages of the settlement of investment disputes under the ICSID Convention. Another advantage, according to numerous authors, is the fact that the Washington Convention comprises its own review mechanism. As stated by Van Houtte, the possibility of annulment of an arbitral award represents an important guarantee of the assurance of the basic requirements for a just decision.65 The possibility of revision functions as an incentive for arbitrators to proceed so as to make their decision as objective as possible while observing all procedural principles considered standard not only in arbitration proceedings, but also in judicial proceedings. It is obvious that the learned Belgian has not been a party to a dispute before a court of first instance in the Czech Republic, where an entirely opposite interpretation of the institute of award revision can be encountered. The approach of the judge to the decision is often highly alibistic: the judge almost advises the parties of the possibility of appeal, as he is actually not sure of his decision.

(b)  Experience with Award Revisions in ICSID

So far the aforementioned ad hoc committee has dealt with award annulment in several cases, for example Klockner v Cameroon,66 Amco v Indonesia,67 MINE v Guinea,68 Vivendi v Argentina,69 and Wena Hotels v Egypt.70 Gruslin v Malaysia was not taken up due to the failure to pay the judicial fees.71

The above-mentioned annulment applications (Klockner I, Klockner II, Amco I, Amco II, MINE, Wena, and Vivendi) show that the parties have usually challenged (p. 1147) the award on several grounds, which is understandable as they wanted to avail themselves of all possibilities. At the same time, it should be noted that two of the aforementioned grounds for annulment have not been pleaded at all. Challenges have not involved allegations that the tribunal was not properly constituted (Art 52(1)(a) ) or that there was corruption on the part of a member of the tribunal (Art 52(1)(b) ). All other grounds have been invoked, although the grounds for contesting the award have, naturally, differed from case to case.

The seemingly laconic text of Article 52 may cover a whole range of possibilities. In five of the seven cases the parties referred to Article 52(1)(b): lack of use or excess of the powers of the tribunal. In other words, the ground for the annulment request was that the original tribunal had failed to act (or did not exercise its power) although it should have done so, and failure to apply the proper law.

Article 52(1)(d) was invoked because of the alleged lack of impartiality, the right to be heard, scope of arbitrators' discretion, and defects of evidence. At least one of these grounds was invoked in all seven cases.

Article 52(1)(e) was invoked in six cases, on allegations of the non-existence of reasons substantiating the award, unsatisfactory and improper reasons, discrepancies between the reasons and failure to deal with the issues submitted to the tribunal.72 Probably the most frequent challenge concerns Article 52(1)(e), lack of reasons. It is in this respect that an extensive interpretation was used in some cases which made the appropriate ad hoc committees revise the merits of the award.

In the field of investment disputes within the framework of the ICSID Convention, more than 110 cases have been dealt with up until now, with almost 40 of them having been settled. In about 11 out of these 110 cases, an annulment application was submitted: two in Klockner v Cameroon and Amco v Indonesia. As noted above, the case of Gruslin v Malaysia was suspended.73 Some authors state that the number of revised cases is low, but the opposite opinion could equally be defended. With regard to the character of the disputes (huge amounts of money at stake, the prestige of the parties to the dispute, in case of states enhanced by political aspects of home policy), I would accept the opinion of those who consider the number of contested awards within ICSID to be low rather than high. In any event, an analysis of the decisions of the ad hoc committees is useful both for the arbitrators—the opinion of outstanding jurists deciding about the annulment of an arbitration award may be an inspiration for other arbitrators—but also for the legal representatives of the parties to the dispute. Moreover, these decisions may also enrich and influence the development of international investment law. Unfortunately, not all decisions have (p. 1148) been published, which can be considered the greatest shortcoming in comparison with the settlement of disputes with the WTO framework.74

What requirements are generally imposed on an arbitral award and what is the position of an ad hoc committee? The arbitration award must satisfy several criteria. In particular, the arbitral award must be effective; the necessary prerequisite for its effectiveness is that it is final. On the other hand, it is obvious that an arbitral award should also be correct. Apart from the fact that investment disputes put pressure on arbitrators to settle them without undue delay and so contribute to the economy of the arbitration proceedings, it is obvious that a conflict may originate between the need for correctness and the need for effectiveness of the arbitral award. In international public law, as suggested in the historical excursus, the mechanism of appeal and revision is not permitted in principle. The only exception is the above-mentioned dispute settlement system of the WTO. We are speaking, naturally, only about the formally regulated, institutionalized system of appeal and revision. This does not mean that everyone must necessarily submit to an incorrect decision under any circumstances without the possibility of refusing to abide by such a decision.

The possibility to annul an award under the ICSID Convention is considered a feature resulting from the semi-private character of investment disputes. The principal task of an ad hoc committee is to find the correct relation between the above-mentioned requirement of finality of the award and its correctness.75 Naturally, it is hardly possible to allow the requirements regarding finality of the decision and its speed to be preferred to its correctness.

Scholars divide the decisions of ad hoc committees into several generations, with most emphasizing the existence of three generations (although a fourth generation is admittedly in the process of formation).76

The first generation of annulment decisions includes the decisions of the ad hoc committees in the Klockner I and Amco I cases. These decisions have attracted the attention of critics particularly since the ad hoc committees did not assess (p. 1149) procedural defects, but endeavoured to apply the appellate principle and tried to review the merits of the decisions of the original arbitration tribunals.

The concern to introduce an element of appeal into the revision of arbitral awards under the ICSID Convention has found a response in the second generation of annulment decisions, including the decisions in the MINE, Klockner II, and Amco II cases. Unfortunately, the latter two decisions have not been published.

According to the prevailing opinion of authors, the third generation of decisions of ad hoc committees represent an equilibrium between the decisions of the first and the second generations. Among these are the decisions in the Wena and Vivendi cases. Even these decisions did not escape some interesting criticism, particularly on the part of the legal representatives of the parties to the dispute.77 According to Professor Schreuer, the last two decisions in particular show that an ad hoc committee will interfere (ie annul the award) only in the case of serious and important faults by the arbitration tribunal.78

While in the Klockner I case (a first generation decision), the ad hoc committee based its decision on the conviction that any procedural defect represents a reason for annulling the award, in the second generation decisions the ad hoc committees refused a purely formal assessment and emphasized the breach of material principles by the arbitration tribunals.

According to the author of this chapter, the above-mentioned equilibrium in the approach of ad hoc committees in the third generation decisions consists particularly in the confirmation of the trend towards moderation in the review of an award on the merits. It is obvious that an ad hoc committee cannot ascertain an alleged serious breach of fundamental procedural rules or alleged failure to state the reasons (which could result in a serious or important fault of the tribunal) unless it has examined the merits of the case. However, this process shifts to a certain extent the revision of awards in investment disputes from the assessment of procedural shortcomings, which is the case in commercial arbitration, to a full appellate review (similar to the settlement of disputes in the WTO dispute settlement system).

The introduction of appellate review in investment arbitration has a number of advocates even at present, although it is necessary to note that those opposing such appellate review have by far no means left the battlefield.79 It is possible that this is a really well-balanced procedure and that this method of review of decisions in investment disputes will assert itself elsewhere also and will become standard in (p. 1150) commercial arbitration80 or will also enrich the methods of dispute settlement in the field of international public law. Much water may run under the bridge before this happens. Until then, heated discussions will rage between the advocates and the opponents of this approach.

However, there are a number of arguments which call for caution and point to the unfeasibility of the appellate principle, if the appellate or revision authority works only on the basis of written documents (particularly decisions, submissions of parties, minutes of hearings, and documentary evidence). Direct contact with the parties, their legal representatives, and witnesses may provide more information on the dispute than written submissions, distorted, moreover, by interpretation and written records provided during the ‘first instance’ hearings.

(c)  Review of Arbitral Awards in (International) Commercial Arbitration—Mixed Views

It is frequently said that the main advantage of arbitration is to allow parties to settle their disputes without having to resort to courts. Permitting parties to contract for expanded judicial review of arbitration awards would take away this advantage by allowing a party back into court for a second try. Generally, parties to an arbitration agreement are trying to avoid the costs and delays inherent in the court system. These goals are abandoned by expanded judicial review. What we have in mind here is the revision of the merits of the case, not the revision of the eventual procedural faults. I believe that national legislation all over the world is pretty similar and enables parties to ask for review of an arbitral award for alleged procedural shortcomings that could have a negative impact on an award. A very similar attitude was chosen by drafters of the ICSID Convention. It is also clear that a similar possibility is given in investment arbitration under UNCITRAL rules. Yet, there might be significant differences in the treatment of review by particular courts all over the world, bearing in mind slight differences in national legislation in this respect as well. In a number of potential national jurisdictions, the outcome of the review might vary substantially. There is hardly any consistency or predictability in revisions of arbitral awards in UNCITRAL investment arbitration. This seems to be the main disadvantage of the investment disputes decided under UNCITRAL. Arbitration is characterized as an informal, inexpensive, fast, and private adjudicative process that may consider custom as well as principles of fairness and equity to reach an outcome that is final and subject to limited revision.

(p. 1151) It seems that the classical reasons for setting aside an arbitral award do not suit the changing needs for the correctness of an award in the 21st century. The difference is particularly visible in investment disputes where there is even more pressure for a ‘just’ award. Requirements of proper and transparent decision overshadow other aspects of arbitration, namely disputable advantages of inexpensiveness, expediency, privacy, etc. Finality is not the only value, especially, if one takes into consideration possible incorrectness as the highest cost. There is growing controversy over the issue of extending the grounds for setting aside of arbitral awards. This controversy is clearly visible not only in the above-mentioned ICSID cases but also in domestic arbitration and revision of awards in regular courts. Interesting discussions may be heard not only at ICSID but also elsewhere. In the USA, for example, Circuit Courts of Appeal try to find an answer to the question whether parties to an arbitration agreement could agree to federal court appellate review of an arbitral award beyond the level of review Congress has set forth in the Federal Arbitration Act.81 Although this question poses no problems in European states with a strong tradition of arbitration, some issues raised by arguing parties may be interesting to explore. What is at stake here is whether a clause expanding judicial review of an award (a) may give either party a right to request a revision of the award on the grounds that the award is not supported by the evidence, (b) may provide for de novo review for ‘errors of law’, or (c) may do both and set out the specific scope of review. Some Circuit Courts have held that parties are not permitted to expand the scope of review beyond the limits of the Federal Arbitration Act, while other Circuit Courts have held that parties may do so.82 If nothing else, the above-mentioned discussion clearly illustrates the danger of the review mechanism common to UNCITRAL investment arbitration. Not only can legislation vary from state to state but also state courts within the same jurisdiction differ substantially in interpreting their own national legislation. The issue deserves thorough analysis from other points of view as well. The discussion about expanding grounds for setting aside an award in investment dispute is reminiscent of a similar discussion at the ICSID level.

Concluding Remarks

Along with numerous authors, I share the idea voiced by Pierre Lalive that in the whole field of dispute settlement there is nothing more important than the finality (p. 1152) of arbitral awards and the possibility of their annulment.83 Nobody believes that the finality of an arbitral award should become a dogma. It is evident that the very possibility of contesting an arbitral award hangs over the arbitrators as a permanent threat and makes them deal with the case with maximum care. It is also obvious, and here it is possible to agree with a number of authors, that the Washington Convention without the review mechanism would be unacceptable for a number of its states parties.

There are a number of procedural rules under which investment arbitration may be conducted. The more widely used are the UNCITRAL Arbitration Rules and the ICSID Convention rules. One of the main differences between these two sets of rules is the post-award remedies in the sense that, while ICSID offers quite a complex and to some extent foreseeable system of award remedy, UNCITRAL rules are silent in this respect and post-award remedy is left to rules adopted in national jurisdictions. In contrast to the ICSID system, UNCITRAL can reach a level of predictability only with great difficulty and national courts having jurisdiction in motion to set aside arbitral awards not only apply a municipal law that differs from state to state, but also vary in their decision-making under similar rules. On the other hand, neither are ICSID ad hoc committees immune from various approaches to interpreting the same set of rules they apply.

It seems indisputable that the settlement of investment disputes under the ICSID Convention has significant advantages when compared with other dispute settlement methods.84 It involves inter alia a combination of flexibility of arbitration proceedings and at least to a certain extent of foreseeability of arbitrator's thinking as reflected in their decisions. The consistency of decision-making is contributed to also by the existence of the revision mechanism and the relative transparency of award publication. The question remains how to interpret the provisions of Article 52 of the Washington Convention in terms of the grounds for contesting an arbitral award. So far there is no agreement over whether interpretation should be restrictive and any review should approach that in commercial arbitration or whether, with regard to the character of disputes, it would not be better to adopt an extensive interpretation with the risk that the remedial system will change into a so-called ‘appellate’ system (within WTO dispute settlement meaning) and that disputes, settlement of which is relatively long as it is, will be further protracted. More controversial in this respect is probably the provision of Article 52(1)(e) which could make it easier to slip into reviewing the merits of the case rather than serious procedural defects that could influence the decision itself. We shall see in the next few years what decisions we shall encounter. It will not be without interest to monitor the interpretation of Article 52(1)(b), especially with reference to the application of governing law. So far there is no doubt that the application of incorrect law or the failure to apply correct law is not identical with the incorrect application of correct law. I believe, therefore, (p. 1153) that this provides a relatively wide field for the enhancement of the review of the merits even in the settlement of investment disputes. The impossibility of reviewing a serious error in the legal assessment of a case may completely undermine trust in the settlement of disputes using this method. And it is obvious that such a fault does not constitute a reason for the annulment of the award,85 at least not for the time being. The discussion of this subject is certainly highly interesting, but the final decision is still some way off. We shall see whether this decision will be correct. Should it be erroneous, perhaps it will be possible to review it.86

At present I cannot but agree with Curtis D Brown87 that the limitations of judicial review provide real benefits to some parties to a dispute.

Select Bibliography

  • Alvarez, A, ‘Setting Aside Additional Facility Award: The Metalclad Case’, in E Gaillard and Y Banifatemi (eds), Annulment of ICSID Awards, IAI International Arbitration Series No.1 (Paris, Juris Publishing, 2004)
  • Bjorklund, A, ‘The Continuing Appeal of Annulment: Lessons from Amco Asia and CME’ in T Weiler (ed), International Investment Law and Arbitration: Leading Cases from the ICSID, NAFTA, Bilateral Treaties and Customary International Law (London, Cameron May, 2005)
  • de Vattel, E, Le Droit des gens, ou principes de la loi naturelle, appliqués à la conduite et aux affaires des nations et des souverains (London, 1758)
  • Gaillard, E, and Banifatemi, Y (eds), Annulment of ICSID Awards, IAI International Arbitration Series No.1 (Paris, Juris Publishing, 2004)
  • Grotius, H, De iure belli ac pacis—libri tres, in quibus ius naturae et gentium item publici praecipua explicantur (1625)
  • Schreuer, C, ‘Three Generations of ICSID Annulment Proceedings’, in E Gaillard and Y Banifatemi (eds), Annulment of ICSID Awards, IAI International Arbitration Series No.1 (Paris, Juris Publishing, 2004)
  • Sornarajah, M, The Settlement of Foreign Investment Disputes (London, Kluwer Law International, 2000)
  • Van Houtte, H, ‘Article 52 of the Washington Convention: A Brief Introduction’ in E Gaillard and Y Banifatemi (eds), Annulment of ICSID Awards, IAI International Arbitration Series No. 1 (Paris, Juris Publishing, 2004)

Footnotes:

∗  I would like to express my gratitude to Hugo Perezcano for his patient and thorough review of my challenging text and also for not setting it aside.

‘Nothing is ever settled until it is settled right’—wise words, attributed to many great men such as Abraham Lincoln, Rudyard Kipling, Frank A Vanderlin, etc, can serve as a motto for the settlement of disputes generally. Can this motto, let us say, in Vanderlin's modification ‘Since nothing is settled until it is settled right, no matter how unlimited power a person may have, unless they exercise it fairly and justly their actions will return to plague them’, serve also as a guide in international investment disputes?

This maxim became known to the world of arbitration probably during discussions at the First. Peace Conference in the Hague in 1899. The issue of the revision of arbitral awards was put on the table by the American delegation and US delegate Holls admitted that the purpose of arbitration is to reach a final award, but according to him, this fundamental principle has some limits, expressed correctly by President Lincoln in the words: ‘Nothing is settled until it is settled right’. Cf L Audry, La Révision de la sentence arbitrale (Paris, Duchemin, 1914) at 29, ff.

V Balaš, ‘Possibilities of Review of Decisions in the Settlement of Disputes in International Economic Law (WTO, NAFTA, ICSID)’ (in Czech), 2 Časopis pro právní vědu a praxi (2004) at 97 ff. For the possibilities of review in investment disputes according to ICSID rules, see 101.

Published as a Communication of the Ministry of Foreign Affairs of the Czech Republic No. 420/1992 Coll.

One can imagine a situation in which a factually correct decision can be successfully challenged on the grounds of procedural defects. This consideration led the ad hoc Committee in the Amco Asia case to assess the quality of the reasons in a claim for nullity based on an alleged failure to state the reasons for the award. Cf A Bjorklund, ‘The Continuing Appeal of Annulment: Lessons from Amco Asia and CME’, in T Weiler (ed), International Investment Law and Arbitration: Leading Cases from the ICSID, NAFTA, Bilateral Treaties and Customary International Law (London, Cameron May, 2005) at 479.

Based on the interpretation of the relevant provisions of the ICSID Convention, the prevailing opinion of jurists continues to support only formal reasons for the vacation of an arbitral award in ICSID arbitration. A similar situation prevails in most national jurisdictions and, therefore, it is possible for national courts to vacate an arbitral award in an investment dispute rendered by an ad hoc tribunal, also mostly for formal reasons.

As is shown below with regard to ICSID, the number of different post-award remedies reflects the complexity of the whole issue.

HJ Schlochauer, ‘Arbitration’, in R Bernhardt (ed), Encyclopedia of Public International Law, Vol I (Amsterdam and New York, Elsevier North Holland, 1992) at 215 et seq.

Ibid.

10  Ibid at 216. One cannot but agree with the comment by Hugo Perezcano who has greatly contributed to this chapter. According to him, ‘this is not necessarily the case. In the field of protection of aliens and their property, this used to be the case, but it has made way to investment protection including investor-State dispute settlement where the investor pursues the claim itself. It is not so in other international economic areas, e.g. international trade. While it is not considered diplomatic protection (although an argument could be made in that respect), it is States that defend their nationals' interests, for instance at the WTO. In other areas, there is increasing participation of private parties, e.g. human rights.’

11  Ibid.

12  Cf M Sornarajah, The Settlement of Foreign Investment Disputes (London, Kluwer Law International, 2000) at 315 ff.

13  Two other driving forces behind the development of the current system of investment arbitration were the following: (a) the need to protect investments in countries that had recently gained their independence (eg African countries in the post-World War II era) and (b) the need to foster development in less developed countries through foreign investment (the purpose behind the 1965 Washington Convention).

14  Introducing an investor-state dispute settlement system was probably also a way around the Calvo Doctrine by avoiding espousal of claims, thus removing one state from the equation.

15  For instance, the Elsi case involved a dispute between two states, the USA and Italy, concerning the application of an FCN treaty from the 1930s.

16  A Zimmermann and R Geiss, ‘Article 61’, in A Zimmermann, C Tomuschat, K Oellers-Frahm, C Tams, and T Thienel (eds), The Statute of the International Court of Justice: A Commentary (Oxford, Oxford University Press, 2006) at 1301 fn 3.

17  Ibid at 1302 ff.

18  This means that the term used here is not limited to the notion generally used in international commercial arbitration, but is meant in a broader sense.

19  We have in mind error of law (procedural or substatntive), fundamental error of fact, plain errors of fact.

20  H Grotius, De iure belli ac pacis—libri tres, in quibus ius naturae et gentium item publici praecipua explicantur (1625) ch. XX, sect. XLVI-2.

21  E de Vattel, Le Droit des gens, ou principes de la loi naturelle, appliqués à la conduite et aux affaires des nations et des souverains (London, 1758) lib II, ch 18, sect. 329.

22  Arbitration as such is naturally an even older concept and the settlement of disputes between city states which can be traced back as early as the ancient pre-classical era. It can be found also in the late Middle Ages, always with varying significance in different historical periods. This type of dispute settlement, however, cannot be considered as the origin of modern international arbitration, since these ancient practices exercised no influence on international arbitration as it is known at present. Cf H Lammasch, Die Lehre von der Schiedsgerichtsbarkeit (1914).

23 ‘Stari autem debet sententiae arbitri quam dixerit, sive aequa sit sive iniqua; et sibi imputet qui comprimiserit, nam et divi Pii rescripto adjicitur: vel minus probabilem solutionem sequo animo ferre debet’ (L 27, sect. 2, D de receptis 4, 8), quoted in B Kućera, International Judgement: Studies in International Proceedings (in Czech) (Prague, Orbis, 1935) at 125.

24 ‘Ceterum si condemnavit vel absolvit, dum arbiter esse desierit, mutare sentientiam non posse’ (L 19, sect. 2, D de receptis 4, 8 (Paulus)), quoted according to Kućera, ibid.

25 ‘Arbiter etsi erraverit in sententia dicenda, corrigere eam non potest’ (Gaius D 4, 8, 20), quoted according to J Kincl, ‘Dicta et regulae iuris or Roman Legal Wisdom’ (in Czech) (Prague, Charles University, 1990) at 43.

26  L Goldschmidt, ‘Projet de reglement pour tribunaux arbitraux internationaux’ paper presented at l'Institut de droit international (Session de Genève, 1874) in Revue de droit international et de législation comparée (VI, 1874) at 449–50.

27  M Novakovich, Les Compromis et les arbitrages internationaux du XIIe au XVe siècle (1905) at 81.

28  Kućera, above n 23.

29  Grotius, above n 20: ‘Nulla enim hic est potestas superior quae promissi vinculum aut impediat, aut solvat. Standum ergo omnino sive aequum, sive iniquum pronuntiaverint, ita ut Plinii illud huc recte aptes: summium quisque causae iudicem facit, quemcunque eligit’.

30  de Vattel, above n 21.

31  E Nys, ‘La revision de la sentence arbitrale’, in Revue de droit internationale et de législation comparee (12 1910) at 600.

32  J G Bluntschli, Das moderne Völkerrecht der zivilisierten Staaten als Rechtsbuch dargestellt (Nördlingen, 3rd edn, 1878) at 277.

33  Cf A Winterová, Civilní právo procesní (Civil Procedure law) (Linde, 1999) at 443 ff.

34  This is a legacy of commercial or international commercial arbitration that also sporadically allows some kind of internal control in institutionalized arbitration, such as scrutiny by the Court in ICC arbitration. Art 27 of the ICC Rules deserves some attention. Any effort to improve the ICSID Rules, it would require attention to the following provision: ‘Before signing any Award, the Arbitral Tribunal shall submit it in draft form to the Court. The Court may lay down modifications as to the form of the Award and, without affecting the Arbitral Tribunal's liberty of decision, may also draw its attention to points of substance. No Award shall be rendered by the Arbitral Tribunal until it has been approved by the Court as to its form’.

35  Dispute Settlement Understanding Art 17.6.

36  The author has problems with terminology and as it was pointed out, expressions such as ‘appellation’, ‘cassation’, and ‘revision’, can have different meanings in different contexts. In the above-mentioned contexts, these terms have the meaning given to them in the continental civil law tradition.

37  It would appear that if judicial review before municipal courts is the rule, it would be systematic. On the other hand, it can be said, what seems to be unsystematic is the fact that the decisions of the tribunal at the international level are reviewed by national judicial authorities. By analogy, one can hardly imagine that, let us say, decisions of the European Court for Human Rights would be reviewed by the national courts. Moreover, another problematic feature is or can be a different standard of review from state to state.

38  Canada offers a good example, as each Canadian province's judicial authorities may have a slightly different approach to the review of arbitral awards. It should be noted that the differences do not concern by any means only the revision of the award, but the whole process. An illustrative example is, for instance, the award issued in the dispute between Mexico and Metalclad Corporation under Chapter 11 of the North American Free Trade Agreement (NAFTA). Cf eg A Alvarez, ‘Setting Aside Additional Facility Award: The Metalclad Case’, in E Gaillard and Y Banifatemi (eds), Annulment of ICSID Awards, IAI International Arbitration Series No. 1 (Paris, Juris Publishing, 2004) at 267 ff.

39  I am grateful to Hugo Perezcano for this observation.

40  Art 52.4–5 ICSID Convention.

41  This happens mostly when some fundamental requirements are not met, eg if there is no legal dispute, there cannot be any award on merit, the dispute does not arise directly out of investment, the parties do not meet certain conditions, the claimant is not a national of a contracting state, or the respondent is not a contracting state etc.

42  See Lauder v Czech Republic and CME v Czech Republic cases.

43  Only future developments will show whether such an attitude is tenable and for how long. The danger of this situation could rest in an erosion of confidence in dispute settlement that does not properly prevent or safeguard against possible problems. It is not of much importance that such errors may happen only sporadically. Just a few controversial awards may disrupt the whole system.

44  See eg Klockner v Cameroon, Decision on Annulment, 3 May 1983, 2 ICSID Reports 119; Amco v Indonesia, Decision on Annulment, 16 May 1986, 1 ICSID Reports 515; MINE v Guinea, Decision on Annulment, 22 December 1989, 4 ICSID Reports 87.

45  AMCO Asia Corp et al v Republic of Indonesia, Annulment Decision, 25 ILM 1439 (1986) at 6, at para 23.

46  Non-application of applicable law or application of selected parts of it does not necessarily mean that the outcome would vary substantively.

47  Svea Court of Appeal, Department 16, Judgment, Case No T 8735-11 (in the matter of a challenge to the arbitral award in CME v Czech Republic) at 90 ff.

48  Ibid at 91.

49  Ibid.

50  Ibid at 92 ff.

51  Ibid.

52  Comments relating to Applicable Law on the Stockholm Tribunal's final Award of 14 March 2003 by Christoph Schreuer, at 1 and 2. Available at <http://www.univie.ac.at/intlaw/schreu.htm>.

53  As concerns governing law or applicable law, one should note the quite different approach of the arbitral tribunal in Saluka v Czech Republic compared to the CME tribunal.

54  Not all legal documents are clear to all readers in the same way. The French Civil Procedure Code in its Book IV—Arbitration, Title IV—Means of Recourse, Arts 1481 ff speaks of appeal available against an arbitral award (Arts 1482, 1483, etc). I suppose that it is not appeal as such and that it concerns only revision and setting aside the arbitral award. Irrespective of some difficulties with terms, the grounds for setting aside under Art 1484 are common to all other systems, including public policy rule (see eg, UNCITRAL).

55  The grounds for annulment of an arbitral award at the request of a party are given in Art 31 of the Arbitration Procedure Act (or Act No. 216/1994 CoL), in case of re-opening (revision) of proceedings, in the provisions of s 228, para 1(a) and (b) of the Rules of Civil Procedure. Article 31 (dealing with the ‘Setting aside of an arbitration award by court and cessation of an ordered enforcement of a decision’) states: ‘Upon the request of any of the parties, the court shall set aside the arbitration award if (a) no arbitration agreement can be concluded in the concerned case; (b) the arbitration agreement is null and void for other reasons, was cancelled or does not apply to the concerned case; (c) any of the arbitrators who took part in the case was not called on to decide on the case on the basis of the arbitration agreement or otherwise or was not capable to become an arbitrator; (d) the arbitration award was not approved of by the majority of arbitrators; (e) the party was not provided with the possibility to discuss the case before the arbitrators; (f) the arbitration award adjudges the party to a performance that was not requested by the entitled party or that is impossible or unlawful under Czech law; (g) it becomes clear that reasons for resumption of civil judicial proceedings are given in the case’ (Civil Procedure Code, Art 228, para 1 (a) and (b))

56  Art 190 (on Finality—Action for Annulment) of the Swiss Federal Statute on Private International Law states as follows: ‘ … (1) The award is final from its notification; (2) The award may only be annulled: (a) if the sole arbitrator was not properly appointed or if the Arbitral Tribunal was not properly constituted; (b) if the Arbitral Tribunal wrongly accepted or declined jurisdiction; (c) if the Arbitral Tribunal's decision went beyond the claims submitted to it, or failed to decide one of the items of the claim; (d) if the principle of equal treatment of the parties or the right of the parties to be heard was violated; (e) if the award is incompatible with public policy; (3) Preliminary awards can be annulled on the grounds of the above paras 2(a) and 2(b) only; the time limits runs from the notification of the preliminary award.’

57  The FAA provides that a court may vacate an arbitration award and, at its discretion, order a re-hearing only (1) where the award was procured by corruption, fraud, or undue means; (2) where there was evident partiality or corruption in the arbitrators, or either of them; (3) where the arbitrators were guilty of misconduct in refusing to postpone the hearing, upon sufficient cause shown, or in refusing to hear evidence pertinent and material to the controversy or of any other misbehaviour by which the rights of any party have been prejudiced; or (4) where the arbitrators exceeded their powers, or so imperfectly executed them that a mutual, final, and definitive award upon the subject matter submitted was not made (9 USC§ 10 (a)). The Uniform Arbitration Act (‘UAA’) provides for review under similar circumstances as does Revised Uniform Arbitration Act (‘RUAA’). The reason for limited review is seen in judicial hostility to arbitration in which a court seemed likely to engage in ‘unlimited review’ of an arbitration decision. Broadening of the scope of reasons was not that frequent but it is notable that the US Supreme Court in Wilko v Swan created an additional standard of review for vacating arbitration awards: manifest disregard of the law. Even where the arbitrator did not exceed the power granted by the contract, a result could be so egregious that a court would not let it stand.

58  UNCITRAL Model Law on International Commercial Arbitration (UN Doc A/40/17, annex I) (as adopted by the United Nations Commission on International Trade Law on 21 June 1985).

59  An arbitral award rendered in Russia cannot be appealed. According to Art 34 of the International Arbitration Act (IAA), upon application of a party an arbitral award may be set aside by the competent Russian court if the party making the application furnishes proof that: a party to the arbitration agreement was under some incapacity, or said agreement is not valid under the law to which the parties have subjected it (or in the absence of such an agreeement—under the law of the Russian Federation); or the party making the application was not duly notified of the appointment of an arbitrator or of the arbitral proceedings or was otherwise unable to present its case; or the award was made in relation to a dispute not contemplated by or failling within the terms of the arbitration agreement, or contains a decision on matters beyond the scope of the arbitration agreement; where only part of an award has been rendered outside the scope of the arbitration agreement, then only that part of the award may be set aside; or the composition of the arbitral tribunal or the arbitral procedure was inconsistent with that agreed by the parties, unless such agreement was in conflict with a provision of the IAA from which the parties cannot deviate; or if the court finds ex officio that: the subject-matter of the dispute is not capable of settlement by arbitration under Russian law; or the award is in conflict with Russian public policy.

60  Chapter VII (on Recourse Against Arbitral Award) of the 1996 Arbitration and Conciliation Act reads as follows:

I.  Recourse to a Court against an arbitral award may be made only by an application for setting aside such award in accordance with sub-section (2) and sub-section (3).

II.  An arbitral award may be set aside by the Court only if—

a.  the party making the application furnishes proof that—

i.  a party was under some incapacity; or

ii.  the arbitration agreement is not valid under the law to which the parties have subjected it or, failing any indication thereon, under the law for the time being in force; or

iii.  the party making the application was not given proper notice of the appointment of an arbitrator or of the arbitral proceedings or was otherwise unable to present his case; or

iv.  the arbitral award deals with a dispute not contemplated by or not falling within the terms of the submission to arbitration, or it contains decisions on matters beyond the scope of the submission to arbitration:

b.  Provided that, if the decisions on matters submitted to arbitration can be separated from those not so submitted, only that part of the arbitral award which contains decisions on matters not submitted to arbitration may be set aside; or

i.  the composition of the arbitral Tribunal or the arbitral procedure was not in accordance with the agreement of the parties, unless such agreement was in conflict with a provision of this Part from which the parties cannot derogate, or, failing such agreement, was not in accordance with this Part; or

c.  the Court finds that—

i.  the subject-matter of the dispute is not capable of settlement by arbitration under the law for the time being in force, or

ii.  the arbitral award is in conflict with the public policy of India. Explanation: Without prejudice to the generality of sub-clause (ii), it is hereby declared, for the avoidance of any doubt, that an award is in conflict with the public policy of India if the making of the award was induced/or affected by fraud or corruption or was in violation of section 75 or section 81.

III.  An application for setting aside may not be made after three months have elapsed from the date on which the party making that application had received the arbitral award or, if a request had been made under section 33, from the date on which that request had been disposed of by the arbitral Tribunal: Provided that if the Court is satisfied that the applicant was prevented by sufficient cause from making the application within the said period of three months it may entertain the application within a further period of thirty days, but not thereafter. On receipt of an application under sub-section (1), the Court may, where it is appropriate and it is so requested by a party, adjourn the proceedings for a period of time determined by it in order to give the arbitral Tribunal an opportunity to resume the arbitral proceedings or to take such other action as in the opinion of arbitral Tribunal will eliminate the grounds for setting aside the arbitral award.

61  Ibid Art 34. Application for setting aside as exclusive recourse against arbitral award:

(1)  Recourse to a court against an arbitral award may be made only by an application for setting aside in accordance with paragraphs (2) and (3) of this article.

(2)  An arbitral award may be set aside by the court specified in article 6 only if:

  1. (a)  the party making the application furnishes proof that: (i) a party to the arbitration agreement referred to in article 7 was under some incapacity; or the said agreement is not valid under the law to which the parties have subjected it or, failing any indication thereon, under the law of this State; or (ii) the party making the application was not given proper notice of the appointment of an arbitrator or of the arbitral proceedings or was otherwise unable to present his case; or (iii) the award deals with a dispute not contemplated by or not falling within the terms of the submission to arbitration, or contains decisions on matters beyond the scope of the submission to arbitration, provided that, if the decisions on matters submitted to arbitration can be separated from those not so submitted, only that part of the award which contains decisions on matters not submitted to arbitration may be set aside; or (iv) the composition of the arbitral tribunal or the arbitral procedure was not in accordance with the agreement of the parties, unless such agreement was in conflict with a provision of this Law from which the parties cannot derogate, or, failing such agreement, was not in accordance with this Law; or

  2. (b)  the court finds that: (i) the subject-matter of the dispute is not capable of settlement by arbitration under the law of this State; or (ii) the award is in conflict with the public policy of this State.

(3)  An application for setting aside may not be made after three months have elapsed from the date on which the party making that application had received the award or, if a request had been made under article 33, from the date on which that request had been disposed of by the arbitral tribunal.

(4)  The court, when asked to set aside an award, may, where appropriate and so requested by a party, suspend the setting aside proceedings for a period of time determined by it in order to give the arbitral tribunal an opportunity to resume the arbitral proceedings or to take such other action as in the tribunal's opinion will eliminate the grounds for setting aside.

62  This note has been prepared by the Secretariat of the United Nations Commission on International Trade Law (UNCITRAL) for informational purposes only; it is not an official commentary on the Model Law. A commentary prepared by the Secretariat on an earlier draft of the Model Law appears in document A/CN.9/264, reproduced in UNCITRAL Yearbook, vol XVI: 1985 (UN publication, Sales No. E.87.V.4).

63  Explanatory note at para 44.

64  Cf eg H Van Houtte, ‘Article 52 of the Washington Convention. A Brief Introduction’, in Gaillard and Banifatemi, above n 38 at 11.

65  Ibid.

66  Klockner Industrie-Anlagen GmbH and others v United Republic of Cameroon and Société Camerounaise des Engrais (Klockner I), Decision on Annulment, 3 May 1985, 2 ICSID Rep at 95; Klockner Industrie-Anlagen GmbH and others v United Republic of Cameroon and Société Camerounaise des Engrais (Klockner II), Decision rejecting the request of parties to annul the award, 17 May 1990.

67  Amco Asia Corporation and others v Republic of Indonesia (Amco I), Decision on Annulment, 16 May 1986, 1 ICSID Rep at 509. Amco Asia Corporation and others v Republic of Indonesia (Amco II), Decision rejecting the request of parties to annul the award and annulling the decision on additional decisions and correction, 19 December 1992.

68  Maritime International Nominees Establishment v Republic of Guinea, Decision partly annulling the award, 22 December 1989, 4 ICSID Rep (1997) at 79.

69  Compania de Aguas del Aconquija SA and Vivendi Universal v Argentina Republic, Decision on the annulment application, 3 July 2002, 42 ILM (2002) at 1135.

70  Wena Hotels Limited v Arab Republic of Egypt, Decision on Annulment, 5 February 2002, 41 ILM (2002) at 933. The case was closed before the ad hoc committee decided.

71  Philippe Gruslin v Malaysia, Resolution on suspension of proceedings was issued according to Arbitration Rule 43, para 1, 4 April 1996.

72  C Schreuer, ‘Three Generations of ICSID Annulment Proceedings’ in Gaillard and Banifatemi, above n 38 at 22.

73  Philippe Gruslin v Malaysia (Case No. ARB/99/3), proceeding discontinued for lack of payment of advances pursuant to Administrative and Financial Regulation 14(3)(d); Order for the discontinuance of the proceeding issued by the ad hoc committee on 2 April 2002.

74  Having regard to the ‘public’ nature of investment disputes, it could be argued that the decisions of arbitration panels and of the ad hoc committees (ie the decisions of the ‘first and second instances’) should be published automatically. The increasing public pressure on the transparency of arbitration proceedings in the cases involving public funds is more than obvious and the investor who wants to protect his interest should not have any reasons to object. As can be seen from a number of arbitration proceedings, it is very often the investor who endeavours to publish information on the progress and the result of such proceedings. It must be admitted, however, that even transparency has certain limits and the burden of proof in the case of a request for non-publication of information on the progress and result of an investment dispute should lie with that party to the dispute which invokes the secrecy or non-transparency of the proceedings. To a certain extent, this feature of the settlement of investment disputes in arbitration proceedings can be contrasted with commercial arbitration, in which all the parties to the dispute are, as a rule, private natural or legal persons. Cf eg V Balaš, ‘Confidentiality of Arbitration Proceedings? Yes, But ! …’ (in Czech), 6 Právní fórum (2004) at 223–8.

75  Van Houtte, above n 54 at 14.

76  Cf C Schreuer, ‘Three Generations of ICSID Annulment Proceedings’, in Gaillard and Banifatemi, above n 38 at 17 ff.

77  Cf eg E Schwarz, ‘Finality at What Costs? The Decision of the Ad Hoc Committee in Wena Hotels v Egypt’ in Gaillard and Banifatemi, above n 38 at 43 and I Suarez Anzorena, ‘Vivendi v Argentina: On the Admissibility of Request for Partial Annulment and the Ground of Manifest Excess of Powers’, ibid at 123 ff.

78  Schreuer, ‘Three Generations of ICSID Annulment Proceedings’, above n 72 at 19.

79  See in detail, Asif H Qureshi, Ch 28 (‘An Appellate System in International Investment Arbitration?’) in this volume.

80  With reference to the character of commercial arbitration, such developments probably cannot be expected in the near future, since the interest of the parties continues to be in speedy settlement.

81  9 USC §§ 1–16 (1994).

82  See CD Brown, ‘Increasing Arbitration's ‘Appeal’: Contracting for Greater Judicial Review of Arbitrator's Decisions’, The Metropolitan Counsel (August 2003) at 28.

83  P Lalive, ‘Concluding Remarks’, in Gaillard and Banifatemi, above n 38 at 298.

84  See, for more detail, August Reinisch and Loretto Malintoppi, ch 18 (‘Methods of Dispute Resolution’) in this volume.

85  Ibid, 311.

86  It should be pointed out that awards in investment disputes are rendered by highly qualified jurists. The only marginal number of awards was challenged successfully. If any kind of quasi- appellate system was allowed, it is quite clear that this would become a regular part of investment dispute settlement. I am not sure that this is a price worth paying.

The author approached this topic with a firm belief that it is time to change the system and allow broader review. My initial enthusiasm has almost disappeared by the end. The outcome of this encounter with the review of arbitral awards in investment disputes is: ‘I really do not know’.