Coventry and others v Lawrence and another

This judgment is concerned with an attack on the compatibility with the European Convention on Human Rights (“the Convention”) of the system for recovery of costs in civil litigation in England and Wales following the passing of the Access to Justice Act 1999 (“the 1999 Act”). The proceedings to which it relates have been the subject of two previous judgments of this court – [2014] UKSC 13, [2014] AC 822 and [2014] UKSC 46, [2015] AC 106. The fact that this is the third judgment of this court in this case is an unfortunate irony, as the issue which has to be addressed arises from the contention that the order for costs made against the respondents at first instance infringed article 6 of, and/or article 1 of the First Protocol to, the Convention, and considerable further costs have been incurred since then.

A detailed summary of the factual and procedural history is to be found in the first two judgments at [2014] AC 822, paras 2-27 and [2015] AC 106, paras 1-4. So far as relevant for present purposes, and at the risk of over-simplification, the facts are as follows.

The appellants, Katherine Lawrence and Raymond Shields, the owners of a residential bungalow in Mildenhall, Suffolk, brought proceedings for an injunction and damages based on the contention that the noise emanating from speedway and other motorsport activities, operated by David Coventry and Moto-Land UK Limited (“the respondents”) on a stadium and track some 800 metres away, constituted a nuisance. After a trial lasting 11 days, HHJ Seymour QC, sitting as a Deputy High Court Judge, found in favour of the appellants, awarding them damages and an injunction limiting the level of noise emanating from the stadium and track, against the respondents. He also dismissed the claim in so far as it had been brought against the respondents’ landlords (“the landlords”).

So far as the figures are concerned, the amount of damages awarded by the judge in favour of the appellants against the respondents was a total of £20,750; and, on the evidence, the value of the appellants’ bungalow was under £400,000, and the maximum diminution in its value if the nuisance had continued (ie if no injunction had been granted) was £74,000.

Of central relevance for current purposes, the judge also ordered the respondents to pay 60% of the appellant’s costs as assessed on the standard basis. We now have fairly precise figures as to what that order means (subject to the points dealt with in paras 7 and 8 below).

The appellants’ “base costs”, that is the costs payable by the appellants to their lawyers on a conventional basis, in connection with the proceedings up to the time the judge made the order, were £307,642, of which the respondents would therefore be liable for 60%, ie £184,585. However, the appellants agreed with their lawyers that they would proceed on the basis that the lawyers would act under a conditional fee agreement, ie on a “no win no fee” basis. This meant that, as the appellants had won before the judge, they would be liable to pay (i) a success fee to their lawyers on top of the base costs, to compensate the lawyers for acting on a conditional fee agreement, and (ii) a so-called ATE premium, a premium to an insurance company in return for that insurance company having agreed to underwrite any liability which the appellants might have had for the respondents’ costs if the respondents had won. The success fees amounted to £215,007, of which the respondents were liable for 60%, ie £129,004; the ATE premium appears to have been in the region of £305,000, of which the respondents were liable for 60%, ie about £183,000.

The respondents accept that they cannot challenge their liability for the £184,585 on the ground that it infringes their Convention rights. However, Mr McCracken QC was understandably anxious to make it clear on their behalf that they would anticipate strongly challenging the appellants’ right to recover that sum, on the sort of grounds on which a paying party is always entitled to seek to challenge the receiving party’s bill of costs when assessed on the standard basis. Those grounds would be that the total sum is disproportionate, and in any event that the total sum includes items which it was not reasonable to have incurred at all, and that the sums incurred in connection with those items which were reasonably incurred were themselves unreasonable.

However, in relation to the £129,004 and the £183,000, the respondents argue that it would infringe their rights under article 6 of the Convention (“article 6”) and/or article 1 of the First Protocol to the Convention (“A1P1”) if they were liable for those sums, and that is the issue which we preliminarily considered in our second judgment and now have to deal with in this judgment. Again Mr McCracken was anxious to emphasise that, if his arguments based on article 6 and A1P1 fail, the respondents would wish to raise, to the extent which they properly can, the sort of arguments described in para 7 to challenge any liability to these two sums.

In order to complete the history, the judge’s decision on the substantive issue was reversed by the Court of Appeal, who decided that the respondents had not been guilty of nuisance. However, the appellants were successful in their appeal to this court, and, following our first decision, we reinstated Judge Seymour’s order (albeit with modifications), including (subject to what we decide in this judgment) the direction that the respondents pay 60% of the appellants’ costs of the proceedings up to and including his judgment. Our second judgment concerned the liability of the landlords for the respondents’ nuisance, and, in paras 32-46, what were in the event preliminary observations on the respondents’ contention that the extent of their liability for costs pursuant to Judge Seymour’s order (as reinstated by this court) infringed article 6 and/or A1P1. For the reasons given in paras 43-45, we did not feel it right to determine that issue until we had heard argument on the issue from the Secretary of State and any other appropriate intervener.

The appellants’ base costs in the Court of Appeal and the Supreme Court were, respectively, £103,457 and £204,226. The appellants’ success fees were £71,770 in the Court of Appeal and £92,115 in the Supreme Court. The appellants’ ATE premia were £70,141 in the Court of Appeal and £126,588 in the Supreme Court. It appears likely that the effect of the order we make following the two judgments which we have given is that the respondents will be liable for all, or a substantial proportion, of these sums, subject to the arguments discussed in paras 6 to 8 above, including in relation to the success fees and ATE premia the contention that any such liability would be contrary to article 6 and/or A1P1.

The issue raised by the respondents based on article 6 and A1P1 has now been fully argued. We received full submissions from the respondents in support of their contention that their liability to pay the success fees and ATE premia infringed article 6 and A1P1. The case to the contrary was presented by the appellants, with supporting arguments from the Secretary of State for Justice, the General Council of the Bar, the Law Society of England and Wales, the Asbestos Victims Support Group Forum, and (in written form only) the Association of Business Recovery Professionals Limited. We also had submissions from the Media Lawyers Association, which were not directed to the point at issue.

Section 17(1) of the Courts and Legal Services Act 1990 (“the 1990 Act”) stated that the general objective of Part II was the development of legal services in England and Wales “by making provision for new or better ways of providing such services and a wider choice of persons providing them, while maintaining the proper and efficient administration of justice”. Section 58 permitted lawyers, for the first time, to enter into conditional fee agreements (“CFAs”, which, as explained above, were “no win no fee” agreements under which a lawyer was only to be paid a fee if the client won the case) subject to satisfying prescribed conditions. The 1990 Act did not, however, permit the successful party to recover any “success fee” (ie the “uplift” or extra payment which the lawyer would receive in return for agreeing to act on a CFA rather than on the traditional basis) from the losing party.

At the same time as CFAs were being developed under the new statutory regime, the Law Society developed a new form of insurance cover known as after the event insurance (“ATE”). An ATE premium was, as explained above, a sum of money paid to an insurer by a person involved in, or contemplating, litigation, in return for the insurer agreeing to underwrite (sometimes subject to a maximum) the person’s liability to pay the costs of another party to the litigation. The ATE premium was not recoverable from the losing party under the 1990 Act.

In March 1998, the Government published a consultation paper entitled Access to Justice with Conditional Fees. In this paper, the Government stated that it considered that wider use of CFAs and legal expense insurance would promote access to justice. The Government expressly sought views on whether the success fee and the ATE insurance premium should be recoverable by the successful party. The majority of those who responded to the consultation supported the proposal that both should be recoverable. In December 1998, the Government published a White Paper, Modernising Justice (Cm 4155) in which it stated that it had decided to make it possible for the successful party to recover the success fee and ATE premium from the unsuccessful party. It said at para 2.44: “This will make conditional fees more attractive and fairer, and allow respondents and appellants whose case is not about money to use them. This will be a further radical expansion of access to justice”.

The 1999 Act gave effect to this policy decision. Part I of the 1999 Act created a new Legal Services Commission (in place of the Legal Aid Board) with power to determine which types of litigation should qualify for public funding. From 1 April 2000, legal aid was no longer available for personal injury cases, although it continued to be available for clinical negligence cases.

As to the recoverability of CFA success fees, section 27 of the 1999 Act repealed section 58 of the 1990 Act and added a new section 58 and a section 58A. Sections 58A(6) and (7) are in these terms:

As to the recoverability of ATE premiums, section 29 provided:

Following the enactment of the 1999 Act, the Lord Chancellor conducted a further consultation on the question of how success fees and ATE premiums (to which we shall refer compendiously as “additional liabilities”) should be recoverable in practice. This included consultation on the content of the statutory instruments (and rules of court) which were necessary to give effect to the 1999 Act. In January 2000, the Government published its response to the consultation. At para 14, it stated:

The Civil Procedure Rules (“CPR”) were amended by the Civil Procedure (Amendment No 3) Rules 2000 (SI 2000/1317) to reflect the new approach to CFAs and ATE premiums.

CPR 43.2(1)(a) provided that “‘costs’ includes … any additional liability incurred under a funding arrangement” and CPR Rule 43.2(1)(o) provided that “additional liability” included the percentage increase or success fee under a CFA and the premium payable for an ATE insurance policy. CPR Rule 44.4 set out the basis of assessment of costs. It provided:

Rule 44.5 set out the factors to be taken into account in deciding the amount of costs. It provided:

The concept of using proportionality to assess costs, on the standard basis, had not played any part in the taxation, or assessment as it is now called, of costs under the Rules of the Supreme Court (as was recognised by the Court of Appeal in Home Office v Lownds [2002] EWCA Civ 365, [2002] 1 WLR 2450, para 2). The only test was that of reasonableness.

Rule 44.3A contained provisions as to costs orders relating to funding arrangements, including CFAs and ATE insurance. Rule 44.3B set limits on what could be recovered under funding arrangements.

As envisaged by the amendments that were made to the CPR, an amended costs practice direction was promulgated to supplement CPR Parts 43 to 48 (“the CPD”) in order to give effect to section 58A(6) and (7) of the 1999 Act. It is not in dispute that practice directions differ from rules in the CPR in that (a) they provide guidance that should be followed but do not have binding effect; and (b) should yield to rules in the CPR where there is a conflict between them: see White Book (2015), Vol II paras 12-15 to 12-18.

Paragraph 9.1 of the CPD stated that “[u]nder an order for payment of ‘costs’ the costs payable will include an additional liability incurred under a funding arrangement”. Section 11 included the following:

Lord Bingham, having referred to the 1999 Act as having “introduced a new regime for funding litigation”, described the three key aims of the funding regime introduced under the 1999 Act in Callery v Gray (Nos 1 and 2) [2002] UKHL 28, [2002] 1 WLR 2000 as follows:

In MGN Ltd v United Kingdom (2011) 53 EHRR 5, para 197 of the majority judgment of the European Court of Human Rights (“ECtHR”) acknowledged that the CFA with recoverable success fees “sought to achieve the legitimate aim of the widest public access to legal services for civil litigation funded by the private sector”. A deliberate policy of the 1999 Act regime was to impose the cost of all CFA litigation on unsuccessful respondents as a class: see per Lord Hoffmann in Campbell v MGN Ltd (No 2) [2005] UKHL 61, [2005] 1 WLR 3394 at para 16. There was to be a fundamental rebalancing of the means of access to justice by resort to the private sector rather than by the use of public (legal aid) funds. Instead of placing a burden on the legal aid fund, legal proceedings were to be funded in the first instance by a party’s lawyers (who would undertake the work “on risk” in exchange for a potential success fee) and then, if the proceedings were successful, the success fee would be transferred to the losing party.

As Lord Bingham pointed out in para 4 of his speech in Callery, the new funding regime faced two contingencies which “had they occurred, could have proved fatal”: (i) lawyers declining to act on a conditional fee basis; and (ii) no accessible market developing in ATE insurance. To counter the former, the maximum permissible percentage uplift was retained at 100% (as had been the case under the 1990 Act). As regards the latter, a healthy market for ATE insurance did in fact continue to develop.

The concept of proportionality lies at the heart of this case. It is important not to confuse two different meanings of proportionality which are in play here. Lord Hoffmann made the same point at para 23 of his speech in Campbell v MGN Ltd (No 2).

The first meaning of proportionality is that with which we are familiar in the context of the Convention. A valuable recent statement of what proportionality in this sense entails is to be found in Bank Mellat v HM Treasury (No 2) [2013] UKSC 39, [2014] AC 700. Lord Reed said at para 71:

And at para 74, he set out the proportionality test in the following terms:

Lord Reed said that the intensity with which this proportionality test is applied “depends on the context” (para 70) and that “the intensity of review … varies according to the nature of the right at stake and the context in which the interference occurs” (para 71). In applying all four stages mentioned in para 74, appropriate weight must be given to informed legislative choices: see, for example, AXA General Insurance Ltd v The Lord Advocate [2011] UKSC 46, [2012] 1 AC 868 at para 131. In the present context, the concept of “proportionality” in this sense focuses on the balance that is struck by the 1999 Act system between the rights of different types of litigant.

The second meaning of proportionality finds expression in the CPR and is relevant to the assessment of costs. Thus where the amount of costs is to be assessed on the standard basis, the court will only allow costs “which are proportionate to the matters in issue” (rule 44.4(2)(a)). An aspect of this is that the court must have regard to the various matters set out in rule 44.5(3) which we have set out at para 21 above. In Home Office v Lownds [2002] 1 WLR 2450, the Court of Appeal gave important guidance as to the application of proportionality in an assessment of costs on the standard basis. The court adopted the following statement at para 23:

The court also said at para 31 that it was necessary to adopt a two-stage approach to the assessment of costs. The first stage was the “global” approach. At this stage, the question for the court was whether the total sum claimed was disproportionate in particular by reference to the considerations in rule 44.5(3). If the total did not appear to be disproportionate according to that test, all that was normally required was that each item of costs should have been reasonably incurred and the cost for that item should be reasonable. If the costs as a whole did appear to be disproportionate, then the court would want to be satisfied that the work in relation to each item was necessary and, if so, that the costs of each item was reasonable.

The point was put clearly at para 28:

In other words, where base costs were incurred which were necessary, they would be treated as being proportionate even if in fact they were not proportionate “to the matters in issue” (CPR 44.4(2)(a)), ie even if the total necessary costs were disproportionate to the value of the claim.

The introduction of additional liabilities made the proportionality issue more acute. If apparently disproportionate base costs were recoverable under the new CPR regime because they were necessary, what was to be done where the base costs were inflated by additional liabilities?

An additional liability was an element of costs: see CPR 43.2(1)(o). Costs which were unreasonably incurred or which were unreasonable in amount would not be allowed (CPR 44.4(1)). Accordingly, a success fee and an ATE premium would only be allowed to the extent that they were reasonably incurred and were reasonable in amount, having regard to the factors set out at CPR 44.5(3). On the standard basis, only costs which were proportionate to the matters in issue would be allowed (CPR 44.4(2)(a)). The proportionality limitation, therefore, applied to additional liabilities as well as to base costs. CPD para 11.5 did not disapply the proportionality criterion, but confirmed that additional liabilities were to be judged by reference to proportionality, albeit separately from the base costs. The criterion of proportionality therefore applied subject only to the limitation imposed by para 11.9.

In Rogers v Merthyr Tydfil County Borough Council [2006] EWCA Civ 1134, [2007] 1 WLR 808, the Court of Appeal had to consider whether an ATE premium was recoverable by a successful claimant. The court addressed the issue of proportionality at paras 102 to 106. The damages were agreed in the sum of £3,105 plus interest. The case went to trial and the claimant won. The deputy district judge assessed the costs in the sum of £16,821. This included an ATE premium of £5,103. The Court of Appeal held that the premium was recoverable as a proportionate expense if it was necessarily incurred, even if the amount was large in comparison with the amount of damages reasonably claimed. The court said:

In other words, the court did not ask whether the premium was proportionate to the importance of the case and what was at stake. Instead it adopted the Lownds approach. If the premium was necessarily incurred, it was proportionate. And it was proportionate even though it was disproportionately high when compared with the amount of damages reasonably claimed. ATE insurance was integral to the fundamental objective of improving access to justice in civil litigation. A premium that was reasonable in amount (having regard to the litigation risk) was necessary and, therefore, proportionate.

By the same reasoning, a success fee that was reasonable in amount was also necessary and, therefore, proportionate. This principle was applied in many cases. It was essential to the viability of the 1999 Act scheme. In determining whether a success fee was reasonable, the courts had to consider whether it was proportionate to the litigation risk. Thus, for example, if a solicitor had only a 50% chance of earning payment by winning, it was commercially appropriate for him to charge a success fee of 100%. In two identical cases, he would statistically be paid only in one, so that to break even he needed to double his charges in the successful case. In other words, a success fee proportionate to the risk of going unpaid was calculated by dividing the risk of losing by the prospect of winning and multiplying the product by one hundred to yield a percentage. This is the so-called “ready-reckoner” approach which was approved by the Court of Appeal in cases such as Atack v Lee [2004] EWCA Civ 1712, [2005] 1 WLR 2643 and routinely applied by costs judges. This approach only allowed a solicitor to charge for the risk he faced of going unpaid on the basis that he would thereby break even (and recover his normal costs) across his overall case-load. The ready-reckoner principles were also reflected in the statutory limit on success fees of 100% of a solicitor’s basic charges. This fully allowed for solicitors taking on CFA cases with merits of at least 50%, but not cases which were weaker than that.

The submissions of Mr McCracken QC can be summarised as follows. The system set out in the CPD was incompatible with article 6 and A1P1 of the Convention in that it unjustifiably interfered with the article 6 and A1P1 Convention rights of “non-rich” respondents who unsuccessfully contested litigation instituted by appellants who had the benefit of CFA agreements and ATE insurance.

The system had a number of shortcomings which were described as “flaws” by Jackson LJ in his Review of Civil Litigation which were summarised by the ECtHR at paras 207 to 210 of its judgment in MGN v United Kingdom. The flaws were (i) the lack of focus of the regime and the lack of any qualifying requirements for appellants who would be allowed to enter into a CFA; (ii) the absence of any incentive for appellants to control the incurring of legal costs and the fact that judges assessed costs only at the end of the case when it was too late to control costs that had been spent; (iii) the “blackmail” or “chilling” effect of the regime which drove parties to settle early despite good prospects of a defence; and (iv) the fact that the regime gave the opportunity to “cherry pick” winning cases to conduct on CFAs. At para 217, the court concluded that:

These flaws were regarded by the ECtHR as sufficiently serious to lead it to conclude that the system was incompatible with article 10 of the Convention. Mr McCracken submits that the same reasoning necessarily requires the court to hold that the system was also incompatible with article 6 and A1P1.

The system was arbitrary. It singled out from the class of unsuccessful litigants a subset of those who happened to have been opposed by CFA/ATE-funded litigants and imposed on that subset the burden of funding other unsuccessful cases which did not involve them at all.

The real vice of the system lay in the CPD. Paragraph 11.7 based the assessment of CFA uplifts and ATE premiums exclusively on the ex ante perspective of the CFA/ATE party; and para 11.9 expressly disallowed any reduction on the basis that the overall total of base costs and uplifts appeared to be disproportionate. Decisions on uplift therefore disregarded the financial circumstances of the payer, the importance to the payer of fighting the case and the reasonableness of his decision to fight.

The system was not redeemed by the fact that costs were subject to assessment at the end of the proceedings. By that stage, it was too late to control what was being spent. Nor is it an answer to say that the court had the power to cap the costs of a CFA-funded and ATE insurance-protected party at an early stage.

The system failed when tested against the questions identified by Lord Reed in Bank Mellat (No 2), at para 74 (see para 30 above). It is instructive to ask: could a system of private funding of litigation for the non-rich have been adopted which was less intrusive of payers’ fundamental rights without unacceptably compromising the achievement of its objective? There must be an affirmative answer to this question. Apart from the current Legal Aid, Sentencing and Punishment of Offenders Act 2012 (“LASPO”) scheme, examples of less intrusive schemes are: (i) a levy on all adverse costs payments by unsuccessful litigants could have funded the payment of additional liabilities under a modified version of the 1999 Act system; (ii) the system could have been limited to claims against defined groups of “non-ordinary” people (such as insured/large organisations/public bodies); and (iii) the system could have incorporated provisions requiring consideration of all circumstances including (a) the proportionality of the total of base costs and uplifts and premiums, (b) those of the payer (such as his means, whether he was insured, the importance of fighting the case and his reasonableness in fighting the case).

In order to render the system compatible with article 6 (and A1P1) of the Convention, all that is required is to read para 11.9 down in accordance with section 3 of the Human Rights Act 1998 so as to provide a system incorporating the provisions set out at para 48(iii) above.

The first question that we must consider is whether the decision of the ECtHR in MGN v United Kingdom requires us to hold that the 1999 Act scheme is incompatible with article 6 and/or A1P1, at least in relation to the respondents in this case. In that case, the claimant sought damages for breach of confidence and compensation under the Data Protection Act 1998 in respect of the publication in The Daily Mirror of an article and photographs of her. She succeeded at first instance, but lost in the Court of Appeal. She entered into a CFA for the purposes of an appeal to the House of Lords. Her appeal was allowed. The respondents challenged the proportionality of the claimant’s costs (including the success fee). The ECtHR held that there had been a violation of article 10 of the Convention (the right to freedom of expression) as regards the success fee that was payable by the respondents. In defending the CFA scheme, the UK Government advanced arguments similar to those that have been advanced by the Secretary of State (as well as by the appellants and some of the interveners) in the present case. The court held that the requirement to pay the success fees constituted an interference with the defendant’s article 10 rights. The central issue was whether the UK authorities had struck a “fair balance” between freedom of expression protected by article 10 and an individual’s right of access to court protected by article 6 (para 199).

It is true that at paras 207 to 210 the court accepted that the scheme suffered from the four flaws identified in the Jackson report. These flaws were of general application and not confined to defamation and breach of privacy litigation. It is also true that the court said at para 217:

However, the context in which the court made these criticisms was its concern about the effect of the scheme in defamation and privacy cases: see paras 211 to 215. The right of freedom of expression is always given particular weight by the ECtHR. As the court said at para 201, the most careful scrutiny is called for when measures are capable of discouraging the participation of the press in debates over matters of legitimate public concern. It concluded that a fair balance had not been struck between the article 10 rights of defendant publishers and the article 6 rights of appellants who allege defamation or breach of privacy. But in our judgment the balancing of the article 6 rights of appellants against those of respondents is an exercise of a wholly different character. There is no basis for concluding that it was implicit in the reasoning of the court that it would have held that the scheme violated the article 6 rights of the respondents in that case. We reject the submission that the decision in MGN v United Kingdom requires us to hold that the 1999 Act scheme is incompatible with article 6. Essentially for the same reasons, we do not consider that MGN v United Kingdom assists the respondents in relation to their case under A1P1.

We have summarised the four flaws to which the ECtHR referred in MGN v United Kingdom at para 43 above. We accept that the first flaw (the regime was unfocussed), the second flaw (the fact that costs were assessed only at the end of the case) and the fourth flaw (the opportunity for lawyers to cherry pick) were flaws in the system, but they could not have adversely affected the article 6 rights, or the A1P1 rights, of opposing parties. It is the third flaw which lies at the heart of the present case. It is described as the “blackmail” or “chilling” effect. Another way of describing it is as imposing a costs burden on opposing parties which is excessive and in some cases amounts to a denial of justice. Whether this flaw rendered the 1999 Act scheme incompatible with article 6 or A1P1 is the central question that arises in this case and which we discuss in detail below.

But before we do this, we need to refer to what Lord Neuberger described at para 37 of his judgment in Coventry v Lawrence (No 2) as four “unique and regrettable features” of the scheme on which the respondents also rely in support of their case. To some extent these features overlap with the four flaws. The first feature was that appellants had no interest in the level of fees which they agreed to pay their lawyers. The second was that in many cases unsuccessful respondents found themselves paying, in addition to their own costs, three times the appellants’ “real” costs. The third was that proportionality was excluded from consideration in relation to the recovery of the success fee or ATE premium. The fourth was that the stronger the respondents’ case, the greater their liability costs would be if they lost, since the size of the success fee and the premium should have reflected the appellants’ prospects of success.

The first feature was undoubtedly present in many cases, but the reasonableness of the base costs and the additional liabilities could always be the subject of scrutiny by the judge charged with the assessment of the costs. It put a particularly heavy responsibility on costs judges, but the fact that a paying opposing party was entitled to have the costs for which he was liable assessed by an independent and expert tribunal by reference to reasonableness must be a very strong mitigating factor. The second feature was present in those cases where there was a 100% success fee and a premium which equated with the extent of the claimant’s “real” costs recovery. This might be the situation with some cases, but it was not a regular feature of the system as a whole. At best, it is no more than an arithmetical description of the 1999 Act scheme at its worst. The third feature (reflected in para 11.9 of the CPD) was an intrinsic part of the regime as a whole: see the discussion at paras 38 to 41 above. The fourth feature was also an inevitable consequence of the regime, but it did not necessarily interfere with the article 6 or A1P1 rights of opposing parties.

Much has been made of the alleged unfairness of the system. But the issue is not whether the system was unfair or had “flaws”. It is whether it was a disproportionate way of achieving the legitimate aim. In Swift v Secretary of State for Justice [2013] EWCA Civ 193, [2013] 3 WLR 1151, Lord Dyson said at para 35:

Although these observations were made in a context far removed from that with which we are concerned, they are apt here too.

It is common ground that the question whether a fair balance has been struck between the interests of those litigants who have CFAs and ATE insurance and those who do not is one for the court to determine. But, even in a field such as access to justice and legal costs, the court, while being vigilant to protect fundamental rights, must give considerable weight to informed legislative choices, at least where state authorities are seeking to reconcile the competing interests of different groups in society. In such cases, they are bound to have to draw a line somewhere in order to mark where a particular interest prevails and another one yields. Making a reasonable assessment of where to draw the line, especially if that assessment involves balancing conflicting interests falls within the State’s wide discretionary area of judgment. As Lord Bingham said in Brown v Stott [2003] 1 AC 681, 703:

The choices made by Parliament in enacting the 1999 Act followed a wide consultation to enable it to evaluate the various interests at stake. Similarly, in formulating the CPR and the CPD, the relevant rule-makers were (following consultation) in the best position to determine how to effect the reforms and how to strike the appropriate balance between the different types of litigant.

An important question is whether the compatibility of the scheme should be judged “by reference to the generality of cases, so that a few unfortunate results are inevitable”: see per Lord Neuberger in Coventry v Lawrence (No 2) at para 44. In Campbell v MGN Ltd (No 2), Lord Hoffmann considered that this approach should be adopted in relation to the question whether the 1999 Act scheme was compatible with article 10 of the Convention. He said:

In MGN v United Kingdom, the ECtHR rejected the Government’s submission that “any disproportionality visited on an individual case by the CFA/recoverable success fee regime was justified by the need to adopt provisions of general application when pursuing broad social and economic policy objectives” (para 202). The court seems to have rejected it because the lengthy public consultation that had taken place since 2003 had exposed flaws in the system that were too serious to survive scrutiny.

Nevertheless, the ECtHR recognises that a legislative or regulatory scheme may in some circumstances be compatible with the Convention even if it operates harshly in individual cases. The issue was considered in some detail in Animal Defenders v United Kingdom (2013) 57 EHRR 21. That case involved a challenge to the UK laws which ban political advertising on TV and radio. There was no dispute that the ban amounted to an interference with article 10 rights, was prescribed by law and pursued a legitimate aim. The issue was whether the interference was proportionate to the legitimate aim. The court made the following preliminary remarks to the effect that a scheme can be compatible with the Convention even though it produces hard results in individual cases. It said:

The court held that there was no violation of article 10 because the impact of the prohibition did not outweigh the convincing justifications for the general measure (para 124). When the same issue had been before the House of Lords, Lord Bingham said the same thing, namely that “[a] general rule means that a line must be drawn, and it is for Parliament to decide where”, and added that this “inevitably means that hard cases will arise falling on the wrong side of it, but that should not be held to invalidate the rule if, judged in the round, it is beneficial” - R (Animal Defenders International) v Secretary of State for Culture, Media and Sport [2008] UKHL 15, [2008] 1 AC 1312, para 33.

In our judgment, there is a powerful argument that the 1999 Act scheme is compatible with the Convention for the simple reason that it is a general measure which was (i) justified by the need to widen access to justice to litigants following the withdrawal of legal aid; (ii) made following wide consultation and (iii) fell within the wide area of discretionary judgment of the legislature and rule-makers to make. On that basis, it is no answer to say that other measures could have been taken which would have operated less harshly on non-rich respondents: the reasoning of the ECtHR at para 110 of the Animal Defenders case is particularly apposite here. Nevertheless, we do not propose to base our conclusion solely on this argument: we bear in mind the fact that it was rejected in MGN v United Kingdom, albeit in the context of an article 10 case. Accordingly, we proceed to examine the position rather more critically.

The withdrawal of legal aid in most areas of civil litigation presented a real problem for the government. It had to decide how to secure access to justice for those who previously qualified for legal aid. Under the first scheme that was adopted (and which was in force from 1995 until 2000), when success fees were permitted for the first time and ATE insurance was first encouraged, the success fee and ATE insurance premium were not recoverable from the opposing party. The problems with this scheme included that (i) it only worked well where appellants sought substantial monetary relief (thereby realising a fund, in the event of success, from which the success fee would be paid) and (ii) damages recoverable by CFA appellants were eroded by the irrecoverable cost of funding and ATE insurance.

These difficulties were overcome by the 1999 Act scheme. The first difficulty was overcome because a substantial fund of damages was no longer necessary to secure the payment of success fees and ATE premiums: inter partes costs orders were sufficient. The second difficulty was resolved because damages (or, in a low money or non-money claim, the litigant’s own funds) were no longer eroded by irrecoverable success fees and premiums. In policy terms, the principal shift from the first scheme to the second scheme was to transfer the cost of financing successful claims from winning litigants to losing litigants. The cost of unsuccessful claims remained with lawyers and ATE insurers.

The potential unfairness of the 1999 Act scheme on unsuccessful litigants was mitigated by the fact that district judges and costs judges would perform the role of “watchdog” as Lord Bingham described it in Callery v Gray (Nos 1 and 2) [2002] UKHL 28, [2002] 1 WLR 2000 at para 6. Lord Bingham said that the courts would be astute to check any practices which might undermine the fairness of the new funding regime, which was to “operate so as to promote access to justice and not so as to confer disproportionate benefits on legal practitioners or after the event insurers or impose unfair burdens on respondents or their insurers” (para 10). Thus the base cost and any additional liabilities were to be assessed by the court. As to base costs, where costs were to be paid on the standard basis they were to be judged by the criteria of reasonableness and proportionality. Where costs were to be paid on the indemnity basis, they were to be judged by the sole criterion of reasonableness. As regards any additional liability, a successful litigant was only entitled to a reasonable success fee and ATE premium and (where costs were assessed on the standard basis) a proportionate success fee (as explained in Lownds). In an appropriate case, the court had the ability to make a cost-capping order as was required, for example, by the Court of Appeal in King v Telegraph Group Ltd [2004] EWCA Civ 613, [2005] 1 WLR 2282.

Nor should it be overlooked that respondents could themselves enter into CFAs and take out ATE insurance.

There was, and indeed there is, no perfect solution to the problem of how best to enhance access to justice following the withdrawal of legal aid for most civil cases. A successful defendant was often better off under the 1999 Act scheme than he had been when legal aid was generally available to appellants. At that time, a successful defendant usually had to bear his own costs of defending a claim. The appellant did not have the means to meet the defendant’s costs and it was a rare case in which a successful defendant would be able to obtain its costs from the legal aid fund. Under the 1999 Act regime, the successful defendant would usually obtain its costs from the ATE insurer. On the other hand, the unsuccessful defendant was unquestionably better off under the previous regime because it was only liable for the claimant’s base costs. This was the policy choice that was made by Parliament.

Mr McCracken submits that the current LASPO scheme (based on the proposals for reform made by Sir Rupert Jackson) is fairer than the scheme that it replaced. The LASPO scheme was intended to readjust the balance which had been adjusted in 1990 and 1999, and it has inevitably curtailed access to the courts in some respects as a result, as is demonstrated by the facts of this case. Appellants of modest means cannot finance litigation without a CFA. But that inevitably requires them to pay a success fee on their solicitors’ and counsel’s basic charges. In a substantial case, these costs are bound to be high. How is the success fee to be paid by appellants who bring claims for non-financial remedies or where the damages claimed are very small? Sir Rupert recognised the problem, when he called for general damages to be increased by 10%. This was effected by the Court of Appeal in Simmons v Castle [2012] EWCA Civ 1039 and 1288, [2013] 1 WLR 1239. But in the present case, this would have benefited the appellants to the extent of only £2,085. Even if the success fees were to be substantially reduced on assessment, this increase in damages would represent a very small fraction of the overall figure. In short, under the LASPO regime, the present litigation would not have been viable. The success fees are almost certainly more than the appellants’ likely damages, and more than the financial value of the rights they are attempting to protect (the diminution in value of their home being, on the expert evidence, no more than £74,000).

Sir Rupert Jackson also recognised the serious impediment that inter partes costs liability would cause if ATE premiums were not to be recoverable. That is why he proposed qualified one-way costs shifting (QOCS). But QOCS was only introduced in respect of personal injury claims. It has no impact on other claims brought by litigants of relatively modest means (such as the present claim).

The reason for referring to the LASPO scheme at some length is not to criticise the Jackson reforms, but (i) to show that there are restrictions on access to justice inherent in the LASPO scheme and (ii) to demonstrate that, at least in the absence of a widely accessible civil legal aid system (which had ceased to exist by 1999), it is impossible to devise a fair scheme which promotes access to justice for all litigants.

At this stage, we should comment on the other “less intrusive” schemes suggested by Mr McCracken. A levy on all adverse costs payments by unsuccessful litigants would be a radically different scheme from anything adopted hitherto in our civil justice system. It would require primary legislation and would probably be highly controversial. It is impossible to predict how politically acceptable such a proposal would be and what consequences it would bring in its train. It is at best a proposal whose potential effect and effectiveness is no more than speculative. We reject the idea that, viewed in the round, it would better promote the Convention rights of litigants than the 1999 Act scheme.

We also reject the suggestion that the 1999 Act scheme should have been limited to defined categories of rich litigants, on the basis that, if it had been limited in this way, it would have been unlikely to interfere with their article 6 or A1P1 rights. But how would a “large organisation” have been defined? Even large organisations become insolvent or at least face financial difficulties from time to time. How (if at all) would the scheme have taken account of the fact that some individuals are wealthier than some large organisations? At what stage of the proceedings would a decision have been made by the court on the question whether the litigant fell within the category of rich litigants? In our view, questions such as these demonstrate that a scheme limited to defined categories of rich litigants would have been uncertain and arbitrary, and would almost certainly have led to a great deal of expensive and time-consuming “satellite” litigation to determine the extent of the means of many actual and potential litigants.

The most sustained argument was that a scheme less intrusive of an unsuccessful litigant’s article 6 rights would have required consideration of all the circumstances including (a) the proportionality of the total of base costs and uplifts and premiums and (b) all of the payer’s circumstances. It is important to clarify that proportionate costs in this context means costs which were proportionate to what was at stake in the litigation. The argument is that, even if the total costs (including uplifts and premiums) were reasonably and necessarily incurred by the litigant in order to have the benefit of legal representation, they should not have been allowed on assessment to the extent that they exceeded a sum which was proportionate to what was at stake.

Much of Mr McCracken’s attack therefore was directed at CPD 11.9. But CPD 11.9 cannot be viewed in isolation. CPD 11 made separate provision for base costs and additional liability. CPD 11.5 provided that the court would consider the amount of any additional liability separately from the base costs. CPD 11.7 provided that, when the court was considering the factors to be taken into account in assessing an additional liability, it would have regard to the facts and circumstances “as they reasonably appeared to the solicitor or counsel when the funding arrangement was entered into …”. By contrast, the assessment of base costs was conducted ex post facto on the basis of what happened in the litigation. The reason for treating the assessment of an additional liability differently was that, if legal representatives knew that the assessment of the reasonableness and proportionality of their success fees would be reviewed by the court on an ex post facto basis (rather than on the basis of what they reasonably thought would happen), this would have been likely to discourage them from entering into CFAs. They would have been concerned that, if they were successful, the success fees payable by the unsuccessful party might be reduced for reasons which they could not reasonably have foreseen when they entered into the CFA. They would have been faced with the unattractive choice between foregoing the unrecovered part of the fee or seeking to recover it from their client. That would have risked undermining the whole system introduced by the 1999 Act.

CPD 11.9 provided that a percentage increase would not be reduced simply on the ground that, when added to base costs which were reasonable and (where relevant proportionate), the total appeared disproportionate. Like CPD 11.5, this provision was necessary to make the scheme work. If legal representatives knew that reasonable success fees were liable to be reduced on the grounds that, when added to the base costs, the total appeared to be disproportionate, this would have been likely to deter them from entering into CFAs. Success fees were calculated on the basis of an assessment of the risk of losing (the so-called “ready-reckoner approach”). That was intrinsic to the CFA system. To use the language of Lownds, a reasonable success fee was necessary to fund the litigation. The reasoning in Rogers is apposite here. As we have seen, the court held that, if an ATE premium was necessarily incurred, it was to be regarded as proportionate and therefore recoverable. It was “integral to the means of providing access to justice in civil disputes in what may be called the post-legal aid world”: [2007] 1 WLR 808, para 105. So too was a CFA success fee. It was necessary in order to secure access to justice. It was therefore proportionate. If it were otherwise, there would have been a real danger (to put it no higher) that litigants who previously qualified for legal aid would have been unrepresented and the fundamental and legitimate aim of the 1999 Act scheme would have been frustrated.

In summary, if the basis upon which Mr McCracken’s attack on section 11.7 and 11.9 was founded were to be accepted, it would have imperilled the whole scheme which had been put in place by the 1999 Act, because lawyers would have been unwilling to enter into CFAs for fear that, even if successful, the uplift which they had agreed on the basis envisaged by the system embodied in the 1999 Act would have been liable to be reduced or disallowed on assessment because it would have been held to have been disproportionate to what was at stake in the litigation.

Nor can we accept that the scheme was incompatible with the Convention on the grounds that the assessment of the successful party’s total costs did not take account of the paying party’s financial circumstances. So far as we are aware, the financial position of the paying party has never been a relevant factor in determining the assessment of reasonable and proportionate costs. As already mentioned, if the position were otherwise, there would be a real risk of protracted and expensive disputes as to the true financial worth of the paying party, often with requests for disclosure of documents and cross-examination. In other words, it would lead to satellite litigation, with the expenses, delays and uncertainties which such litigation normally involves.

We should interpolate that it was suggested in argument that the rules already provided that the financial circumstances of the paying party could (and perhaps should) be taken into account in the assessment of costs. CPR rule 44.5(1) required the court to have regard “to all the circumstances” in deciding costs. CPR 1.2 required the court to give effect to the “overriding objective” of dealing with cases justly when exercising any power under the Rules. CPR 1.1(2) provided that dealing with cases justly and at proportionate cost includes so far as practicable “(c) dealing with the case in ways which are proportionate – (iv) to the financial position of each party”. Finally, para 11.1 of the CPD provided that in applying the test of proportionality “the court will have regard to rule 1.1(2)(c)”.

If para 11.1 of the CPD is left out of account, we are satisfied that the CPR did not require or permit costs to be assessed by reference to the financial position of the parties. CPR 44.5 set out the factors to be taken into account in assessing costs. They did not include the financial circumstances of the parties. That is not at all surprising. If the fact that the paying party had limited resources militated in favour of reducing the amount of costs that it was reasonable for him to pay, then why should the fact that he was wealthy not militate in favour of increasing the costs? The logic of the argument should also mean that the financial circumstances of the receiving party was relevant. We are satisfied that this cannot have been intended by the draftsman of the rules. If the draftsman had intended financial circumstances to be taken into account, he would have included them as a relevant factor in CPR 44.5. We do not consider that financial circumstances could be introduced as an additional relevant factor by means of CPR 1.1(2)(c). If rule 1.1(2)(c) was relevant to the assessment of costs at all, it related only to the manner in which the process of the assessment was conducted. It was not relevant to the amount of costs to be paid.

Paragraph 11.1 of the CPD did, however, purport to go further than the rules. It purported to say that the court would have regard to the financial position of each party in applying the test of proportionality. In our view, this conflicted with the rules which, for the reasons that we have given, did not permit this. It should, therefore, not have been followed or should have been disapplied: see para 24 above.

To summarise. It was undoubtedly a feature of the 1999 Act scheme that the costs awarded to successful appellants who had the benefit of CFAs could be very high indeed. For that reason, it had the potential to place respondents under considerable pressure to settle before even more costs were incurred. This is the third flaw identified by the ECtHR in MGN v United Kingdom and the second of Lord Neuberger’s four unique and regrettable features. We accept that, in a number of individual cases, the scheme might be said to have interfered with a defendant’s right of access to justice. But for the reasons stated earlier (paras 58 to 63 above), it is necessary to concentrate on the scheme as a whole. The scheme as a whole was a rational and coherent scheme for providing access to justice to those to whom it would probably otherwise have been denied. It was subject to certain safeguards. The government was entitled to a considerable area of discretionary judgment in choosing the scheme that it considered would strike the right balance between the interests of appellants and respondents whilst at the same time securing access to justice to those who would previously have qualified for legal aid. It had to find a solution to the problem created by the withdrawal of legal aid. The government has now produced three different schemes. Each was produced after wide consultation. Each has generated considerable criticism. As already indicated, once civil legal aid was constrained to the extent that it was in 1999, it became impossible to come up with a solution which would meet with universal approval. This is relevant to the question whether the 1999 Act scheme struck a fair balance between the interests of different litigants.

For the reasons that we have given, we are satisfied that the scheme was not incompatible with article 6 or A1P1.

For completeness, we should add that it was argued that, in any event, at least one of the respondents had failed to establish that he was not “non-ordinary” or “non-rich” (see para 48), either because there was no evidence of his means or because he was in fact insured against liability for nuisance. For the reasons we have given, it is unnecessary to decide whether that is a well-founded argument. However, the very fact that it has been raised demonstrates the risk of satellite litigation if the respondents’ case is accepted: it would be necessary to assess a party’s means and liabilities, identify the precise terms of an insurance policy that has been mislaid, and then decide whether it covered nuisance by noise.

This only arises if (contrary to our view) the system was incompatible with article 6 and A1P1. No-one is arguing for a declaration of incompatibility under section 4 of the Human Rights Act 1998. There is nothing in the language of the 1999 Act which requires any particular scheme. We cannot read “may” in section 58A as meaning “must”. The statute permits a scheme to be created (by rules and practice directions), but it does no more than that.

As we have said, the real complaint is about paras 11.7 and 11.9 of the CPD, especially para 11.9. If these render the scheme incompatible for the reasons given by Mr McCracken, the question arises whether they can and should be “read down” or disapplied so to require consideration of all the circumstances including (a) the proportionality (to the matters in issue in the litigation) of the total of base costs and uplifts and premiums; and (b) the financial circumstances of the paying party.

We do not consider that it is possible to read down para 11.9 in accordance with possibility (a) or to disapply it. It would involve a radical departure from the Lownds approach to proportionality. The question would not be whether the success fee and ATE premium were necessary (and therefore treated as proportionate), but rather whether they were in fact proportionate. The focus of the former question in a CFA case was on the reasonableness of the fee and premium having regard to the litigation risk. The focus of the latter question was on whether the amount of the costs was proportionate to what was at stake in the litigation. In our judgment, it is impermissible to do this under the guise of interpretation. As we have explained at paras 75 to 77 above, lawyers would have been deterred from entering into CFAs if they knew that the success fees were susceptible to reduction on the grounds that, when added to base costs, they appeared to be disproportionate to the matters in issue. Paragraph 11.9 was integral to the scheme. It cannot be interpreted away.

Even if it was open to us to read down para 11.9 in the way for which Mr McCracken contends we do not consider that it would be right to do so. Nor would it be right to disapply it. As the Bar Council points out, the Court of Appeal actively shaped the law relating to additional liabilities throughout the period from 2000 until 2013. It was implicit in all of the cases that success fees (often substantial success fees) were recoverable. In none of the cases did the court disallow or reduce the amounts payable in success fees on the grounds that they were so high as to amount to a breach of the paying party’s Convention rights. In these circumstances, litigants and their lawyers had a legitimate expectation that the court would not (at least without reasonable notice) decide that these fees were in principle incompatible with the Convention.

This is no mere abstract statement. A decision to declare that the 1999 Act scheme was incompatible with the Convention would have a serious impact on many thousands of pre-April 2013 cases which are in run-off, as well as claims to which the pre-Jackson costs rules continue to apply, such as mesothelioma, insolvency and publication and privacy cases. Any order made by this court in the present case would have no effect on the contractual obligations of litigants to pay success fees to their lawyers and ATE premiums to their insurers. Successful parties would, therefore, still be liable to pay their lawyers and insurers if they won their cases and could not recover them from unsuccessful respondents.

Mr McCracken responds with three points. First, he says that there are good prospects that the lawyers would not seek to enforce the success fees for reasons of (i) decency, (ii) reputational risk and (iii) questionable enforceability in view of advice (in a situation of conflict of interest) about agreements with these problematic terms. Secondly, some appellants entered into CFAs and ATE insurance policies in the knowledge that their terms made them liable to pay the success fee and premium in the event of success even if, for whatever reason, those sums were not in fact recovered from the defendant. It cannot be said that an order disallowing recovery of success fee and ATE insurance premium would impose on such a claimant a burden he never envisaged having to assume. Thirdly, if the lawyer did not advise the claimant that he should at least consider not entering into a CFA or ATE insurance policy unless his liability to pay the success fee or premium was limited to whatever was actually recovered from the paying party, or provide some other assurance that they would not have to pay in uplift more than they could sensibly risk, then they would potentially have a remedy in negligence against the lawyer.

We do not consider that these points provide a sufficient answer. We accept that some lawyers might not seek to enforce the success fees. But others might well do so, particularly if their client had means. It would be quite wrong to assume that only litigants of modest resources would have entered into CFAs. As regards the second point, we accept that the terms of some CFAs and ATE policies imposed a burden on the litigant to pay the success fee and premium in certain circumstances. But it does not follow that all litigants who entered into CFAs and ATE premiums did so in the belief that they were at risk, even if successful, of having to pay the success fee and premium. We suspect that many such litigants would have been most surprised to be told that this was the position. As for the third point, in view of the fact that CFAs and ATE insurance policies had been routinely used throughout the period 2000-2013 and sanctioned by the courts (even the House of Lords) without any suggestion that they were incompatible with article 6 or A1P1 or otherwise unlawful, it would be remarkable if a lawyer who advised his client to enter into a typical CFA and take out a typical ATE insurance policy could be said to have acted negligently.

The result of declaring that the 1999 Act scheme was contrary to the Convention might also be that some appellants would decide to abandon their claims so as to avoid incurring further irrecoverable liabilities.

As regards the financial circumstances of the paying party, to take these into account would involve a substantial change to the CPD and rules as we have interpreted them at paras 78 to 81 above. This too would involve a fundamental change to the 1999 Act scheme and cannot be achieved under the guise of interpretation.

For the reasons that we have given, the 1999 Act scheme was compatible with article 6 and A1P1. We have not addressed A1P1 separately. That is because it has (rightly) not been suggested that, if the scheme was compatible with article 6, it could nevertheless for some other reason be incompatible with A1P1.

If (contrary to our view) the scheme was incompatible with article 6 and A1P1, we would not read it down so as to make it compatible and we would not strike the scheme down or disapply it.

This is an awkward case. The Supreme Court is embarked on an examination of a system of costs which, despite its replacement in 2013 for future litigation by the LASPO scheme, remains applicable in many pending cases, and which appellate courts have from 1999 onwards repeatedly endorsed, developed and enforced: see eg Callery v Gray [2001] EWCA Civ 1117, [2001] 1 WLR 2112 and [2002] UKHL 28, [2002] 1 WLR 2000; Atack v Lee [2004] EWCA Civ 1712, [2005] 1 WLR 2643; Halloran v Delaney [2002] EWCA Civ 1258, [2003] 1 WLR 28; In re Claims Direct Test Cases [2003] EWCA Civ 136, [2003] 4 All ER 508; U v Liverpool City Council (Practice Note) [2005] EWCA Civ 475, [2005] 1 WLR 2657; Rogers v Methyr Tydfil County Borough Council (Law Society intervening) (Practice Note) [2006] EWCA Civ 1134, [2007] 1 WLR 808; Crane v Canons Leisure Centre [2007] EWCA Civ 1352, [2008] 1 WLR 2549; and C v W [2008] EWCA Civ 1459; [2009] 4 All ER 1129.

Lord Neuberger and Lord Dyson make a powerful case in their judgment for a conclusion that the system of costs not only fulfils the legitimate aim of affording access to justice to appellants (that is unchallenged), but does so in a way which falls within the wide area of discretionary judgment which rule-makers must be recognised as having when balancing the interests of those seeking access to justice and respondents faced with the additional burden of costs which the system could impose.

In striking such a balance, the state is entitled to look at the system as a whole, and the possibility of individual hard cases is not itself fatal: see Animal Defenders cases, cited by Lord Neuberger and Lord Dyson in paras 62 and 63. As the ECtHR there put it, “the more convincing the general justifications for the general measure are, the less importance the court will attach to its impact in the particular case”, and Lord Bingham in the House of Lords had said the same thing. See also James v United Kingdom (1986) 8 EHRR 123.

Nonetheless, the European Court of Human Rights was in MGN Ltd v United Kingdom (2011) 53 EHRR 195 persuaded, after an examination of the system and of the flaws in it identified by Sir Rupert Jackson and accepted by the Government, that, despite its legitimate aim to promote access to justice, it operated disproportionately in the context of claims for defamation, because of its effect on the right to freedom of expression protected under article 10 of the Convention.

The appellants, supported by the Secretary of State for Justice and legal professional and other bodies, distinguish this decision and its reasoning on the basis that there is here no competing interest comparable to freedom of expression. Instead, the respondents are relying upon their own right of access to justice under article 6 and/or their right to protection of their property under paragraph 1 of protocol 1 to the Convention (“A1P1”).

Lord Neuberger and Lord Dyson accept this distinction, and regard the balancing exercise involved in the present appeal as being “of a wholly different character” (para 52). In para 79 they discount any suggestion that the scheme in force from 1999 to 2013 was incompatible because it did not take account of the paying party’s financial circumstances, but in para 83 they accept that, in a number of individual cases, it might be said to have interfered with a respondent’s right of access to justice. Nonetheless, they conclude that, viewed as a whole, it was rational and coherent, and not incompatible with either article 6 or A1P1.

While freedom of expression is a particularly powerful interest under the Convention, the interest of any respondent in being able to defend himself or itself in litigation, at a reasonable and proportionate cost is, in my opinion, also one of some weight; and it certainly engages (as I understand Lord Neuberger and Lord Dyson to accept) a balancing exercise, when set against the countervailing interest that appellants should have the access to justice which the system was designed to give.

In this context, the strength of the present respondents’ case lies in their claim to be individuals or small undertakings carrying on modest businesses without insurance and faced with one-off litigation, which has involved them in eye-catchingly large costs exposure. Precisely how compelling this claim is, as Lord Neuberger and Lord Dyson note (para 85), untested. Even small businesses, carrying on motor racing activities, would be expected to carry some forms of insurance, although it is at least open to doubt whether the insurance against nuisance which at any rate the first respondent carried covered nuisance by noise; and whether noise insurance would have been available to, or have been expected to be obtained by, either the second or third respondent has not been investigated.

Whatever the position in this regard, the balancing exercise and any decision as to the validity of the system and the grant of any relief must all be undertaken taking account of the circumstances and competing interests as they stand at the time of the present proceedings. It is unnecessary and indeed inappropriate to scrutinise the scheme in the same way as would or might have been appropriate before or when it first came into force. Since then, much water has flowed under the bridge, in terms of the rules made and practice directions issued under the legislation, and the constant jurisprudence of domestic courts endorsing the system and of litigants and their lawyers acting on the basis that it applied and was valid.

I do not in this context accept the submission, made by the respondents and supported by the Department of Justice for Northern Ireland, that litigants and their lawyers cannot have had a legitimate expectation that the system would apply and be upheld. When appellate courts have repeatedly endorsed the system, it seems to me unrealistic to expect them to have avoided use of the system from concern about what would, if appreciated at all, have been seen as a remote risk that courts might change their attitude. I also consider that their legitimate expectation that the system would be enforced is one which falls to be taken into account at the present stage, and is not merely a matter that might (being itself a protected possession within A1P1) be raised as against the United Kingdom in Strasbourg.

In the above circumstances, I reject the respondents’ challenge to the system of costs whereby they are potentially liable in respect of success fees agreed and ATE premia incurred by the appellants. The position must, as Lord Neuberger and Lord Dyson have said, be considered as a whole. The system had a legitimate aim, the present is on its face an extreme and unusual case. It is difficult to conceive of any solution which would cater for such cases, without imperilling the whole system. The system has been repeatedly endorsed by domestic courts over a decade. Litigants and their lawyers have justifiably relied upon its validity. Legal certainty, consistency and the legitimate expectations which have so been generated all militate in favour of the Supreme Court upholding the system (though it can of course still be challenged as against the United Kingdom in proceedings in Strasbourg); and I would uphold it and refuse any relief accordingly.

I agree with Lord Mance that this is an awkward case. Subject to one critical point, like him, I agree that Lord Neuberger and Lord Dyson make a powerful case that the system of costs under review not only, as is conceded on behalf of the respondents, fulfils the legitimate aim of affording access to justice to appellants, but also falls within the wide area of discretionary judgment which rule-makers must be recognised as having when balancing the interests of those seeking access to justice and respondents faced with the additional burden of costs which the system could impose.

I also agree with Lord Mance (at para 99) that, in striking such a balance, the state is entitled to look at the system as a whole, and the possibility of individual hard cases is not itself fatal: see the Animal Defenders cases, cited by Lord Neuberger and Lord Dyson in paras 62 and 63. Lord Mance notes the general point made by the ECtHR that, “the more convincing the general justifications for the general measure are, the less importance the Court will attach to its impact in the particular case”. He also observes that Lord Bingham had said the same thing in the House of Lords and refers to James v United Kingdom (1986) 8 EHRR 123.

As Lord Mance observes at para 100, in MGN Ltd v United Kingdom (2011) 53 EHRR 195, the ECtHR was nonetheless persuaded, after an examination of the system and of the flaws in it identified by Sir Rupert Jackson and accepted by the Government, that, despite its legitimate aim to promote access to justice, it operated disproportionately in the context of claims for defamation, because of its effect on the right to freedom of expression protected under article 10 of the Convention. As I see it, the question is whether that decision can properly be distinguished from the issue in this appeal on the footing that there is here no competing interest comparable to freedom of expression. Lord Neuberger and Lord Dyson say that it can. I respectfully disagree. In my opinion, the principles identified by the ECtHR apply to the facts of this case, where the respondents are relying upon their own right of access to justice and to a fair trial under article 6 and/or their right to protection of their property under paragraph 1 of protocol 1 to the Convention (“A1P1”).

Lord Neuberger and Lord Dyson summarise the respondents’ case at paras 42 to 49 as follows. For ease of reference I set their case out here in much the same way. The system unjustifiably interferes with the article 6 and A1P1 Convention rights of “non-rich” respondents who unsuccessfully contest litigation instituted by appellants who had the benefit of CFA agreements and ATE insurance. In his Review of Civil Litigation Sir Rupert Jackson identified a number of what he described as flaws, which were summarised by the ECtHR at paras 207 to 210 of its judgment in MGN v United Kingdom. They were (i) the lack of focus of the regime and the lack of any qualifying requirements for appellants who would be allowed to enter into a CFA; (ii) the absence of any incentive for appellants to control the incurring of legal costs and the fact that judges assessed costs only at the end of the case when it was too late to control costs that had been spent; (iii) the “blackmail” or “chilling” effect of the regime which drove parties to settle early despite good prospects of a defence; and (iv) the fact that the regime gave the opportunity to “cherry pick” winning cases to conduct on CFAs. At para 217, the ECtHR concluded that:

The system was arbitrary. It singled out from the class of unsuccessful litigants a subset of those who happened to have been opposed by CFA/ATE-funded litigants and imposed on that subset the burden of funding other unsuccessful cases which did not involve them at all. The real vice of the system lay in the CPD, which, so far as relevant, is quoted by Lord Neuberger and Lord Dyson in para 25 above. Paragraph 11.7 based the assessment of CFA uplifts and ATE premiums exclusively on the ex ante perspective of the CFA/ATE party; and para 11.9 expressly disallowed any reduction on the basis that the overall total of base costs and uplifts appeared to be disproportionate. Decisions on uplift therefore disregarded the financial circumstances of the payer, the importance to the payer of fighting the case and the reasonableness of his decision to fight.

The system was not redeemed by the fact that costs were subject to assessment at the end of the proceedings. By that stage, it was too late to control what was being spent. Nor is it an answer to say that the court had the power to cap the costs of a CFA-funded and ATE insurance-protected party at an early stage.

The system failed when tested against the questions identified by Lord Reed in Bank Mellat (No 2), at para 74, which are quoted by Lord Neuberger and Lord Dyson at para 30 above. It is instructive to ask: could a system of private funding of litigation for the non-rich have been adopted which was less intrusive of payers’ fundamental rights without unacceptably compromising the achievement of its objective? There must be an affirmative answer to this question. Apart from the current LASPO scheme, examples of less intrusive schemes are: (i) a levy on all adverse costs payments by unsuccessful litigants could have funded the payment of additional liabilities under a modified version of the 1999 Act system; (ii) the system could have been limited to claims against defined groups, such as insured, or large corporations, large organisations, or public bodies; and (iii) the system could have incorporated provisions requiring consideration of all the circumstances including (a) the proportionality of the total of base costs and uplifts and premiums and (b) those of the payer (such as his means, whether he was insured, the importance of fighting the case and his reasonableness in fighting the case).

In order to render the system compatible with article 6 (and A1P1) of the Convention, all that is required is to read para 11.9 down in accordance with section 3 of the Human Rights Act 1998 so as to provide a system incorporating the provisions set out at (iii) in para 114 above.

The critical point in this case to my mind is that in the system under review some classes of defendant were treated differently from others. Professor Zuckerman put the point with his usual clarity in the third edition of his book on Civil Procedure in 2013 at para 27.315 in the context of his discussion of the decision of the ECtHR in MGN v United Kingdom:

In my opinion those points have great force. As I see it, the system was unfairly discriminatory against some classes of respondent by comparison with others. I can understand that it might be just to introduce some such system where the respondents are part of a class of respondents who are frequent litigators such that a system which provides the rough with the smooth may be justifiable. An example would be respondents who have relevant liability insurance because liability insurers are not concerned so much with the result in the particular case as with balancing the premium income over a long period. Similar considerations may apply to commercial entities of which the same may be said. They may also apply to organs of the state. Although (for the most part) they do not have the protection of the Human Rights Act, in the nature of things they are respondents in many classes of case. Save in such cases, it seems to me to be discriminatory and disproportionate to burden uninsured respondents with costs which vastly exceed the fair and reasonable costs incurred by the claimant in order to encourage solicitors to act for other appellants against other respondents against whom the claims may fail.

It is a striking feature of a CFA that it is available to rich as well as poor appellants, as the House of Lords held in Campbell v Mirror Group Newspapers Ltd [2005] UKHL 61, [2005] 1 WLR 3394. So, when it was made possible to recover the success fee (and ATE premium) from respondents, it was not only a means of providing access to justice (ie through lawyers) for those who could not otherwise afford it but also a risk free means of providing access to lawyers for those who could afford to fund it in other ways. There are cases in which appellants are richer than respondents. The claimant always had a choice about whether to go to court, whereas once that choice was made, the defendant had no choice not to take part. He must give in, negotiate or fight. Moreover, if the claimant had ATE insurance, which meant that his chances were better (usually substantially better) than evens, the defendant would have little or no prospect of obtaining ATE insurance because his chances would be unlikely to be so rated.

The facts of this case bear much of this out, as can be seen from Lord Neuberger’s description of the costs position in this case, reported as Coventry v Lawrence (No 2) [2014] UKSC 46, at paras 32 to 39. He said this at paras 32 to 34:

Lord Neuberger (with whom Lord Sumption and I agreed) then, in paras 35 and 36, expressed grave concern about the base costs in a case like this. He added:

Lord Neuberger then briefly summarised in para 38 the case which Mr McCracken QC has advanced before us. I appreciate that the majority do not accept the respondents’ case as summarised above. However, as already stated, my own view is that the system described above, as applied to respondents such as the respondents in this case, is discriminatory, disproportionate and unfair and infringes their right to a fair trial under article 6. In my opinion it also breaches their rights under A1P1, although I do not think that adds anything significant to their case under article 6. In these circumstances I am not surprised that Sir Rupert was critical of the system in his Review and or that Parliament has now changed the law.

Nor am I surprised that there has been significant extra-judicial adverse comment about the scheme. In particular, Sir Anthony May, who was then President of the Queen’s Bench Division, said this in Cardiff on 19 June 2009 in a passage cited by Sir Rupert Jackson in his final Report at pp 97-98:

The decision of the ECtHR in MGNv United Kingdom is to my mind of considerable importance. Lord Neuberger and Lord Dyson have set out the facts at para 50 above. As they observe, in defending the CFA scheme, the UK Government advanced arguments similar to those that have been advanced by the Secretary of State (as well as by the appellants and some of the interveners) in the present case. The ECtHR held that the requirement to pay the success fees constituted an interference with the defendant’s article 10 rights. The central issue was whether the UK authorities had struck a “fair balance” between freedom of expression protected by article 10 and an individual’s right of access to court protected by article 6 (para 199).

At paras 206 to 217 the ECtHR accepted that the scheme suffered from the four flaws identified in the Jackson report and set out above. At para 51 above Lord Neuberger and Lord Dyson correctly recognise that these flaws were of general application and not confined to defamation and breach of privacy litigation. In this connection they quote the passage from para 217 set out at para 111 above.

I recognise the point made in para 52 above that the context in which the ECtHR made those criticisms was its concern about the effect of the scheme in defamation and privacy cases: see paras 211 to 215. I also recognise that the right of freedom of expression is always given particular weight by the ECtHR and (as the ECtHR said at para 201) that the most careful scrutiny is called for when measures are capable of discouraging the participation of the press in debates over matters of legitimate public concern. I further recognise that it was in that context that the ECtHR concluded that a fair balance had not been struck between the article 10 rights of defendant publishers and the article 6 rights of appellants who allege defamation or breach of privacy.

The question is whether the same applies to the relative rights of appellants seeking access to justice and a fair trial under article 6 and those of respondents seeking a fair trial under article 6 and the recognition of their rights under A1P1. I respectfully disagree that the balance to be struck in this class of case is of a wholly different character. As Lord Mance says in para 102, in para 79 Lord Neuberger and Lord Dyson discount any suggestion that the scheme in force from 1999 to 2013 was incompatible because it did not take account of the paying party’s financial circumstances. They say that the financial position of the paying party has never been a relevant factor in determining the assessment of reasonable and proportionate costs. However, as Lord Mance points out, in para 83 they accept that, in a number of individual cases, it might be said to have interfered with a defendant’s right of access to justice. Nonetheless, they conclude that, viewed as a whole, it was rational and coherent, and not incompatible with either article 6 or A1P1.

Save that I would go further, I agree with Lord Mance at para 103 that, while freedom of expression is a particularly powerful interest under the Convention, the interest of any defendant in being able to defend himself or itself in litigation, at a reasonable and proportionate cost is also one of some weight; and it certainly engages (as he understands Lord Neuberger and Lord Dyson to accept) a balancing exercise, when set against the countervailing interest that appellants should have the access to justice which the system was designed to give. The respect in which I would go further is that it appears to me that, just as a claimant is entitled to a fair trial, so too is a defendant. It is unfairly to diminish that right to say that it is merely entitled to some weight. It is the duty of the court to ensure that both parties have a fair trial.

As Lord Mance says at para 104, the strength of the present respondents’ case lies in their claim to be individuals or small undertakings carrying on modest businesses without insurance and faced with one-off litigation, which has involved them in eye-catchingly large costs exposure. I agree that precisely how compelling this claim is, as Lord Neuberger and Lord Dyson note (para 85), untested. However, as I understand it, their case has been rejected on the basis that is bound to fail.

It is true that previous challenges to the scheme have failed but the points now taken were not determined in any of them. To my mind, so far as it applies to the class of defendant concerned in this case, the scheme is discriminatory and disproportionate and disregards their rights. So far as I can see, the Government at no stage considered the plight of respondents such as these. In MGN v United Kingdom, having set out the facts, the ECtHR reached these conclusions in paras 219 and 220:

I agree with Lord Neuberger and Lord Dyson (at para 53) that, of the four flaws identified by Sir Rupert Jackson, it is the third flaw that lies at the heart of this case. One way of describing it is as imposing a costs burden on opposing parties which is excessive and in some cases amounts to a denial of justice. In para 54 they set out what Lord Neuberger had described as four “unique and regrettable features” of the scheme on which the respondents also rely in support of their case. As they say, to some extent these features overlap with the four flaws. The first feature was that appellants had no interest in the level of fees which they agreed to pay their lawyers. The second was that in many cases unsuccessful respondents found themselves paying, in addition to their own costs, three times the appellants’ “real” costs. The third was that proportionality was excluded from consideration in relation to the recovery of the success fee or ATE premium. The fourth was that the stronger the respondents’ case, the greater their liability costs would be if they lost, since the size of the success fee and the premium should have reflected the appellants’ prospects of success.

I accept that the question is not whether the system was unfair or had flaws. It is whether it was a disproportionate way of achieving the legitimate aim. In my opinion, it plainly was because it did not treat all respondents in the same way but chose a particular class of respondents on whom to impose liabilities far beyond the bounds of what was reasonable or proportionate.

As Lord Sumption observed in Bank Mellat (No 2) at para 25, “a measure may respond to a real problem but nevertheless be irrational or disproportionate by reason of its being discriminatory in some respect that is incapable of objective justification”. In my opinion this is such a case.

The majority place weight on what they call the “legitimate expectation” of litigants and lawyers that courts will uphold the legality of the costs regime around which they have contracted. They say that, even if the system is incompatible with article 6 and A1P1, litigants’ and lawyers’ legitimate expectation that successful appellants would receive a costs order covering the success fee and ATE insurance premium should be weighed in the balance so that a remedy of reading down the provisions so as to be compatible with respondents’ Convention rights is not appropriate: Lord Neuberger and Lord Dyson, para 89; Lord Mance, para 106. The appellants before us argued that their expectation that they would receive such a costs order was a “legitimate expectation” which was a protected possession under A1P1. Presumably (though they do not say so) the members of the majority intend to use the term “legitimate expectation” in that sense.

The question is what order the court must make. To my mind the answer is clear. By section 6 of the Human Rights Act, the court must not act incompatibly with a Convention right. It is the court’s duty to balance these competing rights as part of the balancing exercise and determine the proper way forward. Insofar as the expectations of litigants and their lawyers are relevant to the article 6 balancing exercise, I consider that they are one factor only and cannot render proportionate the discriminatory treatment of the particular classes of defendant I have already discussed.

As the majority observe, legitimate expectation may be relevant on the issue of remedy. It strikes me that it may be relevant in this way. When deciding what order to make in light of the facts at the present day, considerations of legitimate expectation should cause the court to go back to the balancing exercise it has already undertaken when evaluating the scheme itself. The court would then take account of the parties’ legitimate expectations in a new balancing exercise and decide whether to maintain its previous view. Considerations tending to the conclusion that that the scheme is still incompatible might include the following: (i) the ECtHR was untroubled by this concern in MGN v United Kingdom; (ii) the legitimacy of an expectation that the Government would enforce a scheme which breaches the Convention rights of others must be very limited indeed; (iii) the application of A1P1 to the question of whether a superior court should feel able to disturb settled case-law is an area where the court should act with great caution; (iv) as a matter of fact the scheme was heavily criticised from its inception, so that litigants must have known that there was an issue under article 6; (v) since it is the court’s duty to act compatibly with the Convention, the court should not make an excessive costs order which directly infringes a respondent’s rights, even though this may have a deleterious effect on the claimant; (vi) it may not have such an effect because the solicitors may not enforce their rights against the appellants; and (vii) the appellants may have rights against the United Kingdom. If all these points are borne in mind when striking a balance between the rights of the parties, the correct conclusion is that respondents’ article 6 rights are still breached because leaving the scheme in place is still not proportionate. I would add by way of footnote that it is surely relevant to observe that a respondent is in no way responsible for any legitimate expectation that a claimant might have had.

The question of what precise order the court should make then arises. It appears to me to be at least arguable that CPD 11.9 can be read down as submitted on behalf of the respondents. However, I recognise that Lord Neuberger and Lord Dyson say that it cannot. If that is correct, the appropriate remedy would to my mind be to strike down CPD 11.9, which is not of course primary (or even secondary) legislation, as being contrary to the Convention. There is scope for further argument on these questions and, if it were relevant I would be willing to receive further argument on them, but I recognise that in present circumstances there would be no point in such argument because I am in a minority.

For all these reasons, in respectful disagreement with the majority, I would allow the appeal.