1. Introduction

The idea of an appeal procedure in arbitration is generally regarded with some, if not substantial, scepticism within the arbitration community. This is not entirely surprising, for the concept would appear to fly in the face of the very premise on which arbitration is built, namely that it is a procedure in which parties give the arbitrators a mandate to decide - and terminate - a dispute quickly, quietly and efficiently. Also, the fact that parties are entitled to choose the arbitrators who decide their case is another argument for excluding a review of the award on the merits. Furthermore, the notion of appeal conjures up the uncomfortable prospect of rigid and cumbersome processes, delaying tactics and additional costs.

Yet, true as this may be, the possibility of an appeal in arbitration is an issue that periodically comes under review and various well-argued papers have been published in its favour.2 The matter is also on the desks of many arbitral institutions, several of which have already incorporated it into their rules. Although some jurisdictions appear to be addressing the issue more actively than others, discussions over appeal procedures in arbitration transcend jurisdictional limits and, to a greater or lesser extent, are taking place worldwide.


Spain is one of the jurisdictions that have been receptive to the idea. By summer 2015, no less than four arbitral institutions operating in Spain have rules providing for appeals in arbitration (and more are thinking about it).3 Somewhat inconveniently from a user's perspective, each takes a different approach.

The purpose of this essay is not to provide arguments for or against the appropriateness of appellate procedures in arbitration, but rather to summarize the situation as it exists today and to consider some of the issues such an appeal procedure raises.

As the leading global arbitral institution, the ICC has pioneered many developments in the field of arbitration, which have set benchmarks and helped establish principles recognized worldwide. While admitting that the issue of an arbitral appeal procedure is not an easy one to tackle, if there is a forum where it should above all be debated, it is without doubt the ICC.

2. The underlying discussion: finality vs fairness

Finality first

One of arbitration's main advantages is the fact that, as a general principle, awards of arbitral tribunals are not subject to review on the merits by the courts. In other words, courts cannot review the correctness of the tribunal's evaluation of the facts or application of the law to those facts. The finality of arbitral awards is generally considered to be an important contributing factor to quicker termination of a dispute than in court litigation. Yet, hand-in-hand with finality comes the fear that it might open the door to unfair results. If the merits of the arbitral tribunal's decisions are not subject to review - whether judicial or otherwise - legitimate concerns arise over the possibility of mistaken and/or unfair decisions.

Parties' perceptions of whether finality or fairness should predominate may well coincide at the beginning of an arbitration, but once an award has been rendered, they usually diverge, with the winner praising finality, while the loser defends fairness and tries to ensure that no unfair or incorrect decision has been made.4 In other words, the desirability of finality is absolute when one is on the winning side, but otherwise not necessarily so. It is safe to say that at the present [Page375:] time the prevailing opinion among lawyers, arbitrators and parties in international arbitration is that finality is a principle of arbitration that may not be forfeited. In sum, it triumphs over fairness.

Fairness not forsaken

Yet, there are indications that the concept of an appeal in arbitration is not only alive but progressively gaining ground.

a) From a user's perspective, the lack of a review on the merits continues to be a cause of concern. A survey conducted among Fortune 1000 companies in 20115 provides useful insights, all the more so as a very similar survey was conducted in 1997, which allows interesting comparisons to be made.

The survey contains the following table listing the reasons why companies have not used arbitration:6

The results are mixed. While some of the figures show positive developments for arbitration (e.g. substantial reduction in resistance from senior management, rise in corporate experience, increased confidence in third party neutrals and a better supply of qualified third [Page376:] party neutrals), concerns over cost and time have increased and the lack of appeal, although slightly less worrying to companies in 2011, remains a significant barrier to the use of arbitration and the most important of all the reasons listed. This does not of course mean that over 50% of users wish an appeal system to be introduced. A 2006 study by the School of Arbitration at Queen Mary, London7 showed that only a small percentage of the in-house counsel interviewed (9%) were in favour of some form of appeal in arbitration.

It would seem reasonable to conclude that users are concerned about the threat to fairness if there is no appeal procedure, but do not want to forego the benefits of finality (e.g. speed and limitation of costs). Any arbitral appeal system would therefore need to balance these competing needs.

b) From the perspective of arbitration providers, there has been a slow but significant evolution. In 2012, none of the leading international arbitral institutions (AAA, LCIA and ICC) had an appeal mechanism in their rules. However, three private ADR institutions (CPR, JAMS and the European Court of Arbitration, a private institution based in Strasbourg with a branch in Spain) did. Since then, the AAA, through its international branch, ICDR, and, in Spain, three Spanish institutions have introduced into their rules provisions allowing for appeals. All in all, it would seem that the pressure of demand in the arbitration market is slowly changing the perspective of some arbitration service providers on this issue.

c) Finally, it should be noted that the Pepperdine Survey provided data on the likelihood of the companies interviewed using arbitration in the future. Again, an interesting comparison is made between 1997 and 2011 (where the date related to corporate/commercial disputes):8

Although a clear trend emerges,9 it should be recalled that the data relates essentially to US domestic use of ADR. International arbitration offers substantial advantages in transnational disputes, including the convenience of avoiding foreign courts, the aid offered by the 1958 New York Convention in facilitating the enforcement of arbitral awards, [Page377:] flexibility in the procedure and the appointment of arbitrators, which all help to optimize proceedings. Consequently, it would be inappropriate to consider that the same trend is true of international arbitration. Nonetheless, the above figures serve as a reminder that arbitration - whether domestic or international - is a service industry and, as such, needs to be alert to users' inclinations and needs. Although today it is the best alternative to litigation in foreign courts, this can always change.

3. Commercial arbitration and investment arbitration: two different appellate models

A discussion of the question of an appeal procedure in arbitration should distinguish between commercial and investment arbitration. Although both types of arbitration share the same principles, there are some important differences.10

Commercial arbitration deals essentially with disputes arising from contracts between two or more parties, generally in the private sector. These are often international contracts involving parties of different nationalities, and the applicable law will be either a defined national law or, less often, general principles of international commercial law. The arbitrators who decide the dispute may be of any nationality and legal background, not necessarily related to the substantive law applicable to the contract (although their main task will be to issue an award on the basis of that law). Generally speaking, the principle of confidentiality is fully applicable to commercial arbitrations.

By contrast, investment arbitration deals, on the whole, with disputes between private persons and a state. The claim is generally based on an instrument of international public law, such as a bilateral investment treaty (BIT) or a multilateral agreement between states that provides private parties with standing to claim against a state. Although the wording of such treaties may vary, they share similar concepts, such as that of fair and just compensation. Arbitrators decide investment disputes pursuant to the terms of the relevant treaty, and 'by interpreting and applying these terms, investment arbitrators contribute to the development of the meaning of these substantive terms'.11 Unlike commercial arbitration, investment arbitration is exposed to much publicity, and many of the awards are published.

In light of these differences, Irene Ten Cate proposed two distinct arbitration models:12

Commercial arbitration is best understood as a pure 'dispute resolution', while investment arbitration incorporates elements of a 'public values' model of adjudication.


It follows from this general principle that the goals pursued by the appellate process in commercial arbitration are different from those pursued in investment arbitration. In commercial arbitration, the appeal process is essentially for the purpose of correcting mistakes made by the arbitral tribunal, and so it has above all a corrective function.13 It is argued that in investment arbitration the appellate procedure is focused on law-making, that is to say its purpose is to create arbitral case law that will contribute to certainty in the interpretation and application of the treaties.14

Given this distinction, it can be assumed that the question of an appeal procedure will have different implications in each model. This essay addresses appellate arbitral procedures within the context of commercial arbitration, and the comments made herein are not necessarily applicable to investment arbitration.

4. Overview of existing options

There are today a variety of possibilities that exist for reviewing arbitral awards. They include (i) annulment actions brought in state courts, chiefly based on the formal grounds listed in the UNCITRAL Model Law on International Commercial Arbitration; (ii) appeal on the merits in local courts; (iii) appeal on the merits before an arbitral tribunal; and (iv) simple institutional review of the draft award prior to being issued.

Annulment of award

An annulment action is a judicial remedy aimed at setting aside the award. It is usually foreseen in modern arbitration laws based on the UNCITRAL Model Law. According to Article 34 of the Model Law, a party may seek the annulment of an award on the basis of the six grounds summarized below:

• the arbitration agreement is not valid;

• lack of proper notice of the appointment of the arbitrators or of the arbitral proceedings or inability to present one's case;

• the award deals with matters outside the scope of the submission to arbitration;

• the composition of arbitral tribunal or the arbitral proceedings were not in accordance with the parties' agreement;


• the subject matter of the dispute is not arbitrable under the state's laws;

• the award is in conflict with the public policy of that state.

Annulment actions are not appeal procedures and are not intended to allow a court to review the merits of the award. Their purpose is not to correct errors in the assessment of evidence or facts or the application of the law, but rather to protect the integrity of the arbitral process by ensuring that the parties have validly agreed to submit to arbitration, that the arbitration has been conducted in accordance with their wishes and applicable law, that the matter is arbitrable under relevant law, that each party has had the chance to make its case, and that there is no infringement of public policy.15

When ruling on the annulment action, the court will generally not be able to modify the challenged award but simply either dismiss the challenge and uphold the award or find in favour of the challenger and vacate the entire award.

Article 34(4) of the UNCITRAL Model Law provides that, where appropriate, the court may suspend the setting aside proceedings for a period of time to allow the arbitral tribunal to resume its proceedings or to take other action likely to eliminate the grounds for setting aside the award. This provision offers a means of keeping the arbitration alive so that mistakes are removed and a valid award can be rendered. However, where this provision or a similar provision has not been incorporated into national law, there is no easy return to the arbitral tribunal, as it will be considered functus officio.16

Institutional rules: Of the various institutions considered for the purpose of this note, the ICC is the only one whose rules address the remand of an award by a court to the original arbitral tribunal. Article 35(4) of the ICC Arbitration Rules foresees that in such circumstances the tribunal may seize the award anew and issue an addendum or a revised award pursuant to the terms of the remission.

Judicial appeal

The possibility of an appeal on substantive grounds is provided in some jurisdictions, some of which limit it to domestic arbitration.

Section 69 of the 1996 English Arbitration Act allows appeals in international arbitration under certain conditions:


• The possibility of appeal is an opt-out provision: the appeal procedure will apply unless waived by the parties.

• Appeals can be made only on questions of law, not on issues of fact.

• Leave to appeal must be obtained from the court, unless all parties agree to the appeal.

• The court will grant leave only if it is satisfied that, on the basis of the findings of fact in the award, the decision of the tribunal on the question is obviously wrong, or the question is one of general public importance and the decision of the tribunal is at least open to serious doubt.17

Judicial appeal is not the best option for increasing fairness in international commercial arbitration. Submitting to the review of a state court has the same drawbacks as the submission of the original dispute to state courts, e.g. the matter is entrusted to local magistrates rather than international arbitrators chosen for their specific expertise, confidentiality is compromised, and the parties lose control over proceedings.

Arbitral appeal

The concept of an appeal within arbitration is not new and has indeed been applied to no apparent disadvantage in some specific sectors. For example, international commodity trading has produced various arbitration systems that include appellate procedures.18 The Court of Arbitration for Sport also provides for appeals within its procedural rules. Although the concept is absent from the UNCITRAL Model Law, it is nonetheless evoked in paragraph 45 of the UNCITRAL Secretariat's Explanatory Note:

However, a party is not precluded from appealing to an arbitral tribunal of second instance if the parties have agreed on such possibility (as is common in certain commodity trades).


Several national arbitration laws (e.g. the Netherlands, Israel) allow for an arbitral appeal, subject to the parties' agreement. In essence, the appellate procedure consists of a review of the substance of an award by a second arbitral tribunal.

However, the variety of approaches found in national laws and institutional rules shows that there is no one appellate procedure that has gained universal acceptance. Different positions are taken on such matters as scope (should the appeal cover questions of law only or also questions of fact?), the extent of review (ex novo or limited to a review of the original arbitration proceedings?), whether leave is required to appeal, and the composition of the arbitral tribunal that hears the appeal. Such disparities create uncertainty over the value of an appeal system in international commercial arbitration.

Institutional review

Several arbitral institutions review draft awards internally. Exemplified by the ICC International Court of Arbitration's scrutiny process, this is an opportunity for the institution to make suggestions to the arbitral tribunal and request corrections, essentially of a formal nature. Some institutions sometimes call the arbitrators' attention to aspects relating to the merits, while leaving the arbitrators free to decide on whatever follow-up action they consider appropriate. Institutional reviews aim above all to ensure quality in the formal aspects of the awards, as opposed to the merits of the award. The process takes place between the institution and the arbitral tribunal; the parties are not involved. Hence, it is neither an appellate review nor a procedure designed to increase fairness.

5. Appeals in commercial arbitration

'One shot' as the basic principle in arbitration

Some advocates of finality consider that establishing an appeal procedure in arbitration would 'strike at the heart of the very concept of the arbitral process'.19 They argue that it would be yet another step towards judicializing arbitration procedure. Others, however, contend that finality is only as important as the parties believe it should be,20 and that parties should have the right to a second bite if they consider it appropriate. So, is the lack of an appeal a critical element in international arbitration?


One of the basic principles of arbitration is the control it offers to the parties, which allows them tailor the procedure to their specific needs.21 If, in light of their particular circumstances, the parties wish to incorporate an appellate procedure into their arbitration agreement for a given transaction, there would appear to be no reason, in principle, why they should be denied that right. In the words of Rowan Platt:22

Concerns that appeal mechanisms degrade the principle of finality, and strike 'at the heart of the very concept of the arbitral process' are therefore misplaced. So long as parties agree to prioritize correctness or fairness over finality, this accords with the spirit of arbitration.

Advantages of an appellate system

As positive effects of an arbitral appeal system, the following can be highlighted:

• An appellate procedure would help dispel the doubts and fears of potential users and/or decision-makers (usually in-house lawyers) deterred from resorting to arbitration by its absence.

• An appellate procedure would encourage greater fairness by providing a means of correcting mistakes by the original arbitral tribunal.

• An appellate procedure would most probably enhance the quality of awards. The prospect of a review by a second panel of arbitrators is likely to incite the original panel to give maximum attention to the clarity of its analysis and reasoning and will probably raise the quality of the drafting, as arbitrators will wish to ensure their award is fully understood by the reviewing panel.23

Drawbacks of an appellate system

Opponents of an appellate system insist that it is not what users want. They argue that users are more concerned about limiting costs and the time taken to resolve a dispute and that adding another tier to the procedure would exacerbate these concerns. Furthermore, they point out that it is difficult to predict how the courts at the seat of the arbitration and the place of enforcement would regard the validity of an appealed award and the review and enforcement of the award resulting from the appeal, which would lead to legal uncertainty.


Institutional appeal vs contractual appeal

In theory, nothing prevents parties from agreeing contractually on an appeal procedure of their own devising, provided it complies with the applicable arbitration law. However, this is rarely seen in practice, due to the difficulty of negotiating an arbitration appeal procedure, whether in the context of ad hoc or institutional arbitration. If the parties do wish to agree on an appeal procedure, the safest route is to submit to the rules of an institution that provides for such a procedure.

6. Aspects of an appeal procedure

Authors have identified various factors that an appeal procedure should take into account. William H. Knull and Noah D. Rubins24 mentioned the following in a non-exhaustive list: (i) expedited procedure; (ii) scope and standard of review, (iii) monetary limits, (iv) cost shifting, (v) security for costs, (vi) sanctions, (vii) waiver of judicial remedies, (viii) standing body or ad hoc tribunal, (ix) remedies. They further suggest that the system should include modules allowing parties to choose from a range of options for each aspect of the appeal procedure (e.g. scope, form, costs, speed of appeal), with a default solution in the event of a divergence of views. Of the various aspects generally taken into consideration those discussed below would appear to stand out in particular.

Scope and standard of review

As far as the scope of the review is concerned, should it be conducted ex novo as a full arbitration,25 or be limited to issues of law, or include issues of fact, too? When the review is limited to issues of law, should there be a higher threshold for review, requiring the error to be of a certain magnitude? If the review extends to the facts, should new evidence be admitted? Let us consider these questions in more detail.

ex novo arbitration vs limited review

Although an ex novo arbitration would seem most likely to ensure fairness, this option proves highly unattractive for reasons of time and cost. As we saw above, the users of international arbitration tend to favour finality over fairness. Allowing the dispute to be heard for a second time ex novo would tip the scales in the other direction, giving priority to fairness over finality, which would be inconsistent with the overriding wishes of users.


However, there may be times when a limited review can be justified on grounds of proportionality. It would seem reasonable to limit finality, for instance, in the face of blatant unfairness, where decisions have been taken that are plainly wrong, whether there are mistakes in the interpretation or application of the law, or where facts and evidence have been incorrectly assessed.

Review of issues of law

Requiring the alleged mistake to be sufficiently serious prevents a wholesale review of all the merits. This narrows the scope of the review and reduces its impact on costs and time.

The legal analysis of a given situation may often lead to more than one valid answer. If the conclusion reached in the award is one of several possibilities (even though it may be less convincing than others), substituting one decision for another would not satisfy the requirement that the purpose of an appeal procedure should be to correct an error. The appellate tribunal would therefore be justified in making a different decision only if there is a clear mistake of law in the initial decision.26

The setting of a high threshold for reviewing issues of law:

• limits the appeal review to severe situations that justify a restriction on finality;

• is less likely to undermine the criteria used by the first tribunal to reach its decision;

• reduces the number of appeals filed, which are limited to situations where the appellant has a strong case.

Institutional rules: Some institutions, such as the European Court of Arbitration, provide for full ex novo appeal hearings (i.e. in essence, a second arbitration conducted under an expedited procedure). CPR provides that the original award may be challenged if it 'contains material and prejudicial errors of law of such a nature that it does not rest upon any appropriate legal basis' (Rule 8.2(a)).27 JAMS applies the same standard of review as would be applied by the local appellate court. ICDR does not apply a gravity threshold, requiring only that the underlying award be based upon 'an error of law that is material and prejudicial' (Optional Appellate Arbitration Rules A-10). Of the Spanish institutions, one does not set any limits to the grounds for appeal, while another requires a manifest infringement of the substantive norms on which the award is based.


Review of the facts

An appellate procedure may or may not include a review of the facts (the English Arbitration Act, for example, allows appeals only on questions of law).

As a relevant aspect of case analysis and evaluation, the arbitral tribunal's assessment of the facts of the case and the evidence presented or omitted (where this can be inferred) could be considered as equally eligible for review as a mistake in the law. Here, too, arbitrators are not immune to making mistakes that could lead to unfair results. As with issues of law, it would seem appropriate to set a high standard for reviewing facts and evidence to ascertain whether they support the conclusions reached by the initial arbitral tribunal.

Institutional rules: CPR provides as a ground for appeal the fact that the award 'is based upon actual findings clearly unsupported by the record' (Rule 8.2(a)). One of the Spanish institutions - the Corte Civil y Mercantil de Arbitraje - allows for a review when the award is based 'on a clearly erroneous assessment of the facts which have been of decisive importance' (Arbitration Rules, Article 54). ICDR mentions as a ground for appeal the fact that the underlying award is based upon 'determinations of fact that are clearly erroneous' (Optional Appellate Arbitration Rules A-10).

A review of the facts begs the question of how it is to be conducted. Should the evidence be presented and heard anew? The danger here is that the presentation of evidence may lead to additional cost and delay. The general tendency is to rely on the evidence on record in the first arbitral proceeding so as to avoid a further, time-consuming process of presenting evidence.

Institutional rules: CPR does not allow an appeal unless there exists a record of all hearings and evidence (e.g. including exhibits, depositions, affidavits admitted as evidence) in the arbitration proceeding against which the appeal is made. ICDR defines the record on appeal as being composed of documents (expert reports, depositions, affidavits) that were admitted as part of the arbitration hearing. JAMS provides that the record on appeal consists of the stenographic or other record of the arbitration hearing and all exhibits, depositions and affidavits admitted to the record. These provisions allow rapid access to the evidence produced in the initial arbitration without additional cost and delay.

One last point is whether new evidence should be admitted in the appellate proceedings. In general, rules tend to be reticent.

Institutional rules: CPR provides that 'the Tribunal may request the parties to supplement the Record initially submitted by the parties as it may deem appropriate' (Rules 7.3), while ICDR is more restrictive: 'A party may not present for the first time on appeal an issue or evidence that was not raised during the arbitration hearing' (Optional Appellate Arbitration Rules A-16). In Spain, the Corte Española de Arbitraje, mirroring Spanish judicial appeal procedure, provides that new [Page386:] evidence may be admitted if (i) it was requested and improperly denied in the first arbitration, or (ii) though admitted, could not be presented for reasons beyond the control of the party requesting it, or (iii) it came to light after the facts had been established or the requesting party proves that it had no knowledge of the evidence before that time.

Leave to appeal

The English Arbitration Act requires that leave of court be obtained if the parties do not agree to the appeal. One may ask whether institutional rules should also make it necessary to obtain leave to appeal. It is tempting to think that a quick, initial review of the request for appeal could be made, with the aim of excluding those brought for dilatory purposes. This seems to be the approach adopted in the UK.28

The matter raises important questions. For instance, who should review and decide on leave for appeal? The arbitral institution may be reluctant to take on this responsibility, while the appointment of another person or persons for this purpose raises concerns of time and cost. Another concern is the effect the decision would have on the tribunal hearing the appeal, not to mention the counterparty. Yet, the possibility of having a quick and cost-effective initial barrier to deter frivolous claims is an attractive proposition and deserves to be taken into consideration when contemplating an appeal mechanism.

Auctoritas of the appeal tribunal

One of the main issues in any arbitral appeal system is the appellate tribunal's authority to review the decisions of the initial tribunal. There is no doubt that its mandate, and therefore its authority to amend the original award, derives from the will of the parties. Given that this is also the source of the authority of the initial tribunal, why should the appellate tribunal's review and decision prevail?

In the case of judicial appeal, auctoritas can be argued to be based on questions such as the collegiate nature of the appeal tribunal (as opposed to a first instance court composed of a single judge), which contributes to a more balanced view of the issues, or the fact that appellate judges usually have more experience than those sitting in lower courts.


The situation is different in arbitration. Many cases (and certainly all major cases) are decided by collegiate tribunals, generally composed of three arbitrators. So the fact that the appellate tribunal is composed of three arbitrators is of little relevance in itself.29 Equally irrelevant is the argument of the greater experience of the members of the appellate panel. In the initial arbitration parties will choose the best arbitrators they can find for the case. Consequently, it is difficult to justify the auctoritas of the appellate tribunal on this basis. It may be that rules lay down specific requirements for the appellate tribunal, for example to ensure homogeneity among its members. They may, for instance, require that the arbitrators who hear the appeal have a certain level of experience of arbitration or of a specific sector, or have experience as former judges. Yet this again does not necessarily give the second panel greater professional authority than the first.

According to another argument, the auctoritas of the appellate tribunal stems not from the attributes the arbitrators should have, but from the manner in which they are designated. It is suggested that there should be no party-appointed arbitrators on appellate panels.30 This is based on the belief that a tribunal whose members are all appointed by an arbitral institution are more likely to perform an objective review than a tribunal with party-appointed co-arbitrators. If the initial tribunal was composed of a sole arbitrator, a three-member appellate tribunal would have added auctoritas by force of numbers.

Institutional rules: In general, institutional rules tend to allow parties to appoint arbitrators freely for the purposes of appeals. CPR allows them to choose from a list it draws up from its roster of neutrals.

Expedited procedure

There is no doubt that an appellate procedure should be conducted rapidly. Limiting its scope, using evidence already on record and having the tribunal designated by the institution all contribute to faster procedures. Expedited procedures help to limit costs too, as the arbitrators will spend less time on the case. Rather than calculating their fees on the basis of the time they spend on the appeal, it may be worth agreeing upfront on a fixed amount, as this will be an incentive to performing the mandate as quickly as possible and spare the arbitrators the formality of time-keeping. Consideration may also be given to setting a time limit for the appeal. Spain's arbitration law prior to 2011 contained a provision of this kind, according to which the tribunal's mandate terminated automatically if the award was not issued within a fixed period of time (six months from the filing of the answer to the [Page388:] claim), unless the parties agreed otherwise. Failure to abide by the time limit could leave the arbitrators liable if their failure was wilful or due to gross negligence.

Relationship between appealed and appellate awards

If parties do not agree to an appeal procedure, any partial or final award will resolve the issues they deal with definitively. They will be immediately enforceable and the relevant statutes of limitations will immediately start to run. If, however, the parties agree to the possibility of appeal, the initial resolution of the dispute will not be considered definitive, as it will be open to further review on the merits.

Two important consequences follow from this observation:

• The appealed award cannot be enforced, nor should it have any effect, until the appeal procedure is finished.

• Any statute of limitations applicable to the initiation of an annulment action against the award should start running on the date on which the appeal decision is issued.

Jurisdictions in which private arbitral appeals are possible generally consider the award becomes definitive at the outcome of the appeal procedure. In other jurisdictions these matters may be left unregulated, creating uncertainty when it comes to the application of the law. Arbitral institutions deal with these matters in different ways.

ICDR provides that the filing of a notice of appeal implies that the parties agree not to consider the underlying award as final for purposes of any court actions to modify, enforce, correct or vacate that award. It also provides that the running of any period allowed for the commencement of judicial actions related to the underlying award is suspended during the appeal (Optional Appellate Arbitration Rules A-2(a)).

Similarly, JAMS provides that upon the timely filing of an appeal, the award is no longer considered final for purposes of judicial enforcement, modification or annulment pursuant to the JAMS Arbitration Rules (Optional Arbitration Appeal Procedure, § C).

CPR likewise states that the timely filing of an appeal means that the original award will not be considered final for purposes of confirmation, enforcement, vacation or modification of the award. If the appeal tribunal upholds the original award, it is deemed final as of the date of the appeal tribunal's decision. If the appeal tribunal does not confirm the original award, the appellate award will be deemed the final award in lieu of the original award (Rule 2.3). CPR reinforces this provision by Rule 2.4, which provides that each party irrevocably waives the right to initiate a court action to confirm, enforce, vacate or modify the original award until the appeal process has been completed. Any statutes of limitations applicable to the commencement of court actions are suspended until the appeal proceeding has been completed.


In Spain there is no clear answer to the question of whether an appeal procedure renders the underlying award ineffective. Three of the Spanish arbitral institutions considered for this study do not address this issue at all in their rules.31 The fourth provides that an appeal against the award will not prevent the parties from introducing an annulment action before the competent courts. This suggests that the initial award is considered definitive and open to annulment regardless of the appeal procedure.

It should be pointed out that where institutional rules contain provisions for the appeal of awards, they should cover partial as well as final awards. So long as an award decides an issue in a definitive manner, it should be open to appeal if it contains a mistake in law or fact that needs correcting. Appeals against partial awards should be submitted within a limited period of time following the issuing of the award. If the appeal leads to a reversal of the original decision, waiting until the arbitration has been completed might be wasteful of both time and money if the subsequent stages of the proceedings prove to be unnecessary.

As a final remark on this aspect of the appellate procedure, it might be worth considering extending the appellate tribunal's mandate automatically to any further appeals that may arise in the same proceedings. This would ensure that knowledge gained during the appeal process is available for any subsequent appeal proceedings.

7. Conclusion

International arbitration has evolved and expanded substantially in the past twenty-five years and is today probably the best and most attractive binding alternative to litigation for international commercial transactions. Most users favour efficient, quality-driven proceedings, in which finality does not overlook the need for fairness. The answer is found in a system that offers the means of managing costs, running the process efficiently, and appointing skilled arbitrators who will render high-quality awards, which should not suffer from being unappealable. However, the progressive development of procedures allowing awards to be appealed seems to reveal market pressure for an adequate opportunity to seek redress in certain circumstances. The existence [Page390:] today of an array of different appeal alternatives adds uncertainty to a concept that appears counter-intuitive in arbitration. It remains to be seen in the coming years whether the solutions available today address, to the satisfaction of all those involved, all of the issues raised by the subject.

President of the Madrid Court of Arbitration; asp@arbit.es.

W.H. Knull, III & N.D. Rubins, 'Betting the Farm on International Arbitration: Is it Time to Offer an Appeal Option?' (2000) 11:4 The American Review of International Arbitration 531; E.P. Gleason 'International Arbitral Appeals: What are we so Afraid Of?' (2007) 7:2 Pepperdine Dispute Resolution Law Journal 269; I.M. Ten Cate 'International Arbitration and the Ends of Appellate Review', Marquette Law School Legal Studies Paper No. 12-21, (2012) 44 New York University Journal of International Law and Politics 1110; R. Platt 'The Appeal of Appeal Mechanisms in International Arbitration: Fairness over Finality?' (2013) 30:5 Journal of International Arbitration 531.

The four institutions are Corte Española de Arbitraje; Corte Civil y Mercantil de Arbitraje; Tribunal Arbitral del Ilustrísimo Colegio de Abogados de Valencia; and the European Court of Arbitration, through a branch operating in Valencia. The Madrid Court of Arbitration (Corte de Arbitraje de la Cámara de Comercio de Madrid), Spain's leading arbitration institution, has not included an arbitration appeal procedure in its rules.

See W.W. Park, 'Why Courts Review Arbitral Awards' in Law of International Business and Dispute Settlement in the 21st Century - Liber Amicorum Karl-Heinz Böckstiegel (2001) 595 at 596, quoted by R. Platt, supra note 2 at 534.

T.J. Stipanowich & R. Lamare, 'Living with ADR: Evolving Perceptions and Use of Mediation, Arbitration and Conflict Management in Fortune 1000 Corporations', Pepperdine University Legal Studies Research Paper No. 2013/6; (2013) 19 Harvard Negotiation Law Review 1 [hereinafter 'Pepperdine Survey'].

Ibid. at 53, table P.

International Arbitration: Corporate Attitudes and Practices, Queen Mary School of International Arbitration, University of London & PricewaterhouseCoopers (2006) 15, quoted by R. Platt, supra note 2.

Pepperdine Survey, supra note 5 at 51, table O.

Pepperdine Survey, supra note 5 at 51 ('The 2011 Fortune 1000 survey may be remembered as a tipping point in the modern history of mediation and arbitration, because it marks the point at which reliance on mediation contributed to a drop-off in arbitration…').

For an excellent and comprehensive study of this question, see I.M. Ten Cate, supra note 2.

Ibid. at 1119.

Ibid. at 1114.

Ibid. at 1110 ('In court systems … appellate review fulfills two principal functions: error correction and lawmaking. Error correction protects litigants against erroneous decisions and safeguards the integrity of adjudication. Lawmaking refers to the role of appellate courts in the development and harmonization of norms').

Ibid. at 1122 ('Error correction, at least when applied to legal determinations, is premised on the presumption that the law is relatively settled. In investment arbitration, it is precisely the lack of consensus of fundamental norms that is viewed as problematic').

Occasionally, courts have entered into the merits in annulment actions on the basis of an expansive interpretation of the concept of public policy. However, it is generally accepted in all modern judicial systems that public policy should be interpreted restrictively.

As a consequence, setting aside an award in an annulment process takes the parties back to square one: they will have to start the arbitration anew (unless the arbitration agreement has been found non-existent or invalid or the subject matter is not arbitrable, in which case they will have to resort to litigation in the courts).

English Arbitration Act, section 69(3):'Leave to appeal shall be given only if the court is satisfied-(a) that the determination of the question will substantially affect the rights of one or more of the parties, (b) that the question is one which the tribunal was asked to determine, (c) that, on the basis of the findings of fact in the award- (i) the decision of the tribunal on the question is obviously wrong, or (ii) the question is one of general public importance and the decision of the tribunal is at least open to serious doubt, and (d) that, despite the agreement of the parties to resolve the matter by arbitration, it is just and proper in all the circumstances for the court to determine the question.'

See GAFTA (Grain and Feed Trade Association), CTF (Coffee Trade Federation), among others.

P. Mayer, 'Seeking the Middle Ground of Court Control: A Reply to I.N. Duncan Wallace' (1991) 7 Arbitration International 310, quoted by R. Platt, supra note 2 at 532.

R. Platt, supra note 2 at 560.

Paragraph 35 of the Explanatory Note by the UNCITRAL Secretariat on the 1985 Model Law on International Commercial Arbitration as amended in 2006 states that: 'Autonomy of the parties in determining the rules of procedure is of special importance in international cases since it allows the parties to select or tailor the rules according to their specific wishes and needs, unimpeded by traditional and possibly conflicting domestic concepts…'

R. Platt, supra note 2 at 560.

For a more detailed review of this aspect, see I.M. Ten Cate, supra note 2 at 1147.

See W.H. Knull, III & N.D. Rubins, supra note 2 at 41.

This is the approach advocated by Prof. Mauro Rubino Sammartano, current president of the European Court of Arbitration, whose rules provide for full ex novo appeal proceedings.

This raises the question of the liability of arbitrators. How would the services provided by the original tribunal be regarded in light of a successful appeal based on a clear mistake? Appeals on the merits increase arbitrators' exposure to liability.

The award may also be challenged if subject to one or more of the grounds for vacating an award under section 10 of the Federal Arbitration Act (Rule 8.2(b)).

The Hon. Sir Anthony Colman, 'The Question of Appeals in International Arbitration' (2007), paper presented at the congress to celebrate the 40th annual session of UNCITRAL Vienna, 9(12 July 2007 ('However, the judges gave leave in very few cases. In the interests of achieving early finality, they took a strict line on the meaning of the decision being "obviously wrong"… As word got around amongst the lawyers that it was so difficult to obtain leave to appeal, there were fewer and fewer applications because they were seen as a waste of costs. There were so few in fact that the narrow gateway left open for appeals has by now effectively been virtually closed by lack of use.')

One could imagine an appellate tribunal composed of more than three arbitrators, but this an unrealistic hypothesis and would lead to additional costs and delay.

I.M.Ten Cate, supra note 2 at 1153 ('direct appointment of arbitrators by the parties creates incentives that are at odds with error correction. Appellate review in commercial arbitration can therefore be effective only if parties and institutions are willing to use another appointment model for the appellate panel').

J. M. Santos Vijande, a Spanish university professor, considers it unlikely that a Spanish court would grant interim enforcement of an award that is subject to an arbitral appeal. However, he adds that the decisions taken in the initial award that have not been affected by the appeal may be subject to general (i.e. not interim) enforcement. This would suggest that in his opinion the initial award may be partially definitive. He believes that the time limit for bringing an annulment action would start running from the date of the appeal award. See J.M. Santos Vijande, 'On the legal and constitutional feasibility of appeal in the arbitration procedure: particular analysis of the issues raised in relation to the action for annulment, the enforcement of the arbitration award and the admission and taking of evidence' Revista Internacional de Estudios de Derecho Procesal y Arbitraje, No. 1 2011 [in Spanish].