Leigh v Michelin Tyre Plc

One of the principal objects of the Woolf reforms was the control of costs. The Civil Procedure Rules include a number of innovations which were designed to enable the court to limit recoverable costs and thereby further the overriding objective defined in CPR r. 1.1. One of the innovations was the requirement that parties provide cost estimates at important stages of litigation (notably the allocation and listing questionnaire stages), and the conferring on the court of the power to take the parties’ estimates into account on an assessment of costs. At the heart of this appeal lies CPR 43 PD para 6.6 which provides:

“On an assessment of the costs of a party the court may have regard to any estimate previously filed by that party, or by any other party in the same proceedings. Such an estimate may be taken into account as a factor among others, when assessing the reasonableness of any costs claimed.”

In the present case, the claimant’s solicitors filed an allocation questionnaire in which they said that they estimated the claimant’s solicitors’ profit costs to date at £3000 plus VAT, and their overall profit costs as likely to be £6000 plus VAT. The practice direction in force at that time (February 2000) did not state that the estimate should include disbursements in accordance with the definition of “costs” in CPR 43.2(1)(a). The practice direction has since been changed to spell out that estimates must include disbursements. The claimant’s solicitors never revised their estimates. In the event, the litigation was eventually settled, and they lodged a bill of costs in which they claimed £21,741.28. This comprised £14,482.80 in respect of profit costs, £4314.70 for disbursements and £2943.78 for VAT. The first of these figures included £11,744 for their profit costs in respect of the period after the allocation questionnaire stage. This should be compared with the estimate of £3000 given at the allocation questionnaire stage for future solicitors’ profit costs. The district judge assessed the claimant’s recoverable costs at £20,488.83 inclusive of VAT. He made no deduction to reflect the fact that the claimant’s solicitors had previously given what proved to be a wholly inadequate estimate for their future profit costs. The defendant says that the district judge erred in not reducing the assessed costs to reflect the earlier estimate. It failed to persuade Judge Mitchell that the district judge was wrong. Permission to appeal was given because it seemed that the appeal raised an important point of principle as to the relevance of costs estimates in the assessment of costs.

At the date of the allocation stage in the present case, the practice direction supplementing CPR 26 provided that the allocation questionnaire should be in Form N150, which required estimates to be given of costs incurred by legal representatives to date and of the overall costs. CPR 26 PD para 2.1 provided: “attention is drawn to the Costs Practice Direction, para 4.5(1) which requires an estimate of costs to be filed and served when an allocation questionnaire is filed”. CPR 43 PD section 4 dealt with estimates of costs. Para 4.1 provided:

“This section sets out certain steps which parties must take in order to keep the parties informed about their potential liability in respect of costs and in order to assist the court to decide what, if any, order to make about costs and about case management”.

By the time the listing stage had been reached, some significant changes had been made to the practice direction. In particular, section 4 of CPR 43 PD had been replaced by section 6. The new para 6.1 was in almost the same terms as its predecessor para 4.1. An “estimate of costs” was now given a wider definition: it was an estimate of “base costs (including disbursements)”. “Base costs” were defined by para 2.2 of the practice direction as costs other than the amount of any additional liability as defined by CPR r. 43.2. Paras 6.3 to 6. 6 of the amended practice direction provided:

“6.3

The court may at any stage in a case order any party to file an estimate of costs and to serve copies of the estimate on all other parties. The court may direct that the estimate be prepared in such a way as to demonstrate the likely effects of giving or not giving a particular case management direction which the court is considering, for example a direction for a split trial or for the trial of a preliminary issue. The court may specify a time limit for filing and serving the estimate. However, if no time limit is specified the estimate should be filed and served within 28 days of the date of the order.

6.4(1) When a party to a claim which is outside the financial scope of the small claims track, files an allocation questionnaire, he must also file an estimate of base costs and serve a copy of it on every other party, unless the court otherwise directs. The legal representative must in addition serve an estimate upon the party he represents.

(2)

Where a party to a claim which is being dealt with on the fast track or the multi track, or under Part 8, files a listing questionnaire, he must also file an estimate of base costs and serve a copy of it on every other party, unless the court otherwise directs. Where a party is represented, the legal representative must in addition serve an estimate on the party he represents.

(3)

This paragraph does not apply to litigants in person.

6.5

An estimate of base costs should be substantially in the form illustrated in Precedent H in the Schedule of Costs Precedents annexed to the Practice Direction.

6.6

On an assessment of the costs of a party the court may have regard to any estimate previously filed by that party, or by any other party in the same proceedings. Such an estimate may be taken into account as a factor among others, when assessing the reasonableness of any costs claimed.”

The other important new elements were (a) the requirement that legal representatives should serve on their clients their estimates of costs; and (b) the provision in para 6.6 that cost estimates may be taken into account on an assessment of costs when assessing the reasonableness of any costs claimed. Section 6 of the PD has not been further amended, and governs the position today.

The only rule relating to the assessment of costs to which we need refer is CPR r. 44.5 which provides for the factors that are to be taken into account in deciding the amount of costs:

“(1)

The court is to have regard to all the circumstances in deciding whether costs were –

(a)

if it is assessing costs on the standard basis –

(i)

proportionately and reasonably incurred; or

(ii)

were proportionate and reasonable in amount, or

(b)

if it is assessing costs on the indemnity basis –

(iii)

unreasonably incurred; or

(iv)

unreasonable in amount.

(2)

In particular the court must give effect to any orders which have already been made.

(3)

The court must also have regard to –

(a)

the conduct of all the parties, including in particular –

(i)

conduct before, as well as during, the proceedings; and

(ii)

the efforts made, if any, before and during the proceedings in order to try to resolve the dispute;

(b)

the amount or value of any money or property involved;

(c)

the importance of the matter to all the parties;

(d)

the particular complexity of the matter or the difficulty or novelty of the questions raised;

(e)

the skill, effort, specialised knowledge and responsibility involved;

(f)

the time spent on the case; and

(g)

the place where and the circumstances in which work or any part of it was done.

(Rule 35.4(4) gives the court power to limit the amount that a party may recover with regard to the fees and expenses of an expert.)”

The notes in the current edition of the White Book state at para 44.7.2:

“On completing the allocation questionnaire and the listing questionnaire the party must set out an estimate of costs incurred to date and an estimate of likely future costs, Section 6 of the Costs Practice Direction deals with this. Considerable care and precision is required in the preparation of such estimates since the estimates of opposing parties are likely to be compared one with another. An over generous estimate may result in an opponent recovering a similar amount, while an under-generous estimate may result in a recovery on behalf of the client which does not reflect the actual costs involved.”

The claimant, who at the material times was employed by the defendant, suffered injuries at work in May 1996 and July 1997. On 27 May 1999, he issued proceedings alleging that the injuries were caused by the defendant’s negligence and/or breach of statutory duty. The claim form and particulars of claim were served on 2 September 1999. By its defence served on 2 February 2000, the defendant denied liability, causation and loss. Both parties completed and filed allocation questionnaires, and on 24 February, the claim was allocated to the multi-track. Apart from the costs estimates to which we have already referred, the claimant’s allocation questionnaire included the statement that he intended to use one expert witness, a consultant orthopaedic surgeon. The defendant said that it intended to use two experts, an engineer and an orthopaedic surgeon. It also stated that its estimate of costs incurred to date was £2500, and that it estimated the likely overall costs at £7500.

There was a listing hearing on 6 April 2001. Neither party filed or served on the other side an updated estimate of costs as required by CPR 43 PD para 6.4(2). At the hearing, the claimant was given permission to rely on the report of an employment expert. This was because the defendant’s witness statements asserted that the claimant would have been made redundant and would therefore have lost his employment even if he had not been injured. The trial date was set for 23 July 2001. On 16 July, the claim was settled on terms that the defendant pay £48,000 less CRU benefits plus costs.

As we have said, the claimant’s bill of costs was lodged in the sum of £21,891.28. The defendant submitted Points of Dispute on 6 December 2001. It raised a number of points of detail, but the principal point of dispute was the defendant’s contention that the claimant should not be permitted to receive more than the amount of the costs estimate contained in the allocation questionnaire plus 15%, ie £6900 in total. The argument was put in two ways. First it was said that, since the claimant had not approved an increase in the amount of the estimate given by his solicitor in the allocation questionnaire, the claimant’s solicitor would not be able to recover from him more than that estimate plus 15% (see Wong v Vizards [1997] 2 Costs LR 46), and on an application of the indemnity principle, the defendant was not liable to pay the claimant more than the claimant was liable to pay his solicitor. Secondly, it was said that, regardless of the indemnity principle, the defendant should not be liable to pay more than the amount of the costs that had been estimated, because it was entitled to rely on the cost estimates that had been given by the claimant’s solicitors.

The claimant’s solicitors responded as follows. They explained why the costs were so much higher than had been estimated at the allocation stage. Late disclosure of documents by the defendant’s solicitors had obliged them to carry out unexpected investigations and further work. The redundancy issue had not been foreseen: this too needed to be considered. Further, it had not been foreseen that engineering expert evidence would be required: in the event, an engineering expert had to be instructed. The claimant’s solicitors also explained (in answer to the indemnity principle point) that the prosecution of the claimant’s claim was funded by the TGWU. The Union gives the solicitors general authority to proceed with any claim which has a reasonable prospect of success, and does not require them to provide an estimate of costs.

District Judge Chapman held the assessment on 9 May 2002. At this hearing, he dealt with the points of detail and, subject to the point of principle raised by the defendant in relation to the claimant’s solicitors’ costs estimate, he assessed the claimant’s reasonable and proportionate costs at £20,488.83. At the same hearing, he heard argument on the point of principle, but he reserved judgment on this until 21 May. In his reserved judgment, he distinguished Wong v Vizards on the ground that in that case the client had expressly required the solicitor to give costs estimates. He held that the claimant’s solicitors had failed to comply with CPR 43 PD para 6.4(1) in that they had failed to serve on the claimant or his Union a copy of the costs estimate which they had given in the allocation questionnaire. He also noted that neither party had complied with the requirement of CPR 43 PD para 6.4(2) that there be filed with the listing questionnaire an estimate of costs at that stage. The district judge said that he “did not believe that the defendant’s solicitors or insurers (the paying party as it turns out) relied at any stage upon the claimant’s solicitors’ costs estimate in the AQ in deciding whether to continue to conduct the litigation as they did. Indeed it is not so suggested. The defendant’s decision to agree the compromise settlement figure was not influenced by the claimant’s solicitor’s costs estimates in the AQ”.

The nub of the district judge’s reasoning is contained in the following passage of his judgment:

“(vi)

The requirements to provide costs estimates from time to time are confusing. It has been common practice it seems for Solicitors to provide at AQ stage an estimate of profit costs only excluding vat and disbursements and further not to provide any estimate of costs when filing the LQ. The AQ form itself simply asks what the overall costs are likely to be. CPR43 PD6.2 defines an “estimate of costs” which should include disbursements (but there is no reference to vat). Precedent H provides for costs, vat and disbursements including estimates of trial costs. The precedent includes inter alia counsel’s brief and refresher fees, experts’ fees, expenses of witnesses of fact, attendances on all those people and perhaps others. In many cases, of which this is one, it will simply not be possible at the AQ stage which follows shortly after the defence and is before exchange of experts’ reports, disclosure of documents, exchange of witness statements of fact and many other possible developments in the proceedings, for either party to estimate with any degree of accuracy the costs and disbursements which are likely to be incurred if the claim proceeds to trial. Costs estimates are required by the Court primarily to assist in all aspects, including costs, of the proper management of the case and as a reminder to the parties of the potential costs of the litigation they are undertaking. The Court has the power, not exercised in this case, under CPR43 PD6.3 to direct costs estimates to be prepared at any time in the course of proceedings. I anticipate that a more clearly defined process will evolve as Courts pay more attention, as the present Practice Direction allows, to the costs estimates which are provided during the course of proceedings.

(vii)

However, whereas it is entirely appropriate that estimates of costs already incurred should be accurate, it would in my judgment be an unintended and unfair consequence of complying with PD6 to the extent of filing and serving on the other party (but not the Client) if the Solicitors estimating future costs (and disbursements) at such an early stage of the action were at the conclusion of the action to be tied to that estimate as against the other party unless there is clear evidence that the other party relied upon that estimate.

(viii)

In the absence of any evidence that the Defendant in this case relied upon the Claimant’s Solicitors estimate of further costs and so informed the Claimant’s Solicitors, I have concluded that the Claimant’s Solicitors should not be bound to that estimate and should be entitled to recover from the Defendant the costs which have been assessed at the hearing on 9 May 2002.”

The defendant appealed. Judge Mitchell gave judgment on 27 February 2003. He held that the Solicitors’ Practice Rules and the indemnity principle had no part to play here. So far as CPR 43 PD para 6.6 is concerned, the judge said this:

“16.

The contents of a number of the Practice Directions have served to engender satellite litigation of which the Costs Practice Direction has been a fruitful source. Furthermore, the stipulated procedures have of themselves added to the costs of proceedings. To produce a costs estimate with the detail in Form H requires an expenditure of chargeable time. The purpose and effect of requiring such estimates to be provided is not clearly stated. If the intended purpose is to limit the recoverable costs by reference to the estimate that is something which, in my judgment, requires a clear statement in the Rules themselves rather than something which is to be attempted by a Practice Direction, the effect of which would be to govern or fetter the exercise of the court’s discretion under the statutory rules in the manner which is implied by PD 43.6.6.

17.

I have recited the District Judge’s express finding that, at the allocation questionnaire stage, an estimate could not have been provided with any degree of accuracy. That finding will have been informed by this District Judge’s recent and extensive experience of conducting personal injury litigation, mainly on behalf of defendants. I would observe that, if solicitors are to be bound by such estimates in the manner which is submitted by the Defendants, all that is likely to be achieved is that ever more time will be expended and costs incurred in connection with them. Moreover, it is likely that the estimates of future costs will either be routinely inflated to provide for every eventuality or will be so qualified as to be meaningless.

18.

I reject, therefore, the first and primary basis for Mr Brown’s attack on the District Judge’s assessment. Equally, it follows from what I have said that I do not consider that any material breach of the Civil Procedure Rules themselves has been demonstrated which should have been penalised by a reduction of the Claimant’s costs.

19.

So far as his final fall-back position, I agree that a substantial departure from the costs estimate called for an explanation. I think it is evident from the District Judge’s decision that he was satisfied by the explanation provided by the Claimant’s solicitors.

20.

In conclusion, the Defendants have failed to satisfy me that the decision of the District Judge was wrong. Accordingly, the appeal will be dismissed.”

The provisions relating to the giving of estimates of costs at significant stages of litigation are important in assisting the court to achieve the overriding objective stated in CPR r. 1.1 and to control the costs of litigation. The purpose of requiring costs estimates is, as is made clear by CPR 43 PD para 6.1, to keep the parties informed about their potential liability in respect of costs, and to assist the court to decide what, if any, order to make about costs and case management. Realistic costs estimates will also enable the parties to settle costs issues: they should therefore reduce the need for assessments of costs. In his final report on Access to Justice, Lord Woolf said at para 32:

“32.

It is important that the court is aware of the parties’ estimate of the expenditure which has been or will be incurred when considering the future conduct of a case. The parties’ estimates will be dependent on how they are proposing that the case should be conducted. If one method of dealing with the case would be beyond the resources of one of the parties, then dealing with the case justly may involve not adopting that procedure. This could be particularly important where, for example, one party wishes a case to remain on the fast track but the other is arguing for the case to be transferred to the multi track.

33.

Estimates need not go into detail and would therefore not disclose confidential information which might be of tactical value to an opponent. That would fall far short of the radical proposal set out by Adrian Zuckerman in the issues paper. The estimates would be indications to help the procedural Judge decide the best course of action rather than budgets which limited what parties could recover. My other recommendations need to be “bedded down” before proceeding further in this direction on costs.”

Costs estimates are an important part of the machinery of case management. At the first case management conference, the court will have the parties’ statements of case, and will therefore be aware of the issues in the case. The allocation questionnaires will inform the court how many witnesses, and in particular how many expert witnesses, each party wishes to call at the hearing. The parties’ costs estimates are part of the material that is placed before the court at this early stage of the litigation to enable it to form a view as to what measures it should take in order to manage and control the case in the interests of what is reasonable and proportionate. In Jefferson v National Freight Carriers plc [2001] 2 Costs LR 313 at para 40, Lord Woolf CJ approved the following statement by Judge Alton:

“In modern litigation, with the emphasis on proportionality, there is a requirement for parties to make an assessment at the outset of the likely value of the claim and its importance and complexity, and then to plan in advance the necessary work, the appropriate level of person to carry out the work, the overall time which would be necessary and appropriate [to] spend on the various stages in bringing the action to trial and the likely overall cost. While it was not unusual for costs to exceed the amount in issue, it was, in the context of modest litigation such as the present case, one reason for seeking to curb the amount of work done, and the cost by reference to the need for proportionality.”`

We accept, of course, that it will not always be possible at the allocation questionnaire stage to provide a reasonably accurate estimate of the likely overall costs. But it should usually be possible to do so even at that stage, especially in run of the mill cases. Where it becomes clear during the course of the litigation that the estimate was inaccurate, it is all the more important to comply with the obligation in CPR 43 PD para 6.4(2) to file an updated estimate at the listing questionnaire stage.

If it is true, as the district judge in the present case suggested, that it is common practice for solicitors to provide costs estimates only at the allocation questionnaire stage, then that practice should cease. Until and unless CPR 43 PD para 6.4(2) is removed, solicitors should also file an estimate of costs at the stage of the listing questionnaire, unless the court otherwise orders.

The provisions about the purpose of costs estimates and their relevance in relation to the assessment of costs appear in practice directions, and not in the rules themselves. This seems to have influenced the judge in his interpretation of CPR 43 PD para 6.6. He referred to passages in the judgments of this court in Re C (Legal Aid: Preparation of Bill of Costs) [2001] 1 FLR 602, and Godwin v Swindon Borough Council [2002 1 WLR 997. In the former, Hale LJ said at para 21:

“(21). Unlike the Lord Chancellor’s orders under his ‘Henry VIII’ powers, the Civil Procedure Rules 1998 themselves and the 1991 Remuneration Regulations, the Practice Directions are not made by Statutory Instrument. They are not laid before Parliament or subject to either the negative or positive resolution procedures in Parliament. They go through no democratic process at all, although if approved by the Lord Chancellor he will bear ministerial responsibility for them to Parliament. But there is a difference in principle between delegated legislation which may be scrutinised by Parliament and ministerial executive action. There is no ministerial responsibility for Practice Directions made for the Supreme Court by the Heads of Division. As Professor Jolowicz says, loc cit, p 61, ‘It is right that the court should retain its power to regulate its own procedure within the limits set by statutory rules, and to fill in gaps left by those rules; it is wrong that it should have power actually to legislate’.”

In the latter, May LJ said at para 11:

“Practice directions are not the responsibility of the Civil Procedure Rule Committee, whose responsibility under section 2 of the Civil Procedure Act 1997 is limited to making civil procedure rules. Practice directions are subordinate to the rules: see paragraph 6 of Schedule 1 to the 1997 Act. They are, in my view, at best a weak aid to the interpretation of the rules themselves.”

It is true that the ground rules which set out the relevant criteria for the assessment of costs are contained in the rules, not the practice directions. But the rules are, to some extent, open-textured. In particular, CPR r. 44.5(1) provides that the court is to have regard to “all the circumstances” in deciding whether the costs were proportionately and reasonably incurred, or were proportionate and reasonable in amount. Without prejudice to that general injunction, the court must also have regard to the various factors mentioned in CPR r. 44.5(3), of which the first is the conduct of the parties. In our judgment, the provisions in the practice direction as to the giving of estimates of costs at various stages of the litigation are made pursuant to the power in the court to regulate its own procedure within the limits set by the statutory rules and to fill in gaps left by those rules. CPR 43 PD para 6.6 does not purport to, nor does it, introduce criteria for the assessment of costs which are inconsistent with, or additional to, those contained in CPR r. 44.5 itself. The provision has been drafted conservatively: the court may have regard to any estimate previously filed, and such estimate may be taken into account as a factor among others when assessing the reasonableness of any costs claimed. In our judgment, this merely spells out explicitly what is implicit in the broad power conferred on the court by CPR r. 44.5(1). For completeness, we should also mention CPR r. 44.14 which gives the court the power to disallow all or part of the costs which are being assessed where a party or his legal representative fails to comply with a rule or practice direction.

The judge questioned the purpose of the provision of costs estimates. As we have said, it is to enable all parties to the litigation to know what their potential liability for costs may be. That enables them to decide whether to attempt to settle the litigation, or to pursue it, and (in the latter case) what resources to apply to the litigation. But at least as importantly, it also enables the court to take account of the likely costs in determining what directions to give. In so far as the judge was suggesting that costs estimates are unnecessary, and will merely add to the costs of the litigation, he was wrong to do so. The practice direction is expressed in clear mandatory terms: costs estimates must be provided. It is also to be noted that it requires the legal representatives to serve the costs estimates on their clients. Apart perhaps from cases such as the present where a solicitor acts for a client who makes it clear that he or she does not require such estimates, it is also part of a solicitor’s ordinary professional duty to provide the client with an estimate of future costs.

Nor do we agree with the judge in so far as he seems to have been of the view that CPR 43 PD para 6.6 fetters the exercise of the court’s jurisdiction in relation to the assessment of costs. The language plainly does no such thing.

So how should CPR 43 PD para 6.6 be applied where there is a substantial difference between the amount of the costs estimate and the costs claimed on an assessment? If there is no substantial difference between the two figures, then para 6.6 will have no significance. But, if there is a substantial difference as there was in the present case, the practice direction gives no guidance as to how that difference should be taken into account in determining the reasonableness of the costs claimed. It is clear that some guidance is required. In a valuable article “Costs in Personal Injury Cases” (Journal of Personal Injury Law 2002 issue 2/02) Professor John Peysner said this of costs estimates:

“Cost estimates, like building estimates, have a potential to make costs more predictable and controllable. In project management proper estimating is crucial, contingencies are built in and if the estimate is exceeded the contractor must explain. In cost assessment terms estimation, in effect, should shift the burden of proof onto the potentially receiving party to estimate correctly (always bearing in mind the litigators duty to the court) and to re-estimate. The author was very exercised in the run up to the Civil Procedural Rules about this idea and thought that it would be of great value. In fact it appears to have been a damp squib. Anecdotal evidence suggests that litigators are uncertain as to how the information disclosed is used, judges suspect that the estimating process is not rigorous and there seems little evidence of judges revisiting estimates on assessment.”

If costs estimates have proved to be a “damp squib”, it may be that the reason for this is that judges simply do not know how to take them into account when assessing costs. Another factor may be that, as Judge Mitchell said in the present case, there is a concern that, if para 6.6 is taken seriously, it will merely encourage satellite litigation.

What follows is not intended to provide an exhaustive guide as to the circumstances in which a costs estimate may be taken into account in determining the reasonableness of the costs claimed, but it should assist judges in the application of para 6.6 of the practice direction. First, the estimates made by solicitors of the overall likely costs of the litigation should usually provide a useful yardstick by which the reasonableness of the costs finally claimed may be measured. If there is a substantial difference between the estimated costs and the costs claimed, that difference calls for an explanation. In the absence of a satisfactory explanation, the court may conclude that the difference itself is evidence from which it can conclude that the costs claimed are unreasonable.

Secondly, the court may take the estimated costs into account if the other party shows that it relied on the estimate in a certain way. An obvious example would be where A shows that he relied on the relatively low estimate given by B not to make an offer of settlement, but carried on with the litigation on the basis that his potential liability for costs was likely to be of the order indicated in B’s estimate. In our judgment, it would be a proper use of para 6.6 of the practice direction to take such a factor into account in deciding what costs it was reasonable to require A to pay B on an assessment.

Thirdly, the court may take the estimate into account in cases where it decides that it would probably have given different case management directions if a realistic estimate had been given. To take a rather crude example: suppose that at the allocation questionnaire stage the claimant provides an estimate of overall costs in the sum of £20,000, and claims £50,000 at the assessment. The court might conclude that, if it had known that the claimant’s costs were likely to be of the order of £50,000, rather than £20,000, it would probably have given different directions from the ones it gave, and that these would have had the effect of reducing the claimant’s costs. It might, for example, have trimmed the number of experts who could be called, and taken other steps to slim down the complexity of the litigation in the interests of controlling costs in a reasonable and proportionate manner.

In our view, para 6.6 of the practice direction gives the court the power to take matters such as these into account in deciding whether, and if so how far, to reflect them in determining what costs it is reasonable to order the paying party to pay on an assessment. We do not, however, consider that it would be a correct use of the power conferred by para 6.6 to hold a party to his estimate simply in order to penalise him for providing an inadequate estimate. Thus, if (a) the paying party did not rely on the estimate in any way, (b) the court concludes that, even if the estimate had been close to the figure ultimately claimed, its case management directions would not have been affected, and (c) the costs claimed are otherwise reasonable and proportionate, then in our view it would be wrong to reduce the costs claimed simply because they exceed the amount of the estimate. That would be tantamount to treating a costs estimate as a costs cap, in circumstances where the estimate does not purport to be a cap.

Nor is there any justification for interpreting the provisions in the CPR as equating costs estimates with costs budgets or caps. There is, however, much to be said for costs budgeting and the capping of costs. Some judges have made prospective costs cap orders exercising the general power conferred by section 51(1) of the Supreme Court Act 1981: see, for example, Gage J in AB v Leeds Teaching Hospitals NHS Trust (in the matter of the Nationwide Organ Group Litigation) [2003] EWHC 1034. This is not the place to review these decisions. Suffice it to say that, whatever the scope of the jurisdiction to make such orders, it is quite different from the jurisdiction that is exercised retrospectively at the stage of costs assessment, and when the court is required to decide the amount of reasonable and proportionate costs. Costs estimates can also alert the judge responsible for case management to the need to take appropriate action to prevent disproportionate costs from being incurred.

We acknowledge the concerns about the danger of satellite litigation. It might be said that the guidance that we have sought to give will foster disputes. Did the paying party who alleges that he relied on the estimate in fact rely on it, and to what extent? Would the court in fact have made different case management decisions if it had been provided with a realistic costs estimate, and what effect would that have had on the litigation and the parties’ costs? What is the explanation for the difference between the costs estimate and the costs claimed at the assessment stage, and does the explanation satisfactorily account for the difference? These are all valid questions to ask. But these concerns do not justify setting at nought the important CPR provisions relating to the making of costs estimates. If costs estimates are not taken into account at the assessment stage, then they will be entirely nugatory. It should not be difficult for the court to determine whether, and if so how, the paying party has relied on the costs estimate given by the receiving party without conducting an elaborate and detailed investigation. Likewise, in most cases the court should be able without prolonged investigation to form a judgment as to whether, and if so how, the case would have been managed differently if a realistic costs estimate had been given.

If, applying the guidance given in this judgment, the court is satisfied that the costs claimed should be reduced having regard to the costs estimate, the question remains: by how much should the costs be reduced? This will always depend on the circumstances of the individual case. It is a matter for the judgment of the court to decide what reduction to make. Regard should be had to the costs estimate when considering whether the costs claimed were reasonably incurred and reasonable in amount. Moreover, where justice so requires, specific deductions can be made from the costs recoverable to reflect the impact that erroneous and uncorrected estimates have had on case management or on the conduct of the other party.

We consider that, contrary to what occurred in the present case, the costs judge should determine how, if at all, to reflect the costs estimate in the assessment before going on to decide whether, for reasons unrelated to the estimate, there are elements of the costs claimed which were unreasonably incurred or unreasonable in amount. This will avoid the danger of “double jeopardy” referred to in the context of a discussion about proportionality by Lord Woolf CJ in Lownds v Home Office [2002] 1 WLR 2450, [2002] EWCA Civ 365 at para 30.

We recognise that the use of CPR 43 PD para 6.6 to control costs by taking costs estimates into account at the assessment stage is not the most effective way of controlling the cost of litigation. It seems to us that the prospective fixing of costs budgets is likely to achieve that objective far more effectively. The question of costs budgets was raised before the Civil Procedure Rule Committee in June 2001. It is contentious and important. The committee decided to explore the issue, but has not reached any conclusion about it. We invite the committee to re-examine the provisions relating to costs estimates to see whether they should be amended to make them more effective in the control of costs; and also to reach a conclusion on the issue of cost budgets.

On behalf of the defendant, Mr Guy Mansfield QC does not seek to challenge the finding by the district judge that the defendant did not rely on the costs estimate given by the claimant in the allocation questionnaire, nor does he suggest that, if a realistic estimate had been given, the court would probably have managed the case differently. Furthermore, he does not submit that the costs estimate was a reliable guide as to the amount of costs that it would be reasonable to award at the assessment stage. On the contrary, he submits that the estimate was an obviously unreliable guide as to the costs that it would be reasonable to award: it was, he says, a hopeless estimate. He points out that the estimate of costs to be incurred by the claimant after the allocation stage was based on an estimate of 25 hours’ work, whereas the estimate of costs to be incurred by the defendant was based on 50 hours’ work. It is common ground that ordinarily a claimant’s legal representative shoulders a greater burden than his counterpart. In so far as the district judge was satisfied that there was an explanation for the difference between the amount in the costs estimate and the amount finally claimed, Mr Mansfield submits that he was in error. The only features of the case that were not foreseeable when the claimant completed the allocation questionnaire were (a) the site visit, and (b) the point raised by the defendant that the claimant’s employment would have been terminated by reason of redundancy even if he had not been injured. This latter point led to the appointment of an employment expert. But neither feature explained the gulf between the two figures for solicitors’ profit costs. He accepts that the costs assessed by the district judge were reasonable and proportionate if the costs estimate is left out of account. But the claimant should be bound by that estimate.

Mr John Foy QC submits that the district judge was entitled to conclude that the claimant should not be bound by the costs estimate, that the circuit judge was right not to interfere with that decision, and that this court should follow the same course. He also submits that the costs estimate was made on the footing that the case would be settled at an early stage, so that the overall costs estimated were those up to the time when it was thought that the case was likely to settle.

We should say at once that all costs estimates are required to include estimates of the overall costs to be incurred on the assumption that the case will not settle, and not merely estimates of future costs up to some (unspecified) date on which it is thought that the case is likely to, or might, settle. There is no warrant for interpreting the provisions as to costs estimates in any other way.

Despite the persuasive way in which Mr Mansfield puts his case, we are in no doubt that this appeal must be dismissed. We are prepared to accept his submission that the costs estimate was hopelessly inadequate and that there is no satisfactory explanation for the gulf between the estimate and the final figure. But the essence of his argument is that the claimant should be bound by the estimate for no other reason than that the estimate was made and it was hopelessly too low. In other words, the claimant should be penalised because the estimate was seriously inadequate. For the reasons given at paras 26 – 30 above, that is not a sufficient reason to hold that a party should be bound by his estimate.

It follows that this appeal must be dismissed.

Appeal dismissed

Appellants to pay the respondents costs of the appeal in the agreed sum of £13,032.51

Stay of Execution ordered by dyson LJ on 22 May 2003 to be removed.