First, one could argue that this case is a reaffirmation and extension of the recent trend in English cases of applying the law of the seat to the arbitration agreement. The best example of this approach is the decision in C v D  EWCA Civ 1282.
Although this interpretation has considerable explanatory power, it goes against several observations made in other parts of the judgment. Moore-Bick LJ devotes a large part of his reasoning (most of paragraphs 7 to 26) to explaining that the proper law of the contract would often extend to the arbitration agreement, and that the seat, though important, was not conclusive. There are also an interesting section of the judgment (paragraph 26) where he explains that the concept of severability also does not automatically suggest that the law of the seat governs the arbitration agreement, since severability does not amount to breaking all connections between the substantive contract and the arbitration agreement. However, while these observations indicate that the seat is not crucial in determining the implied choice of law, the express reasoning on the ‘closest and most real connection’ test suggests that the seat is critical and possibly conclusive if there is no express or implied choice.
Second, one may argue that while the seat was a very important (and possibly clinching factor) for the ‘closest and most real connection’ test, it was the effect of Brazilian law on the arbitration agreement that drove the Court’s conclusion on implied choice. This is similar to the principle of ‘effectiveness’, which suggests that the parties are taken to have intended the arbitration agreement to be effective, and hence, impliedly chosen a governing law under which it would be effective.
While there is much to be said for the principle of effectiveness, it is not entirely clear that this is an appropriate case for it to be applied. The effect of Brazilian law was not to render the arbitration agreement ineffective; it was only to make the insured party’s consent a condition precedent to the commencement of arbitral proceedings. While there undoubtedly is force in the learned judge’s reasoning, one may argue that the reason the contract didn’t include this consent as an express condition was because the parties assumed that the arbitration agreement would anyway be governed by Brazilian law, and hence contain this pre-condition. More significantly, the judgment suggests that there was no finding of the Commercial Court on whether this was indeed the effect of Brazilian law (paragraph 30). In effect, the insured party contended that the arbitration agreement was enforceable only with its consent, and the Court used this contention against the insured party’s own case so as to make the arbitration agreement enforceable.
Therefore, neither interpretation of the reasoning and conclusions is particularly satisfactory. Further, the concurring judgment of the Master of the Rolls also raises a couple of thorny issues. First, unlike Moore-Bick LJ, the Master of the Rolls does not draw a distinction between implied choice and the ‘closest and most real connection’ test, saying only that the two factors discussed by Moore-Bick LJ suggest that the arbitration agreement was governed by English law. Second, he agrees with Moore-Bick LJ that the reasoning in C v D is not the foundation of the Court’s decision here; in fact, he thinks a “powerful case” may be made against the reasoning in C v D. Finally, he seems to take a different view from Moore-Bick LJ on the relevance of the principle of severability. He suggests that “the growing awareness of the importance of the principle” may explain the relatively recent trend in English law of giving more importance to the seat of the arbitration.
In conclusion, the decision of the Court of Appeal in Sulamerica is a reminder (if ever one was needed) of the complex and conflicting jurisprudence on the law governing arbitration agreements. If one were to look only at the conclusion of the Court, it appears that the seat of the arbitration, and especially the fact that it was London (which meant the mandatory application of certain statutory provisions) was a very major factor. However, at the same time, large parts of the reasoning point towards the seat not being a conclusive factor. Further, the Court’s apparent extension of the ‘effectiveness principle’ is not entirely convincing. Given these issues, and the conflicting jurisprudence highlighted by the Master of the Rolls, it can be expected that future decisions will shed some more light on this very significant and intriguing issue.