In the judgment today in Kupeli -v- Cyprus Turkish Airlines [2016] EWHC 1478 (QB) Mrs Justice Whipple considered issues relating to costs liability after the trial of a preliminary issue.

“….there is a world of difference between a case which comes to trial after reasonable efforts at settlement have been made but settlement has proved impossible, and a case where one party has simply refused to engage, preferring to take the view that it will see its opponents in Court. This is the latter type of case. That attitude inevitably gets weighed in the balance when it comes to costs, if that party fails.”


  • The question of success was usually measured by who was going to pay the cheque.
  • Where some of the claimants had succeeded, albeit not a majority, it was appropriate to describe this as a win for the claimants.
  • The defendants conduct in failing to negotiate or respond to offers of settlement was a relevant factor.
  • It was appropriate that the defendant pay 33% of the claimants’ costs.


A large number of claimants brought actions relating to damages they suffered after the defendant airline had its operating licence suspended.  At trial the judge found that

  • 49 claimants in “category 1” were good claims in principle.
  • 95 claimants in “category 2” claims were dismissed.
  • 694 claimants in category 3 may be able to proceed if the pleadings were amended (however the judge proceeded on the basis that these claims had failed).

The action did not proceed by way of a group litigation order but by way of joined claims.


  1. There are two main issues to decide on costs. They are: (i) who was the successful party? And (ii), should the successful party’s costs be discounted, reduced or subject to any offset by way of a costs order in the opposite direction, to reflect particular aspects of this case?


  1. The Defendant argues that it is the winner. Mr Adkin reminds me that of the total number of claimants in the group, only a few have succeeded. Further, he reminds me that the trial was listed to determine, amongst other things, a “List of Issues” and that the Defendant has succeeded on every single one of those issues (most of which were abandoned before trial, with the remaining few decided against the Claimants at trial). He submits that this was group litigation, and in that context the fact that two individuals within a very large group succeeded in their claims should not detract from the big picture, which is one of failure for the bulk of the claims. He suggests that common sense demands, taking an overview, that the Defendant must be seen to be the winner, and therefore should have a costs order in its favour.
  2. By contrast, the Claimants argue that they are the winners. The Claimants, appearing by Mr Bradley, rely in particular on A L Barnes Limited v Timetalk (UK) Limited [2003] EWCA Civ 402 where, at paragraph 28, Lord Justice Longmore says:
“In deciding who is the successful party the most important thing is to identify the party who is to pay money to the other. That is the surest indication of success and failure.”
He also relies on the notes in the White Book under the heading “Who is the successful party?” at 44x.3.6, which emphasise that success is not a technical term but a result in real life which is to be determined with the exercise of common sense, but that in money claims, the successful party will be the one who receives money as a result of the case, citing Widlake v BAA Limited [2009] EWCA Civ 1256 and Ward LJ in Day v Day, noted in the White Book, who said that “the question of who is the unsuccessful party can easily be determined by deciding who has to write the cheque at the end of the case…“. There are other more recent cases which are cited in the White Book to similar effect, for example, Fox v Foundation Piling Limited [2011] EWCA Civ 790 where Lord Justice Jackson agreed that a claimant who had only recovered a fraction of the overall claim was the “winner” for costs purposes.
  1. The White Book commentary is consistent with paragraph 22.17 from Cook on Costs 2016 to which I was taken by Mr Bradley.
  2. Mr Adkin argues that these authorities are distinguishable because they are not concerned with group litigation, but rather with individual claims where the assessment as a matter of common sense is different.
  3. I see no reason to distinguish the authorities on which Mr Bradley relies, which apply regardless of the number of claimants involved. The broad principle for which all these authorities stand is that common sense must be applied to reach a realistic evaluation of who has won and who has lost, and that one highly relevant factor is whether, at the end of the day, one party will receive money from the other.
  4. I agree with Mr Bradley that the Claimants are the winners because at the end of the day, they will receive a cheque from the Defendant.
  5. In assessing the extent of that success, I disagree with Mr Adkin’s evaluation that his client has succeeded at least to the extent of 94% (comprising the total number of Claimants in Categories 2 and 3: see paragraph 25.3 of his submissions dated 21 April 2016). The Part 1 Trial litigated the overarching principles applicable to the claims by reference to test claims drawn from each category. The number of claims in each category did not affect the trial or preparation for trial in any meaningful way. A more realistic approach is one or other of the following: two out of ten of the test Claimants succeeded in their actions and walked away with damages in their favour, which equates to 20% success (if a percentage is to be adopted); alternatively, one out of three categories of claims succeeded, namely Category 1, which equates to 33% success. Mr Adkin argues that the Category 1 cases were not disputed in principle, because Atlasjet accepted that a contract had existed with the Category 1 Claimants (cf paragraph 55 of the main judgment). But Atlasjet maintained its defence to all the Category 1 claims on the facts. And so the Claimants in Category 1 still had to come to Court, within the Part 1 Trial, to prove the viability of their claims (accepting of course that the Part 1 Trial did not extend to determining each and every one of the Category 1 claims on their own facts). But as a category of claimants, they were successful. I conclude that the Claimants’ success is not so modest that it can or should be treated as immaterial. I reject Atlasjet’s arguments to the contrary.
  6. For those reasons, I conclude that the Claimants are the winners in the Part 1 Trial. As a starting point, Atlasjet should pay the Claimants’ costs.


  1. It is here that I believe Atlasjet’s arguments have greater traction. There are two broad matters which tend to reduce the amount payable by Atlasjet to something less than 100%. The first is the fact that the Claimants lost every one of the List of Issues, either because that issue was abandoned before trial (true of most of them) or because I determined the issue in Atlasjet’s favour. I accept that the List of Issues occupied a substantial part of the trial preparation, although they occupied only a modest part of the trial itself. Under CPR 44.2(4)(b), these are parts of the case on which the Claimants have not been successful.
  2. The second is the fact that the majority of the contractual claims were indeed lost at trial, with only a minority succeeding. This too is relevant under CPR 44.2(b). However, care must be taken in determining the extent of the discount for this reason: the majority of the evidence and argument I heard at trial would have been required, even to establish just one of the claims in Category 1. That is because the key to the Claimants’ success on the Category 1 claims was the unreliability of Atlasjet’s own records, systems and witnesses. That unreliability was established as a result of the extensive evidence called to deal with booking systems, systems for cancelling reservations, allegations of systemic overbookings, Atlasjet’s claim to the Court of Audit and the documents generated by the Court of Audit. These were generic areas of evidence not related specifically to the claims. But this evidence turned out to be very important to the successful Category 1 claims, and the Claimants should in principle recover their costs associated with it.
  3. I have considered whether either or both of these matters should be reflected in an “issues based” costs order, permitted under CPR 44.2(6)(f). But, mindful of CPR 44.2 (7), I have concluded that it is both practicable and preferable to make an order under 44.2 (6)(a) instead, and to award the Claimants a percentage of their costs. That avoids the difficulty of identifying the issues to which the particular costs attach, which might be a very difficult exercise to perform in retrospect (for the parties and any costs judge in due course). That also avoids the prospect of continuing disputes over costs which might go on for months or years, noting that this is a case which has already been ongoing for a long time, appears to have generated a fair amount of ill feeling between the parties, and which quite clearly needs to be brought to an end. Finally, that avoids the spectre of what I would consider to be an undesirable and unfair outcome, namely of the Claimants’ overall win (as I have found it to be) being eradicated (in effect) by the Defendant’s costs attributable to particular issues. It is much better to determine the end position on costs now. I therefore conclude that a percentage costs order is appropriate.
  4. In determining the correct percentage, and noting the two factors in Atlasjet’s favour which I have recorded above, I turn to consider the Claimants’ complaints about Atlasjet’s conduct of the case, which the Claimants argue militate towards a higher costs percentage in their favour. The Claimants argue that the Defendant, by its conduct of this litigation and its attitude generally to the claims, has caused a great deal of unnecessary cost to be incurred; they argue that its conduct is to be taken into account under CPR 44.2(4)(a) and (5). The Claimants make a variety of complaints about Atlasjet’s conduct. Those which are the more compelling are these: they complain that Atlasjet did not answer the pre-action protocol letter which they sent to Atlasjet on 26 May 2012, a factor specified as relevant by CPR 44.2 (5)(a). Further, they say that Atlasjet’s conduct of the trial so far as evidence and disclosure was concerned was seriously deficient: Atlasjet failed to disclose all the documents which were relevant to these claims, and most notably in that context it failed to disclose the log books, which only came to the Claimants’ and the Court’s attention while Mr Ersoy gave his evidence. They complain that Atlasjet failed to assert at any time prior to trial that the Court of Audit documents, which formed a key part in the Claimants’ case on passenger numbers, were not Atlasjet’s documents, and further were not considered by Atlasjet to be reliable. Further and more generally, the Claimants argue that Atlasjet put forward confused and confusing evidence, the consequence of which was that the Claimants were sent on wild goose chases for information and have been left, even now, not knowing precisely what occurred in their cases. Finally, so far as efforts to settle were concerned, the Claimants rely on the fact that they did make an offer to settle on a Calderbank basis dated 28 April 2015, giving two options for settlement (one in the form of a cash offer). Atlasjet rejected that offer, and made no counter offer. I am told that there was an attempted ADR just seven days before trial but that suggestions of settlements, negotiations or ADR which were put forward by the Claimants prior to that were rejected out of hand by the Defendant.
  5. These points are answered by Mr Adkin in his submissions, and by Mr Zimmer in his latest witness statement. Mr Zimmer accepts that the pre-action protocol letter was not answered, but says that it was only sent days before proceedings were issued by the Claimants, which shows that the Claimants never had any intention of settling prior to issue anyway. He notes that the Claimants have failed on all the matters originally pleaded against Atlasjet, and that the modest success secured by the Category 1 Claimants only arises out of the amended claim which was served in December 2014, by which time this claim (and the costs incurred in relation to it) was well advanced. He says that the Claimants’ Calderbank offer was unrealistically high, that the Claimants have achieved at trial far less than they offered in that letter. He suggests that this case was never going to be possible to settle. He disputes various criticisms advanced by the Claimants as to the manner in which Atlasjet litigated this claim. Finally, he suggests that the fault for confusion over the Court of Audit documents lies with the Claimants, who obtained those documents independently from the Court of Audit and then refused to explain them to Atlasjet.
  6. Some of Mr Zimmer’s points are then contested by Mr Armutlu in his latest email. But there is no need for me to detail his answers, save to record that he disputes many of the points made by Mr Zimmer.
  7. There is, as is clear, a mass of accusation and counter accusation going to each party’s conduct in the case. I am not able to resolve all the points raised, nor do I consider it necessary or proportionate for me to do so. From all of this material, however, I conclude that there are two matters which must be taken into account as conduct issues (under CPR 44.2(4)(a)) in the Claimants’ favour. The first is that the Defendant failed to make full disclosure. It was plain long before trial that the Claimants wanted to establish passenger numbers on Atlasjet’s flights from the UK. It was not until Mr Ersoy came to give evidence at trial that it became apparent that Atlasjet had the log books for each of those flights sitting, undisclosed, in his office (see paragraphs 40-41 of the Judgment). The logbooks were made available to the Claimants during the course of the trial. These logbooks should have been disclosed earlier.
  8. Secondly, Atlasjet resisted all early attempts at discussion or negotiation of this case. This case was crying out for some sensible attempt at negotiation before costs racked up and the parties’ attitudes hardened. But Atlasjet did not answer the Claimants’ pre-action protocol letter (in fact Atlasjet was not served with proceedings until August 2012, over two months after that letter was sent, showing that the Claimants were open to an informal response; the Claimants are justified in saying that they had no option but to serve proceedings, given Atlasjet’s silence). The Claimants’Calderbank letter dated 24 April 2015 was, as things turned out, pitched too high; but it was at least some attempt at settlement. Atlasjet refused the offer and made no counter offer. The Calderbank offer was undoubtedly an admissible offer to settle to which I must have regard under CPR 44.2 (4)(c). Mr Adkin is right to say that we will never know if this case could have been settled, but there are two further points to be made. The first is that even if the case could not be settled, an early meeting would surely have focussed the minds of those involved, and is likely to have led at least to some narrowing of issues, which would in the end have saved costs. The second point is that there is a world of difference between a case which comes to trial after reasonable efforts at settlement have been made but settlement has proved impossible, and a case where one party has simply refused to engage, preferring to take the view that it will see its opponents in Court. This is the latter type of case. That attitude inevitably gets weighed in the balance when it comes to costs, if that party fails.
  1. Weighing these various factors, I conclude that Atlasjet should pay 33% of the Claimants’ reasonable costs of the Part 1 Trial on the standard basis. I believe this percentage reflects the overall outcome of the case, the outcome on particular issues in the case, and the conduct of the parties in relation to the case.