LUEN TAT WATCH BAND MANUFACTURER LTD v. STEPHEN LIU YIU KEUNG AND ANOTHER

HCMP 1567/2021

[2026] HKCFI 5361

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF FIRST INSTANCE

MISCELLANEOUS PROCEEDINGS NO 1567 OF 2021

________________________

________________________

________________________

_______________

D E C I S I O N

_______________

Introduction

1.  On 7 May 2026, I handed down the judgment in this action: see [2026] HKCFI 2564. I made a costs order nisi that the defendants do pay the plaintiff’s costs of the amended originating summons, including all reserved costs, to be taxed on an indemnity basis if not agreed, with a certificate for two counsel: see para 290.

2.  By summons dated 20 May 2026, the 1^st and 2^nd defendants ask for the costs order to be varied such that they should bear only 50% of the plaintiff’s costs instead of 100%.

3.  By consent of the parties, the summons is dealt with by paper disposal.

4.  I shall adopt below the defined terms and paragraph numbers used in the judgment.

Grounds of variation

5.  The defendants say that their proposed order would fully reflect the following:

(1)  The plaintiff failed to succeed in respect of the majority of its claim, which had caused substantial costs and time to be expended unnecessarily.

(2)  The plaintiff failed to obtain any of the substantive relief sought.

(3)  The defendants succeeded on multiple matters of fact and law, which had a dispositive effect on the result of the case.

(4)  It was no longer necessary for the plaintiff to continue with prosecuting the case after the 2023 production. Given that its objective was to seek a return of all the Schedule 1 documents, it could have withdrawn the proceedings after 28 July 2023. Had it done so, the defendants would not have incurred the subsequent costs which were substantial.

(5)  The defendants never had a contumelious intent. They had expended vast amounts of time and resources to comply with the orders and had ultimately purged the contempt at a huge cost.

6.  I shall treat the above as the defendants’ grounds of variation.

Legal principles

7.  Insofar as the defendants are seeking an issue-based costs order, the following principles are applicable:

(1)  The general starting point for deciding costs after trial is that costs should follow the event: Order 62, rule 3(2) of the Rules of the High Court, Cap 4A.

(2)  The general rule does not cease to apply simply because the successful party raises issues or makes allegations on which he fails, but where that has caused a significant increase in the length or costs of the proceedings, he may be deprived of the whole or a part of his costs: see Re Elgindata (No 2) [1992] 1 WLR 1207, 1214A-C.

(3)  See also Order 62, rules 5(1)(e) and 5(2)(a) and (b) which expressly provide that in exercising its discretion on costs, the court should take into account the conduct of the parties, including whether it was reasonable for a party to raise, pursue or contest a particular allegation or issue and the manner in which he did so.

8.  In Momin Lok v Hospital Authority [2021] HKCA 1319, the Court of Appeal cautioned that an issue-based costs order must not be so readily made as to undermine the utility of the general rule that costs follow the event and to lead to the proliferation of unwarranted costs disputes. It was emphasised that the question of costs is one that ultimately depends heavily on the facts and circumstances of each case: see para 10.

9.  Both parties have cited a considerable number of authorities. I believe that many of them are cited to illustrate how the court’s discretion on costs was exercised in other contempt cases. However, each case turns on its own facts and the exercise of the discretion is highly fact-sensitive. I therefore do not find it necessary to deal with these authorities individually in this decision.

10.  I also wish to point out that, as far as the production order is concerned, as a matter of analysis, there was only one breach, and that is the defendants’ failure to return all the Schedule 1 documents to the plaintiff. And that remained so even though the plaintiff had put forward seven heads of complaint. In this respect, the present case is different from other contempt cases involving allegations of multiple and distinct breaches, where issue-based costs orders might be more readily made where contempt for some but not all of the alleged breaches was established after trial.

Discussion

11.  Before discussing the five grounds of variation one by one, I would highlight some special features in this case which, in my view, have a material bearing on the issue of costs.

12.  First, the plaintiff had no first-hand knowledge of how many Schedule 1 documents existed, how the defendants had filed and stored them within their own system, what they actually did in order to retrieve them, and what they knew about them at the time. The plaintiff’s case was built on the defendants’ own narrative on these matters: see para 42.

13.  Second, a massive number of documents had been generated during the seven years when the plaintiff was placed under liquidation. And the review exercise undertaken by the defendants to retrieve these documents involved multiple and elaborate steps, with the involvement of Ernst & Young’s in-house IT and forensics teams.

14.  Third, it follows from the above that, from the plaintiff’s perspective, it was by no means a straightforward exercise to decide whether all the Schedule 1 documents had been returned and, if not, the extent of the omission. By way of illustration, in order to determine whether the 1,260 documents returned in the December production were “new” documents, the court directed the parties to come up with a bundle of selected documents to show the broad categories of documents produced in that round: see paras 115 to 118. It was also not readily apparent whether the 541 emails retrieved in the 2023 production were substantively new emails when compared with the previously returned ones: see the discussion in paras 138 to 141. The conclusions ultimately arrived at by the court could only be reached after the extensive evidence concerning the three rounds of production were examined in depth. The adducing and the examining of the evidence in this regard had taken up much time both at the trial and in the course of these proceedings.

15.  Fourth, for the same reason, it must be correct to say that it was not a straightforward exercise to evaluate whether the breach was eventually purged after the 2023 production, and a view could only be formed after all the evidence was examined.

16.  Fifth, it must be emphasised that even after the contempt was proved to have been purged, the court still retained the discretion to decide whether the defendants should be held liable and, if so, proceed to the next stage of sentencing. When exercising the discretion, the court indeed took into account a number of matters which had been revealed and proved in the evidence. Most notably, the court found that vast resources had been devoted to purge the contempt: see para 286. Again, this was a view which could only be formed after all the evidence was properly evaluated.

17.  In Cathay Pacific Airways Ltd v Kwan Siu Wa Becky FACV 5/2012, 26 July 2013, the Court of Final Appeal said, at para 4:

“Usually, an overall view of the justice of the case will serve as the best guide to determine the question of costs.”

18.  In the present case, when one takes a step back and looks at the matter on the whole, one should have no difficulty in concluding that the plaintiff in fact succeeded on the vast majority of the issues in dispute.

(1)  On the issue of construction, I found in favour of the plaintiff’s construction of the production order and rejected the defendants’: see para 187.

(2)  On the impossibility argument, I found that the plaintiff had discharged the burden of showing that it was within the powers of the defendants to comply with the production order by the deadline of 28 April 2021: see paras 201 and 203.

(3)  On actus reus, I found that the plaintiff successfully made out the breach of the production order under six out of the seven heads of complaint and the breach of the affidavit order: see paras 232 and 233.

(4)  On mens rea, I found in favour of the plaintiff on the law: see paras 236 to 257. Ultimately, I found that mens rea was established for three out of the six remaining heads of complaint: see paras 258 to 262.

(5)  On the last resort argument, the plaintiff won: see paras 269 and 270.

19.  At the same time, the plaintiff failed on the issue of purge of contempt and the issue of the court’s discretion: see paras 276 and 286, respectively. I also expressly rejected the plaintiff’s long-maintained stance that a manual “eyeball” review was effective and necessary: see para 197.

20.  However, overall speaking, there is little doubt in my mind that the plaintiff can be said to be the majority winner. It had embarked on these committal proceedings to show that the defendants were in breach of the orders and to coerce their return of all the Schedule 1 documents. It succeeded on both counts. As observed in the judgment (at para 269), it was only after the committal application was taken out and further complaints were made in the affirmations that the December production and the 2023 production took place, which had the effect of returning all the Schedule 1 documents to the plaintiff. Even though the court made no substantive order on the application in the end, there was an express finding that the defendants were in breach of the orders and hence in contempt of court.

21.  Having stated my overall view on the extent of success of the plaintiff, I now turn to the five grounds of variation put forward by the defendants.

22.  Under the first ground, the defendants contend that the plaintiff was unsuccessful in the majority of its claims, having failed to prove four of the seven heads prosecuted and maintained through the end of the trial, and that such unsuccessful claims resulted in the expenditure of substantial time and costs by the defendants. On their estimate, the 15 custodians issue which required the defendants to agree on the joint bundle of selected documents added an additional cost of about US$240,000. The corrupted documents issue contributed to a 2-day increase in the length of the trial. As the plaintiff had only achieved a success rate of about 43% (ie 3 out of 7), the defendants submit that they should only bear 50% of the plaintiff’s costs.

23.  In my view, in the particular circumstances of the case, the defendants’ reliance on the failed heads of complaint does not properly reflect the true success enjoyed by the plaintiff in this committal application.

24.  As a matter of fact, the raising of these two issues did lead to the return of more documents after the April production. The time spent on the 15 custodian issues, together with other steps taken by the defendants, led to the return of 1,026 documents in the December production, 391 of which could not be considered to be “near duplicates” or otherwise inconsequential: see para 118. The time spent on the corrupted documents issue led to the return of 541 documents in the 2023 production, some of which, I had found, could be properly regarded as emails with new contents: see para 141. Hence, while the plaintiff ultimately lost on these two complaints on the requirement of mens rea, the raising of these issues and the costs spent on them did achieve a substantive result for the plaintiff.

25.  Furthermore, as observed above, the evidence concerning the two complaints was also relevant and material evidence going to other issues, including the impossibility argument (whether the steps taken by the defendants were capable of retrieving all the Schedule 1 documents), the issue of purge (whether all the documents were returned after the December production and the 2023 production), and also the exercise of the court’s discretion (ie the finding that vast resources had been devoted to purge the contempt). In other words, the time spent on the two failed complaints were also time which necessarily had to be spent on the other core issues in dispute. It cannot therefore be said that such costs were wasted or that they contributed to an unnecessary increase in the overall costs.

26.  I therefore find that it is not a proper case to make an issue-based costs order as proposed by the defendants. The first ground fails.

27.  The second and third grounds can be dealt with together. The crux of the defendants’ argument is that the plaintiff ultimately failed to obtain the substantive order which it set out to seek in its amended originating summons and that was because the defendants succeeded in obtaining multiple findings of fact and law in their favour.

28.  As observed above, when one takes an overall view of the case, I consider the plaintiff to be the majority winner. No deduction of its costs is therefore warranted under these grounds.

29.  Under the fourth ground, the defendants say that it was no longer necessary for the plaintiff to continue the committal application after the 2023 production, given that there were no further documents to be returned after that point and there was therefore no practical value or utility in persisting with the committal process. Had the plaintiff withdrawn the proceedings at that stage, the defendants would not have incurred over US$3.1 million in legal costs.

30.  I have highlighted above that it was not a straightforward exercise at all to decide the issue of breach and the extent of omission in the three rounds of production. And it was only after an extensive review of all the evidence that the court was able to come to the various material findings. I do not therefore find it unreasonable for the plaintiff to pursue the committal process to its conclusion. No deduction to its costs should be made under this ground.

31.  Under the fifth ground, the defendants highlight that they never had a contumelious intent, they had expended vast amounts of time and resources to comply with the orders, and they ultimately purged their contempt at a cost in excess of HK$24.8 million.

32.  I accept that these are matters which the court may take into account. But I do not think they carry much weight on the issue of costs in the circumstances of this case.

Conclusion

33.  None of the grounds put forward by the defendants to vary the costs order are valid. I therefore dismiss their summons. The costs order nisi made in para 290 of the judgment be made absolute.

34.  I make a costs order nisi that the plaintiff do have costs of the summons, to be taxed on an indemnity basis if not agreed. Costs should follow the event. As this summons forms part of the committal proceedings, the same basis of taxation should apply.