Xpedior Creditor Trust v. Credit Suisse First Boston (USA), Inc.

*461 OPINION AND ORDER SCHEINDLIN, District Judge. The Xpedior Creditor Trust (“Xpedior”), as a putative class representative, is suing Credit Suisse First Boston (“CSFB”) (as successor-in-interest to Donaldson Lufkin & Jenrette Securities Corp. (“DLJ”)) for breach of contract. Xpedior alleges that DLJ breached its underwriting contracts with the class members by requiring extra payments from investors in return for receiving allocations of IPOs that it w'as underwriting. See generally Class Action Complaint, Xpedior Creditor Trust v. Credit Suisse First Boston (USA), Inc., 02 Civ. 9149(SAS) (“Compl.”) ¶1 (“Plaintiff alleges, in sum, that the IPO securities of Xpedior and absent class members were underpriced, creating the environment in which DLJ allocated the underpriced IPO stock of these issuers to certain of DLJ’s favored clients; and, directly or indirectly, shared in portions of the profits of such favored clients pursuant to side agreements or understandings. Plaintiff claims that DLJ’s misconduct breached the underwriting agreements with IPO issuers and its covenant of good faith and fair dealing implied in those agreements; breached fiduciary duties DLJ owed to its IPO issuer clients; and/or resulted in DLJ being unjustly enriched.”). See also MDCM Holdings, Inc. v. Credit Suisse First Boston Corp., 216 F.Supp.2d 251, 254-55 (S.D.N.Y.2002) (same allegations in related action). Xpedior now moves for an order to compel CSFB to produce certain documents in connection with this breach of contract action. CSFB cross-moves for a protective order that would require Xpedior to bear half the costs of producing certain electronic documents. I. LEGAL STANDARD Federal Rule of Civil Procedure 26(b)(1) specifies: Parties may obtain discovery regarding any matter, not privileged, that is relevant to the claim or defense of any party.... Relevant information need not be admissible at the trial if • the discovery appears reasonably calculated to lead to the discovery of admissible evidence. Fed.R.Civ.P. 26(b)(1) (emphases added). Thus, as a general matter, all potentially relevant material is discoverable. The court may, however, limit or condition discovery where a request imposes an “undue burden or expense” on the responding party. Fed.R.Civ.P. 26(c). See also Zubulake v. UBS Warburg, 217 F.R.D. 309 (S.D.N.Y.2003). II. XPEDIOR’S MOTIONS Xpedior moves for an order to compel CSFB to produce documents that fall into four categoriés, discussed in turn. A. Communications with DLJ Clients Who Did Not Receive IPO Allocations Xpedior seeks records of communications between DLJ and its clients regarding the IPOs of class members. Although CSFB agrees that Xpedior is entitled to such records with respect to DLJ clients who actually received allocations, it seeks to withhold' records of communications with clients who, though they expressed interest, did not receive allocations. CSFB has declined to produce records of communications between DLJ and those investors who were unable to obtain allocations of over-subscribed IPOs. CSFB argues that such communications are not relevant to Xpedior’s claims. Xpedior counters that the rejected investors may have been told, in effect, “I’m sorry, you will not be receiving any shares of this IPO because you are not willing to *462 pay special consideration to DLJ.” If such records exist, the logical inference that can be drawn from them is that customers who did receive IPO allocations paid special consideration. 1 Standing alone, plaintiffs suggestion might be based on pure speculation. However, Xpedior cites to the following evidence as support for its good faith belief of DLJ’s misconduct. First, there is some evidence that Tie-in agreements 2 and the payment of Undisclosed Compensation 3 were required at major investment banks. CSFB (but not DLJ) and Robertson Stevens, for example, entered into consent decrees with the SEC regarding such activity. Media reports recounting alleged laddering practices are ubiquitous. And a number of former CSFB brokers have sued CSFB in the wake of these scandals, and have referenced or documented tie-in agreements in their complaints. • See 7/31/03 Oral Argument Transcript (“Tr.”) at 13-14 (Statement of Linda P. Nuss-baum). Second, economic data suggests that Xpedior and other class members’ IPOs played out similarly to those of issuers where there has been evidence of laddering arrangements. These two facts, taken together, provide some support for Xpedior’s speculation. See id. at 13. Whether there is any documentation of the tie-in and undisclosed compensation requirements, especially in consumer communications, is unknown. Nevertheless, Xpedior has demonstrated that it has a good faith basis to believe that DLJ required such agreements. Therefore, it is equally likely that those who were excluded from the IPO were made aware of the requirements for purchase or were told why they would not receive an allocation. Xpedior’s request thus seeks relevant material. In addition, this request may be probative of damages. The greater the number of potential investors who were excluded from the IPO, the greater the demand for the IPO shares. Consistent with basic economics, as demand rose — while the supply of IPO shares remained constant — the value of each share also should have increased. Xpedior has alleged that DLJ deliberately underpriced the offering, i.e., that DLJ set the IPO price below what *463 supply and demand would otherwise have dictated. See Compl. ¶¶ 1-2. Assuming that one measure of Xpedior’s damages is the extent of the underpricing, investors’ indications of interest are relevant to that calculation. CSFB argues that, even if relevant, this production would be unduly burdensome. “To comply with Xpedior’s request, CSFB would have to define the vague category of customers who expressed an interest whether formally at roadshows, informally, or otherwise in buying IPO shares, identify customers in that category, and search for their communications with DLJ.” 7/16/03 Letter from Peter K. Vigeland to the Court (quotation marks and citation omitted). But an underwriter is required to keep records of investors who express interest in buying offering shares, usually through forms called “indications of interest.” See Note to SEC Rule 144(g)(2)(h) (“The broker should obtain and retain in his files written evidence of indications of bona fide unsolicited interest by his customers in the securities at the time such indications are received.”) (codifying Securities Act Release No. 5306 (Sept. 26, 1972)). Xpedior requested documents evidencing communications with customers who expressed an interest in IPO allocations, “whether formally at roadshows, informally, or otherwise.” It is unclear what Xpedior means by “informal” expressions of interest. To the extent that these informal indications of interest are not documented, they obviously cannot be located and need not be produced. Limiting the request to those customers for whom such forms exist will significantly limit the scope of the search, and thus the burden on CSFB. Accordingly, Xpedior’s motion to compel the production of these documents, limited to written indications of interest, is granted. B. IPOs of Non-Class Member Issuers Xpedior next seeks to compel production of documents pertaining to IPOs, underwritten by DLJ, of non-class members. At oral argument on these motions, CSFB indicated that there are eighteen such IPOs. See Tr. at 17 (Statement of Peter K. Vigeland). CSFB argues that the relevance of these documents is low, and that the burden of producing them is high. By comparing who invested in, and who at DLJ worked on, hot IPOs (for class members) and cold IPOs (for non-class members), Xpedior hopes to determine which investors and employees benefitted from “favored customer relationships.” 4 To further this inquiry, Xpedior has limited its document requests to (i) indications of interest, (ii) pot lists (preliminary and final lists of investors receiving shares), and (iii) staffing information. 5 I begin with a brief discussion of relevance. I have already held that CSFB records demonstrating “indications of interest” are relevant. For the same reasons, the pot lists are relevant because they may identify those investors who were excluded from hot IPOs and whether some investors appeared to receive a disproportionate share of the hot IPOs. Finally, staffing assignments could lead to the *464 discovery of important evidence by identifying the brokers who worked on the class members’ IPOs because these brokers may know how DLJ allocated IPO shares to brokers and customers. In short, “the discovery appears reasonably calculated to lead to the discovery of admissible evidence,” Fed.R.Civ.P. 26(b)(1). As to burden, CSFB represented at oral argument that there may be as many as 100 boxes of documents pertaining to these eighteen IPOs. See Tr. at 21 (Statement of Peter K. Vigeland). Although review of these documents is a significant undertaking, it is not unduly burdensome, especially because Xpedior has voluntarily narrowed its requests to easily-identifiable and easy-to-produce documents (i.e., only pot lists and staffing assignments, but not communications with investors, indications of interest, and the other assorted categories of documents that Xpedior has requested for the class members’ IPOs). C. DLJ’s Proprietary Trading in IPO Shares Xpedior next seeks DLJ’s own trading records in IPO shares, on the theory that some of the undisclosed compensation was paid in the form of investors reselling appreciating IPO shares back to DLJ. Xpedior alleges that DLJ profited both by the brokerage commission on the re-sale, and again through its own trading in the IPO stock. CSFB responds that this material is not relevant, because in its Complaint Xpedior never alleged this particular arrangement as a form of undisclosed compensation. But this argument fails to appreciate the distinction between pleading and proof. Xpedior pled undisclosed compensation generically; the re-selling at issue in this discovery request is a form of undisclosed compensation. Moreover, CSFB has conceded—though it originally argued to the contrary—that “trading and commission records for DLJ’s trading on its own behalf are encompassed in the materials already gathered in response to Xpedior’s requests for other trading and commissions data,” and thus “CSFB’s burden in producing this additional material is minimal.” 8/1/03 Letter from Peter K. Vigeland to the Court. Accordingly, Xpedior is entitled to this discovery. D. DLJ’s “Chinese Wall” Policies Finally, Xpedior seeks information pertaining to DLJ’s “Chinese Wall” policy for separating its investment banking and analyst divisions. Xpedior seeks this information both as general background, and as preliminary damages discovery. One potential theory of damages revolves around the true value of the issuer, a figure that was ostensibly determined by DLJ’s analysts when they issued reports to the public. But reports of conflicts of interest between analysts and investment bankers abound, see, e.g., In re Merrill Lynch & Co., Inc. Research Reports Securities Litigation, 272 F.Supp.2d 243 (S.D.N.Y.2003), and any conflict of interest would undermine the reliability of DLJ’s analysts’ valuation. Thus, Xpedior argues that it needs DLJ’s Chinese Wall policies. But Xpedior has failed to articulate a basis upon which to conclude that this information is relevant to its claims. Even supposing that Xpedior’s damages theory is correct, DLJ’s Chinese Wall policy is in no way probative of actual conflicts of interest. If the policy strictly separates DLJ’s analyst and investment banking divisions, that says nothing about whether anyone adhered to the policy. Given that this lawsuit does not present allegations of analyst conflicts, that Xpedior admits that analyst reports are only “one avenue” for *465 valuing the issuers, 7/9/03 Letter from Steven J. Toll to the Court, and that the Chinese Wall policies themselves are of limited use, these documents are simply not relevant. See generally Fed.R.Civ.P. 26(b)(1). Accordingly, CSFB is not required to produce these policies. III. CSFB’S MOTION CSFB moves for a protective order requiring Xpedior to share the cost of restoring certain DLJ computer files housed on decommissioned systems. A. Background The documents at issue (ie., the majority of the documents that Xpedior has requested from CSFB) reside on one of two systems. Files from DLJ’s Investment Banking Division (“IBD”) reside on approximately 1500 optical disks. 1250 of those disks are “near-line” data: they are stored in large “silos” and accessed via three Unix servers. When data is needed, a robotic arm locates the relevant disk and automatically accesses it, akin to a jukebox. 250 of the optical disks are not in the silos, and are therefore not indexed — these disks are likely “off-line,” but are readily accessible. However, when CSFB merged with DLJ, it decommissioned the optical disk system and recycled DLJ’s Unix servers (which contained the indexes of the optical disks). Accessing the optical disks therefore has required CSFB to reconstruct the entire system. Files from DLJ’s Equities Division are stored on Digital Linear Tapes (“DLTs”). Although the DLTs are also accessible “off-line” data, there is no baseline system or complete backup on which to restore the DLT documents; the entire Equities Division server had to be reconstructed from incremental backups. Thus, while the technology was “accessible” at the time that DLJ used it, the technology has been rendered inaccessible by virtue of the fact that CSFB decommissioned it. There is no question that CSFB has spent hundreds of thousands of dollars to restore DLJ’s systems. As a factual matter, the DLJ records were inaccessible at the time this litigation commenced. That being so, it is appropriate to consider cost shifting by analyzing the factors set forth in Zubulake, 217 F.R.D. at 324. 6 B. The Zubulake Factors 1. The extent to which the request is specifically tailored to discover relevant information Xpedior’s requests are appropriately tailored to discover relevant information. DLJ e-mails and word processing documents are the likely source of information about valuation and pricing of issuers’ IPO shares as well as communications regarding IPO allocations. However, because DLJ’s records are only available on the decommissioned systems, restoration is required in order for CSFB to produce any records regarding IPOs underwritten by *466 DLJ, any trading records of DLJ customers, and any records of DLJ’s own proprietary trading. Thus, this factor does not favor cost-shifting. 2. The availability of such information from other sources The documents sought are unavailable from any other source. Although hard copies may exist of some requested information, {e.g., sets of offering binders and original indications of interest forms), the critical pricing, valuation, and customer correspondence that Xpedior has requested (and that CSFB has agreed to produce, at least with respect to class members) is only available from the DLJ servers. This factor does not favor cost-shifting. 3. The total cost of production, compared to the amount in controversy At oral argument, CSFB estimated that the total cost of production would be approximately $400,000. Plaintiff values the litigation at potentially $7 billion, the amount of the alleged underpricing of class members’ IPOs. Even if a class is never certified, Xpedior alone claims damages of $68.7 million. The cost of production is relatively insignificant in comparison. Thus, this factor does not favor cost-shifting. 4. The total cost of production, compared to the resources available to each party CSFB is a large investment bank. According to its form 10-K, CSFB USA alone had net revenues of $5.7 billion in fiscal year 2002. Xpedior is a bankrupt corporation with virtually no assets. Of course, plaintiffs counsel could contribute to the cost of discovery, if necessary. Nonetheless, CSFB’s assets clearly dwarf Xpedior’s. This factor weighs against cost-shifting. 5. The relative ability of each party to control costs and its incentive to do so The work of restoring the DLJ servers is substantially complete, and the cost is now largely fixed. However, to the extent possible, Xpedior has promised to work with CSFB to minimize costs. This factor is essentially neutral. 6. The importance of the issues at stake in the litigation This litigation involves a contract dispute between sophisticated commercial entities, and therefore does not raise the kind of public policy issues that might affect cost-shifting. See Zubulake, 217 F.R.D. at 321 (“For example, if a case has the potential for broad public impact, then public policy weighs heavily in favor of permitting extensive discovery. Cases of this ilk might include toxic tort class actions, environmental actions, so-called ‘impact’ or social reform litigation, cases involving criminal conduct, or cases implicating important legal or constitutional questions.”). While the underlying conduct at issue here does implicate issues of important social concern — namely the alleged manipulation of the securities market — those issues do not predominate here any more than important issues of racial or gender equality predominate in a garden-variety Title VII discrimination case. See Zubulake v. UBS Warburg LLC, 216 F.R.D. 280, 289 (S.D.N.Y.2003) (holding that the “importance of the issues in the litigation” factor is neutral in a gender discrimination lawsuit). Under these circumstances, this factor is neutral. 7. The relative benefits to the parties of obtaining the information. This may be the rare case where both parties benefit from production. Although Xpedior obviously benefits more than does *467 CSFB — because all the documents that it needs to prove its case are contained on the DLJ system, if they exist at all— CSFB would have been required to restore many of the same systems in connection with its production obligations in the related In re Initial Public Offering Securities Litigation. This factor therefore is neutral. 8. Summary In sum, Zubulake factors one, two, three and four weigh against cost-shifting. Factors five, six and seven are neutral. Accordingly, I conclude that cost-shifting is not appropriate in this case; CSFB must bear its own cost of production. 7 IV. CONCLUSION For the reasons discussed above, Xped-ior’s motion to compel is granted in part and denied in part. CSFB’s motion for a protective order is denied. Footnotes [1] This reasoning is known as the “contraposi-tive.” If “if X then Y” is true, then “if not Y then not X” is also necessarily true. Thus, if “If you didn't pay us, then you didn't get an allocation” is true, then so is “If you did get an allocation, then you paid us.” [2] In a tie-in agreement, “underwriters of hot [i.e., oversubscribed] issues require customers, as a condition of participation in a hot issue offering, either (1) to agree to purchase additional shares of the same issue at a later time and at an increased price, or (2) to participate in another hot issue offering.” See In re Initial Public Offering Sec. Litig., 241 F.Supp.2d 281, 305 (S.D.N.Y.2003) (quoting Report of the Securities and Exchange Commission Concerning the Hot Issues Markets, at 37 (Aug. 19 84)). This practice is also sometimes referred to as “laddering.” See Patrick McGeehan, Hedge Fund Managers Said to Talk to Grand Jury, N.Y. Times, May 11, 2001, at Cl. [3] DLJ allegedly profited under this scheme by "requir[ing] their customers to repay a material portion of profits obtained from selling IPO share allocations in the aftermarket through one or more of the following types of transactions: (a) paying inflated brokerage commissions; (b) entering into transactions in otherwise unrelated securities for the primary purpose of generating commissions; and/or (c) purchasing equity offerings underwritten by these IPO Underwriter Defendants, including, but not limited to, secondary (or add-on) offerings that would not be purchased but for the unlawful scheme alleged herein”. In re Initial Public Offering Sec. Litig., 241 F.Supp.2d at 310 (quoting the complaint in that related case). These payments, collectively, have been referred to as “undisclosed compensation.” See id. [4] A "favored customer” is one who preferentially received shares of hot IPOs. In this case, Xpedior alleges that "favored customers” achieved that status by entering into tie-in agreements or otherwise agreeing to pay extra consideration to DLJ. See Compl. ¶ 22. [5] Xpedior has also asked for more extensive discovery against DLJDirect, a hot IPO. In that case, DLJ was both the lead underwriter of the offering and the issuer. Xpedior seeks full discovery of the DLJDirect IPO as if DLJDirect were a class member. [6] In Zubulake, I held that "in conducting the cost-shifting analysis, the following factors should be considered, weighted more-or-less in the following order: 1. The extent to which the request is specifically tailored to discover relevant information; 2. The availability of such information from other sources; 3. The total cost of production, compared to the amount in controversy; 4.The total cost of production, 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; and 7. The relative benefits to the parties of obtaining the information”. 217 F.R.D. at 324. [7] Xpedior had requested that, "if the Court awards cost sharing against Xpedior, reciprocal cost sharing [of approximately $61,000] should be afforded to Xpedior.” 7/16/03 Letter from Linda Nussbaum to the Court. Because CSFB must bear its own costs, no set-off is appropriate.