Case Details
- Citation: [2012] SGHC 41
- Title: Breezeway Overseas Ltd v UBS AG
- Court: High Court of the Republic of Singapore
- Date of Decision: 28 February 2012
- Case Number: Suit No 114 of 2010 (consolidating Suit No 112 of 2010), Summons No 2443 of 2010
- Tribunal/Coram: High Court; Yeong Zee Kin SAR
- Judge: Yeong Zee Kin SAR
- Plaintiff/Applicant: Breezeway Overseas Ltd
- Defendant/Respondent: UBS AG
- Other Parties (as reflected in the extract): Second and third defendants were UBS employees who later left UBS; the fourth and fifth defendants were UBS branches in Hong Kong and Singapore respectively
- Counsel for Plaintiffs/Applicants: Mr Julian Tay with Mr Freddy Lim
- Counsel for Defendants/Respondent: Mr Tan Shou Min (for the first, third, fourth and fifth defendants); Ms Charmaine Chan (for the second defendant)
- Legal Areas: Civil Procedure — Electronic Discovery
- Key Procedural Themes: Discovery in stages; selection of search terms; discovery by direct exchange of soft copies on finalised optical discs
- Judgment Length: 9 pages, 4,892 words
- Procedural Posture: Plaintiffs sought discovery under an electronic discovery protocol; UBS appealed against the court’s decision on disputed search terms
Summary
Breezeway Overseas Ltd v UBS AG [2012] SGHC 41 is a High Court decision addressing how electronic discovery should be structured in modern litigation, particularly where documents are dispersed across multiple employees’ email accounts, computers, and network storage. The case arose from a dispute involving a family investment vehicle, Breezeway, which maintained accounts with UBS. The plaintiffs sought discovery using an electronic discovery protocol, and the court ordered that discovery be conducted in stages and that the parties use agreed search terms to identify relevant documents.
The central issue in the extract is the court’s approach to “discovery in stages” and, more specifically, how search terms should be selected and delimited so that discovery remains efficient and proportionate. The judge emphasised that traditional discovery principles—developed in an era of centralised paper files—do not translate neatly to electronic records characterised by duplication, decentralisation, and the practical impossibility of “pulling out the relevant file” from a single repository. The court’s reasoning provides a framework for managing electronic discovery through custodians, repositories, and keyword searches with reasonable limits.
What Were the Facts of This Case?
Breezeway Overseas Ltd (“Breezeway”) was a customer of UBS AG (“UBS”). Breezeway was structured as a family investment vehicle controlled by Vasanmal Murli, who was one of its directors. The other directors were his wife and two daughters. Breezeway held accounts with UBS through branches in Hong Kong and Singapore, which were represented in the proceedings as the fourth and fifth defendants respectively. The operational servicing of Breezeway’s accounts was initially handled by UBS employees who acted as client advisers and were employed by the Singapore branch.
At the material time, Breezeway’s accounts were serviced first by the second defendant and then by the third defendant. Both were described as client advisers and employees of UBS’s Singapore branch. Importantly for discovery, by the time the summons for discovery came up for hearing, the second and third defendants had left UBS. They deposed that they no longer had in their possession, custody, or power any discoverable documents because all discoverable documents had been turned over to UBS upon the termination of their employment. UBS, however, was not familiar with how those documents had been organised internally.
Because UBS could not readily locate relevant documents using a centralised filing system, the plaintiffs proposed an electronic discovery protocol. The protocol contemplated discovery in stages and the use of search terms to identify discoverable documents efficiently. The court initially ordered parties to adopt such a protocol, directing them to discuss and agree on the search terms. Liberty to apply was granted for determination of disputed search terms, and the parties subsequently returned to court on the contested terms.
In the proceedings reflected in the extract, UBS appealed against the judge’s decision concerning the search terms that had been ordered. The judge therefore set out detailed reasons for the decision, focusing on the rationale for staged discovery and the legal and practical principles governing the selection and delimitation of search terms in electronic discovery.
What Were the Key Legal Issues?
The first legal issue was how discovery should be conducted in an electronic environment where records are decentralised and duplicated across multiple custodians. The court had to determine whether, and how, traditional discovery principles should be adapted to address the realities of email-based recordkeeping and the absence of a complete central electronic document management system. This included assessing the appropriateness of “discovery in stages” as a mechanism to manage cost, efficiency, and proportionality.
The second issue concerned the selection of search terms. The court had to decide the proper approach to keyword searching in electronic discovery: what search terms should be used, and how the search should be delimited to ensure that the results are reasonably targeted rather than overbroad. This required the court to consider how to avoid excessive retrieval of irrelevant material and duplication, while still ensuring that discoverable documents are captured.
Related to these issues was the question of how the court should structure the discovery protocol in a way that is proportionate to the amounts at stake and the significance of the issues in dispute. The judge’s reasoning indicates that proportionality is not merely a general principle but an operational constraint that shapes the scope of custodians, repositories, and the limits of searches.
How Did the Court Analyse the Issues?
The judge began by explaining the “raison d’être” for discovery in stages. In earlier times, document management was more disciplined: paper records were often kept in a single file maintained by the department handling the transactions, with duplicates maintained by other departments such as finance or accounts. Employees typically did not maintain personal files expected to be preserved for litigation, and working papers were often discarded unless incorporated into the departmental file. This made it feasible to “pull out the relevant file” and identify documents relevant to the dispute.
However, the judge observed that modern recordkeeping—especially email—creates two major problems for discovery. First, there is proliferation of copies because the same email can be sent to multiple recipients and stored in each recipient’s account and attachments. Second, records are decentralised because each employee has a personal email account and may store documents on their computer hard disk or network storage. The practical consequence is that the “relevant file” is no longer centralised; it is distributed across multiple repositories belonging to multiple custodians. This is compounded where employees who were most involved in the relevant transactions have left the organisation, making it harder to locate documents and understand their organisation.
In this context, the judge explained that traditional discovery—if applied without modification—would impose significant costs. Even if de-duplication software can identify duplicates among electronic documents, the effort of identifying and removing duplicates across multiple custodians and repositories should not be underestimated. The court therefore treated staged discovery as a more efficient and cost-effective method of managing electronic discovery.
Turning to how discovery may be conducted in stages, the judge articulated a structured approach. Parties must identify, at the close of pleadings, both the issues in dispute and the key witnesses. In electronic discovery, these key witnesses are treated as “custodians” because they are custodians of both knowledge and documents. Once key custodians are identified, the focus shifts to the repositories where electronic documents are stored—such as email accounts, desktop and notebook hard disks, removable media, and network storage locations. The court stressed that not every repository of each custodian must be searched initially; the scope should depend on proportionality and the significance of the issues.
The judge then described two pathways for the initial stage. Where custodians have organised their electronic records into structured folders, parties may effectively “pull out the relevant file” by producing copies of relevant folders and their contents, after reviewing for irrelevant material. Where records are not neatly organised, parties may conduct a “reasonable search” using agreed search terms with reasonable limits. The judge linked this concept to the practice directions and earlier authority, explaining that a reasonable search is an electronic search on identified repositories using agreed search terms, but delimited in two key respects.
First, parties must decide whether to search entire storage devices or only relevant folders and sub-folders. For hard disks, the judge indicated that parties should usually identify relevant folders and sub-folders, because operating systems and application files occupy many folders that are unlikely to contain discoverable documents. This reduces noise and prevents unnecessary retrieval. Second, parties must agree on the time period during which the relevant documents were created or received. This time period can vary depending on the keyword, repository, or custodian, and serves to exclude responsive documents that are likely irrelevant.
Crucially, the judge also addressed the legal effect of search results. Search results are presumed to be relevant without requiring a further review of each document for relevance. This presumption is supported by earlier decisions, including Sanae Achar v Sci-Gen Ltd [2011] 3 SLR 967 and [2011] SGHC 87, at [23]. The court’s reasoning suggests that the purpose of reasonable search parameters is not only to manage cost but also to justify the presumption of relevance: if the search is properly delimited, it is treated as a reliable proxy for relevance.
Finally, the judge emphasised flexibility. Discovery in stages does not foreclose subsequent stages involving additional custodians and repositories. If needed, parties may proceed to further stages after the initial stage, using the same methodology of identifying custodians, repositories, and search parameters. The judge noted that, based on the court’s limited experience, subsequent stages had not been necessary in any case thus far, but that this did not mean they would never be necessary. Parties may also seek specific discovery under O 24, r 5 based on documents disclosed in the initial stage.
What Was the Outcome?
On the procedural posture described in the extract, UBS appealed against the court’s decision regarding the disputed search terms. The judge’s reasons set out the principles that supported the ordered approach to search terms within a staged electronic discovery protocol. The outcome, as reflected in the extract, is that the court maintained its framework for discovery in stages and upheld the methodology for selecting and delimiting search terms, subject to the protocol’s staged and proportional design.
Practically, the decision confirms that parties in Singapore litigation should expect the court to require structured electronic discovery: identification of key custodians and repositories, agreement on search terms, and delimitation by folder scope and time periods. It also signals that search results obtained through properly delimited searches will be treated as presumptively relevant, reducing the need for document-by-document relevance review.
Why Does This Case Matter?
Breezeway Overseas Ltd v UBS AG [2012] SGHC 41 is significant because it provides a clear judicial framework for electronic discovery in Singapore, particularly for cases where records are dispersed across employees’ email accounts and computers and where a centralised document management system is absent. The decision is valuable to practitioners because it translates abstract discovery obligations into operational steps: custodians, repositories, search terms, and limits. This makes it easier for parties to design discovery protocols that are defensible, efficient, and proportionate.
From a precedent perspective, the judgment reinforces several principles that later electronic discovery practice in Singapore has relied upon: staged discovery is appropriate to address duplication and decentralisation; keyword searches must be “reasonable” and delimited; and search results can be presumed relevant when the search is properly constructed. The decision also aligns Singapore practice with comparative reasoning, citing the English case Goodale & Ors v The Ministry of Justice & Ors [2009] EWHC B41 (QB) for the idea of starting with the most important people and adopting an incremental approach.
For litigators, the case has practical implications for how to negotiate discovery protocols and how to respond to disputes about search terms. Parties should anticipate that the court will scrutinise whether proposed searches are overbroad, whether they target relevant repositories rather than entire storage devices, and whether time periods are appropriately tailored. The judgment also underscores that the absence of centralised records does not excuse parties from discovery obligations; instead, it requires a more disciplined electronic discovery methodology to achieve the same ends at reasonable cost.
Legislation Referenced
- Order 24, Rule 5 of the Rules of Court (specific discovery)
Cases Cited
- [2011] SGHC 61 (Robin Duane Littau v Astrata (Asia Pacific) Pte Ltd)
- [2011] SGHC 87 (Sanae Achar v Sci-Gen Ltd)
- [2012] SGHC 41 (Breezeway Overseas Ltd v UBS AG)
- Goodale & Ors v The Ministry of Justice & Ors [2009] EWHC B41 (QB)
Source Documents
This article analyses [2012] SGHC 41 for legal research and educational purposes. It does not constitute legal advice. Readers should consult the full judgment for the Court's complete reasoning.