AXA CHINA REGION INSURANCE CO LTD AND ANOTHER v. PACIFIC CENTURY INSURANCE CO LTD AND OTHERS

HCA9093/2000

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF FIRST INSTANCE

ACTION NO.9093 OF 2000

---------------------

-----------------------

Before : Deputy High Court Judge Poon in Chambers

HCA9093/2000

香港特別行政區

高等法院原訟法庭

民事司法管轄權

案件編號2000年第9093號

---------------------

---------------------

主審法官：高等法院原訟法庭法官朱芬齡內庭聆訊

判案書

1. 本案第三十被告人歐陽慧敏小姐在2005 年12 月9 日發出傳票，要求擱置2005 年10 月25 日判令原告人申索得值的判決。該判決是基於歐陽小姐沒有提交送達認收書和抗辯書，及按原告人的傳票申請而作出的。

2. 本案的「傳訊令狀」及「申索陳述書」是在2005 年7 月22 日以面交方式送達歐陽小姐。原告人在2005 年10 月17 日提出傳票，以歐陽小姐沒有存檔抗辯書，申請得值判決。該傳票和支持申請的誓章已送達歐陽小姐最後為人所知的地址。2005 年10 月25 日聆訊時，歐陽小姐沒有出席，按原告人的申請，法庭頒下得直裁決。其後應原告人的傳票申請，法庭在11 月29 日修改10 月25 日的命令，把履行命令的時限延長。

3. 2005 年10 月25 日的判決是一項合乎規則的裁決(regular judgment)。歐陽小姐在申請把它擱置時，必須就本身延誤或沒有提交抗辯書等事宜給予良好解釋理由。此外，她亦必須顯示她的抗辯理由具有真實的勝訴機會(real prospect of success)。

4. 就前者而言，歐陽小姐表示她已遷離「傳訊令狀」所載的送達地址，之後亦很少回到舊居，所以在後期才知道原告人已申請及取得勝訴的判決。然而歐陽小姐沒有解釋何以她沒有在收到「傳訊令狀」14 天內提交「送達認收書」，和在之後的14 天內提交「抗辯書」。在歐陽小姐收到的「傳訊令狀」中，有中、英文的指示，清楚說明了這些程序和步驟。它同時說明該文件是一份法律文件，忽視它會有嚴重後果，以及提醒被告人如有疑問可向法院登記處詢問。歐陽小姐的誓章和陳詞都沒有就她沒有提交「送達認收書」和「抗辯書」等事提供解釋。即使她已搬遷，這亦無碍她在指定時間內進行這些程序和提交指定文件。

5. 雖然歐陽小姐沒有就造成延誤給予良好解釋，但法庭仍具有酌情權，在她能顯示有合理勝訴機會的抗辯理由的情況下，擱置判令。

6. 就抗辯申索而言，歐陽小姐在誓章中祇簡單地說，她沒有盜用或濫用原告人公司的資料。在聆訊時，她解釋並沒有刻意要求客人提供原告人公司的文件。她表示她祇是在客人有需要和要求協助時，才閱讀客人提供的屬原告人公司的文件。除此之外，歐陽小姐沒有就申索的具體抗辯理由予以說明。

7. 如代表原告人的朱律師所言，原告人申索所針對的不僅局限於保險單或公司文件，而是廣泛地包括屬於原告人公司的客戶資料及其提供的保險條款等商業資料。歐陽小姐陳詞時表示她明白原告人申索的內容，而她亦已從事保險業一段時間，本席因此相信她應知悉申索非單指保險單的文件。然而除了否認原告人的申索和指控外，她卻沒有就抗辯給予任何詳情和具體說明。歐陽小姐因此未能滿全其在本申請所負的責任，遑論顯示她的抗辯理由具有真實勝訴機會。

8. 基於上述原因，本席撤銷歐陽小姐的傳票申請。

9. 經聆聽雙方陳詞，本席認為應按一訴訟常規，由申請失敗一方支付訴費。本席因此頒令第三十被告人支付原告人是次傳票申請的訟費。如雙方不能就金額達成協議，可交法庭評估。

第一及第二原告人：由史密夫律師行朱秀慧律師代表。

第三十被告人：無律師代表，親自出庭。

HCA 9093/2000

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF FIRST INSTANCE

ACTION NO. 9093 OF 2000

____________

____________

Before: Hon Chu J in Chambers

____________________________

REASONS FOR DECISION

____________________________

1.  On 18 May 2005 I handed down the Decision (“the Decision”) in respect of:

(1)         The plaintiffs’ summons dated 19 August 2003 (“the Questionnaires application”); 

(2)         The plaintiffs’ summons dated 17 December 2003 (“the Non-compliance application”) seeking to enforce the 1^st defendant’s undertaking (“the Undertaking”) to the Court under the Order of Beeson J dated 29 September 2000 (“Beeson J’s Order”), and the Order of Deputy Judge To’s Order dated 24 July 2001 as amended on 23 January 2002 and re-amended on 7 February 2003 (collectively referred to as “DJ To’s Order”).

(3)         The 1^st defendant’s summons dated 13 May 2004 seeking (a) amendments of Beeson J’s Order and DJ To’s Order by inserting an undertaking as to damages,  (b) the discharge of the Undertakings, to the extent that they exceed the scope of DJ To’s Order, and (c) release from further compliance with DJ To’s Order.

2.  At paragraph 76 of the Decision, I made the following orders:

(A)     The Questionnaires application is dismissed.

(B)     On the Non-compliance application:

 (1)    The 1^st defendant is to further comply with the Undertaking by conducting a search of all the non-PMS policies issued on or after 21 July 2000, to the extent this was hitherto not done, with a view to delivering up to the plaintiffs  printouts from SAMS and copies thereof, if any, that are in its possession, custody or power, and that had hitherto not been delivered up to the plaintiffs.

 (2)    The 1^st defendant is to further to comply with paragraphs (3) and (4) of DJ To’s Order by carrying out further search, delivering up and disclosure of (a) white cards, underwriting notifications and daily consolidated reports, (b) the 7 policies and 169 policies respectively referred to in Geoff Wong’s 4th affirmation and Ralph Lau’s 13^th affirmation, and (c) the list of documents set out in the affirmation of Ho Wai Shing Steven.

(C)     On the 1^st defendant’s summons:

          Beeson J’s Order and DJ To’s Order be further amended by inserting the usual undertaking as to damages.  The further amended orders are to be filed and served within 14 days from the date of the handing down of the Decision.

3.  I invited the parties to endeavour to agree on: (1) the timetable for carrying out the further searches, delivery up and disclosure, (2) the costs of the summonses, and (3) the precise terms of the Order, with leave to apply for further directions in the event agreement cannot be reached on one or more of the matters.

4.  Consequently, the case was listed for further directions.  For the purpose of the hearing, the parties had lodged on an agreed list of outstanding issues for determination and direction.  At the conclusion of counsel’s submissions, I gave the directions sought.  My reasons for them appear below. 

5.  For ease of reference, I shall refer to the plaintiffs and the 1^st defendant as AXA and PCI respectively.

Compliance with the undertaking: 

Issues 1 and 2

6.  Issues 1 and 2 stem from paragraphs 58 and 76 of the Decision, which require PCI to further comply with the Undertaking by conducting a search of all the non-PMS policies issued on or after 21 July 2000, to the extent this was hitherto not done, for SAMS documents.   The questions raised are whether PCI is required to search (1) non-AXA PMS policies issued on or after 21 July 2000, and (2) non-WWP NB policies issued after 16 October 2000, other than the Protocol Policies.

7.  When preparing this part of the Decision, I was having in mind paragraph 13 of Ralph Lau’s 13^th affirmation and the chart entitled “ Summary of PCI policies searched for SAMS documents” prepared on behalf of PCI and handed up during the previous hearing.  Insofar as the affirmation is concerned, when it referred to non-PMS policies, the focus in paragraph 13 was on WWP NB policies and non-WWP NB policies.

8.  As for the chart, it divided PCI policies other than AXA PMS policies into two boxes, namely, New Business policies and non-AXA PMS policies.  The latter was labelled as irrelevant.  Although AXA had in the reply submission commented that it was unclear from the evidence that these policies had been excluded from the samples searched, there was no real dispute on PCI’s contention that these non-AXA PMS policies were irrelevant.

9.  In the Decision, by non-PMS policies, I was referring to the New Business policies, and effectively only the non-WWP NB policies, as according to PCI, the WWP NB policies had all been searched.  There was no intention to cover non-AXA PMS policies.

10.  As to the issue of Protocol Policies, I had not addressed my mind to this when dealing with the Non-compliance application in the context of the Undertaking.  Protocol Policies were those policies identified and searched in accordance with the Protocol introduced in November 2001.  The Protocol was accepted by Deputy Judge To as a suitable and helpful aid to the compliance of his orders.  Mr Yu SC submits that the Undertaking ought to be similarly restricted by the Protocol, apart from the temporal limit of 21 July 2000 that I had already ordered.

11.  I had in the Decision said I accepted AXA’s claim over the SAMS documents was proprietary in nature.  This is one of the reasons why I had rejected PCI’s construction of the Undertaking.  However, when considering the Non-compliance application and PCI’s application to discharge the Undertaking, the issue is whether the Undertaking should be strictly enforced or enforced in a reduced or restricted manner.   The fact that the SAMS claim is proprietary in nature is not conclusive of the issue.

12.  As indicated in the Decision, an important consideration in my deliberation is my view that further devotion of substantial resources to these interlocutory steps would carry little benefits to the parties in terms of the final adjudication and resolution of the claim.  The intention of my order on the Non-compliance application and the discharge application has therefore been to limit rather than to broaden the tasks of PCI in complying with the Undertaking.  I had adopted the temporal limit that was imported into DJ To’s Order so as to give a confined scope to the Undertaking as well as to introduce an element of uniformity to the Undertaking and DJ To’s Order. 

13.  As explained in Deputy Judge To’s decision of November 2001, the Protocol was introduced in recognition of the difficulties PCI faced in properly complying with its obligations under the interlocutory order.  It was to assist PCI’s task in the identification of policyholders who were AXA’s clients that had been served by former AXA agents.  These considerations are equally relevant to the compliance with the Undertaking.  I am of the view that it would be appropriate and fair to apply the Protocol to the Undertaking.  Accordingly, the search of non-WWP NB policies for SAMS documents is to be confined to the Protocol Policies.

Compliance with paragraphs (3) and (4) of DJ To’s Order:

 Issues 3 to 5

14.  Issue 3 relates to the further search ordered in paragraph 76 of the Decision in connection with the 169 policies referred to in the Ralph Lau’s 13^th affirmation.  As indicated at the outset of this direction hearing, the reference in the Decision should have been to the one policy out of the 169 policies referred to in the said affirmation.

15.  As for Issue 4, it is no longer in issue in view of AXA’s position at the hearing as conveyed by Mr Shieh SC.  It is accepted by AXA that the search for the Category A documents mentioned in Steven Ho’s affirmation should be subject to both the temporal limit of 21 July 2000 and the Protocol.

16.  On Issue 5, the question that calls for further direction is whether apart from the Category A documents, PCI is also required to search for the Category B documents mentioned in Steven Ho’s affirmation.  AXA’s position at the hearing is that the search of Category B documents will also be subject to the temporal limit and the Protocol.

17.  Steven Ho’s affirmation came about as a result of Mr Yu SC’s submissions on PCI’s application for release from further compliance with DJ To’s Order.  He suggested that AXA should provide PCI with a list of documents that AXA said fell within the same category as the White Cards, Underwriting Notifications and Daily Consolidated Reports, being documents that AXA would not have given to their policyholders.  In his reply submission, Mr Kotewall SC (who appeared for AXA at the previous hearing) indicated that AXA were prepared to provide the list as requested but without prejudice to AXA’s right to raise in the future other types of documents that would not have been provided to their policyholders.

18.  When Steven Ho’s affirmation came to be filed, it contained a reference to two types of documents: Category A documents were documents which were never given to the policyholders, and Category B documents were those that would not be given unless the policyholders   requested them.  As noted by Mr Yu SC, unlike the Category A documents, Category B documents are documents emanated from and submitted to AXA by the policyholders.  It is quite possible for the policyholders to have retained copies of them.

19.  Clearly, the list of documents that counsel suggested and envisaged at the previous hearing is confined to the Category A documents.  Mr Yu SC submits that the direction hearing is not an occasion to ask for documents that had not been applied for at the previous hearing.  I agree.   Category B documents was not the subject matter argued and it will not be right to extend the duty to search to cover them.

20.  Further, given that the objective is to assist PCI by giving a clear idea as to what documents to look for in further complying with paragraphs (3) and (4) of DJ To’s Order, the list of documents must be tied to the definition of Client Data.  At the previous hearing, AXA had accepted that it was appropriate and had agreed to give the assistance. 

21.  Client Data is the subject matter of PCI’s obligations under paragraphs (3) and (4) of DJ To’s Order.   It refers to information obtained by agents in the course of their agency with AXA, not being information previously known to him.   The underlying idea of the list of documents is that if the documents were never given to AXA’s policyholders, such as White Cards, Underwriting Notifications and Daily Consolidated Reports, then they must have been obtained by the agents in the course of their agency with AXA.  Category B documents, however, do not stand in the same category because it cannot be said that they must have been obtained by the agents in the course of their agency with AXA.  Indeed, the point is illustrated by the fact that AXA had sub-divided the documents into Category A and Category B in Steven Ho’s affirmation.

22.  Additionally, I had in paragraphs 73 and 74 of the Decision indicated my views on the strict enforcement versus therelease of the obligations under the interlocutory order.   I had concluded in favour of a limited further compliance with the order.  To extend the further search to the Category B documents would be contrary to my conclusion.

23.  For these reasons, my conclusion under Issue 5 is that PCI is not required to search for the Category B documents as further compliance of paragraphs (3) and (4) of DJ To’s Order.

Timetable for compliance:

 Issues 6 to 9

24.  In view of the outcome on Issues 1 to 5, it is not necessary to deal with Issues 6 to 9.  As to the time for PCI to further comply with the Undertaking and DJ To’s Order, parties are in agreement that PCI should have until 30 September 2005 to complete all the outstanding searches and the verifying affidavit should be filed within 28 days thereafter.  

Costs of the three summonses:

Issues 10 & 11

25.  On AXA’s Questionnaires application, there is agreement that PCI should have the costs in any event.

26.  On AXA’s Non-compliance application, PCI accepts at the direction hearing that AXA should have the costs in any event.

27.  What is in issue is the costs arising out of PCI’s summons. 

28.  In my view, the costs of the application to amend by inserting the undertakings as to damages should be dealt with separately.   PCI has to come to court to make the application in order to secure the undertaking.  Although AXA were prepared to give the usual undertaking, the indication only came about at the hearing.  And although PCI did not succeed in getting a wider undertaking, the time spent on arguing this was admittedly minimal.   The costs of this aspect of the summons should follow the normal rule of costs follow event.  I therefore order that PCI should have the costs of the application to amend Beeson J’s Order and DJ To’s Order to insert the undertaking as to damages in any event, to be taxed if not agreed.

29.  As to the costs on the remaining paragraphs of PCI’s summons, I agree with Mr Shieh SC’s analysis and argument.   In essence, PCI’s applications for a discharge of the Undertaking and release from DJ To’s Order were in response to AXA’s Non-compliance application.   The amended paragraph 6 of the summons was only introduced during the hearing.   Although I had, for reasons given in the Decision, taken the view that Unless orders would not be appropriate, I would not have gone on to consider the manner of further compliance with the Undertaking and DJ To’s Order but for PCI’s summons.   Thus viewed, in applying for the discharge and release, PCI is seeking the court’s indulgence.  It is also incumbent upon PCI to take out the application in order to secure the indulgence.  It follows that AXA should have the costs of paragraphs 5 and 6 (as amended) of the summons.

30.  However, to the extent that AXA had unsuccessfully opposed PCI’s application and the opposition had clearly lengthened the arguments and increased the costs on the dispute as to compliance with the Undertaking and DJ To’s Order, PCI should have the costs of the hearing on the discharge and release application.

31.  In view of the costs orders indicated above, it is necessary to apportion the costs of the hearing on the three summonses so as to facilitate the taxation under the costs order.  Having heard counsel, I consider that one-third of the costs of the hearing should be apportioned to the Questionnaires application, another one-third to the Non-compliance application and the remaining one-third to the application for discharge and release.  Accordingly, one-third of the costs of the hearing on the three summonses should go to AXA with the remaining two-thirds of the costs be to PCI.  The hearing is certified fit for two counsel to attend.

Costs of the application for directions

32.  Finally is the costs of the application for directions, which cover not only the direction hearing, but also the preparations and correspondence associated with it.  In deciding the costs for this, I bear in mind that Issue 3 arose out of my oversight in the Decision, that AXA has basically succeeded in Issues 10 and 11 on costs, whereas PCI’s contentions for the other issues have been upheld.   In the circumstances, I consider it appropriate and fair to award to PCI two-third of the costs of and incidental to the application for directions in any event, to be taxed if not agreed.

Conclusion

33.  To recap, the answers on the agreed list of outstanding issues and the consequential directions are as follows:

Issue 1         PCI is not required to search non-AXA PMS policies issued on or after 21 July 2000 for SAMS documents.

Issue 2         PCI is not required to search non-WWP NB policies issued after 16 October 2000 other than the Protocol Policies for SAMS documents.

Issue 3         In respect of amended paragraph 6 of PCI’ summons, the further search, delivery up and disclosure required is limited to the 169 policies referred to in the 13^th Affirmation of Ralph Lau Kim Ming dated 24 February 2004.                             

Issue 4         Other than WWP NB policies issued on or after 21 July 2000 and the Protocol Policies, PCI is not required to search other non-PMS policies issued on or after 21 July 2000 for the Category A documents referred to in the Affirmation of Ho Wai Shing Steven dated 7 July 2004.

Issue 5         PCI is not required to search for the Category B documents referred to in the Affirmation of Ho Wai Shing Steven dated 7 July 2004.

Issues 6 –9  They are not required to be dealt with.

Issues 10,11 The costs order on the three summonses are:

(a)     Subject to (e) and (f) below, the costs of AXA’s summons dated 19 August 2003 be to PCI against AXA in any event, to be taxed if not agreed.       

(b)     Subject to (e) and (f) below, the costs of AXA’s summons dated 17 December 2003 be to AXA against PCI in any event to be taxed if not agreed.

(c)     The costs of paragraphs 1, 2(a), 3 and 4 of PCI’s summons dated 13 May 2004 be to PCI against AXA in any event to be taxed if not agreed.

(d)     Subject to (e) and (f) below, the costs of paragraphs 2(b) & (c), 5 and 6 (as amended) of PCI’s summons dated 13 May 2004 be to AXA against PCI in any event to be taxed if not agreed.

(e)     One-third of the costs of the hearing on the three summonses be to AXA against PCI in any event, to be taxed if not agreed, with a certificate for two counsel.

(f)      Two-third of the costs of the hearing on the three summonses be to PCI against AXA in any event, to be taxed if not agreed, with a certificate for two counsel.

Timetable for compliance 

PCI was to have until 30 September 2005 to further comply with the Undertaking in Beeson J’s Order and also paragraphs (3) and (4) of DJ To’s Orders.

Costs of the direction hearing   

                   Two-third of PCI’s costs of and incidental to the direction hearing be paid by AXA in any event, to be taxed if not agreed.

HCA 9093/2000

 IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF FIRST INSTANCE

ACTION NO. 9093 OF 2000

____________

____________

Before: Deputy High Court Judge Saunders in Chambers

________________

J U D G M E N T

________________

The Applications:

1.  The Plaintiffs, (AXA), by summons dated 28 December 2004, seeks leave to further amend the Re-Re-amended Writ of Summons with Re-Re-Amended Statement of Claim, and to add a further 31 defendants to the action.  Consequential Orders are also sought.

The background:

2.  The writ was issued on 21 September 2000, initially against the 1^st to 10^th Defendants, following the departure, en mass, of a large number of insurance agents from AXA, to the 1^st Defendant, (PCI).  The 2^nd to 10^th Defendants, were Senior Sales Managers for AXA.  The 11^th to 29^th Defendants were sales agents for AXA, employed by various of the 2^nd to 10^th Defendants.  The proposed 29^th to 60^th Defendants were also, at all material times, sales agents for AXA, similarly employed.  I shall refer to the 2^nd to proposed 60^th Defendants, collectively, as “the individual defendants”.

3.  In the usual way of the insurance, or indeed any business, AXA had collected a large amount of data relating to the personal details of its clients,  AXA policyholders, and held as well, details of the policies taken out by them.  It is AXA’s case that all the documents containing such information and the information itself are secret and confidential and owned by AXA.

4.  AXA stored that information on a computer system known as the Sales and Marketing System, (SAMS), general access to which is restricted to Senior Sales Managers and specific access, in relation to their own clients, is permitted to sales agents who deal directly with clients.  The case for AXA is that in a short period of time prior to the departure of the individual defendants from AXA to PCI, print outs of SAMS client data were made, taken with them to PCI, and have been used by PCI and the individual defendants to poach policyholders from AXA to PCI.  It is contended that this was done by a “plan matching scheme”.

5.  The causes of action against PCI are breach of confidence and the economic tort of unlawful interference with contract.  The cause of action against the 2^nd to 28^th and the proposed additional defendants is based upon an alleged breach of an implied duty of fidelity by the individual defendants in wrongfully removing confidential information belonging to AXA during the currency of their agency with AXA, for use after the termination of their agency, and to the detriment of AXA.

The interlocutory applications:

6.  At the same time as the writ was issued, PCI sought interlocutory injunctions requiring the 1^st to 10^th Defendants to deliver up all SAMS documents and copies thereof in their possession custody and power, to AXA, and to verify compliance of the injunction by affidavit.  On 29 September 2000, Beeson J made an interlocutory order,(ex parte on notice), in AXA’s favour, on the application.  An inter partes hearing came before Deputy Judge To in February 2001, which, between 28 February 2001 and 10 April 2001, occupied 30 days of sitting time, culminating in a 111 page judgement delivered on 24 July 2001.

The conclusions of Deputy Judge To:

7.  In the course of his decision Deputy Judge To came to the following conclusions:

(1)	AXA had raised a serious issue to be tried as to whether the individual defendants were in breach of their duty of fidelity to AXA by copying and removing client data during the currency of their agency for the purpose of using that data to AXA’s detriment after the termination of their agency, (paragraph 85);
(2)	AXA had raised a serious issue to be tried as to whether the individual defendants had removed client data during the currency of their agency for use in competition with AXA after the termination of their agency, (paragraph 133);
(3)	AXA had demonstrated a strong arguable case that PCI had knowledge that the information supplied to it by the individual defendants or other ex-AXA agents had the necessary quality of confidence in it and were imparted in circumstances importing an obligation of confidence to PCI, (paragraph 141);
(4)	AXA had established a serious issue to be tried as to whether PCI was under an obligation of confidence in respect of the information that it received and has been in breach of the duty (paragraph 146);
(5)	AXA had raised a triable issue that the intention of PCI must have gone beyond fair and honourable competition and that PCI intended, at least, to cause some harm to AXA, (paragraph 149);
(6)	There was a serious question to be tried as to whether PCI’s breach of confidence constituted an unlawful means for the tort of breach of confidence, (paragraph 152); and
(7)	AXA had raised a serious question to be tried that in the course of the administration of the plan matching scheme PCI agents had misrepresented AXA policies as “Matching Policies” with PCI policies, to AXA policyholders, (paragraph 158).

8.  That decision, delivery nearly four years ago, has not been the subject of any challenge by the Defendants.

The nature of the present disputes:

9.  It is clear that the definition of what constitute “SAMS Data” and “Client Data” is at the centre of the action.  PCI, through its counsel, Mr Stokes, properly say that they must know precisely what AXA contend in order that they might answer the case.  The most substantive proposed amendments to the Statement of Claim are directed at revisions to certain paragraphs relating to those definitions.  These are subject to strong challenge by Mr Stokes.  Further amendments attempt to particularise relevant information, and the complaint made by Mr Stokes is that the information relied upon by AXA is not sufficiently particularised so as to enable PCI to properly respond to the pleading.

10.  Mr Pun who appeared for the 10^th to 28^th Defendants raised no objections to the proposed amendments to the Statement of Claim, not even to the extent of joining with Mr Stokes in his objections.  He confined his argument to an argument that the proposed additional defendants ought not to be joined.

11.  The issue in respect of the proposed amendments to the Statement of Claim fall conveniently into three areas.  They are first, to give a more accurate and comprehensive description of SAMS data, together with appropriate detail of the case on breach of confidence in relation that to that data, leading to an appropriate definition of “Client Data”.  This revolves around paragraphs 18.2, 20.1 and 22.2 of the draft.  I shall call this “the SAMS and Client Data point”.  Next, objection is taken to paragraphs 28 and 29.4 of the draft in which AXA attempt to particularise, for sufficient pleading purposes, the documents upon which they relied for the case against PCI.  The complaint here is that the documents are not sufficiently particularised.  I shall call this the “Identification Point”.

12.  The third point relates to a further particularisation of the Identification Point.  This was taken in relation to an attachment to the Statement of Claim, referred to as “Schedule C”, which relates to a series of documents which had been recorded on a computer hard disc by an imaging process, allegedly by one of the defendants.  Following a suggestion made by me in the course of argument, I understand that the parties are looking at a means by which those documents may be copied onto a separate disc and made available to the advisors to PCI under controlled circumstances, for security purposes, which Mr Stokes accepted entirely, and that may resolve this issue.  I will not consider it further now, so unless resolved by agreement, this issue remains alive.

The SAMS and Client Data point:

13.  I accept Mr Stokes’s submission that the correct and clear definition of the words “SAMS Data” and “Client Data” is essential, because, to adopt his submission,

“those words determine the scope of information that AXA claim to own beneficially, and by reference to which they contend there has been misuse by the individual defendants in their breach of their duty of fidelity and misuse by PCI in receiving Client Data.  The definition is important in determining the extent of the discovery and the evidence that each defendant will need to obtain.”

14.  It is plain that information is not necessarily Client Data merely because was found in the SAMS data system, or on a document printed out from SAMS.  There were circumstances, AXA accepts, in which SAMS documents were produced and given to a client.  If such document was then handed to PCI, by or via a client, it would not be a confidential document.  It is also accepted by AXA that if a policyholder has material in his position, such as a policy itself, the supply of that information, all of which would be in the SAMS data information base, would not be a breach of confidence.

15.  Mr Stokes’s objection to paragraph 18.2, 20.1 and 20.2 of the draft pleading was based upon a contention that the definition sought to be advanced in the re-re-re-amended Statement of Claim was inconsistent with, and contrary to past statements by the Court, or admissions made by AXA as to the scope of Client Data.

16.  I have come to the conclusion that the objection is unfounded, and that the pleading is sound.  I do so for the following reasons.

17.  The starting point for any pleading in a claim for breach of confidential information is Coco v A N Clark (Engineers) Ltd [1969] RPC 41, a decision of Megarry J.  At page 47 the learned Judge recorded just follow:

“In my judgment, three elements are normally required if, apart from contract, a case of breach of confidence is to succeed.  First, the information itself, in the words of Lord Greene, M.R. in the Saltman case, on page 215, must “have the necessary quality of confidence about it”.  Secondly, that information must have been imparted in circumstances importing an obligation of confidence.  Thirdly, there must be an unauthorised use of that information to the detriment of the party communicating it.”

18.  In the present case, the pleader has set out to establish the two primary elements that must be established before attention is turned to the use of the information.  This is done in two stages.  First, in paragraph 18.1 the plea is made that SAMS is a network system containing secret and confidential information about the Plaintiff’s policyholders and their policies (“SAMS Data”).  Clause 18.2 proceeds to define SAMS data exclusively.  From that paragraph, it is necessary to turn to paragraph 20.1.  There the plea is made that SAMS data was and remains all times a trade secret and/or confidential information.

19.  But that plea is not, as was contended by Mr Stokes, in a grammatical argument based upon the location of commas, a plea that ALL SAMS data constitutes trade secrets and/or confidential information.  The plea carefully limits the data which constitutes confidential information to data which:

“was and is created, obtained, derived and/or collected in the course of and for the purposes of the Plaintiff’s insurance business by or through their offices, employee and/or agents”.

20.  Those two paragraphs, 18.2 and 20.1, when read together consecutively, as they must be, in order to import the necessary limitation factors, are the paragraphs upon which the pleader has relied to establish the first of the elements required by Megarry J.  To adopt the expression used by Mr Strachan in his reply, the pleading establishes “the confidentiality of the corpus of the information” held by AXA.

21.  Megarry J’s second requirement was the information must have been imparted in circumstances importing an obligation of confidence, in other words, the pleader must establish a duty in relation to that corpus of information.  It is apparent in the present case, that in respect of individual agents, the “corpus of information” to which their duties lies will differ for each agent.  That necessarily follows from the type of limitations I have referred to the paragraph 14 above.

22.  In paragraph 20.2 of the draft the duty is expressed, and is appropriately limited to accommodate those different positions by two expressions.  They are first,

“obtained or derived by an agent in the course of his agency with the Plaintiff”,

and secondly,

“save where the information was known to the agent prior to such obtaining or derivation”.

The data thus defined is then internally defined in the draft amendment, by the expression “Client Data”.  The pleading then proceeds to allege that there is a duty upon the defendants to keep Client Data secret and/or confidential. 

23.  When the pleading is analysed in this way, it may be seen that the expression “Client Data” will mean one thing for one sales agent and another thing for another sales agent, and yet another thing for a Senior Sales Manager.  The distinguishing features are however generically common throughout.  They are:

(i)	The SAMS data must have been obtained or derived by an agent or manager in the course of his agency with the Plaintiff: if he obtained information that falls within the definition of SAMS data, not from the SAMS system, but from some other separate independent and outside source, then the data is not Client Data, and the duty will not arise, and;
(ii)	Where the information was known to the agent or manager  prior to his obtaining or deriving the information as well from the SAMS system, then the data is not Client Data, and the duty will not arise.

24.  It must be remembered also that while the data to which the duty will apply may vary from agent to agent, insofar as PCI is concerned, depending upon the source of the information to them, there may be no variation in the data.  It will only be where an agent, who by virtue of the limitations, is exempted from duty, supplies the data to PCI, that PCI will be able to seek refuge in the exemption.  If the information is supplied on a general basis to PCI, for example by a senior manager, who does not have the direct links with clients which may otherwise provide the limitation exemptions to an individual agent, then the information will be confidential and limitations on its use by PCI may well arise.

The identification point:

25.  I turn now to consider the question of the particularisation of documents.  The pleading at the centre of this issue is paragraph 29.3.  This is an issue which appears to arise in most breach of confidentiality cases, where a large bulk of information is taken by an employee who then goes to a new employer and gives that information to the new employer.  It will often be, and is the case here, that it is extremely difficult for the old employer, in this case AXA, to precisely identify what documents or information has been taken.  It may be that electronic trails in computers will often now assist in the resolution of this issue.  That may still be, in part, a solution here, but I am told by Mr Strachan that there is a limit as to the amount to information that is available from the computers to enable precise identification of the documents or information taken.

26.  It is a well-known principle of pleading that it is no answer for a plaintiff the to say to a defendant “you have the information and are well aware of the nature of my complaint”.  The Plaintiff must particularize his complaint.

27.  Mr Stokes relied upon Harbord v. Monk (1878) 38 LT 411.  There, in an action against a stockbroker to reopen accounts on the ground of fraud, the plaintiff alleged that the accounts delivered were untrue and unfair and that brokerage and commission had been improperly charged and excessive and unfair profits had been made.  The stockbroker sought further and better particulars.  The plaintiff amended his statement of claim by stating that all the accounts rendered were untrue, and the defendant had the particulars of the profits made by him, that the improper and excessive charges appeared in the accounts, and that the defendant was well aware of them.  The plea was rejected by Bacon VC who gave the plaintiff four days to amend.

28.  The case is an old one, but while still good law, it must be viewed in the light of modern circumstances.  Where, as so often today, information is stored electronically, and may be printed at will, apparently with no electronic trail to identify precisely the document printed, although as here, the printer and the location from which the data has been retrieved may be identified, there may be no information at all available as to precisely which document has be copied and taken.  This casts matters of the obligation of the identification of information taken in an entirely new light. 

29.  Again, the pleading for which primary complaint has made, paragraph 29.3, and its reference to a particular exhibit, must not be read in isolation.  It is part of a series of pleadings commencing with paragraph 18 and concluding with paragraph 29.6, and must be read in the context of those pleadings.  Each of those previous and subsequent pleadings, when read together, serve to provide particulars of the classes of documents upon which the Plaintiff relies.  It is right that the pleading does not identify, on an individual basis, all of the documents upon which the Plaintiff relies.  Some further progress may be made by the supply of the detail of certain codes referred to on the exhibits identifying the documents, and I am assured by Mr Strachan that AXA will, if it is considered useful in identifying the documents, supply the details of the relevant codes.

30.  What PCI asks however is that individual identification of each document upon which AXA intends to rely should be given.

31.  Having regard to the nature of an action for breach of confidential information, and the manner in which documentation is stored and may be retrieved electronically, I have come to the conclusion that, subject to the provision of details as to the codes available from the computers, there is sufficient particularisation if the class of documents or information allegedly taken is appropriately identified.

32.  I am satisfied the expressions used in the draft re-re-re-amended statement of claim sufficiently identify the relevant classes of documents.  It will be perfectly clear to the defendants, all of whom are involved in the insurance industry, and familiar with the types of documents used, by referring to the classes of documents identified in the pleadings, and bearing in mind the limitations on the classes set out in the pleadings, what will be relevant.

Joinder of Parties

33.  Pursuant to O 15 R 4, AXA seeks to join 31 further defendants.  These proposed additional defendants are all sales agents formerly employed by AXA, who, at a number same time as the 1^st to 28^th defendants left AXA, also left and joined PCI.  They are in precisely the same category as the 11^th to 28^th defendants.  Mr Pun, for a small number, (only 6 of the 17), of this latter group oppose the application for joinder.

34.  I am satisfied that Mr Strachan is correct in his submission that the opposition to the joinder is misconceived.  In the first place while it is right that there will be distinctions between each of the 11^th to 69^th in relation to the clients they served, the extent of common issues are simply overwhelming and demand that all the joined as defendants.  I am confident that by appropriate case management by the trial judge, if any of the 2^nd to proposed 60th defendants found it necessary to instruct separate counsel, it will be not be necessary for that counsel to sit through days of evidence which are irrelevant to his case.  No suggestion is made that there may be any conflict of interest between the individual defendants and so there appears to be no reason why they should not all be represented by the same solicitors and counsel.  To do so would have enormous costs savings and is to be commended to them.

35.  Second, I am satisfied that it is not necessary in this case for the summons for joinder to have been served on the proposed additional defendants.  The very fact that the present defendants, of the same class, have made no application for separate trials for each of their cases demonstrates that this is not a case where it can be suggested that there would be any reasoned opposition to the making of the order.

36.  There will be an order for joinder of the additional defendants in terms of the summons.  All questions of costs are reserved.

HCA 9093/2000

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF FIRST INSTANCE

ACTION NO. 9093 OF 2000

____________

____________

Before: Hon Chu J in Chambers

______________

D E C I S I O N

______________

The applications

1.  There are three applications before the court as follows:

(1)	The plaintiffs’ summons dated 19 August 2003 (“the Questionnaires application”) under Order 24 rules 10, 11 & 11A of Rules of the High Court and the inherent jurisdiction of the Court for an order that the 1st defendant (“PCI”) produces to the plaintiffs (“AXA”) copies of all the questionnaires and responses thereto, references to which were said to have been made in the 5th and 6th affirmations of Ralph Lau Kim Ming (“RL(5A)” and “RL(6A”) respectively) and in exhibits “RLKM-15” and “RLKM-16” to the 8th and 9th affirmations of Ralph Lau Kim Ming.  The discovery sought includes any amendments and supplements to the original responses.
(2)	The plaintiffs’ summons dated 17 December 2003 (“the Non-compliance application”) under Order 42 rule 2, Order 45 rule 6 and the inherent jurisdiction of the Court for an order that PCI should fully and properly comply with its delivery up obligations under its undertaking (“the Undertaking”) to the Court under the Order of Beeson J dated 29 September 2000 (“Beeson J’s Order”), and the Order of Deputy Judge To’s Order dated 24 July 2001 as amended on 10 January 2002 and re-amended on 7 February 2003 (collectively referred to as “DJ To’s Order”) together with ancillary orders.
(3)	The 1st defendant’s summons dated 13 May 2004 under Order 20 rule 11, Order 42 rule 3 and the inherent jurisdiction of the Court for (a) the amendment of Beeson J’s Order and DJ To’s Order by inserting the usual cross undertaking as to damages,  (b) the discharge of the Undertakings, to the extent that they exceed the scope of DJ To’s Order, and (c) release from further compliance with DJ To’s Order.

Background

(1)  The claim

2.  The background leading to the present action had been set out in the Decision of Deputy Judge To dated 24 July 2001. I do not repeat them here.

3.  In this action, AXA claim against PCI injunctions, damages and other relief for breach of confidence in receiving and using confidential information provided by the individual defendants in breach of their duty of fidelity and for unlawful interference with AXA’s business.

4.  The confidential information in question relates to information contained in a computer database of AXA called Sales and Marketing System (“SAMS”) and 21 types of documents listed out in Schedule C to the Amended Statement of Claim (“Schedule C documents”).   The unlawful interference complained of relates to PCI’s activities in marketing a Plan Matching Scheme (“PMS”).

(2)  The Undertaking

5.  By a summons issued on 21 September 2000, AXA applied against PCI and several other defendants for an interlocutory injunction and consequential relief, including delivery up of “Client Data” and disclosure of various information.  At the first hearing of the summons, PCI through leading counsel gave the Undertaking to the Court.  The undertaking as set out in Beeson J’s Order are in these terms:

(1)	Within 7 days of the date of the Order, PCI shall deliver up to AXA or their solicitors all printouts from the SAMS and copies thereof, if any, in his, her or its possession, custody or power; and
(2)	Within 5 days of the expiry of the 7 days referred to in (1) above, PCI shall make and file an affidavit or affirmation verifying compliance with the undertaking given under (1) above.
(3)	DJ To’s Order

6.  The substantive hearing took place before Deputy Judge To.  By his Decision handed down on 24 July 2001, Deputy Judge To granted the relief sought in AXA’s summons subject to revising the definition of “Client Data”.   Paragraph (1)(a) of DJ To’s Order as amended restrained PCI from disclosing, divulging or otherwise making use for the purposes of PCI’s business of, Client Data.  Client Data is defined to cover:

(1)	Any information obtained or derived by an agent in the course of his agency with AXA, not being information known to the agent prior to such obtaining or derivation, and contained in the system SAMS, and any documents whether in hard copy or electronic form containing such information; and
(2)	Any of the Schedule C documents, whether in hard copy or electronic form, and any information contained therein obtained or derived by an agent in the course of his agency with AXA not being information known to the agent prior to such obtaining or derivation.

7.  Under paragraph (3) of the Order, PCI had to deliver up to AXA “all documents containing Client Data, including copies, in its possession, custody, power or control which came into possession, custody, power or control on or after 21 July 2000”, except for “documents containing Client Data of policyholders who have been issued with policy for life insurance by [PCI] on or before 24 July 2001, but only in so far as those documents are required by [PCI] for the purpose of servicing such policies”.

8.  Under paragraph 4(b) to (e) of the Order, PCI had to make disclosure of various matters, including its receipt and divulging of documents containing Client Data and the names and addresses of policyholders and insured persons of AXA whose persons appear in any of the Client Data who have been approached by agents of PCI at a time when such agents were in possession, custody, power or control of Client Data relating to that person.

The Questionnaire application

(1)  The questionnaires and responses

9.  The questionnaires and responses that give rise to the Questionnaire application came about as a result of PCI’s obligations under paragraphs (3) and (4) of DJ To’s Order.   They were first mentioned in RL(5A), which was to give an estimate of the time PCI required to comply with the delivery up and disclosure obligations.  The affirmation mentioned three types of questionnaires and responses, which were designed to enable PCI to search for documents that contain Client Data and to ascertain the nature of solicitation its agents made to holders of AXA policies.  They are:

(1)	Schedule C Questionnaire for agents who had submitted Schedule C Documents to PCI, to ascertaining the source of the documents, namely, whether they were provided by the policyholders or removed/detained by the agent upon leaving AXA;
(2)	Agents Questionnaire to identify those August 2000 PMS policies where the agents had used Client Data, and/or in respect of the 9 pairs of AXA/PCI policies complained of by AXA in these proceedings, where they made representations that the policies are matching or comparable; and
(3)	Solicitation Questionnaire to ascertain the nature of the agents’ solicitation, which was required to be disclosed under paragraph 4(d) of DJ To’s Order.

10.  RL(6A) reveals that the responses received were analyzed and the information obtained was compiled into spreadsheets.  They form the basis upon which decisions on what documents should be delivered up were taken.  RLKM-15 and RLKM-19 are two spreadsheets compiled from the responses to the questionnaires and they were relied upon by PCI for complying with paragraph (4) of DJ To’s Order.  

(2)  The issues

11.  Four questions arise for consideration under the Questionnaire application.  They are:

(1)	Whether the questionnaires and the responses had been referred to in RL(5A), RL(6A) and Exhibits RLKM-15 and RLKM-16 within the meaning of Order 24 rule 10A;
(2)	Whether the questionnaires and responses are relevant to these proceedings;
(3)	Whether the questionnaires and responses are subject to litigation privilege and, if so, whether the privilege had been waived by PCI; and
(4)	Whether the disclosure of the questionnaires and responses is necessary for the fair disposal of the cause or matter in these proceedings.

(3)  Reference

12.  On the first question of “reference”, what is required is a direct

allusion in the pleadings or affidavits to the documents in issue: Dubai Bank v. Galadari [1990] 1 WLR 731 at 738-9.  Given that an exhibit is also part of an affidavit: In re Hinchcliffe [1895] 1 Ch 117, at 120, reference to documents for the purpose of Order 24 rule 10 can include reference contained in an exhibit to an affidavit: Shun Kai Finance Co Ltd v. Japan Leasing (HK) Ltd (In liquidation) [2001] 1 HKC 636, at 641.  Order 24 rule 10 had been applied to cover affidavits made under compulsion and by way of interrogatories: Moore v. Peachey [1891] 2QB 707; the purpose of the affidavit is therefore irrelevant: see the discussions in Zida Technologies Ltd v. Tiga Technologies Ltd [2001] 3HKLRD 698 at 713-4 para. 35.

13.  There can be no doubt that the questionnaires had been referred to in RL(5A) and RL(6A), within the meaning of Order 24 rule 10.  As to the responses to the questionnaires, upon a fair reading of paragraph 5 of RL(6A) and reading it in the context, the deponent must be referring to the replies PCI received from the agents consequential upon being sent the questionnaires. I do not agree with the argument that it is not a reference to documents, but is a reference to an event or conduct.  In my view, there are direct allusions to the questionnaires and responses in the affirmations.  That being the case, whether the direct allusions or references could have been avoided by alternative drafting is irrelevant.

(4)  Relevance

14.  As to the second question of relevance, it is not in dispute that PCI had resorted to the questionnaires and responses in carrying out its delivery up and disclosure obligations under DJ To’s Order, and that RLKM-15 and RLKM-19 were compiled with the aid of the responses to the questionnaires.  The documents sought are clearly relevant to the subject matter of the disputes between the parties in these proceedings.

(5)  Litigation privilege        

15.  The third question of privilege forms the main thrust of PCI’s opposition to the Questionnaire application.  It is PCI’s contention that the responses to the questionnaires are covered either by litigation privilege or common interest privilege.  AXA, while disputing that the documents are privileged, says that even if privilege subsisted, the privilege has been waived through the deployment of the documents in the affirmations of Ralph Lau.  These issues will be dealt with in turns.

16.  An explanation of the doctrine of litigation privilege is to be found in the dissenting judgment of Barwick CJ in Grant v. Downs (1976) 135 CLR 674 at 677, which was approved by the House of Lords in Waugh v. British Railways Board [1980] AC 521 at 532G-533D, 537G-H & 543H-544B as follows:

“[A] document which was produced or brought into existence either with the dominant purpose of its author, or of the person or authority under whose direction, whether particular or general, it was produced or brought into existence, of using it or its contents in order to obtain legal advice or to conduct or aid in the conduct of  litigation, at the time of its production in reasonable prospect, should be privileged and excluded from inspection.”

17.  The burden is on the party asserting privilege to establish that the documents or information in issue is privileged.  It is not necessary that they should have come into existence at the instance of the lawyer.  It is sufficient if they came into existence at the instance of the party himself: Buttes Oil v. Hammer [1981] 1 QB 223 at 234H.  Litigation privilege will attach to confidential communication between a party to a litigation or his legal adviser and third parties: Ventouris v. Mountain [1991] 1 WLR 607 at 611F, per Bingham LJ, and may cover information given by an employee to an employer or by an agent to a principal: Three Rivers District Council v. Bank of England (No.5) [2003] QB 1556 at 1568D. 

18.  The bone of contention is whether the responses can be said to have been brought into existence for the dominant purpose of the conduct or aiding in the conduct of the litigation.  For AXA, it is argued that the rationale behind the doctrine of litigation privilege is the inviolability of the confidential communications between a party and his legal advisers, so that documents that might cast light on the instructions to lawyer or legal advice given to the party regarding the conduct of the case will be privileged from production.  There is thus no general privilege over documents created for the purpose of litigation and the privilege is limited to the furtherance and preparation of a party’s own case.  It is said that the responses to the questionnaires were for the dominant purpose of giving disclosure to the AXA in compliance with DJ To’s Order.  Production of the responses will not violate the confidence between PCI and its legal advisers nor will it reveal any legal advice given to PCI.

19.   AXA draws support from the judgment of Sir Richard Scott VC in Re Barings plc,Secretary of State for Trade and Industry v. Baker [1998] CH 356 at 366C, where it is stated that:

“These citations make clear, in my opinion, that documents brought into being by solicitors for the purposes of litigation were afforded privilege because of the light they might cast  on the client’s instructions to the solicitor or the solicitors’ advice to the client regarding the conduct of the case or on client’s prospects.  There was no general privilege that attached to documents brought into existence for the purposes of litigation independent of the need to keep inviolate communications between client and legal adviser.  If documents for which privilege was sought did not relate in some fashion to communications between client and legal adviser, there was no element of public interest that could override the ordinary rights of discovery and no privilege.”

It is also pointed out that Scott VC’s views had been approved by the English Court of Appeal in Visx Inc v. Nidex Co [1999] FSR 91 at 104-6, 109-111 and also by Lawrence Collin J in ISTIL Group Inc. v. Zahoor [2003] 2 All ER 252 at 262. 

20.  PCI on the other hand argues that the law and authorities are settled in favour of a wider doctrine so that litigation privilege exists if communications or documents are brought into existence for the purpose of litigation.    Reliance is placed on authorities in which the claim of privilege was upheld in circumstances in which there was no suggestion that disclosure would impinge on confidential communications between client and lawyers: Re Highgrade Traders [1984] BCLC 151, Guinness Peat Properties Ltd v. Fitzroy Robinson Partnership [1987] 1 WLR 1027 and Robert Hichins Ltd v. International Computers Ltd (unreported) 10.12.1996 Lexis Transcript.   Reference is also made to several academic writings which point out the other authorities do not support Scott VC’s view in Re Barings and have not limited the privilege to cases where there is a need to protect lawyer/client communication: Matthews & Malek, Disclosure (2001) para.9.028, Hollander on Documentary Evidence (8^th ed) para.13-37 & 13-38 and Passmore, Privilege, p.86.

21.  It is also pointed out that a litigant in person is in principle also entitled to privilege for communication between himself and third parties that are made for the purpose of obtaining factual information for the preparation of his case in pending or contemplated litigation: see 16^th Report of Law Reform Committee, Cmnd 3472, para.17.  It is argued that this shows that the doctrine of litigation privilege is not necessarily connected to the inviolability of client/ lawyer communications: see also Phipson on Evidence (15^th ed) para.20-32.

22.  Having considered the various authorities cited by counsel for both sides, I agree that the early cases on litigation privilege stemmed from concerns to protect confidential communications between a party and his lawyer.  It was the public interest associated with the inviolability of lawyer/client communication that gave rise to the exception to the ordinary rights of discovery.  Subsequent and modern cases on the topic, however, had not restricted the application of the doctrine to cases where there was a need to protect lawyer/ client communication.  The rationale as developed is that a litigant or prospective litigant should be able to prepare properly for the litigation in the confidence that others thereafter will not be entitled to examine and perhaps profit from the preparatory documentation: Robert Hichins Ltd v. International Computers Ltd, op cit. p.5, per Simon Brown LJ.  This will also explain why litigants in person may also claim privilege over communications with third parties.  While human rights considerations and the increasing emphasis on openness in civil litigation may provide cause for reconsidering and challenging the wider doctrine that has developed, the common law position remains that litigation privilege can be claimed in litigation where the inviolability of lawyer/ client communications are not threatened. 

23.  It follows that in the present case, it is not necessary to show that the responses may impinge upon lawyer/client communication in order that PCI can successfully claim privilege over the responses.  That aside, there is little doubt that the responses were intended to be presented to PCI’s legal advisers for advice on compliance with DJ To’s Order.  Thus viewed, they ought to be covered by the privilege even on a narrow application of the doctrine.

24.  As to AXA’s argument that the dominant purpose for which the documents were brought into existence must be to further the party’s claim or the preparation, it is taking too narrow a view of the doctrine and is not supported by the authorities.  The argument is also incompatible with AXA’s submission that litigation privilege is to uphold the inviolability of lawyer/client communication.  This is because the argument will result in confidential communications with third party with a view to complying with a court order, not being protected by privilege. 

25.  In my view, it is immaterial that the responses were brought into existence for the purpose of complying with the delivery up and disclosure obligations under DJ To’s Order.  The questionnaires and responses were to provide information to enable PCI to determine, with legal advice, what should be done to meet the obligations under the Order.  Deliberations over what to do to comply with an order clearly form part of the conduct of litigation. The questionnaires and responses are therefore aids to conduct of litigation.  Even though the objective underlying the design and issuance of the questionnaires is to make discovery and disclosure pursuant to DJ To’s Order, it does not follow that the questionnaires and responses are not confidential communications.   The fact that PCI had indicated in the affirmations of Ralph Lau the use proposed to be made of the questionnaires and responses does not mean that the confidentiality, which is a matter between PCI and its agents, is lost. In short, there can be little doubt that the dominant, if not the sole, purpose of the questionnaires, and in turn the responses to them, are for the conduct or aiding in the conduct of this litigation.  The questionnaires and responses are therefore subject to litigation privilege.

(6)  Common interest privilege

26.  I turn next to common interest privilege.  In Buttes gas and Oil Co. v. Hammer (No. 3) op cit ,at 267, Brightman LJ said:

“There is a privilege which may be called a ‘common interest’ privilege.  That is a privilege in aid of anticipated litigation in which several persons have a common interest.”  -- per Lord Denning MR at 234

“ … if two parties with a common interest and a common solicitor exchanges information for the dominant purpose of informing each other of the facts, or of the issues, or advice received, or of obtaining legal advice in respect of contemplated or pending litigation, the documents or copies containing that information are privileged from production in the hands of each.” 

27.  Cases decided since Buttes suggests that for common interest privilege to arise, it matters not whether the parties share a common solicitor constitutes common interest: The World Era [1993] 1 Lloyds Rep 363, 366 and Phipson, op cit, para.21-05.

28.  AXA submits that common interest privilege is inapplicable.  It points out only documents that are otherwise by themselves confidential and privileged materials, will be privileged when sent to and are in the hands of a party with a common interest: see Commercial Union Assurance Co plc v. Mander [1996] 2 Lloyd’s Rep 640, and Hollander, op cit, para. 14-09.   In AXA’s contention, the questionnaires, when created in the hands of PCI, and the responses, when created in the hands of the agents, were not privileged for reasons discussed above, they cannot attract common interest privilege.  In view of my conclusion that the questionnaires and responses are covered by litigation privilege, notwithstanding they were created for the purpose of assisting PCI to make discovery and disclosure to AXA pursuant to DJ To’s Order, this argument must fail.  There is further no dispute that PCI and the agents have a common interest in the questionnaires and responses.       

(7)  Waiver of privilege

29.  I deal finally with the issue of waiver.  The general principle is as stated by Mustill J in Nea Karteria Maritime Co Ltd v. Atlantic and Great Lakes Steamship Corp (No.2) [1981]Com LR 138 at 139:

“… where a party is deploying in court materials which would otherwise be privileged, the opposite party and the court must have an opportunity of satisfying themselves that what the party has chosen to release from privilege represents the whole of the material relevant to the issue in question.  To allow an individual item to be plucked out of context would be to risk injustice through its real weight or meaning being misunderstood.”

30.  The key to the issue is whether the questionnaires and responses had been deployed in the affirmations of Ralph Lau.  Although I have found that references had been made to the questionnaires and responses in RL(5A) and RL(6A) for the purpose of Order 24 rule 10, it is not determinative of the issue of waiver.  This is because mere reference is insufficient o waive privilege: Marubeni Corp v. Alafouzos (unreported) 6.11.1986 Lexis Transcript and Bourns Inc v. Ravchem Corp [1999] 3 All ER 154.

31.  The position is as stated in Matthews & Malek, op cit, para.10.17:

“The key word here is ‘deploying’. A mere reference to a privileged document on an affidavit does not of itself amount to a waiver of privilege, and this is so even if the documents referred to is being relied on for some purpose, for reliance in itself is said not to be the test.  Instead, the test is whether the contents of the document are being relied on, rather than its effect.  The problem is acute in cases where the maker of an affidavit or witness statement had to give details of the source of his information and belief, in order to comply with the rules of admissibility of such affidavit or witness statement.  Provided that the maker does not quote the content, or summarise them, but simply refers to the document’s effect, there is apparently no waiver of privilege”:

approved in Dunlop Slazengor v. Joe Bloggs [2003] EWCA Civ.901 at para.11, and Lucas v. Barking, Havering and Redbridge NHS Trust [2004] 1 WLR 220, 226H-228B. 

The important point is whether it is the effect or the contents of the privileged materials that has been referred to or relied upon.  The privilege will be waived if there is reference to and reliance on the contents of the privileged materials.

32.  In the present case, the reference to the questionnaires and the responses in Ralph Lau’s affirmations, in particular RL(5A), are in the context of explaining the steps PCI would take and had taken to comply with DJ To’s Order.   There is no deployment of the questionnaires and responses in the sense intended by law.  In paragraph 5 of  RL(6A), it refers to upon analyzing the responses and non-responses to the Agents’ Questionnaire, it was discovered that there were 47 policy applications by AXA policyholders in which the agents had failed to indicate whether Client Data had been used, and that out of abundance of caution, PCI had decided to deliver up these policy applications.  In my view, this does not amount to a reference or use of the contents of the responses.  As for the exhibits to Ralph Lau’s affirmations, notably RLKM-15, RLKM 16 and RLKM-19, they contain no reference at all to the responses.  There is, however, no doubt that they were compiled with the assistance of the information disclosed in the responses.   But that is not the same as a reference to or reliance on the contents of the responses so as to give rise to a waiver of the privilege.

(8)  Fair disposal of the cause or matter

33.  In view of my conclusion on the third question of issue, the fourth question of whether the discovery sought is required for a fair disposal of the cause or matter in issue can be dealt with briefly.

34.  In summary, AXA says that PCI had made a number of retractions and corrections in the process of complying with DJ To’s Order and they were not properly or fully explained.  They demonstrate that PCI has misinterpreted the meaning of Client Data in that it regards user as a necessary condition, and had also committed errors in reading, classifying and handling the responses.   AXA therefore has cause to believe that PCI has not fully or adequately complied with DJ To’s Order.  The questionnaires and responses are therefore required to ascertain whether there been full and proper compliance of the Order.   

35.  For reasons that will be explained below, the concerns of AXA over PCI’s understanding of the scope of its disclosure obligations under the Order are not unfounded.  I am in agreement that the questionnaires and responses, subject to their being privileged, are required to be disclosed for     the fair disposal of the issue as to whether PCI had complied with DJ To’s Order. 

36.  For the reasons that the questionnaires and responses are subject to privilege that has not been waived, the Questionnaires application fails.  Accordingly, AXA’s summons dated 19 August 2003 is dismissed.

The Non-compliance application and PCI’s application for release

37.  It is convenient to deal with AXA’s summons for an unless order to enforce the Undertaking and paragraphs (3) and (4) of DJ To’s Order and PCI’s applications under its summons for release from further compliance with the Undertaking and DJ To’s Order together. 

The Undertaking applications

38.   AXA’s and PCI’s applications insofar as they relate to the Undertaking raise two broad issues, namely,

(1)	What is the scope of the Undertaking and whether PCI have any further obligations under the Undertaking, and
(2)	If PCI has not fully complied with the Undertaking, whether it should be enforced by an unless order or in a reduced manner, or whether PCI should be released from further compliance with the Undertaking.

(1)  The scope of the Undertaking

39.  On the scope of the Undertaking, PCI’s case is that the Undertaking should be confined to SAMS printouts taken away by the 2^nd to 10^th defendants and were used or intended to be used for the PMS implemented by PCI in August 2000.  PCI argues that this is the proper construction of the Undertaking since AXA’s pleaded claim was confined to the PMS and the Undertaking was to give AXA interim protection pending the adjourned hearing of the interlocutory injunction application.  It is also said that it could not have been intended that PCI was to search all the polices files given that PCI started business in 1994 and it must be that only the “recent documents” were contemplated.  AXA’s view is that PCI’s construction is incompatible with the clear terms of the Undertaking and inconsistent with the conduct of PCI in not raising the distinction before Beeson J or Deputy Judge To, in making no reference to the restriction in its memos to agents and staff and in directing enquiries to agents other than the 2^nd to 10^th defendants and searching non-PMS polices.

40.  The starting point in construing the Undertaking must be the fair meaning of the words used.  The Undertaking is couched in general terms and admits no ambiguity.  It refers to “all printouts from the plaintiffs’ Sales and Marketing System (SAMS) and copies thereof, if any…”.   There is no qualification and is not confined to SAMS used or to be used for the PMS.  Indeed, I do not understand from the submissions made on its behalf that PCI is disputing that by its plain wordings, the Undertaking is not confined to the August 2000 PMS.   The question is whether there are circumstances justifying looking beyond the narrow confines of syntax and grammar to give the Undertaking a limited scope.

41.  The first matter PCI refers to is the pleaded claim of AXA.  But it is clear from the Statement of Claim that was before Beeson J that AXA makes a general claim of ownership of the SAMS information and documents and also the Schedule C documents.  AXA’s claim of breach of confidence is also not restricted to use or misuse of the materials for the PMS.

42.  PCI also refers to the opening submissions of AXA’s leading counsel before Beeson J in which reference was made to the August 2000 PMA and the use of the confidential information in connection with it.  When read in context, however, the reference was made in the context of explaining the urgency and necessity of seeking interim protection from the court.  It cannot be said that AXA was only complaining about the PMS or their claim was so confined.

43.  As to the circumstances under which the Undertaking came to be offered and accepted, there can be no dispute that the relief sought by AXA under the interlocutory application is wider than the Undertaking in terms of the documents and information covered.  The Undertaking only relates to SAMS documents.  It is plain from the submissions made on behalf of PCI as well as those on behalf of the other defendants that they disputed AXA’s entitlement to the wider relief, but they were prepared to accept and did acknowledge AXA’s ownership of the SAMS printouts and to return them to AXA.   Leading counsel for PCI had drawn a clear distinction between the physical documents and the information or data contained in them.  The Undertaking was therefore offered on the basis that PCI and the other defendants had no right to retain the SAMS printouts.   In this connection, it matters not that leading counsel for AXA in his written submissions made a claim on breach of confidence.   The Undertaking was not based on an acceptance of a misuse of the SAMS documents and information. 

44.  Thus viewed, the Undertaking is not in the nature of an interim measure to hold the ring until the adjourned hearing of the interlocutory application.  The point is made plain by the fact that at the adjourned hearing before Deputy Judge To, no argument was raised by PCI to dispute AXA’s rights to the SAMS documents.  On the contrary, leading counsel for PCI informed the Deputy Judge that the Undertaking had been fully complied with, a proposition subsequently shown to be incorrect.   In the circumstances, there is no necessary connection between the Undertaking and the August 2000 PMS.

45.  It is true that at the hearing before Beeson J, the Norwich Pharmacal principle was not a relevant consideration and the Undertaking was not offered and accepted on this basis.  PCI’s argument that even on the Norwich Pharmacal principle, a roving inquiry would not be permissible is in my view irrelevant in view of the basis on which the Undertaking was offered.   The fact that PCI might have to plough through voluminous files does not necessarily mean that AXA was undertaking a roving inquiry.

46.  As to the fact that a seven days’ compliance period was stipulated in the Undertaking, that is not necessarily indicative, and certainly not conclusive, of whether the Undertaking is restricted in scope.  It is to be noted that the basis for the seven days’ period was not apparent from reading the transcript of the hearing.   In any event, the subjective belief and expectation of PCI as to whether it had any SAMS documents is irrelevant to the construction of the Undertaking: Investors Compensation Scheme Ltd v. West Bromwich Building Society [1998] 1 WLR 896, at 913B.

47.  On the other hand, PCI’s conduct was inconsistent with a belief or understanding that the Undertaking had a limited scope.  As pointed out by AXA, PCI had not confined its enquiries and search to the 2^nd to 10^th defendants or to only PMS policies files.  More importantly, when faced with AXA’s complaints of non-compliance of the Undertaking in November 2002, PCI responded by undertaking in the course of 2003 a sampling search of 122,950 non-PMS policies that were issued between 1994 and 2001.  This does not sit well with the suggestion that PCI, when offering the Undertaking, only contemplated recent documents and dealing with August 2000 PMS.  The conduct of PCI is indicative of the views PCI took on the scope of the Undertaking.  In the circumstances, the court would hesitate to attribute a different meaning to it: Intergulf Express HK Ltd v. Delta Asia Ltd [1998] 2HKLRD 929, 935G-H per Litton PJ.

48.  In light of the matters discussed above, there is no room for the submission that on a contextual reading, the Undertaking is limited to PMS policies.  As said by Lord Mustill in Charter reinsurance Co Ltd v. Fagan [1997] AC 313, 388B-C:

“There comes a point at which the court should remind itself that the task is to discover what the parties meant from what they have said, and that to force upon the words a meaning which they cannot fairly bear is to substitute for the bargain actually made one which the court believes could better have been made.  This is an illegitimate role for the court.”

49.  As to whether there is any outstanding obligation under the Undertaking.  PCI’s argument is that, having regard to the objective and purpose of the Undertaking, it had done all that was reasonable.  It is also said that the fact that SAMS documents were missed in the Original Search should not be regarded as a breach of the Undertaking in that trifling things done or omitted which do not cause any mischief should not be treated as a breach, relying on a passage in Halsbury Laws of England (4^th Edition Reissue) vol. 24 para. 1009.  In any case, PCI says that the Undertaking had been superseded by DJ To’s Order.  The main thrust of PCI’s argument is that the Undertaking was an interim relief to protect AXA pending the adjourned hearing of the interlocutory application such that the Undertaking became lapsed upon the making of the interlocutory order providing for wider reliefs.

50.  As already noted, the Undertaking is not an interim protective measure to tie over the period before the adjourned hearing.  It is also important to note that the Undertaking requires the performance of an act once and for all.  It contemplates the delivery up of all SAMS printouts and copies to be accomplished within seven days.  In the normal course of events, by the time of the adjourned hearing, the delivery up process should and would have been completed.  There is therefore no question of a lapse of the Undertaking, and it is inapt to view the Undertaking as being superseded by an order made at the adjourned hearing.  The fact that if PCI were to discover further SAMS printouts after the seven days’ period, it would be obliged by the Undertaking to deliver up to AXA does not alter the position and the nature of the Undertaking.  The delivery up in such instance is not so much because of PCI being under a continuous obligation to deliver up SAMS documents, but rather because the Undertaking has not been fully performed and PCI had to remedy the position.  To the extent that DJ To’s Order has granted wider reliefs, it only means that PCI is put to obligations additional to or beyond those under the Undertaking.

51.  On the submission that PCI had done all that was reasonable and it should not be regarded as being in breach of the Undertaking, PCI has produced a summary of the policies searched for SAMS documents.  It suggests that some 200,000 policies, involving about 2,000,000 pages of documents, had been gone into.  All the policies had been searched except the Non-PMS policies.  Out of the Non-PMS policies, all the WWP NB policies had been searched.  In respect of the Non-WWP NB policies, all those issued between 21 July and 16 October 2000 had been searched.  As for the pre-21 July 2000 Non-WWP NB policies, they had been searched on a sampling basis and some 122,950 policies had been covered.  As from 16 October 2000, PCI had set up procedure to screen out SAMS documents in relation to the Non-WWP NB policies.  AXA’s has commented that for the AXA PMS policies issued up to 31 July 2000, the soft copy of 169 policies had not been given to AXA.  The soft copy of SAMS found in the sampling search of Non-WWP NB policies have also not been handed in.  AXA further points out that the screening procedure for post-October 2000 Non-WWP NB policies has proved to be unreliable. 

52.  In my view, although PCI had incurred considerable time and resources to search for SAMS documents, it remains a fact that the delivery up has not been completed.  The failing cannot be said to be trifling and/or had caused no mischief to AXA who have a proprietary claim over the documents.  It is not open to PCI to contend that there is no further obligation under the Undertaking.         

53.   This brings me to the second issue of whether PCI should be released from the Undertaking or whether the Undertaking should be enforced. On the relevant principles, there is no dispute that the court does have jurisdiction to release a party from his undertaking which was proved to have been given under mistake: Mullins v. Howell (1879) 1 Ch D 763, 766 and Cutler v. Wandsworth Stadium Ltd [1945] 1 All ER 103.  The court also has power to refrain from enforcing an undertaking against the party who has given it: Purcell v. Trigell [1971] 1 QB 358, 363-4 & 366-7. 

54.  In contending that it should be released from the Undertaking, PCI relies on the purpose of the Undertaking and the circumstances under which it came to be given, its belief and expectation of the scope of the Undertaking and the substantial efforts taken to comply with the Undertaking.  It is said that any further actions would be disproportionately onerous and unlikely to yield any substantial benefits.

55.  For the discussions and reasons mentioned above, it cannot be said that the Undertaking was to allay the concern over the implementation of the August 2000 PMS and that its purpose had been served with the suspension of the August 2000 PMS.  The suggestion that PCI was under a mistaken appreciation of the scope of the Undertaking and the tasks entailed by it is also not borne out by the evidence before the court and PCI’s conduct.  I accept that PCI must have put in considerable time and resources to search for SAMS documents, but as pointed out by AXA, the events since the giving of the Undertaking do show that the compliance has not been satisfactory.   For instance, contrary to Mr Ralph Lau’s expectation, SAMS documents were found in Non-PMS policies.  There is force in AXA’s criticism of PCI’s methodology for and approach towards identifying SAMS documents.

56.  That said, however, the question of whether to enforce or discharge an undertaking ultimately is a matter of judicial discretion, designed to achieve a fair result: Shepherd Homes Ltd. v. Sandham [1971] 1 Ch 341. In my view, in a case like the present, the court ought not to lose sight of the fact that the action is at an interlocutory stage whereas the primary objective of the parties and the court must be the final adjudication and resolution of the claim.  Plainly on the evidence, both PCI and AXA had devoted substantial time and resources on issues of compliance with the Undertaking and DJ To’s Order.   Mr Yu SC has argued forcefully on the risks of the Undertaking and the interlocutory orders turning into a monster taking over the real disputes between the parties.  Quite apart from the fact that protracted interlocutory proceedings is never in the parties’ best interest, what needs to be considered is whether dwelling further on the issues of compliance will bring any substantial benefits to the parties in terms of the final resolution of the disputes in this litigation.  In my view, it will not, given that there has been considerable, though not complete, compliance by PCI.  It should also be noted that events have obviously moved on since 2000 when the PMS was to be implemented. 

57.  AXA has submitted that they should not be left without a remedy in view of the non-compliance of the Undertaking.  It is pointed out that in Jordan v. Norfolk County Council [1994] 1 WLR 1353 and Charrington v. Simons Ltd [1970] 1WLR 732, the courts did not leave the plaintiffs without a remedy when granting the release sought and substituted orders were made.  As accepted by Mr Yu SC, it is open to the court to impose a restricted undertaking in replacement of the Undertaking. 

58.  In my view, having regard to the reasons of Deputy Judge To and the adoption of the cut-off date of 21 July 2000 by him and the developments since the hearing before Beeson J, it will now be fair and appropriate to similarly limit the scope of PCI’s obligations with regard to the SAMS documents to the date of 21 July 2000.   PCI should therefore be required to further comply with the Undertaking by conducting a search of all the non-PMS policies issued on or after 21 July 2000, to the extent this was hitherto not done, with a view to delivering up to AXA’s solicitors printouts from SAMS and copies thereof, if any, that are in its possession, custody or power, and that had hitherto not been delivered up to AXA.

59.  It follows that it is not necessary to consider the making of an Unless order.  For completeness however, I will deal with it briefly.  As the authorities show, Unless order is an order of last resort and should not be made unless there is a history of failure to comply.  It is a party’s last chance to put his case in order: Hytech Information Systems Ltd v. Coventry City Council [1977] 1 WLR 1666 cited with approval in Lessy SARL v. Pacific Star Development Ltd [1997] 3 HKC 306 at 311-2.  I accept that in principle, an Unless order can be imposed where the default is in relation to a matter that would not deprive the other side of a fair trial.  I am however of the view that the sanction that accompanies an Unless order must be commensurate with the default in question and meet the justice of the case: see for example the approach in Kai Yip Air-conditioning Engineering Co v. Ma Hei Sun [2001] 3 HKC 458. 

60.  In the present case, although the process of compliance has been a lengthy and dilapidated one, I am not prepared to conclude that PCI has been guilty of contumelious and wilful breach of the Undertaking.  Further, the sanction of striking out the Defence appears to me to be inappropriate.  In this regard, it is open to AXA to bring committal proceedings on the breach and even if PCI were found guilty of contempt, it would not necessarily be denied the opportunity and right to defend the claim.   In short, I do not consider it is a fair exercise of the discretion to impose an Unless order as proposed by AXA, based upon PCI’s failure to fully comply with the Undertaking.             

Paragraphs (3) and (4) of DJ To’s Order

61.  I move on to the applications for the enforcement or release of paragraphs (3) and (4) of DJ To’s Order.  Broadly speaking, two issues fall to be determined. They are:

(1)	The meaning of Client Data; and
(2)	Whether the discovery and disclosure obligations under the Order should be further enforced by an Unless order or whether PCI should be released from further compliance with them.

62.  The main contention on the meaning of Client Data is whether it embodies the concept of user.  No issue arises as to the need to search for and deliver up the non-SAMS documents of white cards, underwriting notifications and daily consolidated reports as PCI accepts its obligations extend to these documents and that the failure to search for and deliver them was an error on its part.  On the other hand, AXA does not dispute that documents and information, if provided by the policyholders themselves, do not come within the scope of disclosable Client Data under DJ To’s Order.  Thus the only dispute is whether the use of the documents or information is a requirement for Client Data. 

63.  It is apparent from the questionnaires issued by PCI to the agents that PCI had been focusing on the use of the information and documents by the agents.  Further, the 8^th and 9^th affirmations of Mr Ralph Lau also show that user is a determinative factor when PCI decided on what policy files should be disclosed.  For instance, it was said that where PCI was unable to decide if an agent had used Client Data in relation to a policy, the policy files would be disclosed “out of abundance of caution”.  It was also said that 6 policy files had been disclosed by mistake because the agents involved denied having used Client Data. 

64.  PCI argues that in view of AXA’s claim being based on misuse of confidential information and that the purpose of the Order is to identify wrongdoings, use must be a relevant and important factor.  AXA on the other hand argues that user is not a requirement.  AXA also says that even if use is relevant, the requirement will be fulfilled whenever a document if found on PCI policy files, at least for the purpose of policy application to PCI.  It will have to be disclosed unless it was supplied by the policyholder to the agent or PCI. 

65.  In my view, it is plain from the definition of Client Data in DJ To’s Order that use of the information or documents containing the information is not a necessary ingredient.  Although AXA have made a claim for misuse of confidential information, AXA have also claimed proprietary rights over Client Data.  Further, even though at the hearing before Deputy Judge To, submissions had been made on the question of user, they were in the context of the interlocutory injunction restraining the use of Client Data.

66.  I am inclined to agree with AXA’s point that for document not provided by the policyholder, their inclusion in PCI policy files is indicative of the documents being used by the agent.  In this regard, Mr Yu SC argues that it will not come within the scope of the Order if the policyholder had consented to the agent using the information or document.  I do not agree.  While it is correct that information or document supplied by a policyholder or obtained from AXA with the policyholder’s consent is not caught by the Order, a document that was wrongfully retained and removed by an agent upon leaving AXA remains to be Client Data and falls within the Order notwithstanding the consent of the policyholder to use the document or the information in it. 

67.  As put by Mr Kotewell SC in his reply submissions, the focus is on the identification of Client Data and not the identification of wrongful act.  Thus, the fact that AXA’s interlocutory application was based on the Norwich Pharmacal principle and the purpose of the Order is to enable AXA to identify the wrongdoers does not support reading the concept of user into the meaning of Client Data.

68.    In short, I take the view that PCI’s reference to the use of the information or documents in the identification of Client Data is not justified by the definition of Client Data.  PCI, however, says that the concept of user had not been applied in the search for SAMS and Non-SAMS documents.  In Mr Yu SC’s submission, even if the reference to user is wrongful, AXA has not shown PCI had failed to comply with the Order and the consequences of the breach.  AXA’s case of breach and non-compliance is primarily a matter of inference.  It is argued that with the incorrect understanding of the meaning of Client Data, the test, approach and methodology adopted by PCI for identification of relevant materials must necessarily be flawed, so that the delivery up and disclosure made must be incomplete and inadequate.  I agree with this argument.  It should also not be overlooked that the concept of user had clearly featured in the questionnaires sent to the agents.  It is also evident from the affirmations of Ralph Lau that the admission or denial of use by the agents involved was an influencing factor in PCI’s decision to disclose a document.  I am therefore unable to agree with PCI’s submission that AXA has not shown PCI to have failed to comply with the Order.

69.  I turn next to the second issue of whether the obligations under the Order should be enforced by an Unless order or whether PCI should be released from further compliance with them.  In the course of the hearing, PCI accepted that the terms of its summons, particularly paragraph 6, were too wide.  Mr Yu SC clarified that the release application is confined only to paragraphs (3) and (4) of DJ To’s Order.         An amendment to paragraph 6 of PCI’s summons was proposed.  By the amendment, PCI, accepting that it should carry out further search and make delivery up and disclosure in relation to the white cards, underwriting notifications and daily consolidated reports as well as the 7 policies and 169 policies respectively referred to in Geoff Wong’s 4th affirmation and Ralph Lau’s 13^th affirmation, applies to be released from further compliance with paragraphs (3) and (4) of DJ To’s Order subject to the further search and delivery up and disclosure mentioned above.

70.  At the hearing, PCI also requested AXA to identify and specify documents that, it says should be, but have not been disclosed.  AXA, while not accepting it is under any duty, indicated that it was prepared to do so.  Subsequent to the hearing, AXA had by the affirmation of Ho Wai Shing Steven provided a list of: (a) documents that AXA would not give to policyholders as a matter of course, and (b) documents that are not given to AXA policyholders unless requested for.  The list was provided without prejudice to AXA’s contention that PCI should have delivered up and disclosed all documents containing Client Data. 

71.  As noted above, an Unless order is an order of last resort and should only be made where there is a history of failure to comply with the Rules or court orders.  There is at the same time no dispute that the court has a discretion not to enforce strict compliance of an order.  Hence in deciding whether to compel further compliance of the Order, including the imposition of an unless order, or to release PCI from further compliance of the Order, the court has to look at all the circumstances of the case so as to arrive at a fair and just decision.

72.  In the present case, PCI relies principally on the following matters as negating an Unless order and justifying a release from further compliance with the Order:

(1)	DJ To’s Order was to protect AXA in the interim period before trial and the merits of the claim have yet to be adjudicated upon,
(2)	There had been substantial compliance by PCI and the non-compliance was not contumelious and there was no intention to flout the Order, and
(3)	AXA had delayed in making the complaint about PCI’s misunderstanding of the term Client Data.

73.  As previously observed, PCI had devoted considerable time and resources to the discovery and disclosure obligations.  I am also not prepared to conclude on the evidence that PCI had deliberately flouted its obligations to make delivery and disclosure.  Although the process has taken a very long time, that is not necessarily equated with a history of failure to comply with court orders.  On the contrary, it appears from the correspondence that PCI had been responsive on occasions when failings and non-compliance were identified by AXA.  It is not to be overlooked that the scope of search under DJ To’s Order is not insubstantial.  The mistaken view PCI took on the meaning of Client Data is regrettable.  However, PCI is correct to point out that it must be apparent to AXA since as early as November 2001 that PCI was focusing on user, but AXA only complained in June 2003.  No doubt, compliance with court order is a matter for PCI and the duty must be on PCI to see to it that it fully and properly complies with its obligations under the Order.  But in considering whether to impose an Unless order, the fact that the objection was raised late and after considerable time and resources had already been spent on complying with the Order is a relevant consideration. 

74.  Having regard to the considerations mentioned above, it will not be a fair exercise of the court’s discretion to impose an Unless order that carries the sanction of striking out the Defence.  It also cannot be said that without strict compliance with the Order, AXA runs the risk of being deprived a fair trial.  In my view, the appropriate course is to order further compliance of the delivery up and disclosure obligations by PCI, but restricted only to the aspects set out in the amended paragraph 6 of PCI’s summons.  With the completion of these further searches, discovery and disclosure, the objectives and purposes underlying paragraphs (3) and (4) of the Order should have been substantially achieved.  After all, DJ To’s Order is an interlocutory order and the parties’ ultimate objective must be the final resolution of the claim.  The engagement in protracted interlocutory proceedings will only divert the parties’ attention, energy and resources from the resolution of the real disputes between them.

Cross-Undertaking as to damages

75.  Turning finally to PCI’s application under paragraphs 1 and 2(a) of its summons to amend Beeson J’s Order and DJ To’s Order by inserting a cross- undertaking as to damages.  AXA is prepared to give the usual undertaking as to damages.  In my view, that is sufficient.  The addition of a reference to costs for complying with the Undertaking or the Order, as sought by PCI in the summons, is not called for. 

Conclusion

76.  For the reasons indicated, 

(A)	AXA’s summons dated 19August 2003 is dismissed.
(B)	On AXA’s summons dated 17 December 2003:
	(1)	PCI is to further comply with the Undertaking by conducting a search of all the non-PMS policies issued on or after 21 July 2000, to the extent this was hitherto  not done, with a view to delivering up to AXA’s solicitors printouts from SAMS and copies thereof, if any, that are in its possession, custody or power, and that had hitherto not been delivered up to AXA.
	(2)	PCI is to further to comply with paragraphs (3) and (4) of DJ To’s Order by carrying out further search, delivering up and disclosure of (a) white cards, underwriting notifications and daily consolidated reports, (b) the 7 policies and 169 policies respectively referred to in Geoff Wong’s 4th affirmation and Ralph Lau’s 13th affirmation, and (c) the list of documents set out in the affirmation of Ho Wai Shing Steven.
(C)	On PCI’s summons dated 13 May 2004, Beeson J’s Order and DJ To’s Order be further amended by inserting the usual undertaking as to damages.  The further amended orders are to be filed and served within 14 days from the date of this Decision being handed down.

67. I will invite the parties to endeavour to agree on: (1) the timetable for the further searches, delivery up and disclosure to be made pursuant to the Undertaking and DJ To’s Order, (2) the costs of the summonses, and (3) the precise terms of the Order.  In the event agreement cannot be reached on one or more of these matters, the parties are at liberty to apply for further directions either by way of a short hearing or if both parties agree, by way of written submissions.

Mr Robert Kotewall SC and Mr Stewart Wong instructed by Messrs Herbert Smith for the plaintiffs.

Mr Benjamin Yu SC and Mr David Stokes instructed by Messrs Richards Butler for the 1^st defendant.

HCA 9093/2000

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF FIRST INSTANCE

ACTION NO. 9093 OF 2000

____________

____________

______________

D E C I S I O N

______________

Background

1. On 24 July 2001, I granted an interlocutory injunction order against PCI and the Individual Defendants restraining them from disclosing, divulging or otherwise making use of AXA's Client Data and ordering them to deliver up certain documents containing Client Data and to make certain disclosures. On 8 August 2001, PCI filed an application for a stay pending appeal. The application for stay was set down before me for 15 minutes on 9 August 2001. At that hearing, I adjourned the application for argument and gave PCI extension of time until 30 September 2001 or the hearing of the application for stay, whichever the earlier. I indicated to PCI the need for urgency and that the intervening period of time between then and the hearing will be taken into account in any extension of time for compliance to be granted.

2. The hearing of the application was listed before me on 6 September 2001. Due to the very late production by PCI of "Points for Clarification" which AXA required time for consideration, the application was then adjourned to 6 November 2001 for further argument with one more day reserved. At the resumed hearing, I refused PCI's application for stay, but made some variation to the interlocutory injunction order and granted further extension of time for compliance. Written reasons for my decision was delivered on 16 November 2001. The issue of costs was reserved. The present hearing is to determine certain outstanding matters which the parties could not resolve among themselves and the issue of costs of PCI's application for stay and clarification.

Costs of hearing on 6 September 2001

3. I granted the interlocutory injunction order on 24 July 2001 and indicated to PCI the need for urgency in taking steps to comply with my order and that the intervening time between their application for stay and the hearing of the application will be taken into account in determining the extension of time to be granted. On the late afternoon of 4 September 2001, less than 48 hours before the hearing, PCI gave AXA 8 pages of "Points for Clarification" in draft. Then on the afternoon of 5 September 2001, PCI sent AXA an 11 page revised version of their "Points for Clarification". PCI's skeleton argument was only given to AXA's in-house lawyer just before the hearing on 6 September 2001.

4. The original application was an application for stay or extension of time. The clarification now tagged onto the application for stay has substantially changed the nature of the application. While Mr Fung SC was ready to proceed with the application on 6 September 2001, AXA naturally needed time to respond to the new matters raised and had to ask for an adjournment. In any event, the time reserved for hearing the application for stay would not be adequate for dealing with the "Points for Clarification." The interlocutory injunction was granted on 24 July 2001. PCI has been warned of the need for urgency in compliance with my order on 9 August 2001. Yet it was not until less than 48 hours of the hearing of the application for stay that clarification in draft form was first sought and PCI's skeleton argument was thrust onto AXA just before the hearing. If PCI were earnest in seeking clarification in order to enable compliance, they would have done so much earlier and not within less than 48 hours of the hearing of their application for stay. As I have indicated in my reasons for decision handed down on 16 November 2001, some of the scenarios raised by PCI are unreal and exaggerated and many have been satisfactorily answered. The delay in seeking clarification at the eleventh hour and the fact that many of the scenarios raised by PCI are unreal and exaggerated suggest to me that PCI was not at all earnest in complying with my order. PCI's conduct has put AXA in such a position as to make an adjournment inevitable. To mark the Court's disapproval of such conduct, it is appropriate that PCI shall pay AXA's costs of the adjournment on 6 September 2001 on an indemnity basis.

Costs of hearing on 6 to 8 November 2001

5. Some variation and clarification have been made. Mr Scott SC argues that PCI has obtained a significant measure of success and should not be penalised in costs. In my view, PCI was the major loser. It was unsuccessful in the application for stay. A substantial part of the variation and clarification sought has been obtained by agreement or satisfactorily answered. I have exempted Client Data in respect of ex-AXA policyholders who have joined PCI from the application of my order so that PCI may service their policies. This issue has not been raised by counsel on either side before. I have also amended the cut off date or the start date in paragraph 4(b), (c) and (d) of my order. Had PCI properly addressed me on the volume of the work involved at the hearing of the injunction in March or April last year, I would have set a less onerous start date. The amendment, I must say, was the result of indulgence on my part because I did not wish to make an order which cannot be complied with in view of the tight timing now left available by reason of PCI's delay. If costs have to be determined at this stage rather than deferred to the conclusion of the proceedings, I would consider PCI should pay AXA's costs of the application for stay and clarification including the costs of the hearing on 6 to 8 November 2001 and today.

6. In granting the interlocutory injunction on 24 July 2001, I made the unusual order that PCI shall pay AXA's costs because of the course adopted by PCI (see my reasons for decision handed down on 22 November 2001). Had PCI confined their defence within the American Cyanamid framework, costs in the cause would have been appropriate. Similarly, had PCI not sought a stay but only sought clarification and alteration, then it could be argued that the clarification and alteration were sought for the purpose of holding the ring and costs incurred should be in the cause. However, this is not the case. PCI sought a stay which turned out to be unsuccessful and achieved little success in seeking clarification and variation. The costs incurred were unnecessary. In the circumstances, I consider PCI should pay AXA's costs in any event and not in the cause. A discount of one-quarter of the costs of the hearing on 6 to 8 November 2001 only would be more than generous to give recognition to PCI's partial success.

Costs of hearing on 10 January 2002

7. The costs of the Plaintiffs' application and of the hearing on 10 January 2002 shall be awarded to the Plaintiffs in any event.

Conclusion

8. Accordingly, I order that the costs of and occasioned by the 1st Defendant's summons dated 7 August 2001, including the costs of the hearing on 9 August 2001 and 6 to 8 November 2001 be to the Plaintiffs in any event, with certificate for two counsel, save that the Plaintiffs shall only be entitled to 75% of their costs of the hearing on 6 to 8 November 2001 and that the costs of the hearing on 6 September 2001 be to the Plaintiffs to be taxed on an indemnity basis with certificate for two counsel.

9. The costs of and occasioned by the Plaintiffs' application and of the hearing on 10 January 2002 be to the Plaintiffs in any event, with certificate for two counsel.

Representation:

Mr R G Kotewall, SC and Mr Stewart Wong, instructed by Messrs Herbert Smith, for the Plaintiffs

Mr John Scott, SC, Mr David Stokes and Mr Ling Chun Wai, instructed by Messrs Richards Butler, for the 1st Defendant

HCA 9093/2000

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF FIRST INSTANCE

ACTION NO. 9093 OF 2000

____________

____________

______________

D E C I S I O N

______________

Background:

1. This is an application by the 1st Defendant, PCI, and the 2nd to 10th Defendants, the Individual Defendants, seeking to vary the costs order nisi I made in my Decision handed down on 24 July 2001 granting the Plaintiffs' application for interlocutory relief. Mr Scott SC and Mr Ambrose Ho SC for PCI and the Individual Defendants respectively submit that the appropriate order should be "the Plaintiffs' costs in the cause" which was what the Plaintiffs, AXA, asked for in their summons, or alternatively "costs reserved to the trial judge" or "costs in the cause".

Whether the court may make a costs order beyond what is prayed for in the summons:

2. Mr Scott SC submits that there was no application by the Plaintiffs to amend the summons and the Court should not travel outside the maximum asked for in the summons. I do not consider the Court's power to order costs is in any way fettered by the terms of the summons. Section 52A of the High Court Ordinance gives me, subject to the provisions of the rules of court, full power to determine by whom and to what extent the costs are to be paid. The parties are bound by their pleadings and it is not the function of the court to determine how they should conduct their case. The court may not travel outside the pleadings and force on the plaintiff a case he never intended to pursue and on the defendant a case he never expected to answer. But when it comes to costs, the court has unfettered discretion to make an order which is fair and just in the circumstances. Each case depends on its own facts. Very often, in anticipation that the application will run the usual course, the parties may ask for the usual costs order. If by reason of the conduct of the parties, the matter took an unusual course which renders the usual costs order inappropriate, it is open to the Court to make an order which best serves the interest of justice in the particular case, with or without inviting the parties to make an amendment.

3. With no criticism on counsel, for reasons as given below I do consider the course taken unusual. I was fully aware of the costs order that AXA sought. But after hearing the case and after considering the merits, I do not consider the costs order sought appropriate. In the interest of saving costs, I did not invite AXA to amend the summons as I do not consider it necessary in view of my unfettered discretion. Accordingly, I made an order nisi so that the parties may come back to argue the matter before me and have their views ventilated if they consider necessary.

The legal principle:

4. Both Mr Scott SC and Mr Ho SC refer to Desquenne et Giral UK Ltd v. Richardson [2001] FSR 1. In that case, the English Court of Appeal held where an injunction was granted or continued on the basis of balance of convenience in order to hold the ring until the dispute between the parties could properly be decided at trial, it was inconsistent to say that there were successful or unsuccessful parties for the purpose of the rules relating to costs and the proper order was that the costs be reserved to the trial judge because only then could it be determined which party was successful and which was not. This is a decision after the enactment of the English Civil Procedure Rules, which do not apply to Hong Kong. While I consider the rationale relevant, in the absence of equivalent rules in Hong Kong, I consider the courts in Hong Kong have wider discretion.

5. In Hong Kong, there is no general practice whether a successful plaintiff in an interlocutory injunction shall be granted his costs in the cause or otherwise. The authors in Hong Kong Civil Procedure 2001 at paragraph 29/1/48 write as follows:

"It has for many years been the normal practice for a successful plaintiff granted an interlocutory injunction to be granted his costs in the cause and for a successful defendant to be granted his costs in the cause. But the rationale of that practice is perhaps not clear and the Courts are showing a greater willingness to depart from it."

6. In TKI Ltd v. New Happy Ltd [1995] HKC 551, Nazareth VP held that special reasons are necessary to justify a departure from the usual rule that costs should be in the cause.

7. In Kickers International SA v. Paul Kettle Agencies Limited and Another [1990] FSR 436, Hoffman J considered the different types of costs orders in respect of interlocutory costs before the English Civil Procedure Rules were enacted. At 437-439, he held:

"Mr Wyand accepts that although defendant's costs in the cause is an order commonly made in cases like this, there is no general practice. The court has a complete discretion. No doubt it is desirable to encourage litigants to give up when they realise that a motion is hopeless. But it may be even more desirable to encourage them not to launch such motions in the first place..........

An order for a party's costs to be 'in the cause' is made because the court has not investigated the merits and considers that it would not be fair for him to recover those costs unless he succeeded at the trial. But the fact that the merits have not been investigated, or not fully investigated, does not necessarily mean that the court cannot dispose finally of interlocutory costs. The question cannot be settled simply by reference to practice: it must be answered on the facts of the individual case. Would it be unfair in this case for the defendants to have the costs of the motion even if they lost at the trial? On balance I do not think that it would. Although I cannot decide the merits, I can form a view on whether the plaintiff, on the material it has produced and what it must have known about the defendants, was justified in launching the motion. I do not think it was. Therefore whatever the outcome of the trial, the motion which has now been abandoned has put the defendants to a great deal of unnecessary expense.

One solution sometimes adopted is to reserve the costs of the motion to the trial judge. There is a superficial attraction about this course because the trial judge will have all the facts and be in the best position to decide whether a party should have interlocutory costs despite the fact that he has lost at the trial. But in practice it has serious disadvantages. One is the difficulty of reconstructing for the trial judge how things looked at the time of the interlocutory application, particularly when it involved questions of balance of convenience which were irrelevant at the trial. But the more important reason is that very often there is no trial. In such a case, an order that costs be in the cause or reserved to the trial judge is in practice a decision that there should be no order as to costs. Not many defendants, even with the encouragement of success at the interlocutory stage, are willing to insist on a trial for the purpose of recovering their interlocutory costs. But unless they do, the costs are left in limbo, as in Stratford v Lindley (No. 2) [1969] 1 WLR 1547. This gives the unsuccessful plaintiff a powerful bargaining weapon. He can face the defendant with the alternative to accepting a substantial discount on the interlocutory costs or having to incur the trouble, expense and risk of a trial. Furthermore, the existence of an order for substantial costs contingent upon the outcome of the trial, or reserved to the trial judge, may push the parties into a trial of an action which would otherwise have settled.

Thus it seems to me that in making a costs order at the interlocutory stage, there are two risks of injustice which have to be balanced against each other. On the one hand, a final order might award a party costs which, upon fuller consideration at the trial, he would not have been given. On the other hand, failure to make a final order might have the practical effect of depriving a party of some or all of the costs which he should in fairness have recovered. I think it is possible that in the past the courts may have been more impressed by the first risk than by the second.

The plaintiff says that it is not abandoning the action and that it intends to pursue a claim for damages to trial. Nevertheless, I think I should take into account first, possibility that it may change its mind and secondly, the effect that a contingent order for costs would have upon any settlement negotiations. Having regard to all these matters, I think that the proper exercise of the discretion is to deal finally with the costs now. I shall therefore order the plaintiff to pay the defendants' costs in any event."

Probably, the above represents the rationale for the Hong Kong courts' departure from the practice of making the usual order of costs in the cause. I, for one, adopt the rationale in Kickers International SA v. Paul Kettle Agencies Limited and Another.

8. There is no rule of law that interlocutory costs should be in the cause. If the injunction was granted without contest or for the purpose of holding the ring, then cost in the cause would certainly be appropriate. The type of costs order to be made depends on the facts of the particular case. In essence, there are two questions to be considered where the plaintiff succeeds in the interlocutory application. The first question is whether it was justified for the defendant to resist the application. If "not", secondly, whether it would be unfair for the plaintiff to be awarded interlocutory costs unless he succeeds also at trial. Or, put it in another way, whether it would be fair for the plaintiff to be awarded his costs even if he fails at trial.

9. Much reliance has been placed by counsel for PCI and the Individual Defendants on the recent English decisions in Desquenne et Giral UK Ltd v. Richardson and Picnic at Ascot v. Kalus Derigs. However, the English Civil Procedure Rules do not apply in Hong Kong and there is no rule of law in Hong Kong that a plaintiff may only get his interlocutory costs if he also succeeds at trial. Mr Scott SC also argues on the authority of TKI Ltd v. New Happy Ltd that good reasons are required to justify a departure from the usual order. In my view, existence of good reasons is one particular facet of fairness or unfairness to be considered. I have certainly taken that as well as the rationale of the English Court of Appeal in Desquenne into account when making my costs order nisi.

The peculiar background in this case:

10. This case is of particular importance to PCI. The interlocutory relief sought has serious impact on its business and reputation. AXA claim their damages are in the region of $500 million. PCI's and the Individual Defendants' would be no less, if not more in view of the large number of defendants being continuously added onto the list of defendants. PCI launched a policy matching scheme ("PMS") targeted at AXA's policyholders and recruited ex-AXA agents by paying them hefty welcoming bonus so much that it announced a profits warning. In view of the massive discovery involved, it is not likely that the trial will come up within a year or two. If these agents recruited are restrained during the period pending trial, not only that PCI's investment in these agents will be lost, it suffers enormous loss in business and reputation as well. It is therefore understandable that PCI would wish to resist the application and put in all the ammunition it has now rather than later. Indeed in its holding company's annual report for 2000, the company said the group will contest the action vigorously.

11. The affidavits filed on behalf of AXA and the law, i.e. the threshold required by American Cyanamid are all in favour of AXA. It is well settled that it is no part of the courts' function at this stage of the litigation to try to resolve conflicts of evidence on affidavits as to facts on which the claims of either party may ultimately depend nor to decide difficult questions of law which call for detailed argument and mature consideration: per Lord Diplock in American Cyanamid [1975] AC 396 at 407. Brown LJ held in Alfred Dunhill Ltd v. Sunoptic S.A. [1979] FSR 337 at 373 that it is irrelevant whether the court thinks that the plaintiff's chances of success in establishing liability are 90 percent or 20 percent. But this does not mean the defendant may not resist the application on the strength of his case and the weakness of his opponent's. Thus when holding that the plaintiff's prospects of success are not to be weighed against his prospects of failure, Megarry VC held in Mothercare Ltd v. Robson Books Ltd [1979] FSR 466 at 474 that "Odds against success no longer defeat the plaintiff, unless they are so strong that the plaintiff can have no expectation of success, but only a hope". It is therefore open to the defendant to challenge the plaintiff's case as having no expectation but only a hope of success. For that purpose, they may raise all issues to show that AXA have no expectation of success. In view of the severity of the injunction if granted, they may put into the weighing pan the strength of their defence and the weakness of AXA's case as well as the balance of risk of injustice in granting the injunction. Apparently, this was what PCI and the Individual Defendants were attempting to do in this case. While I consider the Defendants were entitled to take that unusual course, they will face the consequence of costs for travelling outside the American Cyanamid framework should they fail.

Reasons for the costs order:

12. AXA's basis of claim against the Individual Defendants are their breach of implied duty of fidelity in wrongfully removing confidential information belonging to AXA during the currency of their agency with AXA for use after the termination of their agency and to AXA's detriment. Against PCI, AXA's course of action is based on breach of confidence in that PCI received confidential information from the Individual Defendants with knowledge of their breach of duty of fidelity and intending to use or disclose the information without AXA's permission; and unlawful interference with AXA's business through the wrongful use of confidential information and misrepresentation made by PCI's own agents in marketing the PMS targeted at AXA's policyholders.

13. As can be seen from my Decision of 24 July 2001, the Defendants argued on every possible issue: whether the information contained in SAMS belonged to the Plaintiff, whether agents may use information obtained from alternative sources, whether the information are trade secret or confidential information equivalent to trade secrets. PCI went even further. It argued whether the information was imparted to PCI in circumstances importing an obligation of confidence and whether PCI had been dishonest. PCI challenged each and every element constituting the unlawful means for the purpose of the tort of unlawful interference. In the ultimate balancing exercise, the Defendants put in the weighing pan everything they could, including the strength of their case and weakness of AXA's, appropriateness of the injunction against PCI as a third party, delay, status quo, utility of the Client Data after the lapse of time and clean hands. As for the relief, PCI and the Individual Defendants argued on the term of the injunction, whether the relief to be granted should be a springboard relief and the width of the order. Some of these issues were arguable but others did not have even a flimsy chance of success. The width and depth counsel took in their arguments are remarkable. In the end, all these issues were resolved in favour of AXA for the purpose of the interlocutory application. Mr Kotewall SC refers to the conclusion I reached in my Decision about these issues. I do not find it necessary to refer to them here. They are clearly set out in my Decision.

14. Mr Scott SC refers to various passages in my Decision in which I refrained expressly from giving a conclusive view of the fact and of the law. He therefore submits that the interlocutory injunction was granted on a balance of convenience in order to hold the ring until the dispute between the parties could properly be decided at trial and costs should be reserved to the trial judge after investigation into the merit of the case. I think almost all interlocutory injunctions involve balancing exercise and it is not the function of the court hearing such matter to try to resolve conflicts of evidence on affidavits as to facts nor to decide difficult questions of law. While it would have been wrong for me to make definitive finding of fact on the basis of the affirmations and to decide on the law, that does not mean the materials before me could not enable me to form a view as to where the merit lies in so far as interlocutory costs are concerned. The Defendants put everything in issue. Counsel compared and contrasted affirmations filed by the other side. They addressed me fully on the law. In effect, they tried to put everything in the weighing balance, including the strength of their case and the weakness of AXA's. Though my finding of fact was provisional, the questions of law were very thoroughly argued, and having heard counsel's submission during those 23 days, I am able to say with some conviction what the facts were likely to be and what the legal consequences were on those facts as to enable me to determine the question of interlocutory costs.

15. For reasons as given in my Decision of 24 July 2001, I dismissed the arguments of the Defendants. While I consider it is open to a defendant to resist an application for interlocutory injunction by showing that the plaintiff has no expectation of success and while in the particular position of the Defendants, I consider it understandable for them to take that course; they must run the risk that at the end of the day, they may fail nevertheless and in which event the court may find the course they took unreasonable. This is precisely the position the Defendants now find themselves in. I have found the Defendants' arguments flawed and erroneous. In the light of the American Cyanamid principle, the Defendants' pursuit of all the issues is unreasonable. The Defendants' arguments took up substantial time. AXA's opening took six and half days, while PCI and the Individual Defendants took seven and half days and five and a half days respectively in their response. AXA took another three days in reply. In the light of the American Cyanamid principle, if the Defendants sought to take the course they took and failed, I have to find their conduct of the defence unreasonable and that they have unjustifiably prolonged the proceedings. Unnecessary costs have been incurred. Regardless of the outcome at trial, these unnecessary costs should never have been incurred had the Defendants confined their defence within the American Cyanmid framework. It is unfair that AXA could recover these unnecessary costs only if they are successful at trial.

16. I am also able to conclude that the balance of convenience in this case is so clear and the outcome of the hearing of the interlocutory application is so plain to the parties, that an order should be made against the Defendants for wasting time and money in contesting the application. Had the Defendants confined their defence within the American Cyanamid framework, the grant of injunction would have been for the purpose of holding the ring, in which event, cost in the cause would be appropriate. But as the Defendants have traversed outside that framework of convenience and put everything in issue, then they would have to bear the consequence of costs. The course taken by the Defendant was unusual and, as it turned out, unjustified. In the circumstances, I consider the costs order sought by AXA inappropriate. In the circumstances, it is appropriate that the Defendants shall pay the Plaintiffs' costs irrespective of the outcome at trial.

17. Mr Scott SC and Mr Ho SC argue that this is not a case where the costs should be taxed and made payable forthwith. Mr Kotewall SC argues to the contrary. It is my usual practice to make express order for immediate taxation, if immediate taxation is intended. It was not my intention to order immediate taxation in the present case, nor do I see any need or justification to do so. For the avoidance of doubt, I would amend my costs order by adding the words "in any event".

Conclusion:

18. Accordingly, I affirm my costs order made on 24 July 2001 to the effect that the Defendants shall pay the Plaintiffs' costs with certificate for three counsel in any event, to be taxed, if not agreed.

19. As I have affirmed my costs order nisi, hence, I also make an order that the Defendants shall pay the Plaintiffs' costs of this hearing with certificate for two counsel in any event, to be taxed, if not agreed.

Representation:

Mr Robert Kotewall, SC leading Mr Stewart Wong, instructed by Messrs Herbert Smith, for the Plaintiffs

Mr John Scott, SC leading Mr David Stokes, instructed by Messrs Richards Butler, for the 1st Defendant

Mr Ambrose Ho, SC leading Mr M C Law, instructed by Messrs Hoosenally & Neo, for the 2nd to 10th Defendants

HCA 9093/2000

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF FIRST INSTANCE

ACTION NO. 9093 OF 2000

____________

____________

______________

D E C I S I O N

______________

Background

1. This is an application by the 1st Defendant, PCI, seeking a stay of the interlocutory order I made on 24 July 2001 pending appeal, or failing that, extension of time to comply, clarification and variation of my order. The Order I made:

"(1) restrains PCI (whether by its directors, officers, agents, employees or otherwise howsoever) from

(a) disclosing, divulging or otherwise making any use of the Client Data;

(b) accepting any application for a life insurance policy by persons whose name appears in any of the Client Data and who have been approached by any agent of PCI at a time when any of the Client Data was in the possession or under the control of PCI;

(c) soliciting any business by representing that certain of its policies match or are comparable with the corresponding policies of AXA;

(d) accepting applications for certain life insurance products from policyholders of AXA to whom a representation of the kind referred to in sub-paragraph (c) above has been made.

(2) Restrains the 2nd to 10th Defendants from disclosing, divulging or otherwise making any use of any of the Client Data.

(3) Orders each of the 1st to 10th Defendants to deliver up, within 7 days, all documents containing Client Data, including copies, in its or his possession, custody, power or control.

(4) Orders each of the 1st to 10th Defendants, within 5 days of delivery up under the preceding paragraph, to swear an affidavit:

(a) verifying compliance with the order in paragraph 3;

(b) disclosing all documents containing information relating to holders or former holders of life insurance policies, and/or insured persons thereunder, issued by AXA which came into its or his possession, custody, power or control since 1 November 1999 stating, with reference to each such document, the date when and the name and address of the person from whom it came;

(c) disclosing all documents containing information relating to holders or former holders of life insurance policies, and/or insured persons thereunder, issued by AXA which it or he has given or divulged to another person since 1 November 1999 stating, with reference to each such document, the date when and the name and address of the person to whom it or he gave or divulged the same;

(d) disclosing the name and address of every holder and former holder of life insurance policies, and/or insured persons thereunder, issued by AXA whose name appears in any of the Client Data who has been approached by any agent of PCI since 1 November 1999, identifying the agent, the nature of the solicitation and stating what if any such business came to PCI as a result;

(e) disclosing the names and addresses of all holders and/or former holders of life insurance policies, and/or insured persons thereunder, issued by AXA to whom any agent of PCI represented that certain life insurance products of PCI match or are comparable with the corresponding life insurance product of AXA stating what if any business came to PCI as a result."

2. Mr Scott SC, who now appears on behalf of PCI in place of Mr Daniel Fung SC, comments that despite a common and well-recognised practice of the Court in a complex case, the order was made without affording PCI an opportunity to address me on the precise terms of the Order to be made. Counsel for AXA, Mr Kotewall SC, says he is not aware of any such alleged common or well-recognised practice. Neither do I. The Order that AXA seek has been fully set out by counsel for AXA at the hearing of the application. PCI as well as the Individual Defendants were fully represented and by extremely strong and competent counsel who argued on every arguable point leaving, as I said, no stone unturned and no issue unexplored. All questions about problems of compliance have been addressed. There was no suggestion by counsel that I should invite them to address me on the precise terms of the Order to be made. In view of the depth the matter has taken, I did not consider it necessary. With hindsight, though, there are points about the precise terms of the Order which could and should have been argued. I remain open to persuasion and am prepared to make variation or clarification where the occasion is warranted by genuine difficulties and ambiguities.

Grounds for stay

3. The grounds for PCI's application for stay are: firstly, that the Order is vague, ambiguous, oppressive and, secondly, the Order is excessively and unnecessarily wide. The thrust of PCI's first ground is that unless the Order is stayed, there is a serious possibility that PCI would be exposed, through no fault of its own and notwithstanding the exercise of judgment and acting in good faith on legal advice, to the risk of contempt proceedings. By complying with the order, PCI's ability to meet its obligations to existing policyholders and its statutory obligation under the Insurance Companies Ordinance would be severely impaired. PCI also argues that its life insurance business would suffer devastating and irreparable damage not adequately compensatable under AXA's unfortified undertaking as to damages. Secondly, the Order is excessively and unnecessarily wide as to be impossible of compliance, which would expose PCI to risk of contempt proceedings.

4. Prior to this hearing, the parties have exchanged views on the certainty and width of the Order. In his affirmation filed on behalf of PCI, Mr Lau raised a number of scenarios which he said presented difficulties in compliance with my Order. Some of these scenarios have been raised by counsel in the course of hearing the application and have been dismissed by me as being unreal and exaggerated. The others have, in my view, been satisfactorily answered by AXA.

5. As I have held in paragraph 224 of my Decision, "it may not be possible to have a formula which will work in all factual situations." Whether an order is unclear, ambiguous and impossible to comply should not be adjudged on the basis of a few scenarios. Imagination and ingenuity could always come up with non-existent and unrealistic situations suggesting the order is unclear and incapable of compliance. I would not suggest that my Order is so perfect as to leave no room for improvement; and indeed I would be making some variations, which would render compliance much less onerous.

6. In my view, in considering these scenarios as well as in considering whether an order is vague, ambiguous, oppressive and excessively wide as to be impossible of compliance, the test is to look at the order through the eyes of a reasonable and law abiding citizen who is willing and earnest in complying with an order of the court and ask whether on a fair reading of the order it identifies to that person with sufficient clarity the subject matter being protected and whether the defendant is in a position to know what he is being enjoined from doing. In arriving at his conclusion, he should discard from his mind remote and unrealistic possibilities. He should also discard considerations whether the order casts an onerous burden on the defendant to comply. That is a matter to be considered before granting the order, and which must have been considered in balancing the convenience or the risk of injustice to the parties. This is the test I shall apply.

Vague, ambiguous or oppressive ground

7. I accept the logic contained in Mr Scott SC's grounds for stay, but I am unable to agree with him that the Order is vague, ambiguous or oppressive. All these issues have been extensively argued by counsel during the hearing of the application for interlocutory injunction. In paragraphs 220 to 222 of my Decision, I have found that "Client Data" as now formulated is not exceptionally and unacceptably wide as to be impossible of compliance. I do not shrink from that view. Under the present formulation, Client Data must have been, firstly, obtained or derived by an agent in the course of his agency with AXA, not being information known to the agent prior to such obtaining or derivation and, secondly, it must be contained in SAMS. The data are source-specific and well defined, being derived only from policyholders qua policyholders.

8. Mr Scott SC complains that the Order presupposes PCI has access to SAMS or a comprehensive list of all of the former and current AXA policyholders. He submits that I was wrong to assume in paragraph 250 of my Decision that PCI's sophisticated computerized system would help it to discover what information is contained in AXA's SAMS. I think counsel is reading far too much into one part of my dicta as to ignore the rest of it. I was referring to PCI as "a company of substantial means and with a sophisticated computerized system" that "it should be able to locate all relevant documents without undue difficulties." The sophisticated computerized system is only part of what I considered as PCI's substantial means. Indeed even according to Mr Lau's affirmation filed on behalf of PCI, whether a PCI policyholder is a former or current AXA policyholder could be ascertained without difficulties. In respect of PMS polices, Mr Lau has confirmed that the information is readily ascertainable from its computer. This supports my point. As for non-PMS polices, reference can be made to and reliance placed on answer to question 19 of PCI's policy application forms. This may involve going through all of the non-PMS policy files. This is not a tremendous exercise in view of the substantial means of PCI. Again this also supports what I held in my Decision.

9. Further, in the course of this hearing, at the suggestion of Mr Scott SC, AXA have come up with a protocol which if accepted by PCI would help PCI to identify AXA's policyholders for the purpose of enabling PCI to comply with my Order. I consider the protocol useful and do not see any reason for PCI rejecting it, unless it has more convenient means from within its own system to identify these policyholders. If the protocol is adopted, I can hardly see any oppressiveness in complying with the Order.

10. Mr Scott SC also complains that the Order presupposes PCI or the introducing agent knows or is able to find out whether a particular piece of information was "obtained or derived" not only by himself, but by other AXA agents and whether it was obtained "in the course of his agency" with AXA. This is certainly within the knowledge of the agent. Given the peculiar relationship between the agent and his client, the information is not difficult to ascertain. If the policyholder is not his relative or personal friend, then the information about this policyholder must have been obtained in the course of his agency with AXA. In selling that person a policy, he must have first obtained his information in the course of his agency with AXA. Further, an agent must know the kind of information usually stored in SAMS. Once it is ascertained by the above test or by the agent's own recollection that the information had been obtained in the course of agency, he must be able to say whether the information was obtained or derived from SAMS. PCI being in the insurance business and having been shown printouts from SAMS and the Schedule C documents would likewise have no difficulties in determining these questions.

11. In the end PCI has to rely on its agents to determine whether the data were acquired within or outside the course of their agency with AXA. It may issue questionnaires including a declaration to its agents for the purpose of confirming the information and to impress on the agents the seriousness and need for caution in providing correct information. If PCI has taken all reasonable steps to confirm the position with its agents, it could not be held to be in breach of the Order. That some of the agents may have left PCI and become uncooperative is at best a reason for extension of time or in the extreme case would relieve PCI of part of its liability for non-compliance. Thus the problems suggested by Mr Scott SC are neither insurmountable nor difficult of compliance as to render the Order vague or oppressive.

12. Mr Scott SC argues that an order worded to cover the acts of servants or agents as in paragraph 1 of the Order puts PCI at risk of being held in contempt for the action of agents over whom its control is limited. I think PCI's worry is exaggerated. The scenario as posed by PCI is whether PCI would be in contempt if an agent contacts an AXA policyholder only for social reasons, e.g. to play golf. AXA's reply is that there is no distinction between social and business contact in the Order and the question is whether Client Data is used to initiate the contact. I endorse that "initiation test". However, paragraph 1 of the Order restrains PCI from using the Client Data. PCI is not vicariously liable for whatever its agent does in his social activities unless he does so in the course of PCI's business. Thus, even if the agent has breached paragraph 2 of the Order by using the Client Data to contact the client, for that purpose PCI is not using the Client Data. If the golf game is organised as part of PCI's promotion activity, then the agent's use of Client Data in contacting the client would be attributed to PCI. This conclusion could be easily reached on application of general legal principles. I consider there is no lack of certainty in the Order and no clarification or variation is called for.

13. By applying the above test, I am satisfied, as I was, that AXA have defined the scope of the injunction they seek with sufficient clarity such that the Order is not impossible of compliance. The interesting scenarios raised by Mr Lau in his affirmation have all been satisfactorily replied by AXA or are too remote and unrealistic and the difficulties suggested are exaggerated.

Impact on PCI's business

14. Mr Scott SC submits that absent a stay or significant variation to the Order, the damage to PCI, its agents and customers from the injunctive part of the order, is incalculable and irreversible, even if PCI is ultimately vindicated at trial. He suggests that it would be almost impossible to establish whether the agent or policyholder left PCI because of the injunction or for other reasons; that it would be virtually impossible to measure the number or value of policies which were foregone as a result of the injunction and that the customer connections with the agents may be lost forever. These issues have been fully argued before me at the hearing of the application. I have already considered these matters fully when balancing the convenience and injustice to the parties. I do not shrink from my views as expressed in my Decision and it is not open to the parties to re-litigate on these matters now.

15. Mr Scott SC submits that agents who are not parties to the proceedings would lose out on commissions and forgo customer connections. The injunction does not apply to agents who are not parties to the action. Even if it does apply indirectly through PCI, it is not inequitable if AXA's Client Data are in law entitled to protection. Certainly AXA have a greater interest in protecting their Client Data than the agent has.

16. Mr Scott SC argues that PCI's policyholders would be leaving because of PCI's inability to continue to provide the level of service its policyholders are now receiving; agents may find their working environment so oppressive that they would quit PCI and PCI would suffer loss in commercial reputation and consumer confidence. There is no evidence that the injunction is likely to affect PCI's ability in maintaining its level of service and that agents are leaving. Even if there are such evidence, it is unlikely to change the balance. In the event that PCI succeeds, any damage suffered could be recovered from AXA's undertaking.

17. As for the disclosure order, Mr Scott SC submits that PCI would have to be engaged in a memory-searching exercise with each of its 2,000 existing agents and about another 2,000 former agents who have left PCI since 1 November 1999. He also submits that in some cases, the agents may need to go back to the customers themselves to find out or to confirm details of the "nature of solicitation" or representations. The 2nd to 10th Defendants have no apparent difficulties in complying and I am unable to see why PCI would have. At worst, the difficulties could be solved by employing additional manpower, which PCI would no doubt have the means to and/or ask for extension of time to comply. If at the end, the injunction should not have been granted, PCI may look to AXA's undertaking for damages.

Excessive and unnecessary width of the Order

18. Mr Scott SC submits that the confidential information that AXA claim to have been misused is to be defined in their pleading by reference to documents allegedly taken by the Individual Defendants and the information contained in those documents, but "Client Data" does not appear in the body of the pleading. Hence, he argues that the relief sought by reference to "Client Data" is much wider than AXA's pleaded case and is not supported by the pleadings. I think this argument is based on the wrong premise that AXA's claim is restricted to documents taken away by the 2nd to 28th Defendants and the information contained therein. In fact, this is not. AXA have made a general claim to the secret and confidential nature of information contained in SAMS and misuse of the information by PCI and its agents in the implementation of the PMS. "Client Data" as referred to in the prayer is a convenient short hand description of what had been scaled down from the secret and confidential information described by AXA in the pleadings. I do not consider the relief sought by AXA is wider than their pleaded case.

19. Mr Scott SC argues that AXA are seeking to prevent the use of information allegedly taken by the 2nd to the 28th named defendants but evidence was only presented against the 2nd to the 10th Defendants while allegations relating to the 11th to 28th Defendants were not addressed. There is no allegation or evidence that any PCI agents who are not defendants in the proceedings removed Client Data. Hence, Mr Scott SC submits the Order has a wide-ranging effect and suggests to vary the definition of Client Data to limit that to information derived by the 2nd to 10th Defendants only. His argument may have some force as regards the 2nd to 10th Defendants but not as regards PCI. Having seen evidence of such a large scale appropriation of Client Data by the 2nd to 10th Defendants and having regard to PCI's mass recruitment of teams of ex-AXA agents and paying them hefty welcoming bonus, there is a real likelihood that Client Data derived by any AXA agent is at risk. It is unrealistic to restrict Client Data in the way as suggested by Mr Scott SC.

20. Next, Mr Scott SC argues that the documents removed from AXA by their former agents were not created for the purposes of PCI's business and do not perform the function of recording PCI's transactions. He submits that while Clause 9 of PCI's agency contract requires agents to keep books of account and other records for the purpose of showing all his transactions on behalf of PCI and his business pursuant to the agreement, as the documents which AXA now seek from PCI fall outside the above ambit, PCI does not have power over the documents. I think this is all academic. Whether a document is within PCI's power to deliver up is a question of fact. If it is, it should be delivered up; if not, PCI is certainly excused from delivery. No question of contempt proceedings would arise. I think Mr Scott SC's worry is unreal and exaggerated.

Nugatory ground

21. As disclosed by Mr Lau's affirmations filed in support of PCI's application for stay, the PMS has been suspended. That as far as AXA are concerned, paragraph 1 of the Order has been complied with if PCI suspends and does not put into effect the PMS or similar schemes targeted at AXA, and that a proper screening process is put in place.

22. In relation to the delivery up and verification order, i.e. paragraph 3 and 4 of the Order, PCI's grounds for applying for a stay is that the appeal will be rendered nugatory if the execution of paragraphs 3 and 4 of my Order is not stayed because once the documents have been delivered up and information disclosed, they cannot be returned even if the appeal is successful.

23. I must confess for my part that I am quite unable to see why the documents and information delivered up cannot be returned if the appeal is successful. The delivery up is limited to Client Data belonging to AXA and AXA will not be given access to information which they otherwise are not entitled to or do not have.

24. If the information have been "mixed" with other untainted confidential information belonging exclusively to PCI and cannot be physically separated from untainted information, for example, being stored in the same CD-Rom, they could be delivered to the Plaintiffs' solicitors as has been done previously with an undertaking that PCI's confidential information shall not be made available to AXA. As I very much suspect, AXA's Client Data would not have been mixed with PCI's client data except where the ex-AXA policyholders have switched over to PCI. For reasons as I shall give later, the delivery up and verification order do not apply to those Client Data. Thus, the severity of paragraphs 3 and 4 of my Order are substantially mitigated. If for no good reason PCI should "mix" AXA's Client Data in respect of policyholders who have not switched to PCI with their own, then PCI has only itself to blame.

25. If the difficulty is because the process of coming up with the information will take time, then as suggested by AXA's solicitors in their letters dated 2nd and 17th August 2001, the filing of the affidavits required under paragraph 4(b) to (e) of the Order may be delayed until the disposal of the appeal. However, in any event, the documents will have to be disclosed on discovery and exchange of lists of documents has been scheduled to take place in mid November 2001. This will render any argument for stay based on nugatory ground unsustainable. I am not satisfied that compliance with paragraphs 3 and 4 of the Order would render the appeal nugatory or burdensome.

26. Mr Scott SC submits that the 1st Defendant has suspended the PMS further than what the Plaintiffs have sought and with the other 9 Individual Defendants being bound by the injunction order, AXA's interest is adequately protected. I think this is an over simplified view of the action. Mr Scott SC has totally ignored that the mandatory injunction is only part of the remedy sought by AXA. AXA are also seeking the aid under the principle in Norwich Pharmacal Co v Customs and Excise Commissioners [1974] AC 133 to identify likely parties to the wrongful act so that AXA could bring them to justice and have their loss compensated for.

27. Mr Scott SC submits that the benefit to AXA does not justify the onerous burden on PCI. Verifying compliance with paragraphs 1 and 3 of the Order cannot be any real burden, unless there is non-compliance. The burden is only attached to paragraphs (d) and (e). But the burden is not at all onerous as AXA have confirmed that PCI is not required to make any disclosure if no policy has been issued as a result. This would significantly reduce the volume of disclosure required. Where the approach by PCI's agent has resulted in business, AXA are entitled to the disclosure to enable it to bring the wrongdoer to justice and to assess its damages. If PCI should succeed on appeal, the information disclosed could be returned and the expenses incurred compensated for. It could therefore not be said that without a stay the appeal would be rendered nugatory.

Timing

28. Mr Scott SC refers to the fact that the individual defendants were registered as agents of PCI on or after 21 July 2000 and the injunction order was not made until a year later. He submits a stay of my Order until the hearing of the appeal on 30th January 2002 will be inconsequential to AXA's business whereas the effect of the order on PCI's business has been and will be enormous.

29. The suggestion is indeed both convenient and tempting, but in my view, wrong in principle. I have applied the American Cyanamid test and satisfied myself that AXA have raised a serious legal issue to be tried as to whether AXA's Client Data are trade secrets or confidential information amounting to trade secrets as to be protected by the law of confidence. I have directed my mind to the possible result of a postulated trial, first one way and then the other and considered the question of adequacy of damages to AXA and to PCI. I have, in addition, put into the weighing balance considerations such as risk of injustice, appropriateness of injunction against PCI as a third party, the alleged delay in instituting proceedings by AXA, maintenance of status quo and the utility of the Client Data after the lapse of time pending litigation. As I have found after this careful exercise that the balance is in favour of granting the injunction and that damages are inadequate for AXA, there is no reason why AXA should be exposed to the wrongful act a day more, merely because an appeal is pending. Taken literarily, Mr Scott SC's suggestion would mean there shall be automatic stay once leave to appeal against an interlocutory order is granted. I do not understand that to be the law. To take up Mr Scott SC's suggestion, without satisfying myself that PCI has a good prospect of success on appeal, would be to degrade the Court of Appeal to a first instance court. In the light of the above analysis, I am not satisfied that PCI has a good prospect of success. Hence, I do not consider the imminent appeal a good ground for staying the Order.

Variation and clarification

30. Assuming that a PCI policy has been issued as a result of wrongful use of Client Data, PCI queries whether its agent may contact that policyholder for the purpose of providing him service. AXA agree that the Order does not prevent PCI from servicing its policyholders but resist PCI's suggestion to amend the Order.

31. The same difficulty arises also in relation to delivery up of the Client Data of these policyholders under paragraph 3 of the Order which would disable PCI from servicing these clients. AXA's reply is that the Order does not prevent PCI from servicing its policyholders or compel it to deliver up its entire policy files, application forms, policies, claim forms and correspondence with policyholders. Hence, PCI could still service its clients by using information in its policy files. AXA are certainly correct. However, the purpose of the injunction is to protect and not to punish. These ex-AXA policyholders have left AXA. There is nothing to be gained and nothing to be protected by delivery up of Client Data relating to these policyholders. So far as these policyholders are concerned, AXA's position has crystallised. Their only remedy is in damages. Balancing on the one hand the total lack of benefit to AXA by the delivery up and the costs and burden to PCI in servicing these policyholders without the Client Data on the other, I think justice would require that these Client Data should not be delivered up.

32. These two points have not occurred to counsel at the hearing of the application, in fact it is never my intention that the injunction should extend to Client Data in respect of those ex-AXA policyholders who have joined PCI. Accordingly, I shall vary paragraph 3 of my Order by exempting Client Data in respect of ex-AXA policyholders who have switched to PCI. Similarly, for the sake of clarity, paragraph 1(a) as applies to PCI shall be varied accordingly.

33. In relation to the disclosure order in paragraph 4(b) and (c), PCI complains that the issues addressed at the hearing related only to the use of documents allegedly removed from AXA by the 2nd to 10th Defendants. As the Defendants only joined PCI on or after 21 July 2000, a search extending to 1 November 1999 would place an unduly heavy burden on PCI. AXA concede to amend the cut off date to 19 January 2000, the date of commencement of the lease of Room 701 Miramar Tower, as the date to be used to limit the obligation of the 1st Defendant's duty of disclosure. The date of 1 November 1999 was arbitrarily chosen in the case against the Individual Defendants and in my view rightly so in the circumstances. However, as regards the 1st Defendant, it is arguable that to use the same cut off date may be unduly onerous as the August PMS was introduced in August 2000. Bearing in mind that this is only an interlocutory order and having regard to the burden on the 1st Defendant, I would bend in favour of the 1st Defendant. Accordingly, I amend the cut off date in paragraph 4(b), (c) and (d) to 21 July 2000. If the result of the disclosure by the 1st Defendant or indeed by any of the defendants or the result of discovery suggests an earlier cut off date is appropriate, AXA always have the liberty to apply.

34. There are other minor variations agreed upon by the parties during their exchange of views before the hearing. I endorse those variations. Apart from these variations and the variations referred to in the above paragraphs, I do not consider any other variations appropriate.

Stay and extension of time

35. I do not consider a case of stay has been made out on the grounds advanced by the 1st Defendant nor am I satisfied that the appeal will be rendered nugatory in the absence of a stay. Mr Scott SC has raised certain difficulties of compliance. In my view, they are less real than counsel was able to impress on me. Those difficulties are not insurmountable. That apart, much of those difficulties have been removed by my variation order excluding delivery up of Client Data in respect of ex-AXA policyholders who have switched to PCI. I have also reduced the volume of information to be disclosed by adjusting the cut off date. The helpful protocol offered by AXA has relieved much of PCI's burden in identifying the policyholders concerned. The clarification and variation to my Order made as a result of this hearing have given PCI further relief. I am not satisfied that the costs to PCI of compliance outweigh the benefit to AXA. Applying the test I formulated at paragraph 7, the law abiding citizen earnest and willing to comply with the order, I am not satisfied that the Order is vague, ambiguous, oppressive and excessively and unnecessarily wide, nor is it unduly burdensome or impossible to comply. Accordingly the application for stay is dismissed.

36. I have already indicated to the parties at the last hearing on 9 August 2001 that I shall take into account the intervening period between then and the date of hearing in considering any application for extension of time. Four months have now lapsed since my Order of 24 July 2001. As PCI has already complied with paragraph 1 of the Order and as PCI's burden of discovery and disclosure under paragraphs 3 and 4 has been very substantially reduced as a result of the clarification and variation order, I consider an extension of time by three weeks from 2 November 2001 in respect of PMS policies and five weeks from 2 November 2001 in respect of non-PMS policies sufficient to enable compliance. Accordingly, I grant the above extension and with liberty to apply.

37. I shall invite counsel to address me on question of costs.

Representation:

Mr R G Kotewall, SC and Mr Stewart Wong, instructed by Messrs Herbert Smith, for the Plaintiffs

Mr John Scott, SC, Mr David Stokes and Mr Ling Chun Wai, instructed by Messrs Richards Butler, for the 1st Defendant

HCA 9093/2000

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF FIRST INSTANCE

ACTION NO. 9093 OF 2000

____________

____________

_______________

D E C I S I O N

_______________

INTRODUCTION:

Background:

1. This is an application by the plaintiffs for interlocutory injunction, delivery up order and disclosure order against the 1st to 10th defendants in respect of the plaintiffs' confidential information collectively called "Client Data" (see paragraph 9 below). Since the hearing, 18 others have been joined as co-defendants. This decision only applies to the first ten defendants and the word "defendant" in this decision shall, unless the context otherwise requires, be construed to mean the first ten defendants or any of them.

2. The plaintiffs are both members of the AXA group of companies (hereinafter called "AXA") carrying on the business of insurance underwriting, including the underwriting of life insurance policies which the present case is involved. The 1st defendant is an insurance company (hereinafter called "PCI") carrying on a similar business. The other defendants (hereinafter collectively called "the Individual Defendants") were insurance agents appointed by AXA under contracts called "Agent's Contract" to market AXA's insurance policies.

3. In March 2000, PCI launched a policy matching scheme (hereinafter called "PMS") targeted at policyholders of AXA so that they may surrender their AXA policies and switch to PCI policies. PCI offered a hefty welcoming bonus to AXA agents who were willing to join PCI. A large number of AXA agents, including the 2nd to 10th defendants, terminated their Agent's Contracts or have their contracts terminated by AXA and joined PCI. Many of AXA's policyholders who were serviced by these agents surrendered their AXA policies and switched to PCI policies. At about the same time, AXA noticed what they considered as an unusually large amount of printouts of their client information from their computer system called Sales and Marketing System (hereinafter called "SAMS") had been effected under the passwords assigned to the 2nd to 10th defendants prior to the termination of their agency with AXA.

4. On 21 September 2000 AXA instituted the present proceedings. Upon receipt of the writ, PCI voluntarily delivered up a number of SAMS documents in their possession and undertook to deliver up further SAMS documents if discovered and not to process any applications for switching to PCI policies which are accompanied by SAMS documents. The other defendants filed affirmations pursuant to the order of Madam Justice Beeson. All denied that they are still having any SAMS documents.

5. AXA's cause of action against the Individual Defendants is based on breach of implied duty of fidelity by the Individual Defendants in wrongfully removing confidential information belonging to AXA during the currency of their agency with AXA for use after the termination of their agency and to AXA's detriment. As against PCI, AXA's causes of action are, firstly, breach of confidence in that PCI having received the confidential information from the Individual Defendants with knowledge of their breach of duty of fidelity intends to use or disclose the information without AXA's permission and, secondly, interference of AXA's business by unlawful means through the wrongful use of confidential information and misrepresentation made by PCI's own agents in marketing the PMS targeted at AXA's policyholders.

The relief sought:

6. By their summons dated 21 September 2000, AXA seek an interlocutory injunction until trial or further order (1) restraining PCI from disclosing, divulging or otherwise using any of AXA's Client Data, restraining it from accepting any application for a life insurance policy by person whose name appears in AXA's Client Data and who has been approached by any agent of PCI at a time when any of the Client Data was in the possession or under the control of PCI, restraining it from soliciting any business by representing that certain specified life insurance products of PCI match or are comparable with the corresponding AXA product and restraining it from accepting any application for certain specified life insurance product from any person holding a life insurance product issued by AXA and (2) restraining the Individual Defendants from disclosing, divulging or otherwise making use of any of AXA's Client Data.

7. Secondly, AXA seek an order for delivery up to AXA of all documents containing Client Data, including copies, in the possession, custody, power or control of the defendants.

8. Thirdly, AXA seek an order that each of the defendants swear an affidavit:

(a) verifying compliance with the delivery up order;

(b) disclosing all documents containing information relating AXA's policyholders which came into its or his possession, custody, power or control since 1 November 1999 stating, with reference to each such document, the date when and the name and address of the person from whom it came:

(c) disclosing all documents containing information relating to AXA's policyholders which it or he has given or divulged to another person since 1 November 1999 stating, with reference to each such document, the date when and the particulars of the person to whom it or he gave or divulged the documents;

(d) disclosing the name and address of every AXA's policyholder whose name appears in any of the Client Data who has been approached by any agent of the 1st defendant since 1 November 1999, identifying the agent, the nature of the solicitation and stating what if any such business came to the 1st defendant as a result;

(e) disclosing the name and address of all AXA's policyholders to whom any agent of the 1st defendant represented that any of the specified PCI's life insurance products match or are comparable with the corresponding AXA products, and stating what if any business came to the 1st defendant as a result.

9. "Client Data" referred to above means any information obtained or derived by an agent in the course of his agency with AXA, not being information known to the agent prior to such obtaining or derivation, and contained in SAMS, and any documents whether in hard copy or electronic form containing such information and any of the documents listed in Schedule C to the Statement of Claim, whether in hard copy or electronic form, and any information contained therein obtained or derived by an agent in the course of his agency with the plaintiffs not being information known to the agent prior to such obtaining or derivation.

The Agency system:

10. AXA's insurance business, as is customary in the insurance industry, is conducted mainly through insurance agents appointed under agency contracts. These agents are organized into separate groups called agencies headed by a director of agency or senior director of agency with a regional director of agency at the top. Within each agency is a pyramidal structure of senior managers and managers in charge of the various units into which the agents are grouped.

11. Agents including directors of agencies of various levels are not employees of AXA. Clause 1.3 of AXA's Agent's Contract with the 2nd to 10th defendants provides expressly that "it is understood and agreed that there is no employer-employee relationship either expressed or implied between AXA and the agents." They are not entitled to benefits under the Employment Ordinance.

12. The office of the agency is rented by the agency with subsidy from AXA in both rental and remuneration of administrative personnel. These personnel are employees of the agency and not of AXA.

13. Agents are independent and free agents. They are free to move from projects to projects, insurance company to other insurance companies. They are free to choose their principals as they wish, subject to giving one month notice required under the terms of the Agent's Contract. But so long as the Agent's Contract subsists, the contract imposes on the Individual Defendants the obligations of exclusive service to AXA (Clause 1.4), promoting AXA's business faithfully and diligently (Clause 1.5), secrecy (Clause 2.3(b)) and the obligation to return all AXA properties on termination.

14. Agents are sole proprietors operating on their own account. They work from their own agency offices run separately from AXA's office. They receive no wages or salaries from AXA but commission in respect of policies purchased by clients they introduce. They pay their own operational expenses and file their own tax returns as sole proprietors. Their Agent's Contracts require them to keep books and records (Clause 10) and have them available for inspection by AXA during normal business hours and upon termination of the agency to supply such copies of these books and records as AXA may reasonably require (Clause 11.3(a)).

15. Policyholders are usually the agent's friends and relatives or people introduced by them or by his clients. The defendants say that policyholders enjoy a closer relationship with their agent than with the insurance company and wish to be serviced by the same agent irrespective whether the agent stays with the original insurance company or moves to a new one. This view is supported by AXA's former managing director, Mr Terry Jenkins and confirmed in AXA's own training material in 1993 that when an agent moves to service a new insurance company the likelihood of retaining the custom of his policyholders is extremely low, no more than 30%. Thus it is the industry expectation that policyholders have greater affinity to their agents than to the insurance company and the agent's client base represents his inherent commercial value or stock of trade.

16. When an agent leaves an agency, those of his policyholders who do not follow him are left behind. These are called "orphan clients". The director of agency is then at liberty to allocate these orphan clients to agents in his agency. AXA have no right to allocate these orphan clients to other agencies and in fact do not interfere with the allocation within the agency by the director of agency. This suggests that the business relations with policyholders are more the property of the agency than of AXA.

17. The way in which the agents operate within the industry as a whole and the way in which the Individual Defendants operated within AXA are not in dispute.

SAMS:

18. Agents contact clients in the course of marketing AXA's policies. Where policies have been purchased by these clients, their personal particulars, telephone numbers and addresses etc obtained by the agent (hereinafter called "Policyholder's Particulars") and details of the policies, such as policy number, expiry date, currency of account, sum insured, amount of premium, cash value, dividend, loan balance and interest thereon, payment mode and the riders (if any), etc (hereinafter called "Policy Details") are fed into AXA's computer system and stored in a central database, i.e. SAMS. The data stored can be processed and printed out in a number of formats for different purposes.

19. Access to SAMS is restricted. Only senior branch managers, directors of agency, senior directors of agency and regional directors of agency are given access. Each of them is given a unique password to facilitate access. A password holder is obliged to keep the password secret but has authority to authorize a nominated assistant to access SAMS. Access to SAMS could only be made through a dedicated personal computer at each agency office. Each password holder can only access information of policyholders within his responsibility. Thus a senior branch manager may have access to information of policyholders serviced by agents in his units but not those of policyholders serviced by agents in other units. Similarly, a director of agency may access client information in respect of policyholders serviced by agents in his agency but not those serviced by agents in other agencies. All the Individual Defendants were password holders.

20. Password holders are required to sign an agreement called "Hardware Rental Agreement and Software Licence" (hereinafter called "the Licence"). Save and except the 5th and 9th defendant, all Individual Defendants had signed the Licence which provided that all software and data installed on the equipment is or is deemed to be the property of AXA (Clause 4(b)), that the licensee shall treat any data contained in the software as confidential information (Clause 5) and that the licensee will not permit duplication of the program provided.

21. An agent may also access SAMS through his hand-held computer or digital diary but only in respect of information relating to policyholders he services.

22. SAMS data are updated regularly, both on the central database and on the hand-held computers or digital diaries. There are some discrepancies between the plaintiffs' evidence and the defendants' evidence as to the interval between each updating, but there is no dispute that it is a regular updating and that the data in the agent's hand-held computers or digital diaries are replaced every three months and would be erased automatically three months after an agent's termination. The entire SAMS indexation system is also replaced every year in February.

THE PRINCIPLES APPLICABLE TO INTERLOCUTORY INJUNCTION: THE AMERICAN CYANAMID PRINCIPLE:

23. The principles applicable to the grant of interlocutory injunction has been well settled since the House of Lord's decision in American Cyanamid Co v. Ethicon Ltd, [1975] AC 396 per Lord Diplock at 407-409 and as further refined in a number of subsequent leading cases. These principles are as follows.

24. The plaintiff's first hurdle is to satisfy the court that the claim is not frivolous or vexatious; in other words, that there is a serious question to be tried. It is no part of the court's function at the interlocutory stage to try to resolve conflicts of evidence on affidavits as to facts on which the claims of either party may ultimately depend nor to decide difficult questions of law which call for detailed argument and mature considerations. These are matters to be dealt with at trial (American Cyanamid at 407). In this context, it is irrelevant whether the court thinks that the plaintiff's chances of success in establishing liability are 90 percent or 20 percent: Alfred Dunhill Ltd v. Sunoptic SA, [1979] FSR 337 per Megaw LJ at 373.

25. Second, if satisfied that there is a serious question to be tried, the court shall go on to consider whether the balance of convenience lies in favour of granting or refusing the interlocutory relief that is sought. In the balancing exercise, the court shall consider whether damages would be an adequate remedy and whether the unsuccessful party would be in a financial position to pay them. This balancing exercise operates as follows, per Lord Diplock at 408:

"As to that, the governing principle is that the court should first consider whether, if the plaintiff were to succeed at the trial in establishing his right to a permanent injunction, he would be adequately compensated by an award of damages for the loss he would have sustained as a result of the defendant's continuing to do what was sought to be enjoined between the time of the application and the time of the trial. If damages in the measure recoverable at common law would be adequate remedy and the defendant would be in a financial position to pay them, no interlocutory injunction should normally be granted, however strong the plaintiff's claim appeared to be at that stage. If, on the other hand, damages would not provide an adequate remedy for the plaintiff in the event of his succeeding at the trial, the court should then consider whether, on the contrary hypothesis that the defendant were to succeed at the trial in establishing his right to do that which was sought to be enjoined, he would be adequately compensated under the plaintiff's undertaking as to damages for the loss he would have sustained by being prevented from doing so between the time of the application and the time of the trial. If damages in the measure recoverable under such an undertaking would be an adequate remedy and the plaintiff would be in a financial position to pay them, there would be no reason upon this ground to refuse an interlocutory injunction."

26. Third, where there is doubt as to the adequacy of the respective remedies in damages available to either party or to both, then the question of balance of convenience arises.

27. Fourth, where other factors appear to be evenly balanced it is a counsel of prudence to take such measures as are calculated to preserve the status quo.

28. In Centalic Technology Development Ltd v. Worldwide Industrial Ltd [1996] 3 HKC 498, strength of the applicant's case is added by Godfrey JA as a further gloss to the American Cyanamid principles. He held at 510:

"When a judge is called on to exercise his discretion whether to grant or refuse an interlocutory injunction, he has to decide which course is the more just and convenient; and the interests of justice require him to have regard, among other things, to the strength of the applicant's case. If in a case like the present, the judge forms the view that if the action were to go to trial the applicant would succeed, that is a powerful consideration in favour of granting the interlocutory relief sought. If the judge forms the view that the applicant would not succeed, or would probably not succeed, in obtaining a permanent injunction at the trial, that is a powerful consideration against granting such relief."

29. An injunction is an equitable relief. In the exercise of its discretion to grant or to refuse an interlocutory injunction, the court has to take into account all the circumstances of the case as known to the court at that stage. The guiding principle is what the interests of justice require. There are no fixed rules as to when an injunction should be granted or refused. Even Lord Diplock has qualified his general rule that injunction should be refused if damages are adequate and available by the word "normally". At p 409, he also reminds us of the importance that there may be many other special factors to be taken into consideration in the particular circumstances of individual cases.

DUTY OF FIDELITY, DUTY OF CONFIDENCE AND RESTRAINT OF TRADE:

30. AXA's case against the Individual Defendants is based on their duty of fidelity in respect of secret and confidential information owned by AXA, namely Client Data while their case against PCI is based on the duty of confidence. AXA claim that the Individual Defendants removed Client Data in breach of their duty of fidelity during the currency of their agency with AXA for use in competition with AXA after termination of their agency and PCI is in breach of the duty of confidence in using the information. Mr Kotewall SC, on behalf of AXA, relies on a line of authorities starting from Lamb v. Evans [1893] 1 Ch 218, to Faccenda Chicken Ltd v. Fowler [1987] Ch 117, and Lansing Linde Ltd v. Kerr [1991] 1 WLR 251, etc. in support of AXA's case.

31. On the other hand, Mr Fung SC, on behalf of PCI, submits that issues of competition and restraint of trade lie at the heart of this case. Any attempt by an employer or ex-employer to impose a restraint of trade regime on an employee or ex-employee would be regarded in the eyes of the law as anti-competitive and contrary to public policy and must be struck down, as void and of no effect and that the common law has resolutely refused to embrace a tort of "unfair competition", even where one person intends harm to his rival's economic interests. He submits that the employee's duty of fidelity applies during the period of employment only but not afterwards, that in the absence of an express term, AXA can only succeed on the basis of an implied term if they can show improper use of trade secret or confidential information amounting to a trade secret and that the Client Data are neither trade secret nor confidential information equivalent to trade secret. He cites a line of authority starting from Nordenfelt v. Maxim Nordenfelt [1894] AC 535, and Petrofina v. Martin [1966] Ch 146 and Esso Petroleum v. Harper's Garage [1968] AC 269 in support of his proposition. Mr Ho SC, on behalf of the Individual Defendants, echoes Mr Fung SC's opinion.

32. In my judgment, an agent's duty of fidelity, the general duty of confidence and restraint of trade are separate but not mutually exclusive principles. It is for AXA as the plaintiffs to determine what causes of action to found its case. AXA are not relying on restraint of trade and indeed there are no restrictive covenants here against competition or against activities which would be considered legitimate commercial activities. AXA are not seeking to be protected from competition or even from unfair competition. They are seeking protection under the general law of confidence and the duty of fidelity. The line of authorities as referred to me by Mr Kotewall SC is relevant. I do not consider the authorities on restraint of trade helpful to the defendants, save in so far as they may cast light on the extent of an agent's duty of fidelity.

33. The legal issues are (1) what is the nature of the Client Data? Are they protectable information, i.e. whether the Client Data are trade secrets or information of a sufficiently high degree of confidentiality as to amount to trade secrets? (2) whether the Individual Defendants owed the duty of fidelity in respect of the Client Data to AXA while their Agent's Contracts are still extant; and (3) whether PCI owes the duty of confidence in respect of the Client Data coming into its possession towards AXA as owner of the Client Data. It would be convenient to consider these issues in the above order.

NATURE OF CLIENT DATA:

34. "Client Data" for the purpose of this application has the meaning given by AXA in paragraph 9 above. Essentially they are AXA's Policyholder's Particulars, i.e. personal particulars of AXA's policyholders, their telephone numbers, addresses and other contact particulars; and Policy Details, such as policy number, expiry date, currency of account, sum insured, amount of premium, payment mode, cash value, dividend, loan balance and interest thereon and the riders, etc. The information is either obtained by the agent in the course of his agency with AXA or produced by AXA in relation to the policies. They are contained in SAMS or printouts from SAMS, the Schedule C documents and books and records kept by the agents. They are all information used in the trade or business of AXA.

35. Are Client Data trade secrets or confidential information equivalent to trade secrets? In Lansing Linde Ltd v. Kerr [1991] 1 WLR 251, Staughton LJ considered a trade secret as information used in a trade or business which if disclosed to a competitor would be liable to cause real or significant harm to the owner of the secret and that the owner must limit its dissemination or at least not encourage or permit its widespread publication. In this context, trade secret includes secret formulae for manufacture of products (trade secret properly or compendiously so called) and names of customers and goods they buy (equivalents of trade secret or information of a sufficiently high degree of confidentiality as to require the same protection as a trade secret). He set out the three classes of information in Faccenda Chicken and held at 259-260:

"In Faccenda Chicken Ltd. v. Fowler [1984] I.C.R. 589 Goulding J. at first instance had defined three classes of information: (i) information which, because of its trivial character or its easy accessibility from public sources of information, cannot be regarded by reasonable persons or by the law as confidential at all; (ii) information which the servant must treat as confidential ... but which once learned necessarily remains in the servant's head and becomes part of his skill and knowledge; (iii) specific trade secrets so confidential that, even though they may necessarily have been learned by heart and even though the servant may have left the service, they cannot lawfully be used for anyone's benefit but the master's. There the dispute was as to the second class so defined. Goulding J. expressed the view that it could be protected by an express covenant; this court [1987] Ch. 117 was of the contrary opinion. Subsequently other judges have touched on the topic in Balston Ltd. v. Headline Filters Ltd. [1987] F.S.R. 330 and Lock International Plc. v. Beswick [1989] 1 W.L.R. 1268.

It appears to me that the problem is one of definition: what are trade secrets, and how do they differ (if at all) from confidential information? Mr. Poulton suggested that a trade secret is information which, if disclosed to a competitor, would be liable to cause real (or significant) harm to the owner of the secret. I would add first, that it must be information used in a trade or business, and secondly that the owner must limit the dissemination of it or at least not encourage or permit widespread publication.

That is my preferred view of the meaning of trade secret in this context. It can thus include not only secret formulae for the manufacture of products but also, in an appropriate case, the names of customers and the goods which they buy. But some may say that not all such information is a trade secret in ordinary parlance. If that view be adopted, the class of information which can justify a restriction is wider, and extends to some confidential information which would not ordinarily be called a trade secret."

36. Staughton LJ's tests are similar to the ones formulated by Sir Robert Megary V-C in an earlier decision in Thomas Marshall (Exports) Ltd v. Guinle [1979] Ch 227, although expressed in a slightly different language. In Thomas Marshall (Exports) Ltd v. Guinle, the factors taken into consideration by the Vice-Chancellor were the owner's subjective belief that the information is not already in the public domain and that its disclosure would be injurious to the owner's interest. Thus the test according to Sir Robert Megary V-C is subjective, though of course, the owner's belief must be reasonable and the information must be adjudged in the light of trade and industry usage. He held at 248:

"If one turns from the authorities and looks at the matter as a question of principle, I think (and I say this very tentatively, because the principle has not been argued out) that four elements may be discerned which may be of some assistance in identifying confidential information or trade secrets which the court will protect. I speak of such information or secrets only in an industrial or trade setting. First, I think that the information must be information the release of which the owner believes would be injurious to him or of advantage to his rivals or others. Second, I think the owner must believe that the information is confidential or secret, i.e., that it is not already in the public domain. It may be that some or all of his rivals already have the information: but as long as the owner believes it to be confidential I think he is entitled to try and protect it. Third, I think that the owner's belief under the two previous heads must be reasonable. Fourth, I think that the information must be judged in the light of the usage and practices of the particular industry or trade concerned. It may be that information which does not satisfy all these requirements may be entitled to protection as confidential information or trade secrets: but I think that any information which does satisfy them must be of a type which is entitled to protection."

37. In Facenda Chicken, Neill LJ set out the considerations which should be taken into account in deciding whether a particular item of information falls within the implied term so as to prevent its use or disclosure by an employee after his employment has ceased. These are (a) the nature of the employment, (b) the nature of the information itself, (c) whether the employer impressed on the employee the confidentiality of the information, and (d) whether the information can be easily isolated from other information which the employee is free to use or disclose. He said at 137:

"(5) In order to determine whether any particular item of information falls within the implied term so as to prevent its use or disclosure by an employee after his employment has ceased, it is necessary to consider all the circumstances of the case. We are satisfied that the following matters are among those to which attention must be paid:

(a) The nature of the employment. Thus employment in a capacity where 'confidential' material is habitually handled may impose a high obligation of confidentiality because the employee can be expected to realise its sensitive nature to a greater extent than if he were employed in a capacity where such material reaches him only occasionally or incidentally.

(b) The nature of the information itself. In our judgment the information will only be protected if it can properly be classed as a trade secret or as material which, while not properly to be described as a trade secret, is in all the circumstances of such a highly confidential nature as to require the same protection as a trade secret eo nomine....

It is clearly impossible to provide a list of matters which will qualify as trade secrets or their equivalent. Secret processes of manufacture provide obvious examples, but innumerable other pieces of information are capable of being trade secrets, though the secrecy of some information may be only short-lived. In addition, the fact that the circulation of certain information is restricted to a limited number of individuals may throw light on the status of the information and its degree of confidentiality.

(c) Whether the employer impressed on the employee the confidentiality of the information. Thus, though an employer cannot prevent the use or disclosure merely by telling the employee that certain information is confidential, the attitude of the employer towards the information provides evidence which may assist in determining whether or not the information can properly be regarded as a trade secret. It is to be observed that in E. Worsley & Co. Ltd. v. Cooper [1939] 1 All E.R. 290, 307D, Morton J. attached significance to the fact that no warning had been given to the defendant that 'the source from which the paper came was to be treated as confidential.'

(d) Whether the relevant information can be easily isolated from other information which the employee is free to use or disclose. In Printers & Finishers Ltd. v. Holloway [1965] R.P.C. 239, Cross J. considered the protection which might be afforded to information which had been memorised by an ex-employee. He put on one side the memorising of a formula or a list of customers or what had been said (obviously in confidence) at a particular meeting, and continued, at p. 256:

'The employee might well not realise that the feature or expedient in question was in fact peculiar to his late employer's process and factory; but even if he did, such knowledge is not readily separable from his general knowledge of the flock printing process and his acquired skill in manipulating a flock printing plant, and I do not think that any man of average intelligence and honesty would think that there was anything improper in his putting his memory of particular features of his late employer's plant at the disposal of his new employer.'

For our part we would not regard the separability of the information in question as being conclusive, but the fact that the alleged 'confidential' information is part of a package and that the remainder of the package is not confidential is likely to throw light on whether the information in question is really a trade secret."

38. The principle to be distilled from all these well considered authorities is that a trade secret or its equivalent must be information:

(1) used in a trade or business (Lansing Linde and Faccenda Chicken 5(b));

(2) is confidential, i.e. not already in the public domain (Thomas Marshall v. Guinle and Faccenda Chicken);

(3) can be easily isolated from other information which the employee is free to use so that any man of average intelligence and honesty would think it is improper to use the information at the disposal of his new employer (Faccenda Chicken 5(d) and Printers & Finishers)

(4) which, if disclosed to a competitor, would be liable to cause real or significant harm to the owner, (Lansing Linde and Thomas Marshall v. Guinle); and

(5) which the owner of the information must limit its dissemination or at least not encourage or permit its widespread publication or otherwise impress upon the employee the confidentiality of the information (Lansing Linde and Faccenda Chicken 5(c)).

The test to be applied for items (2) and (4) is that of the owner's subjective belief in the light of the usage and practices of the trade and industry. All circumstances must be taken into account, including the nature of the employment, for example whether the employment is in a capacity where confidential material is habitually handled by the employee or agent (Faccenda Chicken 5(a) and Printers & Finishers).

39. For the purpose of the present application, I do not find it necessary to distinguish between trade secret and confidential information equivalent to trade secret, which in ordinary parlance is not a trade secret. Secret formulae for manufacture is an example of the former, while customer's names is an example of the latter. But according to Staughton LJ, both information are protectable as trade secret. The test I set out above applies to information of both categories.

Nature of the agency:

40. I have referred earlier on to the Agent's Contract which imposes on the Individual Defendants an obligation of secrecy in relation to the business of AXA. That seven of nine of the Individual Defendants signed the Licence respecting the data contained in SAMS as confidential, the use of the password system in controlling access to SAMS and that the Client Data in the agent's hand held computer will be erased three months after an agent leaves AXA indicate that their agency is one in which it is likely that the agent would handle secret or confidential information of his principal. These all show that AXA have impressed upon their agents, including the Individual Defendants, the secrecy and confidentiality of the information contained in SAMS.

Nature of the information:

41. It is hardly in dispute that the Client Data are information used in the trade or business of AXA. They are stored in SAMS. AXA have a password system which limits accessibility to SAMS to senior managers and above and even in respect of password holders they could only access information relating to the policyholders serviced by the agents for whom those password holders are responsible. A senior manager may not access information relating to policyholders serviced by other units for which he is not responsible. Similarly, a director of agency may not access information relating to policyholders serviced by other agencies. It is clear that AXA limit the dissemination of the information and does not encourage or permit its widespread publication.

42. The Client Data contain contact details of AXA's policyholders. The information is not available from any public source, except for the policyholder's telephone number but provided that an agent has the full name and address of the policyholder and more importantly that the telephone is listed and registered under his name and not that of any other member of his household. There are no policyholders registers of any kind, whether of AXA or other insurance companies, open to the public. AXA's policyholders are a particular section of the public in that they have purchased insurance policies and are people who are more likely to purchase other or additional insurance products. Not only that these policyholders constitute a fertile target for selling policies and in the present case for selling "matching PCI policies" targeted at AXA's policyholders. The difference between searching a telephone directory, the information therein is within the public domain, and searching AXA's Client Data is like fishing in the open sea and fishing in AXA's own fish farm. In the latter situation, a successful and fuller catch is more likely and AXA would suffer in respect of each catch. If PCI and their agents have access to AXA's Client Data, they would be able to contact AXA's policyholders and market their matching policies specially tailor-made for and targeted at them. I have no difficulty in finding that the Client Data are information used in the trade or business of AXA, the disclosure of which would be liable to cause real or significant harm to AXA and that it is reasonable for AXA to hold such belief.

43. PCI also operates a similar database and enters into similar agency contracts with their own agents containing similar provision as to secrecy. In their solicitors' letter to AXA's solicitors they also impressed upon AXA the importance of confidentiality in respect of their own client information contained in the CDs which they handed over to AXA's solicitors pursuant to their undertaking. This is self-evident that client information is treated as secret or confidential by PCI and the insurance industry as a whole. This reinforces AXA's belief that the Client Data are trade secrets or confidential information equivalent to trade secrets which if disclosed to a competitor would cause serious or substantial harm as being reasonable. AXA's belief is also in line with trade and industry practices. It also shows that the steps taken by AXA in preventing dissemination or publication of their Client Data are entirely reasonable.

44. Mr Fung SC and Mr Ho SC both submit that in Facenda Chicken itself it was held on the facts that customer information did not amount to trade secrets. However, that is a finding of fact only. In my view, that case itself did not lay down any principle that customer lists or customer information cannot amount to trade secret or equivalent to a trade secret. The facts in Facenda Chicken is clearly distinguishable from the present case. In that case, the customer information related to a relatively small number of customers which the employee contacted on his daily round so that the information must have been learnt and become part of the employee's stock of knowledge. In the present case, each agent maintains about 300 policyholders. I am far from being convinced that an agent would be able to recall all these policyholders' contact details without deliberately committing to memory these particulars or without reference to some sort of record which they kept or take away. Further, I do not believe the Individual Defendants would be able to recall all the policyholder's contact details which AXA allege have been removed by them.

45. I do not think the fact that the information could be innocently carried away in the head of an agent by itself is definitive as to whether the information is protectable. In SBJ Stephenson, Bell J held at 298:

"I do not accept that SBJ was being overambitious or unreasonable in protecting information after termination, and I cannot accept that it was unreasonable because it could cover client information which Mr Mandy could use with the client's consent once he had approached the client, simply remembering who he was, after termination of his contract with SBJ. SBJ would still have its own interest in such information as was protectable, whatever view the client took, and I do not see Facenda Chicken as authority for the proposition that information learned without deliberate intent to misuse it is not capable of legitimate protection, even though parts of Neill LJ's judgment at page 146A and H may appear to suggest that, at least in respect of the identity of clients of the employer. In my view, the distinction between information deliberately learned and that which is innocently carried in the employee's head, cannot be definitive as a matter of principle of what information can be legitimately protected and what cannot, after the termination of employment, and I do not believe that what Neill LJ said in the particular circumstances of that case was meant to define such a principle ..."

46. He continued at 300-301:

"In my judgment, the names of SBJ's clients, the details of renewal dates, the cover provided and claims made, premiums charged, fee rates or rates of brokerage earned, construction of insurance programmes, markets used and the manner in which policies were arranged or formulated, was all information which was protected by the terms of Clause 12(A) of Mr Mandy's employment agreement and by his implied terms of fidelity, while he was still employed by SBJ, and (subject to the argument on repudiatory breach by SBJ) by clause 12(A) after the termination of his employment."

47. Though the information in SBJ Stephenson is protected by Clause 12(A) of the employment contract, it appears from the above dicta that Bell J considered the protection equally available under the implied terms of fidelity. I think it could also be so argued in the present case.

48. Thus client's contact details in SBJ Stephenson are protectable even if carried away in the head of the employee. I think the finding in Faccenda Chicken that customer's name list is not trade secret or its equivalent should be limited to its facts. Whether those details are protectable trade secrets or equivalents to trade secrets depends on whether the information meets the test as I set out in paragraph 38 above. In respect of client's particulars, I prefer to follow SBJ Stephenson and Lansing Linde. Further the information which AXA seek protection is not just in respect of client's contact details, but also Policy Details and other SAMS and Schedule C documents. Those information are clearly trade secrets or their equivalents.

49. Mr Fung SC draws a distinction between trade secrets stricto sensu and other forms of confidential information. He submits that an agent's duty of fidelity ceases upon termination of agency and that client's particulars are not trade secrets capable of protection by an implied covenant in restraint of trade but could only be protected by express covenant and only if reasonable. He cited the following dicta from Leggatt LJ from Wallace Bogan v. Cove [1997] IRLR 453 at Paragraph 14:

"The term that is to be implied not to canvass the employer's customers during employment is part of the duty of good faith and fidelity. But when the employment ceases so does that duty. The implied duty not to misuse confidential information endures after the employment has ended, but only if it amounts to a trade secret. Otherwise to govern what is to happen after termination of a contract an employer is dependent on express terms. In default, terms could only be implied in the traditional way, that is, by applying the test of the officious bystander or of business efficacy; and neither test, nor any other such as is contemplated by the Liverpool City Council case, would bind any employee when the employment was over. It is axiomatic that the general law affords no protection to an employer against an ex-employee soliciting the employer's customers."

50. The fact that an agent's duty of fidelity ceases upon termination of his agency is no answer to AXA's claim that the Individual Defendants were in breach of their duty of fidelity by making copies of AXA's Client Data during the subsistence of their agency for use in competition with AXA after their agency has terminated.

51. I do not think the distinction between trade secrets stricto sensu and other forms of confidential information material. It is abundantly clear in Lansing Linde that "trade secret in this context" as used by Staughton LJ includes both trade secret strictly so called and information of a sufficiently high degree of confidentiality as to amount to trade secret, but which in ordinary parlance would not be so described. The test that I have formulated above applies to information of both categories. In this context, I think Mr Fung SC's distinction is semantic. If trade secret is protected, there is no reason why what in substance is trade secret, though not so called in ordinary parlance, is not. In my judgment, by whatever name the information is called, if it is material which in all the circumstances of the case is of such a highly confidential nature as to require the same protection as a trade secret eo nomine, then the law gives it the same protection as a trade secret. It is all a matter of fair and honourable dealing which a person in a fiduciary position must maintain in respect of the information he received in the course of his agency or employment.

52. I also draw support for this proposition from Nourse LJ's dicta in Roger Bullivant Ltd v. Ellis [1987] FSR 172 at 180 where he treated what the court compendiously described as trade secrets on the same footing as materials which while not properly described as trade secrets, are in all the circumstances of such a highly confidential nature as to require the same protection as a trade secret eo nomine. Both trade secrets or their equivalents may not in any circumstances be used by the employee or agent, either during or after the employment, except for the benefit of the employer or principal.

Ownership of the information and source irrelevant:

53. Mr Fung SC and Mr Ho SC submit that another distinguishing feature in the Individual Defendants' relation with AXA is that the policyholders are clients of the agents and not of AXA and the information obtained by the agents belong to the agents. They argue that agents have their own client base which represents part of their inherent commercial value or stock-in-trade. Mr Ho SC submits that the insurance business is operated on a pyramid system in which each agent recruits his own agents and his own clients. Usually an agent recruits his client from among his friends and relatives or people introduced by them. He keeps the name and contact details of these clients in his books and records which he is obliged to keep under the terms of his Agent's Contract with AXA. When a policy is taken out, the agent provides the information to AXA who input the same into SAMS. But the books and records and of course the information contained therein are respected as the agent's property. AXA only has a contractual right to inspect these books and records during normal business hours and as may reasonably require. These books and records are the properties of the agents and do not have to be surrendered to AXA upon termination of their agency. Upon termination, AXA may only request for copies of such parts of the books and records relating to the business of AXA and as AXA may reasonably require. AXA's right to inspect during the currency of the agency and their right to have copies of part of the books and records relating to AXA's business are all subject to the requirement of reasonableness. Thus they submit that the agents are at liberty to use Policyholder's Particulars contained in the agent's books and records which the agent is free to take away and keep after the agency has terminated.

54. Policyholders maintain a closer relationship with the agent than with the insurance company and often follow the agent when he leaves for another insurance company. Even AXA's own training material recognised that 70% of the policyholders would follow their agents. When an agent leaves AXA, those of his policyholders who do not follow him (called "orphan clients") would be reallocated by the agency where the departing agent worked and not by AXA. As policyholders are approached and cultivated by agents, Mr Ho SC argues therefore that the client information in the form of Policyholder's Particulars belong to the agent and not to AXA, especially as the agent is not employed and paid by AXA but is remunerated only on the basis of policy sold.

55. The fact that an agent is not employed by AXA and is not paid monthly wages weighs in favour of the suggestion that whatever the agent obtains in the course of his agency remains the property the agent. However, the cases suggest ownership is not the single or most definitive factor. As Lindley and Bowen LJJ held in Lamb v. Evans, the issue is not to whom the property in the information belongs, but whether the employer or the principal has a sufficient interest or special property in the information to entitle him to restrain the use of the information by the agent against the principal: see dicta cited below in paragraphs 75 and 76.

56. In my view, confidentiality is attached to the protectable subject matter. In Lamb v. Evans, Bowen LJ referred to this protectable characteristic as "special property." In my view, whether the information could be carried away in the head of the agent or in some other records kept by him cannot be definitive as a matter of principle of what information can be legitimately protected and what cannot. The following dicta of Parker LJ in Johnson & Bloy (Holdings) Ltd v. Wolstenholme Rink [1989] 1 FSR 135 at 142 are pertinent:

"If and in so far as it is suggested that the decision of Scott J in Balston Limited v. Headline Filters Ltd [1987] FSR 330 lays down as a proposition of law that anything which is inevitable in somebody's head when they leave the employment is something which they are free to use, I must respectfully say that I do not regard it as laying down any such proposition - and if it does, it is wrong.

The matter may be tested very simply. Somebody may hit upon the combination of two ingredients after many years of research, which produce an immensely valuable result wholly unknown to anybody else. An employee would have no difficulty in holding that knowledge in his head and inevitably carrying it away with him. But it cannot, as I see it (and certainly arguably cannot) be regarded as part of the ordinary skill and experience of the particular employee. It is a secret, whether or not it is carried away in his head, or if that is not so, it is arguably a secret, and that is all that is required for present purposes."

57. In Printers & Finishers, Cross J said at 5:

"The mere fact that the confidential information is not embodied in a document but is carried away by the employee in his head is not of course, of itself a reason against the granting of an injunction to prevent its use or disclosure by him. If the information in question can fairly be regarded as a separate part of the employee's stock of knowledge which a man of ordinary honesty and intelligence would recognise to be the property of his old employer and not his own to do as he likes with, then the court, if it thinks that there is a danger of the information being used or disclosed by the ex-employee to the detriment of the old employer, will do what it can to prevent that result by granting an injunction."

58. The test for trade secrets or their equivalents is as clearly set out by Sir Robert Megarry V-C in Thomas Marshall (Exports) Ltd v. Guinle (see paragraph 36 above) and Lansing Linde Ltd v. Kerr (see paragraph 35 above).

59. On the fact, an AXA agent is employed as an exclusive agent and paid very generous commission upon successful introduction of a sale and thereafter paid continuing commission upon each annual renewal. The policies are issued by AXA and not by the agent. The insured, even if he was a client of the agent, becomes a policyholder of AXA and not of the agent. The policy is renewable annually and usually for substantial durations. Both AXA and the policyholder expect a long and lasting relationship. These support the inference that once a policy is concluded, AXA has an interest in renewing the policy annually with that policyholder and an interest in his Client Data to protect. If so, it matters not who was instrumental in obtaining the information and that the same information could be found in other private books and records kept by the agent, not in the public domain. Accordingly, AXA have an interest in the information relating to this client and policyholder to protect and to restrain its agent from copying or removing Client Data relating to this policyholder during the currency of his agency. It should be noted that the defendants' own "note-books" used to note down clients' information in Lamb v. Evans [1893] 1 Ch 218, and the defendants' own diaries in Peninsular Real Estate Ltd v. Harris [1992] 2 NZLR 216, all formed part of the subject matter falling within the scope of delivery up and injunctive orders.

Information kept in alternative source: Agent's books and records:

60. Mr Ho SC submits that any injunction, if granted, should be "source-specific", i.e. it should only be directed against the misuse of documents or material wrongfully taken away from AXA and should not be wide as to prevent the Individual Defendants from contacting the policyholders by using or resorting to lawful source. He argues that the primary information was collected by the agent first-hand from the client and then compiled into the agent's own record which existed independently of SAMS. They were compiled for the Individual Defendants' own purpose to facilitate their servicing of their clients. He therefore distinguishes between information kept in the agent's record from that contained in SAMS and submits that information from the former source were the Individual Defendants' own property and is not subject to injunction.

61. Both Mr Fung SC and Mr Ho SC refer to clause 10.1 and 11.3(a) of the Agent's Contract. Under the former provision, an agent is obliged to keep books and accounts. Clause 11.3(a) provides:

"The agent shall return to the Company all moneys, policies, receipts, rate books, manuals, literature, all forms and documents held by it bearing the name of the Company, and shall return or supply copies of such parts of the Agent's books and records relating to the business of the Company as the Company may reasonably require."

They submit that this clause envisages that agents would be entitled to keep the originals of their books and records and are required only to deliver copies if reasonably required by AXA.

62. Mr Kotewall SC ingeniously argues that properly read, the word "copies" only apply to "supply" and not to "return", so that agents have a duty to "return such parts of the Agent's books ..." and "supply copies of such parts of the Agent's books ..." He further submits that as a matter of plain English and read in its proper context, "return" which has been used twice in the clause must refer to documents already in existence whether as originals or copies, while "supply" means the making of copies upon request. I think Mr Kotewall SC's point is, at the lowest, arguable.

63. In my view, this clause is not of great assistance to the defendants. AXA are not relying on this clause but rather on the general law of confidence and agent's duty of fidelity. It is at least arguable that confidentiality is attached to the information and it does not matter where the information is kept, be it carried away in the agent's head or written in his books. It should be noted that in Lamb v. Evans and Peninsular Real Estate, the agents' own notebooks were ordered to be delivered up.

Separability of Client Data from unclassified information:

64. On the evidence, it is clear that the Client Data can be easily distinguished from other information which the Individual Defendants are free to use because SAMS is the repository of all the confidential client information owned by AXA to which access is controlled by the password system and to which an individual agent only has limited access. The password system, the Licence and the nature of the information indicate that the confidentiality of the Client Data is so respected that applying the test of the man of average intelligence and honesty, the agent must realise the confidential nature of the information contained in SAMS, so that he is not at liberty to use the information even if carried away in his head or in some books and records kept by him.

Conclusion:

65. To conclude this part of my analysis, I am satisfied that AXA have established a serious legal issue to be tried as to whether the Client Data which they seek to protect are trade secrets or confidential information of a sufficiently high degree of confidentiality as to amount to trade secrets as set out in Faccenda Chicken, Thomas Marshall (Exports) Ltd v. Guinle and Lansing Linde Ltd v. Kerr.

DUTY OF FIDELITY DURING THE SUBSISTENCE OF THE AGENT'S CONTRACT:

66. Duty of fidelity may be created by express terms in a contract or by necessary implication so as to give to a transaction the effect which must have been in contemplation of the parties when they entered into the contract. The duty is readily implied to parties in a fiduciary relation, such as trustee and beneficiary, principal and agent, master and servant and solicitor and client. The nature and extent of the duty of course varies depending on the nature of the fiduciary relation and all the circumstances.

The principle in Faccenda Chicken:

67. For the purpose of this interlocutory application, I do not think it necessary to go into detailed argument about this well established principle of an agent's duty of fidelity towards his principal. The law has been usefully summarised by Neill LJ in Facenda Chicken Ltd v. Fowler [1987] Ch 117 at 135-137, where he said,

"It is sufficient to set out what we understand to be the relevant principles of law. Having considered the cases to which we were referred, we would venture to state these principles:

(1) Where the parties are, or have been, linked by a contract of employment, the obligations of the employee are to be determined by the contract between him and his employer: cf. Vokes Ltd. v. Heather(1945) 62 R.P.C. 135, 141.

(2) In the absence of any express term, the obligations of the employee in respect of the use and disclosure of information are the subject of implied terms.

(3) While the employee remains in the employment of the employer the obligations are included in the implied term which imposes a duty of good faith or fidelity on the employee. For the purposes of the present appeal it is not necessary to consider the precise limits of this implied term, but it may be noted: (a) that the extent of the duty of good faith will vary according to the nature of the contract (see Vokes Ltd. v. Heather, 62 R.P.C. 135); (b) that the duty of good faith will be broken if an employee makes or copies a list of the customers of the employer for use after his employment ends or deliberately memorises such a list, even though, except in special circumstances, there is no general restriction on an ex-employee canvassing or doing business with customers of his former employer: see Robb v. Green [1895] 2 Q.B. 315 and Wessex Dairies Ltd. v. Smith [1935] 2 K.B. 80.

(4) The implied term which imposes an obligation on the employee as to his conduct after the determination of the employment is more restricted in its scope than that which imposes a general duty of good faith. It is clear that the obligation not to use or disclose information may cover secret processes of manufacture such as chemical formulae (Amber Size and Chemical Co. Ltd. v. Menzel [1913] 2 Ch. 239), or designs or special methods of construction (Reid & Sigrist Ltd. v. Moss and Mechanism Ltd. (1932) 49 R.P.C. 461), and other information which is of a sufficiently high degree of confidentiality as to amount to a trade secret. The obligation does not extend, however, to cover all information which is given to or acquired by the employee while in his employment, and in particular may not cover information which is only "confidential" in the sense that an unauthorised disclosure of such information to a third party while the employment subsisted would be a clear breach of the duty of good faith. This distinction is clearly set out in the judgment of Cross J. in Printers & Finishers Ltd. v. Holloway [1965] 1 W.L.R. 1; [1965] R.P.C. 239 where he had to consider whether an ex-employee should be restrained by injunction from making use of his recollection of the contents of certain written printing instructions which had been made available to him when he was working in his former employers' flock printing factory. In his judgment, delivered on 29 April 1964 (not reported on this point in [1965] 1 W.L.R. 1), he said [1965] R.P.C. 239, 253:

'In this connection one must bear in mind that not all information which is given to a servant in confidence and which it would be a breach of his duty for him to disclose to another person during his employment is a trade secret which he can be prevented from using for his own advantage after the employment is over, even though he has entered into no express covenant with regard to the matter in hand. For example, the printing instructions were handed to Holloway to be used by him during his employment exclusively for the plaintiffs' benefit. It would have been a breach of duty on his part to divulge any of the contents to a stranger while he was employed, but many of these instructions are not really 'trade secrets' at all. Holloway was not, indeed, entitled to take a copy of the instructions away with him; but in so far as the instructions cannot be called 'trade secrets' and he carried them in his head, he is entitled to use them for his own benefit or the benefit of any future employer.'

The same distinction is to be found in E. Worsley & Co. Ltd. v. Cooper [1939] 1 All E.R. 290 where it was held that the defendant was entitled, after he had ceased to be employed, to make use of his knowledge of the source of the paper supplied to his previous employer. In our view it is quite plain that this knowledge was nevertheless "confidential" in the sense that it would have been a breach of the duty of good faith for the employee, while the employment subsisted, to have used it for his own purposes or to have disclosed it to a competitor of his employer."

68. The Individual Defendants' Agent's Contracts do not contain any express term providing for the duty of good faith or fidelity whether during or after termination of the agency. The agent's obligations are therefore the subject of implied terms (the second and fourth principles referred to in Facenda Chicken).

69. Clause 2.3(b) of the Agent's Contract provides that the Individual Defendants must, as long as the contract is in force, "observe the strictest secrecy concerning the business of AXA or of persons or companies from time to time dealing with [AXA] or the Agent or which come to the knowledge of the Agent." Under Clause 1.5, they are also required to promote the business of AXA faithfully and diligently. While these provisions do not expressly impose an obligation of good faith or fidelity on the Individual Defendants, the existence of these contractual duties, particularly the duty of secrecy, points strongly to an implied duty of good faith or fidelity on the agents while their agency is still extant. The existence of such an implied duty of fidelity during the currency of the agency is thus hardly disputable.

70. This duty of good faith or fidelity is broken if an employee makes or copies a list of customers of the employer for use after his employment ends or deliberately memorises such a list, even though there is no restriction on the ex-employee canvassing or doing business with customers of his former employer (item (b) of the third principle in Facenda Chicken).

Whether commission agents owe similar duty of fidelity to principal:

71. Mr Fung SC and Mr Ho SC submit that this principle is applicable only as between employee and employer as in the case of Facenda Chicken and all the other authorities on this point cited by AXA and not applicable as between agent and principal. The agents are not employees of AXA but are sole proprietors operating on their own account. They file their own tax returns as sole proprietors. They work from their own agency offices run separately from AXA and pay their own expenses and outgoings. They earn commission from AXA in respect of policies sold. They may recruit agents to work for them. Counsel submit that these features in the Individual Defendants' relationship with AXA distinguish them from the relationship between an employer and employee.

72. Mr Fung SC refers me to Marshall v. N.M. Financial Management Ltd [1995] WLR 1461. In that case, the plaintiff was an exclusive agent employed by the defendant to market its life insurance and pension policies. The contract provided that the plaintiff should be an independent contractor remunerated by commission on business introduced and his relationship with the defendant should be one of agent and principal. In declaring a term of his contract void as being in restraint of trade, Jonathan Sumption QC held at 1465:

"The defendant is seeking to apply a restriction on competition by a self-employed agent. The plaintiff's business as an agent, including the goodwill arising from his reputation and connections, was his own property and not the defendant's. The only interest which the defendant itself could have to protect is its interest in the goodwill attaching to its own business as a seller of investment agreements which is largely if not entirely an interest in the protection of its existing clients with whom it already had contracts."

73. Mr Fung SC submits that the Individual Defendants are in the same position as the plaintiff in Marshall v. N M Financial Management and are in a much stronger position than employees in respect of the liberty in using information obtained while in the course of their agency. Mr Kotewall SC argues that in law their positions are the same.

74. Marshall v. N M Financial Management, however, was concerned with restriction on competition and with balancing a self employed agent's interest in the goodwill arising from his reputation and connection and the principal's interest in the goodwill attaching to its own business as a seller of investment agreements. The case was not about breach of duty of fidelity or about confidential information. It lays down no general principle which allows an agent to use confidential information to the detriment of his employer. The decision certainly does not affect the existence or otherwise of the agent's duty of fidelity.

75. Though the cases cited by Mr Kotewall SC are employment cases, it is obvious from the dicta of those cases that the employee's well established duty of fidelity is just one aspect of his duty as a fiduciary agent of his employer. The following dicta of Kay LJ in Lamb v. Evans [1893] 1 Ch 218 at 235 is illuminating:

"All these materials were obtained while you, the Defendants, were acting as the Plaintiff's agents, while you were in that confidential relation to him and for the purpose for which he employed and paid you, viz., of compiling this book of the Plaintiff's, and therefore to allow you to use any of those materials for your own purposes would be allowing you to use them for a purpose for which they were not compiled - you, while you compiled them, being in the position of the Plaintiff's agent, and there being a confidential relation between you and the Plaintiff."

76. Also in the same case, Lindley LJ held at 226:

"What right has any agent to use materials obtained by him in the course of his employment and for his employer against the interest of that employer? I am not aware that he has any such right. Such a use is contrary to the relation which exists between principal and agent. It is contrary to the good faith of the employment, and good faith underlies the whole of an agent's obligations to his principal. No case, unless it be the one which I will notice presently, can I believe be found which is contrary to the general principle upon which this injunction is framed, viz., that an agent has no right to employ as against his principal materials which that agent has obtained only for his principal and in the course of his agency. They are property of the principal. The principal has, in my judgment, such an interest in them as entitles him to restrain the agent from the use of them except for the purpose for which they were got."

77. In the same case, in relation to whether an employee may use materials obtained by him in the course of his employment and for his employer against the interest of that employer, Bowen LJ held at 229-230:

"Ought not the Plaintiff to be able to restrain them from afterwards using those materials and those documents in competition with the Plaintiff himself? It is not a question of copyright - that must be kept out of sight altogether - nor is it, on the other hand, a simple question of the absolute property at law in the documents themselves or in the blocks themselves. It is a question of whether the Plaintiff, whatever the property in the documents may be, or whatever the property in the materials may be, has not sufficient special property in them to entitle him to restrain the use of them against him when they had been obtained for his use by his agents in the course of their employment. That depends entirely, I think, upon the terms upon which the employment was constituted through which the fiduciary relation of principal and agent came into existence. I think my Brothers have already during the course of the argument expressed what I fully believe, that there is no distinction between law and equity as regards the law of principal and agent. The common law, it is true, treats the matter from the point of view of an implied contract, and assumes that there is a promise to do that which is part of the bargain, or which can be fairly implied as part of the good faith which is necessary to make the bargain effectual. What is an implied contract or an implied promise in law? It is that promise which the law implies and authorizes us to infer in order to give the transaction that effect which the parties must have intended it to have and without which it would be futile."

Thus the English Court of Appeal saw no distinction in the nature of the duty of fidelity imposed on an agent and that imposed on an employee.

78. AXA agents are placed in such a position by AXA, but for which they would not have been able to contact clients and obtain information from them. Similarly, policyholders would not have supplied the information if not for the purpose of purchasing AXA's insurance policies. Agents are remunerated commission for their services in selling policies which included an element of reward for the Policyholder's Particulars so obtained. Under such fiduciary relationship the Individual Defendants must owe a duty of fidelity towards AXA. I cannot see any real distinction between the fiduciary relationship created by a contract of agency and that by a contract of employment. An agent owes the same duty of fidelity to his principal in respect of information obtained by him for his principal in the course of his agency as that owed by an employee to his employer in respect of information obtained by him in the course of employment.

79. In Roger Bullivan Ltd v. Ellis [1987] FSR 498, Nourse LJ, referring to Robb v Green [1895] 2 QB 315 said the principle is "one of no more than fair and honourable dealing [which] should be steadfastly maintained." I am not aware of any authority which decrees that the duty of fidelity is lower in the case of principal and agent than that in the case of employer and employee. This does not accord with the notion of fair and honourable dealing which must exist in any fiduciary relation whether of principal and agent or of employer and employee. In my view, the duty exists as long as the fiduciary relation subsists and it makes no difference whether the fiduciary relation is one between employer and employee or principal and agent.The Individual Defendants were employed as exclusive agents of AXA. During the currency of the agency they may not work for any other insurance company. It must be an implied term of their Agent's Contract that they were to serve their principal, AXA, with good faith and undivided interest and should not do anything which may harm AXA, not at least during the currency of the agency. It goes without saying that they were not to make copies of AXA's confidential information or trade secrets for their use after their agency ends. In my opinion the third principle in Facenda Chicken is applicable to the relation between the Individual Defendants and AXA as if they were employees of AXA.

80. Mr Ho SC submits that Roger v. Bullivant is concerned with an employer and employee relationship in which there was a restrictive covenant and that the customer index removed by the employee was compiled by the employer. It is clear that the above dicta was given by Nourse LJ when considering the question of confidential information by reference to general principles of law and not based on the restrictive covenant. As for the other two points of difference raised by Mr Ho SC, I do not think they amount to material distinction.

81. Mr Ho SC draws a distinction between the primary information and the compilation of such information. He submits whether the product of compilation attracts confidentiality, the primary information, if it can be obtained from other open sources, remains open for use. Attractive as his submission may be, however, from the above analysis, there is really no such open source in respect of any primary information (see paragraph 42 above). AXA's policyholders are not listed in any policyholder register open to the public. Even if an agent knows the name of a policyholder, telephone directories are not useful unless the policyholder is a registered user of a telephone line.

82. Mr Fung SC and Mr Ho SC submit that in the absence of express terms, this protection would not avail AXA in respect of client information, especially client contact details, which an agent could carry away in his head without consciously committing the information to memory. The authorities are clear. In respect of trade secret or its equivalent, an agent may not use it except for his principal's business, even if it may be carried away in his head (see paragraphs 60 to 64 above). It is only in respect of confidential information not amounting to trade secret and which have formed part of the agent's stock of knowledge in his head that he may use it after his agency expires for his own purpose and even in competition with his former principal.

83. Many of the policyholders are personally known to the agents. In respect of those policyholders, an agent would not have difficulties in recalling their contact details. However, the Individual Defendants were senior managers whose function was to recruit and train agents. The substantial part of their income was derived from overriding commission. They did not contact clients save those who are their personal friends and relatives. I do not think the nine Individual Defendants together can recall more than a minute fraction of the client contact particulars among the Client Data which have allegedly been removed by them.

84. In any event, if a policyholder's name is a trade secret or its equivalent which an agent is enjoined from using, then he may not recall that name from his memory and look up the telephone directory for his contact details. Such an exercise may not be fruitful as the policyholder may not be the registered user of a telephone line (see paragraph 42 above). AXA are not trying to restrain any agent from competing with AXA upon termination of his agency. AXA only seek to restrain him from using the confidential information. The agent is at liberty to deal with a policyholder if the policyholder contacts the agent. If the policyholder shows him SAMS documents, the agent is also at liberty to use them or the information contained therein.

Conclusion:

85. I am satisfied that AXA have raised a serious legal issue to be tried as to whether the Individual Defendants would be in breach of their duty of fidelity to AXA by copying and removing Client Data during the currency of their agency for the purpose of using them to AXA's detriment after the termination of their agency.

WHETHER THERE WAS BREACH OF FIDUCIARY DUTY BY THE INDIVIDUAL DEFENDANTS

86. In essence, AXA's case is based on the unusually large volume of SAMS printouts made under the passwords of the Individual Defendants at about the time they left AXA and that some of these printouts or copies thereof were surrendered by PCI. AXA's case is not based on direct evidence but mainly on inferences to be drawn. Some of the Individual Defendants admit authorising some of the printouts and offer a legitimate explanation while the others allege that the printouts were made by somebody else without their knowledge. However, Mr Kotewall SC submits that when these allegations are tested against the evidence from the other agents, the defendants' account or explanation do not ring true in view of the timing and the volume of the printouts.

87. The general thrust of the Individual Defendants' case is that they were senior managers mainly responsible for recruiting and training agents under them. They did not operate SAMS, which operation was invariably delegated to their secretarial staff. Individual agents were entitled to request for printouts in the ordinary course of business and no specific approval is required from the Individual Defendants. Further, as managers, the Individual Defendants do not contact clients. In any event, their evidence is that cold calling policyholders serviced by another agent is seldom successful. Hence they have little or no use of the SAMS printouts.

88. The argument based on the Individual Defendant's senior position cuts both ways. There is no dispute that like clients, agents usually follow their managers or directors of agency. Hence it is useful to have client information of all agents under a manager or a director when he leaves. Possession of an agent's client information also gives the agent pressure to follow his manager or director. As to cold calling, AXA's evidence is that statistically existing policyholders are more likely to take out further policies and agents do service orphan clients which is similar to cold calling. Hence the SAMS printouts are useful.

89. For the purpose of these proceedings the burden AXA need to achieve is to raise a serious question to be tried. AXA do not have to satisfy me that their case is more likely than the defendants'. It would be sufficient if AXA could adduce such evidence as would enable the inference that the Individual Defendants had removed SAMS printouts to be drawn and that the Individual Defendants' explanations are inherently improbable. The specific cases of the Individual Defendants are analysed below.

The case against the 2nd defendant:

90. The 2nd defendant was a regional director of agency to whom password access to SAMS was given. As at 31 March 2000, he had 182 in force agents within his agency of 20 in force units. He resigned on 10 April 2000 and requested leave of absence from the same date. He was registered as a PCI agent on 21 July 2000. Pursuant to its undertaking, PCI delivered five client lists of the 2nd defendant's own direct clients to AXA. These client lists were all printed on 8 April 2000, two days before his resignation took effect.

91. AXA's case against him is as follows. Towards the end of 1999 he had intimated to another agent, Eric Tsoi, that he would leave AXA if he could persuade a sufficient number of AXA agents to leave with him. AXA say that this explains the unusually high incidence of printouts of 34 Audit Reports and 44 Audit Reports from SAMS in December 1999 and April 2000 respectively, but these printouts were not recovered upon a search of the 2nd defendant's office after he left. Inference that he handed over his own client lists to PCI could also be drawn from the delivery up of five client lists by PCI.

92. The 2nd defendant says that the Client Data are compiled using information supplied by agents, they are not confidential information and are available from books and records kept by the agents. These arguments have been considered in some detail above.

93. He also says that he did not personally operate SAMS and did not even know the password even though he was the password holder. Daily operation of SAMS was delegated to his secretarial staff, Ms Catherine Cheung Kin Sheung and Lee Ching Yan to whom agents may request for printouts. He says the high incident of printout in December 1999 is not unusual as Christmas would be a good time for contacting clients to get more business and in any event the Audit Reports are in respect of clients serviced by agents who have not joined PCI. He denies that the printouts in April 2000 were made at his instruction and says he has no idea why his client lists came into possession of PCI. His personal assistant, Ms Phyllis Leung says that she made copies of her own as well as that of the 2nd defendant's client lists without the 2nd defendant's knowledge. When the 2nd defendant's clients asked to switch to PCI upon learning of the 2nd defendant's departure from AXA she gave the client lists to PCI for convenience in order to apply for policies on behalf of those clients. The 2nd defendant says that cold calling other agent's client is unprofessional and seldom fruitful.

94. The 2nd defendant's defence is to a certain degree discredited by Ms Cheung who says that while it is usual to have a lot of printouts of client labels in December for sending out Christmas cards to clients, the high incident of printouts of Audit Reports in December 1999 and April 2000 is unusual and have not been adequately explained. The high incidences of printouts at the same time or within a short period of time when read with the statistical data are unusual and cannot be attributed to individual requests.

95. Audit Reports are used for allocating orphan clients. As Audit Reports are related to orphan clients whose agents have left AXA, they are all the more valuable for marketing PCI matching policies. There is no reason why such a high incident of printing is required all at the same time. Also individual agents have no authority to request for Audit Reports. The fact that the agents who formerly serviced these orphan clients have not joined PCI all the more shows the usefulness of the information in marketing the PMS targeted at these AXA's policyholders.

96. It is well established under the American Cyanamide principle that it is not for the court in an interlocutory application to resolve complicated issues of law or of fact. The explanation given by Ms Leung as to why the 2nd defendant's client lists were delivered up by PCI is far too convenient and coincidental. While part of the 2nd defendant's explanation is credible, for example, the way SAMS was operated in his agency and that he did not personally operate SAMS, the balance of the evidence, in particular, the timing and volume of the printouts and the surrounding circumstances, points strongly to an inference that the 2nd defendant has caused Client Data to be printed out for illegitimate purpose. I am satisfied that AXA have established a serious issue of fact to be tried as against him.

The case against the 3rd defendant:

97. The 3rd defendant was AXA's senior director of agency. He tendered resignation on 10 May 2000 but was terminated by AXA on 18 May 2000. He was registered as a PCI agent on 21 July 2000.

98. AXA's case against the 3rd defendant is that there was an unusually high incidence of printouts of Client Data in April and May 2000 just prior to his resignation, that he imaged a large amount of AXA documents since 16 February 2000 and that some Audit Reports of his agency printed on 5 May 2000 were delivered up by PCI.

99. Upon discovery of the large volume of printouts made in April 2000, Mr Wilson confronted the 3rd defendant on 2 May 2000. The 3rd defendant explained that the printouts were given to the relevant agents and managers. According to Mr Wilson, the agents whom Mr Wilson spoke to denied ever requesting or receiving the printouts.

100. Mr Wilson interviewed the 3rd defendant again on 3 May 2000 and gave him a copy of a note of the meeting of 2 May 2000. After making a few amendments, the 3rd defendant agreed with the contents of the note.

101. On 5 May 2000, Mr Wilson wrote to the 3rd defendant demanding return of all "client files" and enclosing a copy of the note of the 2nd May 2000 meeting as amended by the 3rd defendant which referred to the fact that the 3rd defendant had printed out client data relating to 271 agents. There were no such thing as "client files". The 3rd defendant must have understood that to mean "client lists" or SAMS printouts.

102. On 9 May 2000, the 3rd defendant responded saying that he only had 100 agents and could only have client lists relating to those 100 agents. He said he had authority to keep those client lists. In fact the number of active agents reporting to him were 113, but in addition to printing Client Data relating to those 113 active agents, he had also printed Client Data in relation to 158 terminated agents. The total amount of Client Data he printed out related to 271 agents, but only a portion of the printouts had been returned. Mr Kotewall SC also refers me to various inconsistent allegations made by the then solicitors acting for the 3rd defendant. I do not think it necessary to refer to those inconsistencies here.

103. The 3rd defendant's case is that operation of SAMS was delegated and entrusted to the agency secretary, Ms Fanny Chow, to whom requests for printouts could be made. She was responsible for printing and distributing the printouts to managers and agents concerned through their honeycombs in the office. Ms Chow confirms that client lists printed out in April and not returned were printed out at the request of specific managers without the 3rd defendant's knowledge. This is inconsistent with the 3rd defendant's admission that he printed out or caused to be printed out client lists in April 2000. According to the SAMS log, the printouts were in respect of the same seven units and for exactly the time span for the documents listed in Schedule A. If the 3rd defendant and Ms Chow are referring to the printing of the same documents, then either of them or both of them are not telling the truth.

104. Ms Chow also confirms that client lists printed out in May 2000 and not returned were printed out at the request of specific managers without the 3rd defendant's knowledge. Her allegation is contradicted by three of the managers inquired into by AXA. Her explanation on behalf of the 3rd defendant is all too convenient and coincidental as to be inherently probable.

105. Mr Ho SC submits that it is contrary to common sense for the 3rd defendant to continue printing after he had been confronted by Mr Wilson on 2 May 2000. Mr Kotewall SC replies that it was perfectly in keeping with his wrongful removal of documents to use the time available to print out as much as possible and to engineer a situation whereby it appears that the printouts were being made by others. That was at about the same time when he started asking agents to sign on a document containing just a list of the names of agents, who were given to understand that that would assist him on the question of printouts.

106. While there is no direct evidence that the printouts were authorised or made by the 3rd defendant, the circumstances, including the timing of the printouts and his resignation, the volume of the printouts and that copies are being used for the purpose of enticing serving AXA agents to leave and join PCI, are such that his explanations as well as that of Ms Chow's which are inconsistent with one another and with the other managers of his agency are not to be believed. AXA have therefore established a serious issue on the facts to be tried against the 3rd defendant.

The case against the 4th defendant:

107. The 4th defendant was a sales manager and SAMS password holder. He tendered resignation on 20 June 2000 but was terminated by AXA on 29 June 2000.

108. AXA's case is that client lists of all the in force agents of four units under him, client lists of all terminated agents in his unit and client lists of all agents in Ms Wendy Pan's units were printed out under his password in May and June 2000.

109. His explanation is that the client lists printed in May 2000 were for use by the agents so that they could go through their respective client lists for potential new business because of poor results in competition in March and April 2000. However, his subordinate manager, Ms Wendy Pan, could not recall the 4th defendant having discussed with her about poor results in the competition and the need to seek new business with the use of client lists. As for the client lists of terminated agents, he says they were printed for reallocation of orphan clients. This is not credible as the proper format to be printed for that purpose in the format of Audit Reports.

110. As for the printouts on 20 June 2000, Ms Flora Ng, the 4th defendant's secretary, says that they were printed at her initiative as she was leaving. The explanation as it stands is unconvincing. It offers no reason why the client lists have to be printed indiscriminately merely because she was leaving. That was also the time when the 4th defendant was leaving. The timing could hardly be a coincidence.

111. In the light of the timing and the inherently incredible explanation, the inference is that the 4th defendant caused the printouts to be made for improper motive. AXA therefore have established a serious issue of facts to be tried.

The case against the 5th defendant:

112. The 5th defendant was a senior branch manager of AXA and SAMS password holder. He tendered resignation on 25 July 2000 but was terminated by AXA on 17 August 2000. He was registered as a PCI agent on 28 August 2000. AXA's complaint is that he caused to be printed client lists of all in force and terminated agents in all units in his agency on 9 May 2000.

113. His explanation is that the printouts were made because of poor performance in the competition in March and April 2000 to boost "Smart Medical" and after consultation with Ms Jennifer Wong and Jenny Cheng. Ms Cheng confirmed that while the reason for the printouts had been discussed, no printout had in fact been given to her. His other reasons are that since not all agents had hand held personal computers or digital diary, it is more convenient to issue them hard copies to work with and the printouts would also serve as a reminder to his agents.

114. He gave various reasons for leaving AXA, such as the shrinking in his agency force, the change in AXA's policy requiring a higher threshold to qualify for year-end bonus and in requiring a higher persistence rate, and the decreasing competitiveness of AXA's products. Hence, he says if he was minded to get client lists he would have done so nearer to the time of his departure. However, according to PCI, the offices in Room 701 in Miramar Tower which was rented by PCI's subsidiary, was allocated to the 5th defendant on 24 July 2000, the very day when he tendered his resignation to AXA. Thus the inference is that there were on going negotiations well before July 2000 for his defection. Printing out client lists in April and May would be most timely; in fact printing anytime after February when the indexation has been updated would be timely.

115. The evidence against him for having removed Client Data is so strong that I have no difficulty in finding that AXA have established a serious issue of fact to be tried as against the 5th defendant.

The case against the 6th defendant:

116. The 6th defendant is the wife of the 2nd defendant. She was a sales manager of AXA and SAMS password holder. She resigned on 15 May 2000 and was registered as a PCI agent on 21 July 2000.

117. AXA's case against her is that printouts of client lists of all in force agents in all in force units in her agency were made under her password between 10 and 14 April 2000 and there were other printouts of SAMS documents in April and May 2000. She returned some of the printouts but not all. Subsequently, two of her client lists in respect of Mr Wong Min Kok and Mr Tang Kwok Hung were delivered up by PCI.

118. The 6th defendant's explanation is, like that of the other managers', that she did not concern herself with the operation of SAMS which she entrusted to her secretary, Ms Annie Wong. In brief, Ms Wong would of her own initiative print and distribute all client lists for use by agents in competition and in marketing Smart Medical. Unlike the case of the other defendants, the majority of client lists so distributed were retrieved from the agents and returned to AXA by Ms Wong upon AXA's demand. This made her explanation more credible than those of the other defendants. As for printouts of birthday lists and labels which were not returned, her explanation is that AXA had not made a timely request for their return and had AXA done so, those printouts could have been accounted for.

119. In respect of photocopies of Wong's and Tang's client lists, the 6th defendant's explanation is that Wong is her close friend and in order to provide personal service to Wong, she kept a copy of Wong's client list in her own file, while Tang is her nephew who had defaulted on two instalments of the premium which she paid on his behalf and she kept a copy of Tang's client list for the purpose of obtaining reimbursement from him. After she joined PCI, they requested to follow her and out of convenience, she sent the client lists to PCI.

120. Much of her explanation is credible. The Policyholder's Particulars in respect of Wong and Tang could be outside the terms of Client Data but not the Policy Details. On her own admission she has misused AXA's Client Data in relation to Wong's and Tang's Policy Details. In respect of the birthday lists and labels not returned, she has not offered any credible explanation for printing them in the first place and in accounting for their whereabouts. In respect of those printouts and Wong's and Tang's Policy Details, AXA have established a serious issue of fact to be tried as to whether they have been removed by the 6th defendant for use in competition with AXA.

The case against the 7th defendant:

121. The 7th defendant was AXA's sales manager and SAMS password holder. He resigned on 19 June 2000 and was registered as a PCI agent on 3 August 2000. AXA's complaint is against the unusually high volume of printouts made under his password in December 1999. The printouts were of client lists and client list labels of all in force agents in each of his 12 in force units and of all the terminated agents in 11 of those units and printouts of client list labels in respect of all terminated agents in two other units. In addition, AXA say that he was dissatisfied with Chow Tai Lin's promotion which provided the motive for his leaving AXA and making the printouts.

122. The 7th defendant's explanation is that his motive for leaving AXA had nothing to do with Chow's promotion but was because of changes in AXA's policy, including the introduction of Company Direct, and the declining competitiveness of AXA's products etc. As against that explanation, there are conflicting evidence from AXA that he had intimidated to others his displeasure about Chow's promotion. These disputes are only of marginal utility only.

123. His other explanations are that he did not concern himself with the operation of SAMS which was left to his secretary, Ms Crystal Lam, who would make printouts and distribute the client lists and client lists labels to agents and that the printouts in December was not unusual. He also instructed printouts in respect of his own unit as it was not performing well and he wanted to give the printouts to his agents to push for more business. He kept his own client lists and the client lists of terminated agents and did not take them away upon leaving AXA.

124. At least in respect of the printouts of his own unit and those of the terminated agents there is no dispute he had kept them in April or May 2000. He resigned on 19 June 2000 but was unable to account for them. They could not be found despite a search of his office. A person in his position must recognise the importance of those client lists and the confidentiality attached to them. It is incredible that he would have totally disregarded the safe-keeping of those printouts and is unable to account for them. The inference is that he had taken them away. That also casts doubts on the credibility of his other explanation. This must mean AXA have satisfied me that there is a serious factual issue to be tried.

The case against the 8th defendant:

125. The 8th defendant is AXA's director of agency and SAMS password holder. He tendered resignation on 30 May 2000 but was informed that he was no longer required to report for duty since 7 June 2000. He was registered as a PCI agent on 25 July 2000.

126. AXA complain that large volume of printouts were made under his password between 22 and 27 June 2000. His explanations are that he did not concern himself with the operation of SAMS which was delegated to his office manager, Ms Ivy Lai, and that he was relieved from duty since 7 June and seldom returned to his office after 11 June 2000 while the printouts were made between 22 and 27 June 2000. Ms Lai corroborates his evidence and says that two copies of the 8th defendant's client lists were printed between 22 and 27 June 2000 at the request of Ms Connie Ng because one copy was required for allocating the policyholders of the 8th defendant who have become orphan clients upon his departure and another copy was required by AXA's regional director, Mr Y.K. Chan. In respect of the five incidents of printouts of the 8th defendant's own client list, she explains that they were printed on two occasions on 22 and 26 June 2000 when the 8th defendant was required to answer clients' inquiries. She says that the printing time was only 5 minutes which indicates that the printouts were of a small quantity and could not have related to printouts of all of the 8th defendant's client lists.

127. AXA offer no affirmation by Mr Y.K. Chan or Ms Connie Ng in rebuttal. This together with the timing of the printouts and the 8th defendant's resignation and Ms Lai's affirmation would have been most beneficial to the case of the 8th defendant. However, Ms Lai's answer in relation to the printouts of the 8th defendant's own client lists is obviously not credible. According to Mr Wilson, the 5 minutes referred to in the SAMS log indicates the formatting time and not printing time, which means a lot of data was being formatted for printing. I have no reason to disbelieve this. This casts serious doubts on whether Ms Lai was attempting to conceal the truth. Secondly, her evidence also discredits the 8th defendant in that it becomes obvious that even well after 11 June 2000, the 8th defendant was very active in his agency office. It could well be that he was engaged in removing Client Data after his resignation. I am satisfied that as against him, AXA have raised a serious issue of fact to be tried.

Case against the 9th defendant:

128. The 9th defendant was a sales manager of AXA and SAMS password holder. He resigned on 15 May 2000 and registered as a PCI agent on 21 July 2000. AXA's case is that shortly before he tendered his resignation, printouts were made under his password of client lists of his own unit on 12 April 2000, client lists of Mr Andrew Ng's unit on 12, 13 and 19 April 2000 and client lists of Ms Becky Cheung's unit on 14 and 19 April 2000.

129. The 9th defendant's explanation is that the printouts were made at his instruction because of the competition in March and April 2000, for marketing Smart Medical and because of poor performance in March 2000 or at the request of Mr Andrew Ng and Ms Becky Cheung. His explanation is contradicted by Mr Ng who says that he never requested for printouts on 12 and 13 April 2000 nor has he been consulted about the need for the printouts and he has never been given any printouts. He did request for printouts for his unit and Ms Cheung's unit, but that was on 19 April 2000; and if he had been supplied printouts on 12 and 13 April 2000, he would not have made the request on 19 April 2000. As the evidence stands, an unusually large volume of printouts have been made under his password and his account is inherently improbable. That must mean AXA have raised a serious issue of fact to be tried as to whether he has removed those printouts.

Case against the 10th defendant:

130. The 10th defendant was AXA's sales manager. He tendered resignation on 1 August 2000 but was terminated by AXA on 30 August 2000. Printouts of client lists for all agents in his 14 units were made in November 1999 and for all agents in four units were made in December 1999. AXA's case is that he was unhappy with AXA and was intending to leave in December 1999 which supplied the motive for making the printouts, which coincidentally were made in November and December.

131. The 10th defendant denies harbouring any intention to leave AXA in December 1999 and says it would be more natural to make the printouts in June or July 2000 if he was minded to leave with the printouts. His reason for leaving AXA was because of reduction in the number of agents in his unit and the possibility of a demotion. He says that the printouts were required to promote business around Christmas. Ms Amy Mok corroborates his evidence that printouts of client lists and client list labels were made each year in November and December for the whole agency for sending Christmas cards to policyholders. However, AXA argue that Christmas is not a good time for promoting business as many people are away or have seasonal expenses to incur. In any event, he could not explain why client lists and not just client list labels were printed.

132. In view of the unexplained and extremely high incidence of printouts of client lists and particularly of client list labels, and all the surrounding circumstances, I am satisfied that AXA have raised a serious issue of fact to be tried.

Conclusion:

133. In conclusion, I am satisfied that as against all the Individual Defendants, AXA have raised a serious issue of fact to be tried as to whether they had removed Client Data during the currency of their agency for use in competition with AXA after termination of their agency.

BREACH OF CONFIDENCE BY PCI:

134. There being no fiduciary relation between AXA and PCI, AXA's claim against PCI has to be based on breach of confidence: see Coco v. A. N. Clark (Engineers) Ltd [1969] RPC 41. In accordance with the principle in Coco v. Clark, in order to succeed AXA have to prove:

(1) the Client Data which AXA seek to protect have the necessary quality of confidence about them;

(2) PCI must know that the information is confidential;

(3) it must have been imparted in circumstances importing an obligation of confidence;

(4) there must have been an unauthorised use of the information to the detriment of AXA; and

(5) PCI must have been dishonest in using the information.

135. AXA have already established there is a serious issue to be tried as to whether the Client Data which AXA seek to protect are trade secrets or equivalents to trade secrets. Trade secrets or their equivalents must have the necessary quality of confidence about them. There is no doubt that PCI's use of the Client Data in effecting the PMS targeted at AXA's policyholders is unauthorised and detrimental to AXA. The outstanding issues are therefore whether PCI know that the Client Data are confidential, whether the Client Data have been imparted in circumstances importing an obligation of confidence on PCI and whether PCI has been dishonest in using the information.

PCI's knowledge of confidential nature of Client Data:

136. Like AXA, PCI itself is in the insurance business. It has a well established and sizable operation. Its agency contracts with its own agents contain similar provisions on confidentiality as in AXA's Agent's Contracts. In its letter to AXA's solicitors, it stressed the confidential nature of its own client data delivered over to AXA in accordance with its own undertaking. Such information is the life blood of an insurance company. Therefore PCI itself must appreciate the confidential nature of the Client Data in the eyes of an insurance company.

Whether information imparted in circumstances importing an obligation of confidence:

137. Mr Fung SC submits that PCI is a third party and has no knowledge that AXA's Client Data were being used. He draws a distinction between SAMS printouts and SAMS data and argues that SAMS data are also found legitimately elsewhere, for example, from the policyholders or from the ex-AXA (now PCI) agents' books and records. He further argues that even in respect of the SAMS printouts, PCI could not have known of its confidential nature as the PMS applications, with or without SAMS documents are processed mechanically by clerical staff of PCI.

138. I can well recognize the force of these arguments from the stand point of PCI as a third party recipient of the information as distinct from the case against the Individual Defendants who were in direct fiduciary relationship with AXA. However, PCI is not an ordinary third party. Being in the insurance business itself, PCI must know these information, particularly the Policy Details, wherever they came from, must be trade secrets or confidential information. They could originate from the policyholders as Mr Fung SC suggests. This is a possibility but is hardly supported by any evidence. Quite on the contrary, there is evidence from affirmations filed on behalf of the Individual Defendants explaining how some of these documents have found their way from AXA, not from the policyholders, through the Individual Defendants and their staff to PCI. The inference is that these documents surrendered by PCI and possibly other Client Data used in the PMS applications had been removed by the Individual Defendants in breach of their duty of fidelity owed to AXA.

139. As for whether PCI or its PMS processing staff knew whether the information were supplied in circumstances involving breach of duty of fidelity by the Individual Defendants or ex-AXA agents, Mr Kotewall SC submits that Victor Ko who heads the various departments of PCI processing documentation for PMS applications had been an ex-AXA employee actively involved in launching SAMS and would have no difficulties in recognizing SAMS document, even though he had left AXA before SAMS was fully implemented.

140. In my view, it matters not whether PCI or its clerical staff processing the PMS applications knew the information was supplied under circumstances involving breach of duty. The Individual Defendants' (now PCI agents') knowledge of their own breach of duty to AXA could be readily imputed to PCI. If so, PCI would become bound by the obligation of confidence as well: AG v Guardian Newspaper, [1990] 1 AC 109 at 261 and 281.

141. In my view, AXA have demonstrated a strong arguable case that PCI has knowledge that the information supplied by the Individual Defendants or other ex-AXA agents have the necessary quality of confidence in them and were imparted in circumstances importing an obligation of confidence to PCI.

Dishonesty:

142. Mr Fung SC submits that in order to succeed, the plaintiff must prove dishonesty on the part of PCI in using the information and negligence is not enough. He refers to Royal Brunei v. Tan [1995] 2 AC 378. Mr Kotewall SC does not dispute the correctness of the above proposition of law. However, the test of dishonesty in this context is objective. At p 395, Lord Nicholls held:

"Before considering this issue further it will be helpful to define the terms being used by looking more closely at what dishonesty means in this context. Whatever may be the position in some criminal or other contexts (see, for instance, Reg v. Ghosh [1982] Q B 1053), in the context of the accessory liability principle acting dishonestly, or with a lack of probity, which is synonymous, means simply not acting as an honest person would in the circumstances. This is an objective standard."

143. PCI is in the insurance business. It must have known that the Client Data are confidential information. It planned the PMS specifically targeted at AXA's policyholders. As its annual reports show, PCI recruited on a mass scale teams of ex-AXA agents into their agency force and paid them such hefty welcoming bonus that it declared a profits warning. Between 15 May 2000 and 7 December 2000, 327 AXA agents resigned with 259 joining PCI. The Individual Defendants, in particular, all joined PCI in July or August 2000 when the modified PMS was implemented. All these could not have been coincidental. The inference is that these ex-AXA agents were lured to join PCI to market the PMS targeted at AXA's policyholders formerly serviced by them.

144. It is all too obvious that any of these agents cannot remember the Policyholder's Particulars and Policy Details of the 330 policyholders they serviced. In the circumstances PCI must appreciate there is a real likelihood that the SAMS documents, the Schedule C documents, Policy Details and Policyholder's Particulars, which PCI also recognizes as confidential, filed by these ex-AXA (now PCI) agents in connection with the application for PMS may have been supplied by these agents in circumstances involving breach of their duty of fidelity owed to AXA. If in the circumstances PCI used the information without making any inquiry lest it found out anything it did not wish to know, then such reckless use of the information must be dishonest.

145. Further, from all the circumstances, including the timing of the PMS and the departure of the Individual Defendants, the indiscriminate printout of SAMS data by them, the actual use of SAMS documents in a number of policy applications, the very hefty welcoming bonus paid to the ex-AXA agents, the inference that PCI intended that these ex-AXA agents to bring along AXA's policyholders and if necessary Client Data from AXA to enable these policyholders to switch to PCI's PMS could be drawn. PCI's use of AXA's Client Data under the circumstances must, on an objective standard be dishonest and not merely negligent.

146. Mr Fung SC submits that he knows of no case where an injunction was issued against a third party competitor in receipt of confidential information. I think every case depends on its own facts. The case as presented by AXA is that PCI is not a bona fide recipient of the information. For the purpose of the present application, it is sufficient for AXA to establish a serious issue to be tried as to whether PCI is under an obligation of confidence in respect of the information it received and has been in breach of that duty. I am satisfied that they have so established.

UNLAWFUL INTERFERENCE:

The ingredients of the tort:

147. AXA's second cause of action against PCI is the economic tort of unlawful interference. The basic ingredients of this tort as set out by Henry J in Barretts & Baird v. Institution of Professional Civil Servants, [1987] IRLR 3 are:

(1) there should be interference with the plaintiff's trade or business;

(2) by unlawful means;

(3) with the intention to injure the plaintiff; and,

(4) the plaintiff in fact suffered injury as a result.

148. AXA rely on the PMS as the unlawful interference. The scheme is targeted at AXA's policyholders and induces them to surrender their policies for PCI policies. There can be no doubt that if the scheme is successfully implemented, AXA will suffer injury by way of loss of profits to be derived from premium income in respect of each policy surrendered because of the scheme. The outstanding issues are whether there is intention on the part of PCI to injure AXA with the scheme and whether the interference is by unlawful means.

Intention to injure AXA's interest:

149. Mr Fung SC concedes on the authority of Lonroho plc v. Fayed, [1990] 2 QB 479, that the intention to injure AXA need not be the predominant purpose. It is sufficient so long as the unlawful act is in some sense directed against the plaintiff or is intended to harm the plaintiff. The August 2000 PMS was targeted at AXA's policyholders and was so described in the PMS documents, it must be directed against AXA. Prior to launching the scheme, as disclosed in their annual reports, PCI had planned the mass recruitment of AXA agents and paid them hefty welcoming bonus. Of the 327 AXA agents who resigned by 7 December 2000, 259 joined PCI. A training session was arranged for ex-AXA and intending PCI agents in August 2000, and the client lists of some of the AXA agents were available to the PCI agents recruiting them. It is well known among the industry that policyholders would follow the agent. PCI must know that by planning and implementing the PMS targeted at AXA's policyholders and mass recruitment of their agents would result in injury to AXA. All these surrounding circumstances support the inference that the August PMS was more than a mere competition but an interference directed at AXA with the purpose of inducing AXA's policyholders to surrender their policies thereby occasioning harm to AXA. While this is not conclusive evidence that the PMS was set up with the predominant motive of causing harm to AXA, I am satisfied that AXA have raised a triable issue that the intention of PCI must have gone beyond fair and honourable competition and that PCI intended at least to cause some harm to AXA.

Breach of confidence as the unlawful interference :

150. AXA rely on breach of confidence, representations made by way of deceit and representations made in breach of the Insurance Companies Ordinance, Cap 41 and the Code of Practice for Administration of Insurance as the unlawful means.

151. Mr Kotewall SC refers to Indata Equipment Supplies ltd v ACL Ltd, [1998] FSR 248, as authority that breach of confidence could constitute the unlawful means for this tort. In that case, the defendant finance house disclosed the plaintiff's confidential information to the plaintiff's client with the result of cutting the plaintiff out of the financial deal. Otton and Owen LJJ said obiter that such breach of confidence coupled with the ruthless conduct of the defendant amounted to unlawful means for the tort of unlawful interference with contract. Mr Fung SC argues that Indata concerned a simple bipartite situation involving the original confidee and confidant of the information whereas the present case is concerned with a third party recipient of confidential information. He also submits that the observations of Otton and Owen LJJ are in any event obiter.

152. Mr Kotewall SC refers to the following dicta of Lord Keith of Kinkel in AG v Observer Ltd at 261:

"The third party to whom the information has been wrongfully revealed himself comes under a duty of confidence to the original confider."

He submits that where the conditions required for imposing a duty of confidence on a third party are satisfied, the third party itself comes under its own duty of confidence. I agree. A third party receiving confidential information knowing the information to be confidential and supplied under circumstances involving breach of trust or duty of fidelity is bound in his conscience and is under a duty of confidence himself to the owner of the information. The law imposes on him the same obligation as that obliging the confidant. Similarly, if he receives the information suspecting it to be confidential or supplied to him under circumstances involving breach of trust or duty of fidelity, he is under a duty to enquire. If he uses the information, he may not escape liability by turning a blind eye just in case his inquiry would reveal to him anything that he did not wish to know. In my view, there should not be any distinction between the original confidant and a third party recipient of confidential information with knowledge that the information was supplied to him in circumstances involving breach of duty of fidelity by the original confidant. The dicta of Otton and Owen LJJ are clearly sufficient to raise at least a serious question to be tried as to whether PCI's breach of confidence constitutes the unlawful means for this tort. This is a question of law calling for detailed argument and mature consideration.

Misrepresentation as the unlawful interference:

153. Secondly AXA allege that the deceit practised by PCI on AXA's policyholders through their representation that PCI policies under the PMS match with AXA policies is an unlawful means constituting the tort of unlawful interference. While AXA's expert actuary, Mr Austin, cannot come to the conclusion that the PMS in its entirety is misleading, he is able to reach the conclusion that the scheme in so far as it provides for the making of statements or representations to policyholders that a corresponding policy of PCI "matches" the policy of AXA, is wrong in relation to nine types of policies as listed in their Statement of Claim. In assessing whether a PCI policy could be described as "matching" a corresponding AXA policy, his test is essentially whether there are material differences between the two policies.

154. There is no expert evidence from PCI to contradict the evidence of AXA's expert. Mr Fung SC submits that "matching" is not a term of art that calls for explanation by an expert. PCI's case is that "matching" does not mean identical or that there are no material differences between the matching policy and the policy to be matched. PCI therefore accepts that there may have been material differences between their policies and AXA's. The question is one of degree. Thus any statement by PCI or their agents about "matching" must arguably be false and dishonest. In my opinion, for the purpose of the present proceedings, Mr Austin's expert evidence is sufficient to raise a serious question to be tried.

155. Mr Fung SC submits that the issue is not whether a particular PCI product can be described as "matching" a corresponding AXA product but rather whether PCI has through its agents made false and dishonest representations. He argues that such evidence is lacking and that AXA's case is based on nothing but surmise and speculation. AXA's application is essentially a quia timet application. For the purpose of the present proceedings, AXA need not prove actual instances of the making of misrepresentation, or damages in order to succeed. In my judgment it would be sufficient if AXA can prove a real likelihood of future unlawful interference.

156. While direct evidence of representation to AXA's policyholders is missing, a large number of AXA's policyholders have surrendered their policies and switched to PCI policies following their agents' transfer to PCI. In all but a few cases, the PCI policies were arranged by an ex-AXA agent. There is also evidence from AXA's agents who attended PCI's training courses that agents were asked to approach AXA's policyholders and seek to persuade them to surrender their policies with AXA and replace them with policies issued by PCI on the basis that the PCI policies would match the AXA policies to be surrendered.

157. There is conflicting evidence from AXA and PCI as to what happened during the training courses. In brief, AXA say the agents were trained and told to sell policies on the basis that the basic plans matched or were identical to the AXA products as set out in the handouts. No differences between the policies have been identified or explained to the agents. The agents were not provided with handbooks to enable them to understand the differences. These allegations are denied by PCI.

158. I do not think it necessary for me to compare the AXA policies against PCI policies or AXA's allegations against PCI's, or Mr Austin's report against PCI's criticisms. At this stage of the proceedings, it is not necessary for the court to make any findings as to proof of fraud or deceit, nor do AXA have any burden to adduce such proof save to raise a serious question to be tried. PCI does not dispute there are material differences between the policies, however, they marketed them as matching policies. To describe them or to cause their agents to describe them as matching policies to AXA's policyholders must be a false representation. PCI must know it was false as it recognised there are material differences between its policies and AXA's, or had no belief in its truth or was reckless as to whether it is true. There is evidence that their agents were trained to market them as identical or matching policies with no explanation having been given as to the differences and with no handbook to assist them in understanding the differences between the polices. One of AXA's policyholders recalls that she was approached by a PCI agent and told that the PCI replacement policy would be exactly the same as AXA's; while another AXA's policyholder says that the differences were not explained to her. I am satisfied that AXA has sufficiently raised a serious question to be tried based on misrepresentation on the part of PCI's agents.

Breach of statute as the unlawful interference:

159. Thirdly, AXA rely on breaches of Insurance Companies Ordinance and the Code made thereunder by PCI agents as the unlawful means. Section 56 of the Insurance Companies Ordinance makes it a criminal offence for any person to make any misrepresentation which he knows to be false, misleading or deceptive, or to recklessly make any misrepresentation which is false in order to induce or to attempt to induce another person to enter into any contract of insurance. Article 62 of the Code requires inter alia the agent to explain the differences when making comparisons with other types of polices and not to make any inaccurate or misleading statements about any policies or make inaccurate or misleading statements or comparisons to induce an insured to replace existing long term insurance with other long term insurance to the insured's disadvantage.

160. Even assuming there is evidence of breach of section 56 and the Code, Mr Fung SC argues that AXA's claim under this third limb must fail in limine because as a matter of law breaches of a penal statute cannot constitute the "unlawful means" for the purpose of this tort, unless the enactment in question, on a proper construction, confers a private right of action on the plaintiff. He refers to Lonrho v. Shell Petroleum Co Ltd (No2) [1982] AC 173 at 185 and Mohammed Yaqub Khan v. AG [1987] 1 HKLR 145 in support of his proposition. He also submits that even in Associated British Ports v Transport and General Workers Union [1989] 1 WLR 939 which is relied upon by AXA, Stuart-Smith LJ made a "U-turn" in his view in Associated British Ports.

161. In Associated British Ports, a statutory scheme was set up to regulate the supply of dock labour in ports registered under the scheme. Following a breakdown in negotiation between the union and the employer, the union balloted its members to take industrial action. The employers issued a writ claiming an injunction restraining the union from instructing or inducing its members to act in breach of their contract of employment or in breach of provisions of the scheme (the first tort); and damages for unlawful interference with the employer's business (the second tort).

162. After referring to Lonrho v. Fayed and Lonrho v. Shell, Neill LJ held there was an issue to be tried as to whether the unlawful means can include inducement of breaches of duty which are not actionable. He said at 955:

"Nevertheless, I must proceed to consider the case on the basis that it is arguable that the second species of tort relied on it is not necessary to show that the breach of statutory duty is actionable at the suit of the plaintiff. Accordingly, there is a serious issue to be tried within the American Cyanamid test....

Let me pause to explain the stage which I have now reached. I need make no further reference to the equitable claim. (a) I consider that for the purpose of the first tort it would be necessary for the employers to show that the breach of duty was actionable by them. (b) I consider that in relation to the second tort there is an issue to be tried as to whether unlawful means can include inducement of breaches of duty which are not actionable. (c) Without expressing a final view, I am at present of the opinion that a breach of the obligation imposed in clause 8(5)(b) is not actionable at the suit of the employers.

It follows, therefore that I am expressing no final view whatever as to whether a strike would be lawful or unlawful. At this stage all that can be said is that it is arguable that the employers have a claim in tort which would not require them to prove that a breach of the clause 8(5)(b) obligation was actionable."

163. Butler-Sloss LJ also reached the same conclusion. In fact, she dealt with similar arguments now raised by Mr Fung SC in relation to Lonhro v Shell. She said at 960:

"It is argued for the union that whatever may be the earlier position, since Lonhro v Shell an action under this head, where the unlawful means consists of a breach of statutory obligation, can only succeed if the breach of statutory obligation itself is capable of a right of action between the obligee and obligor. In other words, the same requirement as under the first tort of inducing a breach of statutory obligation. The basis for this submission is the rejection by Lord Diplock, with whom the other members of the House of Lords agreed, of the wide proposition advanced by Lord Denning MR in Ex parte Island Records Ltd [1978] Ch 122. In his formulation Lord Denning MR undoubtedly gathered up and placed in the same bracket both intention to interfere with business by unlawful means and cases where such intention was not required. It was expressed by Lord Deiplock in Lonrho Ltd v Shell Petroleum Co Ltd (No2) [1982] AC 173, 187:

'that whenever a lawful business carried on by one individual in fact suffers damage as the consequence of a contravention by another individual of any statutory prohibition the former has a civil right of action against the latter for such damage.'"

164. She continued at 961:

"If it is necessary for the unlawful means to constitute a cause of action available to the plaintiff where it is a breach of statutory obligation, it would seem to place this member of the group of these torts in an oddly anomalous position. It adds nothing and is virtually indistinguishable from the first tort which we have considered. Further, it is the only member of the group which requires a cause of action by the plaintiff in respect of the unlawful means. There is no logical distinction easily to be drawn between the various members of this group. It is strongly arguable that where the unlawful act relied upon is a breach of statutory duty it is not necessary to be actionable in tort at the instance of the plaintiff."

165. Thus Butler-Sloss LJ drew a distinction between the two torts, namely inducing breach of contract or of statutory duty and unlawful interference with business where the unlawful means relied upon is breach of statutory duty. It is well settled in respect of the first tort, the plaintiff has to prove that the breach of statutory duty is actionable by the plaintiff. It is equally well established that in respect of the second tort, an element of the tort which has to be proved is an intention on the part of the defendant to injure the plaintiff. That element is missing in the first tort. In my view, these are different torts comprising of different elements. The requirement that the breach of statutory duty must be actionable under the second tort would render that tort meaningless and proof of intention to injure superfluous. As Mr Kotewall SC submits, while the question whether a breach of statute gives rise to a civil cause of action by the victim is a question of legislative intention, the tort of unlawful interference is a tort at common law, any reference to legislative intent is, in my view, likely to cloud the issue than resolve it.

166. In the light of the above distinctions, the case of Lonrho v Shell and Mohammed Yaqub Khancan be readily explained as decisions on breach of statutory duty and not on interference of business by unlawful means.

167. Mr Fung SC submits that Stuart-Smith LJ made a "U-turn" in Credit Lyonnais v. ECGD [1999] 1 All ER 929 on his views in Associated British Ports as commented on in a footnote by the editors of Clerk & Lindsell. I do not consider the comments by the learned editors fair. Firstly, Associated British Ports was not referred to or considered or discussed in Credit Lyonnais. Secondly, Credit Lyonnais is a case about conspiracy and not unlawful interference. Thirdly, that the unlawful means, for the purposes of the unlawful act conspiracy, must be actionable at the suit of the plaintiff was a concession by counsel in that case and not a holding by Stuart-Smith LJ.

168. Mr Fung SC refers to paragraphs 24-106 and 24-107 of Clerk & Lindsell and suggests that the editors have castigated Associated British Ports as wrong. The editors write:

"It remains to be seen whether the House of Lords (which expressed no opinion on these matters in [Associated British Ports] will accept this rationalization of the previous precedents, which may be thought rather strained, effected as it was in interlocutory proceedings where views about the law could not be final."

169. Though the Court of Appeal decision in Associated British Ports has been overturned by the House of Lords on different grounds, I do not think this issue is now beyond dispute.

170. It is not appropriate in an interlocutory application to express a concluded opinion as to the law unless it is reasonably clear. The volume of arguments raised by counsel for the parties clearly demonstrates that this matter falls within Lord Diplock's category of difficult questions of law which requires mature consideration. In my opinion, AXA have, at the very least, raised a serious question to be tried as to whether in a case of unlawful interference where the unlawful act relied upon is breach of statutory duty its breach has to be actionable at the instance of the plaintiff. In conclusion, I am satisfied that AXA have raised a serious question to be tried in respect of both causes of action against PCI.

BALANCE OF CONVENIENCE:

171. In this part of my deliberation, I shall direct my attention to the possible result of a postulated trial, first one way and then the other. If in both eventualities, damages after trial are thought to be inadequate, then I shall consider what is commonly described as the 'balance of convenience' or 'balance of the risk of injustice' as more aptly described by May LJ in Cayne v Global Natural Resources [1984] 1 All ER 225 at 237.

Adequacy of damages:

172. AXA place a price tag of $500 million as their potential loss. According to AXA's appointed actuary, Mr Ross, based on the most recent appraisal of the value of AXA policies in force and assuming that the portfolio of policies held by policyholders of the 153 agents who had already left AXA and registered as agents of PCI as at 5 September 2000 exhibits the same characteristics as AXA's whole portfolio of policies, the value to AXA of the policies serviced by those 153 agents is in excess of $300 million. As at 7 December 2000, 259 departed agents have been registered as agents of PCI, thus the value of the policies to AXA which might be lost would be over $500 million. On Mr Ross' valuation, the loss to AXA as a result of each agent joining PCI would be $2 million or $6,000 per policyholder as each agent services on the average 330 policyholders. Though I am not shown Mr Ross' valuation, it does not appear unrealistic had the policies run their full course and provided all the agent's policyholders follow the agent and switch to PCI policies.

173. Mr Fung SC submits that the evidence shows that the damages payable for using SAMS documents would be negligible or of a low order. He is referring to AXA's analysis that 29 out of 297 PMS applications were accompanied by SAMS documents. I think that is a mis-understanding of AXA's case. What AXA say is that they were able to confirm that 29 of the 297 PMS applications were accompanied by SAMS documents, but their claim is not limited to misuse of SAMS documents, but all data contained therein and derived therefrom, as well as Schedule C documents. Further, it is not invariably the case that a PMS application will be accompanied by SAMS documents. Agents may have misused Client Data in identifying those policyholders, obtaining their contact details, their Policy Details and then approach them for the purpose of marketing the PMS.

174. In all fairness to PCI, AXA's estimate of loss is based on the assumption that all policyholders serviced by a departing agent joining PCI will switch to PCI policies. However, as AXA are not seeking to prevent an ex-AXA agent from using any information known to him before the information was obtained or derived by him in the course of his agency with the plaintiff, AXA's loss as a result of the ex-agent's use of such information not covered by the injunction should not be attributed to the defendants. Though there was a general belief among AXA's own agents that 70% of the policyholders would follow the agents, as a matter of fact not all of them do and AXA are able to produce affirmations from policyholders who for good reasons chose not to follow their agents. While I cannot assess what percentage of the policyholders would follow the agents, I do not think AXA's loss would be in the region of $500 million, nor do I think it would be negligible or of a low order. Even if it is half of AXA's estimated loss, it is very substantial.

175. Mr Fung SC submits that both AXA and PCI are very substantial companies. Each is owned by a parent publicly listed company, in the case of AXA on the Australian Stock Exchange and in the case of PCI on the Hong Kong Stock Exchange. PCI has considerable assets and a free surplus of $238 million for the year ended 31 December 1999. Both PCI and its parent company were given an "A" rating by A.M. Best which according to A.M. Best describes companies having "a strong ability to meet their obligations" and maintaining "a level of financial strength that can withstand unfavourable changes in the business, economic or regulatory environment".

176. On the other hand, AXA criticise A.M. Best's lack of credibility in their rating as they are commissioned and paid by the companies they rate and are less independent. A.M. Best's rating was made without taking into account the large amount of welcoming bonus offered to agents as a result of which a profit warning was issued by PCI on 4 September 2000. AXA refer to Standard and Poor's rating in which PCI was given a "B" rating whereas AXA was given an "A" rating.

177. While I am prepared to find that the loss to AXA would be substantial and of the order of millions of dollars, I am far from being convinced that AXA has discharged the burden of proving that PCI is unlikely to be able to pay. In view of PCI's free surplus and the satisfactory rating by both A.M. Best and Standard and Poor, I am satisfied that PCI has the financial resources to meet the damages if awarded.

178. In respect of difficulties in quantification, Mr Fung SC submits that damages for breach of confidence involving commercial information, as opposed to claims for misuse of information relating to the affairs of private individuals are capable of assessment and that difficulties in calculating damages should not inhibit the Court from making a calculation. AXA have an advanced computer system, SAMS, which is capable of identifying which policyholders have switched to PCI and from which an actuary can calculate the loss to AXA. On the other hand, he submits, business foregone by PCI if injunction is granted, would be difficult or impossible to assess.

179. Difficulties of quantification is not just limited to mathematical difficulties, but also applies to difficulties of proof. Damages would hardly be an adequate remedy because of difficulties of proof. The burden for AXA is not just identifying which policyholders have switched and then applying actuary formula. Now that litigation has started, agents who are minded to misuse the information would not do the obviously undesirable thing of annexing SAMS documents to the policy applications. Thus even if AXA can identify the policyholders who have switched, which they would have no difficulties, AXA may not know from their computer system if SAMS documents have been used. Further AXA's claim is not limited to misuse of SAMS documents, but all data contained and derived therefrom as well as the Schedule C documents. AXA have no means of knowing whether an agent has misused Client Data by merely checking SAMS. Without an injunction AXA would be faced with an almost impossible and mammoth task of investigation, let alone the uncertainty of cooperation from the defecting policyholders. In the event that AXA succeed in their claim, it would be wrong for PCI to say at the interlocutory stage that they should not be enjoined from doing what they should not have done if they are in the position to pay damages; and when they are asked to pay, it would be wrong for them to say "sorry, it is too bad that you do not have the means to prove the loss that you have suffered in my hands." This would make a ridicule of our law. To allow PCI to persist in conduct which is prohibited at the price of paying damages, provided AXA could overcome the very onerous burden of proof set up the agents who are in breach of their duty of fidelity, is not anything that I would countenance.

180. Mr Ho SC submits that the Individual Defendants could be asked to give an account. The fact that the Individual Defendants may keep records of their business does not assist. The records will, at the most, relate to businesses obtained by the individuals, but not how the businesses have been obtained and are unlikely to show what information or documents have been used to make the initial contacts with AXA's policyholders, or to assist in the PMS applications or their processing. The account may not provide information relating to business made by agents who are not parties to these proceedings but to whom the Individual Defendants have disclosed Client Data. None of the Individual Defendants has exhibited any of their books and records to satisfy me of the adequacy of those books and records for the purpose of giving account.

181. If an interlocutory injunction is granted, and if PCI is to succeed at trial, they will be free to implement the PMS. The number of policies it obtains thereafter will be a reliable guide to the extent of its damages. The question of SAMS documents is irrelevant because it will then be PCI's case that, even with no injunction, there will not be any use thereof. Their position if an interlocutory injunction is granted would be much better than AXA's if an interlocutory injunction is refused. AXA's position is almost one of impossibility.

182. In the case of the Individual Defendants, Mr Ho SC refers me to the following dicta of Sir Donald Nicholls V-C in Universal Thermosensors Ltd v. Hibben at 855:

"In these circumstances the substance of the matter is that, by misappropriating and misusing the confidential information, the defendants sought to save themselves the trouble and expense of making a tedious trawl through directories and having to spend hours on the telephone. In respect of misuse of confidential information which had occurred by July 1990, the commensurate remedy is that the defendants should pay the plaintiff for that which they wrongly took. They should pay compensation for their use of the plaintiff's property, for the benefit they received. But as to such past misuse, an injunction aimed at preventing the defendants from deriving any future benefit therefrom after July 1990 would not be an appropriate remedy. Such an injunction would unjustly benefit the plaintiff by putting it into a better position as against T.P.L. [the rival business formed by the Individual Defendants] than if there had been no breach of confidence, and it would drive T.P.L. out of business, in circumstances where a monetary award would provide an adequate remedy to the plaintiff for the wrong done to it."

183. Mr Ho SC submits that applying this principle to the present case, since all the SAMS documents have been delivered up, the appropriate remedy should be an award for damages at trial, rather than an interlocutory injunction at this stage which would unjustly benefit the plaintiff. On the basis that all the SAMS documents have been delivered up, Mr Ho SC is probably right. However, the preponderance of evidence suggests this is an incorrect basis. A large quantity of SAMS printouts have been removed unlawfully and unaccounted for. AXA's legitimate business interest is being threatened. In respect of past breach, damages are appropriate; but AXA are not seeking an interlocutory injunction to protect them in relation to past misuse. In respect of future breach, an innocent party is entitled to the protection of his legitimate business interest by an interlocutory injunction, provided the test in American Cyanamide is satisfied. I do not think the Vice-Chancellor in Universal Thermosensors was saying anything more than that an interlocutory injunction was inappropriate "in circumstances where a monetary award would provide an adequate remedy to the plaintiff for the wrong done to it."

184. Mr Ho SC submits that an interlocutory injunction will have disastrous impact on the livelihood of the Individual Defendants. It is obvious that a trial will not take place before a substantial period of time. Clients are the key asset and life blood of every insurance agent. They will be lost to the Individual Defendants if the injunction is granted and it will be difficult to win them back.

185. However, it is PCI's case that their interest in recruiting the 2nd defendant and to a lesser extent the 3rd, 6th and 9th defendants is not for their or their agencies' existing client base, but their ability to generate new business. As for the Individual Defendants, they are senior managers with few direct clients of their own. Their main job was recruiting, training and developing an agency force. On the strength of what PCI and the Individual Defendants say, I cannot see how their livelihood would be adversely affected by the injunction if granted. As senior managers, they earn overriding commission from agents working underneath them and by recruiting new agents and developing the agency force.

186. Mr Ho SC also argues that the loss to the Individual Defendants will be wholly beyond quantification. If they are enjoined from using the Client Data to contact their clients, it is impossible to tell which of them will follow the Individual Defendants and hence assess their loss. Further they will also be deprived of commission earned from business introduced by those clients. I think the case of the Individual Defendants has been grossly exaggerated. They are senior managers. Contacting clients is not their main function and commission from direct business introduced by them is not a significant part of their income.

187. I have already mentioned the difficulties in quantification of AXA's claim. In addition, there is no evidence as to the means of the Individual Defendants and they do not appear to be well placed to compensate AXA for their loss even if apportioned. Without an interlocutory injunction, the reality is that once an AXA's policyholder switches over to PCI, there is no prospect of his return. The injury to AXA may be unprovable, uncompensatable and irreversible. On the other hand, if an injunction is granted and AXA are unsuccessful at trial, AXA are in a very good position to compensate the defendants for their loss.

188. In conclusion, I am satisfied that if the injunction is refused and in the event that AXA are successful at trial, an award of damages against PCI or the Individual Defendants may be inadequate or illusory. On the other hand, if the injunction is granted and AXA are unsuccessful at trial, AXA would be good for the damages likely to be awarded against them. The balance is in favour of granting the injunction than refusing it.

Balance of risk of injustice:

189. Having come to the above conclusion, it would be unnecessary to consider the 'balance of convenience' or 'balance of the risk of injustice' as more aptly described by May LJ in Cayne v Global Natural Resources [1984] 1 All ER 22. Mr Fung SC, however, argues that in deciding whether to grant the injunction, the court should take into account whether the harm resulting from an injunction will outweigh the benefit. Both Mr Fung SC and Mr Kotewall SC rely on the broad principle expressed by Eveleigh LJ in Cayne v Global Natural Resources that the court should do its best endeavour to avoid injustice.

190. Mr Fung SC submits that I should follow that decision and refuse the interlocutory injunction. That was a case involving minority shareholders' action in which the grant of interlocutory injunction has the effect of depriving Global the right of trial merely on the basis of affidavit evidence. I think each case has to be decided on its facts. AXA's case is the reverse of Global's. If no interlocutory injunction is granted to AXA, it may be difficult for them to ask for an injunction at trial because those policyholders who have already switched are unlikely to switch back and the Court has no jurisdiction to compel their return to AXA. On the other hand, if the policyholders are likely to switch, the effect of the interlocutory injunction, if wrongly granted, would merely be to delay the switching and the number of policyholders switching after trial will be a guide to the loss suffered by PCI. I agree with Mr Kotewall SC that the harm to PCI by the grant of an interlocutory injunction does not outweigh or even match the benefit of any interlocutory injunction at this stage. This argument applies with equal force against the Individual Defendants.

191. As for the merits of the parties' case, I think this is a case in which the applicable legal principles involved require mature consideration while the factual disputes could not be resolved on affidavit evidence. I refrain from passing any view on the merits of the case.

Appropriateness of injunction against PCI as a third party:

192. Mr Fung SC submits that the Court should be chary about granting relief that affects third parties like PCI and fourth parties such as policyholders intending to switch to PCI. I do not think PCI can be described as a third party or a bona fide recipient of the information without notice. AXA's case is that PCI is a recipient and user of the secret and confidential information with the requisite knowledge against whom allegations of unlawful interference is made. There is ample evidence on which inference that PCI has knowledge of the agents' breach of duty of fidelity and that PCI dishonestly used the Client Data could be drawn and AXA have successfully raised serious questions to be tried on those issues. If such inference is drawn, PCI is in the same position as an original confidant.

193. In PSM International Ltd v. Whitehouse [1992] FSR 489, Lloyd LJ held at 498:

"I agree that the courts should be chary of granting an equitable remedy which would have the effect of interfering with the contractual rights of innocent third parties. But that equity has power to do so in an appropriate case. I do not doubt....

The present would seem an appropriate case in which to exercise the wider jurisdiction until trial, for two reasons. In the first place all the contracts made with Willenhall were made at a time when Mr Whitehouse was still employed by the plaintiffs, and therefore (it seems) in breach of his duty of fidelity owed to the plaintiffs, as well as clauses 3(1) and 4 of his contract of service.

Secondly it is open to the plaintiffs to argue on the facts that Mr Ryder, of Caradon Rolinx, must at least have been on notice that Mr Whitehouse was acting in breach of contract, even if he was not actively co-operating. That will be an important issue to be resolved at the trial."

If PCI had knowledge of the Individual Defendants' breach of duty and dishonestly used the Client Data they provided, it would be appropriate for the injunction to be granted against PCI, even though it is not the original confidant.

194. As for the policyholders' freedom of choice of insurers, I do not think it is affected by the injunction sought. Provided no unlawful means are used, AXA do not object to lawful competition. What AXA object is the Individual Defendants making copies of Client Data during their agency for the purpose of using the information therein to the prejudice of AXA after termination of their agency. If by proper marketing campaign, such as advertisement on television or newspaper, AXA's policyholders contact PCI's agents and enquire about PCI's PMS and apply for switching, AXA have no complaint. I do not think the injunction has the effect of affecting the rights of any fourth party.

Delay:

195. Mr Fung SC argues that there had been substantial delay in instituting proceedings by AXA and this delay should weigh against grant of interlocutory injunction in favour of AXA. He refers to the following chronology of events:

Date	Event
1994	PCI introduced Policy Protection Plan
1998	AIA introduced Plan Matching
3/2000	PCI started PMS with AXA policies
4/2000	2nd to 10th defendants started leaving AXA
2/5/2000	3rd defendant confronted by Mr Wilson
12/5/2000	3rd defendant's printouts checked by AXA
30/5/2000	AXA's solicitors commenced injunction preparation against 3rd defendant.
7/2000	2nd to 10th defendants registered as PCI agents
8/2000	Revised PMS began operation
21/9/2000	Writ and present summons issued and served

196. AXA do not complain about competition or policy matching as such. What they complain is about breach of duty of fidelity and duty of confidence. They have no knowledge of the March 2000 PMS, presumably because the March scheme was not accompanied by contemporaneous mass departure of AXA agents. Even according to the above chronology, the alarm bells rang in May 2000 when AXA's suspicion was aroused and confirmed in July 2000 with nine of their top directors or senior managers and two hundred of their agents of various grades resigning and joining PCI. Eventually they came to know about the August 2000 PMS only on about 4 August 2000. For AXA, I think time does not start to run until 4 August 2000 when they have knowledge of the August 2000 PMS or at the earliest July 2000 when they have confirmation that the Individual Defendants have joined PCI. Having regard to the volume of documents, the affirmations now before the Court and the complexity of the issues and evidence, I do not think AXA have taken beyond what I consider as a reasonable time in mounting a case of this nature. I do not think AXA is guilty of such delay as to suggest that they are not entertaining any bona fide fear that they will suffer irreparable damage and hence to disqualify them of the injunction sought.

Status quo:

197. In conjunction with delay, Mr Fung SC refers to the House of Lords decision in Garden Cottage v Milk Marketing Board [1984] AC 130 and argues that the status quo to be preserved is that prevailing immediately prior to the issue or, as refined by Dillon LJ in Graham v Delderfield [1992] FSR 313, service of the writ claiming permanent injunction, i.e. the status quo including the implementation of the various PMS. In Garden Cottage, Lord Diplock held at 140:

"The status quo is the existing state of affairs; but since states of affairs do not remain static this raises the query: existing when? In my opinion, the relevant status quo to which reference was made in American Cyanamid is the state of affairs existing during the period immediately preceding the issue of the writ claiming the permanent injunction or, if there be unreasonable delay between the issue of the writ and the motion for an interlocutory injunction, the period immediately preceding the motion. The duration of that period since the state of affairs last changed must be more than minimal, having regard to the total length of the relationship between the parties in respect of which the injunction is granted; otherwise the state of affairs before the last change would be the relevant status quo."

198. Mr Kotewall SC submits that the above dicta is not inconsistent with AXA's position. He argues that according to the above dicta the status quo to be preserved is that state of affairs existing in the period before the issue of the writ on 21 September 2000. Despite the earlier versions of PMS, the PMS now complained of was revised and implemented in August 2000 and hence, Mr Kotewall SC submits, the status quo must be the state of affairs just before the August 2000 PMS was implemented. He draws support for his proposition by relying on the dicta of Megary V-C in Metric Resources Corporation v Leasemetrix Ltd & Another [1979] FSR 571 at 581 where the Vice-Chancellor says:

"As for preserving the status quo, there is the familiar difficulty of saying what that phrase means. Mr Aldous says it must mean the state of affairs existing at the time the court is hearing the case, so that the status quo that ought to be preserved was that of the defendant company carrying on its business under the name "Leasemetrix," and of the plaintiff company having a mere intention to begin a business in this country under the name "Leasametric." On that footing, it was desirable to keep the first defendant in business under its present name. One problem in this approach is that the status quo may materially change between the hearing at the first instance and a hearing on appeal. The term "status quo" is plainly incomplete on the face of it; and as I suggested in Robbie v Fulham Football Club Ltd., unreported, 26 March 1979, I think the full term is "status quo ante bellum." If the issue of the writ is the notional equivalent of the outbreak of war, that would require matters to be tested when the writ was issued, which sometimes would be capricious or unfair. If the metaphor is pursued, then it may well be that the true status quo ante bellum is the state of affairs which existed immediately before the act which constitutes the casus belli, unless hostilities are delayed so long that the act becomes part of the status quo."

199. Metric Resources Corporation is the earliest of the three decisions but has not been referred to in either Graham v Delderfield or Garden Cottage v Milk Marketing Board. In my opinion, Leasemetrix is not inconsistent with either of these two authorities. In fact it defines with sensible precision the meaning of the term "the period immediately before issue or service" of the writ. Megary V-C identified the proper term to be "status quo ante bellum" and the act complained of as the "causus belli". I do not think the gloss put onto Lord Diplock's dicta by Dillon LJ anything necessarily creating a principle of law. It is no more than a specific application of Lord Diplock's dicta to the facts of that particular case involving undue delay, which in the words of Megary VC, has become part of the status quo. Graham v Delderfield is clearly distinguishable as there was clear delay both in issuing the writ and serving the same.

200. Mr Ho SC submits that upon the resignation of the 2nd defendant on 10 April 2000, Mr Wilson immediately suspected that the 6th defendant who is the wife of the 2nd defendant, the 3rd and 9th defendants were planning to leave AXA. Mr Wilson also became aware of the 3rd defendant printing out large amounts of Client Data and had threaten to apply injunctive relief since 19 May 2000. Yet despite this, no application for injunction was made until 21 September 2000. Hence Mr Ho SC submits the relevant status quo should be the state of affairs existing as at the date of service of the writ. He also added that as there are other competitors in the industry, to allow the Individual Defendants to continue their contact with their clients will not significantly affect the status quo or AXA.

201. AXA are not complaining about their agents leaving and joining their competitors. AXA have no knowledge of the March 2000 PMS and they only came to know about the revised PMS on 4 August 2000. Thus back in April or May 2000, AXA could not be blamed for not taking action when its agents started leaving. In any event it is not the Individual Defendants joining PCI or the PMS as such that AXA are complaining but the Individual Defendants' breach of duty of fidelity and/or making of misrepresentation in implementing the August 2000 PMS, which they do not have knowledge of until early August 2000. It is also not until July or August 2000 that AXA have knowledge that the Individual Defendants have joined PCI when the pieces of jigsaw fell into place. Having regard to the difficulties and complexities in mounting a case of this nature, I consider the amount of time taken by AXA and their legal advisers to prepare the matter properly before issuing the writ and summons is by no means excessive. Accordingly, I find that there is no delay by AXA and hence the status quo to be preserved by the injunction should be, in the words of Lord Diplock "that before the last change" or in the words of Megary V-C, that "which existed immediately before the act which constitutes the casus belli", i.e. the status quo before the August 2000 PMS was implemented.

Utility of Client Data:

202. Mr Fung SC submits that the value of the information in SAMS allegedly in the possession of the defendants decline with time and is stale or of little use as a reason for exercise of discretion against the grant of injunction. The SAMS printouts were allegedly removed by the Individual Defendants in April or May 2000 or earlier. The SAMS data are regularly updated, probably on a weekly basis while the indexation increase endorsement is replaced every February. Hence by now, a year later, Mr Fung SC submits, the information has lost it usefulness. Mr Kotewall SC replies that personal and contact details do not change much over time while most policies are renewed essentially on similar terms. The indexation increase endorsement which is replaced annually only marginally affects the sum insured. I agree. Client's contact details seldom change. Hence, despite passage of time, SAMS data retain much of their usefulness. Of course once an agent managed to contact a policyholder, he is able to obtain all the other information from the policyholder.

203. Mr Fung SC also makes the point that all details of policies needed for the PMS are obtainable from policyholders or by the policyholders making telephone inquiries with AXA. The fact that SAMS documents had found their way into PCI's record is evidence that they are useful and have in fact been used. This confirms the usefulness of the information. PCI's argument failed to address the fact that AXA's injunction is not directed at SAMS documents as such, but also at the information contained therein, including contact details whether contained in SAMS documents or not.

Clean hands:

204. Mr. Fung SC submits that AXA's conduct in this dispute disentitles them from invoking the court's equitable jurisdiction. He is referring to some defamatory statements published in the Asian Wall Street Journal on 20 December 2000 by AXA's Chief Executive and their Senior Manager, Agency Resources. The article is titled "Insurer Accuses Rival of Document Theft" and makes reference to documents delivered up by PCI in this action in conjunction with a suggestion that PCI is guilty of a criminal offence of data theft. The part of the article complained of by PCI reads as follows:

"AXA has filed a formal complaint with the police arguing that the alleged data theft is a criminal offence. The police are investigating, but have declined to comment ... PCI late last month returned some 80 pages of documentation to AXA - client lists with full policy details - that were marked with stamps and other notations from PCI's new-business department, AXA says."

205. AXA's Chief Executive is also quoted in the article as saying:

"The issue here is putting together a scheme that is founded on replacing policies to the detriment of consumers and using data that has been appropriated from us."

206. While Mr Fung SC accepts that AXA is not responsible for the title used in the article, PCI accuses AXA of adopting the article by sending copies of the article or its Chinese translation to policyholders.

207. At the time, the interlocutory injunction application had already been heard by the Hon Madam Justice Beeson and the substantive hearing was scheduled to take place on 3 January 2001 before Deputy High Court Judge Longley. The matters referred to in the article are clearly sub judice. Hence, Mr Fung SC submits the timing of the publication cannot have been coincidental and the irresistible inference is that the article was calculated to influence the minds of its readers, which include, inter alia, the Court, current and former AXA agents and policyholders. The allegedly defamatory statements now form the subject matter of another action by PCI.

208. Mr Kotewall SC replies, citing Lord Parker CJ's dicta in R v. Duffy, ex parte Nash [1960] 2 QB 188 that there is no question of any contempt of court as this is a civil matter and juries are not and will not be involved, while the likelihood of professional judges being influenced at all is extremely remote. As held by Lord Parker CJ in that case, the question is "was there a real risk, as opposed to a remote possibility, that the article was calculated to prejudice a fair hearing." As the matter is to be heard by professional judges who are best trained to put out of their mind matters which are not in evidence in the case, I do not think there is any risk of prejudice. Further, to prove contempt, PCI has to establish beyond reasonable doubt an intention by AXA to impede or prejudice the administration of justice, in the sense that the effect will have to be foreseen at the time as being almost inevitable as a result of the publication. For the above reasons, I do not think there is any real risk of prejudice to a fair hearing.

209. As for PCI's complaint about AXA circulating the article or its translation among their policyholders, AXA's reply is that they were not responsible for the circulation, the act complained of was done by AXA's agents on whom they have no control. Policyholders may be influenced in their decision whether to switch their insurers; but as far as the present proceedings are concerned, whether policyholders are prejudiced is irrelevant.

210. PCI also complains that AXA have singled out those of their own clients who intended to switch to PCI for unfavourable treatment. PCI has filed affirmations from AXA's "defecting" policyholders alleging that AXA have gone to the extent of requiring them to attend videotaped interviews, interrogating them as to their reasons for wishing to surrender their policies, delaying the sending out of surrender cheques and so on. In reply, AXA say that on average AXA issue the cheque for the cash value of the surrendered policy 25 days from the date of receipt of the first documentary request for policy surrender or 27 days where the original producing agent or servicing agent is an agent of PCI. They explain that there were incidents involving fraudulent signatures and forged identity cards which delayed the processing. For reasons as stated below, I do not think it necessary to consider the veracity of PCI's accusation and AXA's reply. The present proceeding is not a trial by affidavit. At the highest, PCI may have this issue resolved at trial.

211. The conduct now complained of by PCI, whether as regards the articles published in Asian Wall Street Journal or AXA's handling of their "defecting" policyholders, do not arise out of the transaction, the subject matter of the present proceedings. In Sang Lee Investment Co Ltd v. Wing Kwai Investment Co Ltd, Lord Brightman held at 208:

"One of the more helpful statements to which their Lordships were referred is to be found in the American case of Weegham v. Killefer [1914] 215 Federal Reporter 168, 171, quoting from an earlier authority: 'A court of equity will leave to his remedy at law - will refuse to interfere to grant relief to - one who, in the matter or transaction concerning which he seeks its aid, has been wanting in good faith, honesty or righteous dealing. While in a proper case it acts upon the conscience of a defendant, to compel him to do that which is just and right, it repels from its precincts remediless the complainant who has been guilty of bad faith fraud or any unconscionable act in the transaction which forms the basis of his suit'. Two conditions are therefore to be satisfied by the litigant who seeks to resist equitable relief on the ground of the misconduct of his opponent. First, such conduct must be wanting in good faith. Secondly, it must be 'in the transaction' which is the basis of the suit."

212. AXA's conduct now complained of, assuming it to be wrongful and wanting in good faith, arose out of a wholly separate event, distinct from the matter now before me. On the authority of Sang Lee Investment Co Ltd v. Wing Kwai Investment Co Ltd, this must be fatal to PCI's plea of AXA's want of good faith and clean hands.

213. In view of the impending action relating to the article itself, I do not think it appropriate for me to consider if the content of the article is defamatory of PCI, if I can at all avoid that issue.

THE INJUNCTION ORDER:

214. I am satisfied that AXA have raised serious issues to be tried and that the balance of convenience lies in favour of granting than refusing the injunction. AXA have not been guilty of delay or such unconscionable conduct as would disqualify them from this equitable remedy. The status quo to be preserved by the injunction order is that before PCI's launching of the August 2000 PMS.

Term of injunction:

215. Both Mr Fung SC and Mr Ho SC refer to Roger Bullivant v Ellis [1987] FSR 172 and Sun Valley v Vincent [2000] FSR 825 and submit that where documents have been wrongfully removed the appropriate form of relief is a "springboard injunction". The purpose of the injunction is to prevent the employee from gaining an unfair head start in his competing business by misuse of information. Its purpose is not to punish the employee and should go no further in its scope and last no longer than is necessary to cancel out the unfair advantage. They take the view that by the time of hearing nine months have lapsed and that would have already been a restraint longer than necessary. They submit that an injunction until trial would have the effect of driving the Individual Defendants out of business.

216. On the other hand, Mr Kotewall SC argues that this is not a springboard situation. Both Roger Bullivant v. Ellis and Sun Valley v. Vincent are springboard cases. Indeed in Sun Valley, Parker J distinguished Universal Thermosensors expressly on the basis that it was not a springboard case and in which interlocutory injunction until trial or further order was properly granted.

217. Inherent in the idea of a springboard is that the information is otherwise in the public domain available to anyone who is minded to spend the time and effort to search or compile the information. The springboard injunction prevents an employee from stealing his employer's confidential information entrusted to him and thereby gaining an unfair head start. In Terrapin Ltd v. Builders' Supply Co (Hayes) Ltd [1960] RPC 135 the following dicta of Roxburgh J was cited with approval by May LJ in Roger Bullivant:

"As I understand it, the essence of this branch of the law, whatever the origin of it, is that a person who has obtained information in confidence is not allowed to use it as a springboard for activities detrimental to the person who made the confidential communication and springboard it remains even when all the features have been published or can be ascertained by actual inspection by any member of the public ... The possessor of the confidential information still has a long start over any member of the public ... It is, in my view, inherent in the principle upon which the Saltman case rests that the possessor of such information must be placed under a special disability in the field of competition to ensure that he does not get an unfair start."

218. In the instant case, the information concerned are trade secrets or their equivalents which are not legitimately available elsewhere, except from AXA or AXA's policyholders. On the evidence, the defendants would not be able to compile the Policyholder's Particulars and their Policy Details from any public source. The consideration of unfair head start is irrelevant in the instant case. I agree with Mr Kotewall SC that this is not a springboard situation. Any interlocutory injunction to be granted must be for such period as would last until trial or further order. If AXA fail to establish their claim at the conclusion of the trial, the defendants may look to AXA's undertaking and I am sure AXA would be good for their undertaking.

Scope of relief:

219. Mr Fung SC and Mr Ho SC criticise the relief sought by AXA as being exceptionally and unacceptably wide as to be incapable of complying. Mr Ho SC submits that the injunction should be "source-specific", i.e. it should only be directed against the misuse of AXA's materials wrongfully taken away and should not be so wide as to prevent the Individual Defendants from using or resorting to lawful sources.

220. The injunction proposed in the summons has been narrowed down by the revised definition of "Client Data". In order to be caught within the scope of "Client Data" and hence trigger the injunction, the information must have been, firstly, obtained or derived by an agent in the course of his agency with AXA, not being information known to the agent prior to such obtaining or derivation and secondly, it must be contained in SAMS. Thus the information would only be protected if it is traceable back to the obtaining or derivation. If the information or Schedule C documents are obtained from the policyholder himself, the information or documents would not have been derived by an agent in the course of agency, whether the information happens to be contained in SAMS or not. The present formulation of "Client Data" is already "source-specific", deriving only from policyholders qua policyholders, or from agents qua agents, or from AXA. I am satisfied that the present formulation has addressed the defendants' concern that they be allowed to use information or documents which are legitimately available. I agree with Mr Kotewall SC's submission that it is this misinterpretation of the scope of "Client Data" which undermines the argument of PCI and the Individual Defendants concerning the scope of the relief sought. I do not agree with Mr Fung SC and Mr Ho SC that the relief sought is too wide, oppressive and unworkable.

221. As for PCI's worry that it has no way of knowing if the information provided to it is coincidentally contained in SAMS, I think the difficulties have been exaggerated. The PMS is set up with AXA's policyholders in mind. PCI has set up a special task force staffed by ex-AXA agents to screen the applications for the PMS. PCI only need to ascertain from its agents the source of the information. It must be incumbent upon PCI to ensure that in seeking business on their behalf and for their benefit their agents do not breach the law. PCI is in the position to require full and proper disclosure by its agents of their source of information, specifically whether any has been derived from SAMS. A proper screening process with cross-checking should ensure compliance.

222. In view of the reformulation of the definition of "Client Data", I do not think the Individual Defendants have any difficulties in complying with the terms of the injunction being sought. The present formulation of "Client Data" has sufficiently addressed the Individual Defendants' concern where a policyholder approaches an agent about switching and the process of switching is completed without the use of any documents or information obtained by the agent in the course of his agency with AXA.

223. Mr Ho SC complains that despite the reformulation of the scope of "Client Data", the injunction sought is objectionable. Firstly, he refers to Peninsular Real Estate Ltd v. Harris [1992] 2 NZLR 216, Universal Thermosensors Ltd v. Hibben [1992] 1 WLR 840 and Roberts v. Northwest Fixing [1993] FSR 281 and submits that it is well established that in the absence of a valid restraint of trade clause, an ex-employee may legitimately contact his ex-employer's customers if he happens to recall their contact details. He suggests including a proviso to the effect that the Individual Defendants "shall not be enjoined from calling upon customers remembered by them or customers whose names are acquired from other sources." However, the truth is as Mr Kotewall SC points out, in Roberts v. Northwest Fixing the point about information in the employee's memory was a concession by counsel; while the decision in the latter two cases did not lay down any general rule but were the result of the facts of the cases. The general rule as pointed out in the earlier part of this judgment is that laid down in Johnson & Bloy, Printers & Finishers and SBJ Stephenson. Customer's list and Policy Details are trade secrets or their equivalents. They may not be used by the agent except for the principal's purpose, not even if the information could be carried away in his head. The proviso suggested by Mr Ho SC is based on the terms of the injunction granted in Roberts v. Northwest Fixing, which is the result of concession by counsel. Mr Kotewall SC does not agree to make the concession. In view of the authorities, I think he is right.

224. Secondly, Mr Ho SC raises a number of scenarios as illustrations of how the injunction could operate unfairly to the agent's friends and relatives and a friend who later becomes the agent's spouse. Mr Kotewall SC has replied to these scenarios with admirable clarity showing neither difficulty nor inequity in enforcement. I do not feel it necessary to recite these extreme examples and justify how the injunction works. What is important is that the plaintiff should be in a position to identify with sufficient clarity what information is protected and the defendant be in a position to know what he is being enjoined from doing. It may not be possible to have a formula which will work in all factual situations. I am satisfied that AXA have defined the scope of the injunction they seek with sufficient clarity.

225. Thirdly, Mr Ho SC argues that the injunction has the effect of restraining the Individual Defendants from providing policy details to the Inland Revenue Department. This is a novel argument but is wholly misconceived and unsupported by evidence. Since the Individual Defendants are exclusive agents of AXA, a statement of income from AXA must be sufficient evidence of their total business receipts. As for their expenditure, I fail to see how Client Data would assist the Individual Defendants in proving to the Commissioner of Inland Revenue their meal and travelling expenses, telephone bills, motor car expenses and other outgoings. There is no evidence produced as to what their books and records are, how they would assist the Individual Defendants with their tax returns and what Client Data information has been demanded by the Commissioner.

226. Fourthly, Mr Ho SC complains that the injunction seeks to restrain not only the Individual Defendants but also the agents who work for them. This is the standard formulation of any injunction granted by the court to ensure that the persons enjoined do not escape liabilities simply by asking another to do the prohibited act instead of doing it personally. Insurance agents are agents of the insurance company and not of the managers or senior managers who supervise them. They receive commission from the insurance company and not from the senior managers. Thus an agent's act will not be attributed to an individual defendant unless the individual defendant authorises the agent to perform a prohibited act. Likewise, an act of the agent in the ordinary course of business as an agent would be attributed to the principal, i.e. PCI; and the Individual Defendants do not have anything to worry about.

Conclusion:

227. In conclusion, I am satisfied that the injunction in the terms sought by AXA is appropriate. I grant the injunction accordingly.

DELIVERY UP ORDER:

228. AXA also seek an interlocutory order for the delivery up of documents containing Client Data in the possession, power, custody or control of the defendants. While some of the defendants have made delivery up in respect of SAMS documents pursuant to their undertaking to the Hon Madam Justice Beeson, the delivery up now being sought is not just limited to SAMS documents, but covers any documents containing Client Data, including the Schedule C documents.

The applicable principle: Nottingham Building Society v Eurodynamics Systems Plc:

229. It is well settled that a successful plaintiff is entitled to the delivery up of all materials containing confidential information: Robb v Green. At the interlocutory stage, the principles governing the grant of interlocutory mandatory injunctions as summarized by Chadwick J in Nottingham Building Society v Eurodynamics Systems Plc [1993] FSR 468 at 474 are as follows:

(1) the overriding consideration is which course is likely to involve the least injustice if it turns out to be wrong;

(2) an interlocutory injunction which is mandatory carries a greater risk of injustice than a prohibitory one;

(3) it is legitimate, where a mandatory injunction is sought, to consider whether the court does feel a high degree of assurance that the plaintiff will be able to establish his right at the trial;

(4) even if the court is unable to be so satisfied, there may still be circumstances where the injunction should be granted, where the risk of injustice in refusing to grant the injunction outweighs the risk of injustice in granting it.

Essentially, the court has to balance the risk of injustice and the likelihood of the plaintiff's success at trial.

The case of PCI:

230. PCI's case is that insofar as SAMS documents are concerned, all SAMS documents have already been returned and there is no evidence whatever that PCI has any further documents. Hence Mr Fung SC submits it is futile to order further delivery.

231. While PCI has made some deliveries, the vast majority of the SAMS documents delivered up are copies and not originals. Some of the SAMS documents or copies delivered up are extracts from a larger document. The immediate questions are where are the originals and where are the rest of the documents. There is at least a strong suspicion that not all documents containing Client Data have been delivered up by PCI or some of the Individual Defendants or all of them.

232. There cannot be any genuine dispute as to the confidential nature of the Client Data, the defendants' obligations in relation thereto, and AXA's entitlement to the delivery up of the same, if they are in the possession, power, custody or control of the defendants. There is no risk of injustice that could be caused to the defendants by the delivery up order, as it is limited to the Client Data owned by AXA. If the defendants do not have the data, they will not be prejudiced by the delivery up order at all; if they have, it is only right that they should deliver them now, rather than to hang onto them until conclusion of the trial. There can be no injustice to the defendants if they are ordered to deliver up immediately what they are not entitled to possess, and to which AXA have an undoubted right.

233. On the other hand, the risk of injustice to AXA if no delivery up order is made is obvious. Possession of the Client Data by the defendants would enable them to use the same unlawfully and to the detriment of AXA, whether by way of implementing the PMS or otherwise. As held above AXA would not be adequately compensated by an award of damages at trial. If AXA's policyholders switch to the PMS, it is unlikely that they would return and also AXA would face enormous difficulties in proving and quantifying their damages.

234. I would refrain from giving my opinion whether AXA are likely to succeed at trial. But having regard to the principles in Nottingham Building Society v Eurodynamics Systems Plc, I am satisfied on balance that the risk of injustice to AXA in refusing to grant the delivery up order outweighs the risk of injustice to the defendants in granting it. I dismiss Mr Fung SC's submission that the order would be unworkable as that submission is based on a misinterpretation of "Client Data" as now reformulated by AXA.

235. I do not think PCI's confirmation that they do not have any more SAMS documents would render the delivery up order futile. The delivery up order is an ancillary order to ensure compliance with the interlocutory injunction. For the above reasons I grant AXA the delivery up order against PCI in the terms sought.

The case of the Individual Defendants:

236. All Individual Defendants denied having removed or retained any document containing Client Data and do not have anything to deliver up. There is evidence that an unusually large volume of printouts of Client Data from SAMS had been made under the passwords assigned to them. Except in the case of the 6th defendant, those printouts have neither been delivered up nor accounted for. A small fraction has been delivered up by PCI. I do not think the Individual Defendants' affirmations that they have no documents containing Client Data to deliver up can be taken at face value. The explanations of the 2nd, 3rd and 6th defendants as to why SAMS printouts of their clients are found in the files of PCI are inherently incredible. All these evidence taken together suggest that it is likely that some Client Data are still in the possession of some or all of these Individual Defendants.

237. In the case of the 6th defendant, though the majority of client lists printouts have been returned to AXA, all birthday lists and labels printed have not been returned. They contain Policyholder's Particulars and would enable those policyholders to be contacted for the purpose of marketing the PMS. Her explanation for not being able to return those lists and labels is that AXA had not made timely demand for their return. This is a plausible explanation. But on the other hand, why were they printed in the first place if she had no intention to use them; if she has used them, legitimately or otherwise, why is she unable to account for them; and if she has not used them where are they now? These lists and labels are just as useful and important as the client lists themselves. There is therefore a real likelihood that she is still in possession of those lists and labels which contain Client Data.

238. In addition, as against the 2nd defendant, in his affirmation he said that he checked the agent codes to which the Audit Reports printed out under his password related and found that those agents are not with PCI. Had delivery up been complete, he could not have any more SAMS documents to check with. The inference is that he still has some SAMS documents and has used them for his checking exercise.

239. On the authorities of Lonrho v. Fayed (No 3), The Times, June 24, 1993 and Pacific Link Communications Ltd v. Wong Man Him Melvyn [1996] 1 HKC 474, Mr Ho SC submits that the Individual Defendants' affirmations that they do not have any SAMS documents is final and conclusive and a second delivery up order should not be issued. The authorities referred to by Mr Ho SC are in relation to discovery procedure under Order 24 of the Rules of High Court. That procedure is to ensure all relevant documents in the possession, custody or power of a party are produced and made available for the purpose of determination of the issues at trial. Mr Kotewall SC submits that the delivery up order sought in these proceedings is for a wholly different purpose and if granted, is a recognition of the proprietary right of AXA to the documents which the defendants have undertaken to deliver up. I agree and have this to add. One of the purposes for ordering delivery up is to enjoin a defendant from possibly continuing with a wrongful act and to protect the plaintiff from the likelihood of suffering further loss for which an award of damages is inadequate. The considerations are wholly different from those applicable to discovery under Order 24. In my view, the authorities cited by Mr Ho SC are clearly inappropriate for the instant case.

240. For reasons as explained above, there is a strong prima facie case that delivery up was incomplete and hence the Individual Defendants are in breach of their undertaking to court. Such failure is a contempt of court. Further, the defendants' undertakings are limited to SAMS documents and do not cover fully the subject matter of the delivery up order being sought, i.e. any documents containing Client Data. It would be convenient and indeed appropriate to make a second order to purge the defendants' contempt as well as to encompass all documents containing Client Data. Applying the principles in Nottingham Building Society v Eurodynamics Systems Plc and for similar considerations as in the case against PCI, I think the balance is in favour of granting the delivery up order. Accordingly, I grant the order against each of the Individual Defendants.

DISCLOSURE ORDER:

241. In addition, AXA seek disclosure orders which in summary are as follow. Firstly, it seeks affidavits verifying compliance with the delivery up order. Mr Fung SC submits that all SAMS documents have already been returned and there is no evidence whatever that PCI has any further documents. However, on a proper analysis of the documents and CDs surrendered and in the light of the volume of documents removed by the Individual Defendants (see above), there is a real likelihood that full delivery up has not been made by PCI or some or all of the Individual Defendants. If the defendants have made full delivery up, there is no burden for them to confirm the same by affidavit. If they have not and do not make frank and full disclosure, then they would expose themselves to the possibility of an action in contempt of court. A disclosure order is therefore appropriate to enforce compliance with the delivery up order. In the circumstances, the disclosure must follow as a matter of course.

242. Secondly, AXA seek affidavits disclosing information about documents containing Client Data given to or by any of the Individual Defendants, and the identity of the person giving or receiving the same since 1 November 1999, the date of the first printouts complained of by AXA.

243. In granting the interlocutory injunction against the defendants, I am satisfied that AXA have shown that there is a serious issue to be tried as to whether the defendants were in breach of their duty of fidelity or of confidence to AXA and that an award of damages is inadequate. The evidence shows that large volumes of Client Data had been printed out by the Individual Defendants, but only a small amount has been delivered up or accounted for. It is likely that the defendants have used or will use the Client Data to AXA's detriment. Not only that should be stopped, but those who have been parties or who would be likely parties to the wrongful act must be identified so that AXA could bring them to justice and have their loss compensated for. Under the principle in Norwich Pharmacal Co v Customs and Excise Commissioners [1974] AC 133, the defendants, being prima facie wrongdoers themselves, must assist AXA by disclosing the identity of other wrongdoers or potential wrongdoers as well as full information to AXA. Each such person who supplied such documents to any of the defendants would potentially be a wrongdoer because of the unauthorised disclosure of the confidential information. The same applies to recipients of the confidential information who, by reason of the nature of the data, must have known that the information is confidential and supplied in circumstances involving breach of the duty of fidelity or duty of confidence.

244. At this stage, there is no need for AXA to establish that these other persons had definitely committed wrongful acts against them: P v T Ltd [1997] 1 WLR 1309, nor do they need to make out a strong case: Wellcome Foundation Ltd v Attorney General [1992] 1 HKC 171 at 188. Only the defendants are in the best position to supply the information and the plaintiff should not be put to further expenses in identifying those other wrongdoers.

245. Mr Fung SC argues that the information may have found its way into a document without any misuse of confidential information on the part of PCI or any of its agents. He submits that since no wrong has been committed the Norwich Pharmacal principle does not apply. I cannot agree as the preponderance of the evidence suggests otherwise.

246. Next, Mr Fung SC submits that even if some of PCI's documents contains "stolen" AXA information, it would be highly oppressive and totally disproportionate to require PCI to turn over all its documents whether "clean" or "tainted", on the off chance that they might contain some misappropriated data. He also refers to the practical difficulties for PCI to determine which of its policyholders are or were at some time in the past AXA's policyholders without checking all its documents against SAMS records. For similar reasons as explained above, the difficulties envisaged by Mr Fung SC are being exaggerated. In any event, as Mr Kotewall SC submits, if necessary more time can be allowed for PCI to comply. Further, the fact that some of the documents may contain information "innocently" acquired does not mean that the Norwich Pharmacal principle ceases to apply: see P v T Ltd [1997] 1 WLR 1309.

247. I do not share Mr Ho SC's view that disclosure up to 1 November 1999 too extensive. The disclosure is essentially of the nature of discovery. At this stage there is no burden on AXA to justify a particular cut off date, so long as it is reasonable and not unjust in all the circumstances. A disclosure order is intended to be a powerful weapon to enable a plaintiff to put right the wrong done to him, to prevent further loss and to identify all those who are parties to the wrongful act. For a disclosure order to be effective, it must cover a realistic period of time before the wrong complained of. The evidence shows that at least three of the Individual Defendants were contemplating leaving AXA at that time and Client Data were being printed out from SAMS since that time. It is probable that unlawful use and unauthorised disclosure of Client Data would have started since. In the circumstances, I think AXA are entitled to all information which will assist them to discover if further causes of action exists. Disclosure for a six months period before the Individual Defendants left AXA is not unreasonable or unjust. Accordingly, I grant the disclosure order against all the defendants in the terms sought by AXA.

248. Thirdly AXA seek affidavits giving information as to which AXA's policyholders have been approached and to whom representations as to matching have been made. Mr Fung SC submits that the application for this relief is misconceived as there is nothing wrong with representing to policyholders that PCI's policies "match" or are "comparable" with AXA's policies provided that the relevant differences are explained. There is evidence from some of AXA's agents who attended the training session that they were trained to market PCI's policies as identical or matching policies. There is also evidence from one of AXA's policyholders that he has been approached by a PCI agent and told that the PCI replacement policy would be exactly the same as AXA's.

249. There is at least a serious question to be tried as to whether any "relevant differences" have been explained to AXA's policyholders. AXA's economic interests are being jeopardised. If in fact such material differences have not been explained, it is just and fair that AXA should be informed which of their policyholders have been approached and to whom the representation as to matching have been made so that AXA can take steps to protect their interests and prevent, mitigate or quantify their loss. PCI and the Individual Defendants are in the best position to provide the information. It would be wrong to leave it to AXA to make enquiries from each of their policyholders, which would only have the effect of providing free publicity to PCI's PMS. For the purpose of these interlocutory proceedings, and having balanced the risk of injustice, I consider the evidence sufficient to invoke the grant of disclosure orders against all the defendants.

250. I dismiss counsel's submission that the disclosure order is onerous and impossible to comply in that there is no way for PCI to find out if a policyholder is on AXA's Client Data or not, that PCI itself does not know which of its agents have approached any current or former AXA's policyholders and that the agent may not remember the name of every person they have approached since 1 November 1999. These difficulties are unreal and being exaggerated. PCI is a company of substantial means and with a sophisticated computerised system. It should be able to locate all relevant documents without undue difficulties. This is particularly so as the PMS is targeted at AXA's policyholders. It would not be difficult to identify these policyholders without reference to any SAMS records. It would have no difficulties in identifying the agent responsible for processing the PMS application. Certainly, the Individual Defendants themselves should have no difficulties in recalling details of successful transactions and would have personal records of AXA's policyholders approached. PCI is surely in the position to compel its agents to furnish the information. The difficulties envisaged by Mr Fung SC are being exaggerated.

251. None of the difficulties raised by Mr Fung SC in respect of PCI is applicable to any of the Individual Defendants. According to the Individual Defendants, they maintained books and records which contain Client Data for the purpose of filing tax returns. If so, they would have no difficulties supplying the information required.

252. Mr Ho SC submits that the Individual Defendants are unable to provide the name and address of every AXA's policyholder who has been approached by PCI's agents and to whom it has been represented that the PCI insurance products are matching. There is no reason why the Individual Defendants could not rely on PCI to produce the information. In addition, they can speak for themselves and such other agents in PCI working for them if they have the knowledge. If they honestly do not have the information to disclose, they could so confirm in their affidavit, subject of course to the risk of being cited for contempt if it is shown that they are not telling the truth.

253. Lastly, Mr Fung SC submits that the Insurance Companies Ordinance and the Code of Practice provide a regulatory and supervisory framework in which reprehensible tactics such as twisting or fraudulent misrepresentations are controlled and that it is surprising that neither AXA nor their policyholders have availed themselves of the alternative avenue of redress, namely through the machinery set up under the Code of Practice. I agree with Mr Kotewall SC that there is no rule of law that a party against whom a civil wrong has been committed cannot seek redress in a Court. As Mr Kotewall SC submits, even if the authorities are going to investigate into the alleged misconduct, the likely consequence is the de-registration of the agents concerned. Such a complaint to the authorities is unlikely to put an immediate stop to the acts complained of and, at least, the remedy of delivery up and disclosure are not available to AXA. I do not think the availability of alternative remedy is a bar to AXA's access to the court.

254. The most important purpose of the injunctive relief is to prevent further loss for which an award of damages is inadequate. The interlocutory injunction order granted would be inadequate if not reinforced by the ancillary orders now sought by AXA, including the disclosure order. Having regard to all the circumstances and for similar reasons as given in relation to the other orders, I think it appropriate to grant AXA the disclosure order now sought.

CONCLUSION:

255. Accordingly, I grant the plaintiffs all the orders sought in their summons dated 21 September 2000 subject to the amendment in respect of the meaning of "Client Data". I also make a cost order nisi that the defendants shall pay the plaintiffs' cost with certificate for three counsel, to be taxed, if not agreed.

256. I am grateful to leading counsel and their juniors for their full and fair exposition of the law. The case has been extremely well and ingeniously argued on all sides, leaving no stone unturned and no issue unexplored. The very considerable depth which counsel have taken in the legal arguments is both remarkable and admirable for an interlocutory application of this kind.

Representation:

Mr Robert Kotewall, SC leading Mr Stewart Wong and Mr Sanjay Sakhrani, instructed by Messrs Herbert Smith, for the Plaintiffs

Mr Daniel Fung, SC leading Mr David Stokes and Mr Ling Chun Wai, instructed by Messrs Richards Butler, for the 1st Defendant

Mr Ambrose Ho, SC leading Mr Rimsky Yuen and Mr C M Law, instructed by Messrs Hoosenally & Neo, for the 2nd to 10th Defendants