|
|
Home
| Databases
| WorldLII
| Search
| Feedback
Court of Appeal of New Zealand |
![[AustLII]](/images/tiny-austlii.gif)
Court of Appeal of New ZealandLast Updated: 21 December 2011
IN THE COURT OF APPEAL OF NEW ZEALAND
CA198/05BETWEEN COMMERCE
COMMISSION
Appellant
AND TELECOM CORPORATION OF NEW ZEALAND
LIMITED
First
Respondent
AND TELECOM NEW ZEALAND
LIMITED
Second
Respondent
Hearing: 20 July 2006
Court: Glazebrook, Chambers and Robertson JJ
Counsel: J A Farmer QC and G M Coumbe for
Appellant
J E Hodder
for First and Second Respondents
Judgment: 13 September 2006 at 11 am
|
JUDGMENT OF THE COURT
|
____________________________________________________________________
REASONS OF THE COURT
Introduction
[1] The Commerce Commission (Commission) appeals against part of the judgment of Wild J delivered on 23 August 2005 in the High Court at Wellington in which he made a cost-shifting order that required the Commission to pay half of the costs of the test sample of electronic discovery which has to be undertaken by Telecom. There are various estimates as to the amount involved in this sampling with suggestions that it could be as much as $800,000 although it is more likely to be nearer $200,000 to $300,000.
[2] The Commission disputes Wild J’s jurisdiction to make the cost-shifting order and, if the jurisdiction does exist, it challenges the approach taken in making such an order.
Facts
[3] The Commission, in its substantive proceeding, is seeking a declaration and pecuniary penalties against Telecom for an alleged breach of s 36 of the Commerce Act 1986. The claim relates to Telecom’s introduction of its “0867 package” which created a new charging regime for internet access from internet service providers not using an 0867 prefix. The Commission alleges that this package was anti-competitive. Mr Farmer succinctly described the key issue in the substantive litigation as being: what was the purpose of the new regime – was it anti-competitive or did it have a legitimate business reason?
[4] CallPlus Limited (CallPlus), a competitor of Telecom which was adversely affected by the change, commenced proceedings also.
[5] As part of the litigation, an application was made by CallPlus for further discovery of electronic documents held by Telecom. After CallPlus settled its proceedings against Telecom, the Commission stepped into the shoes of CallPlus on this aspect of the case.
[6] The electronic documents sought are no longer immediately accessible as they have been deleted from Telecom’s active computer system and are now available only from backup tapes. These tapes store data in a random way and are not organised for the retrieval of individual documents or files. Restoration is expensive. It appears that a platform will have to be constructed for the exercise to be carried out. Telecom sought cost-shifting on the grounds that even sample electronic discovery would be unduly burdensome.
Judgment under appeal
[7] Wild J, in a judgment of 22 September 2004, ordered Telecom to make electronic discovery, that is to retrieve and discover identified documents from Telecom’s electronic records.
[8] Following an application for recall, in a further judgment of 23 August 2005, Wild J ordered that a “test” sample of the proposed electronic discovery be completed first. This was to be restricted to communications between and related to named personnel. The Judge ordered that the costs of that initial exercise be shared equally between the Commission and Telecom. Order 4 of Wild J dated 23 August 2005 provided:
Telecom’s discovery of its stored electronic documents shall be limited to documents from 1999, the costs of such discovery to be shared 50/50 between the Commerce Commission and Telecom. Once the results of that limited discovery are known, the parties shall have the opportunity to argue why the results and the expense do or do not justify any further discovery of Telecom’s stored electronic documents.
[9] The Judge did not cap the amount to be spent on the initial electronic discovery.
[10] In order that this matter is seen in context, it is noted that many documents were preserved and made available by Telecom. Some 5,500 documents have already been discovered. It is access to additional electronic material (which was not preserved) that is at the heart of this application.
Contentions of the parties
Appellant
[11] Mr Farmer submitted that Wild J did not have jurisdiction to make a cost-shifting order because:
(a) The High Court Rules do not include a general rule for cost-shifting for discovery (the Rules do provide for cost-shifting but not in these circumstances);
(b) New Zealand does not have an equivalent to r 26(c) of the Federal Rules of Civil Procedure 2004 (US) which permits cost-shifting;
(c) Cost-shifting is not authorised by r 48C of the High Court Rules; and
(d) It would not be a proper use of the Court’s inherent jurisdiction to invoke it to allow cost-shifting.
[12] Alternatively, the appellant submitted that, if the Judge did have the jurisdiction to make the cost-shifting order, Wild J adopted the wrong approach in making the order. In particular:
(a) Cost-shifting should not occur until after the initial sampling phase. The purpose of the sampling is to see whether such discovery is unduly burdensome on the party providing discovery thus warranting some cost-shifting;
(b) Wild J failed to take into account a relevant factor, namely that Telecom’s own defaults have contributed significantly to the alleged costs and difficulties of now providing the electronic discovery sought; and
(c) The Judge, having adopted the eight factors formulated by the American Courts in Zubulake v UBS Warburg LLC, 217 FRD 309 (SD NY, 2003) and Wiginton v CB Richard Ellis Inc, 229 FRD 568 (ND Ill, 2003), then failed to correctly apply them.
Respondents
[13] Mr Hodder contended there was no good reason for this Court to review and amend the exercise of discretion by Wild J in the High Court. He argued that:
(a) There was jurisdiction under the High Court Rules for the Court to make a cost-shifting order;
(b) The choice to import principles from United States case law was correctly made and the factors formulated by that case law were correctly applied;
(c) The High Court had regard to a range of factors properly relevant to its exercise of discretion; and
(d) The non-preservation argument was without weight. Telecom took reasonable steps to ensure the preservation of properly discoverable documents and had merely not retained backup tapes.
[14] Mr Hodder submitted that as the parties were in spirited dispute over whether the electronic discovery the Commission sought would yield documents which would enhance the “just, speedy and inexpensive determination” of the litigation (High Court Rules, r 4), the cost-shifting order was a sensible first step.
Discussion
The costs regimes
[15] The costs regimes applicable in the United States of America and New Zealand contain significant differences. The Federal Rules of Civil Procedure provide express rules conferring a general cost-shifting power for discovery. The High Court Rules do not.
[16] In the United States, when a party has allegedly not made full disclosure, the Court has jurisdiction either to order that reasonable costs be paid to the applicant or make a protective order that makes the applicant liable for costs: Federal Rules of Civil Procedure, r 37(a). Rule 26(c) of the Federal Rules of Civil Procedure provides that a Court may make any protective order “which justice requires to protect a party or person from annoyance, embarrassment, oppression or undue burden or expense”.
[17] While the High Court Rules do not provide express rules relating to general cost-shifting, the New Zealand cost regime gives the Court a wide discretion. Rule 46 provides that “[a]ll matters relating to costs of and incidental to a proceeding or a step in a proceeding are at the discretion of the Court”. Rule 4 states that the rules are to be construed so as to ensure “the just, speedy, and inexpensive determination of any proceeding or interlocutory application”. Accordingly we are satisfied that the differences between the costing regimes in New Zealand and the United States do not provide an immediate impediment to the orders made by Wild J. The discretion with regard to ordering costs is wide enough to warrant further investigation as to jurisdiction.
Jurisdiction to order cost-shifting
[18] The usual rule is that the party required to make discovery bears the cost initially. It is inherent in the rules as a whole that initially each side bears its own costs for every step it wishes or is required to take in the proceedings. Discovery is no different from any other step. The costs schedule is predicated on the premise that the winning party can normally expect a contribution from the losing party towards every step it has been required to take. Discovery is one of those steps for which a contribution can be expected in due course by the winning party.
[19] The High Court Rules do include an express rule for shifting of costs in discovery (High Court Rules, r 303), but this is limited to pre-commencement discovery and discovery against a non-party. Mr Farmer argued that this limited provision in the Rules meant that the Court did not have jurisdiction to order cost-shifting in any other circumstances.
[20] In response, Mr Hodder asserted that r 303 was not a code. He also questioned the relevance, let alone influence, of this new rule as it was introduced in a 2004 amendment so it ought not to apply to the current proceedings which pre-date that time.
[21] Mr Hodder focused his attention upon r 236 which provides that the Court may make an interlocutory order “subject to any terms or conditions that the Court thinks just”. The order made by Wild J fell within the definition of an “interlocutory order”: r 3. Wild J had the jurisdiction to impose any conditions that he considered just. Mr Hodder noted that, at the time of Wild J’s original decision, the old interlocutory regime under the High Court Rules prevailed, although by the time of the recall application the new regime applied. But, he said, his jurisdiction argument held good whichever regime applied, as the equivalent to r 236 in the old regime (r 263) also contained a provision that any interlocutory order could be “on such terms and conditions and subject to such undertakings as the Court thinks just”.
[22] We accept Mr Hodder’s submission. In our view, r 236 (and, for that matter, r 263 of the old regime) provides jurisdiction for a cost-shifting order. The fact that r 303 creates a presumption of cost-shifting in particular circumstances does not preclude the possibility of cost-shifting in other circumstances.
[23] Mr Farmer also noted that, as the Rules Committee was currently considering what specific rules should be made to cover electronic discovery, any ruling by the Court would pre-empt their process. We do not agree. There is a case before the Court which requires adjudication. The possibility of the Rules Committee taking action at some future date is no barrier to relief being granted, in a way that is available and appropriate, in terms of the operative regime.
[24] We reject the challenge that there is no jurisdiction to make a cost-shifting order.
Approach to determining whether a cost-shifting order should be applied
(i) Did the Judge misapply the principles he borrowed from United States law?
[25] Wild J based his assessment of whether to order cost-shifting on United States case law. He noted:
[33] My understanding is that under the American Federal Rules of Civil Procedure relating to discovery, the presumption is that the responding party must bear the expense of complying with discovery requests, but that the Court retains a jurisdiction to protect the responding party from “undue burden or expense” in complying, including by conditioning discovery on the requesting party’s payment of part or all of the costs of discovery.
[34] “Undue burden or expense” is defined as a situation where the burden or expense of the proposed discovery outweighs its likely benefit, taking into account such factors as the needs of the case, the parties’ resources, the importance of the issues in the case and the importance of the proposed discovery in resolving those issues.
[35] In Zubulake, a decision of the United States District Court for the Southern District of New York, Scheindlin U.S.D.J. suggested a three-step analysis in cases where discovery of electronic data is sought. First, where electronic data is kept in an accessible format, the usual rules of discovery apply. Cost-shifting should be considered only where the data sought is relatively inaccessible.
[36] Secondly, in most cases, requiring the responding party to restore and produce a small sample of the inaccessible data/media will be a sensible means of determining what information may be retrieved.
[37] Thirdly, in conducting the cost-shifting analysis, seven factors should be considered. Listed in descending order of importance they are:
(1) The extent to which the discovery is specifically tailored to discover relevant information. The less specific the request, the more appropriate it is to shift costs to the requesting party.
(2) The availability of such information from other sources.
(3) The total cost of discovery, compared with the amount in issue.
(4) The total cost of discovery, compared to the resources available to each party.
(5) The relative ability of each party to control costs and its incentive to do so.
(6) The importance of the issues at stake in the litigation (in cases of public importance, extensive discovery should be permitted), and
(7) The relative benefits to the parties of obtaining the information.
[38] In Zubulake the Court considered that the most important inquiry is the “marginal utility” of the discovery, a combination of factors (1) and (2). The Court in that case was critical of a similar eight-factor test suggested by the same Court in Rowe Entertainment & Ors v William Morris Agency, Inc. & Ors, 205 FRD 421 (SD NY 2002), as favouring cost-shifting inappropriately.
[39] Subsequently, in Wiginton, the United States District Court for the Northern District of Illinois, Eastern Division confirmed that the “marginal utility” inquiry is the most important in determining the appropriateness of cost-shifting. The Wiginton Court added a further factor to the Zubulake test:
(8) The importance of the requested discovery in resolving the issues of the litigation.
[26] The Zubulake test was developed as a response to Rowe, which had been the accepted test for determining whether to make a cost-shifting order. Judge Scheindlin entered into an indepth critique of the Rowe test and undertook significant modifications. Her major concern was that the Rowe test was framed in a way that was not neutral and in fact favoured cost-shifting: Zubulake, 320. Wiginton largely affirmed the Zubulake test but included one additional factor.
[27] The draft Federal Rules, which have been approved by the United States Supreme Court, will take effect on 1 December 2006 unless Congress enacts legislation to reject, modify or defer them. The Rules, in their current form, reflect the approach of the United States cases with the addition of one factor discussed below.
[28] The appellant accepted that the experience of the American Courts in this evolving and developing area was not an inappropriate starting point for a Judge. However, Ms Coumbe, who argued this part of the case, contended that the Judge misapplied the second step of the three-part test. She contended that cost-shifting should not be ordered until after a sampling process had been completed.
[29] As a reason for sampling was to determine whether cost-shifting should occur, Ms Coumbe submitted that the producing party should be required to restore backup tapes at its own cost for the sampling. This would assist in obtaining a more accurate assessment of cost and in determining whether the exercise would produce a sufficient proportion of responsive documents. She said that this was the approach taken in Zubulake.
[30] The Zubulake factors are a commonsense encapsulation of factors which require assessment. The United States approach does not have legislative status in New Zealand. Nor does it provide an exhaustive checklist with any deviation constituting an error in the exercise of discretion. The Zubulake factors are, however, of value until the Rules Committee deals with the subject in this country.
[31] To the extent that the Judge used the United States framework, there is nothing to suggest that he misconstrued his task (subject to what is said in the next section). The general approach of Wild J was sensible in the circumstances.
[32] In our view there is nothing in Zubulake, or in the other United States authorities referred to us, to suggest that there is a fixed rule that cost-shifting can occur only after sampling. Although cost-shifting may be less likely to be ordered at a sampling stage, there is no reason, either as a matter of principle or an exercise of pragmatism, why the factors in favour of cost sharing cannot apply to sampling.
(ii) Assessing all relevant factors
[33] Ms Coumbe placed emphasis on a failure by the Judge to take into account that Telecom failed to preserve documents in an accessible format and had thereby contributed to the high costs of recovery. She submitted that this was a relevant factor that weighed heavily against cost-shifting.
[34] The available evidence suggests that in 1999 Telecom had more than 200 file servers which were scattered geographically throughout New Zealand. They ran on Novell Network file server operating systems. Individual Telecom computer users usually had access to a personal storage area on a particular hard disk in a file server.
[35] In 2002 Telecom updated and consolidated this technology. The major changes included:
(a) migration from Novell Netware to Microsoft 2000 Server (conducted in July 2002);
(b) a significant reduction in the number of servers to approximately 50;
(c) consolidation of the distributed server environment into central locations such as data centres; and
(d) implementation of network storage technologies, rather than hard disks in individual file servers.
[36] As part of the changes many of the Novell file servers were decommissioned or redeployed within a reconfigured infrastructure and deleted data was not generally kept.
[37] This change means that backup tapes containing deleted documents from 1999 to 2002 cannot be accessed by the current operating system.
[38] Ms Coumbe contended that when Telecom made this change it knew that there was litigation in contemplation and that it had a duty to ensure that it maintained its records in a form in which they were conveniently accessible.
[39] To fully appreciate the appellant’s argument, it is important to look at the chronology of events. The 0867 package was devised and developed from February to June 1999 and publicly announced on 10 June 1999.
[40] There was a formal request by the Commission to Telecom on 17 August 1999 to provide information. On 13 September 1999 the Commission served Telecom with a notice to supply documents: Commerce Act 1986, s 98. A further notice under that section was issued on 22 December 1999.
[41] On 3 April 2000 CallPlus commenced proceedings for an interim injunction against Telecom. Potter J granted interim relief on 17 April 2000. The Commission commenced its substantive proceedings against Telecom on 31 July 2000. CallPlus formally requested from Telecom discovery of current and deleted electronic documents and back-up tapes on 5 September 2000.
[42] In a letter from Senior Counsel for Telecom on 28 February 2006 it was noted: “... that Telecom regarded litigation as being in contemplation no later than early June 1999 ...”.
[43] As against that background, the failure to preserve documents (which is not specifically mentioned by Wild J) clearly requires consideration.
[44] Wild J applied the Zubulake test. However, that test is not comprehensive in that it does not include a failure to preserve potentially discoverable documents as a factor.
[45] There has been criticism by some commentators that the absence of easy access to documents has not been treated as important: see Lloyd S Van Oostenrijk “Paper or Plastic? Electronic Discovery Spoliation in the Digital Age” 42 Hous. L. Rev 1163.
[46] The United States Judicial Committee has recognised this and has sought to address it. The proposed r 26(b)(2)(C) of the draft Rules now fills this gap. This rule sets out a seven-part test to be applied when determining whether a cost-shifting order ought to be made. The factors are very similar to the Zubulake factors but include:
- (3) the failure to produce relevant information that seems likely to have existed but is no longer available on more easily accessed sources.
[47] Mr Hodder agreed that Wild J did not specifically advert to preservation, but he made the point that it was clearly on the table. It had been discussed in the submissions and therefore it must be assumed that it was a factor which was weighed in the discretionary decision. It did not require identification or specific discussion in the circumstances.
[48] We are, however, persuaded that Wild J erred in not directly addressing this crucial issue of preservation. It is a salient factor which requires specific consideration even in a sampling exercise which is as extensive as will be required here. It is part of the rules regime proposed for the United States. It is an appropriate factor for evaluation in New Zealand.
(iii) What is the consequence of applying this approach to this case?
[49] There is no doubt that, by the middle of 1999, Telecom knew that these proceedings were on foot and they chose to move to the new platform without preserving the means of access to documents which could be relevant.
[50] Mr Hodder was dismissive of the suggestion that there was any failure on the part of Telecom in preserving ready access in this case. Mr Hodder contended that the ‘non-preservation’ point was a misnomer. A substantial initial discovery was provided by Telecom. Further discovery occurred before and after the recalled judgment, including everything from readily accessible electronic files.
[51] We note, however, that there has been very little electronic discovery and that Telecom should have known that files which have become inaccessible were potentially relevant to the proceedings.
[52] Mr Hodder submitted that it was simplistic, if not misleading, to suggest that the substantive issues in the litigation would be determined by what he called covert activity. He said the case was about law, economic analysis and overt commercial initiatives, and not some cloak and dagger issue.
[53] In our view, there will be many factors which will require assessment, analysis and evaluation, but we do not accept Mr Hodder’s submission that contemporaneous communications between those who were involved in the fashioning of the project and its implementation can be immediately discounted. What relevant senior officials of Telecom were saying and doing during the creation and implementation of the package could be most compelling and persuasive evidence on the issue of whether what occurred was anti-competitive or had a legitimate business reason. Whether those communications are in fact probative will only become apparent once seen and evaluated. The sampling exercise will enable a Court to determine whether it is worth this line of discovery being pursued.
[54] We are satisfied that Telecom should have archived, stored and maintained in an accessible form, information of the sort which is now sought. Its failure to do so must be laid at its door.
[55] When that reality is placed into the equation, it is inevitable that the conclusion is reached that the limited sampling should be undertaken by Telecom in conformity with normal practice at its own expense. When the results of the sampling are known, and if it is determined that more general discovery is still required, then a further assessment can be made as to cost-shifting.
Conclusion
[56] In our view, Wild J had jurisdiction to make a cost-shifting order. Given the lack of authority in this country, it was sensible for him to look overseas and to take guidance from the American authorities, such as Zubulake. There is, however, a flaw in the Zubulake test, as has now been recognised in the revised Federal Court Rules. The flaw is that it omitted an important factor, namely an analysis as to why the material is now not easily available. Has that state of affairs come about because of the discovering party’s failure to preserve when it knew it should be preserving? As it happens, that is the decisive factor in this case.
[57] The appeal is allowed. The orders made by Wild J on 23 August 2005 are varied by deleting the reference to cost sharing. The cost of the sample discovery of Telecom’s stored electronic documents is, in the normal manner, the responsibility of Telecom. The appellant will have costs of $6,000 together with usual disbursements.
Solicitors:
Chapman Tripp Sheffield Young, Wellington, for
Respondents
NZLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.nzlii.org/nz/cases/NZCA/2006/252.html