10:09 PM Nicholas Poon (Assistant Registrar, Supreme Court of Singapore)

    Dual-Institution Provision in Arbitration Agreements and the Principle of Effective Interpretation: Government of the Russian Federation v I M Badprim Srl



    It is not commonplace to have an arbitration agreement that prescribes an institution to administer the putative arbitration in accordance with the rules of a different institution. There are good reasons for the relative rarity of such “dual-institution” provisions in an arbitration agreement, the simplest of which is that the administering institution may not have the organisational capabilities to apply rules devised by another institution that has those capabilities precisely so that those capabilities could be put to good use. But from time to time, such pathological hybrid arbitration agreements – as these have been called – surface in courts. Insigma Technology Co Ltd v Alstom Technology Ltd [2009] 3 SLR(R) 936 (“Insigma”) is one such case.  

    A decision of the Court of Appeal, Insigma affirmed what is known as the “principle of effective interpretation” which outworking was explained by the court in the following terms (at [31]): 

    “where the parties have evinced a clear intention to settle any dispute by arbitration, the court should give effect to such intention, even if certain aspects of the agreement may be ambiguous, inconsistent, incomplete or lacking in certain particulars [citation omitted] so long as the arbitration can be carried out without prejudice to the rights of either party and so long as giving effect to such intention does not result in an arbitration that is not within the contemplation of either party.”

    To the extent that it has become a self-standing principle of arbitration law, the principle of effective interpretation has only been applied by the Singapore court once since, in HKL Group Ltd v Rizq International Holdings Pte Ltd [2013] SGHCR 5 (“HKL Group”), a decision of an assistant registrar sitting in the High Court which also involves the validity of an arbitration agreement containing a dual-institution provision. In other words, the contours of this principle have not yet been tested rigorously.  

    Government of the Russian Federation v I M Badprim S.r.l (“Badprim”), a 2015 Svea Court of Appeal’s decision in which the same type of pathology was at the heart of the application to set aside an arbitral award, should therefore be of interest to users and practitioners of arbitration. Located in Stockholm, the Svea Court of Appeal is one of six appellate courts in the Swedish legal system. 

    Facts of Badprim   

    For the purposes of this commentary, all that needs to be stated is that a contract for the design and construction of border crossing posts was entered into between Badprim, a Moldovan company, and the Federal Customs Office of the Russian Federation. Contained in this contract was an arbitration agreement which provided for arbitration under the auspices of the Stockholm Chamber of Commerce (“SCC”), but in accordance with the arbitral rules of the International Chamber of Commerce (“ICC Rules”). Badprim commenced arbitration under the SCC against the Government of the Russian Federation and the Federal Customs Office, claiming compensation for work performed. The tribunal held that the arbitration agreement was binding on the Russian Government, despite the latter’s protest, and eventually rendered an award in favour of Badprim. 

    The Russian Government then applied to set aside the award in the Svea court. Two grounds for annulment were raised: (a) the Russian Government was not a party to the contract or the arbitration agreement; and (b) the arbitration agreement was in any event invalid because it was not possible for the SCC to apply the ICC Rules. It was not disputed by Badprim that the SCC could not administer an arbitration in full compliance with the ICC Rules as a result of the organisational differences. Badprim argued, however, that the arbitration agreement should be “interpreted in a manner so as to make it enforceable in practice”, given that the parties had clearly agreed that their disputes should be resolved by arbitration before the SCC. 

    Decision of the Svea Court of Appeal

    By a majority decision, the arbitration agreement was upheld. The main factors which influenced the majority’s decision were:

    • the arbitration agreement had not been specifically negotiated by the parties;
    • there was no evidence that there was any particular purpose behind the parties’ selection of the ICC Rules; 
    • the Russian Government, which finalised the contract and was deemed to be in a superior contracting position vis-à-vis Badprim, did not inform Badprim of the unusualness of the arbitration agreement; and
    • the SCC agreed to and did administer the arbitration. 

    In the premises, even though the majority recognised that the arbitration agreement was contradictory in so far as it contemplated an arbitration which could not be undertaken in full accordance with the applicable rules – at least conceptually – because the SCC could not apply the ICC Rules in their entirety, the majority held that the basic agreement of the parties for their dispute to be resolved by arbitration was enforceable.    

    The dissenting judge held that the award should be annulled because he agreed with the Russian Government’s submission on the first issue, namely, that the Russian Government was neither expressly listed as a party to the arbitration agreement nor was there solid evidence that it had consented to the arbitration agreement. The judge was however in agreement with the majority as regards the validity of the arbitration agreement, albeit for slightly different reasons. 

    To him, it was of “determining importance” that the parties had agreed to arbitrate their dispute. In a context such as the present, as long as the administering institution applies the arbitral rules of another institution “as well as is possible”, the parties would have received their bargain, that is, resolution of the dispute by arbitration. In his view, it “ought not to be possible to complain that various measures, assessments and decisions” by the administering institution may not have been exactly as they would have been if they had been applied by the institution which devised the applicable rules. Accordingly, an arbitration agreement is not invalid merely in consequence of such pathology. The only caveat offered by the minority was that the analysis may be different if the administering institution refused to apply the chosen rules. 

    Brief commentary

    Clearly, the principle of effective interpretation is at the core of both decisions, particularly in the minority judgment which appears to suggest, as a general rule, dual-institution provisions do not render an arbitration agreement invalid, even if the full force of some provisions of the arbitral rules may be compromised because the administering institution’s organisational structure and processes simply cannot adapt to the applicable rules. The underlying logic is hard to miss: if the parties had agreed to resolving their dispute by arbitration, that essential agreement is paramount and should be given full effect; practical difficulties about the exact implementation of the arbitral procedure are of secondary importance. 

    While a big picture substance-over-form approach is typically attractive, can it really be said that the fundamental expectation of arbitration is all that matters? If that is indeed the case, a mere sentence in a contract that “All disputes shall be resolved by arbitration.” would have to be upheld, despite the sheer uncertainty as to its implementation. But arbitration agreements have been struck down for uncertainty and ambiguity (see eg, Lovelock EJR v Exportles [1968] 1 Lloyd’s Rep 163). Courts have not gone so far as to say that an intention to arbitration is all that matters, rightly so, because “where is the seat of arbitration”; “do any institutional rules even apply”; or “how should the tribunal be constituted” are not questions which can be swept under the carpet so easily. An arbitration does not materialise spontaneously. 

    Of course, those questions did not surface in Badprim because the arbitration agreement was not threadbare. But the point remains: the operability of the chosen rules matters. This was also adverted to by the majority when they suggested that the analysis might be different if there was evidence that there was a particular purpose for the parties’ selection of the arbitral rules. It should be noted, in this regard, that a particular purpose for selecting a set of institutional rules ought to be irrelevant unless the purpose relates to the administering institution’s capability in applying the chosen rules or a part thereof. For example, evidence that the Russian Government and Badprim had selected the ICC Rules because they wanted a neutral set of rules ought to be inconsequential because as long as the ICC Rules are capable of being applied, whether by the SCC or other institutions for that matter, the bargain for neutrality is preserved. 

    Hence, in relation to the reference to particular purpose, it is likely that the majority had in mind the operational consequences of the SCC applying the ICC Rules. One straightforward example of this is the provision in the ICC Rules on the appointment of arbitrators. Art 13(3) states that where the ICC Court of Arbitration is to appoint an arbitrator, it shall make the appointment upon proposal of a National Committee or Group of the ICC that it considers appropriate. Even if the SCC Board (an entity constituted under the SCC Rules) is deemed to be the equivalent of the ICC Court of Arbitration, there would be no proposal of a National Committee or Group of the ICC to speak of. 

    Thus, if there was evidence that the parties had contemplated the procedure under Art 13 of the ICC Rules in the lead up to the conclusion of their contract, or better yet, intimated that the procedure was of importance to them, the majority might well have decided the other way. And it would not have been surprising because the parties’ intention behind contracting for the ICC Rules would be undermined considerably by SCC’s inability to fully apply the letter and spirit of Art 13. It is of course possible to argue that striking out the entire arbitration agreement in these circumstances is an even grosser disregard of the parties’ basic intention to arbitration their dispute. While persuasive in some respects, the shortcoming of this argument is that it puts the cart before the horse: if the cornerstone of arbitration is consent, can there be consent to an alternative procedure that is radically different from what was originally intended?   

    Therefore, the workability of the arbitration agreement does not and cannot cease to be a relevant consideration just because parties have evinced a clear intention to arbitration. That the Court of Appeal in Insigma thought it necessary to stress (at [35]) that the party seeking to invalidate the arbitration agreement “was unable to explain or to convince [the court] where the uncertainty lay”, and why the arbitration agreement “could not or did not work” supports the proposition that workability of the dual-institution provision is a relevant consideration. It is also consistent with the position under the UNCITRAL Model Law and numerous national arbitration legislations, including the International Arbitration Act and the English Arbitration Act, all of which permit a court to refuse enforcement of an arbitration agreement if it can be shown that the arbitration agreement is “inoperative” or “incapable of being performed”. 

    Last but not least, a little something ought to be said about the relevance of the administering institution’s consent to administering the arbitration. This was a factor in both the majority and minority’s judgments in Badprim, as well as in Insigma (see [40]) and HKL Group (see [37]). If it is evident that the administering institution does not have all the capabilities required by the applicable rules, but nevertheless consents to administering the arbitration, what weight ought to be given to such consent? Quite a bit it seems, as far as the Svea Court of Appeal is concerned. Despite being cognisant of the organisational differences between the SCC and the ICC which made the SCC unable, in their words, “to administer an arbitration fully compliant with the ICC’s Rules”, the court on the whole was comfortable in upholding the arbitration agreement. Undoubtedly, the SCC’s consent to administering the arbitration, and actual – and successful, it seems – administration of the arbitration was sufficient to negative any concerns over the workability of the ICC Rules in an SCC-administered arbitration. 

    The better approach, it is suggested, is to regard the administering institution’s consent as being a gateway consideration. Without an institution willing to administer, the court would not even be able to give effect to an institutional arbitration as required by the arbitration agreement. However, it does not mean that workability of the applicable rules is thrown out of the door once the administering institution consents. The practical difficulties will still matter if it can be shown that the administering institution is incapable of applying the letter and spirit of the applicable rules if not in full at least in substantial compliance. But this must be more than a hypothetical difficulty. Had an appointment issue arisen under Art 13(3) on the facts after the SCC had consented to and was in the process of administering the arbitration, it would be difficult to see how the court could reasonably overcome the contention that any modification by the SCC of the procedure under Art 13 to mitigate its lack of organisational capabilities would not be a true application of the rule as the parties are entitled to expect, with the consequence that the arbitration had proceeded on a footing that was not in accordance with the applicable rules.  


    It is axiomatic that parties ought to be held to their bargains. However, it is equally axiomatic that the process imposed on them by the court accurately reflects the bargain, and not something else altogether. That is why the principle of effective interpretation is not – and should not become – a panacea for all gaps, inconsistencies and pathologies in an arbitration agreement.  Hence, even though Badprim joins the long list of court decisions which are partial to arbitration agreements with dual-institution provisions, it would be a mistake to assume as a given that an arbitration agreement with a dual-institution provision will always be enforced. Consequently, parties would be well-advised to consider whether the benefits of contracting for one institution to administer the rules of another are well worth the litigation that may ensue. 

    * This blog entry may be cited as Nicholas Poon, "Dual-Institution Provision in Arbitration Agreements and the Principle of Effective Interpretation: Government of the Russian Federation v I M Badprim Srl", Singapore Law Blog (9 April 2015) (

    ** The opinions contained in the commentary reflect the author’s own views and are not to be understood as reflecting the views of the author’s employer.

    *** A PDF version of this entry may be downloaded here 

Comment Section