BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

England and Wales High Court (Senior Courts Costs Office) Decisions


You are here: BAILII >> Databases >> England and Wales High Court (Senior Courts Costs Office) Decisions >> RSA Pursuit Test Cases, Re [2005] EWHC 90003 (Costs) (27 May 2005)
URL: http://www.bailii.org/ew/cases/EWHC/Costs/2005/90003.html
Cite as: [2005] EWHC 90003 (Costs)

[New search] [Context] [Printable version] [Help]


This judgment has been obtained from the Supreme Court Costs Office pages on the HM Courts Service web site. The citation used by BAILII is not an officially approved citation.

 

Neutral Citation Number: [2005] EWHC 90003 (Costs)
Case Nos: PTH0310421; PTH0402192; PTH0403390; PTH0401699; PTH0404292

IN THE HIGH COURT OF JUSTICE
SUPREME COURT COSTS OFFICE

Clifford's Inn, Fetter Lane
London, EC4A 1DQ
27 May 2005

B e f o r e :

Senior Costs Judge Hurst
____________________

RSA Pursuit Test Cases

____________________

Mr Timothy Dutton QC and Mr Andrew Mitchell
(instructed by Hextalls) for RSA First Assist
Mr Jeremy Morgan QC (instructed by Charles Russell) for the Claimants
Mr Andrew Bartlett QC and Mr Alexander Hutton (instructed by Beachcroft Wansbroughs and Kennedys) for the Defendants in the cases of Sandiford, Clarke, Anthony Baker, Deborah Baker
Mr Nicholas Bacon (instructed by Leo Abse & Cohen) for the Defendant in the case of Farr
Hearing dates : 14, 15, 16, 17, 18 February 2005 and 27 April 2005

____________________

HTML VERSION OF JUDGMENT
____________________

Crown Copyright ©

Table Of Contents

   

para

Background

1

The Issues

4

The Applicable Law

8

The Pursuit Policy Wording

19

The Test Cases

22

 

Deborah Baker v Addenbrookes

23

 

Anthony Baker v Euromark

27

 

Clarke v Tom James

32

 

Sandiford v Price's Patent Candles

36

 

Farr v Kerslake

40

The Evidence

45

Rsa And First Assist's Witnesses

45

 

Peter William Smith

45

 

Emmanuel Gilbert

79

The Claimants' Witnesses

96

 

Mark Scrivenger

96

 

William John Vallance

116

 

Amanda Stevens

135

 

Kenneth Besfor

156

 

Peter Henry Evan Bennett

161

The Defendants' Witnesses

173

 

Christopher Wait

173

 

Kate O'Reilly

193

 

John Paul Ivory

211

 

Adam Richard Burrell

222

Comments On The Evidence

225

 

First Assist Witnesses

225

 

The Claimants' Witnesses

228

 

The Defendants' Witnesses

229

   

General Submissions

236

 

First Assist

237

 

Test Case Claimants

247

 

Defendants 1 - 4

251

 

(i) Section 29 Access to Justice Act 1999

252

 

(ii) The Costs Judge

254

 

(iii) The Ultimate Question

255

 

Defendant 5 (Farr)

257

General Conclusions

260

Submissions On The Issues

271

Issue 1: Is the contract of insurance void for uncertainty because at the time the contract is made the amount of the premium is insufficiently certain and is the said contract accordingly unenforceable by RSA against the claimant and if so what is the consequence?

271

 

First Assist

271

 

Defendant 5 (Farr)

275

Issue 1 - Conclusions

277

Issue 2: Is the insurance arrangement between the client, the insurer and/or the solicitors unlawful on the grounds of champerty and if so what is the consequence?

283

 

First Assist

283

 

Defendants 1 - 4

291

 

Defendant 5 (Farr)

296

Issue 2 - Conclusions

297

Issue 3 - Is the method of calculation of the premium inherently flawed and if so what is the consequence?

307

 

First Assist

307

 

Test Case Claimants

322

 

Defendants 1 - 4

323

 

1. Constant Relationship

325

 

2. Policy taken out after CFA entered into

327

 

3. The relative sizes of the estimates

328

 

4. Estimates of success

333

 

5. Premium calculated on Claimant's costs as claimed

335

 

6. Premium calculated so as to make up a lack of

premium income in unsuccessful cases

336

 

Defendant 5 (Farr)

338

Issue 3 - Conclusions

342

Issue 4 - What commissions, if any, are payable to the Claimants' legal representatives and/or any other agents of the insurers and if there are any, when and in what circumstances are they payable?

364

Issue 5 - Should the amount of the recoverable premium be reduced on the grounds that an insurance policy ought reasonably to have been taken out at an earlier stage in the proceedings?

365

 

First Assist

365

 

Test Case Claimants

366

 

Sandiford

367

 

Clarke

368

 

Defendants 1 - 4

369

 

Sandiford

371

 

Clarke

372

 

Defendant 5 (Farr)

373

Issue 5 - Conclusions

374

Issue 6 - Has the claimant acted reasonably in taking out the RSA Pursuit policy, and if not what are the consequences?

378

 

First Assist

378

 

Test Case Claimants

380

 

Deborah Baker

382

 

Anthony Baker

388

 

Clarke

392

 

Sandiford

396

 

Farr

400

 

Defendants 1 - 4

410

 

Deborah Baker

414

 

Anthony Baker

416

 

Clarke

418

 

Sandiford

419

 

Defendant 5 (Farr)

420

Issue 6 - Conclusions

423

 

Deborah Baker

424

 

Anthony Baker

430

 

Clarke

432

 

Sandiford

434

 

Farr

435

Issue 7 - What if anything is the recoverable amount of the premium against the defendant pursuant to Section 29 of the Access to Justice Act 1999?

438

 

First Assist

438

 

The Claimants

440

 

Defendants 1 - 4

441

 

Defendant 5 (Farr)

446

Issue 7 - Conclusions

447

 

Deborah Baker

457

 

Anthony Baker

460

 

Clarke

461

 

Sandiford

464

 

Farr

465

Issue 8 - Has the indemnity principle been breached in the case of Farr?

468

 

First Assist

468

 

Test Case Claimants

469

 

Defendant 5 (Farr)

470

Issue 8 - Conclusions

472

Issue 9 - What is the impact of BTE cover in Farr?

475

Summary

476

Chief Master Hurst

BACKGROUND

  1. First Assist, through a policy underwritten by Royal and Sun Alliance (RSA), provides a bespoke after the event (ATE) insurance policy known as "Pursuit" to claimants. The policy is aimed at claimants whose claims render them unsuitable for "mass market" or "delegated" ATE policies. Mr Smith of First Assist states that each case is individually assessed and underwritten with the premium calculated by reference to a particular underwriting process and method. Five test cases have been selected as the vehicle for the trial of various issues arising from the use of the policy. First Assist has been joined as a party to these proceedings. Mr Dutton suggests that the five Test Cases are more complex and risky than those previously dealt with by the court.

  2. In reality the trial of the issues in these Test Cases is part of the battle between the liability insurers and the ATE insurance providers. The insurers standing behind the Defendants are Allianz Cornhill, AXA, Norwich Union, Ensign and NHSLA. (The NHSLA is in fact a statutory body charged with paying out on claims recovered against English NHS Trusts.) The Defendants argue that the five Test Cases are sufficiently representative and that First Assist could easily have added more cases. The Claimants suggest that the Defendant paying parties have absolute control over the cases which go to detailed assessment by the simple expedient of either not challenging, or making reasonable offers on cases which are less favourable to their argument. Mr Bartlett argues that there were plenty of cases available for First Assist to choose from should they have so decided. In my view nothing turns on this particular aspect of the proceedings. The fact is that I have heard full evidence and submissions on the five Test Cases before me, and, should different issues arise in the future, further Test Cases may be brought in order to resolve those issues.

  3. I should say at the outset that I am satisfied that the Pursuit policy is the product of an honest attempt by a reputable insurer to produce an ATE policy to fill a specific gap in the market. Unlike the packages offered by Claims Direct and The Accident Group ([2003] EWCA Civ 136 and [2004] EWCA Civ 575 respectively), the premium payable in respect of the Pursuit policy is exclusively referable to that policy and does not cover the cost of any ancillary benefits for the insured.

    THE ISSUES

  4. On 4 May 2004, in accordance with an Order dated 19 April 2004 the Defendants served a list of seven issues. On 15 July 2004 I made an Order in the case of Farr v Kerslake adding two further issues to the list. The issues are as follows:
    1. Is the contract of insurance void for uncertainty because at the time the contract is made the amount of the premium is insufficiently certain and is the said contract accordingly unenforceable by RSA against the claimant and if so what is the consequence?

    2. Is the insurance arrangement between the client, the insurer and/or the solicitors unlawful on the grounds of champerty and if so what is the consequence?

    3. Is the method of calculation of the premium inherently flawed and if so what is the consequence?

    4. What commissions, if any, are payable to the claimants' legal representatives and/or any other agents of the insurers and if there are any, when and in what circumstances are they payable?

    5. Should the amount of the recoverable premium be reduced on the grounds that an insurance policy ought reasonably to have been taken out at an earlier stage in the proceedings?

    6. Has the claimant acted reasonably in taking out the RSA Pursuit policy, and if not what are the consequences?

    7. What if anything is the recoverable amount of the premium against the defendant pursuant to Section 29 of the Access to Justice Act 1999?

    8. Does the claimant's claim for the RSA Pursuit premium breach the indemnity principle because the claimant's purported liability to pay the premium is not a genuine liability but purely a device to enable recovery of the premium between the parties?

    9. Assuming the method of calculation is not inherently flawed (issue iii) what is the relevance, if any, to the calculation of the RSA Pursuit premium of any pre-existing legal expenses insurance cover incepted and/or in existence prior to the claimant's agreement to enter into the Pursuit policy?
  5. Issue i) and issue viii) are pursued only by the Defendant in Farr v Kerslake. Issue iv), relating to commissions, was not pursued by any of the Defendants once the evidence had been heard. Issue ix) was not pursued.

  6. The issues fall into two categories, the first relating to enforceability and the second to reasonableness.

  7. It is First Assist's case that: (1) there is no merit in any of the arguments raised by the paying parties to challenge the enforceability of the Pursuit policy and that Section 29 of the Access to Justice Act 1999 is therefore engaged; (2) the Claimants' contention that it was reasonable to use the Pursuit policy, and that the premium itself is reasonable, is, on the analysis of the facts of each individual case, correct; and, (3) the Claimants are right that they should recover all, or substantially all, of the premium in the Test Cases.

    THE APPLICABLE LAW

  8. Section 29 of the Access to Justice Act 1999 provides:

    "Where in any proceedings a costs order is made in favour of any party who has taken out an insurance policy against the risk of incurring a liability in those proceedings, the costs payable to him may, subject in the case of court proceedings to rules of court, include costs in respect of the premium of the policy."

  9. The Court of Appeal in Callery v Gray (No.2) [2001] EWCA Civ 1246; [2001] 4 All ER 1 CA, held that the words "insurance against the risk of incurring a costs liability" mean: "insurance against the risk of incurring a costs liability that cannot be passed on to the opposing party" (paragraphs 59 and 60). The court set out the test of what is reasonable:

     

    "11. It was common ground, and rightly so, that the court, when considering whether to award an insurance premium by way of costs, has to consider whether the premium is reasonable. It was also common ground that, insofar as the court finds that the premium is not reasonable, it can and should reduce it. There was debate as to the appropriate approach to the application of the test of what is reasonable.

    12. It is important in this context to draw a distinction between two separate matters. The first is the nature of the benefits to which the litigant is contractually entitled in exchange for the payment of the premium. This falls to be determined from the terms of the contract under which the premium is paid. Section 29 permits the recovery of a premium where this is payment for insurance against the risk of liability for costs. If payment of a so-called premium buys a contractual entitlement to other benefits it is, to say the least, arguable that the premium cannot, to that extent, be recovered under section 29. Thus the court has to consider the terms of the contract under which the premium is paid to see whether it is simply a contract of insurance against liability for costs or whether it is something other than, or additional to, that.

    13. The contractual benefits purchased by the premium must be distinguished from the use made by the insurer of the premium. An insurer will necessarily look to premium income to meet the costs of the business. The primary costs are likely to be those of meeting claims, but the costs will also include matters such as commissions, advertising and, indeed, refurbishing the insurer's premises. The court will not be directly concerned with how, or on what, the insurer spends the premium income. The court will, however, be concerned with the question of whether the premium is a reasonable price to pay for the benefits that it purchases. Ultimately, this should be a question to be considered having regard to experience, or evidence, of the market. If an insurer is conducting his business in a manner which incurs extravagant, extraneous or otherwise unnecessary expenditure, which has to be covered by the premiums, those premiums are likely to be uncompetitive. To pay such a premium where other more reasonable premiums are available may disentitle the litigant from making a full recovery of the costs of the premium.


    ...

    15. It is highly desirable in the interests of justice that an effective and transparent market should develop in ATE insurance. If the litigant is not at risk as to the premium … it is less easy for a competitive market to develop. Nonetheless, we consider that the solicitor advising the client should be in a position to assist him in selecting ATE insurance cover that caters for his needs on reasonable terms. Master O'Hare informed us that there are at present two sources of information as to availability of ATE cover: the magazine "Litigation Funding", published by the Law Society and the web site www.thejudge.co.uk. We would encourage solicitors to take advantage of such sources of information and hope that before long the exercise of choice will result in competition for ATE business which establishes transparent market rates.

    16. In the meantime, where an insurance premium is challenged it must be open to the insurer, whose position is akin to a subrogated underwriter, to place evidence before the court in an attempt to demonstrate that the premium is reasonable having regard to the costs that have to be covered. Satellite litigation involving such an exercise is, however, unsatisfactory. The Judge can only be expected to give broad consideration to such evidence, for it is not part of the function of a judge assessing costs to carry out an audit of an insurer's business."

  10. Sections 27 and 28 of the 1999 Act amended the Courts and Legal Services Act 1990 by introducing entirely new provisions relating to conditional fee agreements (CFAs). Section 29 of the 1999 Act was part of the new litigation funding structure introduced by the 1999 Act. Although Section 29 has not been inserted into the 1990 Act it may safely be assumed that Parliament had in mind the Statutory Objective of the 1990 Act:

     

    "17. (1) The general objective of this Part is the development of legal services in England and Wales (and in particular the development of advocacy, litigation, conveyancing and probate services) 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.

    (2) In this Act objective is referred to as "the statutory objective"."



  11. The second issue, which all Defendants pursue, is the issue of champerty. The most recent definitive decision on the subject of maintenance and champerty is that of the Court of Appeal in R (Factortame) v Secretary of State for Transport [2002] EWCA Civ 932. I quote selectively from the extensive review of the authorities in the judgment of Lord Phillips MR, who gave the judgment of the court:

    "The law of champerty

    31. Champerty is a variety of maintenance. Maintenance and champerty used to be both crimes and torts. A champertous agreement was illegal and void, involving as it did criminal conduct. Ss. 13(1) and 14(1) of the Criminal Law Act 1967 abolished both the crimes and the torts of maintenance and champerty. S.14(2) provided, however:

    "The abolition of criminal and civil liability under the law of England and Wales for maintenance and champerty shall not affect any rule of that law as to the cases in which a contract is to be treated as contrary to public policy or otherwise illegal."

    Thus, champerty survives as a rule of public policy capable of rendering a contract unenforceable.

    32. "A person is guilty of maintenance if he supports litigation in which he has no legitimate concern without just cause or excuse" - Chitty 28th Ed. Vol.1 17-050. Champerty ‘occurs when the person maintaining another stipulates for a share of the proceeds of the action or suit' - ibid 17-054. Because the question of whether maintenance and champerty can be justified is one of public policy, the law must be kept under review as public policy changes. As Danckwerts L.J. observed in Hill v Archbold [1968] 1 QB 686 at 697:

    "…the law of maintenance depends upon the question of public policy, and public policy …is not a fixed and immutable matter. It is a conception which, if it has any sense at all, must be alterable by the passage of time."

    33. In Trendtex Trading v Credit Suisse [1980] 1 QB 629 at p.663 Oliver LJ remarked:

    "There is, I think, a clear requirement of public policy that officers of the court should be inhibited from putting themselves in a position where their own interests may conflict with their duties to the court by agreement, for instance, of so called "contingency fees".

    34. The introduction of conditional fees shows that even this requirement of public policy is no longer absolute. This case raises the question of whether the requirement extends to expert witnesses or others in a position to influence the conduct of litigation and, if it does, whether on the facts of the present case the agreements concluded by Grant Thornton can be justified.

    35. In Trepca Mines Ltd (No.2) [1963] 1 Ch 199 at p.219 Lord Denning MR observed:

    "The reason why the common law condemns champerty is because of the abuses to which it may give rise. The common law fears that the champertous maintainer might be tempted, for his own personal gain, to inflame the damages, to suppress evidence, or even to suborn witnesses. These fears may be exaggerated, but, be that so or not, the law for centuries had declared champerty to be unlawful, and we cannot do otherwise than enforce the law; and I may observe that it has received statutory support, in the case of solicitors, in section 65 of the Solicitors Act 1957."

    36. Where the law expressly restricts the circumstances in which agreements in support of litigation are lawful, this provides a powerful indication of the limits of public policy in analogous situations. Where this is not the case, then we believe one must today look at the facts of the particular case and consider whether those facts suggest that the agreement in question might tempt the allegedly champertous maintainer for his personal gain, to inflame the damages, to suppress evidence, to subborn witnesses or otherwise to undermine the ends of justice.

    37. In reaching this conclusion we have been particularly influenced by the approach of the Court of Appeal and the House of Lords in Giles v Thompson…

    38. In the leading judgment in the Court of Appeal [1993] 3 All ER 321 Steyn LJ at p.328 identified the public policy which renders champertous agreements illegal as resting on the perceived need to protect the integrity of public justice. Later, at p.336, he added that the policy focused on the protection of the party confronted with the maintained litigation, it did not exist to protect the plaintiff. At pp.328-9 he gave a valuable exposition of the history of this area of the law, culminating in the enactment of s.58 of the Courts and Legal Services Act 1990, which we shall have to consider in more detail in due course. As to this, he remarked at p.331:

    "The relevance of s58 is that Parliament has, subject to the requirements of the section, empowered the Lord Chancellor to validate by order agreements for a percentage uplift in the costs in the event of success. The ability to recover fees beyond what was otherwise reasonable was intended to be ‘an incentive to lawyers to undertake speculative actions'. Such agreements were, and in the absence of an order still are, unlawful as being contrary to public policy. The rationale of the common law rule is that such agreements allowed the duty and interest of solicitors to conflict with a resultant risk of abuse of legal procedure. Section 58 evidences a proposed modification in relation to an important species of champerty. It represents at least a concession to the view that the abuses associated with champerty are not the inevitable result of all variants of contingency fee agreements. And there is, of course, no more cogent evidence of a change of public policy than the expression of the will of Parliament."

    Subsequently, he observed at p.332:

    "Contingency fee agreements are nowadays perhaps the most important species of champerty. Such agreements are still unlawful. Yet an English solicitor may share in a contingency fee earned in foreign litigation; see r.8 (contingency fees) of the Solicitors' Practice Rules 1990. This reinforces the point that the doctrine of champerty serves to protect only the integrity of English public justice. It is based not on grounds of morality but on a concern to protect the administration of civil justice in this country."

    He continued, on the following page:

    "Ultimately, it is necessary to consider the questions posed in this case in the light of contemporary public policy. The correct approach is not to ask whether, in accordance with contemporary public policy, the agreement has in fact caused the corruption of public justice. The court must consider the tendency of the agreement. The question is whether the agreement has the tendency to corrupt public justice. And this question requires the closest attention to the nature and surrounding circumstances of a particular agreement. That is illustrated by the well-known decision of the House of Lords in Trendtex Trading Corp v Credit Suisse [1981] 3 AllER 520, [1982] AC 679."

    ...

    40. In the House of Lords [1994] 1 AC 142 Lord Mustill gave the leading speech, in which the other members of the House concurred…

    ...

    42. On these facts Lord Mustill held that it was appropriate to consider whether the mischief was established against which the public policy was directed. As to this, he observed at p.161:

    "It is sufficient to adopt the description of the policy underlying the former criminal and civil sanctions expressed by Fletcher Moulton LJ in British Cash and Parcel Conveyors Ltd v. Lamson Store Service Co. Ltd [1908] 1 KB 1006, 1014:

    "It is directed against wanton and officious intermeddling with the disputes of others in which the [maintainer] has no interest whatever, and where the assistance he renders to the one or the other party is without justification or excuse."

    This was a description of maintenance. For champerty there must be added the notion of a division of the spoils."

    43. Lord Mustill held that in neither case was this mischief established. Summarising the position, he said at p.165:

    "Returning to the company, is it wantonly or officiously interfering in the litigation; is it doing so in order to share in the profits? I think not. The company makes its profits from the hiring, not from the litigation. It does not divide the spoils, but relies upon the fruits of the litigation as a source from which the motorist can satisfy his or her liability for the provision of a genuine service, external to the litigation. I can see no convincing reason for saying that, as between the parties to the hiring agreement, the whole transaction is so unbalanced, or so fraught with risk, that it ought to be stamped out. The agreement is one which in my opinion the law should recognise and enforce."

    44. This decision abundantly supports the proposition that, in any individual case, it is necessary to look at the agreement under attack in order to see whether it tends to conflict with existing public policy that is directed to protecting the due administration of justice with particular regard to the interests of the defendant. This is a question that we have to address. In so doing we revert to the statement of Lord Mustill that ‘the rule, now in the course of attenuation, which forbids a solicitor from accepting payment for professional services calculated as a proportion of the sum recovered from the defendant …survives nowadays, so far as it survives at all, largely as a rule of professional conduct'. With respect, this statement is not correct. The basis of the rule is statutory. It is now necessary to look at the relevant statutory provisions …

    . . . .

    62. More generally, however, s.58[of the 1990 Act] evidences a radical shift in the attitude of public policy to the practice of conducting litigation on terms that the obligation to pay fees will be contingent upon success. Whereas before this practice was outlawed, it is now permissible - subject to the requirements imposed by the section. These requirements do not appear designed to mitigate the mischief that had led to the banning of contingency fees - the undesirability of the interests of officers of the court conflicting with their duties to the court. Rather the requirements appear designed to protect the litigants concluding conditional fee agreements who, when the section was first enacted, were required to pay any ‘uplift' out of their recoveries. Conditional fees are now permitted in order to give effect to another facet of public policy - the desirability of access to justice. Conditional fees are designed to ensure that those who do not have the resources to fund advocacy or litigation services should none the less be able to obtain these in support claims which appear to have merit.

    . . . .

    76. In Giles v Thompson Lord Mustill applied the test of public policy identified by Fletcher Moulton LJ in the British Cash case. That test is appropriate when considering those who, in one way or another, support litigation in which they are concerned. It is not, however, really in point when considering agreements under which those who are playing a legitimate part in the process of litigation provide their services on a contingency fee basis. A solicitor who charges a contingency fee which does not satisfy the requirements of s.58, can hardly be said to be guilty of ‘wanton and officious intermeddling in the disputes of others …where the assistance he renders to one party or another is without justification of excuse'. The public policy in play in the present case is that which weighs against a person who is in a position to influence the outcome of litigation having an interest in that outcome.

    ..."

  12. The parties' submissions on champerty are dealt with below at paragraph 283.

  13. The trial of these Test Case issues takes place during the course of detailed assessment proceedings. All the assessments are on the standard basis and are accordingly governed by CPR 44.4(1) and (2). It is not necessary to set out this rule, but it is necessary to bear in mind that the court will resolve any doubt which it may have as to whether costs were reasonably incurred or reasonable and proportionate in amount in favour of the paying party. CPR 44.5 sets out the factors to be taken into account in deciding the amount of costs, but again there is no need for me to set out the terms of that rule which is familiar to all those involved in these proceedings. For the sake of completeness I mention that CPR 44.5 is supplemented by Section 11 of the Costs Practice Direction from which I quote selectively:

     

    "11.7 Subject to paragraph 17.8(2), when the court is considering the factors to be taken into account in assessing an additional liability, it will have regard to the facts and circumstances as they reasonably appeared to the solicitor or counsel when the funding arrangement was entered into and at the time of any variation of the arrangement.

    ...

    11.10 In deciding whether the cost of insurance cover is reasonable, relevant factors to be taken into account include:

    (1) [omitted]

    (2) the level and extent of the cover provided;

    (3) the availability of any pre-existing insurance cover;

    (4) whether any part of the premium would be rebated in the event of early settlement;

    (5) the amount of commission payable to the receiving party or his legal representatives or other agents."



  14. Paragraph 17.8(2), referred to in paragraph 11.7, applies only to costs only proceedings.

  15. With regard to what constitutes "premium" MacGillivray on Insurance Law (9th Edition 1997 paragraph 72), defines the word "premium" in these terms:

    "The premium is the consideration required of the insured in return for which the insurer undertakes his obligation under the contract of insurance (Lewis Ltd v Norwich Union Fire Insurance Co [1916] AC 509, 519)."

  16. This definition was adopted by the Court of Appeal in the Claims Direct Test Cases [2003] EWCA Civ 136 at paragraph 25. In a different appeal in the same proceedings [2002] EWCA Civ 428 at [44] Arden LJ said:

    "44. ... The expression "premium" is not defined by the Access to Justice Act 1999. The court has been referred to the Civil Procedure Rules and various authorities. … In my judgment the premium is not necessarily limited to payments paid on inception of cover, but could include any further amounts paid by, or on behalf of the insured, pursuant to terms agreed with the insurer."

  17. She went on to refer to the decision of the Court of Appeal in Callery v Gray (No.2) in these terms:

    "46. ... The court specifically added "Satellite litigation involving such an exercise [i.e. examining evidence of insurance cover] is however unsatisfactory. The Judge can only be expected to give broad consideration to such evidence. It is not part of a function of a Judge assessing costs to carry out an audit of the insurance business" ... The court may wish to check the overall result which it reaches by reference to the alternative method of obtaining access to justice. This might involve looking at alternative rates of cover, or the costs which would be involved if litigation were to be funded in some other way. This has been called the "top down" approach. Nonetheless, in making that comparison it may be necessary to bear in mind that like may not be being compared with like ... Nevertheless in my judgment, the comparison between the cover provided by these appellants and other means of financing litigation, including other insurance cover, is a relevant consideration to which the appellants could properly bring ... the attention of the Senior Costs Judge. I say this, bearing in mind the general purposes of the new methods of funding litigation introduced by the 1999 Act and by the fact that it is obviously highly desirable in the interests of justice that these methods should be competitive. A premium may not be reasonable if there are alternative ways of providing the same funding at significantly less expense."

  18. The Court of Appeal has given guidance as to the correct approach to be taken to proportionality in the context of costs assessments in Lownds v Home Office [2002] EWCA Civ 365.

    THE PURSUIT POLICY WORDING

  19. The wording of the Pursuit policy remains constant throughout the five Test Cases. So far as relevant that wording is as follows. Under "Important Notes":

    "The Insurer will provide the insurance described in this Policy in consideration of the Insured's promise to pay the Premium."

  20. Under "Definitions":

    "INSURED

    The term used to denote collectively both the Insured Litigant and Insured Solicitor in their capacity as beneficiaries under this Policy.

    ...

    LIMIT OF INDEMNITY

    The Insurer's liability in respect of Normal Fees and Expenses shall not exceed in the aggregate 100% of the amount specified in the proposal for this insurance as the Insured Solicitor's reasonable estimate of the sum likely to be recovered from the Opponent, excluding costs, if the Legal Proceedings are resolved in the Insured Litigant's favour. This amount appears on the Schedule.

    ...

    ADVERSE COSTS

    The net costs of the Opponent in the Legal Proceedings to the extent that the Insured Litigant is legally liable to discharge them, after taking account of any costs awarded against the Opponent or agreed to be paid by the Opponent. No cover is provided in respect of any success fee to which the Opponent or Opponent's solicitor or Opponent's barrister may be entitled.

    NORMAL FEES

    The costs of the Insured's Solicitor acting for the Insured Litigant under the Conditional Fee Agreement and which form the basis of the calculation of the Success Fee.

    Where the outcome of the Legal Proceedings is not a Success the Insurer shall have the right to have the Insured's Solicitor's bills taxed or assessed on the standard basis.

    EXPENSES

    Expenses and other disbursements paid by the Insured's Solicitor to other parties which are reasonably and properly incurred by the Insured's Solicitor

    a) in connection with the Legal Proceedings,

    b) in appealing or resisting an appeal against the judgment of a court in connection with the Legal Proceedings provided Our prior written consent has been obtained to the appeal.

    The Insurer shall not be liable for Counsel's fees where Counsel is retained to act in the Legal Proceedings under a conditional fee agreement.

    ...

    PERIOD OF INSURANCE

    Cover commences at the later of the inception date shown on the Schedule and the date of signing of the Conditional Fee Agreement.

    Cover ceases when

    a) the Legal Proceedings are concluded by a judgment of the Court of first instance or following an appeal to which We have given Our prior written consent, or

    b) the Legal Proceeding are discontinued with Our prior written consent, or

    c) the Conditional Fee Agreement is terminated whichever is the earliest.

    PREMIUM

    The amount shown on the Schedule which becomes payable when the outcome of the Legal Proceedings is a Success.

    Where the Insured Litigant receives an offer to settle or a payment into Court which equals or exceeds the definition of Success the Insured Litigant may at that time immediately pay the Premium based on the Normal Fees at that date. The Insured Litigant may elect by written notice to Us to continue with the action and defer payment of the Premium to the Conclusion of the Legal Proceedings. The Premium will then be based on the Normal Fees at the Conclusion of the Legal Proceedings. If, having made this election, the Insured Litigant ultimately fails to achieve a Successful outcome to the Legal Proceedings the Premium payable at the date when the Insured Litigant made the election shall remain payable.

    SUCCESS FEE

    The amount specified in the Conditional Fee Agreement as the percentage uplift to the Insured Solicitor's Normal Fees payable on the successful conclusion of the Legal Proceedings.

    SUCCESS/SUCCESSFUL

    The Insured Litigant is offered or obtains at any time a net entitlement to money and/or damages and/or costs which, taking into account any counter-or cross-claim in the Legal Proceedings, equals or exceeds the sum shown in the Schedule as the definition of Success. If the Insured Litigant achieves such an outcome at trial and is required by Us to defend an appeal which results in a net entitlement to money, damages and costs which is lower than the definition of Success then the action will be deemed unsuccessful.

    Where an offer to settle is received from the Opponent at any time which equals or exceeds the definition of Success then the outcome of the Legal Proceedings is Successful even if the Insured Litigant elects not to accept the offer.

    ...

    COVER

    Section A

    Where the outcome of the Legal Proceedings is not a Success the Insurer will, subject to the Limit of Indemnity indemnify the Insured Litigant in respect of

    a) Adverse Costs

    provided that

    1. the Court makes an award of Adverse Costs against the Insured Litigant or

    2. the Legal Proceedings are settled or otherwise discontinued with the prior written agreement of the Insured Litigant, the Insured Solicitor and Us

    3. the Insurer shall not be liable to pay any Adverse Costs until the Legal Proceedings are finally concluded.

    (b) Expenses

    provided that

    1. the Insurer shall only be liable for any Expenses to the extent that the Insured Litigant is not entitled to recover them from the Opponent or any other party

    2. the Insurer shall not be liable to pay any Expenses until the Legal Proceedings are finally concluded.

    ...

    GENERAL EXCLUSIONS

    This insurance does not cover

    1. The Insured Solicitor's Contribution

    2. Any payment by the Insurer under this Policy which is due as a result of a discontinuance or settlement to which We have not given Our prior written consent.

    3. Damages of any kind.

    4. Enforcement proceedings.

    5. Wasted costs or any increased legal or other costs arising from any unreasonable delay or negligence by the Insured Litigant or Insured Solicitor which in Our opinion is prejudicial to the conduct of the Legal Proceedings.

    6. Under Section B only, any liability of the Insurer arising from the Insured Solicitor's failure to comply with procedural directions or pre-action protocols.

    7. Legal Proceedings made, commenced, brought or transferred outside the Territorial Limits or which are not governed by English law.

    8. Any payment by the Insurer arising from Legal Proceedings in respect of which the Insured is or but for the existence of this insurance would be entitled to indemnity under any other insurance policy.

    9. Any amount which the Opponent is obliged to pay but fails to pay to the Insured Litigant or Insured Solicitor for any reason.

    10. Normal Fees, Expenses and Adverse Costs incurred outside the Period of Insurance.

    11. Any costs incurred by the Insured Litigant or Insured Solicitor in providing Us with any information or documentation under this insurance.

    12. Any payment by the Insurer under the Policy where there has been misrepresentation or material non-disclosure by the Insured Litigant or Insured Solicitor.

    13. Any payment by the Insurer under the Policy if the Legal Proceedings are stayed, discontinued, abandoned or withdrawn by virtue of the bankruptcy, insolvency or liquidation of the Opponent.

    CONDITIONS

    1. Premium

    The Premium payable under this Policy and shown on the Schedule is established by reference to the Normal Fees of the Insured Solicitor.

    The Premium is payable even if the Opponent delays or defaults in settling any judgment or agreed settlement.

    The Premium will not be affected by taxation or assessment or any agreement which reduces the Insured Solicitor's Normal Fees or the level of the Success Fee.

    ...

    3. Termination

    The policy will terminate if the Insured Litigant or Insured Solicitor terminates the Conditional Fee Agreement.

    We may cancel the Policy by giving fourteen days' notice in writing to the Insured Litigant and Insured Solicitor if

    a) the Insured Litigant fails to comply with the terms of the Conditional Fee Agreement

    b) the Insured Litigant does not follow the Insured Solicitor's recommendations with regard to settlement of the Legal Proceedings

    c) the Insured Litigant does not follow Our recommendations with regard to settlement of the Legal Proceedings

    d) the Insured Litigant rejects the Insured Solicitor's advice to discontinue the Legal Proceedings.

    In the event that the Policy is cancelled, the Insurer shall be under no obligation to make any payment.

    4. Minimising Claims or Legal Proceedings

    The Insured must take all reasonable measures to minimise the cost of Legal Proceedings.

    ...

    7. Due Observance

    The due observance of and compliance with the terms provisions and conditions of the Policy insofar as they relate to anything to be done or complied with by the Insured Litigant or Insured Solicitor shall be conditions precedent to any liability of the Insurer to make any payment hereunder. In the event that the Policy is terminated the parties to this Policy shall continue to observe the conditions to the extent that they remain relevant.

    8. Provision of information

    The Insured Solicitor must

    a) provide to Us regular progress reports on the Legal Proceedings and associated costs and when specifically requested by Us.

    b) advise Us in writing as soon as an offer to settle the Legal Proceedings or a payment into Court is made by the Opponent. The Insured Solicitor and the Insured Litigant must not enter into any agreement to settle without Our prior written consent.

    c) advise Us in writing as soon as either the Insured Solicitor or Insured Litigant wishes to discontinue the Legal Proceedings.

    d) provide Us with a copy of the judgment of the Court in the Legal Proceedings.

    e) advise Us immediately in writing if the Conditional Fee Agreement for which the Policy has been issued has been terminated or varied.

    f) provide Us promptly with any requested information.

    g) provide Us immediately with a copy of any Counsel's Opinion.

    h) allow Us to inspect the files of the Insured Solicitor at any time, including after the conclusion of the Legal Proceedings.

    The Insured Litigant or Insured Solicitor must advise Us of any material changes to the prospects of success in the Legal Proceedings.

    ..."

  21. The definition of success in each case is set out in the schedule, together with the premium expressed as a percentage of "normal fees". I deal with the particular definition of success in each Test Case in the next section of this judgment.

    THE TEST CASES

  22. The factual background to each of the Test Cases and the chronology relating to each is not in dispute. The chronology in each case is however of importance. I set out below short details of each of the Test Cases.

    Deborah Baker v Addenbrookes

  23. This was a clinical negligence claim arising out of alleged negligence resulting in the surgical removal of a section of the Claimant's bowel. The claim was ultimately settled for ?400,000. Mr Morgan identified the difficulties with the claim under a number of bullet points:
  24. Mr Morgan supplied a full chronology in each case, which I summarise as follows:

    31 July 1999

    The Claimant was admitted to hospital, the alleged negligence occurred in the next few days (limitation 31 July 2002).

    26 September 2001

    Solicitors Scrivenger Seabrook instructed and CFA entered into. Success fee 100%.

    8 January 2002

    Solicitors submit proposal for ATE insurance to LPL.

    18 February 2002

    Solicitors submit proposal for disbursement funding to First National Bank (no response from First National).

    15 March 2002

    Solicitors approach The Judge Broker who in turn (19 March 2002) approaches First Assist.

    2 April 2002

    First Assist correspond with The Judge as to possible cover.

    20 June 2002

    First Assist quotation: limit of indemnity ?18,500 [own disbursements only] premium 172% of normal fees plus IPT.

    20 September 2002

    Inception of cover on terms of quote.

    24 April 2003

    Defendants offer ?100,000.

    16 June 2003

    Defendants offer ?350,000.

    22 July 2003

    Settled by Consent Order for ?400,000 plus costs.

  25. The definition of success contained in the policy schedule reads:

    "Definition of success

    The case is finally decided in favour of the insured litigant whether by a court decision or an agreement to pay the insured litigant damages."



  26. Premium ?52,048.92 plus IPT (?54,787.37), ie normal costs, ?30,261 x 172%.

    Anthony Baker v Euromark

  27. This was a personal injury claim in which the Claimant asserted that he had been hit on the hand with a baton during a training course. Settled for ?1,250.

  28. The difficulties and issues identified by Mr Morgan were:


      • Rejection of the case by the Claimant's union solicitors Thompsons.


      • Rejection of the claim to the CICA on the basis of criminal assault.


      • Adverse evidence of an independent witness.


      • Causation problems because of previous medical history.


      • Various ATE proposals were rejected.

  29. Chronology

    19 August 1999

    Incident occurs.

    1 August 2001

    Thompsons advise client they will proceed no further.

    14 November 2001

    Solicitors AMS Law instructed. CFA entered into, 100% success fee.

    26 July 2002

    After rejection of previous applications for ATE insurance RSA First Assist proposal form submitted.

    12 August 2002

    Quote from First Assist.

    16 August 2002

    Claim form issued (limitation period expires 19 August 2002).

    16 October 2002

    Inception of policy.

    19 August 2003

    Part 36 payment by Defendants ?850.

    25 September 2003

    Settled for ?1,250.

  30. The schedule to the policy put the limit of indemnity at ?2,000 (the limit of indemnity applies only to expenses and excludes counsel's fees). The definition of success reads:

    "Definition of success:

    The claim for damages is finally decided in favour of the Insured Litigant whether by a court order or an agreement to pay the Insured Litigant damages or acceptance of any offer as advised."

  31. The premium is stated to be 74% of normal fees plus IPT. The premium claimed is ?8,536 plus IPT (?8,962) being ?11,535 normal fees x 74%.

    Clarke v Tom James

  32. The Claimant's claim was for an upper limb disorder (tennis elbow) allegedly caused by pulling a heavily laden trolley. The case ultimately settled for ?20,000 after the exchange of expert evidence. The difficulties and issues identified by Mr Morgan were:
      • Causation, the Defendants argued that the Claimant's injuries were caused by the Claimant's karate activities.


      • The Claimant was unsure of the date of the onset of the symptoms, limitation was thought to expire in Summer 2002.


      • The policy was eventually taken up just before the expiry of four months for service, the insurers required further medical evidence.
  33. Chronology:

    23 March 2001

    Claimant instructed solicitors Hart Brown.

    6 February 2002

    The Defendants having denied liability the Claimant instructed Charles Russell.

    7 March 2002

    CFA entered into: success fee 67%.

    27 March 2002

    Proposal submitted to The Judge Broker. Further information requested by First Assist.

    1 May 2002

    Proceedings issued. Further correspondence between Charles Russell and RSA.

    27 August 2002

    Quotation to Charles Russell.

    29 August 2002

    Policy incepted. Limit of indemnity ?7,000 [own disbursements only]. Premium 123% of normal fees.

    10 July 2003

    Acceptance of Part 36 payment, ?20,000.



  34. The definition of success reads:

    "Definition of success:

    The claim for damages is decided in favour of the Insured Litigant whether by a court decision or where an offer is received which the Insured Litigant's Solicitor advises should be accepted or acceptance of any offer."

  35. Premium ?30,772.76 plus IPT (?32,392.38) being normal fees of ?25,018.50 x 123%.

    Sandiford v Price's Patent Candles

  36. This was a claim for damages for stress at work settled for ?44,000 plus CRU benefits. Apart from the normal difficulty associated with a stress at work case the Claimant's solicitors had tried to obtain other ATE insurance but were unsuccessful. They used The Judge broker but only one insurer was prepared to quote.


  37. Chronology:

    August 1998

    The cause of action accrued. The Claimant's condition was diagnosed as a psychiatric injury. Limitation therefore ran to July 2001.

    23 February 2001

    The Claimant instructed solicitors Ormerods in connection with an employment dispute.

    31 July 2001

    Proceedings issued. Subsequent unsuccessful applications for ATE insurance.

    21 August 2002

    The CFA: 70% recoverable success fee.

    23 August 2002

    Inception of policy. Limit of indemnity ?10,000, [own disbursements only] premium 160% of normal fees.

    15 July 2003

    Settled for ?44,000 plus CRU.



  38. The definition of normal fees in the policy wording was to be amended to show that this was based upon the insured litigant's solicitors' total fees and not limited to those under the conditional fee agreement. The definition of success reads:

    "Definition of success:

    The case is finally decided in favour of the Insured Litigant whether by a court decision or an agreement or acceptance of any offer as advised by the Insured Litigant's Solicitor."

  39. Premium ?16,176.75 plus IPT (?16,986) being normal fees of ?10,110.47 x 160%.

    Farr v Kerslake

  40. The Claimant sustained a serious personal injury (his leg was amputated) as a result of being run over by a bus. Settled for ?250,000 on the first day of trial. Mr Morgan identified the difficulties and issues as:

      • The Claimant had amnesia as to the circumstances of the accident.


      • The Claimant had been drinking (which was the purpose of the outing) at the time of the accident.


      • The Defendant bus driver, who was sober, was firm that the Claimant was wholly at fault.


      • There was no independent direct witness evidence.


      • The Claimant had before the event insurance up to ?25,000 which was used up and ultimately applied to disbursements.


      • The defence raised a positive case not indicated in the reply to the pre-action protocol letter.


  41. Chronology:

    23 August 1999

    Accident occurred.

    3 October 2000

    Solicitors (Dolmans) undertake work under the BTE cover.

    18 June 2002

    Proceedings commenced.

    22 July 2002

    Solicitors write to First Assist enquiring about possibility of topping up BTE.

    17 December 2002

    CFA entered into: 80% recoverable success fee.

    18 December 2002

    Proposal to First Assist.

    16 January 2003

    First Assist quote (the quote is based on incorrect figures).

    23 January 2003

    Insurance policy incepts.

    15 August 2003

    Defendants offer ?100,000 otherwise will proceed to trial on liability and quantum.

    29 September 2003

    Settled on first day of trial for ?250,000.


  42. The limit of indemnity for own disbursements was nil (the BTE policy covered disbursements). The definition of success read:

    "Definition of success:

    The claim for damages is finally settled in favour of the Insured Litigant whether by a court decision or where an offer is received which the Insured Litigants Solicitor advises should be accepted or any other offer accepted."

  43. Notwithstanding the definition of normal fees in the policy the premium will be based on the insured litigant's solicitors' total fees from the date of first instruction in this claim and not just those incurred under the conditional fee agreement.

  44. Premium claimed ?153,378.76 plus IPT (?161,047.69) being 257% of normal costs, ?59,680.45. It is common ground that the premium calculation was based on incorrect figures.

    THE EVIDENCE

    RSA AND FIRST ASSIST'S WITNESSES

    Peter William Smith

  45. Mr Smith is the technical director of First Assist Group Ltd. He had been with that company and its predecessor, the Legal Protection Group Ltd ("LPG"), since 1990. He is a Fellow of the Chartered Insurance Institute and has extensive knowledge of the insurance industry, working for some time with the Royal Insurance Company. In 1990 he joined LPG whose business included legal expenses insurance and the provision of associated legal help lines. Royal Insurance and Sun Alliance merged in 1997 to form RSA. First Assist was formed by consolidating LPG with RSA's equivalent subsidiary, the Care Assist Group Ltd. Mr Smith became the legal expenses manager of the First Assist, a post which he held until 2003. In March 2003 First Assist became independent of RSA. Mr Smith was appointed legal expenses and technical claims director and then technical director in late 2004. He has overall responsibility for the underwriting and claims performance of the company's insurance products.

  46. In his witness statement Mr Smith sets out the origins of the Pursuit scheme. He states that the introduction of conditional fees with recoverable success fees created a vastly enhanced need for ATE insurance. He says (paragraph 13):

    "Without ATE the use of CFAs would be impossible (since a litigant who could not afford to meet his own solicitor's costs could also clearly not afford to meet his opponent's costs if unsuccessful in the action)."

  47. Since First Assist and RSA were already one of the market leaders in before the event insurance (BTE): "The decision was made to take on the challenge of developing an ATE product as well." Because of the perceived instability of the mass personal injury market, the company set out to design a product to support the use of CFAs in areas going well beyond that market. The Pursuit policy was designed to cater for multi track actions in any area in which CFAs could be employed. Mr Smith identifies two problems (paragraph 14):

    "Firstly it was obvious that the premium for such cases would have to be much larger than in run of the mill personal injury cases … Secondly it was obvious that a losing litigant would not have the means to pay the premium."

  48. In order to overcome these problems it was decided (paragraph 15):

    "That the premium needed to be deferred and conditional upon success. Those who won their cases would have the means to pay the premium. Those who lost would not need to do so."

  49. It was decided that the policy needed to provide unlimited cover for adverse costs, giving an unsuccessful litigant full protection, even if the case proved more expensive than had been originally expected.

  50. At the time the Pursuit policy was being developed success fees and ATE insurance premiums were not recoverable from the paying party. These items became recoverable from 1 April 2000. The product was therefore developed on the assumption that the insured would be responsible for meeting the whole cost of the premium out of his or her own resources. On 12 March 1999 a letter of explanation and marketing package was sent out to solicitors firms describing the Pursuit policy:

    "A ground breaking "after the event" conditional fee agreement protection plan."

  51. The letter, which was signed by Mr Smith, put forward as the unique benefits of the Pursuit policy:

      • "No premium is payable at the outset.


      • No premium is payable if your client's case is unsuccessful.


      • Unlimited cover is provided for adverse costs.


      • The majority of your costs and disbursements are paid if your client's case is unsuccessful."


  52. Mr Smith explains (paragraph 19):

    "One further feature of the product's development was that we were aware that some solicitors were reluctant to use CFAs for fear of the effect of losing cases on the firm's financial position. We therefore incorporated a section of cover … under which we would cover a proportion of the solicitors own costs. This facility was not used in any of the Test Cases but does have relevance to the rating mechanism explained below."

  53. Mr Smith then goes on to set out the principles of the Pursuit model. He explains that the cover needs to be unlimited because the level of own and adverse costs is inevitably uncertain.

  54. Since First Assist wished to concentrate on the more complex cases they realised that the payment of the premium in advance would cause problems (paragraph 22):

    "We therefore determined to make the premium deferred and we did this prior to any legislation as to recoverability of the premium from the opponent."

  55. Even after recovery of ATE premiums became possible the insured was and is liable to pay any shortfall in the premium.

  56. It was a feature of the Pursuit policy that the insured would not have to pay the cost of the premium if the case was lost. Mr Smith suggests that ATE policies which defer payment of the premium until an order for payment of the insured's costs is made, or the use of a self funding premium in the event of failure, are different ways of achieving the same underlying objective of not leaving an impecunious, unsuccessful claimant with an outstanding ATE premium to pay.

  57. Mr Smith explains the method by which the premium is calculated at paragraphs 26 to 38 of his statement. The cost of the premium had to reflect the fact that because legal fees and disbursements inevitably increase the further the litigation progresses, the risk to the insurer increases in a similar way. When considering how to quantify the value of the premium this had to be based "upon those costs and disbursements that we were able to know about at the time the risk was presented to us" (paragraph 27).

  58. Ideally, the insurers would need to know the amount of the claimant's disbursements, the defendant's costs and the defendant's disbursements. That information not being available however, the premium could not be calculated on that basis. Mr Smith goes on to suggest that the winning party's solicitors (paragraph 30):

    "will always be in a position at any stage in the case to know what their costs are and these will be quantifiable at whatever stage the case is concluded."

  59. Mr Smith and First Assist took the view that, although it is not possible to say whether the opposing party's costs and disbursements are the same as the claimants, they were often similar, and in more difficult cases particularly so. Mr Smith continues (paragraph 32):

    "We felt that it was a reasonable assumption … to make in broad terms, for the purpose of establishing and maintaining a consistent underwriting procedure that the level of the insured's costs and the opponent's cost would escalate for the duration of the case on a roughly equal basis … We thus arrived at the basic fundamentals of a sliding scale premium calculated by reference to own solicitor's fees. It is important to appreciate that each case is rated on the estimated costs of that case. There is no standard assumption, for example, that the insured liability will always be a fixed proportion of own costs. The only workable assumption is that, as costs for both sides start at zero and proceed to the estimated levels at trial, the relationship between own costs and the insured's liability will remain constant throughout the action."

  60. In order to work out the monetary value of the premium a multiplier (the premium rate) had to be ascertained which could be applied to own solicitor's fees. He explains (paragraph 33):

    "In order to do this it was necessary to form a view as to:

    (i) the likelihood of the insured's case succeeding in percentage terms; and

    (ii) if the policy was called upon, how much would have to be paid out in order to work out the break even position, or "burning cost" for the scheme, where the cost of the claims paid exactly matches the value of risk premium income (excluding expenses and commission) received. Given the product was an entirely new concept, the absence of previous underwriting data, and the considerable volatility in rates in the ATE market, we resolved to apply basic principles to arrive at an appropriate rating model."

  61. Mr Smith continued (paragraph 36):

    "In order to form a view as to the burning cost it was necessary for rating purposes to predict what the policy liability, ie own disbursements and adverse costs and disbursements, might come to in each case. This was done by asking the prospective insured's solicitors in the proposal form to provide us with the best estimate of these costs and disbursements to trial … The insured's solicitors would also be in the best position to know the likely rate of the other side in coming to their estimate which we could then use to calculate the premium."

  62. At paragraph 39 and following Mr Smith gives details of the calculation method of the premium.

  63. The premium is expressed as a specified percentage of the claimant's solicitors' basic or "normal" fees. This percentage is arrived at by:

    (a) adding together the opponent's solicitor's estimated fees and disbursements and the claimant's solicitor's own disbursements;

    (b) dividing the total by the anticipated basic fees of the claimant's solicitors (producing an exposure multiplier);

    (c) arriving at a multiple, which is inversely proportionate to the percentage prospects of success of the case at the time the cover is taken out (producing a risk multiplier);

    (d) applying the risk multiplier to the exposure multiplier to reach the relevant burning cost premium rate;

    (e) adding an allowance for profit and administration and an allowance for broker's commission (if a broker is appointed). IPT is then due on the resultant premium (paragraph 41).

  64. The higher the risk of failure the higher the risk multiplier becomes. Likewise the higher the ratio is between insured costs and disbursements and uninsured costs, the higher the exposure multiplier becomes. The view of Mr Smith and First Assist is that this reflects in a fair and logical manner the risk that the ATE insurer was running.

  65. Mr Smith states that in each case there was a period after the quotation was provided by First Assist before cover was incepted. This period was used to encourage the Defendant to settle before the liability for premium was incurred. He goes on to suggest that since some of the liability insurers have already paid Pursuit premiums at the lower levels in other cases, the Defendants have cherry picked the Test Cases for the trial of these issues (paragraph 75). The Defendants deny that assertion pointing out that First Assist and the Claimants were given the opportunity by the Order of 19 April 2004 to identify test cases of their own.

  66. Having seen the evidence of the Defendants' witnesses Mr Smith put in a further witness statement in which he criticises the evidence of Mr Ivory, firstly on the basis that he has put forward a selection of cases of clinical negligence where the costs were always above ?50,000. He comments that had cases below that level been included this would have brought down the overall average in respect of premium. On Mr Smith's figures, which he exhibits, he calculates an average premium of ?21,606 when considering all Pursuit cases won against the NHSLA since the introduction of the policy. This exhibit was taken from the First Assist database of all ATE cases which had been written using the Pursuit policy and the Pursuit rating in respect of clinical negligence cases only, and where, according to their records, the opponent in those cases was the NHSLA. There were 46 cases in all, of which 16 incurred premiums below ?10,000. The average premium rate across all the cases is 115% of normal costs. He explains (paragraph 7):

    "It can also be seen that our exposure is on average approximately ?60,000 compared with the solicitors' exposure of ?38,000. If it is reasonable for a solicitor to recover a success fee in a range of 60% - 100% to reflect the risk, it follows that the required premium rate for our exposure is in the range of 100% - 150% to reflect the risk we take."

  67. Mr Smith also exhibits an analysis of the lost cases in respect of all classes of legal actions which have been unsuccessful since the inception of the Pursuit policy. The exhibit shows for these cases the date the cases were lost, details of the litigant and solicitor, the date that cover was taken out, the estimated exposure as it was presented by the solicitor at the time First Assist were asked to go on cover, and the amount which they have either been obliged to pay under the terms of the policy or which they currently hold as a reserve in the expectation that they will have to pay out. This analysis demonstrates that the maximum exposure to date had been ?170,000 paid out to a successful opponent where the estimate of adverse costs had been ?15,000. Mr Smith explains about one particular commercial case where there had been deliberate misrepresentation and non disclosure by the Claimant and in respect of which the Judge at trial found fraud. The policy was repudiated, but otherwise the policy liability would have been ?520,000. He describes the business as "long tail". That is, that there is a risk that some of the cases which will conclude in future years will not only take longer to go through but will, if lost, turn into heavily losing cases, although expected to win when put on risk. He describes this as a continual cause for concern.

  68. Mr Smith describes the "Watermark" policy which is for routine personal injury cases, and which is offered at a competitive premium. Watermark is a delegated authority scheme allowing solicitors to provide ATE at agreed premiums for personal injury risks within agreed parameters. There is also another policy called "Pursuit PI" or "Small PI" which is written on the Watermark wording not the Pursuit wording. The product was introduced in June 2003. He states (paragraph 11):

    "The intention was to introduce a policy that was cost effective and also was aimed at solicitors who did bulk PI work and had the demonstrable track record of using CFAs effectively. Such solicitors needed to have high quality systems in place for handling such large scale work, and expertise in it. Watermark was designed as a delegated authority scheme for such solicitors as opposed to one off cases only - they were therefore expected to place all eligible risks with us.

    ...

    14. The cover is similar to Pursuit to the extent that premium is conditional and deferred. However there is a limit of indemnity of ?125,000, and in common with other delegated authority schemes in the market there is a stringent set of acceptance criteria for cases to be eligible to be put on cover by the solicitor."

  69. He goes on to explain that, even had the Watermark policy been available, none of the Test Cases would have fallen within the eligibility criteria. The range of premiums for the Watermark policy was from ?195 in a fast track RTA case to ?845 in a non RTA multi track case (excluding IPT). The Small PI Pursuit policy was designed to fill a gap in the market where solicitors had PI cases which they wished to insure, but which fell outside the terms of their scheme with another provider. The base rates (as used in the Watermark policy) were subject to loading when considering factors such as: individual case prospects, the state of the case, limitation issues, multiple defendants and the fact that the case represented a higher risk than Watermark eligible cases. The scheme was introduced in July 2003 and cover was available only for cases which were introduced by The Judge broker. Underwriters retained a discretion on rating and certain parameters were laid down. The underwriting team retained a discretion whether or not to offer cover but the policy wording itself was the same as the Watermark policy and the limit of indemnity remained ?125,000. Mr Smith exhibited a chart showing how the basic Watermark premiums were loaded, depending on the time when the policy was taken out and whether the merits exceeded 60% or 70%. This meant that the range of premiums for a fast track RTA case outside the Watermark scheme could be as high as ?1,755 and for a non RTA multi track as high as ?7,605 (exclusive of IPT).

  70. At paragraph 22 of his witness statement Mr Smith points out that in the past the NHSLA have paid premiums rated at 253% and 212% and actual premiums of ?39,048 and ?38,095 but have declined to pay other cases with much lower premiums, for example ?800 and ?1,900. He suggests that their approach is inconsistent.

  71. Mr Smith also comments on the evidence of Kate O'Reilly and Mr Burrell and finally on the evidence of Mr Wait of Temple Legal Protection Ltd an underwriting agency.

  72. In cross examination Mr Smith was asked about the difficulty which solicitors might have in giving accurate estimates, his response was:

    "It [obviously] depends on the state of proceedings. It depends on whether information has been shared under protocols between the solicitors, and often in our cases, because they are taken on late, there may be reliance on an estimate provided by the other side, but that still suffers from the same propensity for variance as the solicitor's own costs. By the nature of the fact that they are estimates, they are not likely to be absolutely accurate. On the other hand, they may be much more accurate than saying that you want, let us say, ?100,000 cover in every case. I would say there is some difficulty. In the majority of cases those estimates are unlikely to be seriously adrift in terms of their relationship one to the other, which is fundamental to Pursuit rating. In some cases -- in the Gouldens case I mentioned previously, which is such an example -- where because the case had an unexpected turn, all the estimates were similarly affected and were wrong. From my experience there is a tendency for estimates to be inaccurate in individual cases. In most cases, the inaccuracy is not material. In some cases both sides costs will be materially underestimated because the case goes all the way to trial and the trial is more complex than is anticipated, but I have some difficulty with the assertion that there is systematic under - or over - estimating ..."

  73. It was suggested to Mr Smith that the Claimant's costs build up first and Defendant's costs gradually catch up. He stated that in his company's experience settlements tended to be quite late or very early. If the settlement was very early then the premium would be very low because it would be based on very low basic costs. He pointed out that in the Test Cases the settlement was very late. Most of the cases had been running for three years before proceedings were issued and then subsequently continued to run, in one case settling at trial. He agreed the Defendants' costs would rise as trial approached, as it became necessary to focus on the issues. If a case settled late, then the premium would very fairly reflect the proportion which had been assumed at the outset.

  74. With regard to the calculation of the premium rate, Mr Smith pointed out that this had to be done case by case. It is a small book, it is immature and other people have been in that market, have quoted rates, and have had to exit the market because their rates were unsustainable. First Assist have to compete in that market, and, he said, if they are to acquire business, their premium case by case must be seen as realistic and appropriate by the solicitor and the client.

  75. Mr Smith was asked questions in relation to champerty. He was not sure how First Assist could run up solicitors' costs. If the case had a definition of a "win", which is the Law Society definition, as soon as there is a win the cover comes to an end. There is nothing which First Assist can do about it. He did not see how First Assist could force the solicitor to incur costs. He agreed that the company has an ability to control the costs in losing cases, which is, he said, quite different, but does not go to the premium.

  76. The policy contains a provision that a subsequent assessment which reduces the insured solicitor's normal fees, or the level of the success fee, will not affect the premium. Mr Smith explained:

    "… the reason for that is if we charge an up front premium there would be no reduction if solicitor's costs were taxed or assessed down, so why should there be with the benefit of hindsight? We set the premium rate up-front. If we charge an up-front premium amount there would be no adjustment and we think it is inequitable that there should be a retrospective adjustment."

  77. He went on to explain that if the company found that the premium needed to be reduced, because there has been some kind of injustice or error or it facilitated a settlement, they would take a pragmatic approach to that.

  78. At my request Mr Smith clarified the premium rate in the case of Farr using the correct figure for own costs of ?84,000 as opposed to the ?112,000 used in error at the time the policy was set up. According to his calculation the premium rates should have been 193% which would have produced a premium of ?115,182.40 plus IPT (?120,941.52).

    Emmanuel Gilbert

  79. Mr Gilbert is the managing director of The Judge, an insurance broker specialising exclusively in the provision of advice and services in connection with the ATE insurance market. At paragraph 3 of his witness statement, dated 27 January 2005, Mr Gilbert explains:

    "I set up The Judge in April 2000 as an advisory source and started broking in 2001. The Judge is deliberately not a tied agent. Its business purely comes in from solicitors. The Judge has broked about 2,500 cases since November 2001 on behalf of about 700 firms of solicitors. There has been a 50% increase in cases submitted on an annual basis. We reject about 40% of cases on sight of the proposal forms and these do not go to the insurers at all."

  80. He states that the Pursuit policy is the only product which offers an unlimited indemnity with a deferred premium payable only on success. Whilst some policies have premiums payable on a stepped basis he knows of no others which are based on a sliding scale.

  81. On receipt of a proposal he says (paragraph 5):

    "We would usually send a proposal to the most suitable three or so of them [ATE insurers] on any occasion. I would regard it as a tight market so far as the availability of ATE insurance underwriting capacity is concerned. There are only a limited number of insurers active in the market. In most cases it would be rare to get as many as three quotations back accepted … Outside the mainstream personal injury field the market is a declining one and is far from vibrant. This is particularly the case with clinical negligence where there are only a handful of providers remaining in the market."

  82. His assessment of the Pursuit policy is as follows (paragraph 8):

    "Pursuit is a somewhat different product than some of the alternatives in the market in that it operates on a sliding scale and does not have a specified premium figure at the outset. At the time we supply the quotation to solicitors we take pains to make it clear how Pursuit operates, particularly to any solicitors who may be unfamiliar with the product. When in discussion with solicitors, I seek to tell them that there is a parallel between the premium and their success fee, with the risk for solicitors bearing their own fees being comparable to the risk being borne by the insurers. I also explain that the risk for insurers can in fact be substantially greater than theirs, because the policy covers adverse costs, adverse disbursements and own disbursements. I also tell them that the nature of the policy can also be an incentive for the other side to settle, because the cost of the premium rises the longer the case proceeds."

  83. Mr Gilbert explains that The Judge does more business with First Assist than with any other provider. In his experience they deal with claims promptly and the company is a substantial organisation in terms of financial backing and continuity of service. He describes the availability of a deferred premium as a strong selling point. He suggests that banks generally have little appetite for lending in this area and if a loan is arranged it will be made to the solicitor's firm not to the client. The availability of deferred premiums elsewhere is only possible by specific arrangement.

  84. With regard to his own commission Mr Gilbert explains (paragraph 13):

    "We receive a deferred commission of a general rate of 10% from First Assist conditional upon the successful outcome of the case."

  85. He suggests that some brokers may receive up to 20% commission from other insurers. He suggests that the effective rate of commission is below 10% because no commission is payable on unsuccessful cases.

  86. In his second witness statement Mr Gilbert comments on the feasibility of solicitors trying to obtain policies for their clients by individual requests to insurance providers rather than through a broker (paragraph 2):

    "Where insurance cover is provided under a delegated authority scheme, it is a condition of the scheme that the solicitors must insure all their CFA cases using that particular policy and for the policy to be incepted at the same time that the CFA is entered into, usually before any letter of claim is dispatched. Solicitors who are not part of the insurer's panel authorised to work on a delegated basis could not access that type of policy. … The usual period of time for acceptance is 14, 21 or 28 days. Once a claimant has been declined cover by one insurer, that has to be disclosed in any future insurance applications and in my experience if a proposal has been turned down by two insurers it is almost "certain death" to the prospects of sourcing insurance from any other provider."

  87. Mr Gilbert goes on to explain that the majority of policies listed in Litigation Funding or on The Judge website are provided by insurance intermediaries, who are not themselves underwriters, who have the authority to bind underwriters within the limits of the authority which they hold from the underwriter. In respect of his own website he points out that it contains a general warning:

    "Premiums shown here are indicative only and may vary depending on a number of factors which might include the stage the case has reached, whether liability is in dispute, the prospects of success, etc."

    There are also other more specific warnings relating to particular policies.

  88. Mr Gilbert goes on to consider the policies apparently available in respect of Clarke v Tom James, Sandiford v Prices Patent Candles Ltd, Baker v Euromark (although he was not personally involved in this case) and Baker v Addenbrookes. In respect of each policy he sets out the reasons why in his view such a policy would not have been available. It is not necessary for me to set out all the policies and the reasons that in his opinion they would not be available, but the reasons include that: the policy could not be used unless the Claimant's solicitors insured all their business through the particular scheme; the policy was not available or not funded at the particular time; in respect of some policies applications were actually refused; a number of policies were underwritten by NIG so that a refusal in respect of one policy would mean that applications under other policies funded by NIG would also be refused. A number of policies were available only for specific types of claim, such as personal injury, housing disrepair and employment. In respect of Baker v Euromark Mr Gilbert thought that LPL might possibly have provided a quotation but did not think this was likely given the possible criminal element involved. He also thought that M Young Legal Associates and Oracle Legal Services might have provided a quotation. Other possible companies were Saturn Professional Risks, Composite Legal Expenses, Lit Co, Wren Insurance Services, Glen Shield Ltd (in liquidation in summer 2003), NFL, QLP and Vision Underwriting. In respect of Baker v Addenbrookes Mr Gilbert thought that Temple Legal Protection might possibly have provided a deferred premium quotation.

  89. In his first witness statement Mr Gilbert gave details of the individual Test Cases in which he was approached to arrange insurance (paragraphs 16-21).

  90. In the case of Baker v Addenbrookes The Judge received an initial enquiry from the solicitors Scrivenger Seabrook on or about 15 March 2002 and copies of the papers were sent out on 19 March to First Assist for consideration. No other insurers were approached as Scrivenger Seabrook had specifically requested a deferred premium. First Assist requested further information and the solicitors submitted a completed proposal form which was forwarded to First Assist on 15 May 2002. First Assist issued a quotation on 20 June 2002 for a premium rated at 172% of the Claimant's solicitor's normal fees plus IPT. This was accepted by the Claimant on 10 September 2002.

  91. In Clarke v Tom James The Judge received a proposal and supporting documents from the solicitors Charles Russell on 28 March 2002. Copies were sent out to First Assist, Amicus, M Young Legal Associates and Lit Co on the same day. On 11 April 2002 First Assist asked for costs estimates and a medical report. All the other insurers also asked for medical evidence. A medical report was forwarded to all insurers on 22 July 2002. First Assist was the only insurer to provide a quotation, which they did on 27 August for a premium rated at 122% of the Claimant's solicitor's normal fees plus IPT. This was accepted by the Claimant on or about 28 August 2002.

  92. In the case of Sandiford v Price's Patent Candle Co Ltd The Judge received a proposal and supporting documents from the solicitors Ormerods on 31 May 2002. Copies were sent to First Assist only, since no other insurers on The Judge's panel at that time were prepared to consider the case. This was because it was a claim for personal injuries caused by stress at work which, in Mr Gilbert's experience, is the most difficult type of case to arrange insurance for. Further information was requested by First Assist on 6 June and a quotation was issued on 26 July for a premium rated at 160% of the Claimant's solicitor's normal fees plus IPT. This was accepted by the Claimant on or about 20 August 2002.

  93. In cross examination Mr Gilbert was asked about the cost of the Pursuit policy at the early stages in the case of Baker v Addenbrookes. Mr Bartlett suggested that even if the case had settled at the outset the premium would have been about ?20,000 plus IPT with a costs risk of ?4,000. Mr Gilbert explained:

    "insurance policies are not that black and white. It's a mechanism to try and spread the risk because you could have had a premium which is charged on the full amount which is payable from day one and then it settles the day after, and that in certain circumstances can end up being more expensive than a policy like this … If you draw a comparison with a case which is insured early on, say for ?100,000 cover, and you might pay a premium of between 20 and ?30,000 for that typically for a non- personal injury case. Pre- proceedings the only liability for the insurer is always going to be disbursements but the premium is still ?20,000 to ?30,000. So the thing you are describing is not unique to First Assist. "

  94. Mr Gilbert went on to explain that any set of circumstances which deviate from a normal personal injury claim attracts significant loading or causes the insurers to say that they do not want to insure. All insurers have underwriting criteria and they prefer that the cases fit in with the norm. Anything that falls outside that is either rejected or loaded.

  95. In relation to top-up cover Mr Gilbert suggested that if the limit of indemnity is reached, the premium rate for top-up cover is likely to be 50% of the increased limit of indemnity payable in advance.

    THE CLAIMANTS' WITNESSES

    Mark Scrivenger

  96. Mr Scrivenger was the solicitor for Deborah Baker in relation to her claim against Addenbrookes NHS Trust. He has extensive knowledge of clinical negligence claims, his firm having been a member of the NHSLA Defendant panel until 2001. Since 1 April of that year he has undertaken claimant clinical negligence. He practices solely in that field, although his firm also undertakes personal injury work.

  97. Mr Baker contacted Mr Scrivenger in June 2001 some two years after his wife's hospital treatment. Mr and Mrs Baker had been attempting to negotiate compensation direct with the hospital. This attempt had come to nothing. In September 2001 Mrs Baker entered into a CFA and Mr Scrivenger explained to her, and her husband, the principles of ATE insurance.

  98. Mr Scrivenger obtained a strongly supportive expert opinion and he then approached Litigation Protection Ltd (LPL). He discussed the possibility of insurance with LPL on 14 December 2001 and then completed a proposal form and sent it to Mrs Baker, who duly signed it and returned it. On 4 February 2002 LPL replied offering insurance in the following terms:

    "We can advise that for a limit of indemnity of ?100,000 the premium payable, for our Clinical Justice Plan will be ?13,230 being ?12,600 together with insurance premium tax at the current rate of 5% of ?630.

    We can advise that for a limit of indemnity of ?75,000 the premium payable for our Clinical Justice Plan will be ?9,922.50 being ?9,450 together with insurance premium tax at the current rate of 5% of ?472.50."

  99. The letter went on to state that the quotation was open for a period of 21 days and continued:

    "If cover is not taken out in this period it may or may not be available at a later stage depending on the circumstances although the premium required will certainly be higher. If you are unable to accept this quotation within 21 days please advise before the expiration of this period.

    You should ensure that the level of indemnity you have sought will provide full cover for the entirety of the proceedings. Because of relevant actuarial evidence most underwriters are reluctant to consider later increases to the indemnity particularly after the issue of proceedings."

  100. The letter concluded:

    "I trust the premium is acceptable to your client and look forward to receiving your cheque in order that we may issue the policy documentation. If your client is obtaining funding for the premium we need confirmation of their application, in writing, within 21 days."

  101. Mr Scrivenger explained that the Baker's financial position at the time did not allow them to make an immediate payment of a sum which was considerably larger than the cost of experts' fees which they had been able to meet themselves from time to time. Mr Scrivenger states (paragraph 12):

    "I therefore approached a lender with a request for a loan by way of a credit agreement under the Consumer Credit Act 1974. The loan was to be used to pay the premium. Mrs Baker signed such an agreement with the proposed lender First National Bank Plc (FNB)."

  102. The agreement and documentation was sent to FNB by letter dated 18 February 2002. Despite numerous phone calls no response was received from FNB. Mr Scrivenger spoke to Mr Gilbert of The Judge in March and to LPL in April. LPL told him they were looking for alternative sources of funding and that other solicitors had raised concerns about FNB.

  103. In March 2002 Mr Scrivenger sent to Mr Gilbert relevant papers in Mrs Baker's case for consideration and a copy of the expert's report (paragraph 16):

    "I requested he investigate possible funding terms on behalf of Mrs Baker, I also pointed out that I had to issue proceedings by the end of July 2002, so time was short.

    ...

    18. Mr Gilbert returned to me with the possibility of a policy with First Assist. There were discussions about the nature and terms of the policy but it was also clear to me that First National were no longer effective or interested, and in fact never replied to the application for a loan by Mrs Baker to pay the premium to LPL. This left the Bakers with an LPL policy which they would have to pay the premium up-front which, financially, they were not in a position to do and which left Mrs Baker in a thoroughly unsatisfactory position with limitation a few months away."

  104. Mr Gilbert was able to obtain a First Assist Pursuit policy which had the attractions of providing access to justice for Mrs Baker without exposing her to the risk of a potentially crippling financial burden if the case was lost. It did not impose an immediate financial burden because the premium was deferred and there was an incentive for the Defendant to settle as soon as practicable since this would result in a lower premium.

  105. Before Mrs Baker formally accepted the quote on the Pursuit policy Mr Scrivenger wrote to the NHS Trust on 16 August with a letter of claim, which included the following passage:

    "This letter is written to provide an opportunity for settlement of our client's claim without the additional cost of proceeding to trial including the cost of insurance cover which is available to our client."

  106. The letter also included a Part 36 offer of ?75,000 which Mrs Baker was willing to accept in order to put matters behind her. The letter of claim stated:

    "If this offer is not accepted (within 21 days) then our client will exercise the insurance cover available to her … such a settlement would save additional cost particularly for the Trust and we would draw your attention to the savings made in avoiding the insurance premium cover if such a cover is reached."

  107. The letter went on to point out that although the Trust was prepared to agree to an extension of the limitation period, the NHSLA took the view that it was incumbent on the Claimant to issue her proceedings within the three years if she wished to pursue her claim further:

    "You will therefore appreciate that we do not have the luxury of delaying the cost of insurance cover beyond the limit of the Part 36 offer."

  108. The Pursuit policy was incepted on 26 September 2002. The case settled for ?400,000 after the defence and schedule had been served on 22 July 2003. Mrs Baker's costs, including the success fee but excluding the ATE premium, were subsequently agreed by negotiation. The costs were settled at a figure in excess of 90% of the figure claimed.

  109. Other policies identified by Mr Burrell from The Judge website as being available in the case of Deborah Baker included four provided by LPL and one each from Amicus, Mike Young Legal Associates, QLP and Wren. All of these suffered from drawbacks which made it impracticable to utilise them.

  110. In cross examination, when asked whether in clinical negligence cases the claimant's costs are nearly always higher than the defendant's costs, Mr Scrivenger agreed and pointed out:

    "Yes there are different functions. The claimant has to build a case and that takes a lot more time and effort, whereas the defendants try and destroy it and that takes a lot less time."

  111. When First Assist quoted for the policy it stated that the premium would be 158% of normal fees. On the estimated costs of ?17,500 this would produce a premium of ?27,650. Mr Scrivenger wrote to Mr Gilbert on 5 April at some length querying the level of the premium. He explained:

    "My client and I were both astonished about the level of premium but you have to take the product as you find it and it is insurers who run the market not the lawyers. These were observations that I made as a clinical negligence lawyer but of course bear no relation to the rating set by the insurance industry. I might also say that the limit of ?75,000 - ?100,000 I would not be comfortable with these days."

  112. Mr Scrivenger was concerned that there might be an issue of recoverability for the full amount of the premium. The premium rate had in fact increased to 172% of normal fees because of a miscalculation on the part of the underwriter. In respect of recoverability the underwriter at RSA indicated that the company would be willing to send an open letter to the solicitors in these terms (4/126/1088):

    "In the event of the costs and the premium being reduced upon assessment we would be agreeable to taking a commercial view in such a situation and that we would consider either accepting a reduced premium based on the actual assessment of the costs or limiting any sum in respect of the shortfall that we would seek from the claimant from damages depending upon the reason for the reduction in the costs claimed."

  113. Mr Scrivenger did not recall seeing such a letter, although it seems that he had obtained some qualified comfort as a result of a conversation with Mr Gilbert.

  114. In relation to Mr Scrivenger's original estimates of costs for the purpose of the proposal form, it was put to him that his estimate of ?17,500 was quite a long way out from the actual figure of ?30,261 up to the time of settlement. He explained that there were two main factors for that: firstly chasing up the causation problems which had been discovered; and secondly the cost of preparing the schedules. He said he was looking at the case still as a defendant, having only recently changed to undertaking claimant work. He agreed that it could happen that solicitors' estimates of their own costs could be a long way out as a result of an unexpected turn of events "they have a life of their own sometimes".

  115. At the time the policy was taken out Mr Scrivenger did not understand that the estimates which he was giving would affect the premium rating and the size of the premium, nor that they would depend on, among other things, the relative proportion of the estimates of claimant's and defendant's costs. Nonetheless he felt that he had given careful estimates and carried out a thorough exercise, but estimates of costs are often imprecise "because things can happen".

    William John Vallance

  116. Mr Valance is a partner in the firm AMS Law, and has been a personal injury lawyer for 22 years. He handled the claim on behalf of Anthony Baker v Euromark. Mr Baker sustained an injury to his left hand in August 1999 and instructed his union solicitors Messrs Thompsons. They wrote a letter of claim to the Defendants in March 2001 and the Defendants denied liability. In August 2001 Mr Baker was advised by Thompsons that they did not believe there was any reasonable prospect of establishing legal liability for his injury. They were unable to recommend that any further steps should be taken in respect of the claim. Thompsons were particularly concerned with the lack of any independent witness evidence and the Defendants' contention that they were not vicariously liable for the incident, since their employee who had caused the injury was acting outside the scope of his employment.

  117. Having discussed the case with Mr Baker, Mr Vallance agreed to act on his behalf and a CFA was entered into on 14 November 2001 with a success fee of 100%. The limitation period was due to expire on 19 August 2002.

  118. Mr Vallance discussed insurance with his client and found that he had before the event insurance with Zurich Insurance, arranged through Capita Assistance. On 5 January 2002 Capita Assistance wrote to Mr Baker advising him that his cover was restricted to telephone helpline advice only. Mr Vallance entered into further correspondence with them but on 16 May 2002 Capita rejected Mr Baker's claim for insurance cover. Mr Baker also had a household legal expenses insurance policy with MSL Keystone but they too rejected the request for cover on the basis that the claim to them was made outside the 180 days notification period, and because the claim had already been rejected by Mr Baker's union.

  119. Having received the papers from Thompsons Mr Vallance wrote to the Defendants' insurers, AXA, inviting them to reconsider their denial of liability and to confirm that vicarious liability would no longer be raised as an issue in the proceedings. In the event the vicarious liability issue remained live until the defence was served on 7 January 2003.

  120. After Capita's rejection, on 16 May 2002, Mr Vallance realised he would have to obtain ATE insurance cover from another source. He stated (paragraph 13):

    "I am aware of a range of different ATE products available in the market as a significant proportion of my caseload is conducted pursuant to CFAs backed by ATE products. I use a number of different ATE providers and I am not tied to recommending any particular product. I advise my clients on a case by case basis depending on the circumstances. It is not possible to give standardised advice in this regard. I must consider whether the proposed ATE product is suitable for my clients in accordance with my professional responsibility."

  121. Mr Vallance was of the view that in order to have any realistic chance of obtaining ATE cover it would be necessary to have counsel's opinion. This was obtained, on 11 June 2002, and counsel put the prospects of success at 60%.

  122. Having obtained counsel's advice Mr Vallance applied to Amicus Legal Ltd, on 14 June 2002. That application was rejected by Amicus on 20 June and the rejection was affirmed on 27 June on the basis that the company did not believe the Claimant had sufficient prospects of success due to lack of corroborative evidence. Mr Vallance then applied to Eastgate Assistance on 16 July. That application was rejected on 19 July because they were not then accepting any new proposals. By this time the limitation date was approaching.

  123. Forming the view that further applications for standard ATE insurance were unlikely to be successful he submitted an RSA Pursuit proposal form on 26 July 2002. A provisional quotation was given on 29 July, subject to further information being provided, and a formal quotation was given on 12 July which assessed the premium rate at 74% of normal fees. The workings of the policy were explained to the client, who professed himself happy to proceed with it, and, before the proposal was accepted, a further conference with counsel was arranged. Proceedings were issued on a protective basis in the Sheffield County Court on 17 August 2002.

  124. Counsel advised that in his view it was reasonable to take out the RSA policy but suggested that the Defendants should be put on notice of the basis of assessment of the premium. On 4 September 2002 Mr Vallance wrote to AXA Insurance:

    "Our client has received an offer of after the event insurance from Royal and Sun Alliance pursuant to their "Pursuit Protection Plan" this offer of insurance remains open until 11 September 2002. The premium for this policy cannot be identified at the outset, as it is based on a percentage of our normal fees (i.e. basic costs under the conditional fee agreement) calculated upon conclusion of the case. It is therefore in the party's interest to conclude a settlement of this case sooner rather than later.

    To this end we enclose a claimant's Part 36 offer to settle the issue of liability."

  125. No response was received from AXA. Subsequent telephone calls and further faxed copies of the letter to AXA failed to produce any response. On 3 October 2002 the signed quotation was sent to RSA. RSA sent the policy documents to Mr Vallance on 16 October. Mr Vallance continued telephoning AXA on various occasions up to 13 October without making any material progress. Finally, in response to a further letter from Mr Vallance he received a reply from AXA dated 26 November 2002 stating:

    "We see no reason to alter or previous stance in this matter. Both liability and causation are denied and we leave it to your client to take whatever steps he considers appropriate."

  126. The case was eventually settled in the sum of ?1,250 plus costs pursuant to a consent order dated 25 September 2003. Mr Vallance stated (paragraph 39):

    "Quantum was considerably lower than had been anticipated at the outset because of difficulties in establishing medical causation. This was one of the risk factors identified in the CFA with my client."

  127. For completeness it should be added that a claim had been made to the Criminal Injuries Compensation Authority on the basis that the injury might have been caused as a result of a criminal act, but that claim was rejected.

  128. In cross examination Mr Vallance agreed that counsel had thought that fast track was the likely track for this claim, although he said that counsel's view on quantum was potentially between ?15,000 and ?20,000. He personally took the lower of those figures, and described it as borderline between fast track and multi track. There were difficulties because the Claimant had pre-existing problems with his hands, his back and his feet and this had an impact on a potential loss of earnings claim.

  129. Mr Vallance did not accept that the claim arose out of an accident at work because it had occurred on a training course and Mr Baker was not employed at the time. He suggested that this made a difference because it meant that there was no employment legislation to rely on and it was not possible to plead any breaches of statutory duty. He did agree that there would be no need for massive cover for an ATE policy in respect of a claim which counsel thought was a fast track claim.

  130. Asked about estimating the costs he replied:

    "I generally find costs estimates to be difficult whether it is own or other parties costs. But I arrived at that figure on the basis that at the time I was making the application, the defendants had not instructed solicitors at that point, it was still being dealt with by insurers. So I felt that the defendants solicitors would be involved over a shorter time span so that their costs were likely to be less."

  131. When Mr Vallance received the quotation at a premium rate of 74% of normal fees he said he understood how the policy worked but did not know the method of arriving at the percentage, nor did he know the importance of the proportions between his estimates of claimant's as against defendants' costs.

  132. Following the conference with the client and counsel the potential damages claim was reduced to ?5,000. Mr Vallance wrote to RSA on 3 October 2002 (2/30/506)

    "It no longer seems likely that we will be able to attribute a loss of earnings claim to the accident at work and, in the circumstances, our expectation of a successful outcome would be ?5,000."

  133. That paragraph was contained in the letter returning the Pursuit quotation duly signed.

  134. Mr Vallance confirmed that in arriving at his estimates he did his best to give an honest estimate, and the estimates which he gave in the allocation questionnaire six months later were consistent with the estimates he gave to the insurers.

    Amanda Stevens

  135. Amanda Stevens is a solicitor with Charles Russell, of Guildford. She handled the claim of Gary Clarke v Tom James of London Ltd.

  136. The Claimant had originally consulted Hart Brown, solicitors, in March 2001. They were unwilling to enter into a conditional fee agreement. They wrote a pre-action protocol letter to which the Defendants' insurers replied on 24 August 2001 denying liability. As a result of a change of personnel at Hart Brown Mr Clarke consulted Ms Stevens and requested that she act for him. She has specialised in personal injury claims since 1990 and is a member of the National Executive Committee of APIL.

  137. Ms Stevens received Mr Clarke's file in February 2002, and, having ascertained that he did not have any before the event insurance, at a meeting on 7 March 2002 a CFA was signed and instructions were taken concerning ATE insurance. She explained (paragraph 4 of her first witness statement):

    "I advised Mr Clarke to use the legal expenses insurance brokerage service provided by The Judge.Co.UK.Ltd (The Judge) whereby he would complete one proposal form which would be sent to a panel of five or six after the event legal expenses insurance providers. I considered this to be a cheaper and more effective way of obtaining several quotes simultaneously than approaching insurers direct. I also advised Mr Clarke that by requesting five quotes he would be well prepared to counter any subsequent arguments by the Defendants that the premium was unreasonable.

    ...

    5. An insurance proposal form was forwarded to The Judge on 27 March 2002 and on 2 April I received a letter from The Judge advising that the application had been submitted to Amicus, Mike Young Legal Associates (MYLA), Litco and First Assist being the only insurance providers on their panel that The Judge considered might offer insurance for this particular case.

    ...

    Emmanuel Gilbert managing director of The Judge advised me on 11 April that two of the insurance providers had responded to date and that they had explained no insurance quotation would be forthcoming until medical evidence had been obtained."

  138. Following this Ms Stevens obtained a medical report which was forwarded to The Judge on 19 July. She requested that it be dealt with rapidly as she was of the view that proceedings needed to be served by 1 September. She continued:

    "Mr Clarke also asked me to enquire about the option of a deferred premium as he was already paying out money for disbursements and had concerns about the other expenses he would have to pay in pursuing the litigation. I spoke again with Emmanuel Gilbert on 1 August 2002 and he advised that MYLA were not able to quote on this case for their disbursement funding policy. He mentioned one other disbursement funding policy, which might be available with First National Bank, but explained that Charles Russell would have to register with the bank and show the last two years accounts before being accepted, so it was not going to be easy to get cover up and running in time for this claim. This avenue was therefore not pursued."

  139. MYLA were not prepared to offer adverse legal costs cover for Mr Clarke's case and no further quotations were forthcoming from the other insurance panel providers but a quotation was supplied by First Assist following the provision of estimates of costs and disbursements by Ms Stevens. The client was advised and the fact that the premium was deferred was attractive to him because of his personal financial circumstances. The quotation was accepted and the policy was issued on 29 August 2002. On the following day notice was served on the Defendant with the rest of the proceedings.

  140. With regard to the suggestion that the ATE policy ought reasonably to have been taken out at an earlier stage Ms Stevens said (paragraph7):

    "It is not necessary to protect a client from adverse costs before proceedings are served, as there can be no liability for such costs at that time. In this case it took five months for a quotation to be obtained as the insurers were unwilling to provide one without medical evidence. I had applied on Mr Clarke's behalf for insurance as quickly as possible after receiving his instructions to pursue the claim."

  141. Ms Stevens pointed out that the Defendants refused to concede liability or medical causation until the case was already within the first trial window. She herself had tried to bring matters to a speedy conclusion obtaining a default judgment for Mr Clarke on 4 October 2002 and making a Part 36 offer on 21 October to settle the claim for ?13,000. The Defendants rejected the offer and applied to set aside the judgment. Attempts to discuss the matter with the Defendants' lawyer were met with the response that she was without instructions to enter into any without prejudice negotiations. Another employee of the Defendants had recently brought and settled a claim based on similar grounds. This was brought to the attention of the Defendants' solicitors but did not result in a settlement. It was not until the judgment had been set aside, and disclosure and exchange of experts' reports had taken place in June 2003, that an offer to settle in the sum of ?20,000 was made by the Defendants which the Claimant accepted. Ms Stevens points out that if the Claimants Part 36 offer had been accepted the premium payable would have been approximately ?10,455.

  142. Ms Stevens made a second witness statement to deal with the evidence of Adam Burrell, who suggested that there were other policies available which would have been cheaper than the First Assist policy. She said (paragraph 2):

    "It is my belief that the ATE insurance market for RSI cases was virtually non existent. I have formed this view from discussions with colleagues in the personal injury world outside my own firm whom I meet regularly at training and social events and where we naturally discuss our cases on an anonymous basis."

  143. She exhibited letters from the President of APIL and the Head of Underwriting at DAS, both confirming the difficulty in trying to insure RSI cases. The letter from DAS states:

    "The comments I have made in respect of RSI and stress claims would apply equally to not only our own DAS ATE product, called Conditional Fee Insurance Services and subsequently renamed 80E, but also to other products underwritten by DAS namely Litigation Protection, Greystoke Legal Services, QLP and Saturn Professional Risks."

  144. In her statement Ms Stevens goes through a number of insurers identified by Mr Burrell, giving reasons why an approach would not have been successful. These companies were normally not suitable, since they, for the most part, required the firm of solicitors to be a member of a panel or to guarantee a minimum number of cases for insurance, neither of which Ms Stevens was in a position to provide. She explained:

    "The nature of my practice has not really come out through the witness statement but I only handle about 10 cases at any one time. That was true in 2002 and it is true now. They are all very different cases and when I have gone to various providers to ask about being accepted onto their schemes, they have said that they do not find me an attractive proposition for insurance purposes.

    ...

    I seem to have the very difficult cases. I get referrals from other solicitors perhaps when they turn down the case on merits but the client wants to go on and they will ask us to have a look at it. I have catastrophic injury cases, a number of very difficult brain injury cases, RSI cases. The sort of things that you cannot handle too many at one time and do them properly."

  145. She explained that she uses The Judge because she found Mr Gilbert very knowledgeable about the market, and:

    "I like the idea of having one form, one small fee and somebody else who is more knowledgeable about the market than I was who could get simultaneous quotations."

  146. In cross examination Ms Stevens was asked about her risk assessment (60%) of the client's chances of success. She referred to the risks set out in the CFA:

    "The fact that we have not seen your employer's file, which may include documents prejudicial to your claim.

    The claim could fail on the issue of medical causation and we do not yet have a medical report.

    The employer may produce witnesses who deny your method of handling work trolleys was the appropriate and approved method.

    We could fail to beat a Part 36 offer."

  147. She explained:

    "I always find when you're handling claims from employees, you never quite know what is going to happen. You receive the client in the office and they tell you their account, but it's the strange situation where the defendant actually knows them and their working practices much better than you do.

    ...

    You never quite know what is going to come out of the woodwork in terms of other employees coming forward with evidence. That's one big risk factor. On the liability side there was also a risk factor on medical causation."

  148. Ms Stevens was asked whether, had the case come to her 12 months earlier, she would have been able to investigate the case and perhaps get ATE insurance, and then issue proceedings afterwards. She did not believe that this would have made a material difference. Her client was "absolutely strapped for cash" and her firm does not have facilities to bankroll clients, nor at that time any system for deferred payments with doctors. She could not therefore have applied for ATE insurance any earlier.

  149. She had used the services of The Judge on two previous occasions and was expecting about five quotes. In completing the proposal form she was very concerned to work out whether it was going to be a fast track case or not because that would affect the viability of the claim, in addition to which she had to obtain partner approval for any CFA entered into.

  150. Ms Stevens provided The Judge with an estimate of costs through to trial of ?15,000 from the date of the CFA on 7 March 2002. That figure was intended to refer to a one day trial. However in the same email the estimate for expert's fees anticipates a two day trial, and it does not include counsel's fees. This was an error which she cleared up with the insurers in the subsequent email correspondence. On 20 August 2002 Ms Stevens gave an estimate of costs of ?40,000 following a conference with counsel. She asked for a quotation from The Judge based on ?40,000 adverse costs. In a subsequent email of 28 August she explained:

    "Our estimate of costs of ?35,000 to ?40,000 is our estimate of the total costs (including disbursements of ?6,500), that we expect to incur in taking this case to trial. When assessing the adverse costs risk, we have assumed that the defendant's hourly charge out rate will be no higher than our own (frequently insurance companies require their panel solicitors to work at a lower hourly rate than the court norm). We cannot see that they are unlikely to incur any less costs than we will and they will no doubt have to interview witnesses as well and are likely to get their own expect evidence in a case of this type. I am not sure whether this will affect your calculations, but I feel it would be acceptable for the adverse risk to be ?35,000 with ?6,500 cover in respect of own disbursements so potential risk to insurers of ?41,500."

  151. Ms Stevens clarified her email stating:

    "What I actually meant was that I didn't think that they would need any less hours on the job than we would … and that they will no doubt have to interview witnesses as well and are likely to get their own expert evidence in a case of this type …"

  152. Ms Stevens did not know how the First Assist premium was calculated, nor did she have any knowledge of the significance of the estimate for the figure for own costs and the other side's costs in that calculation.

  153. When the offer of insurance came through, at a premium rate of 123%, Ms Stevens first reaction was relief that there was an insurance offer, and the second was that it was high. She wrote to Mr Clarke in these terms and summarised the cover, saying in effect:

    "This is high but we have been out to the market place and that is what it is really."

  154. She also stated that the firm would have to keep an eye on costs. Asked about the potential size of the premium she stated:

    "Well I had no intention for this claim to run a day longer than it had to. But, I had a client who had had two surgical operations on his arm for this injury caused at work. I thought he was entitled to make a claim. He wanted to make a claim. We had no other alternative in the marketplace, and I could never advise the client to go into litigation facing a potential adverse costs order without insurance so I thought I was stuck really. The client was given the information and with informed consent he wished to proceed."

  155. On 28 August a letter was written to RSA asking for a letter of comfort to deal with the question of shortfall and recoverability of the premium. The letter points out that in other cases insured with RSA such a letter had been supplied. RSA replied on 29 August 2002:

    "In the event of the costs and the premium being reduced upon assessment at our discretion we would be agreeable to taking a commercial view of such a situation and that we would consider either accepting a reduced premium based on the actual assessment of the costs or limiting any sum in respect of the shortfall that we would seek from the Claimant to the amount of her damages depending upon the reason for the reduction in the costs claimed."

    Kenneth Besfor

  156. Mr Besfor is a Fellow of the Institute of Legal Executives with Messrs Ormerods Solicitors. He handled the claim of Valerie Sandiford v Prices Patent Candle Company Limited. He was admitted as a Fellow in August 1981 and has considerable experience in personal injury litigation. Valerie Sandiford was originally represented by another firm of solicitors in respect of her claim in the Employment Tribunal against the Defendants. She consulted Ormerods on 23 February 2001. They agreed to act for her on a privately funded basis. At that time the claim for personal injury damages was still to be considered. This was first discussed with counsel in about April 2001. Mr Besfor stated (paragraph 6):

    "I was aware that this would be a difficult and risky claim to pursue on behalf of the Claimant and that therefore it might be difficult to obtain after the event insurance. I was satisfied that the client that she had no alternative method of funding and even applied for public funding from the Legal Service Commission despite the provisions of the Access to Justice Act 1999. However this proved to be unsuccessful …

    ...

    7. A proposal was submitted to Litigation Protection Ltd on 2 May 2002 a company which had successfully provided after the event insurance in the past. Unfortunately the application was returned, it would appear unread, with the blanket reason that stress at work cases were not covered by them … Efforts were also made to contact other after the event insurance providers including Eastgate Assistance, ENA Underwriting Ltd, Temple Legal Protection, Watkins Davies Insurance Consultants Ltd … Eastgate Assistance and Temple Legal Protection simply could not be contacted. I cannot now recall why ENA Underwriting Ltd and Watkins Davies Insurance Consultants Ltd were not of assistance but I am certain that they were not. It was then decided to utilise the services of The Judge.Co.UK.Ltd and proposal forms were submitted to them on 30 May 2002. It was indicated by "The Judge" that a number of potential insurance providers would be nominated. However due to the difficulties of pursuing a case of this nature they were only able to provide a quotation from the First Assist Group Ltd.

    8. Consideration had also been given in addition to the above insurance providers to the services of Accident Line of which my firm were a panel member. There was a delay in pursuing this option as at the time my firm were still putting in place a funding arrangement to compliment that scheme. In any event an application to Accident Line was also refused."

  157. The quotation from The Judge was accepted on 16 August 2002. The quote was put to the Claimant on the basis that it was the only quote which the solicitors were able to obtain.

  158. Mr Besfor confirmed in cross examination that he did not understand fully how the policy worked. He indicated to the client that he thought the premium rate of 160% of normal fees was very high, but that it was the only offer which they had had. Counsel advised that the claim was worth ?50,000 plus and it eventually settled for ?44,000. Counsel had indicated that the maximum award might have been as high as ?100,000.

  159. After settlement Mr Besfor wrote to The Judge on 22 August 2003 stating:

    "Please contact the insurers on our behalf and thereafter inform us whether or not the premium calculation is correct. It would be helpful to establish, in view of the substantial amount of the premium, whether the insurers would accept a premium based on the costs which are ultimately agreed, not on the costs set out in the bill of costs."

  160. He said that this was sent at the prompting of his costs draftsman. There was some confusion about the actual amount of the premium. Mr Besfor explained:

    "This was a sort of premium that I had not dealt with before and I wanted to ensure that I could get any assistance that was necessary."

    Peter Henry Evan Bennett

  161. Mr Bennett is a solicitor with Dolmans, solicitors, who handled the claim on behalf of Howard Farr v Kerslake. Mr Bennett has been with Dolmans since qualifying in January 1995. The claim was initially conducted under a BTE policy issued by the First Assist Group which provided cover up to ?25,000 in respect of all costs and disbursements in the action, including opponent's costs, any surplus being utilised to pay own costs. The BTE policy proved satisfactory until it became necessary to commence proceedings. Once proceedings had been commenced, and the Defendant's defence served in October 2002, it became clear that the BTE policy would be insufficient to cover both party's costs in the action. Enquiries of the BTE insurers received the response that top-up cover was available, but only on the basis that a CFA was entered into with the client. The CFA would then be underwritten by their bespoke ATE cover which was unlimited. This would enable the ?25,000 BTE cover to be used for payment of disbursements. A CFA was accordingly entered into with Mr Farr and ATE insurance arranged with First Assist. On 28 March 2003 Mr Bennett wrote to the Defendant's solicitors stating:

    "Please note that Royal Sun Alliance's premium for issuing this (ATE) policy of insurance has been calculated by reference to the risks inherent in the action and with specific reference to the defence put forward by your clients.

    Further, this premium is also calculated by reference to our costs in conducting the action and the extent of any adverse costs as estimated by us by reference to the information on that issue that you have provided to the court and/or this firm …

    Specifically, the amount of the premium is calculated as a multiple of our base costs, prior to the addition of our CFA success fee, plus insurance premium tax at the prevailing rate."

  162. The letter pointed out that the longer the issue of liability remained outstanding the greater the premium would be. In response to a request from the Defendant's solicitors Mr Bennett informed them that the multiple in question was 2.57, i.e. 257% of normal costs.

  163. On 31 July 2003 a Part 36 offer was put forward on behalf of the Claimant offering to apportion liability on an 85/15 basis in favour of the Claimant. That offer was rejected. On the first day of trial, on 29 September 2003, the case was settled for ?250,000 while the Claimant was in process of giving evidence. The Defendants had made a Part 36 payment of ?100,000 on 28 August 2003.

  164. In a third witness statement Mr Bennett gave further details of his correspondence with First Assist on the subject of insurance cover, both BTE and ATE, and pointed out (paragraph 7):

    "By way of example the Defendant's legal costs up to and including the liability trial were predicted by their solicitors, to be in excess of ?92,000. There was a very real possibility of this case being fought equally hard on quantum, if Mr Farr had succeeded at the liability trial on 29, 30 September 2003 … Upon that basis, with adverse costs of liability only being in excess of ?92,000 it is, in my submission, easy to see how the funding issue could have resurrected itself, with a vengeance, following the liability trial and/or during the preparation of the Claimant's case on quantum. Particularly if adverse costs were "capped" at say ?100,000 which was a fairly common upper ceiling in my experience."

  165. Mr Bennett thought that the ATE provision provided by First Assist was particularly advantageous since they were an insurer who was already familiar with the case through their involvement as BTE providers. First Assist specifically informed him that they would be assessing the case from their current knowledge of the matter as BTE insurers.

  166. Mr Bennett was asked about the policies indicated by The Judge website. He replied:

    "Examples I have seen from The Judge website caused me a number of concerns … in the context of my position as a partner within Dolmans in Cardiff. The first issue is that we as a practice are a predominantly - vastly predominantly - a defendant practice. I would say about 90 per cent of our work is defendant insurer work. To that extent, because of the way the work comes to us, which is predominantly from local authorities, predominantly directly from local authorities as opposed to through insurers … On that basis, we have to be very very careful in terms of the relationships we develop or maintain … with ATE insurance providers. The reason being that local authorities particularly … are very sensitive about any of their advisors, whether they are legal advisors or otherwise having connections to the ATE market."

  167. He explained that any ATE provider who required the firm to join a panel, or fulfil some kind of pre qualification, would cause serious difficulties.

  168. Mr Bennett was asked about the suitability of the policy which Mr Wait said could have been offered in respect of this case. He stated:

    "I would have concerns about the extent of the cover at ?120,000. Because of the way that the costs developed in this case … or the way the case generally developed. I also have concerns in relation to the nature of the policy itself in the sense that Mr Wait appears to suggest that it is possible - even though it is a product offered by another insurer - to use the BTE cover in the manner that it was used in this case. I don't see how one insurers competitor with another insurer would be prepared to allow that second insurer access … to its BTE fund ..."

  169. Mr Bennett accepted that his costs estimate in respect of the figures for the quantum trial were too high, and that it was not a well prepared document having been prepared in some haste.

  170. He advised Mr Farr that he thought the policy was a good product because it had unlimited costs cover and also allowed them to use the ?25,000 BTE cover for disbursements:

    "What I said to Mr Farr at the time was that this was the best option, albeit in my view at that stage, he had very very few options."

  171. In cross examination Mr Bennett accepted that once liability had been established Mr Farr was bound to recover something in respect of damages, and that once a Part 36 payment had been made the only costs risk was of not meeting that payment into court. He felt however that even after the Part 36 payment had been made there was still a substantial risk because the payment had been made prior to the trial on liability. He had put the prospect of success at better than 75%.

  172. Mr Bennett confirmed that when completing the proposal form he did not provide an estimate of the Defendant's costs at all, but only figures for his own costs, which subsequently proved to be incorrect. He did not understand what effect the estimate would have on the calculation of the premium.

    THE DEFENDANTS' WITNESSES

    Christopher Wait

  173. Mr Wait is the owner and founder of Temple Legal Protection Ltd, an underwriting agency similar to First Assist. He has worked continuously in the insurance industry for 19 years since 1985, having been a liability underwriter and continuously employed in the legal expenses market since 1993. In his witness statement, dated 12 January 2005, he explains (paragraph 1):

    "The services provided by Temple are carried out exclusively for the provision of a full range of legal expenses insurance both before the event legal expenses (BTE) and litigation insurance, ie after the event insurance (ATE). Temple provides ATE insurance to all categories and classes of litigation including personal injury, contract disputes, defamation, clinical negligence, professional negligence, intellectual property and marine disputes for both claimants and defendants."

  174. The underwriting and claims management services are carried out under binding authorities.

  175. Mr Wait was approached by the Defendant's solicitors in the case of Farr v Kerslake, who asked him what premium Temple would have quoted had the company been approached by the Claimant's solicitors. He says this:

    "6. I can confirm that Temple would have offered ATE insurance for this case had it been approached for a quote in or about December 2002. We had a Binder from Lloyds of London which allowed us to rate and condition such cases up to a limit of ?250,000. On the basis of the information provided to First Assist in requesting cover I confirm that Temple would have provided a limit of indemnity of ?150,000 for a premium of ?22,400 plus IPT. The limit of ?150,000 allows for an estimate of ?120,000 adverse costs and allows for ?22,400 plus IPT cover for the premium.

    7. In relation to the method of calculating the premium I can state as follows:

    (i) I would pay particular attention to the advice given by the Claimant's solicitors and their counsel;

    (ii) from my own experience of these cases I would then have made my own assessment of the likelihood of success;

    (iii) in this case I would regard the chances of success as high;

    (iv) assuming a loss of about ?112,000 I would have calculated the pure burning costs at about ?14,000;

    (v) I would have added a percentage for profit and insurance administration costs of 60% or ?8,400;

    (vi) the premium would be on the basis that the BTE policy (with a cover of ?25,000) would have been used for disbursements."

  176. Mr Wait was asked what difference it would have made to the premium which he proposed, had he known that the figure for costs was ?84,000 rather than ?112,000. He said it would affect the rating a little bit, because when the company looks at the limit of indemnity they always round up the estimate of adverse costs. He therefore thought the altered figure would reduce the premium, but probably only by a couple of thousand pounds, it would not make a substantial difference. He said that a figure for adverse costs of ?84,000 would be rounded up to ?90,000 or ?100,000. He wished to ensure that the client had enough cover.

  177. He was asked whether, in the case of Baker v Addenbrookes, his company would have offered a policy. Although he had not had the chance to have a good look at all the papers, he believed that it was a case that his company would have underwritten:

    "Our policy would have offered deferred premium and protection for that premium within the policy cover, our policy would also have been a step premium policy. All of our policies since the beginning of 2002 or end of 2001 have all been written on a stepped premium basis, both for personal injury, commercial and clinical negligence, so the key features would have been premium payable at the end, with protection for the insurance."

  178. Mr Wait explained that the policy was self insuring. He went on to explain how the stepped premium works:

    "We would have added Step C premium to the limit of indemnity and then that would have calculated the limit of indemnity for us. Premium A is payable if the case settles before proceedings are issued. Premium B is paid if the case settles between the proceedings being issued and 45 days before trial, and the last window, Premium C, is payable 45 days before trial and including trial. They are the three blocks of risks that we as underwriter look at and think they are the different dynamics of the case."

    The premium is payable at the conclusion of the case.

  179. Mr Wait was of the opinion that Mrs Baker's case against Addenbrookes did not look like a difficult case of clinical negligence:

    "Given that the solicitor involved in that case was a pretty good clinical negligence solicitor we would have probably covered it. The premiums probably would have been for Band A about ?5,000, for B it would have been about ?20,000 and for C it would have been about ?50,000, I would have thought so if this case had gone to trial and settled at trial we would have wanted ?50,000."

  180. He was asked whether there would have been a lower premium had the case settled at the stage it in fact did. He replied:

    "Probably but we would need to look at the case, do our homework on the solicitor and on the medical side of things to work out, would we have covered it first, and then looked at the exposure to do a proper job on it."

  181. Mr Wait was asked in cross examination what his reaction would have been in the Baker v Addenbrookes case had he been approached by Mr Scrivenger, who, at the outset, assessed the overall prospects as low as 33%. He replied:

    "We would have queried that because if he wants to run the case and he thinks it is very unlikely to win we would have queried it ... When a solicitor tells us what the prospects are we use that as a basis for thought but we look because solicitors tell you all sorts of different things, and it is my money that is at risk, not theirs, so even when they tell us it is a great case we sometimes do not insure it because we do not believe it to be."

  182. Mr Wait was taken to a specimen of the Temple Advantage policy, and taken through the various exclusions and conditions, particularly in relation to offers of settlement and cancellation. He agreed that the company could cancel the insurance certificate immediately if there was any deterioration in the prospects of a successful outcome:

    "That is correct, although it is unlikely we would unless it was a significant deterioration, because, remember, we are going to have to go on risk and pay all the adverse costs. It is a similar situation to the top-up thing … where there is no point in a legal expenses underwriter not providing top-up in a case. … If I have got ?80,000 exposed and a solicitor comes along and says: we've got it wrong, the adverse costs order because the other side are messing us around, it may be ?120,000, if I say, no, I'm not going to give you cover then I am going to have to write a cheque out for the costs that have already been incurred, so it is in my interests to give a top- up cover."

  183. When asked about top-up cover in relation to the case of Farr v Kerslake Mr Wait stated that he was not aware that the Defendant had made an offer to settle of ?100,000. He agreed that in such circumstance it might be necessary for an insured with a fixed limit of indemnity to acquire top-up cover to enable the claim to proceed to trial. He disagreed with Mr Gilbert's evidence that the cost of top-up cover could be 50% of the top-up amount. He said that this was not his experience and they provided top-up cover "quite a lot". He explained:

    "I think the point I am making is that there are other underwriters out there, including us, that would see a request for higher limits as something that you have to deal with. If the case is okay - assuming that the case has not gone wrong when that request is made ... we are either going to have to write a cheque because the guy is going to abandon his claim because he will not have sufficient funds to carry on. [Mr Dutton suggested he might accept the offer of ?100,000 and recover his costs, and therefore the premium.] … I think that the solicitors would not be giving him advice to take ?100,000 if they thought that was wrong."

  184. With regard to the suggestion that this put pressure on the Claimant, Mr Wait suggested that the thing to do was to get a robust lawyer on his side who was not going to just cave in, and also a robust ATE insurer.

  185. Mr Wait confirmed that, in the case of Farr, the solicitor's estimate of success had been "very bullish", as was that of counsel. The figure in the proposal form was 75%, whereas Mr Wait's company rated the chance of success as 85% or above. He confirmed that if the case had been rated at 70% or 75% the premium would have gone up.

  186. In respect of clinical negligence insurance Mr Wait said the minimum premium was ?1,250.

  187. It was put to Mr Wait that certain policies, for example those offered by Litigation Protection, might charge high premiums at the outset. He stated:

    "Yes, but the thing about stand alone one-off premiums in those scenarios is that you are … buying cover from now until when the thing finishes. I accept with First Assist it is a sliding scale and with Temple it is a stepped premium similar sort of sliding scale type thing. With them, yes, if it settles early fine. If it settles very late then it is sort of swings and roundabouts for them and us."

  188. Mr Wait gave his views about competition in the ATE market:

    "It is really not in our interests to have so few players as there really are. You heard yesterday Emmanuel Gilbert's evidence about everybody gets to see these things from Litigation Funding. The reality is, when you apply to all these different providers they cannot act. A lot of them are almost quasi-claims management type companies. They are not underwriters in the same sense that First Assist and Temple and DAS and Law Club are underwriters. If First Assist are knocked out of this market, let me assure you, it is not in our interests. "

  189. He denied that Temple is only interested in firms which bring in volume work, although he agreed that they preferred to have schemes, because they do not like cherry picking. He felt, however, that the case of Farr was a genuine case and had not been cherry picked.

  190. Mr Wait was asked about correspondence with Charles Russell in connection with the case of Clarke v Tom James. He was shown a letter from Temple to Charles Russell dated 21 August 2002:

    "One of the requirements of a delegated requirement scheme is that the firm has a sufficient number of risks to insure through the scheme. We look for a minimum of six insured cases a month, so it is viable for us to monitor and audit the scheme. If you do not have a sufficient number of cases then we would be happy to receive individual submissions as long as we are assured you are not cherry picking the marginal risks for insurance."

  191. He explained:

    "We are trying to discourage them from sending us all their rubbish, which, if you come to our office … you will see the enormous amount of rubbish claims that we get all the time. We do not charge a fee for that, so yes we are trying to discourage stuff, but we are also saying, come to us if there are compelling reasons. On this Dolmans one [Farr] there would have been. On the clinical negligence we would have loved to have a dialogue with Charles Russell, they are a good firm."

  192. In relation to Farr it was put to Mr Wait that whilst Mr Farr's solicitors had suggested a 70% prospect of success, having regard to the evidential difficulties, he was using 85% as the basis of his premium calculation. He did not agree that this was too high given the facts of the case.

    Kate O'Reilly

  193. Kate O'Reilly is a Fellow of the Institute of Actuaries. Although an actuary Kate O'Reilly gave factual rather than expert evidence in her statement of 28 January 2005. She has analysed the figures for First Assist premiums in the Test Cases in the light of information given by RSA First Assist, taking into account (paragraph 5):

    "(a) the increase of premiums collected in successful cases in order to allow for the lack of any premiums in unsuccessful cases;

    (b) the deferral of payment of premiums until the end of the case; and

    (c) the fact that cover is unlimited."

  194. Kate O'Reilly states that the premium rating is based on the assumed total of costs at risk under the policy (i.e., adverse costs and disbursements plus own disbursements). Thus, she says (paragraph 10)

    "The rating is based on the assumed total of costs at risk under the policy … ; own costs are then used only as a scaling factor to link increases in own costs with increases in the premium."

  195. Later on she states (paragraph 17):

    "An estimate of costs at risk which is too high relative to own costs will also lead to a premium higher than is needed to match the risk."

  196. Ms O'Reilly demonstrates the increase in ATE premium where no premium is paid for losing cases, as against the premium (significantly lower) where all policy holders pay regardless of outcome. Thus, with a success rate of 95%, a 5.3% extra risk premium is required to provide cover for the lost cases. When the success rate falls to 50% the extra risk premium rises to 100% of the premium which would be payable when all policy holders pay regardless of outcome. She states that the final premium figure is dependent on the assumed failure rate for cases where ATE cover is provided. If the underwriter achieves a lower actual failure rate than that provided for in the premium rating formula, premiums will be higher than required. If the failure rates are higher, the premiums will prove inadequate.

  197. Ms O'Reilly goes on to set out in tabular form the impact on Pursuit premiums of incorrectly estimating failure rates. She shows this by percentages, which are a calculation of variation from the premium needed to match the risks. If the estimated failure rate is different from the actual, she states (paragraph 27):

    "For example if the estimated failure rate, when the ATE policy was incepted was 30% and the actual underlying failure rate is 20% then the premium would be 71% higher than needed."

  198. If the actual failure rate is 30% as predicted the premium will be neither too high nor too low. If the actual failure rate is higher than predicted, say 40%, the premium would be 36% lower than needed.

  199. Ms O'Reilly goes on to deal with the fact that the Pursuit policy has no limit of indemnity. She was asked to explain the actual cost of providing unlimited cover over and above a given limit of indemnity. She sets out a hypothetical example in her witness statement (paragraph 37):

    "In placing a cost on unlimited cover, the important factor is to estimate what percentage of claim has adverse costs in excess of standard market limits and by how much they exceed these limits. For convenience I will refer to these cases as "large losses". … If the probability of large losses is low then the cost of providing this cover will be low. If say 5% of claims go to trial, 30% lose and 1% of these have costs at risk of ?500,000, the additional cost per policy of covering these large losses is ?75.

    38. … Using the simple example in the previous paragraph and assuming that the initial estimates of costs at risk and own costs were accurate, Pursuit scheme methodology would calculate the premium for the large costs at risk cases on the basis of a risk amount of ?500,000. Under the Pursuit scheme, due to the method of calculation of the premium it is already the case that the larger the risk covered, the larger the premium. Therefore, there is no mathematical requirement to add any extra factor to the premium in all cases to cater for large losses."

  200. In cross examination Ms O'Reilly confirmed that, when an insurer is looking at a First Assist case, the methodology is exposure related, rather than experience related, so that the risks are evaluated on the facts of the individual case. She agreed that under the First Assist methodology an underwriter needs an estimate of the potential adverse costs (i.e., adverse solicitors' costs and adverse disbursements) and an estimate of own costs, i.e. own disbursements (because they are at risk) and own solicitors' costs (because that gives a measure of costs). She stated:

    "If he has accurate estimates to trial he will be able to accurately estimate the premium at trial, but it may be overstated before trial.

    ...

    If the assumption of a constant ratio between adverse costs and own costs is correct, then yes ... [if the estimates are reasonable he has a reasonable basis for calculating the premium.]"

  201. She was then asked if she accepted that it was fair that an insurer, within the risk being insured on a book, had got to look to balance his premiums as against the adverse risk across his total book. She agreed. She also agreed that it would be wrong to take a small slice of the book and try to draw conclusions across the whole. She said:

    "That's why it would be good to get cases where there was in fact a swing the other way. The Test cases have … a swing the other way, but we can't tell from the [exhibit PWS3] which ones are.

    ...

    We have seen absolutely no evidence that there are any swings the other way for First Assist."

  202. She explained further:

    "Well, for example, in the First Assist methodology, if the assumption is that own costs are ?25,000 and adverse costs are ?50,000, if the adverse costs go to ?100,000 you might think that is a problem, but within the First Assist methodology if the own costs have also doubled to ?50,000 the methodology will make allowance for that fact.

    ... It is the ratio between the adverse costs and the own costs that is important rather than the absolute amount of the estimate."

  203. In relation to the schedule produced by Mr Ivory of 63 clinical negligence cases with costs exceeding ?50,000 Ms O'Reilly agreed that they were all successful cases where payment is sought under an ATE premium but that there was no information as to what point in each individual claim the ATE insurance had been taken out, and that it could fairly be assumed that the ATE insurance had not been taken out at the same time in each case. She further accepted that there was no information about the existence of any BTE or other insurance, nor is there any evidence of the percentage risk of failure. With the exception of Baker v Addenbrookes no details of procedural history or the underlying facts were available. It was however possible to draw inferences from the sample given the qualifications which she had agreed.

  204. In respect of Baker v Addenbrookes Ms O'Reilly was of the view that Mr Scrivenger's estimate of adverse costs "looked fine" although she thought that his own costs were perhaps not estimated very well. As to the proportion of one set of costs to the other she stated:

    "It can be stepped. It'll be … basically mirroring and always increasing. It can only ever go up.

    ...

    Constantly increasing, and mirroring each other in the way they step forward."

  205. Mr Dutton put to Ms O'Reilly that Mrs Baker had the possibility of taking out a flat rate premium with Litigation Protection which would have required her to borrow money in order to pay the premium at the outset. Miss O'Reilly stated:

    "I think the Baker case is unusual but, maybe it's not unusual I think you're basically saying that Mrs Baker had a choice between a premium with a minimum of ?8,000 to ?9,000 with an unlimited downside premium. Or a flat ?12,000 or ?10,000 or ?14,000. I personally would consider a minimum of ?9,000 with no limit on, how big it can get I personally would go with the flat ?12,000. But that's just me."

  206. She was asked about cases where the estimate of adverse costs is significantly different from the actual out turn, either higher or lower. She responded:

    "If you're rating a flat rate premium, ATE policy, it's looking at how the adverse costs have gone relative to the estimated versus actual that's important. That's the averaging that's important to you. But in the First Assist methodology the averaging needs to be between policies which were rated too high relative to the risk and too low relative to the risk and there is nothing here to say that those policies were not rated correctly relative to the risk. So it is quite possible that all of the First Assist policies are rated too high relative to the risk. Even though some have very high pay outs, some have very low premiums, that isn't proof that they are not excessive relative to the risk presented."

  207. Asked whether the methodology produced the correct premium rate and a fair result, she stated:

    "We have no way of knowing whether that is an unfair or fair premium to either side."

  208. With regard to inaccurate estimates in other types of insurance she stated:

    "Insurance companies are usually quite clever at getting around that. I mean, your bulk carrier example, for example. It could sit in port for the whole year, it could go out on a trip every day of the week and have a very high exposure. What will usually happen is that there will be a statement at the end of the year from the shipping company saying, "This vessel was out on these days. This vessel was in on these days," and an adjustment at the year end to take account of that."

  209. Mr Dutton put to Ms O'Reilly that early settlement would result in a lower quantum of premium and that there is a sliding scale right through to trial. She replied:

    "But again it goes back to the issue of a lower premium does not necessarily mean that it's not excessive. Because of the way the RSA methodology works, it's not an average between high and low, it's an average between excessive for the paying parties and excessive for RSA and I haven't seen any policies which are excessive in terms of fairness to the Pursuit policy, whereas we are seeing the cases where they are excessive relative to the paying parties.

    ...

    I think the concept of the Pursuit policy is great … it gives the claimant access to justice, it in theory offers low premiums if the case settles early and they don't have to pay, they don't have to fund, they don't have to get into bank loans. But that doesn't mean that it shouldn't be priced fairly for the paying parties."

  210. With regard to the cost of providing unlimited cover in the Pursuit policy Ms O'Reilly agreed that it was a great benefit and she accepted that the cost of making that extra provision was not very large. She identified the unique features of the product as the fact that the premium is conditional, it is deferred and the cover is unlimited. Her assumption had been that that must have been what was driving the premiums to be higher, but she said that if that was not the case she accepted the position. With regard to the Temple policy she stated:

    "I think with Mr Wait's policy or any other policy, the reason it's so expensive is because people have got to the ?100,000 limit. They've actually spent the money, that amount is at risk, definitely at risk so they need to top it up. Whereas, when people are taking it out at the start, it may get to ?100,000 or it may settle at ?1,000. So it's more a question of, "this is a definite definite risk as opposed to it's not very likely."

    John Paul Ivory

  211. Mr Ivory is a Fellow of the Association of Law Costs Draftsmen employed by Kain Knight Group Plc. He has dealt with matters on behalf of various NHS Trusts and the NHSLA since about August 1998. Kain Knight is one of three costs draftsmen companies on the NHSLA Costs Draftsmen's Panel. Those companies are responsible for negotiating claimants' solicitors' costs on all NHS clinical negligence claims where the individual claimant's costs exceed ?50,000.

  212. Between August 2000 and December 2004 Kain Knight had been involved in 547 claims on behalf of NHSLA. Mr Ivory produced a schedule giving details of all NHS clinical negligence claims where his company had negotiated and settled the claimant's costs and an ATE insurance premium. He had identified 63 such claims. He suggests that in his general experience claimants' solicitors' costs in clinical negligence cases are "at least double the defendants' solicitors' costs". Of the 63 cases identified, 11 involved First Assist Pursuit premiums, which were claimed in the range of ?9,349.66 to ?72,696.57. In the remaining 52 claims, which were not First Assist Pursuit ATE policies, the premiums claimed were between ?840 to ?36,750. All figures include IPT. In respect of 13 claims, where the limit of indemnity on the policy was ?100,000, the premiums were in the range of ?7,875 to ?36,750 including IPT. Mr Ivory suggests that the average premium for the 11 First Assist cases is ?40,065.73, and for the remaining 52 cases ?10,065.36. In respect of the 13 claims where the limit of indemnity was ?100,000 he calculates the average premium to be ?18,127.69. In his second witness statement Mr Ivory exhibited a further schedule, which was a revision of the earlier schedule which included two additional cases where the Claimant's costs included an ATE premium. Both were First Assist cases and he states that the range of premiums was from ?9,349.66 to ?121,357.47, in respect of which he calculates the average to be ?46,419.27.

  213. Mr Ivory had been involved in Deborah Baker case v Addenbrookes in 2002 when Mr Scrivenger sent in the letter of claim. He had advised NHSLA at that time as to the sort of insurance policy that was being taken on and the sort of premium that was going to result from it. He agreed that he knew the longer the case went on the greater the costs incurred, and that if it turned into a loss for the NHSLA the premium might be higher. He and his company advised the NHSLA on the policy and the premiums that were available at the time.

  214. Mr Ivory was asked in cross examination by Mr Dutton if he could have collated information about cases in which the costs were below ?50,000. He stated:

    "The protocol follows cases over ?50,000 and we have that information to hand … and we produce monthly reports on that information, from which I've been able to produce the schedule. For the cases under ?50,000 that would involve a trawl through many, many files … Other materials, and it would have taken probably weeks, if not months."

  215. Cases over ?50,000 must be referred to one of the panel firms of costs draftsmen. Cases under ?50,000 do not form part of the protocol, so that the individual fee earner at the panel firm of solicitors "can outsource that to whoever". In respect of Baker v Addenbrookes Mr Ivory had prepared the Points of Dispute. He confirmed that Mr Scrivenger's claim for ?30,641 base costs had been settled for a figure of ?27,500 with an uplift which he put at 85%. Mr Ivory was of the view that his clients did not have adequate costs protection by way of their Calderbank offer and therefore he and the claims handler decided that it was appropriate to offer an economic sum:

    "So the way we arrived at the final figure was to allow an element of care and conduct or success fee of 85% … at the end of the day, we wanted an economic settlement."

  216. He described the settlement as one that Mr Scrivenger was happy with, and the client was happy with the overall sum. Mr Ivory confirmed that he had seen the letter from Scrivenger Seabrooke of 27 January 2004 setting out the methodology of the premium calculation. He discussed the calculation and the way it was formulated with his instructing solicitor (Messrs Kennedys) and they decided to take the advice of counsel. Mr Ivory had not personally come across this method of premium calculation before, nor had he come across the Pursuit policy, although during 2004 he did.

  217. Mr Ivory was taken to Mr Smith's second witness statement and the exhibit PWS3 summarising the NHSLA cases which had been won when using the RSA Pursuit policy. Of the 46 cases listed the average premium was ?21,606, although 16 cases had individual premiums below ?10,000. Mr Ivory confirmed that for a case with unlimited cover he had never seen a premium as low as ?10,000 in a clinical negligence case. Even for cover of ?100,000 he said:

    "They tend to be higher. What I would say on that point is that having seen cases where the claims were less than ?50,000, I have no doubt that if the exercise was performed to include those in my schedule that there are not any that I have seen from First Assist certainly and those in Mr Smith's schedule are not cases that I am aware of."

  218. He agreed that in another case of Scrivenger Seabrook, referred to in the exhibit of Mr Smith, that a premium of ?1,067 in respect of costs at risk of ?40,000 was low.

  219. In relation to the premium of ?1,067 Mr Ivory stated in re-examination that the costs at risk, since the case was in the pre action stage, would not be anywhere near ?5,000 at that stage. He thought it was not usual to have unlimited cover in smaller cases.

  220. Mr Dutton took Mr Ivory through a number of entries on Mr Smith's schedule in order to demonstrate that NHSLA had on previous occasions paid First Assist premiums, and, where the claimant had lost, had received payment from First Assist. It was put to Mr Ivory that his average of ?18,127.69 for the 52 non First Assist cases was not very far from Mr Smith's average of ?21,500. He responded:

    "Superficially but again, if I was running arguments on the individual cases I would say that unlimited cover was something that would not be required probably on the lower value ones that Mr Smith has been able to produce."

  221. In re-examination Mr Ivory pointed out that the figures that Mr Smith had produced covered the whole of the range, and that therefore his own figure for the 52 cases, i.e. an average of ?10,065.36 was to be preferred.

    Adam Richard Burrell

  222. Mr Burrell is a Technical Claims Co-ordinator employed by Liberata GIS. He was instructed to deal with three of the Test Cases on behalf of Norwich Union, AXA and Allianz Cornhill. He had carried out a search on The Judge website in order to ascertain what policies were available for each claimant that were appropriate for the type of case, according to a given set of criteria. He exhibited the results of his search in respect of each of the Test Cases except Farr.

  223. Mr Burrell stated that his search was based on the assumptions of the criteria set out in paragraph 4 of his witness statement "the assumption was that the criteria were correct on each of the cases." Although he was familiar with the facts of the cases he said:

    "I was instructed to enter onto The Judge website the criteria given for each case. I wasn't asked to comment on whether or not the fees on the various printouts meant that they were appropriate or not appropriate in each case."

  224. Mr Burrell confirmed, in response to questions from me, that he had a set of criteria which he entered into the search engine and exhibited the results but that he had not applied his mind to those results. He said he was only in a position to confirm what the results of the search on The Judge database and his search of Litigation Funding were. He was not in a position to offer any comment on the availability or suitability of any of the policies identified as a result of his searches.

    COMMENTS ON THE EVIDENCE

    First Assist Witnesses

  225. Mr Smith is an experienced underwriter who gave clear and cogent evidence as to the work and discussions which had gone into the setting up of the Pursuit policy, which I accept. It is, in my judgment, particularly relevant that the Pursuit policy was created before the recovery of the ATE premium became a possibility. None of the Defendants called evidence from any other underwriter who could suggest any other way of creating a sliding scale premium for complex cases. Ms O'Reilly was unaware of any other insurer who had attempted to devise such an alternative model.

  226. Mr Smith's explanation at paragraph 33 of his first witness statement, in which he refers to the burning cost, describes a costs rating which is not equivalent to actual burning cost. Burning cost, as I understand it, is the aggregate of the actual cost of a group of policies divided by the number in the group. The cost rating described by Mr Smith is neither the actual cost, nor the average cost, rather it is the actual potential cost.

  227. Mr Gilbert is an experienced and shrewd insurance broker who again gave clear evidence and answered the questions put to him directly and without hesitation. I accept his evidence. Both Mr Dutton and Mr Morgan take the point that Mr Gilbert was not cross examined about the availability of a Temple policy to Mrs Deborah Baker. They submit that he would have been able to give answers as to whether or not that would have been a suitable policy. Temple is a product referred to on his website and he would no doubt have had knowledge of it, and he would have been able to deal with any assertion put to him that the Temple policy should have been recommended. Mr Bartlett's position is that Mr Gilbert broked to a strictly limited number of insurers and did not act as a broker for Temple policies. In any event Mr Gilbert accepted in his second witness statement that it was possible that Temple would have provided a deferred premium quotation.

    The Claimants' Witnesses

  228. All five solicitors gave clear evidence and were clearly competent in their particular field. Mr Scrivenger and Ms Stevens stood out as being particularly impressive and experienced. It is evident that none of the solicitors understood how the Pursuit Premium methodology worked. Mr Bennett had unfortunately put incorrect figures in the proposal form for Mr Farr but he freely admitted that the proposal was not "a well done document". Mr Morgan raises, in respect of Mr Scrivenger, that the availability or suitability of the Temple policy was not put to him. In respect of this Mr Bartlett points out that Mr Scrivenger said he put the matter in the hands of Mr Gilbert and relied on his expertise. He did not claim to have surveyed the market himself and did not have any personal experience of policies other than the Pursuit policy. Given my decision under Issue 6 the lack of cross-examination of Mr Gilbert and Mr Scrivenger is not material.

    The Defendants' Witnesses

  229. Mr Wait is not an underwriter but works under a cover holder authority using the rating mechanism required by underwriters at Lloyds. He is however well experienced in the insurance industry and although he gave evidence as to the possible availability of alternative policies in certain cases, he did not possess the experience and expertise of Mr Gilbert. Therefore, to the extent that there is any conflict between the evidence of Mr Wait and Mr Gilbert I prefer the evidence of Mr Gilbert.

  230. Mr Wait suggested, in relation to the Baker v Addenbrookes case, that it did not look like a difficult case, but even on that basis he would have wanted a premium of ?50,000 had the case gone to trial.

  231. I do not accept his view that Mrs Baker's clinical negligence case was not a difficult case. For the reasons given by Mr Scrivenger it is quite clear that there were considerable difficulties, particularly surrounding the issue of causation.

  232. Kate O'Reilly although an actuary did not give expert evidence. The Defendants did not apparently instruct her until late in the day and sought permission to rely on her evidence, although, curiously, they suggested that the evidence was not expert evidence. First Assist objected to the late introduction of expert evidence and accordingly the evidence which she was permitted to give was factual only. I found it a helpful explanation of the way in which the premium methodology operated.

  233. Mr Ivory is a costs draftsman with particular knowledge of NHS cases. He produced a number of records from his employers, who are one of three companies on the NHSLA Costs Draftsman Panel. Although the figures which he produced covered all the NHS clinical negligence claims where the individual claimant's costs exceeded ?50,000, this could only give a snapshot of the true position, since his firm only deals with a selection of such cases (probably one third), nor did the figures reflect the position in cases under ?50,000. Thus, although Mr Ivory presented as an honest and competent witness, the evidence which he gave could amount to no more than a pointer towards the fact that the Pursuit Policy premiums might be unreasonably high.

  234. Mr Burrell carried out a mechanical exercise in relation to The Judge website and Litigation Funding, he merely collated the results and did not apply his mind to those results. He was not in a position to express any view as to whether any of the policies listed would actually have been available for any particular case. I can derive no assistance from his evidence.

  235. As to the information contained in Litigation Funding and The Judge website, this is no more than an indication of policies which might be available in certain circumstances. As Mr Gilbert points out, the premiums on his website are "indicative only" and the website contains further warnings. Litigation Funding has similar warnings and reservations. I can derive no firm data from these sources.

    GENERAL SUBMISSIONS

  236. I have had the advantage of reading in excess of 350 pages of submissions, together with numerous additional annexes and papers. I have also heard lengthy oral submissions. In this section of the judgment I summarise the main points raised in those submissions.

    First Assist

  237. Mr Dutton submits that the Pursuit Policy is a move away from the normal insurance precept of the many paying for the few, to the few paying for the fewer. The policy is an individually written contract of insurance which was developed after extensive consultation with the Lord Chancellor's Department. It was designed before Section 29 of the Access to Justice Act was thought of, and therefore before recoverability of the premium had become an issue.

  238. He suggests that at the time of the design of the policy CFAs had not come into use by solicitors in the fields of commercial disputes, professional negligence, clinical negligence, insolvency and complex personal injury. The policy and the premium rating mechanism were designed with these sorts of cases in mind.

  239. With regard to the issues relating to enforceability of the contract of insurance (uncertainty, champerty, breach of the indemnity principle, issues i, ii and viii) he submits that the following facts are not in dispute:
    1. Each Claimant signed a contract of insurance with First Assist. At the point that the contract was entered into it is a binding contract with all the essential features such as consideration.

    2. Each Claimant did so after explanation of the policy and the rights and liabilities which would arise under it from their solicitor.

    3. Each had an insurable interest - namely in the risks and costs consequences of losing their cases.

    4. The premium was calculable during the life of the policy under the formula provided to the insured claimant prior to the inception of the policy.

    5. The terms of the policy clearly set out the rights and duties of each party.

    6. First Assist had a right to agree to any settlement which the Claimant proposed to enter into.

    7. First Assist has a right to recover the premium from the Claimant if the claim succeeds.

    8. The premium is due but not "payable" if the claim is lost. There is no suggestion that the contract fails for want of consideration.

    9. The premium is both due and payable from the client under the terms of the policy in each of the Test Cases.

    10. First Assist has not waived its right to premium in any of the Test Cases.
  240. In those circumstances Mr Dutton argues that there was and remains an enforceable contract of insurance as between each Claimant and First Assist in each of the Test Cases.

  241. He goes on to suggest that all the witnesses accepted that an insurer must seek to set a premium which ensures that he does not make a loss across the various categories of cases on his book, and that, to examine the premium rating mechanism or quantum of individual premiums in only one or two categories of case, as in these Test Cases, and to then seek to draw general conclusions from them adverse to the insurer is unsafe.

  242. He submits that the terms of the Pursuit Policy are straightforward and clear. The policy defines the limit of indemnity, and on an objective test a person reading the policy would have no difficulty understanding the mechanism of premium calculation.

  243. He says that the premium is plainly due from the inception of the policy and it becomes payable in the event of success, or, if the case becomes successful during its life, the insured may elect to pay the premium based upon normal costs at that date. He points out that so far as "success" is concerned it was initially intended that "success" should be defined by reference to a single net monetary amount set out in the schedule, but First Assist subsequently adopted the Law Society definition of success in the schedules to the five polices in the Test Cases.

  244. Condition 1 of the policy provides that the premium will not be affected by assessment or agreement which reduces the insured's solicitor's normal fees or the level of the success fee. Mr Dutton says that this is required because the premium has to be calculated by reference to the proportion of own side's and adverse costs. Own side's normal costs become the multiplicand (and "proxy") for the calculation of the premium. He points out that, if a client and his solicitor were able to agree a reduction in the normal fees, the insurer would have insufficient premium for the risk taken on at the start of the policy. He accepts that this could mean that the premium could ultimately be calculated by reference to costs which the court subsequently determined were unreasonable, but he suggests that this is both simplistic and unrealistic on the facts of the Test Cases.

  245. In relation to proportionality Mr Dutton argues that the question of whether or not the premium is proportionate is to be answered by reference to the risk being written and the costs of the parties in the litigation. The insurer is not insuring against the prospect of a sum of damages that may be recovered, he is insuring against the downside costs risks which the individual litigants may incur. He suggests that the court needs to look primarily at the costs risks being insured against. He submits that it is necessary to answer the question by looking at what the insurer needed and what the individual claimant needed in order to have sufficient costs cover. The insured obtains unlimited costs cover against a future risk, which could run into tens of thousands of pounds. The fact that one of the Test Cases settled for ?1,250, and that the premium was several times higher than that is not the point. In Mr Dutton's submission the point is what is the costs risk that the insurer is being asked to provide for? The answer is he is providing for an unlimited risk.

  246. Mr Dutton tentatively suggests that I should give guidance for Costs Judges in future RSA/First Assist cases, but then goes on to say that it is difficult and unwise for the court to hand down guidance about the detail of future assessments where the policy is bespoke and individually underwritten. In my judgment the appropriate way forward is for me to give my conclusions on the issues based on the evidence which I have heard. To the extent that those issues are resolved, principles will (subject to any appeal) emerge. I do not propose to give further general guidance.

    Test Case Claimants

  247. Mr Morgan's submissions deal mainly with the issues of reasonableness in relation to Claimants' conduct in taking out the Pursuit policies.

  248. He submits that reasonableness and proportionality should be judged by focussing on the reasonable interests of the individual claimant judged by the standard of the sensible claimant without the application of hindsight. He maintains that if the claimant acted on advice, then unless that advice was negligent the claimant should be taken to have acted reasonably. If the claimant reasonably chose a policy (in the absence of realistic alternatives) with a premium which is high or the calculation of which can be criticised, he suggests that the court should nevertheless allow the premium charged by the insurer, since the claimant's choice, when presented with such a premium, is either to take it or to leave it. He goes on to suggest that any reduction of the premium would be "unprincipled and without precedent in the law of costs". This is on the basis that in normal circumstances where a reduction is made in the amount of costs claimed it is because "a cheaper alternative rate was available". Furthermore since the amount of the premium is fixed by contract the insured has no right to apply to the court for a reduction of the premium on the ground that it is unreasonably high. He submits that there is an important policy reason underlying the submission in that, if it fails the claimant who has taken out a policy, relying on such advice, could be held to be unable to recover the premium and unable to have recourse to anyone else for payment. If the claimant is wholly free from blame, that would be an unacceptable result and the potential for such a result would itself impose a substantial barrier to access to justice, as litigants with any case out of the ordinary, would be deterred from making a claim by a well founded fear of incurring an irrecoverable premium.

  249. Mr Morgan goes on to argue that, if the court is of the view that there were elements of the Pursuit Policy which were unreasonable, it would be wrong to disallow any premium representing those elements, retrospectively since the policy was both promulgated by First Assist and taken out by the Claimants acting in good faith. He suggests that the correct procedure in such circumstances would be to direct that any element found to be unreasonable should not be recoverable in policies taken out after a reasonable period from the delivery of the judgment, thereby giving First Assist an opportunity to modify their product.

  250. Mr Morgan submits that the ATE premium is not strictly a disbursement, in that it is not paid by the solicitor on the client's behalf (see Re: Buckwell & Berkeley [1902] 2 Ch 596 CA confirming Re: Remnant [1849] 11 Beav 603 at 611). Given that situation there is no possibility of the amount of the premium being reduced on an assessment under the Solicitors Act 1974, even if the amount of premium was reduced on an assessment between the parties.

    Defendants 1-4

  251. Both Mr Bartlett and Mr Bacon submit that the premiums charged under the First Assist Pursuit policies are excessive and the terms and charging structure of the policies are inappropriate. Mr Bartlett submits that there are three basic considerations which I need to take into account in deciding these issues. They are:

    "(i) The permissive nature of the jurisdiction granted by Section 29 [of the Access to Justice Act 1999] recognises the interests of the Defendants.

    (ii) The vital role of the Costs Judge.

    (iii) The nature of the ultimate question."

    (i) Section 29 Access to Justice Act 1999

  252. Mr Bartlett submits that the Section does not automatically justify any insurance arrangements which make it easier for Claimants to pursue their claims. In particular "justice" involves justice for Defendants as well as for Claimants. Secondly, the Act states that the costs payable "may" include the cost of an ATE premium. In other words it is permissive but does not require reimbursement of the premium whether in whole or in part: see Lord Scott of Foscote Callery v Gray [2002] 1 WLR 2000 paragraph 115:

    "The section does not mandate the inclusion of costs in respect of the premium. And whether a particular ATE insurance premium, or part of it, should be included in the costs ordered to be paid in a particular case should be tested and answered by reference to the same principles of reasonableness that apply under the Rules and Practice Directions to all other items of expenditure."

  253. Similarly the new regime is intended to:

    "Operate so as to promote access to justice but not so as to confer disproportionate benefits on … after the event insurers or impose unfair burdens on defendants or other insurers."

    (Callery v Gray paragraph 10, Lord Bingham of Cornhill).

    (ii) The Costs Judge

  254. The House of Lords was concerned that the new regime should be controlled very carefully. Mr Bartlett submits that the only effective control on the size of ATE premiums is that exercised by the Costs Judge:

    "There is only one restraining force on the premium charged and that is how much the Costs Judge will allow on an assessment against the liability insurer."

    Paragraph 43 per Lord Hoffman, see also paragraph 6 (Lord Bingham), paragraph 32 and 42 (Lord Hoffman), paragraph 60 (Lord Hope).

    (iii) The Ultimate Question

  255. The ultimate question is how much the losing party should be required to contribute to the premium which the ATE insurers have chosen to set? Mr Bartlett relies on this passage from my judgment in an earlier case:

    "It is … necessary to consider whether the claim for … premium … is reasonable and proportionate. It is in my judgment open to any insurer to calculate and charge premiums on whatever basis best suits its business. The question which I have to decide is what amount ought reasonably to be payable by the paying party in accordance with Section 29 of the Access to Justice Act 1999."

    Pirie v Ayling 18 February 2003, paragraph 12.

  256. The main thrust of Mr Bartlett's submission was about what he calls inherent flaws in the Pursuit methodology. This is examined under Issue 3.

    Defendant 5 (Farr)

  257. To the extent that Mr Bacon's extensive submissions support or duplicate what is said by Mr Bartlett I do not repeat them.

  258. The documents demonstrate that the total premium income in respect of clinical negligence cases was a little under ?1 million and the losses paid out of the pool were about ?260,000, therefore, Mr Bacon argues, premium income is greater than is needed to pay for losses, commission, profit and administration and this generates disproportionate benefits for the ATE insurer and places an unfair burden on the liability insurer.

  259. He refers to the set off provisions of the Pursuit Policy and compares them to those of the Temple Policy provided by Mr Wait. He says first that the adverse costs provision in the Pursuit Policy, has the effect of reducing the risk of the insurer of having to pay out under the policy, as it provides a cushion, in the guise of costs awarded in favour of the claimant, before liability for adverse costs is triggered. The difference between the two policies, says Mr Bacon, is that the insured under the Temple Policy does not have to pay a premium of ?115,182 in order to achieve what the Temple Policy provides for approximately ?20,000. Where the adverse costs swallow up both the award of damages and the costs awarded in favour of the claimant, he suggests that under the Temple Policy the net cover provided by the policy would be the amount of the shortfall, whereas under the Pursuit Policy no cover would be provided at all, because the case had resulted in a success for the claimant.

    GENERAL CONCLUSIONS

  260. In my judgment the condition that the premium will not be affected by assessment or agreement which reduces the insured's solicitor's normal fees, or the level of the success fee, can operate to produce a premium which is unreasonable and disproportionate in all the circumstances. Whilst Mr Dutton argues that this conclusion is unrealistic on the facts of the five Test Cases, it must be the case that a premium calculated on the actual costs, when known, more accurately reflects the true position. A distinction must be drawn between the premium rate, which is arrived at on the basis of the estimates put forward by the Claimant's solicitors, and the premium itself, which is calculated by multiplying the Claimant's solicitors actual costs at the date of settlement by the premium rate. Provided that the premium rate has been calculated by a method which is not flawed (see Issue 3 below) I can see no reason why the calculation of the premium itself should not be based on the Claimant's actual, reasonable and proportionate costs as assessed or agreed with the opposing party. In this way the use of hindsight is avoided in the calculation of the premium rate, but reality prevails when the premium itself is calculated.

  261. In relation to Mr Dutton's arguments about proportionality, Mr Bartlett accepts that, whilst it is very important to look at the costs risk compared with the premium, the size of the premium in relation to the size of the claim is an important matter of proportionality, because it is an element of the costs and it is necessary to keep some sense of proportion about what costs are appropriate to a given size of claim. I accept his submission that the court should look both at the costs risks and at the size of the claim when considering the premium.

  262. As to Mr Dutton's suggestion that it would be unsafe to draw general conclusions from an examination of the premium rating mechanism and the individual premiums, the purpose of the Test Cases is to try and achieve some element of certainty for those who elect to use the Pursuit policy. It may well be that where the policy is used in other types of cases different considerations may apply. Nevertheless it is necessary to reach general conclusions in these Test Cases.

  263. With regard to Mr Morgan's argument that if the claimant, acting on advice, has reasonably chosen an expensive policy then no reduction is permissible, Mr Bartlett argues that this is wrong in law relying on the decision of the Court of Appeal in Sarwar v Alam [2000] EWCA Civ 1401:

    "51. We deprecate any attempt to equate the question of reasonableness that a Costs Judge has to decide with the question of whether the Claimant's Solicitor was in breach of his duty of care to his/her client. If a solicitor gives advice which proves unsound, it will not necessarily follow that the advice was negligent. The advice will necessarily be based on information provided by the client. If the information is inadequate or inaccurate the advice may prove to be unsound without any question of fault on the part of the solicitor."

  264. In the case of ATE insurance, of course, the information upon which the advice is based does not come from the client. I do accept however that the question of reasonableness which I have to decide, does not equate with the question of whether the Claimant's solicitor was in breach of the duty of care. It follows that, simply because a client has acted on advice, is no answer to the question of reasonableness which I have to decide.

  265. In these Test Cases the Court is required to assess the costs on the standard basis applying tests of reasonableness and proportionality in accordance with CPR 44.4. If the costs as a whole, or a single item of costs, are or appear to be disproportionate, the test of necessity must be used (see Lownds v Home Office [2002] EWCA Civ 365; [2002] 1 WLR 2450 and Giambrone v JMC Holidays Ltd [2002] EWHC 2932 (QB) Morland J.). Mr Morgan suggests that following Lownds all costs which are necessary should be allowed, and that if, as the Claimants and First Assist contend, it was necessary for the Claimants to take out a First Assist policy, then the premium should be allowed. In my judgment Mr Morgan's submission must fail. The purpose of assessment is not merely to ensure that the receiving party recovers reasonable and proportionate costs but also to ensure that the paying party is not required to pay more than an amount which is reasonable and proportionate. If Mr Morgan's argument were to be upheld it could result in the defendants in these cases having to pay the full amount of the premiums, notwithstanding a decision (for example because the calculation methodology was flawed) that the premium was too high. If, however, the paying party can demonstrate that a cheaper policy or policies were available, the burden is then upon the Claimant to justify why the Claimant, or legal representatives, selected a more expensive one.

  266. Mr Morgan suggested that if any part of the premium was found to be unreasonable and therefore irrecoverable the correct approach, avoiding hindsight, would be to make a finding but to state that that element of the premium should not be recoverable in policies taken out after a certain date in the future. That approach is correct in respect of any severable element of the premium. Where, however, there is a finding that errors have been made in calculating a premium, thereby increasing it, the court should allow only the reduced amount which would have been payable if the errors had not been made.

  267. Mr Morgan's argument that the premium is not a disbursement does not, in my view, prevent the court from making a reduction in the amount of the premium payable by the Defendant. He argues that if the premium, or part of it, is disallowed then the innocent client will suffer absent a clear claim of negligence against a solicitor (there is no such allegation). Mr Dutton does not support Mr Morgan's submission on this point. The extent to which the premium is payable by a Claimant to First Assist is governed by the contract between them, in which I do not interfere. Mr Dutton informs me that the contracting parties have agreed that each Claimant is bound to pay the sums in respect of premium set out in the schedule which has been prepared at my request. A Claimant should not expect a Defendant to indemnify him or her fully in respect of any bad bargain agreed direct or through the legal representatives with the insurer.

  268. Mr Morgan relied on the Court of Appeal decision in Callery v Gray (No.2) and in particular paragraph 13 of that judgment, which I have quoted earlier, to argue that in assessing recoverable premium there are only two possible approaches. First to examine the insurer's costs and to award a premium which reflects those costs, plus a level of profit which the court considers reasonable; and the second is to look at the market. He suggests that in Callery the Court of Appeal rejected the first option and opted for the latter. In the particular circumstances of the Claims Direct Test Cases and The Accident Group Test Cases the amount described as premium was deconstructed and a figure arrived at which reflected the court's view of the premium recoverable under Section 29. In the case of the Pursuit Policies, deconstruction has not proved necessary, since Mr Smith has given a very full explanation of how the premium calculation works. The fact remains that, as was common ground in Callery (No.2), "in so far as the court finds that the premium is not reasonable it can and should reduce it" (paragraph 11). With regard to evidence of the market, it is clear that even now the ATE market is still developing. It is not yet possible, at least in respect of cases such as these five Test Cases, to look at a selection of available policies and identify the one with the cheapest premium commensurate with the required cover.

  269. With regard to Mr Bartlett's reliance on Sarwar Mr Morgan seeks to distinguish the Test Cases from Sarwar. It seems to me however that the passage relied on by Mr Bartlett can be read as being of general application when the court is approaching issues of this kind. I accept Mr Bartlett's submission that there is a difference between unreasonableness and negligence as between solicitor and client and the question of reasonableness as between claimant and defendant.

  270. Although Mr Bacon submitted what he saw as shortcomings in the Pursuit Policy, these arguments go, in the first instance, to the question of whether the premium claimed is reasonable and proportionate and, in the second instance, as to whether the policy was a reasonable choice given the circumstances of the particular case. Those are matters which are dealt with under the relevant issues. The level of First Assist's apparent profits is not relevant, since the purpose of all businesses is to earn profits.

    SUBMISSIONS ON THE ISSUES

    ISSUE 1: Is the contract of insurance void for uncertainty because at the time the contract is made the amount of the premium is insufficiently certain and is the said contract accordingly unenforceable by RSA against the claimant and if so what is the consequence?

    First Assist

  271. Mr Dutton submits that the method by which the premium is to be calculated is quite clear under the policy, namely a percentage of the normal fees as defined in the policy, i.e. the costs of the insured's solicitor acting under the CFA which formed the basis for the calculation of the success fee.

  272. Mr Dutton argues that, although the premium is not a fixed amount at the time the policy is taken out, the insured has agreed the relevant premium rate percentage in relation to his own costs. He relies on Clarke: The Law of Insurance Contracts at 13:2:

    "… [a] case in which the premium is not set at the time of contract but left to the hazard of future agreement between the parties, the contract may be void for uncertainty unless (a) the contract has an effective mechanism for deciding premium later …"

  273. He also relies on Ivamy: General Principles of Insurance Law at page 202:

    "It does not however appear to be necessary that the precise amount of the premium should have been fixed and agreed to, provided that there is a definite agreement to postpone the fixing of the amount and to enter into the contract notwithstanding. Thus, the insurance may be in consideration of the premium to be arranged."

  274. On that basis Mr Dutton argues that a contract of insurance can come into existence where the premium is fixed by reference to a mechanism expressed in the contract itself as a percentage of a calculable sum.

    Defendant 5 (Farr)

  275. The liability insurer (Ensign Policies) behind the Defendant in Farr v Kerslake pursues this point, the Defendants in the other Test Cases do not. Mr Bacon's submission is that the whole policy is void for uncertainty, but that in the alternative, if the policy is not void, the court should depart from "the uncertain terms" and allow a reasonable premium. Mr Bacon argues that the policy definition of "normal fees" and the formula by which the premium is to be calculated are such that it is impossible to know, at any given time, what the normal fees are, because the meaning of "normal fees" in the policy is ambiguous and inherently contradictory. He argues, relying on MacGillivray on Insurance Law, 10th Edition, paragraph 2-6, that where there is no fixed figure for premium it is necessary for the parties to the contract of insurance to be ad idem as to how the premium is to be calculated. He says it is the ambiguity and uncertainty as to the calculation of the normal fees that makes the formula sufficiently uncertain to render it unenforceable.

  276. He puts forward a somewhat technical argument in an attempt to demonstrate the difficulty in understanding the meaning of "normal fees", particularly the words in the definition "and which form the basis of the calculation of the success fee". He correctly states that this is a reference to the solicitor's base costs, but those base costs are, he says, subject to the implied term of reasonableness as applied by CPR 48.8. The client has a right to seek an assessment of the base costs under the terms of the CFA. He argues that base costs must mean base costs properly and reasonably chargeable to the insured, whether as a consequence of an implied term in the CFA or CPR 48.8. On that basis, he says, the meaning of the words "normal fees" become inherently uncertain and contradictory within the policy terms.

    ISSUE 1 - CONCLUSIONS

  277. Mr Dutton's submissions as to certainty are correct and I accept them.

  278. Mr Bacon's submission that the Pursuit Policy is void for uncertainty, or that its inherent uncertainty is such that the court is unable to impose "a reasonable premium" is without merit. The method by which the premium is to be calculated is set out in the policy and is, in my judgment, clear. The resulting premium may well be higher than expected but that is nothing to do with lack of certainty in the policy. The suggestion that the policy might be void because it refers to success by reference to an "amount" in the schedule also fails. The schedules in the Test Cases set out an adequate and clear definition of success, albeit not by reference to an amount. The reason for this change was explained by Mr Smith in his evidence. There is, in my view, nothing in this point. Mr Dutton makes the point that the contracting parties are Mr Farr and First Assist and there is no dispute between them as to the certainty of the policy or the definition of success.

  279. The passage in MacGillivray on Insurance Law, relied on by Mr Bacon, states:

    "All the parties have to do is to commit themselves to a certain arrangement for ascertaining the rate of premium."

    And a little further on:

    "In default of agreement between the parties the amount of the premium will be settled by the court or an arbitrator."

  280. It is clear that the author of the work does not envisage an insurance policy being held to be unenforceable should uncertainty arise, but rather that any differences should be resolved by the court (as in this case) or by an arbitrator.

  281. Mr Bacon expanded his basic argument as to the inherent uncertainty and contradictory nature of the definition of "normal fees" by referring me to a number of authorities on the law of contract. I do not need to refer to those authorities because in my judgment his basic point is not made out. Giving the definition of "normal fees" its ordinary English meaning there can be no doubt what the parties are agreeing to. Where a client has entered into a CFA, the client will have had the agreement fully explained to him as well as having been given the required information under Rule 15 of the Solicitors Practice Rules. The client will almost certainly have been told that he will not be required to pay anything in respect of the solicitor's costs. One of the underlying problems with CFAs with recoverable success fees is that the client has no real interest in what he is being asked to agree to. Although it is technically possible for a client to seek to have his solicitor's base costs assessed, the likelihood of him doing so is almost negligible. Moreover some CFAs may be contentious business agreements thereby excluding the right to assessment.

  282. I accept that, in most cases, there will be an implied term of reasonableness applying to the amount of the solicitor's base costs. The fact that the Pursuit Premium will not be affected by any assessment which reduces the solicitor's normal fees, or the level of the success fee, is not a matter which goes to the certainty, or lack of it, in the meaning of "normal fees". Any disagreement arising out of the calculation of the premium will be resolved by the court or by arbitration. Although the premium rate methodology may be open to criticism there is, in my judgment, no uncertainty about the meaning of the policy wording.

    ISSUE 2: Is the insurance arrangement between the client, the insurer and/or the solicitors unlawful on the grounds of champerty and if so what is the consequence?

    First Assist

  283. Mr Dutton advances the argument that the paying parties have been both paying premium claims and receiving payment under the First Assist scheme where cases have been lost by insured claimants. He argues that the paying parties have therefore waived any objection to the enforceability of the policy and/or are estopped from contending that the policy is unenforceable. He points out that the paying parties have taken the benefits of the policies without arguing in those cases, that the policy was contrary to public policy because it was champertous. In those circumstances, he submits, it is inequitable for them to say, in cases where they have lost, that they wish to argue that the premium is too high and that the policy is contrary to public policy. He says the paying parties have acted inconsistently, whereas First Assist has paid out in the belief that the policies are lawful. In relation to this submission I take the view that the purpose of these Test Cases is to air the arguments and resolve the issues which have been raised. To exclude argument on the basis that the paying parties are in some way estopped would not be helpful. I therefore take the pragmatic view that the issue should be aired. In any event, the paying parties are individuals and companies whose arguments are not estopped by the conduct of the insurers who support them.

  284. With regard to the issue of champerty Mr Dutton submits that the current statutory provisions relating to CFAs restrict significantly the application of the common law doctrine of maintenance and champerty. He suggests that the only statutory provision which actively prohibits a possible champertous arrangement is the Solicitors Practice Rule 8 which prohibits a contingency fee agreement in respect of litigation services. He argues that there is no statutory prohibition on insurance policies which indemnify parties for the cost they incur in litigation. He suggests that such policies are a fundamental part of the legal landscape and are not just lawful, but the premium is expressly recoverable under Section 29 of the Access to Justice Act 1999.

  285. With regard to common law champerty and the enforceability or recovery of expenses incurred between a party and an assisting non party, Mr Dutton submits that the first, and in his view decisive, point is that the premium is not recoverable out of the damages or any compensation. Second, First Assist takes what steps it realistically can to recover the premium from the insured. Third, the fact that a party carries insurance as protection from adverse costs consequences does not mean that the insurance is champertous. Fourthly, he quotes Lord Mustill in Giles v Thompson [1994] 1 AC 142 at 164:

    "the law on maintenance and champerty has not stood still."

  286. He suggests that the paying parties therefore have to argue for common law champerty in respect of the Pursuit policy under the well established definition of champerty to which I have already referred and against a legislative background which does not favour their argument.

  287. To the extent that the Defendants assert that the contract of insurance is champertous, because it is an agreement to share in the spoils of litigation, Mr Dutton says that the answer is that the contract of insurance does not contain any such agreement. The insured agrees to pay a premium calculated by a defined formula; in the event of success the insured is more likely to have the means to pay the premium, but this, he says, is very different from an agreement to share the spoils of the litigation. He suggests that the question which the court needs to ask is whether, in the light of all the facts, there is a real risk that the agreement would tend to undermine public justice (see Factortame above).

  288. With regard to the Defendants' further assertion that there is a tendency to undermine public justice, because, for example, the insurer has the power to veto settlements and might want to make the parties take the case to trial simply to obtain a higher premium than would be payable if the case settled earlier, Mr Dutton points out that an insurer, whose receipt of a deferred premium depends upon success, has an interest in seeing that the insured obtains the funds to pay the premium. It is, he says, the same interest as in the present cases and neither is champertous. He points out that the Defendants' veto argument assumes that there is a veto, which is an assumption which Mr Dutton does not accept, but, if it became necessary to argue about the existence of the veto provision, he would say that the power of veto must have an implied provision of reasonableness inserted into it, because it is an insurance contract containing obligations of reasonable co-operation on both sides.

  289. The Defendants further argue that there is a risk under the policy that success might be achieved, and an election to pay premium arise when an offer to settle in terms which constitute a success has arisen. The Defendants argue that the premium becomes payable at that point and that if the case goes to trial the insurer would not be on risk under the policy because the case will be deemed to have been a success. (This argument is also used in relation to the indemnity principle issue.) Mr Dutton submits that the situation proposed is technically possible but no more than that, nor did it occur in any of the five Test Cases. Mr Smith, when being cross examined, could not recall any cases where RSA had gone to trial but had then refused to pay out on an adverse costs order because the case had been deemed a success and there had been an earlier election not to pay the premium. Should the situation arise where an offer is made and the right to elect to pay the premium arises, there would no doubt be discussions between the insurer and the insured as to how to proceed, and, if the offer is not accepted, whether RSA would continue to support the case to trial. The question to be resolved is: whether the insurer agrees not to treat the making of the offer as a success. As Mr Dutton correctly states, how that question will be resolved depends on the facts of the individual case. Among the Test Cases there is only one (Clarke) where this might have happened, and it did not.

  290. Mr Morgan makes no submissions on this issue, which is between the Defendants and First Assist. The individual Claimants remain neutral.

    Defendants 1 - 4

  291. Mr Bartlett suggests that in a successful case the "proceeds of litigation" include the sum awarded in respect of the ATE premium. The longer the litigation goes on the greater are the normal costs incurred, the greater is the premium and the greater are the profits of First Assist. He suggests that the premium is akin to a contingency fee, without the statutory safeguards which control solicitors' conditional fees. He suggests that the relevant test for champerty is to be found in Factortame paragraph 36 quoted above.

  292. While the Defendants do not allege actual conduct by First Assist that undermined the ends of justice, Mr Bartlett suggests that the question for the court is whether the agreement "might tempt" the ATE insurer to undermine the ends of justice (Factortame paragraph 38).

  293. The fact that the premium payable increases the longer the case goes on, leads Mr Bartlett to argue that, the greater the share of the spoils that a provider of legal services will receive, the greater the temptation to stray from the path of rectitude (Factortame paragraph 85). This consideration, he says, applies equally to ATE insurers as to the providers of legal services.

  294. Mr Bartlett suggests that First Assist has the means to interfere in the conduct of the litigation given that they have an unqualified right of veto on any settlement. He suggests that there can be no legitimate reason for First Assist to exercise such a veto, since acceptance of an offer of settlement terminates any risk of First Assist having to pay out under the policy. He suggests that the reasonable conclusion is that the agreement might tempt First Assist for their own gain to undermine the ends of justice by withholding or delaying consent to a just settlement.

  295. He argues that the wording used in the definition of success will, in the event of an offer being made, either tempt First Assist to use their veto, so that costs will continue to increase and the premium will escalate, or tempt them to encourage the litigant to elect to continue with the action and defer payment of the premium so that it will escalate. He suggests that these circumstances establish both opportunity and motive for interfering.

    Defendant 5 (Farr)

  296. Mr Bacon adopts Mr Bartlett's arguments and suggests that, once an offer has been made which equals or exceeds the definition of success (in monetary terms), First Assist is no longer at any risk if the claimant elects to continue with the action and defer payment of the premium until the conclusion. He suggests that the only risk to First Assist of not being paid, is if the case does not result in a damages award or costs award for the insured from which the premium can be taken (success having already been achieved). As a result he says First Assist would be profiting from the litigation because the premium would be rising even though, from the date of the offer, First Assist would no longer be acting as the ATE insurer.

    ISSUE 2 - CONCLUSIONS

  297. In Factortame the court was faced with a potentially champertous agreement in that the accountants had agreed to act for an 8% share of the proceeds of the litigation. The court went on to ask the question whether, having regard to all the facts, the agreement might tempt the maintainer, for his personal gain "to inflame the damages, to suppress evidence, to subborn witnesses or otherwise to undermine the ends of justice". The court found that the agreement was not champertous.

  298. With regard to the suggestion by the Defendants that there is a tendency to undermine public justice, such a tendency is at best a far fetched hypothesis which is not made out in any of the five Test Cases.

  299. The Defendants argue, in relation to champerty, that there is a risk, under the policy, that success might be achieved and an election to pay premium arise when an offer to settle on terms which constitute success, has arisen. The argument is that the premium becomes payable at that point, and that if the case goes to trial the insurer would not be on risk under the policy because the case would be deemed to have been a success. This again is a hypothetical possibility which, if it did arise, would have to be decided on the facts of the particular case. In the Test Case of Clarke the situation did not in fact arise (nor did it in any of the test cases) and in my judgment the submission is not made out.

  300. In none of the situations put forward by the Defendants can it be said that First Assist or RSA are an "officious and wanton intermeddler". There is no agreement to share the spoils of litigation and there is no tendency to undermine public justice.

  301. Mr Bartlett argues that the "proceeds of litigation" include the sum awarded in respect of the ATE premium. It is clear from the authorities which have been quoted that the "proceeds of litigation" refer primarily to damages (see Factortame). Legal representatives are not permitted to act on a contingency fee basis (i.e., a percentage of the damages if successful) save to the extent permitted by the statutory provisions relating to conditional fee agreements. After the event insurance is a comparatively new phenomenon and the recovery of the ATE premium has only been possible since the implementation of Section 29 of the 1999 Act. It does not seem to me that the ATE premium can properly be regarded as "proceeds of litigation" although I do accept that such a premium might only become payable in the event that the litigation is successful.

  302. Mr Bartlett points out that, in his evidence, Mr Smith accepted that the escalation of the premium after the achievement of "success" was not in respect of the risk of the insured incurring a costs liability. He therefore argues that it is, to that extent, not a premium within Section 29 recoverable from the paying party. Mr Smith in fact said:

    "The risk we take is that the case fails, that he does not obtain the damages he expects and he is incapable of paying us the premium. Would I rather have a premium of ?20,000 cash when he wins or the possibility that he will be able to pay me the premium of a higher amount or not at all at some later date? I will take the cash every time. It is the bird in the hand policy."

  303. It was put to him that the escalation of premium after the date of the offer was not in respect of the risk of his incurring a costs liability, because no cover was being provided, it was because the company was taking a risk on the payment of the premium. Mr Smith thought that was correct: "but you have to read it in conjunction with the definition of success." Mr Bartlett put to him that he was in fact sitting back and reaping a further reward, even though there was no legitimate interest in the continuation of the proceedings. Mr Smith responded:

    "I have a legitimate interest, I want to be paid my premium. That is the contract which I have with the client that he will … pay me the premium. I have given him the choice. He can take the money which is on offer, which has hit the definition of success, and he can pay me the premium. Alternatively he can carry on and I do not wish to fetter him, and had we done so in the case of Farr, Mr Farr would have been deprived of ?150,000 which evidently he is entitled to."

  304. Mr Bartlett's suggestion that Mr Smith's written comment in the Farr case "costs estimates look odd - but rating benefits!" demonstrates that First Assist is not beyond temptation, goes beyond the limits of reality.

  305. Mr Bacon's argument, based on an offer which met the definition of success (in monetary terms), is hypothetical only. In none of the Test Cases is the definition of success a specified figure and in none of the cases has the situation which he describes arisen. Mr Bacon's submission does not advance the argument in relation to champerty in any way.

  306. I am satisfied that the insurance arrangement between the client, the insurer and/or the solicitors is not unlawful on the grounds of champerty.

    ISSUE 3 - Is the method of calculation of the premium inherently flawed and if so what is the consequence?

    First Assist

  307. The methodology used to set the premium when the policy is written depends in part on three estimates given by the Claimants' Solicitor:
    1. the percentage prospects of success;

    2. an estimate of own costs to trial;

    3. an estimate as to adverse costs and disbursements to trial.
  308. The premium rate formula is sensitive to these estimates and to their accuracy. Mr Dutton submits: if it was reasonable for the Claimant to take out the Pursuit Policy for the particular case, then on the face of it the Pursuit Policy was a reasonable one for the litigant concerned. The policy is particularly suited to cases where the Claimant:
    1. has a difficult or complex case;

    2. where there is no alternative form of funding;

    3. costs at risk may be substantial; or

    4. whether the claimant does not have the means to fund any alternative policy (assuming there is one).
  309. Mr Dutton submits there is no flaw in the premium rate calculation methodology and none which prevents the Claimants from recovering the premium under Section 29 of the 1999 Act. He points out that none of the Defendants have called underwriting evidence so as to set out an alternative method of individually underwriting the risk in any of the Test Cases. Ms O'Reilly said that she was unaware of any other insurer who had even attempted to use an individual methodology for individually broked cases.

  310. Mr Dutton points out that the objectives which the Pursuit Policy sets out to achieve are:
    1. access to the court for a meritorious claim;

    2. an affordable premium commensurate with risk (the quantum of the premium and whether or not it is "affordable" is of course at the heart of this issue);

    3. a deferred premium to enable the Claimant to pursue a claim which might otherwise have to be abandoned;

    4. a premium sufficient to reflect the risk at the time when the case is underwritten;

    5. a premium which allows for the risk increasing;

    6. a premium which meets the increasing exposure risk the longer the case goes on;

    7. a premium which reflects the risk (in percentage terms) at the point when the policy is written (he says that this is what makes the solicitor's and counsel's risk estimate important information for the insurer who needs to make an assessment of prospects of success);

    8. a premium which both encourages all the parties to keep costs low and to settle a case as soon as possible commensurate with the merits;

    9. a policy which provides sufficient cover for the case if it goes to trial without the need for the purchase of expensive top-up cover as costs increase beyond early estimates.
  311. It is necessary for the insurer to make sure that the premium income is sufficient to meet losses on losing cases. Mr Dutton suggests that the methodology by which the premium is calculated brings all these factors to bear when the premium rate is set, at the point when the risk is underwritten. He suggests that the real argument is about the reliability of the estimates used by solicitors and insurers when arriving at the premium rate. In his submission, if the risk assessment and costs assessment used at the point the policy is written are correct, then the premium charged will accurately represent the risk underwritten. He does however accept that the rating mechanism is susceptible to an estimate being wrong.

  312. He submits that in order to set a premium in an individually written case the insurer needs to know what are the prospects that he will be called upon to indemnify the insured and how much he could be called upon to pay, if the risk becomes a reality at trial or at any stage when the case may fail.

  313. He submits that the correct approach is not to ask whether another insurer could have written the policy more carefully, but rather, was the policy written on a reasonable insurance basis? No insurance premium can ever be 100% perfect in the way it is rated. In a fixed mass market, premiums are rated across the board of risk, and if a case settles quickly the premium may well turn out to have been too high for the risk that was written. In a tiered premium case, a stage by stage premium case, a premium is likely to be too much at each point where the money comes in at the tier and it may then reach a point, as the litigation progresses and the risk becomes harder and harder to insure, where the amount required at each tier may go up dramatically. The Pursuit policy is the only sliding scale premium product. He submits it has been designed very carefully with two particular features in mind, firstly to give unlimited cover for opponent's costs and disbursements; and secondly, it is calculable because of the use of the own side's costs proxy at every point of the case along the way. It is calculated on the basis of bespoke individual risk underwriting, i.e. on the basis of exposure risk for the individual case.

  314. In relation to estimates Mr Dutton submits that any insurer who has to underwrite a risk needs to know what his exposure might be. Thus the insurer has to have an estimate of the potential exposure and, in the circumstances of these cases the people best placed to give such estimates are the Claimants' solicitors. The Claimant's solicitor must advise the client as to the likely cost risks of embarking upon a piece of litigation. With regard to the inaccuracy of the estimates, he submits that insurance underwriting works on the basis that estimates have to be given, they have to be reasonable, but they may be wrong. They may be too optimistic or too pessimistic. The Defendants suggest that if the estimates appear unreasonable or wrong, it is permissible to look again at the policy and put in the figure for costs which should have been estimated. In Mr Dutton's submission the court should only embark upon that exercise if the estimate was unreasonably wrong at the point when the estimate was given.

  315. Mr Dutton submits that the approach of First Assist to the assessment of insurance risk is not unreasonable and that if, on the individual facts of any of the Test Cases, something has gone wrong in the calculation of the percentage rate to cover the risk, which means there should be an adjustment at the end of the day, that should be done, but, in his submission it is only in the case of Farr where things have gone demonstrably wrong and, by concession, the appropriate adjustment has been made.

  316. Mr Dutton argues that all insurance is written upon working assumptions and the Pursuit Policy methodology is a reasonable working assumption for calculating individual risk premiums for each case, taking account of the various risk factors which apply in litigation. The insurers have to know what their risks are as to adverse costs (including disbursements) and own disbursements. Mr Dutton argues that this can only be based on an estimate, since the costs have yet to be incurred and the litigation has yet to be run. There is no one better placed than the solicitor for the claimant to give such an estimate, but in any event the insurers do not simply adopt those estimates, but apply their own minds to them, varying the estimates of the prospect of success.

  317. In estimating the prospects of success the solicitor has to work out his own risk assessment when taking a case on under a CFA. The insurers in considering the ATE position adjusted the solicitor's risk assessment in three out of the five Test Cases.

  318. No premium, however set, will accurately reflect the ratio of own to adverse costs at all points along the way in the litigation. Mr Dutton suggests that a stepped premium policy would start too high in relation to adverse costs risks, assuming the case settled early, and may become insufficient as the case progresses, possibly requiring top-up cover. He argues that in more complex litigation, own and adverse costs much more nearly reflect each other than in simple running down cases of the type considered in Callery v Gray. He suggests that own and adverse costs are likely to increase proportionately to each other as the case goes on. Furthermore, he suggests that the cases insured under the Pursuit Policy tend to be more difficult and risky and thus settle later, meaning that the costs will have caught up with each other by the time the case is concluded.

  319. The premium charged by RSA does not assume that the case will inevitably go to trial. The premium is based on the costs incurred to the date of settlement. The insurer needs to know his worst case position and must ensure that the premium rate charged ensures that a sufficient premium is obtained for the risk of the case failing at trial. Mr Dutton argues that the estimate of adverse costs to the end of the trial enables the insurer to set the appropriate rate.

  320. There remains the question which may occur where the premium multiplier is very high. In deciding this question, in common with other questions relating to the reasonableness of costs, it is not permitted to use hindsight. It is extremely difficult, even for experienced litigation solicitors, accurately to predict the future risks of litigation, the likely response of the other side, what costs each side may incur in pursuit of their respective cases and what witnesses may or may not be credible. Accordingly, if the solicitor's estimates appear reasonable, as judged by the facts known at the time of the estimate, it should not be necessary for the court to embark upon a minute examination of those estimates.

  321. In those cases where the court is not satisfied as to the reasonableness of the Pursuit Policy and the premium, the court should:
    1. ask whether it was reasonable to take out the Pursuit Policy;

    2. ask whether the multiplier used produces a premium which is significantly above the range of any comparators. If the premium falls into such a range it is unlikely that any further examination will be needed;

    3. if the premium falls significantly outside the range, the court should ask whether, on the facts known or which ought to have been known to the Claimant's Solicitors at the time the estimates were made, whether those estimates were so unreasonable as to fall outside the range which any reasonably competent solicitor should have estimated;

    4. if the court concludes that the solicitors' estimates were unreasonable, and were used by the insurer, the court should examine the premium figure with a view to awarding a reasonable sum. In doing this the court should apply the estimates as they should have been given to reach a reasonable and recoverable premium for the purposes of Section 29.

    Test Case Claimants

  322. Under this head Mr Morgan adopts Mr Dutton's arguments. He additionally argues that the Claimants are merely litigants seeking to purchase a product in a market. They are not free to haggle, they simply have to choose between rival products offered at the prices which insurers charge for them. In his submission, if there is no other product available for a claimant their only possibility, if they do not wish to take the risk of unprotected litigation, is to accept the offer of insurance at the price which the insurer charges for it. Thus, he argues, even if the court were to form the view that it was over priced, unless the Claimant unreasonably rejected some cheaper suitable alternative product, the court should allow recovery of the premium.

    Defendants 1 - 4

  323. This issue is at the heart of Mr Bartlett's case. He identifies six flaws in the premium methodology. He suggests that the size of the premiums is not accounted for either by the fact that the indemnity for adverse costs is unlimited or by the fact that payment of premium is deferred until the end of the case. He says that the premium charged to an insured should fairly reflect the risk that the particular insured brings to the common pool, and in the Test Cases the premium does not fairly reflect that risk.

  324. The flaws identified by Mr Bartlett are as follows:
    1. The methodology assumes a constant relationship between costs at risk and own costs.

    2. Where the policy is taken out some time after the CFA was entered into the premium starts at a high level and goes upwards.

    3. The premium rate is critically dependent on the relative sizes of the estimates of own costs and defendant's costs in every case. Where the ratio of defendant's costs to own costs is overstated the premium rate is too high.

    4. First Assist adopt estimates of success which are the same as, or too close to, those given by solicitors for a different purpose and on a different basis and which are therefore too low for the purpose of rating an insurance premium.

    5. The final premium is calculated by reference to claimant's costs as claimed however unreasonable or excessive.

    6. The premium in successful cases is calculated so as to make up for the lack of premium income in unsuccessful cases.

    1. Constant Relationship

  325. Mr Bartlett argues that the evidence shows that the assumption of a constant relationship between costs at risk and own costs is contrary to normal experience. Normally adverse costs will start at or close to 0% of own costs (because at the outset the claimant has incurred some costs and the defendant has incurred none) and the proportion will gradually rise. Since the First Assist premium rate is based on the estimated proportion to be reached at the end of trial, if the case settles before trial, the effect will be that the premium will be overcharged.

  326. The use of the solicitor's own normal costs for the rating mechanism as the basis for the calculation of the premium is flawed since the defendant's costs and own disbursements will never normally be known to the claimant's solicitors.

    2. Policy taken out after CFA entered into

  327. In three of the five Test Cases (D. Baker, A. Baker and Clarke) there was a substantial time lag between the date of the CFA and the inception of the Pursuit Policy and the policy was taken out at around the same time that the defendant solicitors were instructed. As a result, Mr Bartlett says, when the policy was taken out costs had already been incurred under the CFA but at the outset there would have been only minimal adverse costs. He argues that while a flat rate premium might also appear high in the early stages of the case, such a premium is set on the basis that it is not possible to tell when the case will settle and the premium must make a reasonable contribution to the pool, whether settlement is early or late. With a sliding scale premium, however, this should reflect the differing levels of risk during the case. To do that fairly, he says, it should start far below the level that would be appropriate for a flat rate premium.

    3. The relative sizes of the estimates

  328. It is the ratio between the adverse costs and own costs that is important rather than the absolute amount of the estimate. The premium rate calculation is very sensitive to that ratio. Mr Bartlett argues that estimates may very easily be skewed in a way which produces premiums which are much too high. Since the estimates are so erratic they do not provide a proper basis for a rational estimate of a premium to reflect the risk.

  329. He suggests that there is an incentive for claimants' solicitors to over estimate adverse costs in order to ensure adequate cover for the client. He points to Ms Steven's evidence that she would not wish to give a false indication of the adverse costs for fear of the insurer pleading material non disclosure. He points out that of the solicitors' estimates in the Test Cases, that given by Mr Scrivenger was meant as an estimate of cost to settlement rather than to trial. Ms Stevens confused her figures for a one day trial with figures for a two day trial and had to correct them later, and Mr Bennett's figures were wholly incorrect. Thus, he says, since the evidence shows that the estimates used are not inherently reliable, there is no basis for believing that the premium rate at the outset is the right rate for the risk.

  330. Since First Assist treats VAT in the same way for both sides, whereas in the Test Cases the Defendant's cost are in fact exclusive of VAT, this tends to overstate the Defendant's costs in relation to Claimant's costs still further.

  331. Mr Bartlett argues that the problem is illustrated by comparing the estimated proportions of costs at risk to own costs with the actual proportions in the Test Cases. In every case except Sandiford the estimated proportion was at least double the actual proportion at the time of settlement (in Sandiford the estimate of Defendant's costs came from the Defendant).

  332. Mr Bartlett accepts that the use of a proxy in assessing risk is a recognised technique in underwriting. Normally a proxy would be used where there is evidence to demonstrate that it is an accurate predictor of experience. He suggests that the evidence does not demonstrate that the Pursuit methodology is based on any such evidence.

    4. Estimates of success

  333. Mr Bartlett points out that all the Test Cases in which he is instructed settled before trial and should have resulted in very substantially lower premiums than if the case had gone to trial. The solicitor is required to give an estimate of the prospect of success at a particular damages figure. Self evidently, he says, the chance of achieving a substantial award of damages at a particular figure is lower than the chance of the ATE insurer avoiding having to pay out. Using the solicitor's figure as a starting point therefore has the effect that the success rates are estimated too low for insurance purposes. In addition he suggests, solicitors and counsel tend to be conservative in their estimates of success. He suggests, correctly, that the First Assist premium is very sensitive to the estimate of success. Getting the success rate too low, even by a little, makes the premium too high, as demonstrated by Ms O'Reilly in the table which she produced. Mr Bartlett suggests that on the evidence produced by Mr Smith (PWS3 and PWS4) the failure rate was 18% and the success rate 82%. He therefore argues that in clinical negligence cases the success rate must be at least 82%. He also argues that exhibit PWS3 demonstrates habitual under-estimation. He suggests that the flaw is accentuated by the move away from the original policy concept that the underwriter would set, as the definition of success, a realistic amount for financial settlement. If that amount was not achieved there was no success and no premium was due under the policy. The more flexible wording for the definition of success used subsequently means that the chances of achieving at least some success are far higher than the solicitors original risk estimate for CFA purposes.

  334. Mr Bartlett also points out that since no commission is paid on unsuccessful cases the premium calculation is set at a level to cover the whole of the absence of a premium (including commission). This, he suggests, may amount to a significant over compensation.

    5. Premium calculated on Claimant's costs as claimed

  335. The premium is only calculated after the substantive litigation has been successfully completed for the claimant. It is calculated on the costs claimed by the claimant's solicitors, rather than the costs agreed or allowed on detailed assessment on the standard basis. Mr Bartlett submits therefore, however unreasonable or disproportionate the costs claimed may be, they still form the basis of the calculation of the premium and tend to produce an over estimation of the amount of premium necessary.

    6. Premium calculated so as to make up a lack of premium income in unsuccessful cases

  336. The premiums claimed do not represent simply the risk undertaken in the particular case in which they are collected. The calculation of the premium also allows for the lack of any premium income in cases that are lost. Mr Bartlett submits that the element of the premium attributable to this is irrecoverable from the defendants as a matter of law. He argues that this element of the premium is not a "consideration paid or payable for an insurance against the risk of incurring a costs liability in the proceedings" (Callery v Gray (No.2) paragraph 8). He argues that that element of the premium relates to other proceedings which are lost and does not come within Section 29 as interpreted by the Court of Appeal. It is therefore irrecoverable. He distinguishes the premium element which was held to be recoverable at paragraph 63 of Callery v Gray, which provided cover for the claimant against the risk of being unable to recover the premium from the defendant as a consequence of losing the action. He argues that, whilst the legislation authorises the recovery of success fees to cover the lack of income for losing cases, there is no such explicit authority for recovery of an insurance premium to cover a lack of premium in losing cases.

  337. Mr Bartlett argues that the ordinary concept is that everyone pays a premium appropriate to their risk. Those premiums go into a pool so that there are funds from which to pay the claims in the cases where the insured loses. In First Assist cases, he suggests, the insured is in effect paying someone else's premium in addition to his own in order to keep the pool up to the correct level.

    Defendant 5 (Farr)

  338. Mr Bacon again adopts Mr Bartlett's submissions in relation to the giving of estimates. He points out that, in the case of Farr, Mr Bennett confirmed that the estimate which he provided to First Assist was "not a well done document". He argues that a premium calculation which relies on estimates being reliable and accurate is, by its very nature, one likely to lead to inaccurate estimates of risk.

  339. He argues that the premium calculation uses both actual up to date information and estimated data to arrive at the amount payable. This is an inherent flaw.

  340. In the case of Farr, Mr Bennett estimated his own base costs at ?21,500. The actual costs incurred were ?59,680. He put the prospect of success as greater than 75% (which was not accepted by First Assist) and he omitted to give any estimate of the Defendant's costs. The premium was originally rated on notional defendant's costs of ?112,000. It has since been conceded that this was incorrect. Utilising Mr Bennett's own risk assessment of 75% and utilising a figure of ?84,000 for the Defendant's costs (which is the figure Mr Bacon says should have been used) the resultant premium rate is 149%, giving a premium of ?88,953. Utilising the actual costs figures produces a premium rate of 53% giving a premium of ?31,641. When the VAT element of the Defendant's costs is eliminated, as Mr Bacon argues it should be, the premium rate is reduced to 49%, producing a premium of ?29,253. Mr Bacon argues that the actual adverse costs exposure could be obtained from estimates provided to the court and the Claimant's solicitors in accordance with the CPR or from estimates ordered to be produced by the court.

  341. It is wrong and unreasonable, Mr Bacon submits, for First Assist to rate the premium in the Farr case by reference to both the likely adverse costs of the liability trial and the likely adverse costs of a quantum trial. He argues that, where a split trial had been ordered prior to the inception of the policy, there would be no liability on the part of First Assist to pay out any adverse costs in respect of the quantum trial. He says that success is achieved within the meaning of the policy if the eventual outcome of the case is an order in the Claimant's favour, whether or not the solicitor has advised against the acceptance of an offer at an early stage. Therefore, in the case of Farr, where success on liability is known, at the time of the rating of the premium, to be highly likely, and where no issues of causation would prevent at least some recovery, the policy has little if any prospect of being required to pay out. The rating of the policy should therefore be based on the risks of not winning on liability alone in such a case.

    ISSUE 3 - CONCLUSIONS

  342. The use of estimates by solicitors has been brought very much more into focus by the Court of Appeal decision in Leigh v Michelin Tyres [2003] EWCA Civ 1766, in which the court requested the Civil Procedure Rule Committee to consider further the Practice Direction relating to the giving of costs estimates. It may be that in the future Section 6 of the Costs Practice Direction will be amended, but at the moment the position is as described in Leigh.

  343. Ms O'Reilly in her evidence did not dispute that if the estimates given are accurate the insurer will be obtaining the correct risk premium for a case which it is envisaged will go to trial:

    "If he has accurate estimates to trial he will be able to accurately estimate the premium at trial but it may be over stated before trial … At any point in time if he knows exactly what the ratio is [between own side and adverse costs] and he knows the own costs, then by definition he knows the adverse costs … If the assumption of a constant ratio between adverse costs and own costs is correct then yes you are right [if the estimates are reasonable] he has a reasonable basis for calculating the premium."

  344. With regard to the estimates of own costs, adverse costs and own disbursements given by the claimants' solicitors these are clearly susceptible to being wrong, however careful the solicitor may be in arriving at the estimate. Although the insurers consider and sometimes alter the estimate of the prospect of success they are not in a position to alter the estimates of costs and disbursements. The cost estimate in Farr was queried but this was because the solicitor had failed to provide any figures for Defendant's costs.

  345. An underlying element of the premium rating methodology is the assumption that the ratio between Claimants and Defendants costs will remain constant as the litigation progresses. Mr Scrivenger, who had great experience in clinical negligence acting both for Defendants and Claimants, was of the view that in practice the Claimant's costs would mount up quickly at first and the Defendant's costs would catch up later.

  346. Solicitors find it difficult accurately to estimate their own costs and disbursements (see Leigh v Michelin [2003] EWCA Civ 1766; [2004] 2 All ER 175) and will find it still more difficult to estimate their opponent's costs and disbursements. In some cases (e.g., Sandiford) the estimate for Defendant's costs given by the solicitor was in fact a figure provided by the Defendant. Clearly the Claimant's solicitor could not have given a more accurate estimate for the Defendant's costs, but there is no way of knowing whether the Defendant's own estimate was accurate or not. I accept Mr Bartlett's submission that, given the unreliability of such estimates, a calculation based on them is inherently likely to be arbitrary.

  347. In my judgment it cannot be reasonable to require the paying party to pay a premium based on costs claimed which may be higher than those which the court has found to be reasonable and proportionate. It must follow that if the costs claimed have been found to be unreasonable and disproportionate, the premium calculated on the basis of those costs must itself be unreasonable and disproportionate.

  348. Mr Bartlett submits that the element of premium attributable to the lack of any premium income in cases that are lost is not a "consideration paid or payable for an insurance against the risk of incurring a costs liability in the proceedings" (Callery v Gray (No.2) but relates to other proceedings which are lost and therefore does not come within Section 29 of the 1999 Act.

  349. He argues, by reference to the case of Deborah Baker where the estimated prospect of success was 70%, that, if premium were received in losing cases, the risk multiplier would be 30/100, but in order to allow for the non receipt of premium income in losing cases the risk multiplier is instead 30/70, thereby increasing the premium payable. Mr Dutton and Mr Morgan argue, in my view correctly, that there is no difference in substance between the First Assist method of insuring that claimants do not have to pay the premium in losing cases and the self insuring policies.

  350. Ms O'Reilly said:

    "It is the ratio between the adverse costs and the own costs that is important rather than the absolute amount of the estimate."

  351. But as Mr Bartlett submits, since the estimates are erratic they do not provide a sound basis for a rational estimate of a premium to reflect the risk. The estimates used are not inherently reliable, particularly the risk estimate, and there is no basis for believing that the premium rate at the outset is the right rate for the risk.

  352. When one compares the estimated proportions of costs at risk to own costs with the actual proportions in the Test Cases, the estimates are at least double the actual, at the time of settlement (save in the case of Sandiford when the Defendant's own costs estimate was adopted). Not only are the Claimants' solicitors' estimates of future costs and disbursements inaccurate, the use of the solicitor's estimate of success for CFA purposes as a starting point for assessing the risk for insurance purposes may also be flawed. I accept Mr Bartlett's submissions to the effect that the risk to the ATE insurer normally is lower. As Ms O'Reilly demonstrated in her table, if the success rate is fixed too low this will make the premium much too high. An examination of the NHSLA cases in PWS3 and PWS4 demonstrates a success rate of at least 82%.

  353. Without exception the Claimants' solicitors in the Test Cases all thought the premium rate calculation would produce an extremely high premium, but felt that they had nowhere else to go.

  354. Mr Morgan submits that, provided the Claimant had behaved reasonably the court should allow recovery of the premium, but that would not, in my judgment, produce a fair result. The Defendants are required to pay reasonable and proportionate costs and to require them to pay premiums which do not meet those tests cannot be correct. Given my findings that the premium rate methodology is inherently flawed because of the inaccurate estimates, this will undoubtedly raise issues between the Claimants and First Assist which will have to be resolved either by agreement, or, as a last resort, by further litigation.

  355. With regard to the question of hindsight, Mr Bartlett argues that paragraph 11.7 of the Costs Practice Direction applies only to the assessment of a success fee. I reject that argument. The Direction refers to "an additional liability". This is defined at CPR 43.2(1)(o) as including an insurance premium. It is a general principle of costs law that hindsight should not be used (see Francis v Francis & Dickerson [1956] p.87, Sachs J, Lownds v The Home Office [2002] EWCA Civ 365 and KU v Liverpool City Council [2005] EWCA Civ 475 quoting Argyle (Duchess) v Beuselink [1972] 2 LLR 172 Megarry J with approval). Given, however, my finding that the methodology is flawed, one is left with no alternative but to use an element of hindsight when arriving at a figure for premium which is fair to the parties. First Assist reserved to itself a right to have the insured's solicitor's bill of disbursements assessed on the standard basis (see the definition of "normal fees" set out in the policy terms quoted at paragraph 20 above). Similar protection should have been provided for the insured claimant in respect of the premium calculation. It is not of course possible to reserve a right to have the Defendant's costs assessed. Nevertheless, in my judgment, some protection as to inaccuracy in respect of the Claimant's Solicitor's profit costs and in respect of the proportion it bears to the costs at risk, ought reasonably to have been included in the policy terms. In my judgment the need for some contractual provision allowing for adjustment with the benefit of hindsight is essential where the methodology of premium calculation is exposure related (as here) rather than experience related (as with most other ATE policies).

  356. I accept that, the Pursuit Policy having been launched in March 1999, First Assist should have built up considerable experience by the time that the policies in these Test Cases incepted (August to October 2002). One could reasonably expect that experience to demonstrate that the methodology was not working satisfactorily. It is clear from the evidence of Mr Smith that the policy was designed at a time when premium recovery was not permitted and the policy was principally designed for use in commercial cases. Once the premium became recoverable the use of the policy was extended to personal injury and clinical negligence cases. The methodology was not however adjusted to make it suitable for the new situation in that class of case.

  357. Dealing with Mr Morgan's submission that, provided the Claimant has acted reasonably the court was bound to allow the full premium charge, Mr Bartlett argues that the approach is fundamentally wrong as a matter of law and unrealistic as a matter of fact. He relies on Callery v Gray (No.2) paragraph 69:

    "When considering whether a premium is reasonable the court must have regard to such evidence as there is, or knowledge that experience has provided, of the relationship between the premium and the risk and also of the cost of alternative cover available."

  358. In these Test Cases the evidence of the availability of alternative cover is not satisfactory or persuasive. It is clearly very difficult, in cases of the type before the court, to obtain cover at a reasonable cost, which does not have severe and unwanted restrictions. There is no evidence before me of any efficient market for ATE products dealing with this type of case. Had I found that the premium calculation methodology was without flaw, then Mr Morgan's submission would succeed. In the light of what I have found it clearly cannot do so. First Assist has not sought to explain its premium rate on the basis of market forces, but rather on the basis of a genuine pre-estimate of the sums it needs to make a fair return on its business. I accept First Assist's submissions as to commission rates and administrative costs needed to produce a fair return, but I do not accept its submission that its system of costs estimates has had the effect of producing a fair return.

  359. Mr Bacon suggests that none of the Rules or Practice Directions specifically rule out the use of hindsight when assessing an ATE insurance premium and he submits that it is permissible for the court to apply actual cost figures incurred, rather than estimates, to calculate the reasonable premium, because the premium is not itself calculated until the end of the case, when the own side's costs figure is actually known. Furthermore, where, as in the case of Farr, the estimates are clearly wrong the court must, he says, be entitled to arrive at a figure for premium which reflects the eventual outcome as being a more reliable basis on which to judge reasonableness. Mr Dutton accepts that hindsight can be used as a check to test the reasonableness of what a solicitor did at the time. In the case of Farr there was an obvious error in the solicitor's estimate and First Assist have agreed to adjust the premium. To that extent, therefore, hindsight has been used in that case. Mr Dutton suggests that when testing a professional judgment it is permissible to look at events as they have subsequently turned out, but when testing professional judgment as to assessments of future costs to trial or settlement, it is possible for capable and careful solicitors to reach different judgments. He submits that it is not permissible then to superimpose onto the premium calculation, figures which have turned out to be different at the end of the day, save to the extent that an unreasonable estimate has been given.

  360. When completing the proposal form in the Farr case, Mr Bennett estimated the likelihood of his client recovering ?200,000 as greater than 75%. Leading Counsel was firmly of the view that Mr Farr would succeed. Mr Bacon argues that there was no good reason for First Assist to calculate the premium based on a 70% likelihood. Although Mr Smith said that they did not believe 75%, Mr Bacon says he gave no credible reason why. He submits it was in First Assist's interest to increase potential premium by reducing the chance of success. Ms O'Reilly explained that a 5% reduction increased premium by 28%.

  361. With regard to Mr Bacon's submission that the premium should be rated only up to the liability trial, I accept Mr Dutton's submission that a final settlement of the claim for damages in the Farr case would not have occurred until there was a judgment in his favour at the conclusion of the trial on quantum, not just on liability, or a settlement by which Mr Farr recovered compensation for the injuries suffered by him. It is not for this court to speculate whether a trial which leads to a judgment for the Claimant but nil damages, or damages below the level of a Part 36 offer or payment, where the court makes a costs order against the Claimant, is nevertheless a "final settlement in favour of the insured litigant" and therefore a success. Such a decision must depend on the individual facts of each case and this situation has not arisen, either in the case of Farr or in any of the other Test Cases. In the Farr case Mr Bennett discussed the Part 36 offer which had been made with the insurers and corresponded with them about it. Mr Bennett, his client, and the insurers considered that the Part 36 offer was too low and should be rejected, and the insurers gave their continued support of Mr Farr to trial. In the event Mr Farr was wholly successful and the possible difficulties which I have mentioned did not arise.

  362. In summary therefore in relation to the flaws put forward by Mr Bartlett I accept the Defendants' submissions that the premium rating methodology is inherently and seriouslyfundamentally flawed in that it assumes a constant relationship between costs at risk and own costs; and, depends too heavily on the relative sizes of the estimate of own costs and defendant's costs, which are very likely to be inaccurate; and the final premium is calculated by reference to the Claimant's costs as claimed, however unreasonable or disproportionate. With regard to the fact that the estimates of success adopted by First Assist may be too low in that they do not accurately reflect the likelihood of the insurers having to pay out; the premium starts at a high level where the policy is taken out some time after the CFA was entered into, these are fact sensitive and do not in my judgment less seriousamount to fundamental flaws. As to the argument that the premium in successful cases is calculated so as to make up for lack of premium in the unsuccessful cases, I do not regard it as a fundamental flaw.

  363. Given that I have found that there are inherentfundamental flaws the consequence is that the recoverable premium in each of the Test Cases (subject to my findings in respect of the remaining issues) will be such sum as it is reasonable and proportionate to expect the paying party to pay having regard to all the circumstances of each case and the factors as set out in CPR 44.5. In calculating the sum to be taken into account, in respect of the Defendant's costs to date, the parties should, in my view, make some sensible approximation on the basis that both parties' trial costs (including counsel's fees and expert's fees) are likely to be similar, but that the Claimant's pre trial costs are likely to be substantially greater than the Defendant's. As a general rule the Claimant's pre trial costs are likely to be as much as twice the Defendant's. Of course, this may not be true where each side has incurred heavy expert's fees before trial. In all cases reliable information provided by the Defendants would limit the need to rely upon such approximations.

    ISSUE 4 - What commissions, if any, are payable to the Claimants' legal representatives and/or any other agents of the insurers and if there are any, when and in what circumstances are they payable?

  364. Having heard the evidence it is accepted by the Defendants that no point arises under this issue.

    ISSUE 5 - Should the amount of the recoverable premium be reduced on the grounds that an insurance policy ought reasonably to have been taken out at an earlier stage in the proceedings?

    First Assist

  365. This point was abandoned in respect of Anthony Baker and was not run in the case of Deborah Baker. Mr Dutton suggests that the appropriate point to take out insurance is when a Claimant is facing a costs risk and has explored avenues of resolution of the case. He says that this argument was put forward by Norwich Union in Callery v Gray (No.2). In that case it was found that, in simple straightforward road traffic personal injury claims, it is not unreasonable to insure early. This did not amount to a requirement that Claimants should insure early. In any event there is little evidence on the facts of the five Test Cases that it would have made the slightest difference beneficial to the Defendants had they done so.

    Test Case Claimants

  366. Mr Morgan denies that a claimant is under any obligation to take out ATE insurance at the earliest possible moment. The Court of Appeal in Callery v Gray (No.1) merely indicated that it would be permissible for a claimant in a low value low risk RTA case to take out insurance at the outset. He makes submissions about Sandiford and Clarke.

    Sandiford

  367. Mr Morgan argues that it was reasonable not to take out ATE insurance before the CFA was entered into on 21 August 2002. In any event, he says, relying on the evidence of Mr Gilbert, delay in this type of case would not have any substantial effect on the availability of insurance. He points out that Mr Besfor made extensive enquires direct of insurers and used the brokerage service of The Judge. In the end the Claimant was left with no alternative policy.

    Clarke

  368. Mr Morgan goes on to submit that the Claimant's first solicitors, Hart Brown, cannot be criticised for failing to obtain ATE insurance at an earlier stage. That firm was not prepared to enter into a CFA at the outset because of doubts about the merits of the case. He submits that they acted reasonably in sending a protocol letter and giving the Defendants as much time as necessary to respond. He suggests that it is clear from Ms Stevens' evidence that the Defendants were not interested in settling without proceedings being issued. She applied to The Judge for insurance within three weeks of being instructed. Any subsequent delay in obtaining a quote was because of the delay in obtaining medical evidence. She specifically considered with counsel in conference, the likely costs on the other side and the proportionality of taking out this particular policy in this case, and she reached the view that it was proportionate and reasonable to do so.

    Defendants 1 - 4

  369. Mr Bartlett argues that since the Court of Appeal in Callery v Gray (No.2) at paragraph 87, accepted that policies ought to be taken out early in order to keep premiums down, it follows that policies ought not to be taken out late, because it has the result that premiums are unnecessarily high since the many are no longer paying for the few. He suggests that waiting until there is a risk of a liability for adverse costs cuts across the approach of the Court of Appeal in Callery v Gray. It also ignores the fact that cover may be needed before that time. The risk of incurring disbursements which may not be recovered from the defendant commences as soon as the claimant's solicitor is instructed. He suggests that in no class of case should the policy be taken out after proceedings have been served.

  370. Mr Bartlett submits that, in general and as a matter of principle ATE policies should reasonably be taken out at a stage when there is likely to be more choice and lower premium prices available, rather than at a later time nearing the end of the limitation period and/or when proceedings have already been issued.

    Sandiford

  371. He points out that Mr Besfor had received a report on the Claimant from Professor Hirsch which he discussed in conference with counsel on 19 June 2001. Proceedings were issued on 31 July 2001 and served four months later. It was not until May 2002 that any attempt was made to find an ATE policy. He suggests there was culpable delay in investigating ATE policies with no satisfactory explanation from Mr Besfor.

    Clarke

  372. Mr Bartlett suggests that Mr Clarke's first solicitors, Hart Brown, should have investigated ATE policies early in 2001 when limitation was not about to expire and proceedings did not have to be issued urgently. Whilst he accepts that there is no positive evidence as to available policies at that time he argues that the burden remains on the claimant throughout to prove that a reasonable choice was made at a reasonable amount.

    Defendant 5 (Farr)

  373. Mr Bacon argues that in the particular circumstances of the Farr case it was not reasonable to utilise the BTE insurance cover before going to the ATE market. He suggests that it ought to have been obvious to Mr Bennett, by the time the letter of claim was dispatched, that the limit of cover of ?25,000 was not going to be sufficient in a serious and potentially high value claim. He points out that if an ATE product had been take out in September 1999 the premium would not have been recoverable under Section 29 of the 1999 Act. Mr Bacon does not take that point, but suggests that a realistic date to go to the ATE market would have been at about the time of the letter of claim in June 2001. He suggests that a range of premiums between ?205 and ?3,413, with a limit of indemnity of ?100,000, would have been available. He relies on the evidence of Mr Burrell in this regard.

    ISSUE 5 - CONCLUSIONS

  374. In my judgment the Court of Appeal in Callery v Gray (No.2) did not state that ATE insurance ought not to be taken out late because this would have the result that premiums would be unnecessarily high, since the many were no longer paying for the few. The court in fact stated that it was not unreasonable for ATE insurance to be taken out early. The correct test is whether the individual claimant has acted reasonably in taking out ATE insurance when he or she did so. That is a decision which can only be taken having regard to the individual facts in each case.

  375. In the case of Sandiford Mr Bartlett relies on Mr Gilbert's evidence that the Sandiford case might have been eligible for an Accident Line Protect Policy, provided the application had been made before the letter of claim was sent. He suggested that the premium would have been ?3,045 including IPT. Mr Besfor was concerned because his firm had not put in place the complementary disbursement funding which was deemed necessary. Thus, although there was a possibility of obtaining ALP insurance, I am satisfied that on the facts as they appeared to Mr Besfor at the time it was reasonable not to do so.

  376. Having regard to the facts in the other Test Cases where this issue arises (Clarke and Farr) I am satisfied that the individual Claimants acted reasonably in taking out insurance when they did.

  377. Given the view which I have taken of Mr Burrell's evidence, Mr Bacon's submission that suitable cheaper alternative policies would have been available is not made out.

    ISSUE 6 - Has the claimant acted reasonably in taking out the RSA Pursuit policy, and if not what are the consequences?

    First Assist

  378. Mr Dutton submits that the evidence shows that there was nothing unreasonable in the individual Claimants taking out an RSA Pursuit Policy, nor is there any evidence to show that there was an alternative policy which was both suitable and available.

  379. He points out that the definition of success in each of the Test Case policies is similar (but not always identical). I have set out those definitions under the factual descriptions of the individual cases above. He suggests that, in the case of Farr, a final settlement of the claim for damages would not have occurred until either there was a judgment in his favour at the conclusion of the trial on quantum (not just on liability) or a settlement by which Mr Farr recovered compensation from the Defendants.

    Test Case Claimants

  380. In respect of the individual claims Mr Morgan submits that each had one or more of the following characteristics:
    1. all were claims which, by their nature or on the evidence, were difficult to insure;

    2. in all cases proceedings had to be issued because of limitation considerations;

    3. in three cases (Sandiford, Clarke and Baker v Euromark) reasonable enquiries had revealed no alternative policies on any terms;

    4. in Baker v Addenbrookes there was one alternative policy but at a substantial premium payable immediately which the client could not afford;

    5. in Farr the Claimant was in a difficult position because his BTE limit had been used up, he reasonably selected a policy with the same insurer which enabled him to maximise the benefits of the BTE policy;

    6. in three of the cases (Baker v Addenbrookes, Farr and Baker v Euromark) the Defendants were warned that insurance policies were about to be taken out and declined the opportunity to make/accept offers or admit liability before this was done;

    7. in Clarke the Defendants turned down an offer made within two months of insurance being taken out and in Baker v Addenbrookes turned down an offer made before the insurance was incepted;

    8. in all the cases the conduct of the Defendants in failing to make or accept offers was a factor without which the premium would have been nowhere near as high as they were;

    9. in all cases the Claimants were advised by solicitors and in three (Sandiford, Clarke and Baker v Addenbrookes) by a reputable broker before taking out the policy.
  381. As to the reasonableness of the choice of policy, Mr Morgan submits that the Claimants have to show only that they acted reasonably, and that on the evidence they did so. The availability of a deferred premium, of no premium being payable in the event of failure, and of unlimited liability for adverse costs, thereby eliminating the risk of onerous top-up conditions, were important attractions for the Claimants.

    Deborah Baker

  382. Mr Scrivenger acted for Deborah Baker. Mr Morgan points out that Mr Scrivenger had obtained an offer of insurance from LPL but this was not acceptable because Mr and Mrs Baker could not afford to pay a premium at that time. A deferred premium was needed. It was suggested in cross examination that Mr Baker's company might have been able to fund the premium, but, in Mr Morgan's submission, which I accept, the figures exhibited in the company's documents, which were produced at a late stage in the proceedings, do not demonstrate that there was money available to fund the premium. In addition Mr Scrivenger was of the view that there were too many conditions attached to the LPL policy and a level of interference by the insurer, which he would now certainly find unacceptable, and probably did then. "They probably had less clinical negligence knowledge than I did." Mr Scrivenger did what he could to overcome the funding difficulty, approaching First National Bank without any success. He also approached Greystoke who were unable to help. Finally he consulted Mr Gilbert at The Judge and relied on his expertise. The only product which was available and suitable was the First Assist Pursuit Policy. He was extremely concerned at the level of premium but informed the Defendants of the availability of the policy in his protocol letter and also made a Part 36 offer on behalf of the Claimant.

  383. Mr Morgan also points out that Mr Scrivenger had tried to arrange with NHSLA to extend the limitation period so that the inception of the insurance could be delayed until a response to the protocol letter was forthcoming. The NHS Trust was willing to agree to this proposal but the Litigation Authority was not. Had the request been granted the Defendant could have fully investigated the circumstances without incurring any liability for insurance premium at all. In addition the Litigation Authority had obtained specific advice from Mr Ivory, the costs draftsman.

  384. Mr Scrivenger's comments on the Pursuit Policy, as an experienced clinical negligence practitioner, were that it was extremely good, because it gave unlimited cover, it avoided the risk of top up premium and the deferred premium was essential on the particular facts of the case.

  385. Mr Morgan submits that although Mr Scrivenger did not know the precise mechanism of the First Assist premium calculation, he could not have done a more thorough estimating exercise if he had known more. He accepted that his estimate of overall costs had been more for settling the case rather than taking it to trial, but Mr Morgan submits that as long as he was estimating on the same basis for each side the ratio would remain constant and would not affect the calculation of the premium rate.

  386. With regard to Mr Wait's evidence, that his company might have offered a deferred premium policy, Mr Morgan points out that his evidence was given on the basis that he had not had time to consider the papers properly, having only had a brief look at them. He submits therefore that Mr Wait's evidence should be given little weight. Mr Morgan also points out a feature of the Temple policy, which, in his submission, means that it is not a reasonable alternative policy in the circumstances of the Test Cases. The Temple policy provides that in the event of the insured failing to beat a Part 36 offer she/he faces not just the usual set off of costs against costs before the insurer pays out, but a set off of the entirety of his/her damages. This, says Mr Morgan, renders the Temple policy virtually worthless in the case of a successful Part 36 offer. There is no comparable provision in the Pursuit Policy.

  387. Mr Morgan also submits that it is clear from the results of The Judge search that premium would be individually assessed and be very fact dependent. The search results included at least one scheme requiring panel membership.

    Anthony Baker

  388. This is the case which was turned down by the Claimant's union solicitors, and in respect of which an application to the CICA failed. In Mr Morgan's submission this was not a straightforward accident at work and the solicitor, Mr Vallance, was diligent in trying to find funding for the case, including trying to take advantage of any BTE insurance which the client might have. Mr Vallance himself applied to Amicus and Capita/Eastgate, companies to which he was known. These applications were rejected in June and July 2002 and limitation was due to expire in August. He knew there would be further difficulty because of those rejections. At the end of July he applied to First Assist who, he said: "appeared to be prepared to take on more risky cases". Mr Morgan submits that Mr Vallance had done all that he could to investigate alternatives and that the Claimant had either to take the Pursuit Policy or drop the claim.

  389. Having obtained an offer of insurance Mr Vallance put the Defendant on notice of this before the Claimant took out the policy. He was unsuccessful in his efforts to obtain a response from the Defendant.

  390. With regard to the estimates which Mr Vallance had given, he said that he had done his best to give an honest estimate and the estimates which he had later given in the allocation questionnaire were consistent with his estimate provided for insurance purposes. He did not fully understand the detail of the calculation of the Pursuit premium, but even had he done so the estimates would have been no different.

  391. In Mr Morgan's submission both the Claimant and Mr Vallance acted diligently and reasonably in the circumstances. Mr Vallance specifically drew the attention of the Defendant's insurers to the basis on which the premium was calculated, thereby warning them explicitly of the costs risk they faced.

    Clarke

  392. Mr Clarke was represented by Ms Stevens, whom Mr Morgan puts forward as a careful solicitor who has taken active steps to be as up to date as possible in funding matters. Because of the type of work she undertakes delegated authority schemes were not a possibility and she had previously relied on the brokerage services of The Judge, because she was impressed by Mr Gilbert. He had more knowledge of the market than she. She had to complete only one form to enable the broker to obtain quotes "from a variety of insurers" which would in normal circumstances result in competitive quotes and the difficulty of having to admit rejection to other insurers was bypassed. The Judge put the proposal to four insurers but only First Assist would offer cover. Mr Gilbert said in evidence:

    "They are difficult to insure, you have to take these out of the normal personal injury pool because there are reasons why these are excluded from delegated authority schemes. They are viewed universally by insurers as risky cases."

  393. Mr Morgan points out that Ms Stevens' view, having adopted a reasonable approach to investigating insurance, was that her client had no alternative but to take out the Pursuit Policy. Furthermore the deferred premium was attractive to her client, who was "strapped for cash".

  394. Ms Stevens did everything she could to settle the case, including drawing the attention of the Defendant's insurers to another very similar case in which settlement had been achieved. She made a reasonable Part 36 offer at an early stage which was rejected. She made careful estimates of the costs but accepted that her knowledge of the Defendant's costs was limited. She put a figure for Defendant's costs to First Assist for them to consider. It was her view that the Defendant's solicitors would certainly need to do as much work as she in the particular case.

  395. Mr Morgan points out that both Ms Stevens and Mr Gilbert in their evidence considered the alternative policies which were said to be available, none of which was in his submission suitable. There was no suggestion that the Temple policy could have been used.

    Sandiford

  396. Mr Morgan points out that Mr Gilbert in his evidence stated:

    "Stress at work - everything I have just said about manual handling cases you can apply a factor of 5 or 10 to stress at work cases. They are universally loathed by insurers. They are extremely difficult to arrange cover for."

  397. Mr Besfor was the Claimant's solicitor in this case, and although his firm was a member of Accident Line Protect, this route could not be pursued since the firm did not yet have in place a complementary funding scheme. Mr Besfor did accept that it might have been a good idea to obtain ATE insurance before proceedings were issued, but he could not say whether he would have had a wider choice of insurers had he done so. Mr Morgan relies on Mr Gilbert's evidence to submit that it was unlikely to have made any difference and that if it did it would have been marginal.

  398. In the event Mr Besfor approached a number of companies without success, including LPL. He also applied unsuccessfully for legal aid and used the broking service of The Judge. The only quote which he was able to obtain was from First Assist. It was in Mr Morgan's submission the only option if the client was to be able to pursue her case.

  399. Mr Besfor's estimates were such that the estimated costs at risk after trial to normal costs was understated, being 2:1 as opposed to the actual 2.96:1. In this case Mr Morgan suggests that the premium was too low, and that this served to support his argument that in the premium rating mechanism there would be a tendency for errors to cancel each other out.

    Farr

  400. This is a case in which unlimited cover was thought to be a necessity. Mr Wait's evidence was that Temple could have provided the Claimant with cover for adverse costs of ?120,000 for a premium of ?22,400 plus IPT. A smaller limit of indemnity could have been provided for a couple of thousand pounds less.

  401. Mr Morgan put to Mr Wait in cross examination that he was not a disinterested witness but had a general commercial interest in knocking out a competitor and boosting his own products. This he denied. Mr Morgan argues that Temple would not have been interested in considering an approach for a one off personal injury case given the advice which Temple publish on applying for ATE insurance:

    "We do not consider one off personal injury cases unless there are compelling reasons to do so. We want to encourage solicitors to commit to a delegated authority scheme and work with us as a partnership rather than submit individual PI cases where there is a risk that adverse selection is applying."

  402. Temple list reasons why the application for insurance is likely to be declined. These reasons include: if the case has already been rejected by any other insurer; and, if the incident date was more than two years ago. Mr Bennett gave evidence that his firm was not interested in being on ATE panels because of work which they already did on behalf of various local authorities.

  403. Mr Wait in giving evidence as to the policy which Temple could have provided, assessed the prospect of success as greater than 85%. Mr Morgan points out that this risk assessment was given in spite of the evidential difficulties and the knowledge that the case was being vigorously defended. The solicitor's own prospects of success on the CFA were 55.5% and in the case of counsel's CFA 50%.

  404. When Mr Bennett first accepted Mr Farr's case he did so with the benefit of Mr Farr's BTE policy. He reasonably utilised that policy until it was exhausted. Mr Morgan argues that had ATE insurance been sought at the outset it would either not have been available at all, or, given the evidential difficulties, only available at a very high premium.

  405. Mr Morgan submits that it was reasonable to seek ATE insurance from the BTE provider, i.e. RSA. The cover offered was unlimited and although the case was being hard fought First Assist did not require counsel's opinion as to merits.

  406. Mr Morgan argues that Mr Bennett's decision not to go to any other ATE providers, on the basis that he was firmly of the view that he would not get a premium elsewhere, was, in the particular circumstances of the case, reasonable and correct.

  407. With regard to the LPL policy, which was said to be available, Mr Bennett did not accept that this was "an RTA case". He pointed out that there were other issues involved in terms of intoxication and the client not remembering anything about the incident. So far as the Temple policy was concerned he stated he would have had concerns about the limit of ?120,000 for adverse costs. We now know that Mr Bennett had made mistakes in the costs estimates and that the Temple limit of indemnity might have been adequate. Mr Morgan however argues that because of the Part 36 provision, by which adverse costs are only paid after setting off the Claimant's inter partes costs and damages, this would have made the policy unacceptable.

  408. When the Pursuit Policy was offered to Mr Farr, Mr Bennett advised that it was "a very very good product" because of the unlimited costs cover, and the opportunity to use up the BTE. He stated that this was Mr Farr's best option "albeit in my view - at that stage - he had very very few options".

  409. Mr Morgan points out that Mr Bennett put the Defendant on express notice of the existence and nature of the policy.

    Defendants 1 - 4

  410. Mr Bartlett argues that in each of the Test Cases the choice of the First Assist Pursuit policy was an unreasonable one. In the case of Anthony Baker the premium is more than the estimated total costs at risk.

  411. He argues that just because a claimant solicitor did not happen to be a member of a delegated scheme it is not reasonable for the defendant to have to pay an extra premium as a result of this. In his submission the prices of delegated scheme premiums are relevant when considering whether a reasonable choice was made.

  412. He repeats his delay argument, arguing that if a more expensive policy was chosen as a result of delay the extra amount of the premium should not be recoverable against the Defendant.

  413. He suggests that the fact that a claimant does not have to pay the premiums in advance (or in a losing case at all) cannot be relied on to make the choice reasonable, particularly where the size of the premium was such that no reasonable claimant of moderate means would incur such a liability.

    Deborah Baker

  414. Mr Bartlett argued that at the time the Pursuit Policy was chosen Mr Scrivenger's costs had already begun to mount up, meaning that at the outset the premium would still be in excess of ?20,000, whilst the Defendant's lawyers had incurred no costs at all.

  415. With regard to the offer of cover from LPL and Mr Scrivenger's efforts to obtain funding to cover the premium Mr Bartlett argues that further effort should have been made to raise money from the Baker's own property and their business.

    Anthony Baker

  416. The original estimate for damages given by Mr Vallance was ?15,000, but by 3 September 2002 that estimate had been reduced to ?5,000. The policy incepted on 16 October 2002, the claim being settled on 25 October 2003 for ?1,250. Mr Bartlett points out that if the case had fought, and had Mr Vallance's estimate of costs been correct, the premium alone would have been more than the total damages the Claimant was likely to receive. The premium, charged at 74% of normal costs, was he suggests inordinately large as soon as the policy was entered into and continued to be so.

  417. He suggests the choices facing the Claimant were:
    1. to incur a premium which was disproportionate at the outset and could very well exceed the total damages he was likely to recover;

    2. if all the costs estimates were correct spend ?5,180 on a premium to protect himself against ?7,000 worth of estimated costs at risk;

    3. decide to litigate without such a premium and thereby avoid incurring a potential liability of ?5,180;

    4. decide not to litigate since the cost of doing so appeared disproportionate when compared with the likely benefit.

    Clarke

  418. Mr Bartlett argues that since the estimated costs at risk total ?41,000 there was no need for a policy with unlimited cover. In addition, in completing the proposal form Ms Stevens answered "maybe" to the question about whether her client required funding in respect of the premium. Relying on his earlier arguments about delay, Mr Bartlett submits that the Claimant has not established, as between Claimant and Defendant, that the Pursuit Premium was reasonably incurred.

    Sandiford

  419. Mr Bartlett argues that there was no need for unlimited liability. There was no express request for a deferred premium and the client had funded the case privately for 1? years. Furthermore her solicitors took responsibility for payment of the premium on her behalf. Mr Bartlett repeats his delay argument which I have already rejected. He also cites two further policies (Collegiate Underwriting and Litigation Protection) mentioned by Mr Gilbert. There is however no indication of what the premiums would have been in either case or whether those policies would in fact have been available.

    Defendant 5 (Farr)

  420. Mr Bacon points out that Mr Bennett provided an estimate to the Defendant and to the court of the likely quantum of the premium which he put at ?24,610. The estimate was made on 1 August 2003 eight weeks before the listed trial. He suggests that Mr Bennett clearly had no idea what the premium figure was going to be and that it was unreasonable for him to recommend the policy to Mr Farr, even at the revised premium of ?115,182.

  421. With regard to the availability of alternative insurance in the case of Farr, Mr Bacon relies on the evidence of Mr Wait, to the effect that he would have offered a policy for a premium in the region of ?22,000 with a limit of indemnity of ?100,000. Mr Bacon suggests that, because Mr Bennett's firm had relationships with a number of local authorities, it put Mr Bennett in conflict with the best interests of his client. Mr Bennett confirmed that it was the first case in which he had acted for a claimant and had been required to obtain ATE insurance. The relationship with the local authorities affected the advice which he gave to Mr Farr about alternative products. Mr Bacon points out that Mr Bennett made no attempt to investigate alternative products to the Pursuit Policy.

  422. Mr Bacon argues that a premium of ?115,182 compared to total adverse costs (less VAT) of ?78,600 demonstrates that the choice of the Pursuit Policy was an unreasonable one. Even when Mr Bennett realised that the BTE cover was going to be insufficient, he did not undertake any search for alternative ATE cover, preferring to rely on RSA who knew the facts of the case.

    ISSUE 6 - CONCLUSIONS

  423. It is a feature of these Test Cases that the arguments in the issues tend to overlap. I have already dealt with the issue of delay and with the premium methodology. I approach this issue "has the claimant acted reasonably in taking out the RSA Policy?" in its simplest form: i.e., on the basis that the recoverable premium will be reasonable and proportionate.

    Deborah Baker

  424. Mr Bartlett suggested that Mr Ivory's figure on Mr Scrivenger's costs should be preferred to Mr Scrivenger's own evidence, on the basis that Mr Ivory had had a better opportunity to check the figures and his job was to scrutinise and understand bills of costs. Mr Scrivenger is a shrewd litigator who has great experience of defence clinical negligence work, and, by the time he gave evidence, also of claimant work. To suggest that he would not have an intimate working knowledge of his own costs, or not understand his own bills of costs is not realistic.

  425. The argument that when the Pursuit Policy was taken out the Claimant's solicitor had already incurred significant costs, whereas the Defendants had not, overlooked the fact that Mrs Baker and her husband had been trying for the previous two years to settle their case without recourse to litigation. Had the Defendant's solicitors been instructed at an earlier stage, and a proper investigation of the circumstances been carried out, this claim might well have settled without the need for proceedings or ATE premium. To argue that the choice of the policy is unreasonable, based on the Defendant's failure to instruct their own solicitors, is insupportable.

  426. Mr Scrivenger gave clear evidence about his efforts to obtain funding from First National Bank and also clear evidence about the lack of funds in Mr and Mrs Baker's business and their family home. Whilst it might have been possible to obtain adequate cover from LPL, had it been possible to raise funding for the premium, that was not in fact a possibility and the Claimant acted reasonably in accepting the offer of the First Assist Policy.

  427. Following a suggestion from Mr Gilbert, Mr Fallon of First Assist suggested a flat rate, up front, premium of ?17,550 plus IPT. Leaving aside the funding difficulties, no formal proposal was never put to Mr Gilbert or Mr Scrivenger, and accordingly the Claimant cannot be criticised for not accepting it.

  428. With regard to the offer of a Temple Policy, Mr Wait thought it possible that cover would have been offered, but he had not had a good look at the papers and, as I have stated elsewhere, in my view, his assessment of the risks was far too optimistic. In any event the premium of the Temple Policy would have been ?20,000 on issue of proceedings and about ?50,000, 45 days before trial. I regard it as mere speculation whether a Temple Policy would in fact have been available. In my judgment the Claimant's acceptance of the Pursuit Policy was reasonable.

  429. Mr Bartlett takes issue with Mr Morgan's assertion that the set off provisions in the Temple policy render it virtually worthless in the case of successful Part 36 offer. He submits that in a case where the damages are sufficient to cover the costs there would be no difference between the Pursuit and Temple policies because there is no cover under the Pursuit policy, and the Temple cover is in effect eaten up by the set off or prevented by the set off. In a case where the costs exceed the damages, however, the insured would, he submits, be worse off under the Pursuit policy than the Temple policy in the event of not beating a Part 36 payment, because under the Temple policy he will have cover, and under the Pursuit policy he will not, because, by definition, the cover does not apply because he has won some damages. Mr Dutton points out that this point does not arise in any of the Test Cases. Even in the case of Farr, where a Part 36 offer was made, it was rejected on advice and therefore there was no success at that stage. I accept Mr Dutton's submission that I should not speculate on what might happen in a case where a Part 36 offer was rejected on advice and then not beaten, such a situation would be fact sensitive and the outcome would depend on the evidence. In any event it seems that the policy has been subject to re-writing since December 2002 and there are now different provisions in certain clinical negligence policies.

    Anthony Baker

  430. As to the case of Anthony Baker there are grounds for stating that the Claimant has not acted reasonably. Once it had been identified at the conference with counsel, that the maximum potential damages were ?5,000, immediate steps should have been taken to revise the RSA proposal or to apply, through a broker, for appropriate ATE cover. The position with regard to damages had been identified at the conference on 3 September 2002 which would have given adequate time for an appropriate policy to have been identified.

  431. I am in no doubt that this premium (and the potential premium when the policy was taken out) were disproportionate given the maximum level of damages obtainable. Mr Gilbert did identify a number of companies which might have provided a quotation. I find therefore that it was unnecessary and unreasonable for Anthony Baker to select the Pursuit Policy.

    Clarke

  432. For the reasons which I have given under the delay point, I am satisfied that this Claimant acted reasonably in taking out the Pursuit Policy. Ms Stevens acted reasonably in approaching Mr Gilbert at The Judge for quotations. Although he indicated the possibility of alternative policies being available, the fact is, when approached, the only quote which he was able to obtain was that from First Assist. It is not reasonable to expect Ms Stevens, having approached an acknowledged expert in the field, to carry out further enquiries to ensure that every possible option had been investigated.

  433. In the case of Clarke Mr Gilbert was of the view that it was impossible to say whether any particular insurer definitely would or would not have issued a policy. There is no criticism of Ms Stevens' handling of the case, and, in the circumstances facing her when the ATE insurance was taken out, the choice was reasonable.

    Sandiford

  434. Mr Besfor's estimates of costs were inaccurate resulting, in Mr Morgan's submission, in a premium that was too low. Given the difficulties faced by Mr Besfor, the Claimant was given only one option for ATE insurance and she acted reasonably in accepting it.

    Farr

  435. Although First Assist queried whether Mr Bennett's estimate was accurate, Mr Bennett did not deal with the point when he responded.

  436. With regard to Mr Wait's assessment of the prospects of success in the case of Farr as being in excess of 85% this is clearly far too optimistic given the evidential difficulties and the legal representatives own risk assessments of 55.5% and 50% respectively. I am forced to the conclusion that Mr Wait reached his risk assessment with the benefit of hindsight secure in the knowledge that Mr Farr had been successful in his claim.

  437. On the basis of the approach I have taken to this issue (i.e., assuming that the Pursuit Policy premium was reasonable) I am satisfied that the Claimant acted reasonably in selecting the Pursuit Policy. In the case of Farr however, the inherent flaws in the premium calculation were exacerbated by Mr Bennett's failure to provide any figures for Defendant's costs and getting his own estimated figures wrong.

    ISSUE 7 - What if anything is the recoverable amount of the premium against the defendant pursuant to Section 29 of the Access to Justice Act 1999?

    First Assist

  438. Under this heading Mr Dutton deals with Mr Bartlett's point as to the irrecoverability of part of the premium under Section 29 because it relates to the fact that no premium is recovered in losing cases. He suggests that Callery v Gray (No.1), at paragraphs 93 to 95, makes clear that there is nothing impermissible in principle about transferring the burden of unsuccessful claimants' costs to defendants. There is, he says, no conceptual difference between the cost of meeting an adverse costs order in losing cases, and any extra costs involved thereby, by not having premium from such losing cases to off-set this cost. This is, he says, simply a component in the calculation of what it will ultimately cost underwriters. He says the court must distinguish premium (which is the payment to the insurer to take on the insured's risk) and the actual cost of paying out (which is the insurer's). If the insured's risk materialises the absence of a premium on a losing case affects the latter not the former.

  439. Mr Dutton points out that no part of the premium calculation is specifically designed to cover the risk of not receiving premium in losing cases. There is, he says, no extraneous benefit under the Pursuit Policy for which the premium is due or payable. The insured on a losing case is not being covered for the risk of having to pay the premium. There is no such risk. Finally, he says, there is no difference in substance between the First Assist method of ensuring the Claimants do not have to pay the premium in losing cases and the self insuring policies. He points to the evidence of Mr Wait in relation to self insuring premium which he said "gets lost in the ether".

    The Claimants

  440. Mr Morgan for Mr Farr submits that the premium should be ?109,176.72 plus IPT, that is a premium rate of 193% applied to normal costs of ?56,568.25. The normal costs figure is calculated by taking the total profit costs in the bill, ?95,463.75, and deducting the success fees in Parts 4 and 5 and the cost of the work on the bill. Mr Morgan submits that these items should be left out of account bearing in mind the policy wording. The premium is payable "when the outcome of the legal proceedings is a success". He says success is achieved when a settlement is made. It follows in his submission that calculation of the premium must take place at the point when it becomes payable, i.e. normal costs to the point of settlement and the costs of the bill which are incurred after that time must be left out of account.

    Defendants 1 - 4

  441. Mr Bartlett repeats his contention that the part of the premium paid or payable in respect of the fact that other cases are lost by other claimants, without payment by anyone of premium in those cases, is not insurance against the risk of incurring a costs liability in these proceedings and cannot be recovered under Section 29 of the 1999 Act. He further argues that any element of the premium which is attributable to the fact that the premium is deferred until the end of the case is, for the same reasons, irrecoverable. This is in accordance with the principle that the cost of funding litigation is not recoverable. Similarly, costs incurred by delayed receipt of the premium are not recoverable. He argues that the proper and recoverable elements of the premium are those identified by Master O'Hare and adopted by the Court of Appeal in Callery v Gray (No.2) paragraphs 21 to 26, namely the burning cost, risk/profit costs, administrative cost and distribution commission.

  442. Mr Bartlett points out that, in four of the five Test Cases, the premium claimed is substantially higher than the actual costs at risk:
    1. In Deborah Baker the actual costs at risk were ?24,708 (estimated ?58,500) and the premium was ?54,651 (?52,048.98 plus IPT - 172% of normal costs)

    2. In Anthony Baker the actual costs at risk were ?5,752 (estimated ?7,000) and the premium was ?8,963 (?8,536 plus IPT - 74% of normal costs)

    3. In Clarke the actual costs at risk were ?15,267 (estimated ?35,000) and the premium was ?32,311 (?30,772.76 plus IPT - 123% of normal costs)

    4. In Sandiford the actual costs at risk were ?29,885 (estimated ?50,000) and the premium was ?16,986 (?16,176.76 plus IPT - 160% of normal costs)

    5. In Farr the actual costs at risk were ?80,233 (no estimated figures were given. First Assist used a figure of ?112,000) and the premium (adjusted) was ?115,182 (?109,697 plus IPT - 193% of normal costs).
  443. Mr Bartlett points out that in Deborah Baker and Clarke the claim for premium is very nearly the same as the total estimated costs at risk to trial, even though both settled before trial. In Anthony Baker the premium is more than the total estimated costs at risk, in a case which settled after exchange of medical evidence and before trial. The case of Farr settled on the first day of trial. In short Mr Bartlett argues that because of the way the premium is calculated the premium payable is grossly excessive and does not fairly reflect the risk undertaken by the insurer.

  444. Mr Bartlett relies on Lownds v The Home Office [2002] EWCA Civ 365; [2002] 1 WLR 2450 and Giambrone v JMC Holidays [2003] 1 All ER 982. He relies specifically on paragraph 39 of the Lownds judgment:

    "Whether the costs incurred are proportionate should be decided by having regard to what it was reasonable for the party in question to believe might be recovered. Thus (i) the proportionality of the costs incurred by the Claimant should be determined having regard to the sum that it was reasonable for him to believe that he might recover at the time he made the claim …"

  445. He argues that even where a claimant establishes that the Pursuit Policy was the only policy available, it does not follow that the Pursuit premium should be allowed in full. The limits on recoverability - a proportionate and reasonable amount - remain. He suggests that the evidence of other premiums generally, even if not available in the particular case, is helpful in arriving at an appropriate amount.

    Defendant 5 (Farr)

  446. Mr Bacon argues that the premium payable should be calculated in order to exclude costs associated with there being no premium income in unsuccessful cases. In the case of Farr, with a 25% risk of having to pay out, an estimated exposure of ?84,000, he submits that the true burning cost is ?21,000 plus commission, making a total premium of ?24,150 plus IPT. He suggests three possible methods of arriving at a reasonable premium in Farr:
    1. using a simple flat rate formula producing a premium of ?14,303.12 plus IPT;

    2. adopting the premium which Mr Wait said would be payable for a Temple Policy, ?22,000 - ?23,000 including IPT;

    3. using the figures which ought to have been inserted in December 2002 when the premium was being rated, which produce a figure of ?18,507 plus IPT.

    ISSUE 7 - CONCLUSIONS

  447. I have already dealt with Mr Bartlett's submissions in relation to no premium being payable in unsuccessful cases at paragraph 348 above.

  448. The figures set out by Mr Bartlett comparing the actual costs at risk with the premium claimed, amply demonstrate the effect of the flawed premium rate methodology. In this judgment I allow amounts for premiums which appear to me reasonable and proportionate in all the circumstances of each particular Test Case.

  449. Mr Bartlett dealt with the suggestion by Mr Dutton and Mr Morgan that the Defendants had, by reason of their conduct, increased the size of the premiums. I have already commented on the Defendant's delay in the case of Deborah Baker and there is evidence of deliberate delay by the Defendants in other cases. Nonetheless I accept Mr Bartlett's submission that the fact that a particular defendant may have acted unreasonably in not settling the case earlier cannot justify charging unreasonable and/or disproportionate premiums.

  450. Mr Bartlett submits that, on the basis of his arguments, the following figures could be used to arrive at a premium that is reasonable and proportionate in each case.
    1. Deborah Baker. He refers to the two policies from LPL at ?9,922 and ?13,230, as well as the flat rate First Assist premium of ?17,550 and the Temple Policy which he put at ?20,000. He suggests a figure of ?15,000 and no more than ?20,000.

    2. Anthony Baker. He suggests no more than ?750.

    3. Clarke. Based on items appearing in The Judge and Litigation Funding he suggests a premium between ?220 and ?3,413. He suggests this compares favourably with the case of Sandiford where, if the Accident Line policy had been utilised, the premium would have been ?3,045.

    4. Sandiford. He argues that the Accident Line premium provides the most appropriate figure at ?3,045. Alternatively, on the basis of the range of premiums apparently available for stress at work claims, the premium should not exceed ?7,323.
  451. Of the three options for calculating a reasonable premium in the case of Farr put forward by Mr Bacon, I prefer the third: utilising the actual exposure ratio which should have been used in the first place.

  452. The premiums in respect of Anthony Baker, Clarke and Farr have the appearance of disproportionality. For the reasons I have already given the premium in the case of Anthony Baker is disproportionate and fails the test of necessity. In the cases of Clarke and Farr, given that I propose to allow reasonable and proportionate premiums, the disproportionality point disappears. The premiums which I allow in respect of the remaining Test Cases are those which I consider to be reasonable and proportionate.

  453. In his witness statement Mr Smith explained:

    "We felt that it was a reasonable assumption … to make in broad terms, for the purpose of establishing and maintaining a consistent underwriting procedure that the level of the insured's costs and the opponent's costs would escalate for the duration of the case on a roughly equal basis … We thus arrived at the basic fundamentals of a sliding scale premium calculated by reference to own solicitors fees. The only workable assumption is that, as costs for both sides start at zero and proceed to the estimated levels at trial, the relationship between own costs and the insured's liability will remain constant throughout the action."

  454. Ms O'Reilly confirmed that:

    "If the assumption of a constant ratio between adverse costs and own costs is correct then … [if the estimates are reasonable the insurer] has a reasonable basis for calculating the premium."

  455. She added:

    "It is the ratio between the adverse costs and the own costs that is important rather than the absolute amount of the estimate."

  456. Because of the inaccuracy of the estimates of costs put forward by the solicitors the exposure multiplier in each of the Test Cases, except Sandiford, was far higher than the actual exposure multiplier ascertained at the end of the case. Given that the theory upon which the premium methodology was based was that the ratio would remain constant throughout, it is, in my judgment, appropriate to use the actual ratio to calculate the exposure multiplier in arriving at a figure for recoverable premium.

    Deborah Baker

  457. The estimate of normal costs given by Mr Scrivenger was, on his own evidence, intended to reflect a settlement before trial. It is not clear whether his estimate of own disbursements and Defendant's costs were intended to reflect the matter going to trial. In the event the estimates given did not reflect the actual outcome, save to the extent that Mr Scrivenger's estimate of his own costs to settlement was almost exactly the actual figure for Defendant's costs. His own actual costs (?30,261) were greatly in excess of his estimate, whilst his disbursements were less than half his estimate.

  458. Utilising the First Assist formula, and the figures given by Mr Scrivenger, an exposure multiplier of 3.34 is produced. The premium rate calculated by First Assist is 172%. The actual exposure multiplier was 0.82, a premium calculation based on that ratio produces a figure of ?13,042.49 plus IPT (?13,695).

  459. [24,708 ¸ 30,261 = .82 x 30 ¸ 70 (0.43) = 35.26% + 10% = 38.79% + 10% brokerage = 43.10%, 30,261 x 43.10% = ?13,042.49 + IPT = ?13,695]

    Anthony Baker

  460. Mr Vallance's estimates were reasonably accurate save that he significantly under estimated his own costs. The prospect of success at 60% appears reasonable given the failure of the claim to CICA and the union solicitors' refusal to take the case any further. Had the claim been for a significant amount of damages the premium might have been justified. In fact this claim was never anything other than a fast track claim and before the policy proposal was accepted counsel had advised that the maximum damages would be ?5,000. Even if Mr Vallance's estimate of his normal costs had been accurate, this would still have produced a premium in excess of the potential damages. On any reading the premium was clearly going to be disproportionate and the use of the Pursuit Policy not justified. Mr Bartlett says, given the finding that ?798 premium was disproportionate in the case of Samonini (a ?2,000 case) "no more than ?750 should be allowed, if a premium is allowed at all". I allow ?750 plus IPT.

    Clarke

  461. This case settled for ?20,000 after exchange of expert evidence on 10 July 2003, some 11 months after the policy incepted. Settlement took place after trial bundles had been prepared. Although Ms Stevens' estimates were carefully given in respect of her own disbursements and the Defendant's costs, they were very inaccurate, as shown by the ratio of adverse costs to own costs both estimated and actual (own disbursements estimated ?7,000, actual ?1,905; Defendant's costs estimated ?35,000, actual ?13,362). First Assist applied a slightly higher percentage for prospects of success. Ms Stevens explained the potential difficulties and why she had arrived at her figure of 60%. There is no statistical evidence to show First Assist's claims experience in this type of case.

  462. The ratio of estimated adverse costs to estimated own normal costs (the exposure multiplier) is 1.66. The actual is 0.64. Using that figure to calculate the premium produces a premium of ?11,110.26 plus IPT (?11,666).

  463. [15,267 ¸ 24,838.50 = 0.61 x 37.5 ¸ 62.5 (0.6) = 36.6% + 10% = 40.26% + 10% brokerage = 44.73% ?24,838.50 x 44.73% = ?11,110.26 + IPT = ?11,666]

    Sandiford

  464. In this case the costs incurred by the Defendants were double those of the Claimant. The costs estimates were inaccurate in that the exposure multiplier based on the estimates was 2. The actual was 2.96. A calculation based on that ratio would produce a figure for premium higher than that claimed. The actual premium claimed, ?16,176.75 plus IPT (?16,986), is reasonable and I allow it.

    Farr

  465. The position in Farr was made more difficult by the fact that Mr Bennett did not supply any figures for the estimated Defendant's costs. On the basis of subsequent concessions, and calculating the premium on the basis of estimated Defendant's costs of ?84,000, the exposure ratio is 3.9 producing, on the basis of a 70% prospect of success, a premium rating of 193%, giving a premium of ?109,176.72 plus IPT (?114,636). Mr Morgan argues that the correct figure for actual Claimant's costs should be ?56,568.25. Using Mr Morgan's figure the actual exposure ratio is 1.42. This produces a premium of ?39,722.22 plus IPT (?41,708).

  466. [80,233.18 ¸ 56,568.25 = 1.42 x 30 ¸ 70 (0.43) = 61.06% + 15% = 70.22% = ?39,722.22 plus IPT = ?41,708.]

  467. On that basis I allow the following amounts as representing reasonable and proportionate amounts in respect of ATE premiums recoverable from the Defendants.
    1. Deborah Baker - ?13,042.49 plus IPT = ?13,695.

    2. Anthony Baker - ?750 plus IPT = ?787.50.

    3. Clarke - ?11,110.26 plus IPT = ?11,666.

    4. Sandiford - ?16,176.75 plus IPT = ?16,986.

    5. Farr - ?39,722.22 plus IPT = ?41,708.

    ISSUE 8 - Has the indemnity principle been breached in the case of Farr?

    First Assist

  468. It is Mr Dutton's position that legally binding contracts of insurance have been entered into in each of the five Test Cases under which the individual insured is liable to pay the premium to the insurer. To the extent that Mr Bacon argues that the premium is not premium because it is paid only by successful parties, Mr Dutton says the point is bad. The premium is the consideration which the insured agrees to pay to the insurer and which is clearly identified as such in the policy. He points out that Temple charge a premium which is payable only in the event of success, as do Cornhill. In each case it is a valid premium.

    Test Case Claimants

  469. This is an issue between the Defendant in Farr and First Assist. There is no evidence that the Claimants, as individuals, have fraudulently conspired with others to dress up a non existent claim or liability for a premium.

    Defendant 5 (Farr)

  470. In this issue Mr Bacon seeks to argue that the sum charged by First Assist is not a Section 29 premium at all, and therefore not recoverable. He suggests that the premium payable is not an insurance premium in the ordinary sense, but a sum payable when it is certain that no insurance cover is required. He argues that the consideration moving from the insured is a conditional promise to pay a sum of money "if it turns out that no cover is required". This is not the sort of premium envisaged by Section 29. In his closing submissions Mr Bacon quotes from MacGillivray on Insurance Law at 1.1:

    "A contract of insurance is one whereby one party (the "insurer") promises, in return for a money consideration (the "premium"), to pay the other party (the "assured") a sum of money or provide him with some corresponding benefit, upon the occurrence of one or more specified events."

  471. He describes this quotation as a "definition of premium" and suggests that the promise of the insurer to pay a sum of money (the adverse costs) is dependent upon money consideration passing from the insured to the insurer. He suggests that the specified event under the Pursuit Policy is the claim being unsuccessful. In that event however no money consideration passes from the insured to the insurer.

    ISSUE 8 - CONCLUSIONS

  472. Mr Bacon has relied on MacGillivray's definition of a "Contract of Insurance" as if it were a definition of "premium". It is not. MacGillivray does define premium as I have set out earlier in this judgment:

    "The consideration required of the insured in return for which the insurer undertakes his obligation under the contract of insurance."

  473. Under the Pursuit contract of insurance the insured is required to pay a premium in return for which the insurer will pay the adverse costs and own disbursements in the event of the claim being unsuccessful. The fact that the insured is relieved from the obligation to pay the premium if the claim is unsuccessful does not thereby mean that the premium ceases to be premium under Section 29 of the 1999 Act.

  474. I accept Mr Dutton's submissions. There is no breach of the indemnity principle.

    ISSUE 9 - What is the impact of BTE cover in Farr?

  475. This issue is no longer pursued by Mr Bacon.

    SUMMARY

  476. In summary therefore the answers to the issues are as follows.

  477. Issue 1 - The contract of insurance is not void for uncertainty. Mr Dutton's submissions as to certainty are correct.

  478. Issue 2 - The insurance arrangement between the client and/or the solicitors and the insurer is not unlawful on the grounds of champerty.

  479. Issue 3 - The premium rating methodology is inherently and seriouslyfundamentally flawed, in that it assumes a constant relationship between costs at risk and own costs and depends too heavily on relative sizes of the estimate of own costs and Defendants costs which are very likely to be inaccurate. The estimates of success adopted by First Assist may be too low in that they do not accurately reflect the likelihood of the insurers having to pay out, and the final premium is calculated by reference to the Claimants costs as claimed however unreasonable or disproportionate they may be. In consequence the recoverable premium in each Test Case will be such sum as is reasonable and proportionate to expect the paying party to pay having regard to all the circumstances of each case and the factors set out in CPR 44.5.

  480. Issue 4 - The issue of commissions was not pursued.

  481. Issue 5 - In the cases of Sandiford, Clarke and Farr the individual Claimants acted reasonably in taking out insurance when they did. This issue does not arise in the cases of Deborah Baker and Anthony Baker.

  482. Issue 6 - Each Claimant except Anthony Baker acted reasonably in taking out the RSA Pursuit Policy.

  483. Issue 7 - I allow the following amounts as representing reasonable and proportionate amounts in respect of ATE premiums recoverable from the Defendants:
    1. Deborah Baker - ?13,042.49 plus IPT = ?13,695.

    2. Anthony Baker - ?750 plus IPT = ?787.50.

    3. Clarke - ?11,110.26 plus IPT = ?11,666.

    4. Sandiford - ?16,176.75 plus IPT = ?16,986.

    5. Farr - ?39,722.22 plus IPT = ?41,708.
  484. Issue 8 - There is no breach of the indemnity principle.

  485. Issue 9 - The issue of pre-existing legal expenses insurance was not pursued.

  486. I will hear submissions as to the appropriate consequential orders to be made in the light of these decisions.


BAILII:
Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/ew/cases/EWHC/Costs/2005/90003.html