"51. Having regard to the guidance provided by these authorities the position may be summarised as follows: (1) For jurisdiction underCPR r 25.13 (2)(a) to be established it is necessary to satisfy two conditions, namely that the claimant is resident (i) out of the jurisdiction and (ii) in a non-Convention state. (2) Once these jurisdictional conditions are satisfied the court has a discretion to make an order for security for costs underCPR r 25.13 (1) if “it is satisfied, having regard to all the circumstances of the case, that it is just to make such an order”. (3) In order for the court to be so satisfied the court has to ensure that its discretion is being exercised in a non-discriminatory manner for the purposes of articles 6 and 14 of the Convention: see the Bestfort case [2017] CP Rep 9 , paras 50–51. (4) This requires “objectively justified grounds relating to obstacles to or the burden of enforcement in the context of the particular foreign claimant or country concerned”: see Nasser's case[2002] 1 WLR 1868 , para 61 and the Bestfort case at para 51. (5) Such grounds exist where there is a real risk of “substantial obstacles to enforcement” or of an additional burden in terms of cost or delay: see the Bestfort case at para 77. (6) The order for security should generally be tailored to cater for the relevant risk: see Nasser's case at para 64. (7) Where the risk is of non-enforcement, security should usually be ordered by reference to the costs of the proceedings: see, for example, the orders in De Beer's case[2003] 1 WLR 38 and the Bestfort case. (8) Where the risk is limited to additional costs or delay, security should usually be ordered by reference to that extra burden of enforcement: see, for example, the order in Nasser's case. 52. I would add the following observations: (1) The relevant risks are of (i) non-enforcement and/or (ii) additional burdens of enforcement. A real risk of either will suffice to meet the “threshold” test. (2) Some of the authorities refer to difficulties of enforcement. Mere difficulty of enforcement in itself is not enough (save in so far as it results in additional costs and therefore an extra burden of enforcement). The relevant risk is non-enforcement, not difficulty in enforcement, and this is the risk to which the test of “substantial obstacles” is directed. The obstacles need to be sufficiently substantial to amount to a real risk of non-enforcement. Difficulties may, however, be evidence of the “substantial obstacles” required for there to be a real risk of non-enforcement. (3) Delay is mentioned as a relevant additional burden of enforcement, but it is difficult to see how this can be quantified in terms of security unless it is likely to result in some additional cost or interest burden."