 |
[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback] | |
Jersey Unreported Judgments |
||
|
You are here: BAILII >> Databases >> Jersey Unreported Judgments >> Imperium Trustees (Jersey) Limited v Jersey Competent Authority [2024] JCA 014 (18 January 2024) URL: http://www.bailii.org/je/cases/UR/2024/2024_014.html Cite as: [2024] JCA 014, [2024] JCA 14 |
||
[New search] [Help]
|
Before : |
Sir William Bailhache KC JA President The Rt Hon James Wolffe KC JA Mr Paul Matthews JA |
|
Between |
Imperium Trustees (Jersey) Limited |
Appellant/Applicant |
|
And |
Jersey Competent Authority |
Respondent |
Advocate J. Harvey-Hills for the Appellant/Applicant.
Advocate G. G. P. White for the Respondent.
HM Attorney General in person.
judgment
Matthews JA:
1. Although this is my judgment rather than that of the Court, all its members have contributed to it, and most of it is agreed by the Court as a whole. The result is that there will be no order as to costs, because of the application of the International Cooperation (Protection from Liability) (Jersey) Law 2018 (the "2018 Law"), but we will make a declaration of incompatibility under Article 5 of the Human Rights (Jersey) Law 2000. The President has written a short concurring judgment, adding some observations of his own. Wolffe JA disagrees with paragraphs 173 to 187 of my judgment, which discuss whether the 2018 Law pursues a legitimate aim, and he has accordingly written a separate judgment addressing that issue and adding some further observations on proportionality.
2. On 20 May 2022 the Royal Court refused the Appellant's application for leave to apply for judicial review: see [2022] JRC 300. The Appellant's complaint was that the Respondent had unlawfully demanded that it provide information said to be relevant to foreign taxation under the Taxation (Exchange of Information with Third Countries) (Jersey) Regulations 2008 and the Taxation (implementation) (Convention on Mutual Administrative Assistance in Tax Matters) (Jersey) Regulations 2014. On 27 May 2022 the Royal Court, by consent, ordered that there be no order as to costs.
3. On 9 June 2022 the Appellant gave notice of appeal, submitting that this Court, rather than the Judicial Committee of the Privy Council, had the relevant appellate jurisdiction. On 22 September 2022, this court held that it had jurisdiction, and ordered that the costs of the jurisdiction appeal be dealt with on paper: [2022] JCA 196. On 11 October 2022 this court directed that the costs of the jurisdiction appeal should be determined after the substantive appeal. The substantive application for leave to appeal and the substantive appeal were heard together on 24 January 2023. On 1 June 2023, this court granted leave to appeal, allowed the appeal, and gave leave to apply for judicial review: see [2023] JCA 057.
4. The Appellant now submits that this court should order the costs of both the jurisdiction and substantive appeals to be the Appellant's costs in the cause, and also that (seeking to amend the original order for no order as to costs by consent) the costs of the leave application should be the Appellant's costs in the cause. The Respondent resists that, on the basis that the 2018 Law protects the Respondent from liability for costs in these circumstances.
5. In its written contentions (at [28]), the Appellant counters that this legislation violates the European Convention on Human Rights (the "Convention") and accordingly under the Human Rights (Jersey) Law 2000 (the "2000 Law") should be read down or declared to be incompatible (although at the hearing the Appellant said it was not seeking a declaration, merely a reading down). The Attorney General was notified of this under Article 6 of that Law, and he has been joined to the proceedings accordingly.
6. We heard the parties at a hearing on 20 September 2023. In advance of the hearing the Attorney General filed an affidavit by Tom Leveridge, an Assistant Legal Adviser in his Department with a summons seeking leave to adduce the affidavit in evidence. Advocate Harvey-Hills, on behalf of the Appellant, intimated his intention to seek leave to cross-examine Mr Leveridge. In light of this, the Attorney General proposed that Mr Leveridge's affidavit simply be treated as adducing the documents referred to in it, with the exception of two specific paragraphs which he proposed to deal with by way of submissions. Advocate Harvey-Hills confirmed his agreement to the Attorney General's proposal and we have dealt with the affidavit on that basis.
7. After the hearing, we invited and received further written submissions on the Supreme Court decision in Coventry v Lawrence [2015] 1 WLR 3485. Those from the Appellant also included reference to the decision of the European Court in Coventry v United Kingdom [2022] ECHR 816.
8. I will address, first, certain submissions advanced by the Appellant in relation to the general law of costs, before considering the submissions advanced before us on the 2018 Law.
9. Article 16 of the Court of Appeal (Jersey) Law 1961 provides:
There are no specific provisions in the Court of Appeal (Civil) Rules 1964 (or, for that matter, in the Royal Court Rules 2004) dealing with costs, other than rules regulating the procedure for taxation of costs (for the Court of Appeal, see rule 18 of the 1964 Rules).
10. The general principles which apply to the Court's costs jurisdiction, both in the Royal Court and in this Court, are not controversial. In MB & Services Ltd v United Company Rusal plc [2020] (1) JLR N-11, Sir Michael Birt, Commissioner, said (para. 6):
11. Although these are the general principles which structure the court's discretion as regards costs, they do not limit the considerations (so far as relevant to the exercise of the Court's costs jurisdiction) which the Court may properly take into account when exercising its discretion as regards costs. So, whilst the general rule is that costs "should follow the event", principle (ii) envisages that there may be circumstances in which some other order should be made.
12. The Appellant drew our attention to three decisions of the Royal Court (Jersey Financial Services Commission v AP Black (Jersey) Ltd [2007] JLR 1, Volaw Trust and Corporate Services Ltd v Comptroller of Taxes [2013] (2) JLR 203 and AG v Rosenlund [2016] (1) JLR 348), in which the Court concluded that the fact that one of the parties was a public authority exercising public functions was a relevant consideration which justified departure from the general principle that costs should follow success. The Respondent did not grapple with these submissions, relying simply on the 2018 Law. Indeed in argument before us, Advocate White, for the Respondent, accepted that, if it were not for the 2018 Law, the costs award sought by the Appellant in relation to the Court of Appeal proceedings would be appropriate. Nevertheless, I should express my views on the Appellant's submission.
13. The background to the approach taken in the three Jersey cases to which I have referred was the decision of the (English) Divisional Court in Bradford Metropolitan District Council v Booth [2000] 164 JP 485. This was an appeal against an award of costs which had been made by justices against a local authority when the justices had overturned the authority's refusal to renew a licence. The question of costs was governed by section 64 of the Magistrates Courts Act 1980 which provides:
14. The Magistrates posed two questions for the opinion of the Court:
15. The Divisional Court answered these questions: a) No; and b) Yes. Lord Bingham CJ held that the proper approach in such cases could be summarised in three propositions:
16. This decision accordingly supports the conclusion that, in the type of case with which it was concerned, the "need to encourage public authorities to make and stand by honest, reasonable and apparently sound administrative decisions made in the public interest without fear of exposure to undue financial prejudice if the decision is successfully challenged" is a factor relevant to the determination of costs. But it is not a determinative factor and, in particular, the financial prejudice to the particular complainant in the particular circumstances is also identified as a consideration which should be taken into account.
17. In Jersey Financial Services Commission v AP Black (Jersey) Ltd, Mrs Black had been joined to proceedings alleging a collective investments fraud scheme. She successfully applied to strike out the claim as against her and sought her costs. Page, Commissioner, dismissed the application for costs. The Commissioner referred to Bradford Metropolitan District Council v. Booth and other English cases, as well as an earlier Jersey decision, Ani v. Barclays Private Bank & Trust Ltd [2004] JLR 165. He drew the following conclusions:
He held that Mrs Black should bear her own costs of the strike-out application.
18. In Volaw Trust and Corporate Services Ltd v Comptroller of Taxes, the Comptroller had served four successive notices on the applicant requiring the same or similar information to be provided under the Taxation (Exchange of Information with Third Countries) (Jersey) Regulations 2008, at the instance of the Norwegian authorities. The Comptroller withdrew the first notice. The applicant sought to set aside the second, third and fourth notices. The second and third were then withdrawn by the Comptroller but the application to set aside the fourth failed. The Comptroller applied for his costs in relation to the second to fourth notices, even though he had withdrawn the second and third and had been successful only in relation to the fourth. The applicant argued that Jersey Financial Services Commission v Black had been wrongly decided, or at least that it should be confined to cases involving regulatory or disciplinary bodies. Commissioner Page re-affirmed the approach which he had taken in Jersey Financial Services Commission v Black and awarded the Comptroller 90% of his costs.
19. In AG v Rosenlund, at the request of the Danish authorities under the European Convention on Mutual Assistance in Criminal Matters, the Attorney General sought a saisie judiciare of the first Respondent's realisable assets in Jersey. The application was refused on the ground that it was plain and obvious that it could not succeed. The first Respondent applied for costs against the Attorney General. The Commissioner (Clyde- Smith) held that it would not be appropriate to make a costs order against the Attorney General for the period before he had been provided with evidence showing that the application was without possibility of success.
20. The Commissioner said:
21. In arguing that these three cases were wrongly decided and should not be followed, the Appellant relies on the judgment of Lord Neuberger MR (with whom Hallett and Stanley Burnton LJJ agreed) in R (M) v Croydon London Borough Council [2012] 1 WLR 2607, CA. That was a case where the claim for judicial review was settled without a contested hearing, but the question of costs was left over to be dealt with by the court. In answering the question, the court looked at the costs position both in cases where there was a settlement, and in cases where there was not and the court had to give a decision.
22. Lord Neuberger said:
23. R(M) was a case governed by the England and Wales Civil Procedure Rules, which contain detailed provisions structuring the Court's discretion as regards costs, notably by stating explicitly that if the court decides to make an order as to costs, "the general rule is that the unsuccessful party will be ordered to pay the costs of the successful party": CPR r. 44.2(2)(a). In that regard it contrasts both with the statutory regime which governs costs in Jersey and with the terms of section 64 of the Magistrates Courts Act 1980 which was at issue in the Bradford case.
24. R(M) may be contrasted with another decision of the Court of Appeal (Lord Neuberger MR, Maurice Kay and Stanley Burnton LJJ) issued just two years before R(M): R (Perinpanathan) v Westminster Magistrates' Court [2010] 1 WLR 1508. The claimant's daughter was stopped on arrival at Heathrow Airport and found to be carrying cash to the value of £150,000. The police suspected that the money had been intended for use to finance terrorism and seized it. The matter came before the magistrates' court, which held that the police had reasonable grounds for their suspicions, but was not satisfied, on the evidence, that the money was indeed intended for terrorist purposes and ordered its return. However, the court refused to make an order for the claimant's costs. The claimant sought judicial review of that decision. She failed at first instance and appealed to the Court of Appeal.
25. Stanley Burnton LJ (with whom Maurice Kay LJ agreed) gave the first judgment. He discussed the then authorities, including Bradford Metropolitan District Council v Booth and Baxendale-Walker v The Law Society [2008] 1 WLR 426, in which the Court of Appeal had applied the Bradford approach to professional disciplinary proceedings. He then said this:
26. Lord Neuberger MR set out his own reasons (which, he said, "largely reflect those of Stanley Burnton LJ") as follows, again agreed in by the third member of the court, Maurice Kay LJ. After referring to the general rule applicable to cases governed by the CPR, he stated:
27. In light of the observations in these judgments about the CPR, it is unsurprising that the Court of Appeal in R(M), which was concerned with the application of the CPR, did not mention the Bradford line of authority. Equally, R(M) cannot be taken to have over-ruled, or to have cast doubt on, that line of authority in the cases to which it applies.
28. The Bradford line of authority has recently been considered by the UK Supreme Court in Competition and Markets Authority v Flynn Pharma Ltd [2022] 1 WLR 2972. There, the question was what costs order ought to be made by the Competition Appeals Tribunal where it allowed the appeals of companies against a decision of the Competition and Markets Authority imposing fines on them for infringements of competition law. The Tribunal had applied a starting point that costs followed the event. The Court of Appeal overturned both this order and the starting point, on the basis that the Tribunal had disregarded the principle derived from Bradford Metropolitan District Council v Booth. The Supreme Court allowed a further appeal, restoring the costs order of the Tribunal.
29. Lady Rose JSC (with whom all the other judges agreed) said:
30. Lady Rose went on to explain at [120] that whilst the Competition Appeal Tribunal had generally adopted a starting point of costs following the event, it had, in relation to certain types of proceedings, taken the chilling effect into account by adopting a no order as to costs starting point. Further, even where the Tribunal adopted a starting point of costs following the event, in appropriate cases it would make no order as to costs. She noted at [149] the variety of factors which the Tribunal took into account, including, in appropriate cases: "the great disparity in resources between the appellants and the competition authority and the fact that the costs though high were small relative to the total turnover of the appellants".
31. The decision of the UK Supreme Court accordingly endorses the view that the risk that there will be a chilling effect on the conduct of a public body, if costs orders are routinely made against that body, may, in certain types of case, be an "important factor" in considering how a Court should approach awards of costs. The UK Supreme Court did not doubt the correctness of the Bradford line of authority in the cases to which it is properly applied and confirmed that, in relation to certain types of proceedings, the risk of a chilling effect is a consideration which may even justify adopting as a starting point the proposition that the Court should not make any costs order against a public authority.
32. At the same time, the UK Supreme Court rejected the proposition that it is enough to justify departing from the usual starting point that costs should follows success that one of the parties is a public authority. A real risk of a chilling effect "cannot be assumed to exist" (at [98]). An assessment must be made, in the context of any particular jurisdiction, not only "as to whether a chilling effect is sufficiently plausible to justify a starting point of no order as to costs" (at [98]) but as to whether the risk of a chilling effect is something which should be taken into account at all. And even where the court concludes that no order as to costs is the right "starting point", the question of what order as to costs is appropriate in any given case would require to be addressed in light of all the relevant facts and circumstances of which the "chilling effect" would only be one.
33. The statutory regime in Jersey gives the Court a wide discretion as regards costs. It does not, in terms, mandate a particular starting point to be applied in all cases; nor does it limit the considerations which may properly be taken into account by the Court. In that respect it is analogous in nature to the statutory provision which was at issue in the Bradford line of cases. As I have shown, the authority upon which the Respondent relies, R(M), was a decision on the application of the England and Wales Civil Procedure Rules, which, by contrast with the Jersey statute, contains an explicit statutory provision that: "the general rule is that the unsuccessful party will be ordered to pay the costs of the successful party".
34. A fuller examination of the caselaw from England and Wales discloses that the Courts in that jurisdiction have recognised that the fact that a party is a public authority exercising public functions is, in certain types of case, a relevant consideration when deciding whether or not costs should be awarded against that party. Further, in certain types of case, the risk that the routine award of costs against a public authority which has acted reasonably in bringing or defending an application will have a chilling effect may even justify a starting point that the Court will make no order as to costs against the public authority, where the statutory regime permits that.
35. I see no good reason why a like approach should not, in principle, be applicable in Jersey in appropriate cases. The underlying rationale for taking into account, in the context of costs, the fact that a party is exercising public functions is, as Lord Bingham explained in Bradford, the desirability that public authorities should make and stand by honest, reasonable and apparently sound administrative decisions in the public interest. Public authorities should not be deterred from taking decisions in the exercise of their functions by the risk of litigation; nor should they be deterred from defending apparently sound decisions by the potential to face awards of costs.
36. However, as the judgment of the UK Supreme Court in Flynn Pharma explains, it cannot and should not be assumed that in all cases involving public authorities there is a real risk of a "chilling effect". Indeed, the proper starting point, it seems to me, should be a working assumption and expectation that public authorities will, on the one hand, exercise their functions properly and without fear or favour and, on the other, take robust decisions when deciding how to respond to challenges to the lawfulness of their actions. The weight to be given, in any particular case, to the risk of a "chilling effect" should, it seems to us, be assessed against that background. There must, in other words, be some basis - some special factor beyond the fact that the case concerns a public authority - for concluding that there is, in relation to any particular case or type of case at issue, a real risk of a chilling effect, before that consideration should be given weight.
37. The general rule that costs follow success has a salutary effect in encouraging good practice in the conduct of litigation - including, in the case of public authorities, good practice in assessing whether a particular decision should be defended. Public law litigation is an important mechanism of accountability and if the Court were, too readily, to accede to applications by public authorities that they should not face adverse costs orders, this could discourage well-founded challenges from being brought forward and lead to public authorities defending cases which should be conceded. These systemic considerations are part of the background, which justifies requiring some special factor beyond the fact that the case concerns a public authority, when the Court is considering whether, in a particular context, the risk of a chilling effect should be given significant weight.
38. Where the Court does conclude that the risk of a chilling effect falls to be given weight, it must also, of course, take full account of the other considerations relevant to the exercise of its costs jurisdiction in any particular case. These will include, as Lord Bingham made clear in the City of Bradford case, the financial prejudice to the successful claimant in the particular circumstances, if an award of costs is not made in its favour. There is no question, when applying the Court's general costs jurisdiction, of a blanket rule such as that which is enacted in the 2018 Law.
39. The present case is not one of those categories of case (licensing, disciplinary and like cases) to which the Bradford line of authority has been applied in England and Wales - and I do not need to address in this case whether costs awards against public authorities in those types of cases in Jersey would involve a risk of a chilling effect. Were I free to apply our general costs jurisdiction, I would need to address whether in cases of the particular sort before us there is some special factor which would justify taking into account a potential chilling effect. Since Advocate White did not seek to contend, other than by reference to the 2018 Law, that the costs order sought by the Appellant in relation to the appeal proceedings would be inappropriate, I need not address that question.
40. Nor would it be appropriate for us to review in detail the three Jersey cases to which I have referred. Each of them involved the exercise of the Court's discretion as regards costs in particular circumstances. At the level of principle, I do not consider that Commissioner Page can be criticised for treating the need to encourage public authorities to make and stand by honest, reasonable and apparently sound administrative decisions made in the public interest as a "relevant factor", as he did in Jersey Financial Services Commission v AP Black (Jersey) Ltd) and in Volaw Trust and Corporate Services Ltd v Comptroller of Taxes. He rightly emphasised that "the matter is best dealt with simply on that basis - as one element relevant to the court's exercise of discretion in any particular case - rather than treating that body's status as automatically giving rise to a hard-and-fast special rule, or, for that matter, even a prima facie rule. ...". For the reasons which I have explained, it would be wrong to proceed on the basis that this is always a consideration which falls to be given "considerable weight", to use the words of Commissioner Clyde-Smith in AG v Rosenlund. Whether or not it is a factor which should be taken into account - and, if it is, the weight to be attached to it relative to other considerations - will depend on the nature of the case, and its particular circumstances.
41. As I have explained, the Respondent did not engage with the submissions on the three Jersey cases to which I have referred, but instead relies on the provisions of the 2018 Law, to submit that there should be no order for costs. That law, so far as material, provides as follows:
42. Schedule 1 specifies the following enactments:
· Bankers' Books Evidence (Jersey) Law 1986
· Civil Asset Recovery (International Co-Operation) (Jersey) Law 2007
· Competition (Jersey) Law 2005
· Criminal Justice (International Co-Operation) (Jersey) Law 2001
· Financial Services (Jersey) Law 1998
· International Criminal Court (Jersey) Law 2014
· Investigation of Fraud (Jersey) Law 1991
· Proceeds of Crime (Jersey) Law 1999
· Taxation (Implementation) (Jersey) Law 2004
43. The Respondent is a public authority within the meaning of this Law and was acting in the discharge or purported discharge of its functions under an enactment specified in schedule 1 to the law to assist a relevant authority outside Jersey. It has not been suggested, and neither has it been held, that it acted in bad faith. Paragraph (2) of Article 2 does not apply, because this case concerns an order for costs rather than an award of damages, though it may raise the question whether costs refused under this Law could nevertheless be recovered as damages.
44. In response, the Appellant submits that the 2018 Law is incompatible with the European Convention on Human Rights. It relies on a line of European Court of Human Rights jurisprudence, culminating in Zustovic v Croatia [2022] 74 EHRR 3, in which it was held by the ECHR that legislation that required the parties to judicial review proceedings involving public bodies to bear their own costs of litigation violated Article 6(1) of the Convention.
45. Before examining these arguments, I should describe the material which set before us about the passage of the 2018 Law. It is appropriate to look at this material for two reasons. The first is to identify whether there is anything outside the terms of the statute which assists us in identifying the policy objective of the measure. It is clear from the decision of the House of Lords in Wilson v Secretary of State for Trade and Industry [2004] 1 AC 816 that, in considering the compatibility of primary legislation with the European Convention on Human Rights, it is legitimate for the courts to look at legislative travaux préparatoires when considering the aims of the legislation and its proportionality to those aims - see Lord Nicholls at [61] to [67] and Lord Hope at [116] to [118].
46. The second reason is to enable us to weigh better the Attorney General's submission that this legislation falls within the margin of appreciation that the European Court of Human Rights allows to an individual member state when the compatibility of its legislation with the Convention comes to be considered and/or is a proportionate response to the problem which the policy is intended to address. As the European Court explained in Animal Defenders International v United Kingdom [2013] 57 EHRR 21, paras. [106]-[111], [113]-[116], the quality of the consideration of the issues by the domestic legislature and courts may bear on the Strasbourg margin of appreciation. In that case, which concerned the UK's prohibition on political advertising, the Court attached "considerable weight" to the "exacting and pertinent reviews, by both parliamentary and judicial bodies" of the relevant issues at [116].
47. If the legislature has been directed to the relevant issues and appears to have addressed them in a thorough fashion, it will be much easier for the domestic court to hold that the measure is within the Strasbourg margin of appreciation. Where, on the other hand, it is apparent that the legislature has neither been advised about the key issues, nor appears to have considered them, it will be much harder to accept that it has balanced the various factors on either side and made an informed political choice as to the particular needs of the island's community such as to justify a wide margin of appreciation, although the court may nevertheless determine, on the basis of the materials before it, that the legislation is a proportionate response to a legitimate aim.
48. It is important to recognise, as I do, that in this exercise I am not carrying out a judicial review of the work of the legislature in a conventional sense on any of the standard grounds, and still less am I reaching any view as to the merits of the legislation from any political perspective. To the extent that I comment upon the work done in the legislature, this is solely for the purpose of assessing whether this material assists us in our consideration of whether the costs rule in the 2018 Law is compatible with Convention rights - an exercise which the States themselves have mandated in enacting the 2000 Law.
49. The Projet de Loi which became the 2018 Law says in the Report to the States (in part):
50. I note that the opening section of the Projet de Loi states that: "Where a person (usually the object of the investigation) has any complaint as to the reasonableness of the investigation, then such complaints can be addressed to the home jurisdiction of the investigators". But the challenger will not necessarily be the foreign taxpayer (and the costs rule is not limited to tax information cases) and a person in receipt of a notice in Jersey would not necessarily have any locus to challenge the making of a request in a foreign jurisdiction. The facts of this case illustrate the point. The Appellant would not necessarily have locus to challenge the requesting competent authority in Belgium; nor indeed is it obvious that the ultimate beneficiary of the trusts administered by the Appellant would have such a right either.
51. There is a statement of compatibility from the then Chief Minister, which accords with Article 16 of the 2000 Law, and there is no accompanying reasoning as to what had been considered. However, the Report is followed by the Human Rights Notes, to the detail of which I return later.
52. In the legislative debate in the States of Jersey, where the draft law was adopted unanimously, the Constable of St Ouen (Assistant Chief Minister and rapporteur) referred to existing provisions of Jersey law which contained similar protections. Following a reference to the Schedule, he is, rather surprisingly, recorded as stating:
53. In his submissions, the Attorney General invited us to conclude that the word "not" must be a mis-transcription of what the Assistant Chief Minister had said, since it is plain that these are bodies to which the Law does extend protection. I accept that submission. It may just be a typographical error in the preparation of the Hansard report and should have read "now" for "not".
54. The Assistant Chief Minister had introduced the draft law in these terms:
55. The Hansard record of the debate reveals only two contributions from members on the proposed law. Senator Moore referred to examination of the draft law by the Corporate Services Scrutiny Panel and consultation with the Jersey Law Society. She reported that the Panel had concluded it should support the Law and said:
56. Deputy Morel said:
57. In an Appendix to the Projet de Loi there is a Note prepared by the Law Officers' Department. In part this reads as follows:
58. The note from the Law Officers Department, which curiously is described as not providing legal advice, signally does not address the matters which have become apparent in the course of argument before us and which are addressed in this judgment. Indeed, the thrust of the comments in that note is that there are "no human rights issues" at all - indeed, that is stated expressly - and, as we will see, that is not a correct statement of the legal position. In the circumstances, it is hardly surprising that there was no debate on those issues. As I have explained above, this is relevant to our consideration of the Attorney General's submission that the Law is within the Strasbourg margin of appreciation. Where, as here, it is apparent that the legislature has neither been advised of the Convention rights issues which arise, nor appears to have considered the underlying substantive issues which would require to be addressed to determine whether any interference with Convention rights is justified, it is very hard to accept that the legislature has balanced the various factors on either side and made an informed political choice that the impact on Convention rights is justified by the particular needs of the island's community. In these circumstances, the legislative travaux préparatoires do not provide a basis for considering that the margin of appreciation should be a generous one.
59. The Appellant contends that the 2018 Law is incompatible with Article 6 of the European Convention on Human Rights. Advocate Harvey-Hills, supporting this contention, relied on two aspects of Article 6: the right of access to a court; and the principle of equality of arms. He submitted that the costs rule in Article 2 is a restriction on access to the court. He argued that it does not pursue a legitimate aim and, on that account alone, is therefore incompatible with Article 6. He contended that, in any event, the rule is disproportionate. He relied strongly on Zustavic v Croatia (2022) 74 EHRR 3. In any event, he contended that the disparity of treatment to which Article 2 of the 2018 Law gives rise breaches the principle of equality of arms. Advocate Harvey-Hills accordingly invited the Court, pursuant to its obligation under Article 4 of the 2000 Law, to read down Article 2 of the 2018 Law. He offered various possible ways of achieving what he contended would be a Convention-compliant reading of the Article.
60. The Respondent contends that Article 6 does not apply to the present dispute. Before us, Advocate White relied on Ferrazzini v Italy [2001] 34 EHRR 45, in which the Grand Chamber of the European Court of Human Rights held that fiscal disputes do not involve the determination of "civil rights and obligations", such as to engage Article 6. He pointed to Lindgren v Sweden, where, on facts which he contended are similar to the present ones, the Court held that Article 6 did not apply. In response, Advocate Harvey-Hill invited us to conclude that the present case did concern "civil rights and obligations" for the purposes of Article 6; which failing that it involved the determination of a "criminal charge".
61. The Attorney General supported Advocate White's submission that Article 6 does not apply on the basis of Ferrazzini, but he accepted that, were it not for Ferrazzini, Article 6 would apply. The main focus of his submissions was to contend that, if Article 6 does apply, the costs rule contained in Article 2 of the 2018 Law pursues a legitimate aim and is proportionate to that aim.
62. Article 2 of the 2000 Law provides that certain Articles of the Convention, as set out in schedule 1 to the Law, should have effect for the purposes of that Law. These Articles are known as the "Convention rights". Article 3 provides that the court determining a question which is arisen in connection with a Convention right must take into account certain things, including any judgment or decision of the European Court of Human Rights. Article 4 provides that, so far as possible, legislation must be read and given effect to in a way which is compatible with Convention rights.
63. Article 4(1) is modelled on section 3(1) of the UK Human Rights Act 1998. Jersey courts, considering the application of Article 4(1), will find assistance in the approach which the UK courts have taken to section 3(1) of the 1998 Act. As Lord Nicholls of Birkenhead explained in Ghaidan v Godin-Mendoza [2004] 2 AC 557, at [30], the language used (which is also used in Article 4) imposes on the Court an interpretive obligation of "an unusual and far-reaching character". It authorises, and obliges, the Court to read and give effect to principal legislation in a way which is compatible with Convention rights whenever "it is possible to do so". This does not depend on finding an ambiguity in the language of the statute. Lord Nicholls went on to observe at [30] that "to an extent bounded only by what is 'possible' a court can modify the meaning, and hence the effect" of legislation, provided only at [31] that the court may not "adopt a meaning inconsistent with a fundamental feature of the legislation".
64. Article 4(2) makes clear that if the Court reaches the conclusion, after fulfilling the interpretive obligation placed on it by Article 4(1), that a provision of principal legislation is incompatible with a Convention right, this conclusion does not affect the validity and continuing force and effect of the provision in question. In such a case, the Court may exercise the power given to it by Article 5 of the 2000 Law which provides:
65. This court is one of the courts which may make a declaration of incompatibility under this Article: Article 5(3). But such a declaration:
It is for the States Assembly to decide whether to amend legislation which the Court has declared to be incompatible with Convention rights.
66. Article 7(1) provides that it is unlawful for certain public authorities to act in a way which is incompatible with a Convention right. Article 7(2) makes clear that a court is a public authority for these purposes. But Article 7(6) provides Article 7(1) does not apply if -
67. It follows that the court would not be acting unlawfully under Article 7, if it were to act in a way which is incompatible with a Convention right, if, in effect, it is compelled to do so in order to give effect or to apply or enforce principal legislation, including Laws enacted by the States Assembly.
68. Article 8 provides that a person claiming that a public authority (including a court) has acted or proposes to act incompatibly with a Convention right may rely on that right in legal proceedings, "but only if the person is ... or ... would be ... a victim of the unlawful act". Article 9 provides that, where the court finds that the act or proposed act is or would be unlawful, "the court may grant such relief or remedy, or make such order, within its powers as it considers just and appropriate".
69. We are obliged by Article 3 of the 2000 Law to take decisions of the European Court of Human Rights into account. They are not, strictly speaking, binding on us. However, in deciding what weight to attach to decisions of the Strasbourg Court, we must bear in mind: (i) that the purpose of the 2000 Law was to enable the Convention rights to be enforced in the domestic law of Jersey; (ii) that the content of those rights is the same as the content of the rights articulated in the European Convention on Human Rights; and (iii) that it is the European Court of Human Rights which authoritatively defines the scope of those rights.
70. In my judgment, we should take the same approach to the authority of Strasbourg caselaw as the UK Supreme Court takes, when fulfilling its duty under section 2 of the UK Human Rights Act 1998, on which the 2000 Law is modelled. That section is, after all, in similar terms to article 3 of the 2000 Law. In Manchester City Council v Pinnock [2011] 2 AC 104, Lord Neuberger summarised the position as follows:
71. In R (Chester) v Secretary of State for Justice [2013] UKSC 63, Lord Mance added this:
72. I agree that we should follow any decision of the Grand Chamber, or any clear and constant jurisprudence of the Strasbourg Court, unless there is some very good reason not to do so. The observations which I have just quoted indicate the kind of thing which might, exceptionally, justify not following a Grand Chamber decision or a clear and constant line of Chamber authority. Whilst we are obliged to take into account any decision of the Strasbourg Court, we have greater scope to depart from a single Chamber decision or even from a line of decisions which has not crystallised into a clear and constant jurisprudence of the Court.
73. Article 8(1) of the 2000 Law provides that a person seeking to rely on a Convention right in legal proceedings may do so only if that person is a "victim of the unlawful act". Article 8(5) provides that:
74. In Coventry v United Kingdom [2022] ECHR 816, to which Advocate Harvey-Hills very properly drew our attention after the hearing, the European Court held that the original conditional fee agreements scheme in force in England and Wales under the Access to Justice Act 1999 was incompatible with Article 6 of the Convention. The applicant, as an uninsured unsuccessful party, complained of being exposed to very high costs liability as a result of the operation of the scheme. The applicant argued that this was a breach of the right of access to a court and also that it was a breach of the requirement of a fair trial, by reason of its incompatibility with the principle of equality of arms.
75. The court held that the applicant was not a victim in relation to the right, inherent in Article 6, of access to a court. It said:
On the other hand, the court considered that the applicant was a victim in relation to the right to a fair trial - it joined the question of "victim" status to the merits, and concluded that there had, indeed, been a breach of Article 6.
76. In the present case, it can be said that in fact the Appellant has brought this case and has raised the question of its rights before the court. It has therefore had access to a court. Indeed, it has been successful in what, in procedural terms, it has sought, so far. The Attorney General pointed us to the sums at stake in the underlying tax proceedings to suggest that the Appellant could be taken to be very well-funded. It is also evident that the Appellant has enjoyed a practical equality of arms in the hearings before us, in that it has been well-represented by a skilled advocate.
77. Although the Attorney General relied on considerations such as those which I have mentioned in support of his contentions that the 2018 Law is not incompatible with Article 6, he did not submit that the Appellant has no standing to advance the arguments which it has presented. The question of standing was accordingly not the subject of detailed analysis in the submissions before us but, for my own part, I consider that the Attorney General was right not to take that point.
78. I acknowledge that lack of funds does not seem to have constrained the Appellant in its approach to this litigation. Nevertheless, I also accept what Advocate Harvey-Hills said in argument, namely that every pound the Appellant spends on these proceedings by the Respondent is "dead money". The Appellant knows that, if it loses, it may expect to be ordered to pay the Respondent's costs. At the same time, if the Respondent is right and can rely on the 2018 Law, the Appellant also knows that it will not obtain an order for its costs even if it wins. If the Appellant loses, there is thus no barrier to an award of costs against the Appellant. There is, therefore, a clear difference of treatment between the parties, which is already a significant disadvantage for the Appellant.
79. Further, Advocate White accepted, before us, that in the absence of the 2018 Law, the appropriate order in relation to the appeal proceedings would be Appellants' costs in the cause. As I have explained, he did not seek to rely on the Jersey cases to which I have referred in support of any contention that, if the 2018 Law did not apply to exclude such an order, we should not make any order for costs against the Respondent in this case. It follows that, unless the 2018 Law can be read down by reason of Article 4(1) of the 2000 Law, the application of the costs rule in the 2018 Law has, at this stage of proceedings, a direct and detrimental effect on the Appellant, in that it will result in this Court rejecting the Appellant's application as to costs and making a different order from the order which it would have made if that costs rule did not apply.
80. I am satisfied, for these reasons, that the Appellant has "victim" status and has standing to advance the contentions which it has advanced.
81. I do not consider that Coventry prevents us from taking that view. There is a well-established line of Strasbourg jurisprudence to the effect that the incidence of costs at the end of litigation can constitute a restriction on the right of access to the court, and these were cases in which the point was allowed to be taken even though the victim had participated in the proceedings. The point may be illustrated by Černius and Rinkevičius v Lithuania, Applications 73579/17 and 14620/18, 18 June 2020.
82. In Černius and Rinkevičius each of the applicants had brought successful proceedings to challenge the validity of an administrative fine. The costs which they incurred for legal representation exceeded the amounts of the fines. The statutory regime governing such proceedings did not explicitly provide for the reimbursement of a successful applicant's cots of legal representation, and each of the applicants subsequently brought separate proceedings seeking to recover those costs. In each case, the Administrative Court rejected that application on the basis that the invalidity of the administrative decision to impose the fine did not amount to an "unlawful act" such as to give rise to a right to compensation. The first applicant's appeals against that decision were rejected.
83. The applicants applied to the European Court of Human Rights. The Government argued that the applicants had in fact participated in the proceedings and "[i]t followed that the applicants' right of access to court had not been impaired" (para. 58). The Strasbourg Court rejected that argument and held that the refusal to reimburse their legal costs amounted to a restriction on their right of access to a court. It observed (para [68]):
84. A similar approach was taken in Stankov v Bulgaria [2009] 49 EHRR 7, which concerned court fees. The payment of the court fees was not a precondition for initiating proceedings but became due only once the judgment of the court had become final (at [53]). As the Strasbourg Court observed, the "applicant thus had 'access to all states of the proceedings and the courts examined his case on the merits" (at [53]). Nevertheless, "the imposition of a considerable financial burden due after the conclusion of the proceedings may well act as a restriction on the right to a court", and, held the Strasbourg Court, the costs order against the applicant constituted such a restriction which, in the circumstances, breached Article 6 (at [54]).
85. It follows, in my view, that the fact that a litigant has initiated and pursued proceedings to a conclusion and has fully participated in those proceedings with the assistance of a lawyer, does not, as such, prevent the application of costs rules from being held to be a restriction on the right of access to a court.
86. The Appellant's contention is that the costs rule in the 2018 Law is, as such, an unjustified restriction on the right of access to a court. In our courts, costs orders are made at the end of the litigation. In terms of monetary consequences in costs, a party will not know until the end whether he faces the possibility of an adverse order or at best no order. If a party could not raise a contention such as that advanced by the Appellant if the party has participated in the proceedings, no one would ever have standing to advance that argument, and this does not seem to me to be correct. I will need to address whether the Appellant's contention is right or wrong, but it seems to us that the Appellant has standing to advance it. In any event, Coventry would not prevent the Appellant from arguing that the 2018 Law breaches the principle of equality of arms, and thereby incompatible with Article 6.
87. The first question is whether Article 6 of the Convention is engaged at all. That Article relevantly provides as follows:
88. I deal first with the "civil rights and obligations" limb of Article 6(1). Whatever the scope of this provision, it applies to a person only in "the determination of his civil rights and obligations". So, the immediate question is whether anyone's civil rights and obligations are being determined in these proceedings. The Respondent says not. It says that this is a tax dispute and relies on a line of cases from the European Court on Human Rights. The Appellant however says that Article 6(1) does apply. It refers to, inter alia, Jersey decided cases on the exchange of tax information where the court proceeded on the basis that Article 6 was engaged. These include Temple v Attorney General [2015] (1) JLR 203, [19] and Larsen v Comptroller of Taxes [2015] (2) JLR 209, [46]-[53].
89. The Respondent relies on Ferrazzini v Italy [2001] ECHR 464, 34 EHRR 45. In that case, the applicant had challenged assessments to tax by the tax authority. The domestic proceedings took over ten years to reach a first instance decision and over twelve to reach a second instance decision. He complained that this breached his Article 6 rights to a hearing "within a reasonable time". The Grand Chamber, by a majority of 11 to 6, held that disputes about tax liability did not concern "civil rights and obligations", and thus fell outside Article 6, and there was no violation of that Article.
90. The majority of the court said:
91. The minority took a different view, arguing that in 2001 there was no longer a good reason for treating disputes with the state about whether and if so how much tax is owed to the state as outside the protection of Article 6. They were disputes about money obligations. But the concept of "civil rights and obligations" is "autonomous", within the meaning of Article 6. It is therefore for the European Court, and not for us, to decide what it means. I proceed on the basis that disputes about tax liability do still fall outside Article 6(1).
92. Even so, the question is whether, as the Respondent submits, the present is a case about tax liability. Here the Appellant submits that it is not a dispute of that kind. It points out that no assessment to tax has been made against anybody. Furthermore, if one were made by the Belgian tax authorities, it would not be against the Appellant. The Jersey courts are not here asked to adjudicate on anyone's tax liability. Instead it is a dispute about the production of confidential documents. Is there then a "civil right" in dispute?
93. In Grzęda v Poland [2022] 53 BHRC 631, the European Court said:
94. Here, says the Appellant, what is in issue concerns the rights of confidentiality that otherwise exist in private documents held by a professional trustee in relation to a particular trust. In Viscount and Price Waterhouse Coopers v Attorney General [2002] JLR 268, Birt DB said:
(See also Heerema v Heerema [1985-86] JLR 293.)
95. The duty of confidentiality of the trustee towards the beneficiaries of the trust has its co-relative in the beneficiaries' right against the trustee that it keep the trust's affairs confidential. But the trustee is also prima facie entitled to confidentiality in its private dealings, whether in business or otherwise. In the present case, the dispute is whether the obligations and rights of the trustee and the rights of the beneficiaries have been lawfully overridden by the tax information legislation. Those rights and obligations, the Appellant says, are civil rights and obligations within Article 6(1).
96. This view is supported by reference to other decisions on Article 6(1). In National Provincial Building Society v United Kingdom [1997] 25 EHRR 127, the applicant had paid a tax which was later held to have been unlawful, as imposed by regulations which were ultra vires the principal statute. It sought to recover the overpayment by way of a restitutionary claim against the UK government. However, Parliament then enacted retrospective legislation, in effect denying those claims. The applicant claimed that this infringed various provisions of the Convention, including Article 6(1).
97. In relation to the Article 6(1) point, the UK government took a threshold point, that this was a tax claim and not the determination of civil rights and obligations. The European Court disagreed. It said:
98. In the present case, we are concerned not with a private law action like that at issue in National Provincial Building Society, but with a challenge to the lawfulness of investigatory steps which are, in turn, directed to establishing whether or not there is a tax liability and, if so, its amount. We were shown two decisions of the European Court of Human Rights concerned with whether a challenge to the lawfulness of investigatory steps undertaken by a tax authority engages Article 6: Ravon v France (18497/03) 21 February 2008; and Lindstrand Partners v Sweden [2016] ECHR 1139.
99. In Ravon, the French authorities obtained and executed court orders authorising them to conduct a "dawn raid" on residential premises, without warning, in connection with a suspected tax fraud. Under French law, such orders could be challenged only in subsequent substantive proceedings (if any) relating to a tax claim based on the documents obtained. The judges who granted the orders had no jurisdiction to review them or their execution.
100. The European Court distinguished Ferrazzini, and held that Article 6(1) was engaged. It said:
101. The reference to Article 9 of the Code Civil is to the right to respect for one's private life. That Article reads:
So, the rights in domestic French law to private life were civil rights for the purposes of Article 6(1).
102. A different conclusion was come to by the European Court in Lindstrand Partners v Sweden [2016] ECHR 1139, where, so far as one can tell, Ravon was not cited. The Swedish tax authorities conducted audits of three companies, and in 2008, suspecting tax evasion by one of them, obtained search and seizure orders from the local court, not communicated to the companies until they had been executed.
103. They were first executed at a flat occupied by (but not belonging to) a Mr Jurik, who had been in charge of the bookkeeping of the companies, and thereafter at his office, which was inside the premises of the applicant law firm. In fact a lawyer from the firm, representing Mr Jurik, was present when the orders were executed. Certain paper files and electronic media were seized at the flat. Nothing was found at the law firm.
104. After the searches, a request was made for the electronic media to be exempted from seizure, as protected by attorney-client privilege. The request for exemption was refused. Domestic legal proceedings were instituted by the companies, the law firm and (later) Mr Jurik, both in relation to the original orders and the refusal to exempt, but these substantially failed. There was also an issue about the lack of independence of one of the judges involved, but we are not now concerned with that.
105. The principal ground of complaint to the European Court was under Article 8, the right to respect for private life. The court ultimately held that there had been an interference with the applicant's (the law firm's) Article 8 right in searching its offices, but not by reason of the search and seizure at the flat. However, the interference constituted by searching the office was justified in the circumstances as in accordance with the law, in pursuance of a legitimate aim and proportionate.
106. There was then a claim that the seizure of the electronic media, and also the damage to the applicant's professional reputation caused by the searches having been carried out, amounted to a violation of Article 1 of protocol 1. The court dismissed both aspects of this latter claim, as these had been taken into account in relation to the Article 8 claim.
107. Thirdly, the applicant made a claim under Article 6(1). As to this, the court relevantly said:
108. On the face of it, there is a conflict between the decisions in Ravon and in Lindstrand. Both cases concerned the lawfulness of searches and seizures undertaken in the context of a tax investigation. In Ravon, the Court explicitly recognised that such a question engaged civil rights and obligations recognised in French domestic law. The judgment in Lindstrand does not identify any such right in domestic law; and proceeds on the basis that the search and seizure powers formed part of the Swedish tax code. The question is complicated by the fact that the applicant was the law firm (where nothing was found) rather than the three companies. If they had been the applicants, perhaps the point would have been more obvious. Whilst Lindstrand could be distinguished on that footing, it seems unlikely that the powers in question did not involve any invasion of rights recognised in the domestic law of Sweden.
109. As I have observed, Ferrazzini is authority for the proposition that a dispute between a taxpayer and a taxing authority as to whether tax is due, and, if so, as to the amount of tax, does not engage Article 6. But it does not, in my view, follow that disputes about the lawfulness of coercive investigative steps, which involve an interference with property rights or rights of privacy recognised in domestic law, are excluded from the scope of Article 6 simply because those steps have been undertaken as part of a tax investigation or pursuant to powers contained in the tax code. I do not consider that the decision of the Grand Chamber in Tetrazzini compels such a conclusion.
110. Nor do I consider that it should be given that effect. To do so would tend to undermine both the protection afforded by the Convention to Article 8 rights (including the right to protection of the home and the right to protection of privacy) and property rights, protected by Article 1 of the First Protocol. Further, given the constitutional principles which underlie the central role of fair trial rights in a democratic society, it seems to me that Ferrazzini should not be applied so as to exclude the Court's supervision of the use by the state of coercive measures which interfere with property and privacy rights.
111. I am fortified in that conclusion by the decision of the Third Section of the Court in Ravon. Although that case involved the search of premises, the principle which was applied by the Court seems to me to be equally applicable to other coercive measures which interfere with rights recognised in domestic law. Whilst I have taken the decision of the same Section (albeit with a different judicial composition) in Lindstrand into account, I do not consider that it provides a reasoned basis for the alternative view and I decline to follow it. Not only was Ravon not cited, but there is a notable lack of explanation in the judgment for the conclusion that a dispute about the lawfulness of coercive measures does not engage Article 6 simply because those measures are undertaken in the context of a tax investigation.
112. The Respondent also relies on the decision of the Tax Chamber of the UK First Tier Tribunal, in Duncan v HMRC [2018] UKFTT 296. This was a taxpayer's appeal (conducted by her accountant) against a notice under the Finance Act 2008, Sch 36, para 1, requiring her to provide certain information within a certain time, under penalty of a fine in case of failure to do so. The taxpayer did not comply, and became liable to the penalty. The taxpayer claimed that the notice should be set aside on various grounds. Some parts of the notice were set aside, but most of was upheld by the tribunal. The penalty was upheld.
113. One of the points raised, but dismissed, related to Article 6 of the Convention. The tribunal put it this way:
114. It will be seen that this case does not address the question currently being considered, which relates to the 'civil rights and obligations' limb of Article 6(1). Instead, it addresses the question of a criminal charge and Article 6(2). I will return to the question of a "criminal charge" later.
115. As for the paragraph cited from R (oao APVCO 19) v HMT [2015] EWCA Civ 648, that does not assist either. That was a case about whether SDLT was payable or not. The facts of that case were far removed from those in this. What the English Court of Appeal did there was to emphasise the conclusion in Ferrazzini that a tax dispute fell outside Article 6(1). But it does not help to tell us what are the limits of a tax dispute.
116. The Respondent also refers to another decision of the English Court of Appeal, R (Derrin Brothers) v HMRC [2016] EWCA Civ 15. This case also involved a Schedule 36 notice addressed to a UK resident, but this time concerning the tax affairs of an Australian resident. The notice had been served by HMRC pursuant to a request from the Australian tax authorities. Sir Terence Etherton C (with whom Davis and Vos LJJ agreed) said:
117. What is clear from paragraph [64] of Derrin is that there was actually no argument as to whether or not there were any 'civil rights and obligations' within Article 6(1) in play. Both before the judge and the Court of Appeal, the case proceeded on the assumption that there were indeed such rights and obligations, but without making any decision to that effect. The courts at first instance and on appeal instead considered the complaints in detail, and concluded that, on the assumption that Article 6(1) was engaged at all, on the facts of the case no violation of it had been established.
118. The Respondent points to the distinction referred to in paragraph [64] "that Ravon concerned the sanctity of the home which is a core value at the heart of Article 8". But, subsequently, the Court of Appeal also referred to "other important differences" with that case. These included "judicial monitoring and judicial review" as well as "the opportunity ... to make representations indirectly or directly to the FTT". Ultimately, basing itself on those differences, the court concluded (at [114]) that the scheme of Schedule 36 combined with judicial review was sufficient to satisfy any Article 6 rights combined with Article 8, and that the scheme in Ravon was distinguishable, which is why t failed.
119. I accordingly do not find the two England and Wales authorities to which we were referred of assistance.
120. It is clear from the decision in Ravon that there can be cases where some civil right or obligation is in dispute over and above any possible tax liability owed to the state, and in such cases Article 6(1) applies, notwithstanding Ferrazzini. For the reasons I have explained, I do not consider that Lindstrand compels a different view. My conclusion in the present case is that the rights and obligations of confidentiality between trustee and beneficiaries, and the Article 8 rights (which it has been accepted were engaged here) constitute civil rights and obligations recognised in the domestic law of Jersey. Thus, if there is (as there was here) a dispute about whether or not a notice lawfully interferes with those rights, Article 6(1) applies, even though the issue arises in the context of a tax investigation. My conclusion in this respect is reinforced by the structure of the legislation, which provides in Jersey for no other method of reviewing the exercise of state power than the time limited right to apply for judicial review pursuant to the Taxation (Exchange of Information with Third Countries) (Jersey) Regulations, as amended.
121. I can take more shortly the criminal limb of Article 6(1). For this to apply, it requires that the court be determining a "criminal charge". In order to make clear the scope of that expression we were referred to the decision of the European Court in Jussila v Finland [2007] 45 EHRR 39. In that case, the relevant Tax Office found deficiencies in the VAT returns of the applicant taxpayer, and "ordered him to pay, inter alia, tax surcharges ... amounting to 10% of the reassessed tax liability (the additional tax surcharges levied on the applicant totalled 1,836 Finnish Marks, corresponding to 308.80 euros)."
122. The applicant appealed to the local administrative court which, however, dismissed his appeal without a hearing, because it said it had all the necessary information on paper and so an oral hearing was "manifestly unnecessary". The applicant was refused permission to appeal to the Supreme Administrative Court. The applicant claimed a violation of Article 6(1), in not holding an oral hearing. The Finnish government argued that this was not a criminal charge and therefore Article 6(1) did not apply.
123. The European Court said:
124. In the present case, it would be a criminal offence in Jersey not to comply with a notice served under the regulations. It appears that penalties could also be imposed in Belgium in certain circumstances. But the present case is not about whether an offence has been committed or a penalty incurred. It is about whether there is a Convention-incompatible restriction imposed on access to a court in challenging the validity of a notice served on the Appellant. No criminal charge is being determined. In the circumstances, I do not need to take this question any further.
125. In Flynn v Reid 2013 (2) JLR 280, the then Deputy Bailiff (William Bailhache) stated (para. 27):
These considerations have, for example, led to the development, in Jersey as elsewhere, of the jurisdiction to make protective costs orders.
126. It is, therefore, unsurprising that costs rules are capable of giving rise to issues under Article 6 of the European Convention on Human Rights. The Appellant relies on two aspects of Article 6: the right of access to a court which implicit in the right to a fair trial; and the principle of equality of arms. I will discuss some of the cases which were placed before us before turning to my own analysis of the issues.
127. The Appellant relied heavily on Zustovic v Croatia (2022) 74 EHRR 3 and it is therefore convenient to start with that case. It concerned Croatian legislation which provided that each party in judicial review proceedings had to bear their own costs. The applicant had been denied a disability pension by the relevant public authorities. She brought judicial review proceedings against those authorities. She was successful, and the decision to deny her pension was quashed. However, the court refused to award her the costs of the proceedings, on the basis of the legislation to which we have referred. The applicant complained to the Constitutional Court, but her complaint was held inadmissible. Subsequently, in other proceedings, the Constitutional Court held that the legislation was unconstitutional, and invalidated the legislation. It concluded (para. [49]) that the relevant provision "did not have a legitimate aim, and was aimed at protecting the financial interests of the State (since it is precisely the State that must bear the costs of the proceedings in the event of losing the case". The applicant made a second complaint to the Constitutional Court, which this time was upheld.
128. In the meantime, the applicant had also applied to the European Court of Human Rights making a complaint concerning access to a court under Article 6 of the Convention. The Court referred to a number of previous decisions, and stated:
129. The case of Zubac v Croatia [2018] ECHR 306, referred to in paragraph [101] of the judgment in Zustovic, was a Grand Chamber decision. It was not a case about costs, but about a value threshold for cases to be taken to the Supreme Court. The Grand Chamber explained the general principles concerning the right of access to a court as follows:
130. It is instructive to read Zustovic along with two other decisions which were placed before us: Černius and Rinkevičius v Lithuania, Applications 73579/17 and 14620/18, 18 June 2020, and Dragan Kovačević v Croatia [2022] ECHR 364.
131. I have already referred to Černius and Rinkevičius in the context of standing. The Court held, for the reasons I have already quoted, that the absence of any right to reimbursement of the costs of legal representation in proceedings challenging the validity of a fine was a restriction on the applicants' right of access to a court. On the question of legitimate aim, the Court stated:
132. The Court went on to hold that "in the particular circumstances of" the case, the refusal to reimburse the applicants' legal costs was disproportionate. It set out a number of considerations in that regard (see [70]-[73]). In particular, it noted that the domestic court had rejected the claim on the basis of its interpretation of the domestic statutory law, rather than by reference to the proportionality of the costs incurred (see [73]). It also referred, in the context of its consideration of proportionality, to "the principle that the risk of any mistake made by the State authority must be borne by the State itself and errors must not be remedied at the expense of the individual concerned" (see [71]).
133. In Dragan Kovačević the applicant had been deprived of his legal capacity by administrative proceedings instituted by the relevant social welfare authority. His appeal to the relevant appellate court was dismissed. But his complaint to the Constitutional Court was upheld, and the decisions of the lower courts were quashed. However, the Constitutional Court refused his claim for reimbursement of his costs, under section 23 of the Croatian Constitutional Court Act, which - by contrast with the legislation at issue in Zustovic - provided that, unless the court decides otherwise, each participant in proceedings before it has to bear its own costs. Thereafter the applicant made the present complaint to the European Court of Human Rights, arguing that this amounted to a violation of his right of access to a court.
134. The court held first of all (at [69]) that a rule that each participant in proceedings has to bear its own costs, unless the court decides otherwise, cannot be regarded as incompatible per se with Article 6(1) of the Convention. This will depend on whether the effects of the application of the rule in question are compatible with Article 6. The imposition of a considerable financial burden after the conclusion of proceedings may constitute a restriction of the right of access to the court (at [70]), as may ex post facto refusal to reimburse successful applicants' own costs in disputes against the State arising from the decisions of public authorities (at [71]). Given that the costs of applying to the constitutional court was more than the average salary in Croatia at the time, the Court ruled that the refusal to award the applicant the costs of his complaint amounted to a restriction on his right of access to the court (at [73]).
135. But a restriction will not violate Article 6(1) if it pursues a legitimate aim and there is a reasonable relationship of proportionality between the means employed and the aim sought to be achieved (at [74]). On the question of legitimate aim, the Court said this:
136. The court nevertheless concluded that the restriction was not proportionate to and justified by the aims pursued in the circumstances of the particular case. This was because (i) the proceedings before the Constitutional Court were of existential importance for the applicant (at [79]), (ii) the costs involved constituted a significant financial burden even for an average citizen, let alone for a person of low income like him (at [80]), (iii) there was no possibility of obtaining legal aid for the proceedings (at [81]), (iv) since the proceedings were unilateral, there was no risk that the relevant public authority, not being a party, would have to pay the costs out of its own budget, and therefore there would not be "a chilling effect on social services in the performance of their duties" (at [82]), and (v) the Constitutional Court did not give any meaningful reasons for its decision to deny him the costs of his constitutional complaint (at [83]).
137. I should say something about the "general principle" relied on in Zustovic (see [100]) "that the risk of any mistake made by the State authority must be borne by the State itself and that errors must not be remedied at the expense of the individuals concerned ... ", and what that means in the present context.
138. The "general principle" appears to find its origins in recent caselaw on Article 1 of Protocol 1. It was applied in that context in Beinarovič and Others v Lithuania, applications no. 70520/10 and two others, 12 June 2018. There the state had reversed the Soviet era nationalisation of property, returning land to the applicants. The state later realised that it had made a mistake, that parts of the land should not have been returned because they were forested, and the law required that they belong to the state on grounds of national interest. The forested parts were retaken by the state, and the deprived owner complained of a breach of Article 1 of Protocol 1.
139. The Court said (emphasis added):
140. This comment was then referred to, in passing, in an Article 6 case, Černius and Rinkevičius, where the court said:
141. This was then picked up and relied upon, as I have noted in Zustovic. Indeed, in its conclusions in that case, the Court stated at [106]:
Unsurprisingly, Advocate Harvey-Hills founded strongly on this passage in support of the proposition that the costs rule in the 2018 Law did not pursue a legitimate aim.
142. Stated as an abstract proposition, this "general principle" might be taken to require that a public authority should always be found liable in costs whenever it is the subject of a successful judicial review. By definition, the public authority would, in such a case, have acted unlawfully and - so it might be said - should therefore always bear the costs of the successful claimant. But in my view, that would not be a correct reflection of what Article 6 requires. In Dragan Kovacevic the Court acknowledged that a rule that each party should bear its own costs, unless the court decides otherwise, is not as such incompatible with Article 6. The question, in any given case, is whether a particular costs rule which restricts access to a court is proportionate to a legitimate aim.
143. The problem with such an approach to the "state should remedy its own mistakes" principle would not only be that it is new. It would also cut across the well-established principles - articulated at Grand Chamber level and repeated in numerous cases - by reference to which the court decides whether or not to strike down laws which restrict access to a court. If a law restricts such access but pursues a legitimate aim and is proportionate to that aim, it does not violate Article 6. But, if the "state should remedy its own mistakes" principle has the effect that the state does not pay litigation costs when the state was originally in the wrong, it still violates Article 6. This would be unsatisfactory, and incompatible with the principles to which I have referred. Were it the law, it would also imply that the UK Supreme Court, in Flynn Pharma, erred in endorsing the possibility that the risk of a "chilling effect" may, in appropriate cases, justify refusing a costs order against an unsuccessful public authority.
144. Nor would it be correct, in my view, to read Zustovic as saying that a rule which insulates a public authority from costs would, as such, never pursue a legitimate aim. As is clear from the extract of its judgment at [131] above, the Strasbourg Court was content in that case to endorse the view of the Croatian Constitutional Court that the law in question in that case did not pursue a legitimate aim.
145. In any event, what appears to have rendered the aim not legitimate in that case was not simply that the rule involved the State in shifting the expense of remedying its own errors to party challenging its mistake but that it did that "with a view to protecting its own financial interests". So stated, the proposition does not exclude the possibility that such a rule could be justified in pursuit of other aims, if proportionate to do so. It is, in any event, apparent from paragraph 69 of the judgment in Černius and Rinkevičius, and paragraph 76 of Dragan Kovačević, which I have quoted above, that a policy limiting reimbursement of litigation costs in administration proceedings might sometimes be justified by public interest financial considerations.
146. The better approach, it seems to me, is to treat the "general principle" stated in Zustovic as a consideration to be taken into account when undertaking the proportionality analysis. I note that it was considered in that context by the Strasbourg Court in Černius and Rinkevičius, para. 71. When assessing whether or not a costs rule which restricts access to the court for the purpose of public law proceedings, is proportionate, it seems to me to be right to start from the proposition that a public authority which has been found to act unlawfully should, unless there is a good reason to the contrary, bear the costs of having that error pointed out and corrected. That consideration - and its virtuous effect in supporting the court's role in supervising the lawfulness of administrative actions - seems to me to be a consideration which should be given appropriate weight in the proportionality analysis.
147. Coventry v United Kingdom, to which I have already referred, shows that the structural effects of a particular costs regime may fall to be assessed for their compatibility with the principle of equality of arms, even if their application has not, in the particular circumstances, been found to be an unjustified restriction on access to the court. That case concerned the former CFA regime in England and Wales. The losing party in a litigation was ordered to pay a very large sum, including success fees and insurance premiums, quite disproportionate to the damages awarded. He challenged this in the domestic courts, but the UK Supreme Court held that the regime was compatible with Article 6: Coventry v Lawrence [2015] 1 WLR 3485.
148. On the subsequent application to the European Court of Human Rights, the Strasbourg court took a different view. It concluded that the regime was incompatible with the principle of equality of arms. It set out the applicable general principles in the following terms:
149. The court declined to consider the scheme as a whole, as the UK government asked, looking at everyone involved. But nor did it focus on the position of the particular applicant himself. Instead it looked at the operation of the scheme from the point of view of the particular class of uninsured defendants: see at [78]. The court relied on four flaws in the CFA scheme which had been identified in Sir Rupert Jackson's report on its operation in January 2010, which had also been relied on by the Strasbourg court in MGN Ltd:
150. The court concluded:
151. The case of Stankiewicz v Poland, to which the Court in Coventry referred, illustrates that a costs rule which privileges a public authority may give rise to a breach of Article 6. The application concerned civil proceedings between a District Prosecutor and the applicant. The applicant was successful and the first instance court ordered the prosecutor to pay the applicant's legal costs, in accordance with the usual Polish rule that the unsuccessful party should bear the costs of litigation. The costs order was ultimately reversed on appeal on the basis of a statutory provision which provided that the participation of the public prosecutor in a civil case would not give rise to reimbursement of litigation costs either to or from the State Treasury.
152. Although the Court acknowledged that the applicant's access to a court was not at issue, it observed at [60] that "there may be situations" in which the issues linked to the determination of litigation costs can be of relevance for the assessment as to whether the proceedings in a civil case seen as a whole have complied with the requirements of the Article 6(1). It observed at [68] that the rule gave the prosecutor a privileged position with respect to the costs of civil proceedings. Whilst such a privilege "may be justified for the protection of the legal order ... it should not be applied so as to put a party to civil proceedings at an undue disadvantage vis-à-vis the prosecuting authorities". In the circumstances of the case, it had that effect and there was accordingly a breach of Article 6.
153. The Law Officers' Department Note in the Appendix to the Projet de Loi which became the 2018 Law did not identify the potential for costs rules to operate as a restriction on the right of access to a court or to infringe the principle of equality of arms. I should address the cases to which that Note did refer. The cases are Masson v The Netherlands (1996) 22 EHRR 491, (Ashendon v United Kingdom [2012] 54 EHRR 13 and R (Henderson) v Secretary of State for Justice [2015] EWHC 130 (Admin).
154. In Masson v The Netherlands, the applicants had been arrested and charged with serious criminal offences. They were kept in pre-trial detention and had to pay legal fees to their lawyers. Ultimately, both of them were acquitted of all the charges. They complained to the European Court of Human Rights that (inter alia) they had been denied compensation for their pre-trial detention and also denied reimbursement of their legal costs. They submitted that this amounted to a violation of their right under Article 6(1), in that their requests for compensation for pre-trial detention and reimbursement of legal costs had not been dealt with in public by an impartial tribunal. It was not argued that requiring them to pay their own (substantial) legal costs amounted to a restriction on their right of access to a tribunal.
155. Much of the argument centred on the question whether there was a "dispute" over a "right" arguably recognised in domestic law. The Convention was directly effective in Dutch law, in priority to domestic law. The court held:
156. Having considered the Dutch domestic law, the court concluded (at [52]):
157. The potentially relevant point in this case concerned the question of whether the Convention gives a right to costs to an accused who has been acquitted of a criminal charge. It says nothing about the potential for costs rules in civil cases to operate as a restriction on access to the court or to engage the principle of equality of arms.
158. In Ashendon v United Kingdom each of the applicants was acquitted of their respective criminal charges, but refused reimbursement of the legal costs of their defences. In England and Wales at the time, the trial judge had power on such an acquittal to make a costs order in favour of the accused, to be paid out of central funds, in such amount as the judge considered reasonably sufficient to compensate the accused for any expenses properly incurred by such accused in the proceedings.
159. The relevant Practice Direction then provided (and the current one in substance still provides) in part as follows:
160. In each case the judge gave reasons, relating to the particular circumstances of the case, for the refusal to exercise that power. The applicants claimed a violation, not of Article 6(1) of the Convention, but of Article 6(2), dealing with the presumption of innocence:
161. The court held that there had been no such violation in either case. It said:
162. The issue in that case was, accordingly, the specific one of whether a decision declining to make a costs order for the reasons given was incompatible with the presumption of innocence. It was not argued that, and the court did not at any time consider whether, the non-exercise of the power to award costs to an acquitted accused amounted to a restriction on the right of access to a court.
163. Lastly, in R (Henderson) v Secretary of State for Justice, the claimant was acquitted on criminal charges after trial at the Crown Court. He had engaged his lawyers privately, mistakenly believing that he was not entitled to (means-tested) legal aid. The claimant applied for a defendant's costs order, but the judge held that he had no jurisdiction to make one, Parliament having legislated in 2012 to limit the power to order payment of such costs. The claimant sought (inter alia) a declaration of incompatibility with Articles 6 and 14 of the Convention.
164. Burnett LJ (as he then was, and with whom Goss J agreed) said:
165. This case, too, despite the width of the language of Burnett LJ (referring to no "violation of article 6"), has nothing to do with Article 6(1) and the question of restriction on access to the court. It is entirely concerned with whether the refusal of reimbursement of legal costs amounted to a denial of the presumption of innocence under article 6(2).
166. I conclude accordingly that none of these three cases, on which the Note in the Appendix to the Projet de Loi depends, is of any relevance when it comes to considering whether Article 6(1) impacts on a refusal to award a successful litigant its costs in public law proceedings on the basis that this is a restriction on access to a court or incompatible with the principle of equality of arms. They are dealing with a different point.
167. At the outset, I observe that Advocate Harvey-Hills' challenge is to the compatibility with Convention rights of the costs rule in the 2018 Law. He does not found on any specific circumstances of this case to argue that a regime, which may in itself be compatible with Article 6, nevertheless results in a breach of Article 6 by reason of its effect in this particular case. The Animal Defenders International case to which I have already referred, shows that a challenge to the compatibility with Convention rights of a general measure is one which may properly be advanced. In that case, the Grand Chamber gave the following guidance:
168. The Attorney General emphasised that the Strasbourg cases in this field are typically highly fact-sensitive. They usually focus not on whether a rule of law is incompatible with Convention rights, but on the impact which that rule has had in the particular circumstances. So, for example, in Dragan Kovačević v Croatia, the Strasbourg Court observed that a rule that each participant in proceedings has to bear its own costs, unless the court decides otherwise, cannot be regarded as incompatible per se with Article 6(1) of the Convention. But as Zustovic v Croatia demonstrates, cases can arise where a costs rule is, as such, incompatible with Convention rights, albeit in that case it was because the rule did not serve a legitimate aim.
169. The rule in the present case is quite different in character from the rule at issue in Dragan Kovačević. Indeed, it is a rule which is different in character from the rule which was under consideration in Zustovic. That rule required each party to bear its own costs, without (by contrast with the rule at issue in Dragan Kovačević) any judicial discretion to modify the rule in particular circumstances. Those two rules treated both parties equally. By contrast, the rule in the present case systematically treats the two parties differently as regards costs. Subject to bad faith, the Respondent would never be liable for a successful applicant's costs; whilst the applicant would always be at risk of being required to meet the Respondent's costs. The costs rule in the 2018 Law does not admit of any judicial discretion to modify that rule in light of the particular circumstances.
170. There was no disagreement that the issues which fall to be addressed in order to decide whether that rule is an unjustified restriction on the right of access to a court are as follows:
(i) Is that rule a restriction on the right of access to the court?
(ii) Does that rule pursue a legitimate aim?
(iii) Is there a reasonable relationship of proportionality between the means employed and the aim sought to be achieved?
171. The Attorney General did not dispute that the costs rule in the 2018 Law is, in principle, a restriction on the right of access to the court. I consider that he was right to take that position. I recognise that the rule may not be much of a restriction for a well-financed corporation that is managing the financial and other affairs of wealthy, international business people and which treats legal costs simply as an expense of its business. But there is nothing in the 2018 Law to confine the costs rule to such parties. It may, for other potential applicants, represent a very significant barrier to access to justice. The argument before us proceeded on the basis that the 2018 Law concerned only judicial review of tax information requests. But that is palpably not the case; and for potential appellants against other administrative directions, for example, of the Jersey Financial Services Commission or the Jersey Competition Regulatory Authority, it may be especially true that the 2018 Law represents a barrier to access to justice.
172. Further, even in relation to tax co-operation cases, any person challenging the issue of a Notice, however well-resourced, will have to take into account the fact that whether successful or not, he will have to bear his own costs of the proceedings, and that he may in addition face an adverse costs order if the proceedings are unsuccessful. That litigant would also have to recognise that the costs which may be incurred might be affected by the conduct of the public authority, that he would have no control over that, and that, short of bad faith, would have no opportunity to seek costs from the public authority.
173. Advocate Harvey-Hills characterised the aim of the costs rule in the 2018 Law as being to insulate public authorities from the costs consequences of their own illegality. He pointed out that the rule in the 2018 Law is generally only relevant where a public authority has been unsuccessful. In the context of a judicial review, that will imply a finding that the public authority has acted unlawfully and that the proceedings were justified. Relying in particular on Zustovic, he contended that this aim was per se illegitimate and that the costs rule in the 2018 Law was accordingly not justified.
174. The Attorney General, on the other hand, argued that the aim of the Law was to protect public authorities in the particular classes of case to which the Law applies from the chilling effect which would flow from a potential liability to costs. He pointed out that the Law protects public authorities in Jersey which are acting in aid of authorities in other jurisdictions and emphasised the importance of Jersey being seen to be a reliable partner in law enforcement activity internationally. He submitted that the targets of international investigations are typically extremely well-resourced individuals or organisations. He pointed to the evidence about the sums at stake in the present case by way of illustration, and, although there was no direct evidence as such before us, also described the costs at stake in the recent Tantular litigation.
175. I need, first, to identify the aim of the costs rule in the 2018 Law. That is a question of fact, though not of the traditional kind, and only limited kinds of evidence are admissible. I must then decide whether that aim is a legitimate one. If the costs rule does not pursue a legitimate aim then, as Advocate Harvey-Hills contended, the costs rule would be an unjustified restriction on the right of access to a court. It is in this part of the case that I understand that not all the members of the court are in substance agreed.
176. The starting point is how one assesses the aim of the legislation. This is a question to be determined on all the admissible materials before us. In my judgment, the approach set out in in Wilson v Secretary of State for Trade and Industry [2004] 1 AC 816 is the approach we should follow. There are several passages of direct relevance in the speech of Lord Nicholls of Birkenhead, as follows. (I have emphasised some words by underlining.)
177. Lord Nicholls of Birkenhead said this:
178. Thus, the primary tool for ascertaining the policy objective of the statute is a construction of the statute itself. Similar comments are to be found in the speeches of the other of their Lordships who sat in that case: see at paragraphs 116-118 (Lord Hope), 140-142 (Lord Hobhouse), 173 (Lord Scott) and 178 (Lord Rodger). This approach is closely linked to a recognition that it is not for the courts to review judicially the proceedings of the States where they are acting legislatively rather than administratively. We should be careful not to take the ministerial or other statements as reflecting the objective intention of the States. What one cannot do is to form a conclusion as to the aim of the legislation on the basis of any extraneous material which is different from the aim or purpose established from its proper construction on usual principles.
179. I therefore take from this summary that the court is to use the construction of the statute as the basis for ascertaining its purpose, but it is permissible when considering the question of compatibility to have regard to external material such as the Report accompanying the proposition and the speeches of members when the legislation was debated.
180. In this case, we have to decide first what the aim of the legislative provision in question is and second whether that is a legitimate aim. The provision has as its obvious purpose a prohibition against a court making a costs order or an order for damages against one of the public authorities listed in the statute. One can probably imply from that provision an intention to protect the States' budget or the budget of the public authority concerned. Indeed, the Attorney General asserted as much in his skeleton argument.
181. The Attorney General, however, wants us to go further and hold that the avoidance of the chilling effect in responding to tax requests from foreign competent authorities was the aim of this legislation - in other words that the social policy underlying the statute was to address a problem which arose because officials in Jersey's competent authority were not minded to give assistance in some cases out of fear of an adverse costs order. But there is nothing in the terms of the statute which suggests that this was its object or purpose, and it would not have been difficult to find some language which made such a purpose clear. It is right to accept that although not an inevitable consequence of the aim set out in [180] above, it is not inconsistent with it. No material has been put before us to indicate that there had been a particular problem in responding to tax information requests which this legislation was intended to address. Indeed, the various cases where a decision has been challenged suggest that the competent authority is perfectly content to give assistance. I also note that the legislation is drafted more widely in its application than merely responding to tax requests. It extends also to the Jersey Financial Services Commission and the Jersey Competition Regulatory Authority.
182. Accordingly, attractively though it was advanced, this part of the argument rests upon the proposition that this is one of those rare cases envisaged by Wilson where it is legitimate to use the travaux préparatoires. Thus, we should have regard to the Report accompanying the Projet de loi and the Hansard report of the debate. I am not sure that it is necessary to do so in this case, because it seems to me that the aim of the legislation is clear enough without any such reference. Nonetheless, assuming it is legitimate to use this material, the Report contains only one sentence which could support the conclusion that the avoidance of the chilling effect was the aim of the legislation when it sets out, as part of the background, that "The provision of such assistance to other jurisdictions should not be constrained by considerations regarding the risk to public funds in Jersey arising from claims for costs, damages or consequential losses."
183. In the Hansard report of the debate, this sentiment was repeated by the Assistant Minister when introducing the legislation who said this:
184. One member out of 49 - the Scrutiny Panel chairman Senator Moore - supported this in these terms:
185. And Deputy Morel referred to the need for:
186. That is all there is. Even if it were appropriate to look at the travaux préparatoires at all, there is not enough here to satisfy me that the purpose of the legislation in preventing the court from making costs orders in these cases was to avoid any possible chilling effect that such orders might have on the performance of the island's international obligations. In my judgment the real purpose was to protect the States' budget. In some circumstances that might be a legitimate aim, but where the result would be an interference with Convention rights, more is needed to point to the particular social or economic problem which explains why the legislation was to be enacted.
187. While there may be something in the argument in connection with orders for damages - which has not been argued out before us and on which I therefore express no opinion - there is no sufficient justification for this aim in this case, even if it were to be accepted, in respect of costs. The States budget on the figures provided to us runs to just under £1 billion per annum, with a margin for a surplus of some £50 million. Orders for costs which have a significant effect on a budget of that size will be extremely rare. In any event, there is a step between the protection of the budget and officials feeling constrained by the potential for costs orders which is not inevitable. We have no real evidence that that was the problem the Law was intended to avoid.
188. It is convenient at this stage to deal with a further argument that the Attorney General put forward, namely that regardless of what the international conventions actually say about the sharing of expenses in exceptional cases, in practice Jersey as a small jurisdiction is unable to recover contributions from requesting states. It is suggested that this emphasises the exposure the island has to large claims for costs and would lead to an unwillingness to provide the mutual assistance the international agreements contemplate. We had no formal evidence of this but, assuming it to be true, the court should not in my judgment take this into account.
189. I note that the European Convention on Mutual Legal Assistance 1959 contained no relevant provision for reimbursement of extraordinary costs. However, the reservation entered by the United Kingdom when extending the 1959 Convention to the island in 2008 was in these terms:
190. Only Italy and Spain have responded to this reservation in any negative sense. The remaining States parties to the Convention (including, incidentally, Belgium) have therefore accepted it.
191. By contrast, Article 26 of the Convention on Mutual Administrative Assistance in Tax Matters, which is the relevant provision of the applicable convention in this case, is in these terms:
192. Without of course deciding the point, it would certainly be at least arguable that any costs award large enough to unbalance Jersey's £1 billion budget would qualify as "extraordinary costs". Probably there would be room for discussion even below that high level.
193. The Court is entitled to assume that requesting authorities will stand by the international agreements into which they have entered. It would not be honourable if they were not to do so. For a court to be constrained in its decision by what was described by the Attorney General as the reality of international politics, which on this account would be a distortion of the position under international law, would be wrong in principle.
194. Although I accordingly proceed on the basis that the aim of the costs rule in the 2018 Law is not a legitimate one, I have also concluded that it is not proportionate either, and, accordingly, that it is not compatible with Article 6 of the Convention.
195. I approach the question of proportionality in this case by applying the guidance set out in Animal Defenders International, which was quoted above. I recognise that we are called upon to assess a legislative choice which has been made by the States. I also recognise that the question is not whether a less restrictive rule could have been adopted, or whether it has been proved that the aim could not be achieved without legislating in the terms set out in the 2018 Law. I acknowledge that the "core issue" may be said to be whether, in adopting the general measure and striking the balance it did, the States acted within the margin of appreciation afforded to it.
196. In Animal Defenders International, the European Court observed that the quality of the parliamentary and judicial review of the necessity of the measure is of particular importance when assessing the legislative choice which has been made, in particular as regards the margin of appreciation, and I accordingly need to say something about the consideration of the issues by the States.
197. At the outset, I acknowledge that the written record of legislative deliberation is not a complete record of the consideration given to a measure. In the present case, I can see that the 2018 Law was considered, for example, by the Corporate Services Scrutiny Panel. We know that most of the members of the States Assembly did not speak in the debate, and we cannot make assumptions about the considerations which may have motivated them to support the Law.
198. Nevertheless, when I consider the travaux préparatoires, there is a striking absence of any recognition of the potential for costs rules to affect access to justice and the equality of arms. This was, in relation to the proposed costs rule, one might have thought, the key consideration which the legislature required to have in mind when deciding how to strike the balance between the legitimate aims which it was seeking to advance and interests and rights protected by the Convention.
199. On the contrary, the Note from the Law Officers Department stated that "there are no human rights issues. The Convention leaves the question of when costs should be obtained in litigation ... as matters for national law". As the cases to which I have referred demonstrate, that was incorrect. Although the Note from the Law Officers Department did state that: "The question of whether liability for costs or damages will tend to restrain abuse of state power, or will instead provide an obstacle to the beneficial use of important regulatory powers, is not a matter on which the Convention has an opinion" it did not frame the issue in relation to costs by reference to the impact which costs rules may have on fair trial rights.
200. This error as to the legal framework would not matter if the travaux préparatoires showed that the States had in fact addressed the issues which I have identified. But that is not the case. There is no evidence that the legislature was mindful of the potential for the incidence of costs to restrict access to justice or to affect the equality of arms between participants in a litigation. There is, moreover, evidence in the speech of the Assistant Chief Minister, when introducing the Law, that the States were proceeding on the view that similar protections are "common in other jurisdictions". In the proceedings before us, the Attorney General was unable to assist us as to the basis for that statement and did not point us to analogous legislation in any other jurisdiction. Indeed, as I have already noted, the costs rule in the 2018 Law is quite different in character from rules which have been considered by the Strasbourg Court.
201. In these circumstances, the question which I must address is whether, on the assumption that, contrary to my conclusions earlier, the aim of the legislation was legitimate, the legislative choice made by the States can be supported as a proportionate measure, on the basis of the material before us. For the following reasons, I do not consider that it can. In addressing the issues, I keep firmly in mind the purpose which I have assumed the legislation is intended to advance.
202. At the heart of the Attorney General's submissions before us was the proposition that those affected by the costs rule in the 2018 Law are generally very wealthy, and unlikely to be deterred from litigating by that rule. Although he did not refer us to Flynn Pharma, his submissions were to the effect that the cases to which the 2018 Law would apply were typically ones where (to quote Lady Rose at para. 149 of that case) where there was a "great disparity in resources between the appellants and the [public] authority and ... the costs though high were small relative to the total turnover" of the litigant.
203. I readily accept, from my own experience, that the cases to which the 2018 Law applies may often involve very well-resourced litigants; and the present case would, indeed, appear to be illustrative of the point. But the Attorney General did not place before us any statistical evidence that it is necessarily or always the position that the 2018 Law could only be relevant to litigants of this character. Indeed, anticipating that the Court would be concerned about the potential for the costs rule to operate unfairly in the case of less well-resourced litigants, he placed before us materials about the availability in Jersey of legal aid.
204. There is, though, a wide gulf between cases in which a litigant is sufficiently impecunious to qualify for legal aid and the class of litigant which was the focus of the Attorney General's submissions. Anyone with experience of litigation will appreciate the impact of costs on litigants even of relatively substantial means, including individuals and corporations who could well, on the face of it, be the subject of requests under the various measures listed in Schedule 1 to the 2018 Law. Accordingly, I do not consider that the potential availability of legal aid addresses the potential effect of the costs rule for a foreseeable class of litigants - namely, that very large class for whom the costs rules may have a significant impact but who would not qualify for legal aid.
205. In any event, I am not satisfied that legal aid would be generally available to bring judicial review applications to which the 2018 Law would apply. Whilst the scheme makes it possible to obtain legal aid for some judicial review claims, it does not do so in respect of "corporate or business activity".
206. The difficulty with the costs rule in the 2018 Law is that, like the rule under consideration in Zustovic, it is a blanket rule. As I have noted, it is liable to have more extreme effects than the rule in that case - which the Strasbourg Court held was incompatible with Article 6 - because it not only insulates the public authority from liability for the costs of the private litigant, but imposes on the private litigant the risk of a costs award in favour of the public authority. By contrast with the rule under consideration in Dragan Kovačević, it does not contain any scope for judicial modification in light of the circumstances of the particular case - a qualification which, in that case, the Strasbourg Court considered provided "a necessary flexibility" (see [78]).
207. For a foreseeable class of litigants (those whom I have described above), the rule is liable to be a significant deterrent to the bringing of potentially well-founded applications for judicial review. The costs rule applies, of course, specifically where a public authority has, in fact, been found to have acted unlawfully. Judicial review is a key safeguard for the rule of law. Public authorities are protected from unmeritorious judicial reviews by the requirement to obtain leave. Although I do not consider that the "general principle" that public authorities should bear the cost of putting right their mistakes requires that public authorities should necessarily always be found liable in costs even of meritorious claims, the fact that the costs rule in the 2018 Law excludes the possibility of a public authority of ever bearing those costs is, as I have explained, a factor which weighs against that rule being regarded as proportionate.
208. Further, the rule applies regardless of the way that the public authority has conducted itself in relation to the litigation - provided that the public authority does not act in bad faith. The potential to incur a liability for costs imposes a discipline on litigants. Whilst I acknowledge that, as the Attorney General submitted, public authorities may be expected to be mindful of their own costs when considering how to conduct a litigation, the fact remains that, in principle, the costs rule in the 2018 Law would apply even if the public authority has acted unreasonably in the way that it has litigated, provided only that it does not act in bad faith. I observe that in AG v Rosenlund, the Commissioner found that the Attorney General's conduct of the litigation in that case would, if he had been a private litigant, would have justified an award against him of indemnity costs.
209. The Attorney General pointed to features of the 2018 Law which, he submitted, mitigated the blanket impact of the costs rule - specifically: (i) the exception in Article 2(3) for bad faith; (ii) the power of the Minister in Article 2(4) to exclude "any type of damages, costs or consequential loss in respect of assistance in any legal proceedings" from the application of the Law; and (iii) the provision in Article 2(2) that the Law does not prevent an award of damages in respect of an act which is unlawful by reason of Article 7 of the Human Rights (Jersey) Law 2000.
210. I do not consider that these address the fundamental problems with the costs rule which I have identified. Those problems may arise in cases where the public authority is acting entirely in good faith. I doubt if the Minister's power to exclude a "type of ... costs" could be used to qualify the costs rule in a way which would mitigate its blanket nature appropriately. In any event, that power has not in fact been exercised. And Article 2(2) is concerned with the damages rule in the 2018 Law, not the costs rule. Whilst it might be said that a litigant whose Convention rights are infringed by the application of the costs rule could bring a claim for damages, I do not consider that this would answer the reasons why I have found the rule to be disproportionate. In particular, it does not address the deterrent effect which the costs rule may have on well-founded challenges to unlawful action.
211. The Attorney General did not shrink, in his submissions, from asserting that the risk of liability in costs could have a chilling effect on the way in which public authorities will deal with requests from their international counterparts. By its nature, of course, that potential chilling effect is difficult to demonstrate. The risk of a chilling effect does, though, it seems to us, require to be put in its proper context. First, as I have observed, public authorities can and should generally be expected to exercise their powers robustly. Secondly, the 2018 Law applies in cases where the public authority is acting pursuant to Jersey's treaty obligations where they may be expected to seek, so far as lawful, to assist their international counterparts. Thirdly, at least in the case of tax information exchange treaties, there is provision for extraordinary costs to be borne by the applicant State. Fourthly, an applicant for judicial review requires to obtain leave. The chilling effect could not reasonably come into play at the point when a decision is made to give assistance because no significant exposure to costs would arise at that stage and a review of the risks of litigation, if leave were given, could lead to the withdrawal of the Notice without any real expense.
212. The Attorney General's submissions as regards a chilling effect were, as I have observed, focused on the exceptionally well-resourced litigant, who might be expected to take every point, and for whom the costs involved in litigation would be insignificant relative to his resources. The existence of such litigants does not justify a blanket rule such as that in the 2018 Law. As I have explained, such factors could, in particular cases, be relevant to the exercise of the Court's costs jurisdiction under the general law of Jersey. Indeed, as the UK Supreme Court's judgment in Flynn Pharma acknowledges (at [149]), such considerations might, in appropriate cases, justify a starting point, in considering costs, of no costs award against the public authority, subject to the judicial power to modify the position in light of the circumstances of the case. The fact that the particular case which the Attorney General identified as giving rise to the risk of a chilling effect can be accommodated within more nuanced rules is itself a reason for holding that the blanket rule in the 2018 Law is disproportionate.
213. In any event, I consider that the costs rule in the 2018 Law is incompatible with the principle of equality of arms. That principle requires a 'fair balance' between the parties in which each party is "afforded a reasonable opportunity to present his case under conditions that do not place him at a substantial disadvantage vis-à-vis his opponent or opponents".
214. I readily acknowledge that the equality of arms principle is not absolute. The cases on civil legal aid, to which this Court referred in Useni v Attorney General [2022] JCA 197, paragraphs [69]-[74], illustrate the point. I also recognise that the costs rule in the 2018 Law may not place the particular type of litigant on whom the Attorney General focused at a substantial disadvantage.
215. Nevertheless, the costs rule in the 2018 Law structurally and systematically differentiates between the two parties. Given the significance of costs and the risk of costs liabilities for litigants - to which the Deputy Bailiff referred in Flynn v Reid - it seems to us that there is a readily foreseeable, and potentially significant, class of litigants for whom that structural and systematic difference does indeed place the claimant at a substantial disadvantage vis-à-vis the public authority defendant. I accordingly conclude that the costs rule in the 2018 Law is incompatible with Article 6 on that ground also.
216. In his written submissions, Advocate Harvey-Hills noted that in Zustovic, the Strasbourg Court also found that there had been a breach of Article 1 of the First Protocol and contended that if we were to conclude that Article 6 is not engaged, we should address the costs rule in the 2018 Law under reference to Article 1 of the First Protocol. Since we have concluded that the costs rule is incompatible with Article 6, we do not require to consider Article 1 of the First Protocol.
217. As I have explained, Article 4 of the Human Rights (Jersey) Law 2000 requires us, so far as it is possible to do so, to read and give effect to the 2018 Law in a way which is compatible with Convention rights.
218. Advocate Harvey-Hills suggested three possible ways in which the 2018 Law could be "read down" so as to be compatible with Article 6.
(i) He invited us to "read in" an exception such that any act of a public authority under the 2008 Regulations or the 2014 Regulations did not attract the protection of the Law. He relied in that regard on the apparent statement of the Assistant Chief Minister that the 2018 Law would "not" provide protection in tax information exchange cases.
(ii) He invited us to read Article 2(1)(c) of the 2018 Law as limited to costs in legal proceedings concerned with damages or consequential loss as referred to in Article 2(1)(a) and (b).
(iii) He invited us to interpret the exception for "bad faith" so as to apply whenever the public authority had acted "in error".
219. The Attorney General, for his part, accepted that if the costs rule was not compatible with Article 6, it could not be "read down".
220. I agree with the Attorney General. None of the proposed readings offered by Advocate Harvey-Hills is persuasive.
(i) I have accepted the Attorney General's submission that the word "not" in the Hansard report of the Assistant Chief Minister's speech is an error. The 2018 Law clearly applies to tax information exchange cases.
(ii) This suggestion is not, in my view, consistent with a "fundamental feature" of the 2018 Law. That Law was clearly intended to insulate public authorities not only from liability in damages and for consequential loss, but from costs risks. That intention is evident not only on the face of the Law but in the travaux préparatoires.
(iii) I do not consider that "bad faith" can reasonably be given the expansive meaning suggested by Advocate Harvey-Hills. In any event, this proposed reading would, likewise, not be compatible with a "fundamental feature" of the 2018 Law, which was to insulate the public authority from the risk of costs liabilities even when it had acted unlawfully.
221. For all these reasons, I have concluded that the costs rule in the 2018 Law is incompatible with Article 6. Nevertheless, it remains the law of Jersey. Article 4(2) of the Human Rights (Jersey) Law makes clear that our interpretive obligation does not affect the validity, operation and enforcement of any incompatible principal legislation. Article 7(6) further makes clear that we would not be acting unlawfully in giving effect to the Law in accordance with its terms. We are obliged to do so and we shall accordingly make no costs order.
222. The only question which remains is whether we should exercise our power to make a declaration of incompatibility under Article 5 of the Human Rights (Jersey) Law 2000. Having concluded that the Law is incompatible with Convention rights, it seems to me to be appropriate to reflect that conclusion in a formal declaration. Whilst the declaration has no effect on the validity and continuing effectiveness of the Law, it is right that this Court's view as to the compatibility of the Law with Convention rights should be authoritatively stated in order that the legislature can consider if it thinks fit whether the 2018 Law should be amended or repealed. This is consistent with the comments of Baroness Hale of Richmond in Barclay (No 2) [2015] AC 276 at [31] where she said: " A declaration of incompatibility ... leaves the incompatible law intact although it sends a clear message to the legislature that the state will be in breach of its international obligations unless and until it is put right ...".
223. Subject to the following, I agree with the judgment of Matthews JA and I would like to add only these brief comments.
224. First, while I agree that the chilling effect is a factor which is capable of properly being taken into account by a court when exercising its discretion as to costs, it is crucially important, as the judgments cited by Matthews JA indicate, that close regard is paid to the nature of the decisions of the public authority which has come under the court's review. In tax information cases of the present kind, there is a low bar which needs to be hurdled before a request for assistance can properly be granted as this court has indicated previously. It follows that this type of case is not one where the competent authority will regularly set off, further to its statutory obligations, without any firm foundation as to its course. That makes it all the more questionable whether the Bradford line of authority should be applied and more likely that these cases fall within the R(M) and Flynn Pharma exceptions. Furthermore, there are generally other factors than the chilling effect to go into the balance for the court's consideration. One of the other factors is a direct counterpoint, namely that the decision not to award costs against the public authority which ex hypothesi has acted wrongly could be seen to reward the authority for unlawful, incompetent or oppressive behaviour in some cases; and to provide legislatively that such costs can never be awarded can be seen as a disincentive to good decision making because the knowledge that costs cannot be awarded could easily lead to the authority making no proper enquiry of the requesting state, taking a sloppy decision in the first place, and, worse, then defending it to the end without any justification for doing so. This possibility should have been considered as part of the debate, and thus could go to the question of proportionality; but if it had been so considered, then in my view it would have inexorably led to a serious question as to whether, if this were the aim of the legislation, there was a sufficiently rational connection between the legislative provision and that aim so as to make it legitimate in the first place.
225. Secondly, I think it important to add a qualification to the approach set out by Matthews JA at [195] of his judgment. I understand it to be based on the Animal Defenders case at [110].
226. In my judgment, the Grand Chamber was there distinguishing the narrower submission of the applicant from the true approach (the core issue) which was whether overall the legislature had acted within the margin of appreciation. However, to adopt the broader approach does not mean that the narrower questions are irrelevant or in some cases cannot be considered. If the position were that the legislature could have achieved a legitimate aim by a less intrusive breach of the Convention, that can be properly taken into account by the court in concluding that the legislation was disproportionate to the problem. Indeed, that seems to me to be the common sense way to look at the issue. If that is right, it would follow that it is legitimate for a court, when considering the proportionality of legislation which breaches a Convention right, to have regard to the other options which were available to the legislature.
227. To the extent that it is said the court cannot consider the other options for achieving the intended aim, I would therefore not agree. That is material to the current debate. If the aim were to have been the avoidance of the chilling effect in order that there was no defensive approach taken to requests for international assistance in tax matters, there were a number of non-legislative ways of addressing it. They would include a Ministerial direction to Revenue Jersey to ensure that there was a policy requirement that the request of a foreign authority should be given effect unless it were clearly unlawful to do so. Such a direction would give comfort to officials who would realise that in it there was political cover for their well-meaning but unlawful implementation of the request. At the same time, no question of breaching the Convention would arise because neither the legislature nor the Minister had taken any step in that jurisdiction. The Ministerial direction could ensure that there was a full review of unlawfulness if the court were to grant leave for judicial review, thus exonerating the officials from criticism for their action in implementing the request up to that date.
228. These and other legislative possibilities - there are many possibilities but one example might be the imposition of a requirement on the court when considering the question of costs to have regard to the extent to which the competent authority has reasonably relied upon what may have turned out be false assurances from a requesting state - fall within an assessment in my judgment of whether the legislature had "acted within the margin of appreciation afforded to it". In this case, there is no evidence that any alternative ways of tackling the issue of the chilling effect, if contrary to my view that had been the intended aim of the legislation, were considered; and that is relevant to the question whether the legislature acted within its margin of appreciation.
229. I agree with the judgment which Matthews JA has handed down, with the exception of paragraphs 173 to 187, which deal with the question of "legitimate aim". The President and I have contributed to the text of Matthews JA's judgment, and it may, with the exception of those paragraphs, be regarded as the judgment of the Court as a whole. I write this separate judgment to explain why, for my own part, I consider that the costs rule in the 2018 Law pursues a legitimate aim. Since I disagree with the President and Matthews JA on that issue, I will also explain, in supplement to the observations at paragraphs 194 to 216 of Matthews JA's judgment (with which I otherwise agree), why I have concluded that the costs rule is nevertheless incompatible with Article 6 of the European Convention on Human Rights.
230. As Matthews JA's judgment explains, we consider that the costs rule in the 2018 Law can, in principle, operate as a restriction on the right of access to the court. In order to determine whether that rule is compatible with Article 6 of the European Convention on Human Rights, we must accordingly address whether it pursues a legitimate aim and, if it does so, whether it is justified by reference to that aim. The first of these issues requires us to address two questions: (i) what is the aim of the measure? and (ii) is that aim a legitimate one? The second requires us to address two further questions: (i) is there a rational connection between the means employed and the aim sought to be achieved? and (ii) are the means employed proportionate to the legitimate aim pursued? In this context, this last question can be addressed by asking whether, allowing such margin of appreciation as is appropriate in the circumstances, the costs rule strikes a fair balance between the general interest (as reflected in the policy objective) and the protection of fundamental rights.
231. The aim of the costs rule was a central area of dispute in the debate before us. Advocate Harvey-Hills characterised its aim as being to protect the public purse and contended that this is not a legitimate aim. In paragraph 45 of his written submission, the Attorney General stated that: "For the purposes of this case, the aim of the 2018 Law is to protect public authorities from liability for damages, costs and consequential loss where they provide assistance pursuant to a request made by a relevant authority or country outside Jersey". But in his oral submission, he advanced squarely the contention, under reference to the legislative materials to which I refer below, that the aim was to minimise the risk of a "chilling effect" and thereby to promote sound decision-making in foreign cases and to protect Jersey's reputation internationally.
232. The 2018 Law provides the relevant Jersey public authorities, in cases to which the Law applies, with an immunity from liability to damages, costs and consequential losses. At one level, that can be said to be the purpose of the legislation. But as Lord Nicholls of Birkenhead explained in his speech in Wilson v First County Trust [2004] 1 AC 816 (at paragraph [61]):
233. The critical question is not what the measure does (and may be taken to have been intended to do), but why the States Assembly enacted it. In the present case, that is not evident from the Law itself, but is explained in the Projet de Loi in the following terms:
It seems to me that this statement is the best evidence of the social policy which the 2018 Law was intended to advance. It appears in a document which, on any view, is one to which the court may properly look should it need to identify the mischief to which the Law was directed, in the context of statutory interpretation or otherwise.
234. In presenting the Law to the States Assembly, the Assistant Chief Minister used very similar language:
This statement by the responsible Minister confirms that the purpose in bringing forward the 2018 Law was indeed that set out in the paragraph from the Projet de Loi to which I have referred.
235. I take it from these materials that the over-arching policy objective which the 2018 Law was intended to advance was the proper fulfilment of Jersey's international obligations, and the consequent protection of Jersey's reputation as a reliable law enforcement partner. I also take it, both from the Projet de Loi and from the Assistant Chief Minister's speech, that the basis upon which it was considered that an immunity from claims for costs, damages or consequential losses would advance that over-arching objective was because it was thought that the risk to public funds which such claims presented could "constrain" the provision of assistance to other jurisdictions.
236. Although the Projet de Loi does not use the words "chilling effect", it is, in my view, apparent from these statements of purpose that the concern, or mischief, which the immunity was intended to address was, as the Attorney General submitted in oral argument, the possibility that the risk to public funds presented by claims for costs, damages and consequential losses would lead the relevant public authorities to be unduly risk averse or defensive when exercising their functions in the context of international co-operation.
237. That understanding of the social policy which the measure was intended to promote is further supported by the observation of Senator Moore that:
238. I recognise the caution which the court must exercise in taking, as an indication of legislative purpose, an individual comment by a single legislator during a debate in which most of the members of the Assembly did not speak. But this statement was made by a member apparently speaking for the Corporate Services Scrutiny Panel, which had scrutinised the Bill and is consistent with the aim articulated in the Projet de Loi and by the Assistant Chief Minister. It accordingly provides some additional support for the conclusion which I have reached about the social purpose of the 2018 Law. These various explanations of the purpose of the 2018 Law draw no distinction between liability to costs, damages and consequential losses. There are important distinctions between these different types of liability, not the least of them the potential significance of costs rules in the context of Article 6. The failure, during the legislative consideration of the 2018 Law, to focus on those differences - and, in particular, to address the impact which costs rules may have for Article 6 rights - bears (through its effect on the margin of appreciation) on our assessment of proportionality. But at this point in the analysis, the question is simply what social policy the costs rule was intended to advance, and neither the travaux nor the terms of the 2018 Law itself provide a basis for considering that, so far as the policy objective was concerned, any distinction falls to be drawn between the costs rule and the other immunities granted by the 2018 Law.
239. Was this aim a legitimate one? There can be no doubt that the over-arching aim - of promoting the proper fulfilment of Jersey's international obligations, and thereby maintaining its reputation as a reliable law enforcement partner - is legitimate. Indeed, it may properly be regarded as a policy objective of very high importance, both generally, through its support for the international rule of law, and specifically for this jurisdiction. It follows, in my view, that the objective of supporting sound decision-making on the part of Jersey public authorities engaged in international co-operation by excluding the risk of a "chilling effect" is a legitimate policy objective. At this stage, of course, I proceed on the assumption that there is a rational connection between the means employed and the stated policy objective. The question of whether that assumption is well-founded falls to be addressed at the next stage of the analysis.
240. I take support for the view that, in the context of costs, excluding a "chilling effect" on the exercise of public functions is a legitimate policy objective from the decisions discussed in the first part of Matthews JA's judgment. Thus, the courts of England & Wales have, following Lord Bingham CJ's analysis in Bradford Metropolitan District Council v Booth (2000) 164 JP 485, acknowledged that it is a relevant consideration, for the purposes of costs awards under section 64 of the Magistrates Courts Act 1980, to take into account:
241. In Competition and Markets Authority v Flynn Pharma Ltd [2022] 1 WLR 2972 Lady Rose, giving the judgment of the UK Supreme Court, explained (paragraph [97]) that:
Indeed, she acknowledged that the Competition Appeal Tribunal could legitimately conclude, in certain types of case, that the potential chilling effect would justify taking as a starting point that no order as to costs should be made against the Competition and Markets Authority.
242. These decisions proceed on the footing, first, that the risk of exposure to costs can, in certain types of case, have a chilling effect on the proper exercise of public functions (including on the willingness of public authorities to stand by and defend honest, reasonable and apparently sound decisions) and, secondly, that where this is so, avoiding such a chilling effect is a relevant consideration when the court makes decisions about costs - in other words that avoiding such a chilling effect is a legitimate aim which may, in appropriate cases, properly modify, in favour of a public authority, the approach which would otherwise be taken to awards of costs.
243. We were not shown any Strasbourg case in which avoiding a "chilling effect" has been specifically identified as the legitimate aim of a costs rule. But in Dragan Kovačevic v Croatia, Application no. 49281/15, 12 May 2022, the European Court of Human Rights, in considering a qualified general rule that parties before the Croatian Constitutional Court bear their own expenses, said this (paragraph [82]):
244. In that case, the legitimate aims advanced for the costs rule were (paragraph [77]): (i) to secure the smooth functioning of the court; and (ii) to protect the state budget. The observations at paragraph [82] of the Court's judgment which I have quoted are made in the context of the Court's discussion of proportionality. Nevertheless, in those remarks, the Court acknowledged that the exposure of public authorities to litigation costs may be capable of having a "chilling effect". That being so, I see no reason why avoiding or minimising such an effect should not be a legitimate aim of a costs rule or a costs decision for the purposes of determining whether the rule or decision is compatible with Convention rights.
245. I have accordingly concluded that the costs rule in the 2018 Law pursues a legitimate aim and turn now to consider whether the costs rule is justified by reference to that aim. The aim was expressed compendiously in relation to liability to costs, damages and consequential losses, but it is the costs rule, and specifically the effect of the costs rule on Article 6 rights (ie the right of access to the court and the equality of arms), which must be justified.
246. As I have already explained, the first question which requires to be addressed at this stage is whether there is a rational connection between the costs rule and the policy objective. If there is no rational connection between them, the costs rule cannot be justified by reference to that objective. In his submission to us, the Attorney General explained that Jersey's international co-operation work often involves investigation into very well-resourced parties, who are willing and able to pursue aggressively litigation which will impose significant costs on the relevant Jersey public authorities. He instanced the recent Tantular litigation where, he advised us, the Attorney General's costs exceeded £900,000 and he was required also to meet some £160,000 of Tantular's costs. He contended that the risk of such litigation, and exposure to the costs associated with litigation of that sort (including the potential costs liability to the other party) may have a chilling effect on the relevant public authorities.
247. I would be prepared to accept, at least at the level of generality, that in relation to such cases, there is a rational connection between the costs rule in the 2018 Law and the stated objective. In Lady Rose's judgment in Flynn Pharma, she observed (paragraph 149) that, among the considerations which the Competition Appeal Tribunal may take into account, when considering whether to depart from the usual rule that costs follow success, is "the great disparity in resources between the appellants and the competition authority and the fact that the costs though high were small relative to the total turnover of the appellants". There are a number of reasons why this should be so. Among them is the consideration that for such a litigant, costs and the exposure to costs liabilities may not have the same disciplining effect which they have for most litigants. There is accordingly a rational basis for apprehending that such litigants may be particularly ready to bring proceedings, and then to pursue those proceedings vigorously, taking every point which can be taken.
248. Of course, the public authority should be prepared, where appropriate, to defend proceedings at the instance of such litigants robustly and will have the benefit of costs awards in its favour (including in appropriate cases indemnity costs awards) if the proceedings are ill-founded or are pursued unreasonably. I agree with the observation in paragraph 36 of Matthews JA's judgment that the proper starting point should be a working assumption and expectation that public authorities will exercise their public functions robustly and properly, will defend decisions which should be defended and will concede decisions which should be conceded. But as Matthews JA observes (at paragraphs 37 and 209), the potential to incur a liability in costs can have a salutary discipline on litigants, including on public authority litigants - in other words, the risk of costs liabilities can and should influence their decisions. Judges of the greatest experience have acknowledged the risk that routine exposure to costs could have a chilling effect on the behaviour of public bodies. It would accordingly, in my view, be unrealistic to say that, in the case of litigants of the sort described by the Attorney General, there is no rational connection between the costs rule and its stated aim.
249. Although I accept that there is a rational connection between the costs rule in the 2018 Law and the articulated policy objective, I consider that the rule is nevertheless, for the following reasons, plainly disproportionate and is, accordingly, incompatible with Article 6.
250. First, the costs rule is overbroad. As Matthews JA has explained, the costs rule in the 2018 Law is quite capable of applying to litigants who are not of the character described by the Attorney General. Indeed, the Attorney General made a positive submission to us that legal aid would be a "safety net" which would catch what he characterised as any "unusual cases". As Matthews JA observes at paragraph 205 of his judgment, the material placed before us about Jersey's legal aid arrangements does not suggest that legal aid would indeed respond in all such cases. But more pertinently, there is a significant class of litigant which would not qualify for legal aid but for whom litigation costs are likely to be a real and significant consideration in deciding whether or not to challenge the decision of a public authority. For those litigants, the costs rule may be a real deterrent to bringing proceedings. In my judgment, this is not a case of a generally justifiable general measure which gives rise to "individual hard cases", such as the Grand Chamber envisaged in Animal Defenders International v United Kingdom (2013) 57 EHRR 21, paragraph [106]. Rather, it is a case of a measure which is materially over-broad relative to the objective which is said to justify it.
251. Second, the rule structurally and systematically differentiates between the two parties to the litigation. The public authority is always (subject only to the qualification for bad faith) released from the discipline which the risk of liability for costs imposes. Indeed, the rule protects not only the public authority which has acted reasonably in the defence of proceedings, even if the public authority has ultimately been unsuccessful; it also insulates from any liability for costs the public authority which has acted unreasonably or incompetently, provided only that it is not in bad faith. On the other hand, the party bringing the challenge, however reasonably and justifiably, always remains subject to costs risks. Other than in relation to litigants for whom costs risks are immaterial, that is a systemic difference which is liable to place the private litigant at an unfair disadvantage relative to the public authority. Matthews JA's judgment makes this point in relation to the equality of arms principle. In my view, it is equally relevant in assessing whether the rule is a proportionate response to the policy objective. On that ground alone, I would hold that the costs rule does not strike a fair balance between the general interest and the requirement to protect fundamental rights.
252. Third, that structural imbalance necessarily (and indeed deliberately) benefits a public authority which has acted unlawfully. It will be apparent from what I have said above that I agree with Matthews JA that the "principle that the risk of any mistake made by the State authority must be borne by the State itself" referred to by the European Court of Human Rights in Zustovic v Croatia (2022) 74 EHRR 3 does not mean that a public authority must always be found liable for the costs of a successful challenge to its actions. Such a conclusion would not be consistent with the approach to costs explained in Booth and Flynn Pharma. But I consider that the effect of the costs rule in insulating (save only in cases of bad faith) a public authority which has acted unlawfully from ever bearing the costs of a necessary challenge, and the fact that this may operate as a practical deterrent to wronged parties bringing proceedings to challenge unlawful state action, is a significant consideration in support of the conclusion that the rule is disproportionate.
253. Fourth, the costs rule in the 2018 Law is a blunt rule which does not admit of any qualification (other than bad faith), regardless of the circumstances. The general rule in Dragan Kovačević which required each party to proceedings before the Croatian Constitutional Court to bear its own expenses was qualified by the Court's power to decide otherwise. The Court observed (at paragraph [78]) that this introduced "a necessary flexibility allowing the Constitutional Court to adapt its decisions on costs to the circumstances of each case". Whilst it found a breach of Article 6 "in the specific circumstances" (see paragraph [84]), the Court did not suggest that the law itself was objectionable. Any experienced advocate or judge knows just how various can be the considerations which bear on costs, and the potential for an unqualified and inflexible rule to operate unjustly in particular circumstances. The blunt nature of the costs rule reinforces the conclusion that it is disproportionate.
254. Fifth, the aim to which the costs rule is directed (in my view, legitimately) would be capable of being addressed in other ways. Where there is a basis for apprehending a "chilling effect", that is, as Matthews JA explains in the first part of his judgment, a consideration which may be taken into account under the general law of costs. Indeed, as Lady Rose explained in Flynn Pharma, it is a consideration which may sometimes justify taking as a starting point a presumption against making a costs award against the public authority. On the face of it, then, there are ways in which the legitimate aim which the costs rule is said to pursue could be advanced, whilst retaining the flexibility necessary to respond fairly to the circumstances of particular cases. In Animal Defenders International, the Grand Chamber observed (paragraph [110]) that when the court is considering whether a general legislative measure is incompatible with Convention rights, the question of whether less restrictive rules could have been adopted is not the "central question". Nevertheless, I agree with the President's observation at paragraph 226 that the availability of alternative, less restrictive, measures - which would also introduce the flexibility to respond to particular circumstances - is a relevant factor, even if not a decisive one, which may bear on whether the provision under consideration has struck a fair balance between the general interest and the protection of fundamental rights.
255. The Human Rights (Jersey) Law 2000 places a responsibility on the Court, where an issue is properly raised, to assess whether or not legislation enacted by the States is compatible with Convention rights. That is a responsibility placed on us by the law, enacted by the States of Jersey. In Animal Defenders International, the Grand Chamber stated (paragraph [110]) that when considering the compatibility with Convention rights of a legislative provision:
For our purposes, the question is, as the UK Supreme Court explained in R (Elan-Cane) v Home Secretary [2023] AC 559, whether the costs rule in the 2018 Law is within the margin of appreciation which the Strasbourg Court would allow. For the reasons which I have set out above, as well as those given by Matthews JA at paragraphs 194 to 215 of his judgment, I do not consider that it is. I agree with Matthews JA that the 2018 Law cannot be "read down", pursuant to our obligation under Article 4 of the 2000 Law, so as to be compatible with Article 6. Although that conclusion does not affect the validity of the Law, I also agree that we should make a declaration of incompatibility.