Royal Courts of Justice, Rolls Buildings,Fetter Lane, London, EC4A 1NL
Before :
MRS JUSTICE MOULDER
- - - - - - - - - - - - - - - - - - - - -
Between :
ROMAN PIPIA Claimant
- and –
BGEO GROUP LIMITED Defendant
(FORMERLY KNOWN AS BGEO GROUP PLC)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Mr P Burton (instructed by Blake Morgan LLP) for the Claimant
Mr A Hunter QC (instructed by Freshfields Bruckhaus Deringer LLP) for the Defendant
Hearing date: 15 February 2019
- - - - - - - - - - - - - - - - - - - - -
APPROVED JUDGMENT
Mrs Justice Moulder :
This is the judgment on the defendant’s application for security for costs made on 20 September 2018.
In support of the application I have two witness statements from Mr Patrick Swain dated 20 September 2018 and 9 November 2018. The witness statements of Mr Swain exhibit two memoranda from Professor Kereselidze and a third memorandum dated 7 February 2019 was also before the court.Professor Kereselidze is currently Head of the Academic Board at New Vision University in Georgia, head of the PhD programme in law and the masters programme in comparative private and international law. From 2005-2017 he was at Tbilisi State university in Georgia responsible for the law faculty structure and curriculum. During this time he provided advice on the evaluation of draft laws on harmonisation with the legislative instruments of the European Union.
In response I have a witness statement from Ms Sarah Rees dated 18 October 2018 and a report from Mr Irakli Adeishvili and a supplemental report dated 4 January 2019. In Ms Rees’ witness statement Mr Adeishvili is described as a senior lawyer at Geocell Ltd, a large telecommunications company in Georgia, and from 2009 to 2015, a judge of the Tbilisi court of appeal. He was a member of a number of commissions working on judicial issues at the Council of Europe.
In the proceedings the claimant is seeking to recover damages for losses arising out of an alleged delict under Georgian law. The amount claimed is just under US$286.5million.
Relevant rules
The relevant provisions of the CPR on security for costs are as follows:
CPR 25.12
A defendant to any claim may apply under this Section of this Part for security for his costs of the proceedings.
An application for security for costs must be supported by written evidence.
Where the court makes an order for security for costs, it will –
determine the amount of security; and
direct –
the manner in which; and (ii) the time within which the security must be given.
The court may make an order for security for costs under rule 25.12 if –
it is satisfied, having regard to all the circumstances of the case, that it is just to make such an order; and
one or more of the conditions in paragraph (2) applies, or
…
The conditions are –
the claimant is –
resident out of the jurisdiction; but
not resident in a Brussels Contracting State, a State bound by the Lugano Convention, a State bound by the 2005 Hague Convention or a Regulation State, as defined in section 1(3) of the Civil Jurisdiction and Judgments Act 1982
On this application it is common ground that the claimant is resident in Georgia. However it was submitted for the claimant that:
the condition in (2)(a) is not satisfied by reason of the Association Agreement (the
“Association Agreement”) between the European Union and the European Atomic Energy Community and their Member States, and Georgia that took effect from 1 July 2016.
it is not just to make such an order- there is no proper evidence before the court which establishes a real risk that a costs order would not be enforced in Georgia.
even if the threshold test had been passed the court should not exercise its discretion in favour of making an order.
Absence of jurisdiction
Counsel for the claimant submitted that the condition in CPR 25.13 (2)(a) was not satisfied by virtue of the Association Agreement:
Counsel submitted that the Association Agreement was to be interpreted as “tantamount” to the conventions identified in CPR 25.13 (2)(a)(ii) i.e. Brussels, Lugano and the Regulation states.
Counsel submitted that the Association Agreement impliedly excluded any application for security for costs against a Georgian party. He referred to the fact that as noted in the White Book at 25.13.7 conventions and agreements may expressly or impliedly exclude the jurisdiction to request security for costs. Counsel relied by analogy on the 1960 Convention on Third Party liability in the field of nuclear energy (the“1960 Convention”) and the provision in that convention against discrimination based upon nationality, domicile or residence.
Counsel relied, amongst others, on Article 21, Article 414 and Article 416 of the Association Agreement.
Article 21 headed “Legal cooperation” specifically refers to the Hague Convention and states that:
“1. The Parties agree to develop judicial cooperation in civil and commercial matters as regards the negotiation, ratification and implementation of multilateral conventions on civil judicial cooperation and, in particular, the conventions of the Hague Conference on Private International Law in the field of international legal cooperation and litigation as well as the protection of children.”
Article 21 merely refers to “developing judicial cooperation” as regards the ratification and implementation of the Hague Convention. The stated aims of the Association Agreement are set out in broad terms in Article 1. They include:
“(f) to enhance cooperation in the area of freedom, security and justice with the aim of reinforcing the rule of law and the respect for human rights and fundamental freedoms”.
The Association Agreement does not provide for the enforcement of judgments either on a bilateral basis or through the Hague Convention. I do not accept therefore that there is any basis on which the Association Agreement can be interpreted as falling within the express terms of CPR 25.13 (2)(a)(ii).
As to the submission that the Association Agreement impliedly excluded any application for security for costs against a Georgian party, counsel relies on Articles 414 and 416. Article 414 “Access to courts and administrative organs” states:
“Within the scope of this Agreement, the Parties undertake to ensure that natural and legal persons of the other Party have access free of discrimination in relation to its own nationals to the competent courts and administrative organs of the Parties to defend their individual rights, including property rights”
Article 416 “Non-discrimination” states:
“1. In the fields covered by this Agreement and without prejudice to any special provisions contained therein: (a) the arrangements applied by Georgia in respect of the EU or the Member States shall not give rise to any discrimination between the Member States, their nationals, companies or firms; (b) the arrangements applied by the EU or the MemberStates in respect of Georgia shall not give rise to anydiscrimination between nationals, companies or firms ofGeorgia.” [emphasis added]
It is noteworthy that unlike the 1960 Convention which refers to discrimination based on residence, the Association Agreement refers only to discrimination based on nationality. In my view the provisions of CPR 25.13 providing for security for costs do not give rise to discrimination between nationals, companies or firms of Georgia. The rules on security for costs are based on the residence of the party, and not on nationality, and thus apply equally to any party who is not resident in the UK or one of the states which is party to the Brussels or Lugano convention or in a Regulation state. The nationality of the party is irrelevant for this purpose and the residence condition can encompass therefore nationals of other states who are resident in Georgia as well as nationals of Georgia. In this regard the observations of Gloster LJ in Bestfort Developments LLP v Ras Al Khaimah Investment Authority [2016] EWCA Civ 1099 at [59] are equally applicable in my view.
Accordingly I do not accept that the Association Agreement impliedly excludes any application for security for costs against a Georgian party.
Is it just to make an order?
An order for security for costs is intended to protect a defendant put to the cost of defending themselves against those claimants for whom the residence conditions are satisfied.
It was common ground that the test to be applied on an application for security for costs is as set out in the Court of Appeal decision in Bestfort Developments LLP v Ras Al Khaimah Investment Authority [2016] EWCA Civ 1099.
Allowing the appeal Gloster LJ held that :
“[77] In my judgment, it is sufficient for an applicant for security for costs simply to adduce evidence to show that “on objectivelyjustified grounds relating to obstacles to or the burden ofenforcement” , there is a real risk that it will not be in a positionto enforce an order for costs against the claimant/appellant andthat, in all the circumstances, it is just to make an order forsecurity. Obviously there must be “a proper basis for considering that such obstacles may exist or that enforcement may be encumbered by some extra burden” but whether the evidence is sufficient in any particular case to satisfy the judge that there is a real risk of serious obstacles to enforcement, will depend on the circumstances of the case. In other words, I consider that the judge was wrong to uphold the Master’s approach that the appropriate test was one of “likelihood”, which involved demonstrating that it was “more likely than not” (i.e. an over 50% likelihood), or “likely on the balance of probabilities”, that there would be substantial obstacles to enforcement, rather than some lower standard based on risk or possibility...” [emphasis added]
“[88] …The evidence which I have summarised above clearlyshowed, notwithstanding the respondents’ expert evidence to thecontrary, a real and serious risk that an order for costs might notbe enforced in Georgia. Indeed, Mr Millett correctly accepted as much before this court and properly conceded that, if the test was one of real risk, then an order for security should have been made. … This is a case where in my view it was obvious on the evidence that an order for security was justified.” [emphasis added]
As to the nature of the evidence which is required and the approach which a court should take, Gloster LJ in Bestfort said:
“[81]…The inherent uncertainty with which the court is having to deal in the context of an interlocutory application for security for costs — a future risk, or a potential difficulty — supports a risk-based, rather than a likelihood-based, approach.”
82. In my judgment, and as Mr Marshall submitted, an analogy can be drawn with the test applied by the court in the context of freezing injunctions. In that context the jurisdiction arises where:
“the court concludes, on the whole of the evidence then before it, that the refusal of a Mareva injunction would involve a real risk that a judgment or award in favour of the plaintiffs would remain unsatisfied.”
see Ninemia Maritime Corp v Trave Schiffahrts GmbH & Co KG (The Niedersachsen) [1983] 1 W.L.R. 1412 , 1422E-H per
Kerr LJ. A claimant has to adduce “solid evidence of risk of dissipation by the defendant” to support his assertion that there is a real risk that the judgment or award will go unsatisfied. As the authors go on to state, “since each case depends on its own facts it is impossible to lay down any general guidelines on satisfying this evidential burden”, although they then outline some of the factors which may be relevant. Likewise, in the context of an application for security for costs, I consider that asimilar – and necessarily flexible — test is appropriate for thepurposes of deciding whether an order for security should bemade. The analogy with the freezing order jurisdiction is particularly apt, in my view, because it reflects the test which a claimant has to satisfy in order to obtain protection for satisfaction of any judgment which it might obtain against a defendant. An application by a defendant for an order for security for his costs is the converse side of the coin. There should, it seems to me, be an appropriate symmetry between the two tests that respectively entitle a claimant to a freezing order to satisfy any judgment, and a defendant (or appellant) to security for its costs. There are further similarities. On the making of a freezing order, the court makes an interim finding on the merits (the existence of a good arguable case) which is later tested at trial; on the issue of risk of dissipation, however,it makes a determination on an issue that is never tested at trial,namely: is there, on the whole of the evidence then before thecourt, a real risk of dissipation? As Mr Marshall submitted, thatapproach reflects the perceived justice of protecting the applicantagainst the risk of his being unable to enforce any judgment hemay later obtain because of unjustified dissipation, when a trialon the risk of dissipation is not practicable or proportionate.It isdirectly comparable to the security for costs jurisdiction whichprotects against “the risk of being unable to enforce any costsorder they may later obtain”. It follows that the tests should besimilar.” [emphasis added]
Counsel for the claimant submitted that:
The memoranda of Professor Kereselidze and the comment on it in the witness statements of Mr Swain is inadmissible.
The fact that security for costs was ordered in Bestfort and related to the enforcement of orders in Georgia is irrelevant and offends the principles of evidence in Hoyle v Rogers [2014] EWCA Civ 257. The court can read the decision in Bestfort but neither the court nor the claimant has seen the evidence relied upon in that case.
Counsel for the claimant submitted that the evidence in the witness statements and the memoranda could not be adduced in this way as it was expert evidence which should therefore comply with CPR 35. Counsel relied on the decision of Marcus Smith J in New Media Distribution Company Sezc Limited v Kagalovsky [2018] EWHC 2742 (Ch):
“10. But the short point is that these statements are, or purport to be, expert opinion. And this is the second problem with these paragraphs in Kagalovsky 4. It is not right for a factual statement(that is, Kagalovsky 4) to be used to adduce expert, when thereare clear procedural rules of this court that no party may call anexpert or put in evidence an expert's report without the court'spermission. It is not right for these provisions in CPR 35 to becircumvented simply by attaching the expert statements to astatement of fact.[emphasis added]”
11. Indeed, there are a number of problems with this course. One loses, in their entirety, the safeguards that exist regarding the adduction of expert evidence. I have in mind, for example, an expert's duty to the court, the expert declarations that one normally sees and the fact that experts will be cross-examined. Here, the statements of Professor Butler and Mr Rievman contain none of the requirements and provisions that expert evidence ought to have.
12. Furthermore, it has been impossible to test whether these two gentlemen are actually third party experts. That is something that would be dealt with on an application to adduce expert evidence.
13. Finally, there is no provision for the cross-examination of these experts. Now, that may or may not be necessary in this case - Mr Ramsden suggested not - but it seems to me that this court really cannot proceed on the basis that evidence being adduced in this way cannot be tested. The reason it cannot be tested is because these statements appear as annexes to a witness of fact who himself, even if he were medically competent to give evidence (which Mr Kagalovsky is not), could not sensibly be cross-examined on this material.
Counsel for the Claimant also relied on the decision in Al Nehayan v Kent [2016] EWHC 623 (QB) of Mrs Justice Nicola Davies that evidence on an application for security for costs must comply with CPR 35.
Counsel for the claimant also submitted that the court in setting the evidential standard, should bear in mind thatit makes a determination on an issue that is never tested attrial; counsel sought to distinguish the distinction drawn by Lord Nicholls (and cited by Gloster LJ at [80]) between evidence required at an interlocutory stage to assess risk and the final stage of proceedings as confined to cases where there would be a final determination.
As to the evidence being adduced in the form of the witness statement of Mr Swain and the memoranda, counsel for the defendant stressed the need for a flexible approach as identified by Gloster LJ and referred the court to Appendix 10 of the Commercial Court guide:
“An affidavit or witness statement in support of an application for security for costs should deal not only with the residence of the claimant (or other respondent to the application) and the location of its assets but also with the practical difficulties (if any) of enforcing an order for costs against it.”
The decision of Marcus Smith J in New Media Distribution Company was a decision on an application to exclude evidence in proceedings but does not appear to have been in the context of an application for security for costs. The decision in Al Nehayan was before the decision of the Court of Appeal in Bestfort and was therefore applying a different test. It seems to me therefore that the position is set out in Bestfort and in particular at [82] quoted above: the application for security for costs is akin to an application for a freezing order and a flexible approach is required. In the circumstances of such an application it is not just or proportionate to require expert evidence to be adduced which complies with the rules on expert evidence.
As to the submission that, the standard should be set having regard to the facts, the evidence will never be tested at trial, Gloster LJ expressly acknowledged this but nevertheless endorsed a flexible test based on the risk and not the likelihood of the defendant being unable to enforce a costs order.
Accordingly, in my view this court can have regard to the evidence in the witness statements of Mr Swain including the memoranda of Professor Kereselidze exhibited to the witness statements. This conclusion seems to me to be reinforced by the dicta in the recent decision of the Court of Appeal in Danilina v Chernukhin [2018] EWCA Civ 1802. Hamblen LJ at [58]:
“As a matter of authority, this court has held in the Bestfort case [2017] CP Rep 9 that the appropriate “threshold” test when considering the issue of whether there are “substantial obstacles” to enforcement is one of real risk rather than likelihood. Various reasons are given for reaching that conclusion, including theneed for a simple and clear approach to issues which will beconsidered at an interlocutory hearing on the basis of what“necessarily and proportionately, will be limited evidence”: atpara 48.”
Accordingly Hamblen LJ adopting that approach went on to hold:
“59. Whilst the court in the Bestfort case was concerned with the “threshold” test for exercising the discretion to order security for costs in a non-discriminatory manner rather than any discretion as to the amount of security to be ordered, I consider that the court’s approach should be consistent. If, for the reasons given in the Bestfort case, it is not appropriate to require that more than a real risk be established for the purpose of nondiscrimination, it is equally inappropriate to do so for the purpose of quantum. The consequence of adopting a sliding approach is in effect to require the defendant to establish likelihood of nonenforcement (if not more) if security for the entirety of the costs is to be obtained.” He observed at [60]:
“Further, it would lead to the type of detailed evidentiary exercise which the court was keen to avoid through its decision in the Bestfort case. It would allow in via the back door all the evidence and evidential inquiries which the court in that case took care to shut out via the front door.”
It was submitted for the claimant that the court must consider the evidence before this court and not the evidence before the court in Bestfort.
Counsel for the defendant submitted that it was open to the court to rely on the evidence which is set out in the judgment in Bestfort at [45]. Counsel for the defendant relied on the decision of Patricia Robertson QC (sitting as a Judge of the High Court) in JSC BTA Bank v Ablyazov and Khrapunov [2018] EWHC 1368 (Comm) at [19]:
“The following principles appear to be settled and uncontroversial:
i) This Court cannot rely upon a bare finding of a prior Court in a matter in which Mr Khrapunov was not a party or privy: Hollington v Hewthorn & Co Ltd [1943] KB 587. The rationale for that rule is that fairness requires that this Court must decide on the basis of the evidence before it, rather than simply adopting the opinion of another Court (including when that other Court was making its findings of fact on the evidence that was before it): Rogers v Hoyle [2014] EWCA Civ 257 at [39]-[40] per Christopher Clarke LJ (approving the reasoning of Leggatt J at first instance, at [2013] EWHC 1409 (QB), in particular at [93], [101] and [104]).
ii) However, this Court can take into account the substance ofthe underlying evidence as set out in prior judgments (such as the contents of documents or the evidence of witnesses), givingthis such weight as is appropriate (and on the basis that it isentirely open to Mr Khrapunov to challenge that evidence andadduce other evidence): Rogers v Hoyle [2013] EWHC 1409 (QB) at [115]-[117] ; [2014] EWCA Civ 257 at [54] and [99]; and JSC A Bank v Ablyazov and another [2016] EWHC 3071 (Comm) at [24]. Furthermore, as detailed below, some of the same underlying evidence was, in fact, also before me” [emphasis added].
Although in oral submissions counsel for the claimant appeared to accept this authority, he nevertheless submitted that if the court did have regard to the evidence set out in the judgment it must be in the context that the court does not know precisely what evidence was before the court in Bestfort.
It seems to me that in its judgment at [45] the Court of Appeal made clear findings about the evidence before it and whilst this court has to consider all the circumstances of this case, given the test is one of “risk” and the flexible approach advocated by Gloster LJ, the court should take into account the findings set out in Bestfort whilst also having regard to the evidence submitted in this case, including the evidence in opposition of Mr Adeishvili.
In my view therefore, the indications given by the Commercial Court Guide are correct that evidence on this type of interlocutory hearing can be given by witness statements and does not have to comply with Part 35, although the court will still need to be satisfied having regard to the totality of the evidence before the court, that the evidence is sufficient to establish the risk of non-enforcement.
Evidence of Georgian law as to risk
Turning then to consider whether the evidence before this court shows that on objectively justified grounds relating to obstacles to, or the burden of, enforcement, there is a real risk that the defendant will not be in a position to enforce an order for costs.
The evidence as summarised in Bestfort was as follows:
“45. The relevant evidence relating to the difficulties of enforcement in Ras Al Khaimah and Georgia of any order of the English court in relation to costs may be summarised by reference to the account set out in the appellants' skeleton argument in this court from which Mr Millett did not demur. There was, not surprisingly, given the interlocutory nature of the application, no cross-examination of the experts on what were relatively short reports.”
i) …
ii) So far as enforcement of any costs order in Georgia was concerned, the appellants/defendants' evidence, supported by their Georgian law expert, Professor Kereselidze, was to the effect that there was a real risk that the appellants would not be able to obtain recognition in Georgia of any costs order made in their favour in these proceedings, due to the terms of art.68 of the Law of Georgia on International Private Law (the "IPL") which provided the basis upon which Georgian courts may recognise a foreign decision. That showed that a foreign judgment may not be enforced in Georgia:
a) under art.68(2)(e), where the foreign country does not recognise court decisions of Georgia;
b) under art.68(2)(f) if "proceedings are pending in Georgia between the same parties on the same issue and on the same basis"; and
c) under art.68(2)(g) if "the decision contradicts the basic legal principles of Georgia".
iii) As regards art.68(2)(e), Professor Kereselidze opined that, given that there are no multilateral or bilateral enforcement treaties in place between Georgia and England, there was a real risk that an English judgment for costs would not be enforced in Georgia. Professor Kereselidze drew attention to two decisions of the Georgian Courts where money or property judgments had not been recognised given the absence of any international agreement, and, consequently, the absence of any obligation on the foreign state to recognise Georgian court decisions: the first (Ruling No. A-2046-SH-57-2010 of 20 December 2010), a decision in respect of a judgment of the courts of Israel, and the second, an application to recognise a foreign judgment dealing with the distribution of matrimonial property.
iv) In response, the respondents' Georgian law expert, Professor Ninidze, opined that the view expressed by Professor Kereselidze that a Georgian court would not recognise an order made in a foreign country due to the absence of a bilateral or multilateral treaty was outdated and no longer followed by the Supreme Court of Georgia. He expressed the view that more recent authority demonstrated that the Georgian court was very willing and prepared to recognise legally effective foreign judgments, unless there was positive proof that Georgian judgments would not be recognised in that jurisdiction. Accordingly, he said that the lack of a bilateral or multilateralconvention did not prevent recognition of a foreign judgment. In support of this view he exhibited four matrimonial cases in which the Georgian courts had recognised foreign divorces; two German and two Greek.
v) In reply Professor Kereselidze expressed the view that family law cases (such as recognition of a marriage or divorce) were cases which arose in a completely different context and were not relevant for present purposes. He expressed the view that familycases were treated differently from cases concerning commerciallaw or property, and that in a family law context the Georgian courts had a greater expectation that their judgments would be enforced abroad and, as such, were more willing to enforce foreign family law judgments in Georgia, which usually involved recognition but did not require enforcement. However, he went on to say that outside the family law context, the position was uncertain. He said:
"One line of authority…… states that absence of bilateral or multilateral treaties is about recognition. It is true that there is a different line of authority, even outside family law cases, which appears to acknowledge that foreign court decisions may be recognised despite the absence of multilateral bilateral treaties (see for example cases..). However it is not correct to say, as Professor Ninidze does, that the decision of A-2046-SH-57-2010 is "outdated and "no longer followed". The reality is that there isno discernible consistency in the approach of the GeorgianCourts, and as such it is very difficult to predict the outcome inany particular case (See for example case of #a-1369-sh-30-2012 where a foreign court decision was again not recognised). The decisions of the Georgian Courts do not always contain a fully reasoned judgment and so this further increases the difficulty in deducing their approach to this issue with any certainty."
vi) So far as art.68(2)(f) of the IPL was concerned, Professor Kereselidze's evidence was that recognition might be refused on this basis, on the grounds that the s.25 proceedings in England and the claims brought by the respondents currently pending in the Georgian courts were arguably "between the same parties on the same issue and on the same basis". In his response, Professor Ninidze rejected that view, stating that that was "simply incorrect" since the English proceedings were proceedings for worldwide freezing orders and for the appointment of receivers and, as such, there could "never be proceedings on the same issues and on the same basis" in Georgia. In response, Professor
Kereselidze disagreed, stating that in circumstances where the English s.25 proceedings were ancillary proceedings commenced by the respondents/claimants in support of their Georgian claims, it was at least arguable that the Georgian courts would consider it appropriate to refuse recognition on the basis of art.68.2(f), at least until such time as the Georgian claims were resolved in Georgia.”
The evidence accepted by the Court of Appeal in Bestfort came from Professor Kereselidze. In the memorandum prepared in connection with this application Professor Kereselidze states that he remains of the view which he expressed in Bestfort that there is a “real and serious risk that a party to English court proceedings will be unable to obtain recognition and enforcement in Georgia of an English court order or judgment including a judgment or order as to legal costs”.
In the memoranda Professor Kereselidze (as in Bestfort) refers to Article 68(2)(e) of the Law of Georgia on International Private Law and states that pursuant to paragraph 2 (e) Georgia does not recognise a foreign court decision if a foreign country does not recognise Georgian court decisions. He identifies two scenarios: where there is no bilateral or mulitilateral agreement between Georgia and the foreign country and where the foreign country does not recognise Georgian court decisions.
As to the former, counsel for the claimant submitted orally that the existence of the Association Agreementcould amount to such a multilateral agreement. It was put to counsel for the claimant by the court that neither Professor Kereselidze nor the claimant’s expert had made any reference to the Association Agreement as having this effect, to which counsel replied that the report of his expert was merely responsive to the memoranda of Professor Kereselidze.
In my view if the Association Agreement had amounted to such a multilateral agreement one or both of the experts on Georgian law would have addressed this in their reports. The reports have apparently considered the case law extensively and thus both experts have spent some time considering the issue on this application in issuing both their original reports and the supplemental reports. Further on its face it would appear that the Association Agreement does not go so far as to establish a basis for recognition of judgments. As set out above, Article 21 headed “Legal cooperation” specifically refers to the Hague Convention but refers only to developing judicial cooperation in this regard. I do not accept therefore the submission that the Association Agreement has any impact on the test of whether there is a real risk.
As to whether the English courts failed to recognise Georgian decisions, counsel for the claimant submitted that Professor Kereselidze failed to identify why Article 68 applied.
Although counsel for the claimant submitted that the evidence of Professor Kereselidze was “defendant centred”, there is no basis in my view for such a submission and also no basis to prefer the evidence of one expert over the other. Both appear to be experts in their field and appear to have considered the case law, albeit that they disagree in their conclusions.
In his third memorandum Professor Kereselidze responds to the cases identified by the claimant’s expert and gives reasons why he disagrees with the views expressed. He maintains his view that there is considerable uncertainty as to whether enforcement will be effective at all. He states that this is primarily because there is no bilateral or multilateral treaty on recognition and enforcement and further that the Georgian courts do not always contain a fully reasoned judgment and therefore it is difficult to predict the outcome of any particular case in circumstances where there is no consistency in the approach taken and no reasonable number of cases in order to draw a conclusion.
Gloster LJ in Bestfort concurred with the proposition that as on a freezing injunction where a claimant has to adduce “solid evidence of risk of dissipation by the defendant” to support his assertion, nevertheless “since each case depends on its own facts, it is impossible to lay down any general guidelines on satisfying this evidential burden”. At [77] Gloster LJ said:
“Obviously there must be a proper basis for considering that such obstacles may exist…but whether the evidence is sufficient in any particular case to satisfy the judge that there is a real risk of serious obstacles to enforcement will depend on the circumstances of the case”
There is inherent uncertainty in this type of application but the court cannot resolve the disputed issues as to Georgian law. In my view the defendant has adduced sufficient evidence by the evidence of Professor Kereselidze to show that on objectively justified grounds relating to obstacles to enforcement, there is a real risk that the defendant will not be in a position to enforce an order for costs against the claimant.
Exercise of discretion
Counsel for the claimant submitted that the terms of the Association Agreement are also relevant as to whether the court can make an order which is non-discriminatory, the exercise of the discretion and the extent of any security granted.
As discussed above, in my view the nature of the Association Agreement does not alter the risk which the defendant faces on enforcement: it does not on its face amount to an agreement to enforce judgments nor do either of the experts concur with the submission made on behalf of the claimant in this regard.
Counsel for the claimant further submitted that the exercise of the discretion must comply with Articles 6 and 14 of the ECHR.
Counsel appeared to accept in written submissions that the discretion can be exercised in a non-discriminatory way if there are objectively justified grounds relating to obstacles to or to the burden of enforcement (Nasser v United Bankof Kuwait [2002] 1 WLR 1868 at [61]).
The issue of discrimination and Articles 6 and 14 of the ECHR was considered at length in Bestfort and Gloster LJ concluded at [67] and [68] that what was required was some rational justification for the system that treats certain categories of litigants differently from others and Gloster LJ held that there was a rational justification for the system contained in CPR 25.13. The discretion must be exercised on objectively rational grounds by reference to the difficulties of enforcement but that exercise is not subject to the “severe scrutiny” justification.
There are in my view no other factors which militate against the exercise of the discretion. I note the submission that the claimant has paid sums on account of previous costs orders but this is insufficient in my view to outweigh the risk which the defendant faces. I also note that the evidence of Mr Swain (paragraph 25 of his second witness statement) is that the claimant has refused to provide information as to whether he has assets in the U.K. or elsewhere in the EU against which a costs order could be enforced and the claimant has not identified any prejudice to him from being ordered to provide security.
Accordingly for these reasons I conclude that in all the circumstances it is just to order security for costs in this case.
Quantum
Given the conclusion that the defendant has shown a real risk, the defendant is entitled to security for the full amount: Chemukhin at [64]:
“In my judgment, once it has been established that there are
“substantial obstacles” sufficient to create a real risk of nonenforcement, the starting point is that the defendant should have security for the entirety of the costs and there is no room for discounting the security figure by grading the risk using a sliding scale approach.”
Accordingly for the reasons set out above, the application for security for costs is granted and the claimant must provide security for the full amount of the costs until the close of pleadings. The precise amount of such costs will be determined by the court having heard submissions at the consequential hearing on the defendant’s costs estimate.