16 Rev. Litig. 631

The Attorney-Client Privilege Meets the Common Interest Arrangement: Protecting Confidences While Exchanging Information for Mutual Gain

James M. Fischer

Common interest arrangements, when recognized, permit parties with common legal interests to share privileged information without losing protection on that information. Such arrangements are akin to the “joint defense” privilege but encompass a greater scope, as the joint defense privilege requires the information-sharing occur within the context of actual or threatened litigation. This article examines the common interest arrangement and whether preserving the confidentiality of information-sharing arrangements should be restricted to matters in litigation. The author believes that no such limitation should apply but recognizes that as the number of persons exposed to confidential information expands, prevention of defection becomes more difficult. The article explains the common interest arrangement and the limits that apply to information-sharing arrangements, and considers the consequences of an expansive approach to such arrangements.


   I. Introduction On January 9, 1997, the Wall Street Journal[reported that tobacco company Liggett Group, Inc., proposed to turn over its own lawyers' notes taken during three decades of meetings with lawyers from other tobacco companies regarding the industry's tobacco defense. n1 Liggett's decision reportedly was motivated by a desire to extricate itself from the massive litigation hanging over the tobacco industry. n2 Not surprisingly, Liggett's decision was not well received by the other tobacco companies. Charles Wall, deputy general counsel for Philip Morris Companies, objected to Liggett's strategy: "If Liggett's notes reflect the legal discussions at a joint meeting, Liggett has no right to turn those over to anyone without the consent of everyone elip Just because something isn't a litigation matter doesn't mean it isn't privileged." n3    The legal issue raised in the Wall Street Journal article concerns what has become known as the "common interest arrangement." This arrangement, when recognized, permits parties who possess common legal interests to share and exchange attorney client privileged information without that information losing its protected status. n4 The arrangement bears many similarities to what is traditionally referred to as the "joint defense" privilege. n5 Indeed, in some cases the two terms are used interchangeably. n6 The two concepts, however, are distinct in that the joint defense privilege requires the information exchange to occur within the context of actual or threatened litigation. n7 Although the "joint defense" privilege originated as a method by which defendants could share information and maintain a common defense without sacrificing the privileged status of the materials shared, the privilege also applies to cooperating plaintiffs. n8 Thus, I will use the term "joint litigant" privilege to reflect the actual practice. In contrast the common interest arrangement, in theory, can be applied to both litigation and nonlitigation matters. It is the broader application of the common interest arrangement to non litigation matters that many courts have rejected as an impermissible gloss on the law of attorney client privilege.    This Article will examine the common interest arrangement and whether preserving the confidentiality of information sharing arrangements should be restricted to matters in litigation. I believe that no valid reasons exist for such a limitation as long as our legal system continues to recognize the underlying privilege itself. However, as the number of persons exposed to confidential information through a common interest arrangement expands, it may be more difficult to police and prevent defections from the information sharing arrangement, leaving the continued confidentiality of the material vulnerable to a claim of waiver when the defector discloses the shared information to third parties.    II. Terminology That parties may find it in their joint interests to share information, but will be dissuaded from doing so if disclosure causes the information to lose its protected status, is obvious. The law of privilege has responded to this problem by recognizing both "joint client" and "joint litigant" privileges. The "joint client" and "joint litigant" concepts are distinct, although as some commentators have noted, the two concepts are occasionally "mangled." n9 The "joint client" privilege attaches when the clients are represented by a common lawyer. Communications among the clients and their common lawyer remain privileged as against third parties, and the joint client privilege applies to both litigated and nonlitigated matters. The joint litigant privilege, on the other hand, preserves the confidentiality of attorney client privileged matters when they are shared with co parties, even though those parties are represented by separate counsel. This privilege permits co parties to coordinate strategy and present a unified front to a common adversary, but the privilege protects only communications that occur in a litigation context.    The common interest arrangement takes the joint litigant concept one step further. It permits persons to share information outside of the formal structure of litigation in which the information sharers are co parties. The Wall Street Journal article discussing Liggett's efforts to resolve its tobacco related liability exposure illustrates the distinction. Liggett did not offer to turn over material that had been exchanged while Liggett was defending a claim as a co defendant. Instead, Liggett proposed to disclose information gained at meetings held by a group called the Committee of Counsel, which consisted of "in house lawyers from the leading U.S. tobacco marketers." n10 The committee was not formed to defend any single claim but, according to the Journal, was "formed under the aegis of the Tobacco Institute, the industry trade association." n11 The committee reportedly "conferred on a wide variety of issues including health research, public relations, legislation and marketing of "safe cigarettes.' n12 According to the report in the Journal, it was notes of these meetings, rather than discussions of joint defense strategies, that Liggett was planning to divulge. n13    The seminal decision advancing the common interest arrangement is Duplan Corp. v. Deering Milliken, Inc. n14 The case involved complex litigation over the validity of certain patents. During the course of the litigation, the defendants sought access to several thousand documents claimed by the plaintiff to be privileged. A number of these documents, however, had been reviewed by others. Unless that review itself was privileged or within the scope of the existing attorney client privilege, the review by third parties would normally be deemed to waive the privilege. n15    The decision in Duplan Corp. is less important for its holding, which was largely predicated upon its invocation of the now rejected "control group" test, n16 than for its articulation of the standard for the common interest arrangement. The court stated:     A community of interest exists among different persons or separate corporations where they have an identical legal interest with respect to the subject matter of a communication between an attorney and a client concerning legal advice. The third parties receiving copies of the communication and claiming a community of interest may be distinct legal entities from the client receiving the legal advice and may be a non party to any anticipated or pending litigation. The key consideration is that the nature of the interest be identical, not similar, and be legal, not solely commercial. The fact that there may be an overlap of a commercial and a legal interest for a third party does not negate the effect of the legal interest in establishing a community of interest. n17 I will address the definition of "community of interest" later in this Article. The important message for now one that has been accepted by many but not all courts is that the common interest arrangement can extend to nonparties to litigation. n18    III. Sword or Shield When the common interest arrangement is recognized, the effect is to protect privileged intra group communication from discovery by third parties but still allow members of the group to use the communications in any litigation among themselves. n19 In this sense the common interest arrangement acts as a shield to prevent disclosure and use of disseminated privileged materials by nonmembers of the group. Occasionally, however, the common interest arrangement is used as a sword to get access to information by a "member" of the arrangement. For example, a policyholder may assume control of the defense and instruct and direct counsel regarding the defense of the claim. May the carrier compel disclosure to it of policyholder attorney communications on the ground that they relate to a matter, the defense of the claim, as to which the carrier and policyholder share a common legal interest here, reducing or defeating the claim against the policyholder? Most courts that have considered the matter have refused to permit the common interest arrangement to be used as a sword to compel information sharing, offering two alternative rationales for this result. One approach rejects the "common interest" argument on the ground that the policyholder and carrier could not be co parties. n20 The other approach, reaching the same result, treats the policyholder carrier relationship as adversarial. n21 Nevertheless, a few courts have held to the contrary, allowing the common interest arrangement to be used as a sword to access privileged information. n22    Both of the reasons commonly given for not extending the common interest arrangement beyond the confines of the joint litigant privilege (1) the potential for the parties to become adversaries and (2) the lack of co party status fail to support the position taken. Co parties often have divergent interests; indeed, the rules of procedure accommodate the interests of co parties who may wish to assert claims against one another. The fact that parties are adverse in some respects should not preclude them from uniting around shared interests or from freely deciding that their interests in exchanging information outweigh their differences. As one commentator has noted:     Potential adversity exists in virtually every joint defense cases because (1) every defendant can be tempted by the possibility of a superior settlement if it breaks ranks and cooperates with the plaintiff against the other defendants, and (2) most cases present situations where a given defendant, without settling, may attempt to obtain absolution in the fact finder's mind, or to minimize its own liability (as when damages will be apportioned), by pointing the finger at another defendant during trial. In addition, many litigation situations present the prospect that joint defendants could have cross claims against each other. n23 The common interest arrangement represents an interpretation of the attorney client privilege that avoids forcing defendants to elect between either common counsel, with the associated "conflicts of interests" such representation presents, or separate counsel, with its cost of forfeiting the confidentiality of the shared information. n24 Permitting co parties to share information obviates this dilemma. Yet, it would be foolish to suppose that co parties do not have divergent interests. It is the existence of divergent interests that makes co party representation through common counsel difficult. These divergent interests are also what the common adversary seeks to exploit through divide and conquer tactics. n25 Finally, divergent interests also cause co parties to band together to avoid a "race to the bottom" as competition forces each party to seek to advance its own self identified, short term interests. Information exchange cures the information deficit that drives this variant of the "Prisoners' Dilemma" and enables co parties to cooperate and thereby maximize their long term interests. When squarely presented with the issue in this form, courts have not imposed stringent requirements that the common interests dominate; rather, courts have been satisfied where the subject matter of the information was related to an interest that the parties have identified as both common and substantial. n26 Thus, the argument that the parties have "divergent" interests is a weak reed on which to reject recognition of common interest arrangements.     The second reason given for not recognizing the common interest arrangement as a sword is the lack of co party status. n27 This requirement is essentially a variation of the tendency to conflate the "joint litigant" privilege and the "common interest" arrangement. n28 The better view is to recognize that the invocation of the common interest arrangement does not require that all parties to the arrangement have litigation party status. The attorney client privilege itself has not been limited to communications made in connection with or in anticipation of litigation, but has been applied to communications when no litigation is threatened. n29 Persons may as easily find that their legal interests would be served by information sharing in nonlitigation contexts as in litigation contexts. There is nothing sacrosanct about litigation related communications that suggests a greater need for protection from compelled disclosure than would be applied to attorney client communications that preceded the litigation. n30 Cases that have suggested that co party status is a prerequisite to invoking the joint litigant privilege have usually involved situations where only co party assertions were at issue. n31 Since the dominant reason for recognizing the joint litigant privilege and common interest arrangement is to permit individuals to communicate confidentially with their respective attorneys and with each other to advance their shared legal interests, there is no compelling reason to limit those communications to co parties when the parties define their common legal interests more broadly. n32 Of course, in a nonlitigation setting the danger is greater that the underlying communication will be for a commercial purpose rather than for securing legal advice. n33 It is this factor, not the presence or absence of litigation, that more properly limits the application of the privilege.    Returning to a prior example, where a policyholder communicates with a carrier, there is little reason to strip that communication of its preexisting protected status simply because of the nonparty status of the carrier when that nonparty is interested in the matter and the communication advances the interests of both. It is overly formulaic to find that the communication has lost its confidential status because it was communicated to an interested party who was represented by separate counsel, but to maintain confidentiality when both parties are represented by common counsel. The interests advanced by the communication are distinct from the nature of the parties' representation with respect to that communication. Requiring common counsel as a necessary linchpin for maintaining confidentiality requires the policyholder and carrier to use a common counsel arrangement in situations where their interests, as they define them, would be better served by separate counsel. n34    Decisions such as First Pacific Networks, Inc. v. Atlantic Mutual Insurance Co. n35 in which the court refused to recognize the common interest arrangement, represent a distinct situation where the common interest arrangement is used strategically by a nonparty to access confidential information. This application of the common interest arrangement is substantially different from the concern here the ability of parties to an information sharing arrangement to protect shared information from disclosure to third parties. The common interest arrangement is a device that allows parties to agree among themselves how best to use confidential information to advance their mutually agreed upon common interests.    Efforts to use the common interest arrangement as a sword compelling disclosure, rather than as a shield against disclosure, have been consistently and properly rejected because compelled disclosure is the antithesis of the cooperation that underlies the arrangement. The more appropriate basis for distinguishing between common interest situations is to recognize the arrangement when the parties affirmatively seek to cooperate, but to reject forced cooperation as a means of accessing communication between parties who have common interests but have not agreed to share information. n36 This approach retains the benefits of information sharing and avoids the need to engage in a balancing or weighing of "shared" and "divergent" interests to determine whether confidentiality will be maintained or abrogated. n37 The parties themselves, rather than the courts, are in a better position to engage in any balancing. Therefore, the parties' decision should be respected, absent additional, separate factors suggesting that the privilege should be deemed lost.    IV. Nature of the Common Interest Arrangement Although the joint litigant privilege has been widely adopted and the common interest arrangement is gaining in acceptance, no consensus has developed as to the relationship of those doctrines to the attorney client privilege. One view is that the information sharing arrangements are exceptions to waiver rules that normally apply in privilege adjudications. n38 Under this view the requirement of confidentiality flows from the underlying communication. If the communication is not privileged, for example, because it is related to business rather than to legal advice, n39 or because the communication must be disclosed to a third person, n40 no privilege is conferred by sharing the communication with other parties or their lawyers. Under the alternative view, the joint litigant privilege and the common interest arrangement are seen as extensions of an existing privilege. n41 The decisional law is somewhat undefined here. It is unclear how, if at all, this latter view would influence development of legal doctrine in this area. How the common interest arrangement represents an extension of the attorney client privilege and how far reaching the extension is remain open and unanswered questions. One possible, albeit unlikely, application of this view would be to transform nonprivileged individual information into privileged group information when it was shared within the group for the purpose of obtaining legal advice.    While the idea of transforming public information into confidential information runs counter to accepted thinking, it is not completely alien to principles of confidentiality that have come to surround the attorney client relationship. For example, public information may become subject to the professional duty to maintain confidentiality simply because it is acquired by the attorney as a result of the attorney client relationship. n42 Nonetheless, it is probably unwise to urge a court to recognize a new, distinct privilege for shared information. In jurisdictions where evidentiary privileges have been codified by the legislature, such as California, courts have adopted the position that new privileges should not be created through the common law process. n43 In jurisdictions where the legislature has delegated privilege formulation to the courts, as in the federal system, we should expect courts to be reluctant to add to the existing edifice. n44 Moreover, any need to protect shared information does not exist for independent, substantive reasons, as may be the case with "self critical analysis" n45; rather, the need exists simply because of the waiver concept in the current law of privilege. If it is deemed desirable to permit persons to share attorney client privileged information, and if that identified good is deemed to outweigh the interests that underlie the "waiver" principle, then common interest arrangements are justified as an exception to the rule that communication of privileged material to or in the presence of a nonclient "waives" the privilege.    Waiver, as applied to attorney client privilege materials, rests on the idea that disclosure is inconsistent with the privilege's core concept of confidentiality. This easily coheres with cases of voluntary disclosure, but co exists less easily with the inadvertent disclosure cases. Nonetheless, the underlying theme in both situations is that a privilege holder must remain vigilant and consistent with respect to the protected nature of the information. Selective, albeit inadvertent, "disclosure" may introduce discontinuities that cause the privilege to expire. n46 Even in cases of voluntary disclosure, the fact that disclosure advances the privilege holder's legal position does not always prevent a finding of waiver.    This approach to waiver may create problems for persons who wish to create a common interest arrangement. The decision to share information is clearly selective, even as it is intended to further the legal position of the parties to the arrangement. As such, it raises the prospect that a court would find that disclosure defeats the privilege because it is inconsistent with the confidential nature of the materials, even if the receivers of the disclosed materials agree to maintain the confidentiality of the materials.    Arrayed against this argument are the numerous decisions that have identified the mutual interests of the parties to disclosure in sharing materials with each other. n47 Yet, most voluntary disclosures may be assumed to advance the interests of the discloser. This is particularly true the more we assume that individuals act for self interested rather than altruistic reasons. To find a waiver only when the parties to the disclosure share an identical legal interest creates a classification scheme that is self defining but not authenticated by any external criterion other than the definition itself. All other things being equal, how is disclosure to one sharing an identical legal interest to which the privileged materials relate, under circumstances that the discloser believes advances her interests, functionally different from a disclosure to one with a differing legal interest? Disclosure to an adversary certainly can be in the interests of the discloser where, for example, it is undertaken to facilitate settlement; nonetheless, such disclosure usually is deemed to waive the privilege. n48    The argument for treating selective disclosure as a waiver is that it is unfair to allow a person to disclose only favorable information while continuing to treat harmful information as confidential. Such a form of "spin" is deemed to be contrary to the legal system's search for truth. Thus, once helpful confidential information is disclosed, the privilege holder must release the whole of the information, "warts and all." n49    Disclosures among the parties to a common interest arrangement are designed to advance the interests of the discloser and retard or hinder the interests of another, albeit usually someone other than the members of the information sharing consortium. Such disclosures aim to promote the adversary system and enhance the attorney' s ability to represent the client competently. Disclosure, however, is inconsistent with a formal regard for confidentiality, while preserving confidentiality advances the values that underlie the adversary system. n50    Validating information sharing arrangements requires going beyond the core concept of confidentiality and adopting practical goals that justify continued protection of disclosed materials under the attorney client privilege. Such arguments are universally accepted, either expressly or implicitly, in the more limited exceptions, such as the protection for disclosures to interpreters or agents of the lawyer. n51 System enhancing goals underlie both the joint client and joint litigant arrangements. In each of these cases, there has been a disclosure outside of the narrow attorney client relationship without a corresponding finding that disclosure defeated confidentiality. Yet, the disclosure is itself no different from a disclosure to a third party that defeats the privilege. The true difference is not that confidentiality has been preserved, for once the "secret" has been disclosed, it is no longer a secret. Rather, confidentiality is preserved because the values associated with the disclosure to the third person outweigh the interests in treating the privilege as having been waived. Different treatment of the joint litigant privilege on the one hand and the common interest arrangement on the other rests on a formulaic approach to the problem. Rather than analyzing and weighing the interests behind the costs and benefits of information sharing, we have adopted by default a classification scheme that is based on insubstantial criteria for determining when privileged information may be shared without losing its privileged status.    Requiring that an "arrangement" precede a finding that a person can access confidential attorney client information prevents the doctrine from eviscerating the very interests it is designed to protect. This preserves the rule that, as between participants to a common interest arrangement, there is no bar to disclosure. n52 For example, in Worthan & Van Liew v. Superior Court, n53 the court permitted one client to obtain disclosure of information that the co client had shared with their common attorney. In that case, however, the underlying arrangement that created the coclient status put each client on notice that the representation was joint in nature, that information would be shared between the clients, and that it could be used if the parties became adversaries. Yet, acceptance of the principle that parties to an information sharing arrangement do not have an expectation of confidentiality among themselves does not suggest that the members cannot have a reasonable expectation of confidentiality with respect to third parties. Nor does it suggest that the expectation of confidentiality will be maintained by joint agreement and enforced against defectors.    It is for this reason that commentators advise participants in a joint litigant or common interest arrangement to memorialize their understanding. n54 It is not required that such arrangements be in writing, n55 but it is unlikely that a court will read a prior agreement as providing for shared information unless that prior agreement is fairly specific, particularly when, as is normally the case, compelled disclosure is sought in an adversary setting. n56 Thus, as one court noted: "The common interest doctrine, then, has both a theoretical and a practical component. In theory, the parties among whom privileged matter is shared must have a common legal, as opposed to commercial, interest. In practice they must have demonstrated cooperation in formulating a common legal strategy." n57    V. What Limits Apply to Information Sharing Arrangements? There is an inherent tension between information sharing and confidentiality. The more parties there are who have access to information, the more difficult it is to contend that the information is confidential. It is often difficult to define the line that separates "public" from "private." I have argued here that because information sharing arrangements facilitate the sharers' joint interests and because disclosure is limited to the members of the information sharing consortium, the protected status of the information should be maintained. The factors that lead to the basic decision to treat the information as privileged are not substantially altered by an information sharing arrangement. Information exists, after all, to be conveyed. The value of information lies not in its intrinsic self, but in its capacity to be communicated and used. Of course, recognition of the privilege means that information relevant to the decision making process is not accessible by all interested individuals. Since the making of decisions with full information is generally understood to be more desirable than the contrary, the cost of realizing the benefits of privilege recognition is that the decision making process may be less accurate than if decision making were based on all relevant information. The critical issue then is ascertaining the desired balance between (1) respecting a need to exchange information to maximize one's position within a system in which rights are largely defined by law and (2) resolving disputes between parties based on a truncated record because critical, relevant evidence is not admissible. As one commentator perceptively noted, "At some point widespread circulation of privileged information threatens to make a mockery of justice if, due to his inability to obtain the information or offer it in evidence, the opponent is subjected to a judicial result that many others (who do have the information) know to be wrong." n58    This tension is reflected in two maxims that are often found in privilege adjudication cases. The first maxim is that the privilege should be liberally construed in order to further full communication and discussion between attorney and client so that the client may be fully advised of his rights. n59 The second maxim is that the privilege should be narrowly construed because it leads to the exclusion of relevant evidence. n60 These antagonistic positions reflect our ambivalent commitment to the privilege. We should, of course, expect that limits will be placed on any doctrine that operates to exclude information from the decision maker. We should insist, however, that the limits be consistent with the reasons for recognition of the privilege, rather than simply ad hoc responses to the perceived costs that privilege recognition entails.     The proper limitation on information sharing arrangements is that the information shared must be for the purpose of furthering the legal interests of the members of the arrangement. It is impractical to extend the traditional elements of privilege to an information sharing arrangement because the reasons for information sharing are different from the reasons for client lawyer communications. Client lawyer communications are designed to provide the client with competent legal advice so that the client realizes his legal rights and interests. Information sharing is not done primarily for the purpose of securing legal advice, but is engaged in so that the client can maximize his legal rights and interests in settings where joint, coordinated action is preferable to individual action. In many respects, the relationship between the underlying privilege and information sharing arrangements is functionally similar to that between privilege and work product. The objective in both cases is to use and develop information to advise the client. Attorneys will be discouraged from sharing information unless they have assurances that sharing will not waive applicable privileges. We generally recognize that sharing with certain intermediaries, such as physicians, accountants, or investigators, does not waive the privilege. Sharing information among persons with common legal interests falls logically within the same category. The focus of any effort to restrict information sharing arrangements should be directed toward the permissible lengths to which persons may go in maximizing their legal rights and interests. n61 Admittedly, this approach would validate many more information sharing arrangements than has been the case historically.     As noted previously, one limitation on an expansive validation of information sharing arrangements has been the focus on "identical legal interests." This requirement has often been applied with substantial rigor. Thus, the failure of the parties "to coordinate litigation strategy" or engage in a unified defense has been cited as demonstrating the absence of an identical legal interest even though both parties shared the common economic interest of defeating or minimizing exposure on a claim to which both were liable under a reinsurance contract. n62 In another case, the court held that common legal interests could exist when "the parties asserting the privilege were co parties to litigation or reasonably believed that they could be made a party to litigation." n63 Similarly, where a nonexclusive license under a patent was held to have an interest, for business reasons and convenience, in the continued validity of the patent, the court deemed that interest to be different from that of a patent holder who has a monopoly interest in the validity of the patent; hence, no common legal interest protected information shared between a licensee and the patent holder from disclosure through discovery by third parties. n64    Other courts have taken a more expansive and, in my opinion, better approach to the commonality issue. In the patents context, these courts have found that the shared interests of parties in obtaining the greatest protection for or in exploiting patents satisfy the common legal interest requirement. n65 Similarly, in Schachar v. American Academy of Ophthalmology, Inc., n66 the information exchange involved parties and their attorneys who were suing some of the same defen dants for antitrust violations in separate actions in the federal district courts for the Northern District of Illinois and the Northern District of Georgia. The court emphasized that "the joint privilege is meant to recognize "the advantages of, and even, the necessity for, an exchange or pooling of information between attorneys representing parties sharing such a common interest in litigation, actual or prospective.'" n67 In addressing whether the requisite community of interests existed, the court further noted:     Defendants maintain that there is insufficient mutuality of interest between the Schachar plaintiffs and the Vest plaintiffs for a joint privilege to arise regarding work product and attorney client confidences. The court disagrees. Vest was initially filed as a class action; Rule 23 certification was postponed while the parties explored settlement. The present plaintiffs putative members of the Vest class were in effect parties to that litigation and may still stand to gain if that case successfully proceeds as a class action. Moreover, the factual bases for the two cases are sufficiently similar that the attorneys should be allowed to coordinate prosecution efforts to some extent without jeopardizing the confidentiality of their work product or their shared confidences. n68 In In re Celotex Corp., n69 the information exchange occurred between the debtor in bankruptcy and certain defendants who were considered "insiders" and stood to constitute eighty percent of the equity ownership under the debtor's proposed plan of reorganization. Celotex was in bankruptcy as a result of asbestos litigation. Bankruptcy law allows the debtor (Celotex) a period of time in which the debtor has exclusive right to propose a reorganization plan. The various creditor committees sought to end this period of exclusivity and propose competing plans. During the ensuing litigation, the plaintiffs (the creditor committees) sought discovery of information shared by the debtor and the defendants, relating to the reorganization plan proposed by the debtor in the general bankruptcy case. The court found that the defendants and the debtor shared a common legal interest in the bankruptcy case because the information shared was "created postpetition in contemplation of Debtor's confirmation in which Defendants are legally and commercially engulfed." n70    A last example is Durham Industries, Inc. v. North River Insurance Co., n71 in which a policyholder sought discovery of information shared between the carrier and the reinsurers. The information consisted of correspondence between the carrier and its attorneys regarding the policyholder's claim. The court, after quoting the "identical legal interests" test from DuPlan Corp v. Deering Milliken, Inc., n72 held that "where the reinsurers bear a percentage of liability ... their interest is clearly identical to that of [the carrier]." n73    This second group of decisions reflects a greater willingness to permit parties to engage in a cooperative effort to meet a threat that implicates their interests. How coextensive the parties' interests must be in relation to the threat is the point of distinction between the two groups of decisions. The more rigid and narrow approach seen in the first group reveals an unwillingness to give parties much latitude to self define their shared interests and to act in a manner that they deem will best advance their common goals. This approach requires that the "legal" interest refer to a substantive legal interest, i.e., an exclusive right to exploit patents or a right to control the defense of a claim asserted against another for which an indemnity has been promised, in order for information exchanged to retain its protected status as privileged material. Under this view the presence of litigation is an integral component of the common interest doctrine because litigation, actual or threatened, presents the fulcrum on which the parties' common litigation position demonstrates the identical nature of their common legal interests.    Although the decisions in the second group contain some qualifying language, they exhibit a greater willingness to look at economic realities and the strategic considerations that encourage parties to cooperate for mutual protection. In this regard, these cases adopt the approach, suggested by a student commentator, that a common interest exists when the information disclosed is germane to and can be reasonably expected to further an interest that is related to the participants in the information exchange. n74 The common interest arrangement, when practically applied, simply acknowledges that the parties' efforts to cooperate through information exchange will not be deemed inconsistent with the parties' desires to have the information retain its confidential status as to others outside the information sharing arrangement. These decisions give the parties the flexibility to respond to common problems for which legal advice is desirable and permit the parties to share that advice in order to increase the likelihood that their interests will be realized.    The narrow approach to the common interest arrangement is predicated on the notion that protection should only be afforded to information exchanges when those exchanges arise out of the need for a common defense or prosecution, rather than the need to address a common problem. n75 However, the need for legal advice is not limited to litigation settings, and the range of parties who are interested in the resolution of a "problem" is not limited to those who may be made co parties or who have the legal right to assume control of the defense or claim. n76     The narrow view articulated in the first group of decisions suffers from an overly formal and dated view of litigation and the legal system. As disputes become more complex and relationships more interwoven, it becomes increasingly anachronistic to insist on using formal litigation models as guidelines to determine whether information exchanges, which are beneficial to the exchangers, will be encouraged or discouraged. None of the cases in this area identifies any unfair advantages that information exchanges achieve vis a vis their common adversaries. n77 Only a profound enthusiasm for the wooden application of law justifies an approach that discourages information sharing for mutual gain, absent a showing of harm to others.    VI. What Consequences Follow from an Expansive Approach to Information Sharing Agreements? Any consortium composed of members having some divergent interests faces the risk of defection. n78 Indeed, the information gained through membership in an information sharing arrangement makes the defector more valuable than a similarly situated nonmember if the defector can deliver information otherwise unavailable, or available only at greater costs than those associated with procuring the defection. As the defector's value goes up, the pressure on the defector to capture that value by defecting correspondingly increases. To ensure stability, the consortium will insist on some form of security to protect against defection. Not surprisingly, practitioners who have written on this topic have emphasized the need for a written agreement that conditions admission to information sharing arrangements on each member's agreement to maintain the confidentiality of shared information unless relieved of the obligation by agreement of all members of the arrangement. n79 Left unaddressed is the enforceability of these "unanimous consent" provisions. If the provisions are enforceable against a potential defector, the arrangement functions effectively as an agreed to compulsory association from which defection is not permitted. This is accomplished by depriving the defector, and those who would induce defection, of the benefits of defection. n80    A number of courts have refused to enforce provisions that bar a member of an information sharing arrangement from defecting to capture the value of the information. The rationale of these decisions is that a disclosure, even one in breach of an agreement, constitutes a "waiver" of confidentiality. n81 In effect, once the interests of the members of an information sharing arrangement become adverse, their joint ability to enforce the confidentiality provisions inherent in the arrangement lapses.    It should be observed that analysis of the enforceability of confidentiality provisions in common interest arrangements raises two closely related issues. The first issue is the enforceability of the provision as a matter of contract law. The second issue is whether, even assuming that the provision is enforceable, waiver nonetheless occurs upon a breach of the agreement.     There is some authority for the proposition that "secrecy" obligations will be enforced even outside the context of trade secrets. n82 Specifically, the cases dealing with common interest arrangements fall along a continuum that ranges from no enforcement of confidentiality at one end, n83 to limited enforcement in the middle, n84 to full enforcement at the other end. n85    The majority approach, which denies enforcement of secrecy provisions in common interest arrangements, states the weaker view. Permitting defectors to disclose information they have obtained while a member of an information sharing arrangement imposes substantial costs on information sharing consortiums. It is somewhat inconsistent to recognize common interest arrangements as valuable, permissible methods by which persons may advance their common interests, and yet create incentives for members to breach their understanding and their commitment to confidentiality so that they may capture the profit in their newly acquired information. As noted previously, no consortium can withstand defection, particularly when defection is indirectly encouraged and promoted through vigorous judicial enforcement of waiver rules.    It may be argued that permitting enforcement of information exchange agreements would come perilously close to validating obstruction of justice by impeding voluntary testimony by individuals with relevant evidence. It might also be suggested that lawyers who facilitate such arrangements could also be subject to discipline. n86 I believe these concerns are exaggerated unless there is some concrete proof that the information sharing arrangement is a ruse, designed to prevent the parties to the arrangement from voluntarily giving relevant information about each other, rather than to advance their mutual common interests. It is the very nature of privileges that they restrict the decision maker's access to relevant information. Unless the arrangement is entered into for an improper purpose, the fact that it does not facilitate an adversary's access to privileged information is simply a consequence of the decision to afford protection to the information in the first place.    I recognize and do not minimize the strength of the counterargument that parties to an information sharing arrangement cannot have any "legitimate expectation that the attorney client privilege will prevent use of shared materials if one of the parties later becomes an adverse litigant." n87 However, the strength of this assertion rests solely on its premise that the confidentiality provision will not be respected. The suggestion or statement that parties have no reasonable expectation of confidentiality is only meaningful in the context of information sharing arrangements where confidentiality has been expressly bargained for, but the applicable legal rules are such that the parties cannot create a cone of confidentiality around themselves. If, as I assert, confidentiality should be respected because it is more fully consistent with the reasons for allowing information exchange arrangements in the first place, then one would anticipate the opposite result parties would have an expectation of confidentiality because their confidences would be respected by the courts and defection would be deterred.    VII. Conclusion The majority approach, in effect, holds that the parties to an information sharing arrangement assume the risk of defection and cannot count on judicial assistance to minimize the risk even after the parties have done all that they could contractually to prevent defection, or at least deprive the defector of the profits of defection. This approach substantially depreciates the value of information exchange arrangements. However, most courts seem to recognize that such arrangements are useful to participants. It is therefore difficult to square this view with the unwillingness of courts to hold parties to information sharing arrangements to their word and enforce the confidentiality agreements to which they have agreed.    Nevertheless, the danger that a party to an information sharing arrangement may decide to defect cannot be minimized. Indeed, as information sharing arrangements grow in membership and this may be the chief distinction between such arrangements and the more traditional joint client and joint litigant doctrines the ties that bind the parties together may loosen. Absent judicial willingness to enforce promises of joint action, common interest arrangements risk disruption due to individual self interest, i.e., the "Prisoner's Dilemma" in another form.    I appreciate that this approach favoring disclosure is neither perfect nor flawless. It acknowledges that the values that underlie the attorney client privilege are better advanced by encouraging disclo sure in common interest situations than by discouraging disclosure through the threat of waiver. This approach will allow privilege holders to share and use information while denying access to that same information to their opponents. It may also encourage retroactive claims of privilege to shield past disclosures. Nonetheless, on balance the pro disclosure approach is superior because it furthers the fundamental reason for the attorney client privilege: the encouragement of disclosures by clients not for their own sake, but to enable clients to realize fully their legal rights in a society dominated by legal rather than cultural privileges and constraints.    Charles Wall's comment at the beginning of this paper was that the Liggett Group could not unilaterally defect and deliver. That position may be overly optimistic. Rather than saying, "They can't do that," perhaps we should add, "can they?" One may not believe they should be able to defect and deliver, but the case law is sufficiently diverse to support the contrary conclusion. Thus, participants in information sharing arrangements must assume some risk that while the arrangement usually will be respected as against attacks by third parties, the arrangement is subject to attack from within.
Past Issues Subscriptions Submissions Editorial Board Journal Staff
    - TROL Home
- Advisory Board
- Abstracts
- Sponsors
- News