In re Omnicom Group Inc. Securities Litigation

MEMORANDUM & ORDER

DOLINGER, United States Magistrate Judge.

Plaintiffs in this securities-fraud class action are challenging a wide array of attorney-client privilege claims asserted by the defendants. In particular, plaintiffs seek a large number of written communications between defendants and the law firm Jones Day concerning the corporate transactions that are the target of plaintiffs’ fraud claims in this case. They also seek a host of documents that were either authored by or sent or copied to an attorney named Robert Profusek during the period when he was working at Omnicom Group, Inc. Defendants have resisted these demands, although they have provided some additional documents to plaintiffs during the pendency of this motion. We have invited and received an ambitious volume of argumentation and some evidence in support of the parties’ respective positions, and we now grant plaintiffs’ application in part and deny it in part.

The Positions of the Parties

Plaintiffs press four theories in support of their application for more disclosure by defendants. They seek the broadest production of documents based on their contention that documents reflecting communications between Omnicom executives and corporate counsel at Jones Day are producible, despite their otherwise privileged status, because they contain communications that were intended by defendants to facilitate or conceal fraudulent conduct. The fraud in question is the same securities fraud as plaintiffs allege in their complaint — that is, an attempt by defendants to avoid including on Omnicom’s books substantial losses incurred by entities that were under the control of Omnicom, thereby concealing the actual financial losses that the company sustained through the downturn in the financial fortunes of those related entities.¹

Defendants respond by denying that they engaged in any fraud, and they contend that in any event plaintiffs have failed to proffer sufficient evidence to sustain their burden of proving a sufficient likelihood of fraud. Instead, defendants suggest that they complied with pertinent accounting rules, and they note that two major accounting firms approved their treatment of those transactions on Omnicom’s books. Defendants point specifically to memos by its accountants Arthur Anderson and KPMG analyzing the relevant accounting issues and concluding that Omnicom should not consolidate Seneca on its financial statements. (Def.’s Memorandum of Law in Opposition to Plaintiffs Motion to Compel at 12-13 & Exs. 13 & 20). Defendants further note that the Securities Exchange Commission reviewed the matter and chose to take no action. Finally, they assert that even if they deviated from proper accounting practices, they did not act with fraudulent intent.

Plaintiffs’ second theory for compelling additional production is that, as shareholders of Omnicom, they are entitled to view the Jones Day communications despite defendants’ invocation of the privilege, because the directors and officers of Omnicom and corporate counsel were acting as fiduciaries for the company and, by extension, for the shareholders. Plaintiffs thus seek to invoke what has been referred to as the fiduciary exception to the privilege. Defendants oppose, arguing principally that the fiduciary exception either does not apply to non-derivative shareholder suits or applies only in very narrow circumstances, which are not found in this case.

Third, plaintiffs argue that some subset of the universe of Jones Day documents should be required to be produced because defendants have waived the privilege. The basis for this argument is plaintiffs’ contention that defendants have, in effect, asserted an adviee-of-eounsel defense, which would constitute a waiver both for the advice itself and for documents reflecting information made known to the attorneys before they offered their advice. Plaintiffs further assert that even if defendants are relying on the advice of their accountants rather than their attorneys, waiver must follow because the accountants themselves relied on two opinion letters from Jones Day.

Defendants counter that they do not assert an advice-of-counsel defense, but rather are relying, at most, on opinions of the accountants, and that in such a circumstance the advice-of-counsel waiver does not apply. They further assert that both the attorneys and the accountants, in formulating their opinions, did not rely on privileged information, but rather rested explicitly on stated assumptions as to the facts, thus negating any need to inquire into the attorneys’ communications with the client. Finally, they argue that in any event the accountants never saw any attorney-client privileged materials, thus further underlining the absence of any waiver.

The last issue that we address concerns an array of communications either authored by or sent or copied to Mr. Profusek while he worked as a corporate executive at Omnicom. Plaintiffs say that, although a trained attorney, he was not acting in that capacity when he served as an executive vice president at Omnicom, and that hence there is no basis for defendants’ contention that these communications are privileged, since Profusek was not acting as an attorney whose advice was being solicited by, or offered to, a client. Defendants respond by insisting that Mr. Profusek, despite having been hired by Omnicom to perform business functions, in fact served principally as a legal advisor to the company, and hence documents sent by him or to him or on which he was copied reflect communications that were intended to, or did in fact, elicit legal advice from him.

ANALYSIS

In assessing the parties’ respective arguments, we start with a reiteration of some well-settled principles that govern the core issues triggered by an invocation of the attorney-client privilege. The governing law is federal, since the claims in this case arise under the federal securities laws. See Fed.R.Civ.P. 501; American S.S. Owners Mut. P & I Ass’n, Inc. v. Alcoa S.S. Co., 232 F.R.D. 191, 194-96 (S.D.N.Y.2005); Falise v. American Tob. Co., 193 F.R.D. 73, 79 (E.D.N.Y. 2000). The party invoking the privilege has the burden of proving the facts on which the privilege claim is based, and must do so by competent and specific evidence, rather than by eonclusory or ipse dixit assertions. See, e.g., United States v. Doe (In re Grand Jury Proceedings), 219 F.3d 175, 182 (2d Cir. 2000); von Bulow v. von Bulow, 811 F.2d 136, 144 (2d Cir.1987); In re Bonanno, 344 F.2d 830, 833 (2d Cir.1965).

The privilege covers communications between the client (or the client’s representative) and his attorney (or the attorney’s representative) that are maintained in confidence and that are undertaken to facilitate or request the rendition of legal advice or the performance of other legal services by the attorney or that constitute the communication of such advice by the attorney. See, e.g., United States v. Int’l Bhd. Of Teamsters, 119 F.3d 210, 214 (2d Cir.1997); United States v. Construction Prods. Research, Inc., 73 F.3d 464, 473 (2d Cir.1996); In re Richard Roe, Inc., 68 F.3d 38, 39-40 (2d Cir.1995). Communications that seek or involve principally the performance by the attorney of non-legal functions are not protected. See, e.g., In re Grand Jury Subpoena Duces Tecum, 731 F.2d 1032, 1037 (2d Cir.1984).

In this case, the parties focus, with one exception, on a series of issues that concern limitations placed on the availability of the privilege despite literal compliance with the general definition of the privilege. We address each of these issues seriatim.

A. The Crime-Fraud Exception

The federal courts have recognized an exception to the privilege for those communications between client and counsel that “are ‘relate[d] to client communications in furtherance of contemplated or ongoing criminal or fraudulent conduct.’ ” United States v. Jacobs, 117 F.3d 82, 87 (2d Cir.1997) (quoting In re John Doe, Inc., 13 F.3d 633, 636 (2d Cir.1994); In re Grand Jury Subpoena Duces Tecum, 731 F.2d at 1038). As further elucidated by the Second Circuit and other courts, the controlling question is whether the communications at issue were undertaken to facilitate or conceal the commission of a crime or fraud. See, e.g., In re Richard Roe, Inc., 168 F.3d 69, 70 (2d Cir.1999) (quoting In re Richard Roe, Inc., 68 F.3d at 40); Liberty Environmental Systems, Inc. v. County of Westchester, 1997 WL 471053, *5 (S.D.N.Y. Aug.18,1997). The pertinent intent is that of the client, not the attorney. See, e.g., In re Grand Jury Proceedings (Gregory P. Violette), 183 F.3d 71, 79 (1st Cir.1999)(quoting, inter alia, Clark v. United States, 289 U.S. 1, 15, 53 S.Ct. 465, 77 L.Ed. 993 (1933)).

According to the Second Circuit, to justify invocation of this exception to the privilege, the discovering party bears the burden to demonstrate that there is “probable cause to believe that a fraud or crime has been committed and that the communications in question were in furtherance of the crime or fraud.” Jacobs, 117 F.3d at 87. Accord, e.g., In re Richard Roe, Inc., 168 F.3d at 70 (quoting In re Richard Roe, Inc., 68 F.3d at 40). This burden is not satisfied by a showing that the material in question “might provide evidence of a crime or fraud.” Id. at 71. Rather, the communication itself must have been in furtherance of a fraud or crime and must have been intended to facilitate the fraud or crime. Id. See, e.g., Jacobs, 117 F.3d at 88 (quoting United States v. White, 887 F.2d 267, 271 (D.C.Cir.1989) (Ruth Bader Ginsburg, J.)).

As for the meaning of the term “probable cause” in this context, the Second Circuit has described it as equivalent to “a reasonable basis for believing that the objective was fraudulent.” In re Grand Jury Subpoena Duces Tecum, 731 F.2d at 1039. Whether described as a “probable cause” test or, as in some circuits, a “prima facie showing”, see, e.g., In re Grand Jury Proceedings, 689 F.2d 1351, 1352 (11th Cir.1982), this standard “require[s] that a prudent person have a reasonable basis to suspect the perpetration or attempted perpetration of a crime or fraud and that the communications were in furtherance thereof.” In re Grand Jury Subpoena, 731 F.2d at 1039. In applying this test, the Second Circuit has emphasized the broad discretion of the trial court:

A party wishing to invoke the crime-fraud exception must demonstrate that there is a factual basis for a showing of probable cause to believe that a fraud or crime has been committed and that the communications in question were in furtherance of the fraud or crime. This is a two-step process. First, the proposed factual basis must strike “a prudent person” as constituting “a reasonable basis to suspect the perpetration or attempted perpetration of a crime or fraud, and that the communications were in furtherance thereof.” In re John Doe, 13 F.3d at 637 (quoting In re Grand Jury, 731 F.2d at 1039). Once there is a showing of a factual basis, the decision whether to engage in an in camera review of the evidence lies in the discretion of the district court. Zolin, 491 U.S. at 572, 109 S.Ct. at 2630-31. Second, if and when there has been an in camera review, the district court exercises its discretion again to determine whether the facts are such that the exception applies.

Jacobs, 117 F.3d at 87.

Bearing these general standards in mind, we conclude that plaintiffs have not met their burden, at least on the current record. In approaching this dispute, we first review a mode of analysis suggested by the court in Laser Inds. Ltd. v. Reliant Technologies, Inc., 167 F.R.D. 417 (N.D.Cal.1996). In that case, the court concluded, in substance, that in applying the “prima facie” test (or, by implication, the probable-cause standard) to an argument for invocation of the crime-fraud exception, it has discretion to reject such a claim unless the movant has demonstrated that it is more likely than not that the otherwise privileged communications were used by the client to assist in perpetrating a common-law fraud. See id. at 431-40.

We focus initially on the degree of certainty needed to set aside the privilege because plaintiffs’ motion triggers a number of concerns about the ease with which the attorney-client privilege may be invaded. First, we note that the effect of granting the motion would be to eliminate any protection for a wide array of otherwise privileged documents. There is no question that the courts have placed considerable emphasis on protecting the legitimate use of the attorney-client privilege because of the significant policies that the privilege embodies. As the Supreme Court stated in Upjohn Co. v. United States, 449 U.S. 383, 101 S.Ct. 677, 66 L.Ed.2d 584 (1981), “the attorney-client privilege is the oldest of the privileges for confidential communications known to the common law” and is enforced in order “to encourage full and frank communication between attorneys and their clients and thereby promote broader public interests in the observance of law and administration of justice.” Id. at 389, 101 S.Ct. 677.² Necessarily, then, court findings that would deprive the client of that protection should be based on an adequate record and bear some assurance of reliability. Second, any findings by the court that would suggest a strong enough basis to infer the perpetration of a fraud when such fraud is an essential elemerit of the plaintiffs’ underlying claims in this case would, at the very least, potentially tilt the playing field of this lawsuit at a relatively early stage in the litigation. Third, the plaintiffs’ application in this case is made at a relatively early phase of pretrial proceedings, that is, after document production but before the commencement of depositions, and of course before the proffering of expert-witness testimony. The need for a comprehensive record on such a motion is apparent in any ease, but it is particularly so in the context of a lawsuit that involves, at the very least, technical issues of accounting as to which expert opinions may well differ and concerning which the court has little or no independent information. All of these considerations speak, as a policy matter, in favor of some caution in setting the height of the bar that the plaintiffs must surmount to pierce the defendants’ privilege.

Recognizing the pertinence of these concerns, we note that the Second Circuit and a number of other circuit courts have not addressed in meaningful detail the actual burden that the movant carries in seeking to pierce the privilege on the basis of the crime-fraud exception and, most importantly, the degree of discretion available to the trial court in assessing such an argument by the movant. The terms “probable cause” and “prima facie ease” have certain recognized meanings in a variety of contexts, but it is not self-evident that they are to be applied without further gloss in the current circumstances. Indeed, the Supreme Court has noted that the actual burden to be imposed on a movant seeking to defeat a privilege in this way is still an open question. See United States v. Zolin, 491 U.S. 554, 563 n. 7, 109 S.Ct. 2619, 105 L.Ed.2d 469 (1989) (referring to Clark, 289 U.S. at 14, 53 S.Ct. 465, which first mentioned the term “prima facie case” in dictum relating to the attorney-client privilege). Moreover, even though we of course accept the general formulation by the circuit court of the movant’s burden, it is not all clear that the court lacks discretion, in appropriate circumstances, to require more then literal compliance with those minimum standards. See, e.g., Jacobs, 117 F.3d at 87 (emphasizing trial-court discretion both as to whether to conduct in camera review and in “determining] whether the facts are such that the exception applies”).

As Magistrate Judge Brasil points out in Laser Industries, the courts that have invoked the “prima facie” test — which the Second Circuit has deemed the rough equivalent of its “ ‘probable cause’ test”, e.g., In re Grand Jury Subpoena, 731 F.2d at 1039— have differed significantly in explaining how that standard is to be applied and what it means substantively, that is, in terms of the degree of certainty to be achieved by the movant in order to prevail. See 167 F.R.D. at 432-36.³ Moreover, the Second Circuit’s linkage of the terms “probable cause” and “prima facie case” in this setting suggests the appropriateness of not merely adopting the usages of these terms from their more traditional contexts, since in those contexts they appear to have distinct and somewhat different meanings.⁴

At least one circuit court has held, in substance, that under the “prima-facie” test, a court need not invoke the crime-fraud exception even if it concludes that there is enough evidence in the record to permit a trier of fact to find that a crime or fraud was committed with the assistance of attorney-client communications. See Duplan Corp., 540 F.2d at 1220 (discussing Wright v. Rockefeller, 376 U.S. 52, 57, 84 S.Ct. 603, 11 L.Ed.2d 512 (1964)). The cited passage from Wright, which was a discrimination case, stated that although the plaintiff had satisfied his prima facie burden, the record as a whole allowed the trier of fact to decide the case either way, and the Supreme Court observed that this meant that the inferences suggested by the prima facie case would not bind the trier of fact. 376 U.S. at 55-57, 84 S.Ct. 603. The significance of this passage, as cited in Du-plan, is that “[w]hile such a showing may justify a finding in favor of the offering party, it does not necessarily compel that finding.” 540 F.2d at 1220 (italics in original). As summarized in Laser Industries, the Du-plan analysis

empowers a trial court, even in the face of a showing sufficient to make rational a finding of the disputed crime or fraud, to decline to make such a finding as long as, taking into account “the entire record,” the judge could rationally conclude that the persuasive support for the contrary inference was at least equal to the persuasive support for the crime/fraud inference. Stated differently, in these evidentiary settings, the making of a “prima facie case” would not compel a court to find that the challenger has established the crime/fraud exception to the privilege.

167 F.R.D. at 435 (italics in original; footnote omitted).

It bears further emphasis that the Fourth Circuit in Duplan took particular note, as we have done, of the timing of the crime-fraud motion. The Duplan court pointed out that a motion made during discovery is not the movant’s last opportunity to seek to pierce the privilege on that basis. Rather, that party will have the opportunity to do so again at the time of trial, when he may have additional evidence to proffer in support of his contentions. 540 F.2d at 1222. At that point “[t]he district judge will have the advantage of seeing the witnesses and hearing them testify.” Id. As Judge Brasil observed:

the judge will have more direct and systematic exposure to all the pertinent evidence — and therefore will be in a better position to rule reliably on the important issues presented by such a challenge. These considerations further support the view that judges should exercise considerable caution when they are pressed during the discovery stage of complex litigation to find that a showing of crime or fraud that is sufficient to justify penetrating the privilege has been made.

167 F.R.D. at 436. This caution is particularly appropriate when, as here, the motion is made prior to any depositions and prior to the generation of what will almost certainly be pertinent expert-witness reports.

In assessing the specific standards that are to be brought to bear in that setting, and seeking in the process a specific basis for imposing a more robust standard for breaching the privilege, the court in Laser Industries looked to the terms of Fed.R.Evid. 104(a), which provides that “[preliminary questions concerning ... the existence of a privilege ... shall be made by the trial court ____” In making such preliminary determinations under this portion of Rule 104(a), the courts have held that the trial judge is required to use a “preponderance of the evidence” standard. See, e.g., Bourjaily v. United States, 483 U.S. 171, 175-76, 107 S.Ct. 2775, 97 L.Ed.2d 144 (1987); In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 743 n. 9 (3rd Cir.1994); Laser Industries, 167 F.R.D. at 436. This legal authority also suggests that the issue for the court in this context may not be simply what a-hypothetical juror might be able to find based on the proffered evidence, but rather what the judge concludes about the pertinent facts based on that evidence. See id. at 437-38 & n. 30. Accord In re Heritage Bond Litig., 2004 WL 1970058, *3 (C.D.Cal. July 23, 2004); Medical Laboratory Mgt. Consultants v. American Broadcasting Cos., 30 F.Supp.2d 1182, 1206 (D.Ariz.1998).

We recognize that some of the language used by the Second Circuit may be read as suggesting the contrary. See In re Grand Jury, 731 F.2d at 1039; see also Jacobs, 117 F.3d at 88. The wording in question, however, appears to have been dictum, and other language from the Second Circuit can be understood as allowing the judge to determine whether he is persuaded by the evidence. See, e.g., id. at 87 (judge is to “determine whether the facts are such that the exception applies”); In re Richard Roe, Inc., 68 F.3d at 40 (“a party seeking to invoke the crime-fraud exception must at least demonstrate that there is probable cause to believe that a crime or fraud has been attempted ...”) (emphasis added). Moreover, our circuit court has apparently never been called upon to determine the significance of Rule 104(a) for this question. See Laser Industries, 167 F.R.D. at 436 (discussing Laurins, 857 F.2d at 541). See also In re Sealed Case, 754 F.2d at 399. In addition, these cases all involved criminal proceedings, in which the balance of interests favoring disclosure, even in the absence of proof of a probability of criminal conduct, may be different from what it is in civil litigation. See Laser Industries, 167 F.R.D. at 438.

The invocation of this somewhat more stringent test for probable cause would serve a number of important policy interests. First, this approach would enhance the reliability of the court’s ruling. “If it is important to preserve the vitality of the privilege, it is important to make sure that decisions that strip parties of privilege protections that otherwise would attach are reliably made.” Id. at 440. Second, this concern is particularly acute “when the fraud allegation is based on technically dense or esoteric evidence (matters well beyond the common experience of generalist judges).” Id.

It further bears emphasis that invasion of the privilege — an event of some moment — is difficult to justify if the judge deciding the matter is unable to say, after review of the record, that he believes that it is more likely than not that the client misused the privilege to engage in communications intended to facilitate or conceal a crime or a fraud. To accept a looser standard would mean, in effect, that in any case in which a fraud claim is asserted and the plaintiff proffers enough evidence to survive a hypothetical summary judgment motion on the issue of fraud — no matter how early in -the proceedings — the defendant would likely lose any protection for communications made with an attorney during the period of the alleged fraud. This result seems implausible in view of the functional importance of the privilege, which the federal courts have repeatedly recognized.

The remaining pertinent considerations were well summarized in Laser Industries:

Using a less substantial, poorly defined, and only vaguely understood standard also makes rulings on challenges to the privilege less predictable. When the rulings are less predictable, the incentives to launch challenges increase — as do the burdens on the courts. More importantly, less predictable outcomes undermine, fundamentally, clients’ confidence in the permanence of the confidentiality of their private communications with their counsel---- [Wjithout that confidence, clients are less likely to share all pertinent information with their lawyers. And the core purpose of the law of privilege is defeated when clients withhold pertinent information from their clients.

Id. at 441.⁵ With these considerations in mind, we briefly address the record before us, which includes a substantial quantity of exhibits, some affidavits and an in camera review that we have conducted of a limited number of documents currently in dispute.

As we understand the matter, plaintiffs contend that Omnicom and its officers engaged in an attempt to defraud stock purchasers and others by not including on the books of Omnicom the losses being sustained by the so-called e-services assets that the company’s subsidiary Communicade owned or controlled in late 2000 and early 2001. (PL’s September 9, 2005 Letter Motion to Compel at 6-8). Plaintiffs assert that accepted accounting rules required that Omnicom write down its own earnings at the end of 2000 and in the first quarter of 2001 to take into account those losses (id.), and that even after Omnicom transferred ownership in these entities to a newly created company — the so-called Seneca transaction — in early May 2001 and supposedly relinquished control of those troubled entities, it was required to take the losses on its books since it in fact continued to control those entities. (Id. at 8-10). Pointing to the conceded fact that Omnicom did not take such a write-down either at the close of 2000 or at the conclusion of the first quarter of 2001, plaintiffs assert that defendants’ failure to disclose these losses and their decision to engage in what they characterize as improper accounting measures and a sham transfer of the assets to Seneca to conceal these losses constituted a fraud on the investing public. (Id.) Finally, plaintiffs assert that, to the extent that Jones Day was involved in the structuring of the Seneca transaction, it is reasonable to surmise that Omnicom’s communications with that law firm had the purpose of facilitating the claimed fraud by eliciting legal advice as to how to structure the deal to make it appear that Omnicom was relinquishing control of the transferred entities when in fact it did not do so. (Id. at 10-12).

In responding to these allegations, defendants argue that their treatment of the losses was proper under accepted accounting standards (Def.’s Opp’n Mem. at 9-11) and that Omnicom did in fact relinquish control — as that term is understood for pertinent accounting purposes — of these entities through the Seneca transaction. (Id. at 13-14). Defendants further note that the validity of the challenged treatment of the losses in question was the subject of examination and opinion memoranda by both Arthur Anderson and KPMG, as well as an inquiry by the Securities Exchange Commission and that both accounting firms found the accounting in question to be proper,⁶ and the Commission took no action whatsoever after investigating the matter thoroughly. (Id. at 7). In addition, in challenging the claims of fraud, defendants assert that all of the pertinent facts were disclosed to both accounting firms and to the Commission in the form of documentation about the financial status of the affected entities, and the pertinent terms of the Seneca transaction. (Id. at 9,12).

We have reviewed in some detail the evidentiary submissions of both sides in connection with this matter, and find that, at least at this stage, the plaintiffs have not demonstrated a sufficient likelihood that defendants engaged in a fraud. It follows that they have not demonstrated that the otherwise protected communications with counsel were intended to facilitate such a fraud. Moreover, we so conclude regardless of whether we apply a “more likely than not” standard or a more forgiving version of the probable-cause test.

In reaching this conclusion, we emphasize that the record does clearly reflect that the defendants engaged in actions that were intended to avoid attribution of the losses in question to Omnicom. It is certainly possible, when the differences over proper accounting practices are resolved, that plaintiffs may be able to sustain their assertions (1) that defendants deviated from accepted accounting practices in treating the e-services asset losses (including their use of Seneca to shield those losses from Omnicom’s books), and (2) that they acted with fraudulent intent. Whether that is the case, however, is certainly not clear on the current record. Moreover, the actions (or inactions) of both accounting firms and the Commission suggest the contrary, at least absent some clear indication that defendants concealed key information from them. Although plaintiffs so assert, the record before us — including the documentation apparently supplied to those entities — does not clearly reflect that the key facts were hidden. In short, we do not, at this coneededly preliminary stage, see a probability that defendants engaged in fraudulent conduct and that they used their communications with outside counsel to carry out the intended fraud.⁷

Finally, even if we were to apply a less demanding standard than preponderance of the evidence, we would come to the same conclusion. As we have noted, the record at the moment, while voluminous in terms of poundage, is not especially illuminating in terms of the facts central to proof or disproof of plaintiffs’ fraud claims, a weakness that is not especially surprising in view of the current status of discovery. Further proceedings may unearth information that will enhance the clarity and persuasive power of plaintiffs’ case on this point, but that must await additional discovery.

For now we conclude that plaintiffs have not demonstrated an adequate basis for setting aside the privilege with respect to the Jones Day documents based on their invocation of the crime-fraud exception to the privilege.

B. The Fiduciary Exception to the Privilege

Plaintiffs proffer, as an alternative argument for obtaining access to the Jones Day documents, the notion that the officers of Omnicom and the Jones Day attorneys owed a fiduciary duty to the company. According to plaintiffs’ reasoning, as fiduciaries the defendants acted in service both to the company and, by extension, to its shareholders, and since the plaintiff class includes shareholders, defendants cannot withhold from them the fruits of an attorney-client relationship of which they and the company were the intended beneficiaries.

The origin of this fiduciary exception to the privilege is found in the principle that a trustee may not withhold from the beneficiary any communications by the trustee with an attorney that were triggered by the trustee’s need for advice on how to carry out his fiduciary responsibilities. See, e.g., Martin v. Valley Nat’l Bank, 140 F.R.D. 291, 322-23 (S.D.N.Y.1991)(citing eases and II Scott on Trusts § 173 at 1407 (3d ed.1967))(referring to obligation of trustee to disclose attorney-client communications to beneficiary on demand). That principle has been applied with some consistency in a variety of contexts in which a litigant who served as a fiduciary to his litigation adversary may be required to produce these communications with his attorney that were intended to guide his conduct in carrying out his fiduciary responsibilities. See, e.g., American S.S. Owners Mut. P & I Ass’n, Inc., 232 F.R.D. at 201-02; Martin, 140 F.R.D. at 322-23. At the same time, the courts have protected communications by a fiduciary that concerned legal advice personal to the fiduciary, for example, advice as to how to avoid liability in performing his fiduciary duties. See, e.g., United States v. Mett, 178 F.3d 1058, 1063-64 (9th Cir.1999); In re Long Island Lighting Co., 129 F.3d 268, 271 (2d Cir.1997); Lawrence v. Cohn, 2002 WL 109530, *5 & n. 2 (S.D.N.Y. Jan.25, 2002); Hudson v. General Dynamics, 73 F.Supp.2d 201, 202-03 (D.Conn.1999); Martin, 140 F.R.D. at 325.

This principle was adapted by the Fifth Circuit in Garner v. Wolfinbarger, 430 F.2d 1093 (5th Cir.1970), to apply in shareholder derivative litigation against a corporation’s officers. The circuit court held that in a lawsuit alleging that corporate management betrayed the interests of the corporation, the defendants’ attorney-client privilege may have to yield in view of the defendants’ fiduciary obligations to the corporation and its shareholders. In so holding, however, the

court conditioned invasion of the privilege on the requirement that the plaintiffs show good cause for such disclosure.⁸ The Court was careful, moreover, to limit the application of the fiduciary exception in terms pertinent here. Thus, it observed:

The attorney-client privilege still has viability for the corporate client. The corporation is not barred from asserting it merely because those demanding information enjoy the status of stockholders. But where the corporation is in suit against its stockholders on charges of acting inimieally to stockholder interests, protection of those interests as well as those of the corporation and the public require that the availability of the privilege be subject to the right of the stockholders to show cause why it should not be invoked in the particular instance.

Id. at 1103-04.

In the wake of Garner, most courts have demonstrated a willingness to apply this principle beyond the scope of a derivative action, indicating that the exception may apply, in appropriate circumstances, even to shareholder class-action lawsuits. See, e.g., Fausek v. White, 965 F.2d 126, 132-33 (6th Cir.1992); Ward v. Succession of Freeman, 854 F.2d 780, 786 (5th Cir.1988); In re Bairnco Corp. Secs. Litig., 148 F.R.D. 91, 99 n. 4 (S.D.N.Y.1993). Contra, Weil v. Investment/Indicators Research & Mgt., Inc., 647 F.2d 18, 23 (9th Cir.1981). As for the specific scope of the exception in such a context, the courts have not adhered to a common approach.

Some courts appear to assume that in any securities fraud case, the plaintiff class — insofar as it consists of current shareholders— will be entitled, on a showing of good cause, to pierce the privilege otherwise applicable to

communications between corporate management and its attorneys, even if the communications predate plaintiffs’ status as shareholders. See, e.g., RMED Int’l, Inc. v. Sloan’s Supermarkets, Inc., 2003 WL 41996, *4 (S.D.N.Y. Jan.6, 2003). Those decisions have not entirely ignored the distinction between a derivative action and a private securities-fraud lawsuit, but have suggested that this distinction is but one factor to be considered in assessing whether the plaintiffs have shown good cause for piercing the privilege. See, e.g., id. at *4 n. 13 (citing cases). In contrast, other courts have suggested that the rationale for the fiduciary exception limits its application, at the very least, to plaintiffs who were shareholders at the time of the privileged communication, see, e.g., In re Kidder Peabody Secs. Litig., 168 F.R.D. 459, 475 (S.D.N.Y.) (citing Moskowitz v. Lopp, 128 F.R.D. 624, 637 (E.D.Pa.1989); In re Atlantic Financial Mgt. Secs. Litig., 121 F.R.D. 141, 146 (D.Mass.1988)), and to claims of misconduct directed against the interests of the corporation itself. See, e.g., Weil, 647 F.2d at 23.

The rationale of Gamer is focused on the nature of the claims asserted by the plaintiffs. More specifically, it addresses the appropriate handling of communications between the corporation’s managers and its lawyers concerning matters as to which the managers had a fiduciary obligation to the corporation and, by extension, to its then-current shareholders. In a case in which the plaintiffs are challenging actions of management on the basis that those actions were “inimical to stockholder interests,” Gamer, 430 F.2d at 1103, the privilege will be subject to a good-cause analysis to determine whether the privilege should be overridden. Moreover, the insertion of the good-cause requirement as a limitation on the availability of this exception to the privilege is justified by the fact “that there may be divergences of interest between the plaintiff shareholder in a derivative action and the corporation itself.” Lawrence, 2002 WL 109530, *5 (citing Martin, 140 F.R.D. at 326); 2 J. Weinstein & M. Berger, Weinstein’s Evidence 1Í 503(b)[05] at 503-56.14 (2d ed.1990).⁹

Given this focus, we question whether the fiduciary exception should apply routinely in a securities-fraud lawsuit. First, the plaintiffs in such a case are seeking personal benefit and are not seeking to benefit the company, which is the intended beneficiary of fiduciary obligations owed by corporate management. Second, plaintiffs are complaining of alleged misconduct injurious to them as members of the investing public rather than injurious to the corporation and its shareholders at the time of the misconduct. Third, in the typical class-action suit for securities fraud, there is no reason to assume that the class members will have been shareholders at the time of the targeted communications, and if not, they may not be in a position to claim any relevant fiduciary obligation on the part of corporate management in any event.¹⁰ See, e.g., In re Kidder Peabody Secs. Litig., 168 F.R.D. at 475; Moskowitz, 128 F.R.D. at 637; In re Atlantic Finan. Mgt. Secs. Litig., 121 F.R.D. at 146; Quintel Corp. v. Citibank, N.A., 567 F.Supp. 1357, 1363-64 (S.D.N.Y.1983).

In this ease, the claims of the plaintiffs do not target any actions taken by defendants that were assertedly contrary to the interests of the corporation or of its shareholders at the time; rather, they are complaining about conduct that allegedly injured members of the investing public who were considering the purchase of Omnicom shares, and defendants did not have any identified fiduciary obligations to them until they made their purchases. Moreover, the transactions that are at the heart of the complaint and that formed the trigger for the targeted attorney-client communications were undertaken in the absence of a fiduciary relationship to a substantial portion of the class members.¹¹ Under these circumstances, the Garner good-cause analysis does not apply, and the documents in question have therefore not been shown to be subject to the fiduciary exception to the attorney-client privilege.

In any event, even if we were to undertake some form of good-cause analysis, we would come to the same conclusion. In a securities-fraud case, the interests of the plaintiffs are still less congruent with' those of the corporation (and its other shareholders) than is the case in a derivative suit. Hence, the justification for production must perforce be still more compelling. Plaintiffs here fail to make that elevated showing. Given the nature of their claims, we infer that disclosure of privileged documents concerning the transactions in question likely would not be to the benefit of the company and its shareholders. We also do not see any compelling need shown by the plaintiffs. In addition, the plaintiffs have not shown that they represent a substantial portion of the shareholders, nor have they shown, as yet, a “clearly colorable” case (that is, one that is more than simply colorable, as reflected in the denial of a Rule 12(b)(6) dismissal), and the wrongful conduct alleged, although potentially unlawful, is plainly not criminal.

In sum, even under a good-cause analysis, modified to take account of the fact that plaintiffs are suing on their own behalf and not for the corporation, we conclude that they have not, as yet, justified invasion of the privilege. They will be free, however, to revisit the issue on a more substantial showing,

C. Waiver of the Privilege

Plaintiffs next assert that defendants have waived the protection of the attorney-client privilege based on their reliance on two opinion letters authored by the Jones Day firm and their further invocation of advice tendered by the accountants Arthur Anderson and KPMG. They seem to advance two related arguments in this respect. First, they argue that an opinion-of-counsel defense should waive the privilege not only for the opinion in question, but for all related documents. Second, they assert that reliance on the accountants’ opinions should trigger a waiver because the accountants relied on the attorneys’ letters.

Defendants deny that they are invoking a defense of reliance on counsel’s opinion, and they further assert that even if they did, that would not waive the privilege, since the attorneys’ letters — which have been disclosed— were not privileged. They further assert that, even though they may be relying on the accountants’ opinions, such reliance does not trigger a waiver of the attorney-client privilege since the accountants are not attorneys and did not see, much less rely upon, any otherwise privileged materials.

The attorney-client privilege may be waived if the holder of the privilege discloses or consents to the disclosure of a significant part of a privileged communication to someone not a party to the communication. See, e.g., Weizmann Institute of Science v. Neschis, 2004 WL 540480, *3 (S.D.N.Y. March 17, 2004); In re Kidder Peabody Secs. Litig., 168 F.R.D. at 468. Furthermore, if a litigant discloses an otherwise privileged communication, he may be deemed to have waived the privilege with respect to other communications on the same topic. Such an implied waiver is recognized because “[a] defendant may not use the privilege to prejudice his opponent’s case or to disclose some selected communications for self-serving purposes.” United States v. Bilzerian, 926 F.2d 1285, 1292 (2d Cir.1991). Such subject-matter waiver is based on “fairness considerations at work in the context of litigation.” In re von Bulow, 828 F.2d 94, 103 (2d Cir.1987). See, e.g., Falise, 193 F.R.D. at 84-85.

Implied waiver of the privilege will also be recognized — even in the absence of a party’s disclosure of privileged material — when that party “makes factual assertions, the truthfulness of which may be assessed only by an examination of the privileged communications or documents.” Id. at 84. See, e.g., Bilzerian, 926 F.2d at 1292 (citing Hearn v. Rhay, 68 F.R.D. 574, 581 (E.D.Wash.1975)); Weizmann Institute of Science, 2004 WL 540480 at *4-6; In re Kidder Peabody Secs. Litig., 168 F.R.D. at 470. Thus, for example, if a defendant invokes a good-faith defense purportedly based on his having consulted an attorney before engaging in the challenged conduct, invocation of that defense will be deemed to waive the privilege that would otherwise protect his conversations with his attorney on this topic. See, e.g., Bilzerian, 926 F.2d at 1292. See also Sanofi-Synthelabo v. Apotex, Inc., 299 F.Supp.2d 303, 308 (S.D.N.Y.2004); Johnson Matthey, Inc. v. Research Corp., 2002 WL 1728566, *2-4 (S.D.N.Y. July 24, 2002).

In applying such a waiver analysis, we are guided by the Second Circuit’s observation that “because the doctrine is rooted in fairness, we have also cautioned against broad generalizations, stressing that ‘[w]hether fairness requires disclosure ... is best decided on a case by case basis, and depends primarily on the specific context in which the privilege is asserted.’ ” John Doe Co. v. United States, 350 F.3d 299, 302 (2d Cir. 2003) (quoting In re Grand Jury Proceedings, 219 F.3d 175, 183 (2d Cir.2000)). With this caution in mind, we turn to the controversy in this case.

As noted, plaintiffs premise then-waiver argument on the theory that defendants have proffered a defense that is ultimately, if not directly, premised on opinions offered by Omnicom’s counsel. When a party invokes a defense based on the notion that he relied on the advice of counsel, the courts generally treat that stance as a waiver of both the substance of the attorney’s opinion — assuming that it has been given and maintained in confidence — and of all communications by the client to the attorney pertaining to the matters that are addressed by the attorney’s stated opinion. See, e.g., Joy v. North, 692 F.2d 880, 893 (2d Cir.1982); Weizmann Institute of Science, 2004 WL 540480, at *3. This conclusion is justified as a product of several related theories. First, by asserting the opinion-of-counsel defense, defendant has placed the substance of that communication in issue — that is, made access to the opinion essential in order to evaluate the defense — and hence waived its privileged status, if any. Second, if the substance of the opinion — whether privileged or not— must be produced, so too must any otherwise privileged communications that would be needed to judge its provenance and reliability. In effect the defendant has placed those communications in issue as well. See Bilzerian, 926 F.2d at 1292; Garfinkle v. Arcata Nat’l Corp., 64 F.R.D. 688, 689 (S.D.N.Y. 1974). Furthermore, if the opinion was arguably privileged, to permit the defendant to disclose only that opinion and not any communications of information by the defendant to the attorney that may or should have influenced the cited opinion would permit the defendant to engage in selective disclosure of privileged materials, allowing that party to reveal only information that is favorable to its case while concealing what may be unfavorable. Id.

In this case defendants note that the opinion letters by Omnicom’s counsel were not held in confidence or intended to be confidential and hence were never privileged. Rather, they were apparently intended to provide some comfort for corporate management and a basis for Omnicom’s accountants to opine on the propriety of the company’s accounting treatment of the losses of its e-services assets and the Seneca transaction. Defendants further assert that they are not relying on the law firm’s letters, but rather on the opinions of the accountants, who never saw any privileged materials. Accordingly, they say, there has been no waiver of Omnicom’s attorney-client privilege.

Defendants have explicitly abjured any intention to assert a defense based on reliance on the opinions of counsel. Rather, they assert reliance on the opinions of the two accounting firms that reviewed the pertinent transactions. This alone does not necessarily avoid a waiver problem, because it appears that the accountants may have relied, in some respect, on the opinions of the attorneys. If so, considerations of fairness may compel disclosure of confidential communications from defendants that influenced or were otherwise pertinent to those attorney opinions. Such a waiver could follow from the necessity for access to client communications to the attorneys in order to determine the bona fides of the attorneys’ opinions. Cf. Granite Partners L.P. v. Bear, Steams & Co., 184 F.R.D. 49, 52-56 (S.D.N.Y.1999); In re Kidder Peabody Secs. Litig., 168 F.R.D. at 472 (noting affirmative use by defendant of arguably privileged report by attorneys and finding waiver for underlying communications that led to report); Coleco Indust., Inc. v. Universal City Studios, Inc., 110 F.R.D. 688, 689-91 (S.D.N.Y.1986).¹²

Nonetheless, the current record does not adequately support plaintiffs’ waiver arguments. It is not at all clear in what respect and to what extent the accountants relied on the opinions of counsel. Until that question is clarified, we are left with insufficient information to determine whether a waiver may properly be inferred, and, if so, its extent.

We assume that further elucidation on this point may be obtained in the course of depositions, and if so plaintiffs will be free to pursue their application on this ground. Accordingly, this aspect of their motion is denied without prejudice.

D. The Profusek Documents

The remaining issue concerns a set of documents withheld by defendants on the basis that they were either authored by, or addressed or copied to Robert Profusek, who was assertedly providing legal advice to the company and its management. The parties dispute whether Mr. Profusek was in fact functioning as an attorney for Omnicom and whether, in any event, the documents in question reflect confidential attorney-client communications.

As noted before, communications are not protected unless they are intended to facilitate a lawyer’s rendition of the sort of services professionally rendered by attorneys and also are intended to be kept, and actually kept, confidential. See, e.g., Int’l Bhd. Of Teamsters, 119 F.3d at 214; Construction Prods. Research, Inc., 73 F.3d at 473; In re Richard Roe, Inc., 68 F.3d at 39-40. There are occasions when an attorney — particularly one steeped in business affairs — may perform a range of services, some legal and some of a commercial or business nature. If the communication in question is intended to elicit business advice or commercial services by the lawyer or reflects only such advice or services, it is not within the scope of the privilege. See, e.g., In re Grand Jury Subpoena Duces Tecum, 731 F.2d at 1037; see also Lawrence, 2002 WL 109530, *8. It also bears emphasis that a document created for business purposes does not acquire protected status merely because a copy of it is sent to an attorney, even if the attorney may ultimately render legal advice based on it. See, e.g., United States v. International Business Machines Corp., 66 F.R.D. 206, 212-13 (S.D.N.Y.1974); cf. Note Funding Corp. v. Bobian Investment Company, 1995 WL 662402, *3 (S.D.N.Y. Nov.9, 1995).

In this case plaintiffs emphasize that when Mr. Profusek was hired by Omnicom as an executive vice president, both it and he represented in public announcements that he was going to be performing purely business functions, and they further note that his title and placement within the corporate hierarchy were fully consistent with those representations. (Pl.’s September 9, 2006 Letter Motion to Compel at 3; Pl.’s Mem. In Further Support of Motion to Compel at 29-31 & Ex. 56). In addition, they suggest that the descriptions of many of the documents on defendants’ privilege log do not accord with the notion that they come within the privilege. (Pl.’s Motion to Compel at 4-5).

Defendants, while acknowledging that Mr. Profusek was initially hired with the anticipation that he would perform business functions, insist that he ultimately was required to function as a lawyer, offering legal advice to his corporate colleagues. In his own declaration, Mr. Profusek also insists that, despite his original expectations, he ended up performing principally legal functions as transaction counsel. (Declaration of Robert A. Profusek, Esq., sworn to Sept. 23, 2005, at H 5). He also addresses the specific documents in question on a document-by-doeument basis. (Id. at 1t 7).

We have reviewed the disputed documents in camera and conclude that some reflect either the rendition of what appears to be legal advice or, viewed in context, what amounts to a request for legal advice from Mr. Profusek. Others, however, bear no such indication, and reflect solely the performance by him of the functions of a business advisor. The former are protected by the privilege and the latter are not. In addition, we find some documents that are plainly of a business nature, authored by business personnel and sent to other business personnel (including Mr. Profusek), with no indication that the creation of the document was motivated by a desire or reed for legal advice from him. These too are not protected.

Based on our review, defendants will be required to produce to plaintiffs the following documents:

Withheld documents

14, 22, 23, 24 25, 26, 28, 30, 32, 57, 61, 62, 63, 65 (except paragraph encompassing “But ... representation.”), 67, 72, 73, 79, 91, 121, 176, 220, 449, 507, 567.

Redacted documents

2, 3, 4, 8, 38, 70 (disclose January 11, 2001 e-mail from Randall Weisenburger and paragraph 3(b)), 84, 90 (disclose October 12, 2001 e-mail from Randall Weisenburger).

CONCLUSION

For the reasons noted, plaintiffs’ motion is granted to the extent noted, and is otherwise denied.

1 . In brief, as of both December 31, 2000 and March 3, 2001, Omnicom declined to write down losses attributable to its so-called e-services assets, which were for the most part owned by Omnicom's subsidiary Communicade, because defendants deemed those losses to be other than permanent impairments. Then in early May 2001, it orchestrated the creation of an entity known as Seneca, in which it took a non-voting interest, and it transferred a number of the e-service assets from Communicade to Seneca and eliminated them from its books on the theory that it had surrendered control of them, purportedly in compliance with accepted accounting practices and SEC regulations.

Plaintiffs contend that the losses afflicting the e-service assets reflected permanent impairments, a fact known to defendants at the time, and that the Seneca transaction was a sham, because Omnicom retained effective control of Seneca notwithstanding its representations to the contrary.

2 . This conclusion is not at all inconsistent with the observation found in many court decisions that the privilege should be confined to its original purpose, and thus narrowly construed, because it stands in derogation of the principle that cases should be decided, to the extent possible, on their merits. See, e.g., United States v. Goldberger & Dubin, P.C., 935 F.2d 501, 504 (2d Cir.1991); In re Horowitz, 482 F.2d 72, 81 (2d Cir.1973).

3 . These differences include questions as to (1) whether the court should impose a shifting burden of proof on the parties once the movant has proffered sufficient evidence to meet the prima facie requirement; (2) whether the relevant test requires the court (a) to decide whether a reasonable juror could find the facts necessary to pierce the privilege or, alternatively, (b) to determine for itself whether the evidence persuades it to find the necessary facts; and (3) whether the test would justify the court in refusing to pierce the privilege unless persuaded that the facts material to the exception have been proven to its satisfaction as more likely true than not. See, e.g., In re Sealed Case, 754 F.2d 395, 399 (D.C.Cir.1985); Duplan Corp. v. Deering Milliken, Inc., 540 F.2d 1215, 1220 (4th Cir.1976); Matter of Feldberg, 862 F.2d 622, 626 (7th Cir.1988); United States v. Laurins, 857 F.2d 529, 541 (9th Cir.1988).

4 . The term "probable cause”, traditionally invoked in a criminal arrest context, is generally referred to as a sufficient basis for a judge to infer that it is probable that a crime has been committed and that the arrestee committed it. See, e.g., Maryland v. Pringle, 540 U.S. 366, 370-71, 124 S.Ct. 795, 157 L.Ed.2d 769 (2003). The most common usage of the “prima facie case” standard is in employment-rights cases, and it refers to the plaintiff’s initial burden under the so-called McDonnell Douglas test, which requires, inter alia, a proffer of circumstances suggesting that adverse actions against the plaintiff were taken for statutorily impermissible reasons. 411 U.S. 792, 802, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973). Significantly, the courts have characterized the latter test as "minimal”, see, e.g., Fisher v. Vassar College, 114 F.3d 1332, 1335 (2d Cir. 1997) (en banc), a characterization that we would not expect to find in the context of the probable-cause requirement for an arrest or search.

5 . We note that we have had a number of occasions in past cases to address the crime-fraud exception and in several have concluded — without a detailed examination of the weight of evidence needed to invade the privilege — that the movants had satisfied their burden. See Specialty Minerals, Inc. v. Pleuss-Stauffer AG, 2004 WL 42280 (S.D.N.Y. Jan.7, 2004), aff'd, 220 F.R.D. 41 (S.D.N.Y.2004); S.E.C. v. Herman, 2004 WL 964104 (S.D.N.Y. May 5, 2004); Lawrence v. Cohn, 2002 WL 42280 (S.D.N.Y. January 7, 2002). Those cases did not provoke the type of analysis in which we have engaged here because in those cases the evidence of fraud was very clear.

6 . Plaintiffs respond to this point by noting, correctly, that the opinions of the two accounting firms were based on stated assumptions, and they contend that the assumptions were false. On the other hand, it appears that the factual bases for testing these assumptions were available in the documents provided to those firms, as well as to Jones Day, which supplied its own opinion letters, on which the accounting firms both apparently relied as well.

7 . We have not undertaken a full in camera review of the privileged documents both because plaintiffs have not made a sufficiently compelling evidentiary case to trigger such review and because we do not see how those documents are likely to answer the open questions that lead us to view plaintiffs’ case as not adequately proven.

8 . The Fifth Circuit suggested some criteria for assessing good cause in the derivative-suit context:

the number of [plaintiff] shareholders and the percentage of stock they represent; the bona fides of the shareholders; the nature of the shareholders’ claim and whether it is obviously colorable; the apparent necessity or desirability of the shareholders having the information and the availability of it from other sources; whether, if the shareholders’ claim is of wrongful action by the corporation, it is of

action criminal, or illegal but not criminal, or of doubtful legality; whether the communication related to past or to prospective actions; whether the communication is of advice concerning the litigation itself; the extent to which the communication is identified versus the extent to which the shareholders are blindly fishing; the risk of revelation of trade secrets or other information in whose confidentiality the corporation has an interest for independent reasons.

Id. at 1104 (brackets added).

9 . Judge Weinstein notes:

The facts of life are that antagonism between the derivative plaintiff and those who really run (i.e. are) the corporation is a common phenomenon. Unless the shareholder owns a substantial block of shares, the advantages stemming from his victory in the litigation will outweigh any harm that would accrue to him from a decrease in the value of his stock. His decision to prove his claim may lead him to seek information which would be highly detri-

mental to the corporation if it were available to the public.

Id.

10 . In this case, the class, as alleged in the complaint, consists of individuals who purchased shares between February 20, 2001 and June 11, 2002. (Corrected Compl. at II1).

11 . At least, plaintiffs fail to demonstrate the existence of such a fiduciary relationship at the relevant time.

12 . We recognize that the lawyers, in their opinion letters, purportedly premised their opinions on certain stated assumptions as to the facts. Whether their conclusions were based solely on those assumptions, as stated, is not self-evident, however, and in any event if the client passed on information to the attorneys that either contradicted the assumptions or influenced them, that would be highly pertinent to a judgment about whether the defendants effectively manufactured a scenario in which the accountants came to certain pre-ordained conclusions.