When the UNCITRAL Model Law on International Commercial Arbitration was approved by the United Nations General Assembly in 1985, “uniformity of the law of arbitral procedures” was a stated purpose. The uncertainty produced by the disparity among the national laws was one of the drafters’ concerns. The other was the inadequacy of domestic laws to govern arbitration for international disputes. In the Explanatory Note (Section A), national laws were described as “outdated”, “fragmentary” and/or “too domestic”, making them inappropriate for arbitration of international commercial disputes.

In 2006, the Model Law was modified on two substantive aspects: the form of the arbitration agreement under article 7 and provisional measures under article 17. All other provisions of the Model Law were left intact, save for a provision on the interpretation of the law. Did that reform go far enough?

This post is an attempt to spark a debate on whether the time has come for a more general reform of the Model Law or whether, to the contrary, the international arbitration community has evolved beyond the need for a pattern to follow.

As reported by GAR recently, since 2010 there have been legislative reforms in Australia, Hong Kong, Spain, France, Portugal, and Singapore. Within the same time frame, Costa Rica, Mexico, and Colombia have also modernized their international arbitration regimes, and The Netherlands is currently revising theirs. If we add legal reforms since the year 2000, there are more than 30 modern international arbitration legal regimes.

Of these, most are based on the Model Law, but how harmonious are they really? Generally, one can say that the Model Law brought about harmonization regarding two crucial aspects: a) enforcement of arbitration agreements and b) setting aside as the only recourse against the award, with limited grounds. As regards adequacy, it is not far-fetched to affirm that the Model Law introduced two hallmarks of modern arbitration: a) limited court intervention and b) respect for party autonomy. Indeed, the majority of modern international arbitration regimes, even those that are not based on the Model Law, such as the French and the Brazilian laws, share these characteristics.

However, besides sharing those characteristics — admittedly very important ones—, legal regimes in different jurisdictions may be quite diverse. Thus, when drafting arbitration clauses, diligent parties usually dedicate significant time and thought to the selection of the seat of the arbitration in order to minimize the risk of facing undesired of unexpected results.

Several questions are begged. Does the difference among the legal regimes of the XXIst century reflect a sophisticated arbitration culture, where each seat has its own brand and reputation and competes against the others? Does the post-modernism of arbitration laws also mean that there is no need for the Model Law, since practitioners are expected to be sensitive to these differences? Is the sign of times: “Let the jurisdictions (and practitioners) compete!”? In other words, Are we beyond the Model Law?

Or should we ask, rather, While “improvement” of domestic laws has been achieved, is not “harmonization” still a valid desire? Surely, the more harmonized the local regimes, the more effective and secure arbitration is in the international sphere. Is the disparity due, not to evolution alone, but to evolution without a model framework that reflects current best practices? In other words, Do we need to revise the Model Law?

Nathalie Voser recently commented on the need to reform the Model Law. I agree with her, because harmonization is part of the reason why arbitration as a neutral means is so appropriate to settle international commercial disputes. The final beneficiaries of arbitration are the users, not the practitioners, academics, or arbitrators, and the less uniform the legal regimes are, the higher the “transactional costs” are likely to be for those users.

What follows is a list of some of the issues that I submit must be tackled in a new, complete revision of the Model Law, together with an invitation to comment thereon.

1. Scope of application: the Model Law should include two options: a) monist regime for all arbitrations, or b) dualist system, whereby the law is only applicable to international arbitration. In my view, international arbitration practices are being adopted and welcome in domestic arbitrations in many countries, so they will eventually prefer a sole regime applicable to both, without any difference. In a truly monist regime, the nationality of arbitrators, for example, should not be an issue, as well as the grounds for setting aside of awards.

Further, in the dualist system option, the definition of “international arbitration” should be revised. A possibility is to define international arbitration as an arbitration which object involves more than one state. The current definition is based on the contract, not on the arbitration, and a contract can evolve: one that starts as domestic can become international, and vice versa, if there is a change in the composition of the parties for example. In addition, the Model Law leaves the parties the option for them to expressly agree if the subject matter relates to more than one country, which is indirect. Article 3(c) should be modified to allow parties in an otherwise domestic arbitration to expressly choose to be governed by the international arbitration law (a sort of waiver of dualism).

2. There should be no formal requirements to the arbitration agreement (this should be a matter of evidence only) but substantive requirements for the validity of the agreement should be spelled out, such as consent and arbitrability of the dispute. For clarity and finality, these requirements should be governed by the law of the seat only. Capacity of the parties may be included as an additional requirement, in order to mirror the ground for setting aside of Art. 34 –and the New York Convention–, keeping in mind that it is generally governed by the personal law of the relevant party.

3. The negative effect of the principle of kompetenz-kompetenz should be recognized, so that judges refer parties to arbitration upon a prima facie verification of an arbitration agreement when the defendant so requests. With that, arbitral tribunals should decide on their own competence as a preliminary matter whenever possible, and local judges should be able to revise the relevant decision within a short time frame (as provided in the second sentence of Article 16(3) of the Model Law). Such judge should ideally be the same as the judge for the setting aside procedure.

4. Judicial support regarding evidence (Art. 27 of the Model Law) should also apply when the place of arbitration is not in the same jurisdiction, so it should be included under the exceptions of Art.1(2).

5. The provisions on the constitution of the arbitral tribunal should be revised to include a) an odd (or “uneven”) number requirement, b) the possibility of having the Permanent Court of Arbitration act as appointing authority (as an alternative to local courts), and c) a mechanism for multi-party situations. Also, the obligation of arbitrators to be and remain independent and impartial should be stated out clearly.

6. Where parties have not agreed on the applicable law, the voie directe practice should be reflected in Article 28(2), so reference to the conflict of laws should be deleted.

7. It might be useful to include a definition of “award”, so as to define the situation of partial awards and awards that decide on provisional measures (which should be considered awards, in my view; otherwise, for the sake of consistency, arbitral tribunals should not be free to decide the form of their decisions regarding provisional measures).

8. Consider adding the sécret du déliberé and the requirement that all awards be reasoned, save for awards on agreed terms.

9. Setting aside should be considered waived if the bases on which it is raised were not raised timely, such as party capacity, validity of the arbitration agreement, and improper constitution of the arbitral tribunal. Further, if a party has applied to recourse under Article 16(3), it should not have a second shot at setting aside the award on the same grounds.

10. The provisions on recognition and enforcement of awards should be revised to a) clarify whether “recognition” is one step and “enforcement” is another and, if so (I read the Model Law as giving de jure recognition to arbitral awards but practice has proven me wrong), whether international awards rendered in the seat of the law need to be recognized (I submit they should not, since they can be set aside in the same jurisdiction), and b) to designate the competent court/s.

11. Change the “public policy” ground for setting aside and denial of enforcement to “international public policy”.

12. Consider adding provisions on confidentiality of the arbitral proceedings.

An updated Model Law will surely further the development of international commercial arbitration.


To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here. To submit a proposal for a blog post, please consult our Editorial Guidelines.

Kluwer Arbitration
This page as PDF


  1. Contributive, thank you. As you pointed out at the beginning the desired purpose of the Model Law was the unification rather than harmonization of arbitration law. Therefore in the revised draft of the Model Law (neither Convention nor other form of treaty law) – and I absolutely agree with you that changes are of wide practical relevance – comparative analyses of existing arbitration practice and theory should be taken into account.

    1. Thank you for your comment, Natalia. I agree with you that any reform must be preceded by comparative analyses.

      1. Among the reforms of national arbitrations Laws, you may add to the list Saudi Arabia New Arbitration Law 2012

  2. This is a well argued, thought-provoking and, indeed, provocative contribution.

    I agree that consideration should be given to updating the Model Law if a sufficient case can be made out for UNCITRAL Working Group II being charged with this task. To get rid of the Model Law completely would, however, be tantamount to throwing out the baby with the bathwater. Whilst the Model Law’s influence would probably (indeed, undoubtedly) live on in a ‘post-Model Law’ world, do we really wish to risk seeing a refragmentation of arbitration laws worldwide in the longer term?

    Surely it is the harmonisation or unification (or whatever one wants to call it) wrought by the Model Law that has kept in the forefront of practitioners’ minds the fundamental things they need to take into account, both when making arbitration agreements and further down the line, after disputes have arisen.

    The beacon of the Model Law will certainly always be needed to guide those (hopefully fairly few) practitioners who may not be sufficiently familiar with arbitration and, in particular, neophytes to the area, such as newly qualified practitioners.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.