I would like to offer a few reflections regarding arbitrators' discretion in making determinations of applicable law. Choice-of-law decisions are notoriously challenging, and the 'right' decision is not always easy to ascertain. However, getting the analysis right is essential, as the choice of law can have significant effects on the parties' rights, the proceedings and the legal and practical outcomes of the dispute. In arbitration, choice-of-law issues are made more complex by the fact that, whereas national courts are generally bound by a pre-determined set of procedural rules, arbitrators may be called upon to determine both procedural and substantive applicable law. Therefore, this power should be exercised judiciously and with an even-handed and objective approach that will create uniformity and predictability, to the extent possible. This is one of several reasons why selecting good arbitrators is one of the most critical things a party does in an arbitration.

My concern is that in some cases, arbitrators may take short cuts or make decisions based on apparently compelling but improper justifications. I do not mean to indicate that there is a widespread problem in this regard, but merely to note that it does sometimes happen, and to advise caution given the large impact such decisions can have. I have observed by my own experience as counsel in arbitrations, and in reading about others' arbitrations, that in cases where the proper decision in a choice-of-law analysis is murky or where there may appear to be countervailing considerations, such as policy, justice or convenience, arbitrators are sometimes tempted to take short cuts or a pragmatic approach and apply laws that are, for instance, more familiar to the arbitrator or more consistent with what the arbitrator believes to be better policy or a better ground for giving parties a 'just' opportunity or result.

In facing choice-of-law questions, an arbitrator may find that several different bodies of law present themselves as potentially viable contenders. In such cases, arbitrators may find themselves gravitating, when faced with such a choice, toward laws with which they are more familiar or that provide the result they believe is proper in light of the circumstances or policy considerations. In other words, they may be tempted to avoid the messiness and inconvenience of a proper conflicts-of-law analysis, or may feel that issues of pragmatism, justice or policy, or other external considerations, should override the proper legal result of such an analysis.

I clearly understand the difficulties. It is not easy in some cases to apply an unfamiliar law. Indeed, I have in mind one particular case in which I was counsel for one of the parties, in which the sole arbitrator struggled a lot to learn and apply unfamiliar legal principles. In that case, there was no choiceof- law issue, but if there had been, it would have been understandable for the arbitrator to be tempted to select a law with which he was familiar, since it was clear that he had difficulty ruling on the unfamiliar principles. However, in such cases, it is the duty of the arbitrator to obtain, either from the parties and/or from an arbitrator-appointed expert, sufficient knowledge of the law to allow for a proper decision in keeping with the applicable law. In the end, that purpose was accomplished, but it was anything but smooth. I have seen other cases (as counsel) in which I was convinced that the arbitrators were influenced by factors of familiarity in making their choice-of-law decisions. Regardless of difficulty, however, lack of familiarity with the applicable law should never be a factor in arbitrators' decisions on choice-of-law matters.

Similarly, the arbitrators should not be driven by policy or other outcomecentric factors, as it appears the US Supreme Court found of the arbitrators in the Stolt-Nielsen S.A. et al. v. AnimalFeeds International Corp. case, a 2010 slip opinion discussed in this dossier by William Park. The Court was considering a case in which the arbitrators had determined that class arbitration was permitted under the scope of the relevant arbitration clause (which was silent on the question). The Court concluded that "the panel proceeded as if it had a common-law court's authority to develop what it viewed as the best rule for such a situation" and that "[t]he panel's few references to intent do not show that the panel did anything other than impose its own policy preference." However, "because the parties agreed that their agreement was 'silent' on the class arbitration issue, the arbitrators' proper task was to identify the rule of law governing in that situation."

I will note that the procedural history of the above case reveals the complexity of these issues. After the arbitrators issued an award stating that class arbitration was permitted, the objecting party appealed. The district court vacated the award on grounds that it was in "manifest disregard" of applicable law and failed to undertake a proper choice of law analysis. The Second Circuit court of appeal, however, reversed, holding that no authority had been cited that the applicable law prohibited class arbitration. This, in turn, was reversed by the Supreme Court, for the reasons I mentioned. Clearly, the decision was not a simple one. However, as both the district court and the Supreme Court found, a proper choice-of-law analysis cannot be dispensed with for any reason, no matter how compelling.

My objective here is to point out the danger of short-cutting choice-of-law analyses or allowing them to be hijacked by external factors. I would admonish arbitrators to take great care not to let their choice of law analysis be driven by a predetermined desired outcome for whatever reason, be it convenience, policy or other considerations.