Diag Human SE & Anor v Volterra Fietta

Historically, maintenance and champerty were crimes and could give rise to civil liability. That changed in 1967 when section 13(1)(a) of the Criminal Law Act 1967 abolished the criminal offences of maintenance and champerty and section 14(1) provided that no person should be liable in tort for any conduct on account of its being maintenance or champerty. Section 14(2), however, provided that the abolition of criminal and civil liability for maintenance and champerty did not affect any rule of law as to the cases in which a contract was to be treated as “contrary to public policy or otherwise illegal.”

Until 1990 it was axiomatic that contingency fee agreements entered into by lawyers with their clients were illegal and contrary to public policy, because they gave lawyers an interest in the outcome of the litigation that could create conflicts between the lawyer’s interests and those of their client. That axiomatic position was authoritatively restated by Buckley LJ in Wallersteiner v Moir (No 2) [1975] 1QB 373 at 401D-E.

Because one of the foundations for the public policy rule was the need to avoid conflicts of interest, what is prohibited at common law is not merely entering into an agreement tainted by maintenance or champerty but also acting in accordance with such an agreement. “If anything is against public policy it is the solicitor undertaking or continuing to act for a party in litigation in circumstances where the solicitor stands to gain more from the action if it is won than if it is lost.”: see Awwad v Geraghty & Co [2001] QB 570, 594B-C per Schiemann LJ (emphasis added).

Section 58 of the 1990 Act prevents a limited class of agreements from being unenforceable. As relevant to the present proceedings, section 58 provides:

Conditional fee agreements.

(1)

A conditional fee agreement which satisfies all of the conditions applicable to it by virtue of this section shall not be unenforceable by reason only of its being a conditional fee agreement; but … any other conditional fee agreement shall be unenforceable.

(2)

For the purposes of this section … —

(a)

a conditional fee agreement is an agreement with a person providing advocacy or litigation services which provides for his fees and expenses, or any part of them, to be payable only in specified circumstances;

(b)

a conditional fee agreement provides for a success fee if it provides for the amount of any fees to which it applies to be increased, in specified circumstances, above the amount which would be payable if it were not payable only in specified circumstances; and

(c)

references to a success fee, in relation to a conditional fee agreement, are to the amount of the increase.

(3)

The following conditions are applicable to every conditional fee agreement —

(a)

it must be in writing;

(b)

…

(c)

it must comply with such requirements (if any) as may be prescribed by the Lord Chancellor.

(4)

The following further conditions are applicable to a conditional fee agreement which provides for a success fee —

(a)

…

(b)

it must state the percentage by which the amount of the fees which would be payable if it were not a conditional fee agreement is to be increased; and

(c)

that percentage must not exceed the percentage specified in relation to the description of proceedings to which the agreement relates by order made by the Lord Chancellor.

The following points may be noted:

The solicitors repeatedly emphasised the seriousness of the financial consequences for them if their appeal were to be rejected. This is not a new submission in cases involving challenges to CFAs. It was answered by Dyson LJ giving the judgment of the Court in Garrett v Halton BC [2006] EWCA Civ 1017, [2007] 1 WLR 554 at [27]-[30] where he said:

“27.

… The starting point must be the language of section 58(1) and (3) of the 1990 Act. It is clear and uncompromising: if one or more of the applicable conditions is not satisfied, then the CFA is unenforceable. Parliament could have adopted a different model. It could, for example, have provided that where an applicable condition is not satisfied, the CFA will only be enforceable with the permission of the court or upon such terms as the court thinks fit. There is nothing inherently improbable in a statutory scheme which provides that, if the applicable conditions are not satisfied, the CFA shall be unenforceable with the consequence that the solicitor will not be entitled to payment for his services. Such a scheme can yield harsh results in certain circumstances, especially if the client has not suffered any actual loss as a result of the breach. It can also produce results which, at first sight, may seem odd: … . But the scheme is designed to protect clients and to encourage solicitors to comply with detailed statutory requirements which are clearly intended to achieve that purpose. The fact that it may produce harsh or surprising results in individual cases is not necessarily a good reason for construing the statutory provisions in such a way as will avoid such results.

…

30 … To use the words of Lord Nicholls, Parliament was painting with a broad brush. It must be taken to have deliberately decided not to distinguish between cases of non-compliance which are innocent and those which are negligent or committed in bad faith, nor between those which cause prejudice (in the sense of actual loss) and those which do not. It would have been open to Parliament to distinguish between such cases, but it chose not to do so. The conditions stated in section 58(3)(c) and in particular the requirements prescribed in the 2000 Regulations are for the protection of solicitors’ clients. Parliament considered that the need to safeguard the interests of clients was so important that it should be secured by providing that, if any of the conditions were not satisfied, the CFA would not be enforceable and the solicitor would not be paid. To use the words of Lord Nicholls again, this is an approach of punishing solicitors pour encourager les autres. Such a policy is tough, but it is not irrational. The public interest in protecting solicitors’ clients required that the satisfaction of the statutory conditions was an essential prerequisite to the enforcement of CFAs.”

Underlying most of the solicitors’ submissions was the contention that public policy has changed and that we should recognise that change. Exactly what was the change for which the solicitors contend is not clear to me, save that it should have the effect of allowing them to be paid to at least some extent by one means or another. However, the contention must be rejected on high authority and because there is no material on the basis of which we could properly and confidently assert what any new and adjusted public policy might be.

Turning first to authority, in Awwad at 593E-F Schiemann LJ (with whom Lord Bingham LCJ agreed) expressed reluctance to develop the common law at a time when Parliament was addressing the same problems:

“It is clear from the careful formulation of the statutes and regulations that Parliament did not wish to abandon regulation altogether and wished to move forward gradually. I see no reason to suppose that Parliament foresaw significant parallel judicial developments of the law.”

To the same effect, at 600D-E May LJ (with whom Lord Bingham also agreed) said:

“In my judgment, where Parliament has, by what are now (with section 27 of the Access to Justice Act 1999) successive enactments, modified the law by which any arrangement to receive a contingency fee was impermissible, there is no present room for the court, by an application of what is perceived to be public policy, to go beyond that which Parliament has provided.”

In Sibthorpe v Southwark LBC [2011] EWCA Civ 25. [2011] 1 WLR 2111, Lord Neuberger MR (with whom Lloyd and Gross LJJ agreed) reaffirmed the principle at [40]:

“In my judgment, when it comes to agreements involving those who conduct litigation or provide advocacy services, the common law of champerty remains substantially as it was described and discussed in Wallersteiner v Moir (No 2) … and Awwad’s case … . This is for two main reasons…. . The second reason, articulated in Awwad’s case [at] 593, 600, by Schiemann and May LJJ, is that, in section 58 of the 1990 Act (as amended) the legislature has laid down the rules as to which previously champertous agreements may be entered into by those conducting litigation and those providing advocacy services, and which may not.

Most recently, these principles were reiterated and reaffirmed by Arnold LJ (with whom Phillips and Simler LJJ agreed) in Farrar v CANDEY Ltd [2022] EWCA Civ 295 at [51]-[52].

Even if it were open to us in the light of these authoritative statements to enter the choppy waters of public policy, which in my view it is not, I can detect nothing other than the solicitors’ speculative hope upon which it would be possible to base an assertion that public policy has changed since either Wallersteiner v Moir (No 2) or the more recent cases of Awwad, Sibthorpe and Farrar which reaffirmed it. Any further modification of public policy in this area is for Parliament to establish and define: it is not for this Court to attempt. This must be borne in mind when considering the issue of severance, not least because many of the authorities about severance are cases about restrictive covenants in contracts of employment, where the nature and scope of public policy are not the same as in the field of champertous retainers of solicitors.

The solicitors submit that severing paragraphs 3 and following of the side letter can be achieved and that, in consequence, they should be entitled to recover their fees for work done after 6 September 2017 at the discounted rate of 70% pursuant to paragraph 2 of the side letter. It is common ground that there is a three-stage test to be applied, and that the first two stages of the test are capable of being satisfied in the present case. The clients, however, submit that the third stage cannot be satisfied and that, in any event, it would be contrary to public policy for the solicitors to recover fees pursuant to the September 2017 Agreement after severance of the offending provisions.

It is not necessary to examine the origins and development of the three-stage test for severance. It is sufficient that in Beckett Investment Management Group Ltd v Hall [2007] EWCA Civ 613, [2007] ICR 1539 at [40] Maurice Kay LJ (with whom Sir Anthony Clark MR and Carnwath LJ agreed) approved the test in the following terms:

“… a contract which contains an unenforceable provision nevertheless remains effective after the removal or severance of that provision if the following conditions are satisfied: 1 The unenforceable provision is capable of being removed without the necessity of adding to or modifying the wording of what remains. 2 The remaining terms continue to be supported by adequate consideration. 3 The removal of the unenforceable provision does not so change the character of the contract that it becomes ‘not the sort of contract that the parties entered into at all’.”

To understand the application of these principles in Beckett, it is necessary to outline the facts briefly. The two defendants were employed by Beckett pursuant to contracts which included covenants that restrained them, for 12 months after the termination of their contracts of employment, from supplying advice to any “client” of the company of a type provided by the company in the ordinary course of business. The definition of “client” included not only those persons, firms, companies or organisations with whom the defendants had dealt in the course of their employment for a specified period, but also any individuals acting in a representative capacity on behalf of a firm, company or organisation with whom they had so dealt. The Court of Appeal held that these covenants were unreasonable and unenforceable to the extent that they deemed to be Beckett’s clients those individuals who had only dealt with the defendants in a representative capacity on behalf of firms, companies or organisations. Applying the three stage test, at [44] Maurice Kay LJ said:

“… I have no difficulty in coming to the conclusion that clause 17.1.3 is susceptible to severance. I would sever so as to leave only the primary wording that I set out in para 10 above. That can be done without the necessity of adding to or modifying the remaining language. The remaining restriction continues to be supported by consideration. Moreover the severed part does not change the character of the surviving part to the extent that the character of the contract becomes “not the sort of contract that the parties entered into at all”.”

To my mind it seems clear that in the last sentence of this passage Maurice Kay LJ was referring to the fact that, after severance, the surviving part of the contract as a whole was of the same character as the contract as a whole before severance: it had the character of a contract of employment with restrictions on post-termination employment elsewhere. All that had changed was that the scope of the restrictions had been reduced. It was essentially the same sort of contract that the parties had entered into originally. Neither the formulation of the test nor its application to the facts as set out above lends any support to a submission that what must be compared is the character of what would become the surviving part before and after severance.

The Beckett three-stage test was approved in EgonZehnder Ltd v Tillman [2019] UKSC 32, [2020] AC 154, which was another case involving covenants in restraint of trade in an employment contract after the employee’s employment had finished. The clause that was the subject of the appeal provided that she could not “directly or indirectly engage or be concerned or interested in any business carried on in competition with any of the businesses” of the claimant for a six-month period after the termination of her contract. The Supreme Court, in a judgment given by Lord Wilson (with whom the other Justices agreed) held that, on the assumption that the word “interested” in the restrictive covenant purported to restrain the claimant from holding shares in the specified businesses, it was an unreasonable restraint of trade. The Court endorsed the Beckett three stage test and held that severing the words “or interested” would not generate any major change in the overall effect of the restraints. Accordingly, severance was ordered. In the course of his judgment, Lord Wilson made observations on all three stages of the test.

In addressing the first stage, Lord Wilson recognised that the requirement that it should be possible to take a “blue pencil” to the text without adding to or modifying the wording of what remains can act capriciously and arbitrarily. However, he regarded the requirement as settled law and inherent in the word “severance” itself. He regarded it as a significant but “appropriate brake on the ability of employers to secure severance of an unreasonable restraint customarily devised by themselves.”

Turning to the second stage, Lord Wilson said at [86]:

“The second criterion is that “the remaining terms continue to be supported by adequate consideration”. It goes without saying that an employer who sues on a covenant made otherwise than under seal must show that he provided consideration for it. But why is it said to be a prerequisite of his ability to sever? The answer is surely to be found in the unusual circumstances of the Sadler and Marshall cases, which generated the criteria adopted in the Beckett case. In those two cases it was the claimant employee who secured severance of unreasonable obligations cast by the contract upon himself. In that situation the court needed to satisfy itself (and in each case it did so) that, were his unreasonable obligation to be removed, there would nevertheless remain consideration passing from him under the contract such as would support the obligation which he was seeking to enforce. In the usual post-employment situation, however, the need to do so does not arise. A claimant employer who asks the court to sever and remove part of a covenant made by the defendant employee is in no way proposing to diminish the consideration passing from himself under the contract such as is necessary to support the obligation which he seeks to enforce. In the usual situation the second requirement can be ignored.” [Emphasis added].

The present case is readily distinguishable from the first of the highlighted examples: the solicitors are seeking to enforce the client’s obligation to pay the 70% discounted fees in any event while securing the severance of further obligations to which the clients (not the solicitors) are subject. The solicitors’ performance under the contract is unaffected and is capable of providing the consideration for the client’s obligation to pay the 70% discounted fees. Nor is the adequacy of this consideration affected by the provision in paragraph 3 of the side letter that the conditional fees set out in paragraphs 5 and 7 were in consideration of the discount to 70% referred to in paragraph 2: see [10] above. That provision might have been relevant if what was being severed was the discounted fee provision in paragraph 2 and it was being submitted that there was therefore no consideration for the conditional fees set out in paragraphs 5 and 7; but it is not relevant to the application of the second stage criterion in the present case. For this reason I do not consider it necessary to refer in any greater detail to either Sadler v Imperial Life Assurance Co of Canada Ltd [1998] IRLR 388or Marshall v N.M Financial Management Ltd [1995] 1 WLR 1461, [1997] 1 WLR 1527 (CA).

Turning to the third stage, at [87]-[88] Lord Wilson endorsed and expanded on the formulation in Beckett:

“87 The third criterion is that “the removal of the unenforceable provision does not so change the character of the contract that it becomes ‘not the sort of contract that the parties entered into at all’”. This is the crucial criterion … In my view this third criterion was rightly imported into the general jurisprudence by the Beckett case … . But I suggest, with respect, that the criterion would better be expressed as being whether removal of the provision would not generate any major change in the overall effect of all the post-employment restraints in the contract. It is for the employer to establish that its removal would not do so. The focus is on the legal effect of the restraints, which will remain constant, not on their perhaps changing significance for the parties and in particular for the employee.

88 Application of the severance principle to Ms Tillman’s restraint covenants now becomes straightforward. First, the words “or interested” are capable of being removed from the non-competition covenant without the need to add to or modify the wording of the remainder. And, second, removal of the prohibition against her being “interested” would not generate any major change in the overall effect of the restraints. So those words should be severed and removed.”

Lord Wilson’s partial reformulation in [87] did not involve rejection of the criterion that he had just said was rightly imported by the Beckett case. Taken in conjunction with the observations of Maurice Kay LJ at [44] of Beckett (set out at [29] above), I understand Lord Wilson to mean that it is necessary to look at the overall effect of the post-employment restraints in the contract, with and without the severed part, to see whether severance would change the character of the contract so that it becomes “not the sort of contract that the parties entered into at all.” In that context, his conclusion that severance of the word “interested” would not generate any major change in the overall effect of the restraints had the consequence that severance would not make the contract as a whole “not the sort of contract the parties entered into at all.” His approach to the legal effect of the restraints is most apposite where, as in Beckett and Tillman, what is proposed to be severed is one of a number of restraints. However, the question to be answered in any given case is whether the proposed severance would change the character of the contract as a whole so that, if severance was implemented, it would cease to be “the sort of contract” into which the parties had originally entered.

Tillman provides no support for the submission, somewhat tentatively made by the solicitors, that what must be compared before and after severance in the present case is only the terms of the contract which are not proposed to be severed. I can see no logic or purpose to support such an approach. It will be a rare case where severance of part of a contract will change the effect of the unchanged terms that are left behind when they are viewed in isolation; and, since the purpose of the exercise is to see whether the contract after severance is not the sort of contract that the parties originally entered into, what must be compared is (a) the pre-severance contract and (b) the proposed post-severance contract.

The mere fact that there are some provisions (e.g. here the provision for discounted fees) that are common denominators in both pre- and post-severance contracts does not necessarily or even probably mean that severance should be permitted. That does not mean that the existence of common denominators is necessarily irrelevant. The greater the overlap between the terms of the pre- and post-severance contracts, the more likely the conclusion may be that the nature of the contract has not changed; but this will require an exercise of judgment based on the substance of the overlap in the specific case rather than anything else.

Before Costs Judge Rowley, the clients relied upon a passage in Freshasia Foods Ltd v Lu [2019] F.S.R. 18, [2018] EWHC 3644 (Ch), a decision of Mr Daniel Alexander QC sitting as a Deputy Judge, which involved the proposed severance of post-termination restrictions on employment. At [53] the Deputy Judge looked favourably on a submission that the Court should look at substance rather than form. He contrasted a case where “relatively minor and reasonably isolable surgery can be performed”, which would indicate that severance was appropriate, with a case where “a great deal of slimming is required … or the effect is a radical change”, which would be regarded as not the sort of contract that the parties entered into and severance should not be done. It follows from what I have said that I endorse what the Deputy Judge said about concentrating on substance not form; so too did the Supreme Court at [78] of Tillman where Lord Wilson referred to the value of the Deputy Judge’s observations, specifically citing paragraph [53] of his judgment. I would also agree that it may be helpful as part of the evaluative judgment required by the third stage of the test to consider how radical is the surgery (or slimming) that is being proposed. That said, as with the degree of overlap to which I have just referred, it is unlikely that a purely mathematical or mechanistic comparison will be decisive.

Applying these principles, I would hold that to implement the severance proposed by the solicitors would fundamentally change the nature of the contract so that, upon severance, it would cease to be the sort of contract into which the parties had originally entered. The September 2017 Agreement, whether it was a new contract or a variation of the February 2017 Agreement, was a CFA with a substantial proportion of the solicitor’s proposed remuneration being conditional upon the contingencies outlined in paragraph 3, that being the stated consideration for the discounting of the solicitors’ normal fees under paragraph 2 of the side letter. Upon severance, it would become a conventional retainer providing simply for the solicitors to charge at a discounted rate, with no conditional element at all. The fact that severance would remove the stated consideration for the solicitors’ agreement to discount their fees emphasises the difference in the nature of the contract before and after severance – before severance the solicitors discounted their fees in return for the prospect of success fees; after severance they discounted their fees for no consideration. The present case cannot be moulded so as to be analogous to either Beckett or Tillman: there is no question of severing one part of the provisions for conditional fees and leaving the rest in place since that would be an exercise in futility: the character of the contract would remain that it was an unenforceable CFA.

The solicitors seek to avoid this conclusion by relying upon Garnat Trading & Shipping (Singapore) Pte Ltd v Thomas Cooper (a Firm) [2016] EWHC 18 (Ch), [2016] 1 Costs LR 45, a decision of Mr Edward Murray sitting as a Deputy High Court Judge. The solicitors submit that the case is indistinguishable from the present case and that, though not bound to do so, we should follow it. In Garnat the solicitors and their client initially entered into a retainer in 2007 which provided for the solicitors’ professional fees to be charged on the basis of hourly billing. After success in the Commercial Court and when an appeal was pending, the retainer was varied in April 2011 exclusively to make specific provision for the costs of work carried out in relation to the appeal. The agreement as varied was unenforceable for want of compliance with section 58 of the 1990 Act. The variation did not affect the basis upon which the solicitors should charge for work other than their conduct of the appeal and, but for the variation, the original retainer would have covered the solicitors’ work on the appeal: see [22]. As the Deputy Judge put it at [24]: “the natural construction of the 12 April 2011 Agreement is that it is intended to vary the Retainer as to the fees payable by the [client] in relation to the appeal. Otherwise, the [original] Retainer continues to apply to issues such as staffing, money laundering compliance, complaints handling, disbursements and so on.”

The question for the Court was therefore “whether the Retainer as a whole is unenforceable by virtue of the variation effected by the 12 April 2011 Agreement or whether the provisions of the Retainer as varied that constitute the unenforceable conditional fee agreement may be severed.” Relying on Beckett the Deputy Judge held that the offending provisions were severable. The kernel of his reasoning was at [28], where he said:

“ 28. Thirdly, the removal of the unenforceable provisions of the 12 April 2011 Agreement does not change the character of the Retainer or cause it to become “not the sort of contract that the parties entered into at all”. It is unenforceable in respect of work carried out by the defendant relating to the appeal, but it continues to apply, without amendment, to work carried out within the scope of the Retainer that does not relate to the appeal. There is independent and unimpeachable consideration for this work, namely, the promise to pay the defendant on the hourly basis set out in the Retainer. The 12 April 2011 Agreement cannot, in my view, be construed as having amended the basis for charging in the Retainer in relation to work other than that relating to the appeal.”

The solicitors submit that we should follow decision in Garnat because there is no distinction between charging fees for part of one’s work (as in Garnat) and charging partial fees for all of one’s work (as in the provision for discounted fees in the present case).

I am unable to accept the proposed analogy or that we should reach a conclusion similar to that reached in Garnat. The critical distinction is to be found at [28] of Garnat, which contemplates work being carried out after the date of the 12 April 2011 Agreement which did not fall within the scope of variation because the variation applied only to work on the appeal. Any other work and the terms on which it was done were unaffected by the 12 April 2011 Agreement, which gave rise to the CFA but only applied to work done on the Appeal.  Once the “blue pencil” was applied to the offending provisions in Garnat (which all related to the work done on the Appeal and nothing else) the parties were left with the original Retainer, which continued to govern all other work and was unchanged: the original terms (and the consideration provided for them) remained unaffected by the severance. As the Deputy Judge put it at [28], the Retainer continued to apply, without amendment, to work carried out that did not relate to the appeal. The Deputy Judge was therefore able to treat the 12 April 2011 Agreement as not affecting the Retainer except in relation to work on the appeal.

In the present case, by contrast, the February 2017 Agreement as varied covers all work from 6 September 2017 and it is a CFA.  If the “blue pencil” is applied to the paragraphs that provide for a success fee, the agreement between the parties in relation to all work carried out by the solicitors is converted from being a CFA into an agreement for payment on a conventional (if discounted) hourly basis.  On any view, this is a major change in the overall effect of the provisions as they existed before applying the “blue pencil”; alternatively it may be said that the agreement after the “blue-pencilling” (a standard contract for the payment of fees on an hourly basis) is not the sort of contract that the parties entered into (a CFA) at all.

It was not and could not be suggested that after “blue-pencilling” in Garnat, the solicitors were left with the original Retainer and could therefore have charged their full hourly rate for their work on the appeal.  They recovered nothing for their work on the appeal at all.

Before Foster J and before us the solicitors relied upon the decision of this Court in Lexlaw Ltd v Zuberi (Bar Council Intervening) [2021] 1 WLR 2729, [2021] EWCA Civ 16, which was decided after the decision of Costs Judge Rowley in this case. The issue in Zuberi was stated with typical succinctness by Lewison LJ at [1]:

“A client enters into a contract of retainer with solicitors to prosecute a claim. The contract provides that in the event of success the solicitors will be entitled to a share of the recoveries. The client achieves success by means of a settlement of the claim; and the solicitors claim their share. But the contract also contains a clause which says that if the client terminates the retainer prematurely (which she did not), she must pay the solicitors’ normal fees and disbursements. Does the existence of that clause invalidate the whole contract?”

Lewison LJ rejected the submission that if a contract of retainer contains any provision which entitles the lawyer to a share of recoveries, then the whole contract of retainer is a “damages-based agreement” (or “DBA”). Instead, he adopted the alternative interpretation, namely that “if a contract of retainer contains a provision which entitles a lawyer to a share of recoveries; but also contains other provisions which provide for payment on a different basis, or other terms which do not deal with payment at all, only those provisions in the contract of retainer which deal with payment out of recoveries amount to the DBA.”: see [33]. Coulson LJ agreed with this interpretation: see [77].

Newey LJ, concurring in the result, adopted a different approach, namely that the relevant regulation had no application to provisions in a contract of retainer which dealt with termination of the retainer; and that, if the termination clause was part of the DBA between the parties, it did not fall foul of the Regulations: see [71]. Coulson LJ agreed with this interpretation if the correct view were that Lewison LJ’s interpretation of what constitutes a DBA were too narrow: see [78].

Early in his judgment, Lewison LJ expressed the view that the conditions for severance of the allegedly offending clause were “amply fulfilled”: see [8]. He recognised, however, that the question of severance was not in issue in Zuberi: see [49].

The solicitors seek to build on Lewison LJ’s narrow approach to what constitutes a DBA by analogy. They submit that, where a retainer contains provisions which entitle a solicitor to payment of part of their fees in specified conditional circumstances but also contains other provisions which provide for payment for other parts on a different basis, only those provisions of the retainer which deal with conditional payment strictly amount to “the CFA” for the purposes of the 1990 Act. Pursuing this line of submission, they submit that severing the offending clauses in the present case would “remove all of the CFA, just as was done in Garnat.”

This submission was rejected by Foster J. She distinguished Zuberi on the facts, there being no “termination” (or equivalent) provision in the present case; and on the basis of the issue that was raised in Zuberi, which was simply whether Parliament intended by the relevant regulation in Zuberi to make the agreement in that case unenforceable as a whole because it included an obligation to pay incurred time costs and expenses if the client exercised a contractual right to terminate; and on the basis that different public policy considerations applied which tended to favour enforceability in Zuberi and which were not transferable to the proscription against continuing to act for a client under a conditional fee arrangement.

In my judgment, Foster J was right to reject the solicitors’ submissions based on Zuberi, essentially for the reasons that she gave. The starting point is the terms of section 58(2)(a), which I have set out above at [19] above. As I have already indicated, that definition of a CFA precludes splitting off the provisions for payment of the solicitor’s discounted fees and treating them as not forming part of the (unenforceable) CFA. I accept the client’s submission that the discounted fee provisions which the solicitors seek to enforce are part of the core agreement that make the September 2017 Agreement a CFA. Second, the provision for discounted fees is not analogous to the “termination” provision in Zuberi. Third, the considerations of public policy which supported Lewison LJ’s narrow construction of the meaning of a DBA are absent in a case involving CFAs such as the present.

Before us, the solicitors contend that the Courts below erred in four main respects.

First, the solicitors assert that the Courts below wrongly “approach[ed] the third limb as a question of ‘how much text’ of the contract was being severed”. It is said that the Costs Judge accepted a submission that the removal of “19 of 21” paragraphs of the Side Letter was indicative of the severed contract being “different” from “the one entered into on 6 September.” They submit that the third stage of the test is a question of substance, not form and that it cannot be decided by a mathematical analysis of the number of terms to be removed.

I agree that the third limb is a question of substance not form and that it is unlikely to be decided by a mathematical and mechanistic analysis of the number of terms to be removed: see [38]-[39] above. However, the Costs Judge was astute not to fall into a mathematical or mechanistic trap as suggested by the solicitors’ submission.

At [62]-[63] of his judgment, the Costs Judge accepted the client’s submissions based on Freshasia, stating that “purely numerically, a considerable amount of slimming is proposed by [the solicitors] to save the contract here.” However, it is plain on a fair reading of his judgment that he took the need for “a considerable amount of slimming” as one indicator of the substance of the change that would be effected by severance and not simply the application of some mechanistic mathematical formula. What immediately followed is to the same effect and concentrates on substance: “Moreover it removes all of the conditionality in the agreement.” I do not accept that the Costs Judge’s reference to and reliance upon the need for a considerable amount of slimming or the removal of all of the conditionality in the agreement shows that he was applying the wrong test. Far from it, these were simply strands that contributed to his justifiable and justified answer to the question of substance he was addressing. His reasoning was adopted by Foster J who concluded that “the nature of the agreement between the parties evidenced by the Side Letter was materially different from that evidenced by the [February 2017 Agreement]”, pointing to the fact that “the central features of the [September 2017 Agreement] were the conditional elements.” I agree.

Second, the solicitors submit that the Costs Judge and Foster J erred in taking into account the solicitors’ evidence that they would only have given the 30% discount because of the possibility of a substantial upside in the event of success. Once again, the Costs Judge did not put this forward as a decisive feature on its own: he regarded it as an indication that the CFA was not the same sort of contract as one under which the solicitors discounted their fees for no return. So did Foster J at [84] of her judgment where she correctly rejected the submission that the evidence was irrelevant.

Third, the solicitors submit that the Courts below focused on the wrong material and that they should have focussed solely on the material that remained, which was unchanged. For the reasons I have set out at [35]-[37] above, this submission involves a misunderstanding of the principles applicable to the third stage test as explained by Beckett and Tillman.

Fourth, the solicitors rely upon Garnat. In my judgment this reliance is misconceived for the reasons I have given at [41]-[44] above.

I therefore reject the criticisms of the judgments below on Ground 1. In my judgment the Courts below were correct to conclude that the third stage of the Beckett test was not satisfied and that, accordingly, severance was not available.

Even if I were wrong in this conclusion, I would hold that severance is precluded as contrary to public policy. The principal effect of severance would be to permit partial enforcement of the unenforceable CFA. As was pointed out during submissions, if the client lost the arbitration, the effect of allowing severance would be that the solicitors would recover precisely the same amount of their fees as if the CFA had been held to be enforceable. That is not, in my view, a tolerable outcome. Nor is it any answer to submit that there is no disadvantage to the client in enforcing the discounted fee element in respect of work carried out for and at the client’s request. The regime imposed by the 1990 Act is concerned withconflicts of interest giving rise to potential harm to clients: see Garrett per Dyson LJ at [38]-[39].

The effect of implementing public policy, as explained by Dyson LJ at [27]-[30] of Garrett, cited at [21] above, is that “if the applicable conditions are not satisfied, the CFA shall be unenforceable with the consequence that the solicitor will not be entitled to payment for his services”.

Awwad, like the present case, involved a CFA of the type that has been described as a “discounted fee agreement”. The solicitor in Awwad agreed to charge the client at her normal rate if the client won the litigation and at a lower rate (£90 per hour) if he lost. The agreement was not sanctioned by the 1990 Act. After acceptance of the opponent’s Part 36 offer, the client declined to pay the solicitor’s bill of costs. The Court of Appeal held that it was contrary to public policy for the solicitor to have acted in pursuance of a fee agreement that was not sanctioned by statute and that the agreement would not be enforced. The solicitors in Awwad, as in this case, submitted that there could be no objection to enforcing the agreement in respect of the discounted element of their fees that were to be payable in any event: see 574F. That argument too was rejected on grounds of public policy: see 594C-E per Schiemann LJ.

It would therefore be contrary both to principle and to authority to allow partial enforcement of the unenforceable CFA in the present case, on grounds of public policy. That precludes any residual arguments in favour of severance.

I would for these reasons dismiss Ground 1 of the appeal.

In the light of the outline of principles set out above, I can state my conclusion on Ground 2 very shortly. It would be contrary to the public policy that forbids partial or total enforcement of the CFA and severance to permit the solicitors to recover on a quantum meruit basis.

Not only is this clear as a matter of principle based on the scope of the public policy prohibition, it would also be contrary to authority. In Awwad the solicitors contended that they should recover fees on the basis of a quantum meruit assessment in respect of services actually rendered: see 574C. That argument was rejected by Schiemann LJ (with whom both Lord Bingham and May LJ expressly agreed on this point) at 596C-E:

“Mr Dutton attempted to make use of that part of the decision in the Mohamed case [2000] 1 WLR 1815 which ruled that the interpreter was entitled to be paid a fair fee for his work as interpreter notwithstanding that his agreement to work as such was part of a champertous agreement which the court refused to enforce. In my judgment this attempt should fail. If the court, for reasons of public policy refuses to enforce an agreement that a solicitor should be paid it must follow that he cannot claim on a quantum meruit. … In the present case, what public policy seeks to prevent is a solicitor continuing to act for a client under a conditional normal fee arrangement. That is what Miss Geraghty did. That is what she wishes to be paid for. Public policy decrees that she should not be paid.”

If further citation of authority is required, it is to be found at [27] of Garrett: see above.

Accordingly I would dismiss Ground 2 of the appeal.

Ground 3: restitution

Mr Bacon KC, who fervently advanced every argument that could possibly be raised on behalf of the solicitors in this appeal, recognised that he would be in difficulty on Ground 3 if he failed on Grounds 1 and 2. I agree.

This case involves preliminary issues arising on an assessment under section 70 of the Solicitor’s Act 1974. So far as material, the section provides:

“(1)

Where before the expiration of one month from the delivery of a solicitor’s bill an application is made by the party chargeable with the bill, the High Court shall, without requiring any sum to be paid into court, order that the bill be assessed and that no action be commenced on the bill until the assessment is completed.

…

(7)

Every order for the assessment of a bill shall require the costs officer to assess not only the bill but also the costs of the assessment and to certify what is due to or by the solicitor in respect of the bill and in respect of the costs of the taxation.

…”

The procedure by which a bill comes to be assessed and, where appropriate, enforced is set out in CPR Part 47 and Practice Directions 46 and 47. The rules make provision for the issuing of interim costs certificates. Paragraph 6.19 of PD46 provides that, after the detailed assessment hearing is concluded the Court will complete the Court copy of the bill so as to show the amount allowed (described in rule CPR 47.17 as a “completed bill”) and determine the result of the cash account. The Court will then award the costs of the detailed assessment hearing in accordance with Section 70(8) of the Solicitors Act 1974, and issue a final costs certificate, which will “include an order to pay the costs to which [the completed bill] relates, unless the Court orders otherwise”: CPR 47.17(5). The final costs certificate will show the amount of any costs which have been agreed between the parties or which have been allowed on detailed assessment: 47PD.16.7. Either an interim or a final costs certificate may be enforced as if it were a judgment for the payment of an amount of money: 47PD.16.12.

The overall operation of this scheme of assessment is not in dispute. The costs officer assessing a bill will allow or disallow disputed items of costs. If at the conclusion of the assessment there is a balance due and owing to the solicitor, that will be reflected in the final costs certificate and the cash account; and the sum assessed as owing will be enforceable by the solicitor as if the final costs certificate were a judgment for payment of the amount of money that the solicitor is owed. Conversely, if there is a balance due and owing by the solicitor (because, typically, the client has paid more on account than is assessed to be recoverable) then that too will be reflected in the final certificate; and the sum assessed as owed by the solicitor will be enforceable by the client on the same basis.

In my judgment this detailed scheme leaves no room for the solicitors’ argument in the present case that sums paid on account by reference to an agreement that is held to be unenforceable shall only be repaid if the client justifies repayment on restitutionary principles. To my mind it must be irrelevant whether an item of costs is disallowed because it is unreasonable or because it is claimed pursuant to an agreement that is unenforceable. In the present case it is common ground that the solicitors’ bill should be assessed at nil. In other words, no sums at all should have been paid to the solicitors at any stage pursuant to the September 2017 Agreement.

The consequences that would follow if Ground 3 were to succeed are startling to the point of absurdity. First, where a solicitor’s bill is reduced to below the sums already paid on account by his client because his fees are held to be unreasonable, the client would be entitled to repayment; but where a solicitor’s bill was reduced to nil because the agreement was unenforceable (on public policy grounds), the client would not. The direct effect of such a conclusion would be to subvert the public policy prohibition on enforcement that I have described above. Second, such a conclusion would only serve to heighten the conflict of interest inherent in a non-compliant CFA: it would be in the financial interests of a solicitor possessed of a retainer that he knew or suspected to be unenforceable on public policy grounds to get in as much money as possible from the client so as to obviate the financial risk of the retainer being held to be unenforceable. As before, there is no suggestion that such a consideration weighed with the solicitors in this case; but what public policy is concerned with is the existence of potential conflicts of interest, not actual disadvantage to the client in a particular case. That said, the point is illustrated by the fact that the client in the present case claims repayment of over US$1.5 million which the client has paid pursuant to the unenforceable September 2017 Agreement.

The solicitors rely upon the decision of Garland J in Aratra Potato Co Ltd v Taylor Joynson Garrett [1995] 4 All ER 695. There the solicitors acted for their clients pursuant to a champertous contingency fee agreement that was unenforceable as contrary to public policy. One issue was whether sums that had been paid to the solicitors could be recovered. Garland J found this to be a matter “of the greatest difficulty”, but concluded, at 710c-e:

“At the end of the day I take the view that, subject to any question of severance, where services have been rendered and paid for under an unenforceable contract in circumstances where it cannot be suggested that the payee has, apart from entering into the agreement, acted unconscionably towards the payer or been unjustly enriched at his expense, it is unreal to hold that the consideration, albeit one contrary to public policy, has wholly failed and that the plaintiff is entitled to recover the price of those services while retaining the benefit of them. The better rationale is that the champertous agreement is unenforceable rather than void or voidable.”

Foster J held that this reasoning and conclusion could not stand in the light of the later decisions in Awwad, Sibthorpe, and Garrett, citing in particular the explanation provided by Dyson LJ at [27]-[30] of Garrett. I agree. Furthermore, it is notable that the issue was raised in Aratra Potato without reference to the statutory scheme for assessment pursuant to section 70 of the Solicitors Act 1974. If fees were disallowed on assessment because they were excessive, there could be no suggestion that the solicitors were entitled to retain them because of the happenstance that the excessive fees had been paid on account. There is no good reason, particularly in the light of the public policy considerations that I have already outlined, why a different result should obtain where, as here, no fees at all should have been received by the solicitors on account.

I would therefore dismiss Ground 3 of the Appeal.