Breezeway Overseas Ltd and another v UBS AG and others
[2012] SGHC 170

Case Number	:	Suit No 114 of 2010 (Registrar's Appeal No 412 of 2011)
Decision Date	:	16 August 2012
Tribunal/Court	:	High Court
Coram	:	Lee Seiu Kin J
Counsel Name(s)	:	Freddy Lim (Lee & Lee) for the plaintiffs; Tan Shou Min (Drew & Napier LLC) for the first, third, fourth and fifth defendants; Charmaine Chan (Legis Point LLC) for the second defendant.
Parties	:	Breezeway Overseas Ltd and another — UBS AG and others

16 August 2012

Lee Seiu Kin J:

1       This was an appeal against part of the decision of the learned senior assistant registrar (“SAR”) regarding certain disputed keyword search terms in the electronic discovery (“e-discovery”) process. The grounds of the SAR’s decision may be found in Breezeway Overseas Ltd v UBS AG [2012] SGHC 41 (“Breezeway”). After hearing both parties, I allowed the appeal in part. I now set out the reasons for my decision.

Background facts

The parties

2       The present appeal arose from a summons (“SUM 2443/2011”) taken out in Suits No 112 of 2010 and 114 of 2010 (“the Suits”) for an order that the parties conduct e-discovery in accordance with the plaintiffs’ draft e-discovery protocol dated 19 May 2011.

3       The first plaintiff, Breezeway Overseas Ltd, is a company registered in the British Virgin Islands, while the second plaintiff, Mr Vasanmal Murli, is a director of the first plaintiff and the person who exerts effective and complete control over the first plaintiff. The first defendant is UBS AG, a global financial services firm that is headquartered in both Basel and Zurich, Switzerland, while the fourth and fifth defendants are the Hong Kong and Singapore branches of the first defendant. For convenience, the first, fourth and fifth defendants are henceforth collectively referred to as “the Bank”. The second defendant, Susan Abraham, and third defendant, Vikrant Kanyal, were former employees of the Bank and had confirmed that they did not have in their possession, custody or power any discoverable documents. The present e-discovery application is therefore between the plaintiffs and the Bank.

4       The first plaintiff is a long-time customer of the Bank. The plaintiffs alleged that sometime in or about 2008, they relied on the Bank’s representations and took loans from the fourth defendant to purchase the following bonds (the “leveraged bonds”):

(a)     6.625% ICICI Bank Bonds on 13 February 2008;

(b)     7.335% Bank of Moscow Bonds on 14 February 2008;

(c)     7.335% Bank of Moscow Bonds on 16 April 2008;

(d)     6.609% VTB Cap Bonds on 15 April 2008;

(e)     8.25% VTB Cap Bonds in July 2008; and

(f)     5.75% Kaupthing Bank Bonds on 6 March 2008.

5       The leveraged bonds were subsequently sold on or about 13 August 2008, and the loan that had been taken to purchase these bonds was applied towards the purchase of 13.625% Venezuela Bonds on or about 13 August 2008.

6       In a meeting between the second plaintiff and the Bank held at the Bank’s office at Suntec City, the Bank’s employees had allegedly represented that the loans were fixed loans and/or fixed to maturity (which meant that the loans could not be recalled prior to maturity). The second plaintiff claimed that he was not told that the first plaintiff had to provide “collateral” to secure the loans, nor was he informed of which assets were allegedly collateralised.

7       Sometime in or about March 2009, the Bank issued margin calls on the plaintiffs’ account, which required the plaintiffs to raise huge sums of money within a short period of time, failing which the Bank threatened to liquidate the plaintiffs’ assets against the plaintiffs’ will. During this period of time, the plaintiffs protested strongly against the Bank’s decisions to reduce the quantum of the loans and to make the margin calls.

8       It was the plaintiffs’ claims against the Bank for, inter alia, misrepresentation, breach of fiduciary duties, gross negligence, wilful misconduct and breach of contract which gave rise to the Suits and, accordingly, the present appeal.

The decision below

9       SUM 2443/2011 was heard substantively over two separate hearings on 28 July 2011 and 12 December 2011.

10     At the first hearing on 28 July 2011, the SAR made orders concerning the categories of documents to be disclosed (“the Documents”), as follows:

(a)     The emails in the Mailboxes (“Mailbox Emails”), emails in the Personal Network Profiles (“P-Drive Emails”), the documents in the Personal Network Profiles (“P-Drive Documents”) and Instant Messaging Records (“IMs”) of Anandraj Jain, Kurt Kumschick, Susan Abraham and Vikrant Kanyal (referred to as the “Four Employees”) for the period of February 2008 to August 2008.

(b)     The Mailbox Emails, P-Drive Emails, P-Drive Documents and IMs of the Four Employees plus Vikram Malhotra and Andreas Reber (referred to as the “Four Employees Plus Two”) for the period of February 2009 to June 2009.

11     The SAR also ordered that:

(a)     The Bank was to conduct reasonable keyword searches (ie, electronic searches of electronic documents with specified words or strings of words) on the Documents.

(b)     The plaintiffs and the Bank were to try to come to an agreement on a set of keywords to be used to conduct the searches.

(c)     If an agreement could not be reached, the Bank was to perform preliminary keyword searches (viz, searches intended solely for the purpose of identifying the number of hits of each proposed keyword: see Robin Duane Littau v Astrata (Asia Pacific) Pte Ltd [2011] SGHC 61 (“Robin Duane Littau”) at [19]) using the disputed keywords, before returning to court for a determination on the disputed keywords.

12     After the 28 July 2011 hearing, the plaintiffs proposed 30 keywords for the conduct of the keyword searches. Of the 30 keywords, the Bank agreed to seven and objected to 23. The seven keywords that the Bank agreed to were keywords which fell within the categories of unique reference numbers (ie bank account numbers) and keywords which identified key witnesses, as follows:

(a)     “Murli”;

(b)     “Vee”;

(c)     “Vasanmal”;

(d)     “Breezeway”;

(e)     “mvee@saniva.com.sg”;

(f)     “207760”; and

(g)     “910879”.

13     The Bank rejected the remaining 23 keywords because, in the Bank’s view, those keywords were too generic and broad considering that the Bank was in the banking business, and that the employees whose Documents would be searched were banking professionals in the wealth management business. The Bank therefore conducted the preliminary searches and provided the search results to the plaintiffs.

14     Parties returned before the SAR on 12 December 2011 for a determination on the 23 disputed keywords. At the hearing, the SAR allowed ten keywords and disallowed the remaining 13, as follows:

(a)     The word “collateral” would be used in a keyword search of the Mailbox Emails and the P-Drive Emails of the second and third defendants for the periods stated in [10] above. The SAR permitted the keyword “collateral” because one of the issues in dispute was whether the loans which the first plaintiff had taken were secured against collateral, and if so, whether there was any identification of which banking assets had been marked as collateral against the loans. There were also disputes of fact in relation to requests by the Bank for additional collateral or margin call. In addition, the SAR took into consideration the number of hits in the preliminary search.

(b)     The following words or phrases would be used in a keyword search of the Mailbox Emails, the P-Drive Emails, the P-Drive Documents and the IMs of the Four Employees Plus Two for the periods stated in [10] above:

(i)       “Fixed Loan”;

(ii)       Proximity search of “fixed” within ten words of “maturity”;

(iii)       “Protest”;

(iv)       “Bank of Moscow”;

(v)       “Kaupthing”;

(vi)       “Kuznetski”; and

(vii)       “Republic of Venezuela”.

(a)     The SAR permitted the keywords “Fixed Loan” and the proximity search because one of the issues in dispute was whether the terms and interest rates of the loans taken by the first plaintiff were “fixed to maturity”. The keyword “Protest” was allowed because although it was a common word, its use in the context of the correspondence between the first plaintiff and the Bank was particularly relevant to a key issue, viz, the Bank’s alleged unilateral decision to reduce the “loanable value” of the leveraged bonds which was in turn used by the Bank as a basis for the margin call. The first plaintiff had protested to the Bank against this reduction and margin call. As for the keywords “Bank of Moscow”, “Kaupthing”, “Kuznetski” and “Republic of Venezuela”, these keywords were permitted because the leveraged bonds were fairly unique to the banker-customer relationship between the Bank and the first plaintiff.

(c)     The terms “ICICI” and “VTB” would be used in a keyword search of the Mailbox Emails and the P-Drive Emails of the second defendant, and the Mailbox Emails of the third defendant, for the periods stated in [10] above. The SAR permitted these keywords because they referred to the ICICI and VTB leveraged bonds which were heavily traded by the India desk and which had been recommended to the first plaintiff. However, in view of the number of preliminary hits, these were restricted to repositories just mentioned.

The appeal

15     The present appeal concerned nine of the ten keywords that the SAR allowed. Only the proximity search was not part of the appeal because the SAR had indicated at the 12 December 2011 hearing that this keyword would only be allowed if the Bank’s search engine could perform the necessary proximity searches. As the Bank had since deposed on affidavit that its search engine was not capable of performing proximity searches, this keyword was not part of the appeal.

16     The parties suggested that the key issue in the present case was whether the keywords ordered by the SAR should be used to perform keyword searches on the Documents. The thrust of the Bank’s arguments was that the disputed keywords were of low relevance and that the searches were not necessary for the fair disposal of the Action or for saving costs. It should be noted that the sub-text to this appeal was the Bank’s concern that – on its interpretation of Robin Duane Littau – it would not be permitted to review the search results for relevance, and therefore that the search results may be over-inclusive by including material irrelevant to the Suits (see [27] et seq).

17     Given that the plaintiffs agreed that the alleged “protest” only arose on or after February 2009, I allowed the part of the appeal relating to the time period to be used for a keyword search on the term “protest”. I therefore ordered that the keyword search for the term “protest” should be restricted to the time period of February 2009 to June 2009. As for the remaining search terms, I agreed with the SAR’s order. However, it should be noted that the Bank’s concerns on the over-inclusiveness of the search results were substantially assuaged because I disagreed with the Bank’s interpretation of Robin Duane Littau (see [28]-[33] below).

18     The present grounds of decision focuses on addressing issues related to keyword searches in the context of e-discovery, as well as the entitlement of a party giving discovery to review the search results for relevance, privilege and confidentiality.

Discovery and e-discovery generally

19     In the discovery framework under O 24 of the rules of Court (Cap 322, R 5, 2006 Rev Ed), a party to a suit has an obligation to disclose to the opposing party all documents which are, or have been, in his possession, custody or power that are relevant to the issues in dispute, subject to the requirement in O 24 r 7 of the rules of Court that discovery must be “necessary either for disposing fairly of the cause or matter or for saving costs” (see Sanae Achar v Sci-Gen Ltd [2011] 3 SLR 967 (“Sanae Achar”) at [7] and [9]).

20     The perennial tension in the law of civil procedure, viz the attempt to achieve both justice and efficiency, comes to the forefront in the discovery process. On the one hand, it is ex hypothesi in the interest of justice that all relevant material is discovered, while on the other, there is a pressing need to ensure efficiency lest injustice be occasioned through the well-meaning but disproportionate attempt to ensure that all relevant material is disclosed. As Jacob LJ succinctly observed in Nichia Corp v Argos Ltd [2007] EWCA Civ 741 at [50]-[51]:

50. … “Perfect justice” in one sense involves a tribunal examining every conceivable aspect of a dispute. All relevant witness and all relevant documents need to be considered. And each party must be given a full opportunity of considering everything and challenging anything it wishes. No stone, however small, should remain unturned. …

51. But a system which sought such “perfect justice” in every case would Actually defeat justice. The cost and time involved would make it impossible to decide all but the most vastly funded cases. The cost of nearly every case would be greater than what it is about. Life is too short to investigate everything in that way. So a compromise is made: one makes do with a lesser procedure even though it may result in the justice being rougher. Putting it another way, better justice is achieved by risking a little bit of injustice.

21     The proliferation of information technology has resulted in a burgeoning volume of discoverable electronic documents, causing a plethora of problems for the traditional discovery framework (see, eg, Sanae Achar at [14] and Surface Stone Pte Ltd v Tay Seng Leon and another [2011] SGHC 223 at [1]). Information technology is now deeply entrenched in every aspect of business and personal life. There are more and more documents generated, duplicated and stored in various physical and virtual locations. Although humans are the primary producers of documents, machines are also generating original documents out of data from humans or other machines. The traditional discovery method of manually reviewing every document (“ocular review”) pits the ability of humans to review documents against the ability of machines to produce them. It is a war lost unless we can harness the ability of the same machines to review them.

22     In this connexion, the Supreme Court Practice Direction No 3 of 2009 (“the e-discovery PD”), recently amended on 1 March 2012, attempts to supplement the traditional tests of relevancy and necessity (see Sanae Achar at [10]). It works within the framework of traditional discovery, seeking a pragmatic compromise between the obligation to give discovery of all relevant documents and the potentially prohibitive costs of identifying those documents by way of ocular review. The e-discovery PD provides some guidance on how technology can be utilised to identify, as far as possible, relevant documents, whilst keeping in view the requirements of necessity, proportionality and economy (see para 43E of the e-discovery PD and O 24 r 7 of the rules of Court). It should be emphasised that the e-discovery PD does not set out to change the law on discovery (see Deutsche Bank AG v Chang Tse Wen and others [2010] SGHC 125 at [14]); a fortiori, it does not aim to resolve the problems associated with traditional principles of discovery.

Keyword searches for identification of relevant material

23     Specifically on the issue of keyword searches, it should be noted that electronic searches of electronic documents using keywords are permitted under paras 43D(1) and 43D(4) of the e-discovery PD. Those paragraphs provide as follows:

(1)    A class of electronically stored documents may be described by specifying or describing a search term or phrase to be used in a search for electronically stored documents which shall be reasonable in scope (“reasonable search”). A request for the giving of discovery by describing a class of electronically stored documents with reference to search terms or phrases must specify or describe limits on the scope of the search; such limits shall include at least the following:

(a)    specifying or describing custodians and repositories, eg physical or logical storage locations, media or devices; and

(b)    specifying the period during which the requested electronically stored documents were created, received or modified.

…

(4)    An application for discovery of any electronically stored document or class of electronically stored documents which specifies or describes a search term or phrase to be used in a reasonable search for electronically stored documents must specify or describe limits on the scope of the search to be conducted.

24     Keyword searches are potentially both over- and under-inclusive. As any user of search engines would realise, false positives and false negatives are an inevitable result of attempting to identify relevant material through keyword searches. This should be contrasted with ocular review, which could theoretically ensure zero gaps in the identification of relevant material. However, when a large number of documents have to be reviewed, discrepancies inevitably arise due to fatigue and variances in each reviewer’s subjective appreciation of the issues in dispute and threshold for relevance. When the documents to be reviewed exceed a certain volume, accurate ocular review becomes prohibitively costly and impracticable.

25     Imperfect as they are, therefore, keyword searches present a practical trade-off between achieving a theoretically complete set of relevant material and keeping costs proportionate to the value of the claim. As part of the e-discovery process of identifying relevant material, keyword searches provide a pragmatic solution where the costs of ocular review would be way out of proportion to the stakes in the case.

26     In my view, it would be helpful to conceptualise the process of identifying relevant material through keyword searches as an iterative sieving process. This coheres with para 43B of the e-discovery PD, which contemplates the conduct of general discovery in stages (see also Sanae Achar at [14], Breezeway at [4]-[15] and Gavin Goodale and ors v The Ministry of Justice and others [2010] EWHC B40 (QB) at [22]-[23]). Under this iterative sieving process, the court and the parties endeavour to select the best possible keywords that would avoid sieving out relevant material whilst simultaneously ensuring a practical and workable manner of processing the material at hand. Parties would thereafter clarify and/or narrow search terms as necessary with a collaborative spirit and in good faith, resorting to applications to court only when parties require an arbiter to break the impasse. The court will eventually sanction a final set of search criteria for the purposes of e-discovery (“court-sanctioned search”).

Post court-sanctioned search reviews

27     Before me, counsel for the Bank expressed the concern that the court-sanctioned search may throw up irrelevant, privileged and/or confidential material. In this regard, he queried if the Bank would be permitted to conduct a “post court-sanctioned search review”, viz, to review the search results of the court-sanctioned search and withhold irrelevant, privileged and/or confidential material.

Post court-sanctioned search review for relevance

28     The parties suggested that Robin Duane Littau, by deeming (at [32]) the search results of the court-sanctioned search “prima facie relevant”, precluded post court-sanctioned search reviews for relevance. The parties also drew my attention to the recently amended e-discovery PD which, in their view, precluded post court-sanctioned search reviews for relevance. The relevant parts of the e-discovery PD provide as follows:

(a)     Para 43D(3):

The obligations of a party responding to a request for reasonable search for electronically stored documents is fulfilled upon that party carrying out the search to the extent stated in the request and disclosing any electronically stored documents located as a result of that search. The party giving discovery shall not be required to review the search results for relevance.

(b)     Appendix E, part 2, para 1(g):

No review for relevance. Subject to paragraph 3 (Review for privileged material) below, each party’s obligation to conduct a reasonable search is fulfilled upon that party carrying out the search to the extent agreed in this plan; the party giving discovery shall not be required to review the search results for relevance.

(c)     Appendix E, part 1, para 2(c):

Review of search results

Where search terms are used, the search results are deemed to be relevant and discoverable subject to review for the purpose of identifying privileged documents. Parties should not expend additional time, effort and resources to review search results for relevance.

[emphasis in original]

29     I disagreed with the view that Robin Duane Littau and the e-discovery PD precluded post court-sanctioned search reviews for relevance, for the following reasons:

(a)     First, while the court in Robin Duane Littau did express the view (at [32]) that “the search results are prima facie relevant”, this must be read in light of the court’s carefully qualified view (at [33]) that “further review for relevance will usually be unnecessary in view of the costs that are likely to be incurred” [emphasis added]. In other words, the court did not preclude post court-sanctioned search reviews for relevance.

(b)     Second, para 43D(3) and Appendix E, Part 2, para 1(g) of the e-Discovery PD suggest that the party giving discovery “shall not be required to review the search results for relevance” [emphasis added]. These paragraphs did not preclude post court-sanctioned search reviews for relevance; they merely note that the party giving discovery shall not be required to do so.

(c)     Third, Appendix E, part 1, para 2(c) of the e-Discovery PD merely suggests that parties “should not” review search results for relevance, possibly in recognition that post court-sanctioned search reviews for relevance would usually incur disproportionate costs. In any case, Appendix E, Part 1, para 2(c) is simply part of a “Check List of Issues for Good Faith Collaboration” which is intended to function as a list of items to be traversed (initially) by counsel and client and (subsequently) when parties meet and confer; it is an exhortation and is not strictly part of the e-discovery PD proper.

30     The upshot of the above observations is that the concept of “prima facie relevance” refers to the notion that the party giving discovery is not required to review the search results of the court-sanctioned search for relevance. The search results of the court-sanctioned search are “prima facie relevant” in the sense that the party giving discovery will be deemed to have complied with his obligation to provide all relevant documents under the general discovery process (see O 24 r 1 of the rules of Court).

31     In this regard, one should be keenly aware of the conceptual distinction between the obligation to give discovery and the concept of relevance in the context of discovery. Although both concepts are related because the party giving discovery has the obligation to give discovery of all relevant material, the e-discovery PD makes it possible for that party to fulfil his discovery obligations by giving discovery of the results of a court-sanctioned search, regardless of whether such search results are over- or under-inclusive vis-à-vis the identification of relevant material.

32     With regard to the obligation to give discovery, as long as the party giving discovery complies with the terms of the court-sanctioned search, as well as with all the necessary requirements as stated in the rules of Court, the party entitled to discovery “would have to accept that [the party giving discovery] had fulfilled [his or] her discovery obligations, notwithstanding the fact that there could well be [documents] not caught by the search engine employed” (Sanae Achar at [23]; and see [30] above). This is because the discovery rules do not require that no stone should be left unturned – an attitude “justified by considerations of proportionality” (Sanae Achar at [23], citing Digicel (St Lucia) Ltd v Cable & Wireless Plc [2008] EWHC 2522 (Ch) at [46]).

33     However, the fact that the obligation to give discovery is fulfilled by the party giving discovery of the results of a court-sanctioned search does not mean that the results of the search are deemed relevant in the sense that the party giving discovery is not entitled to conduct post court-sanctioned search reviews. At bottom, the e-discovery PD is designed to keep costs proportionate, relieving the party giving discovery of the need to conduct costly and time-consuming ocular review of all the documents in his possession, custody or power (see [22] above). The e-discovery PD was not intended to prevent the party giving discovery from undertaking a post court-sanctioned search review to remove documents that are irrelevant to the issues in dispute. But any such further review would be outside the ambit of the e-discovery PD and the decision to remove any document on the ground of irrelevance must be done by way of ocular review. This means that every document a party removes in a post court-sanctioned search review on the basis that it is irrelevant must be processed in the traditional manner, ie, manually examined and subsequently considered irrelevant by a solicitor familiar with the issues in dispute (or the party, in the case of a litigant in person). As this process is usually an expense unreasonably incurred the party electing to do this will not generally be entitled to recover the costs of the post court-sanctioned search review in the event that costs are eventually awarded in his favour.

Post court-sanctioned search reviews for privilege and/or confidentiality

34     Other than a post court-sanctioned search review for relevance, it is clear that the party giving discovery may conduct a post court-sanctioned search review for privileged and/or confidential material (see Robin Duane Littau at [35]). Again, the same conditions apply as in a post court-sanctioned search review for relevance, ie, any documents may only be withheld from discovery on the ground of privilege and/or confidentiality pursuant to ocular review, and costs for this extra step will not generally be allowed.

Application to the facts

35     In the present case, the Bank did preliminary searches on the keywords ordered by the learned SAR. The parties made a summary of the preliminary searches available to me in their submissions. I note that the number of hits for each of the ten search terms was generally not excessive. Furthermore, upon substantive perusal of each keyword ordered by the SAR, I found that – for the reasons stated by the SAR (see [14] above) – it was reasonable to believe that the keywords ordered would identify material that was potentially relevant to the issues in the Suits.

36     For the avoidance of doubt, I ordered that the Bank was entitled to conduct post court-sanctioned search reviews of the documents retrieved from all searches for relevance (including documents that would lead to a “train of inquiry”: see O 24 r 5(3)(c) of the rules of Court, and see The Compagnie Financiere et Commerciale du Pacifique v The Pruvian Guano Co (1882) 11 QBD 55) as well as privilege and/or confidentiality.

Conclusion

37     The appeal was therefore allowed to the extent that the search for the term “protest” would be restricted to the period of February 2009 to June 2009. Given that the plaintiffs opposed the Bank’s application for a stay of proceedings pending the present appeal, although the Bank succeeded in part at this appeal, I made no order as to costs in both the stay opposition and the present appeal.