Background

The English court’s jurisdiction over a defendant domiciled outside the EU is based on the court’s permission to serve proceedings abroad. To obtain permission, the claimant must show: (i) a serious issue to be tried; (ii) a good arguable case that the claim falls within one of the jurisdictional gateways in CPR PD6B; (iii) that England is the appropriate forum to try the claim.

There has been considerable debate about the ‘good arguable case’ test. The extent of the confusion was made clear in a recent judgment from the Court of Appeal, Kaefer Aislamientos v AMS Drilling Mexico [2019] EWCA Civ 10: ‘A test intended to be straightforward has become befuddled by “glosses”, glosses upon glosses, “explications” and “reformulations”’. The court identified nine ‘glosses’ which had been used in various cases for the same test (para. 59 per Green LJ).

The leading case on the ‘good arguable case’ test was formerly Canada Trust Co v Stolzenberg (No 2) [1998] 1 WLR 547. Waller LJ held at p. 555 that the test included the relative concept of who had ‘a much better argument on the material available’. That had led to further arguments on what that phrase meant.

To provide clarity, the test was reformulated by the Supreme Court in Brownlie v Four Seasons Holdings International [2017] UKSC 80 and Goldman Sachs International v Novo Banco [2018] UKSC 34. In Brownlie, at para. 7 Lord Sumption explained the test in three limbs:

(i)‘that the claimant must supply a plausible evidential basis for the application of a relevant jurisdictional gateway;

(ii) that if there is an issue of fact about it, or some other reason for doubting whether it applies, the court must take a view on the material available if it can reliably do so; but

(iii) the nature of the issue and the limitations of the material available at the interlocutory stage may be such that no reliable assessment can be made, in which case there is a good arguable case for the application of the gateway if there is a plausible (albeit contested) evidential basis for it.’

This was strictly obiter but it was approved unanimously by the Supreme Court in Goldman Sachs at para. 9 per Lord Sumption.

However, neither case provided guidance on how the test works in practice, nor what ‘plausible’ evidence is, nor how it relates to the original ‘good arguable case’ test, nor how it intersects with the the Canada Trust test. The Court of Appeal attempted to answer those issues in Kaefer.

Kaefer: the test for permission to serve out

In Kaefer C sought to recover sums due under a contract for refurbishment of an oil rig, owned by D3 (a subsidiary of D4) and chartered by D2. The contract was made between C, D1 and D2 and contained English law and jurisdiction clauses. D3 and D4 were Singapore companies and challenged jurisdiction. C argued that they were undisclosed principals of D2 and thus parties to the contract, so that they could be sued in England under the jurisdiction clause. The Judge (Peter MacDonald Eggars QC, sitting as a Deputy Judge) set aside permission to serve out. He found that C had a good arguable case that D3 was an undisclosed principal but Ds had the better argument that it was not. There was no good arguable case in relation to D4. C appealed.

Brownlie and Goldman Sachs were handed down after the Judge’s decision. In the Court of Appeal, Green LJ (with whom Davis and Asplin LJJ agreed) explained the reformulated test:

  • Limb (i): A ‘plausible evidential basis’ was one in which the claimant has the better argument. The claimant has the burden of proof but the standard is only plausibility (not the balance of probabilities). The claimant does not have to show a ‘much’ better argument. The test is context-specific and flexible. In ruling on jurisdiction, the court must take care not to express a view on the ultimate merits.
  • Limb (ii): The court should to try to overcome the evidential difficulties of interlocutory proceedings and to reach a conclusion if it could do so ‘reliably’. Not every evidential dispute is relevant or unresolvable, and judges should use common sense and pragmatism.
  • Limb (iii): Where the court cannot determine who had the better argument, this limb provides a flexible test which combines ‘good arguable case’ and plausibility of evidence.   

The court found that the Judge had wrongly considered ‘good arguable case’ separately from ‘a better argument’. However, he had in fact carried out the right exercise, in examining the plausibility of the evidence and considering the relative merits of the arguments. The appeal was dismissed.

Other issues

The Kaefer judgment also discusses two other important issues:

(i)In the case of jurisdiction clauses, art. 25 of the Judgments Regulation (EU 1215/2012) applies, even where none of the parties are EU-domiciled. Art. 25 requires a ‘clear and precise’ demonstration that the parties had agreed to the jurisdiction clause. This had to be melded with the ‘good arguable case’ test. ‘Clear and precise’ was hard to define but gave an indication of the sort of evidence required to meet the test.

(ii)The contract also contained an entire agreement clause. The Judge had considered that this clause was neutral in deciding whether D3 and D4 were undisclosed principals of D2. The Court of Appeal disagreed: it was a clear indication that D2 did not intend to act on behalf of undisclosed principals.

Conclusion 

Much of the confusion in earlier cases had been about phraseology and labels. As the Court of Appeal remarked, labels do not matter. The test can still be called the ‘good arguable case’ test so long as it is properly understood and applied.

There are limitations to what can be achieved at an interlocutory hearing with evidence on paper. The test is flexible and, in unclear cases, the third limb allows a claimant to establish jurisdiction with plausible evidence even if the judge cannot resolve the evidential dispute. The parties still have to put in detailed evidence and supporting documents to assist the decision. But Davies LJ criticised claimants who fish for the fullest and widest pre-action disclosure from defendants, and then use non-disclosure to establish a plausible case. Ironically, however, the court’s finding that an entire agreement clause points against the existence of undisclosed principals may well turn out to have greater significance for the law than its detailed explanation of the test for permission to serve out.