Gatekeepers Inside Out SUNG HUI KIM* TABLE OF CONTENTS I. FRAMEWORK FOR ANALYSIS . . . . . . . . . . . . . . . . . . . . . . . . . . . 415 A. CLASSICAL GATEKEEPING THEORY . . . . . . . . . . . . . . . . . 415 B. TWO COMPLICATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416 1. DISTINGUISHING “MONITORING”. . . . . . . . . . . . . . . . . . . . . . 416 2. MODELING “WILLINGNESS”. . . . . . . . . . . . . . . . . . . . . . . . . 418 FOUR QUADRANTS OF GATEKEEPING . . . . . . . . . . . . . . . . 421 II. WILLINGNESS TO INTERDICT . . . . . . . . . . . . . . . . . . . . . . . . . . . 422 C. A. RATIONAL CHOICE FACTORS . . . . . . . . . . . . . . . . . . . . . . 423 1. HIGH COST OF COMPLICITY . . . . . . . . . . . . . . . . . . . . . . . . . 423 a. Loss of Reputational Capital . . . . . . . . . . . . . . . . . . . . 423 b. Private Civil Liability. . . . . . . . . . . . . . . . . . . . . . . . . 426 c. SEC Enforcement . . . . . . . . . . . . . . . . . . . . . . . . . . . 429 LOW COST OF RESISTANCE . . . . . . . . . . . . . . . . . . . . . . . . . 430 BEHAVIORAL REALIST FACTORS . . . . . . . . . . . . . . . . . . . 436 2. B. * Associate Professor of Law, Southwestern Law School. B.A., M.A., Emory University, 1988. J.D., Harvard Law School, 1992. Special thanks to Mitu Gulati and Mitt Regan. I am grateful for the insightful comments of Ron Aronovsky, Scott Baker, Stephen Bainbridge, Jeffrey Bauman, Stephen Choi, Deborah DeMott, Michael Dorff, Silvia Faerman, David Fagundes, Jim Fischer, Andrew Glickman, Susan Saab Fortney, Bryant Garth, Warren Grimes, Michael Guttentag, Jerry Kang, Ann Kappler, Janine Kim, Kimberly Kirkland, Kimberly Krawiec, Arthur Laby, William Lerach, Robert Lupone, Austen Parrish, Andrew Perlman, Jeffrey Rachlinski, Carol Rakatansky, Robert Rosen, Byron Stier and Kelly Strader. I’d like to thank participants at the Duke Law School Law & Finance Colloquium, sponsored by the Duke Global Capital Markets Center; the symposium on Corporate Compliance: The Role of Company Counsel, co-sponsored by the Georgetown Journal of Legal Ethics and the Center for the Study of the Legal Profession; and the Southwestern Law School Faculty Colloquium. Excellent research assistance was provided by Kyndra Joy Caspar, Nicole Charney, DaBin Kim, Andrew Rawcliffe, Greg Shentwu, and Ledena Veloria. I am also grateful for the efforts of Linda Whisman, David McFadden, Carole Weiner and the dedicated staff of the Leigh H. Taylor Southwestern University Law Library. This article is dedicated to the loving memory of my mother, Kwang Ja Kim. 411 Electronic copy available at: http://ssrn.com/abstract=969536 412 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 1. MERE EMPLOYEE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437 2. TEAM PLAYER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439 3. FAITHFUL AGENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441 III. WILLINGNESS TO MONITOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446 IV. CAPACITY TO MONITOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 448 A. FORMAL COMMUNICATION CHANNELS . . . . . . . . . . . . . . 448 B. INFORMAL COMMUNICATION CHANNELS . . . . . . . . . . . . 453 C. OBJECTION: KNOWING THE LAW . . . . . . . . . . . . . . . . . . . 455 V. CAPACITY TO INTERDICT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457 VI. CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 460 “If we want companies to fuse high performance with high integrity, the place to begin—and to be most effective—is inside the company itself. Outside regulators and gatekeepers can never be as potent and preventative as internal governance on the front lines from the CEO on down.” Ben W. Heineman, Jr., former general counsel, General Electric Company1 INTRODUCTION On September 28, 2006, a visibly exhausted Ann Baskins stood before a Congressional committee investigating the Hewlett Packard (“HP”) spying scandal. Baskins, who had just resigned as general counsel, raised her right hand and swore to tell the truth. Then, on the first question, she exercised her Fifth Amendment privilege against self-incrimination.2 While Baskins sat mute nearby, the former HP chairwoman Patricia Dunn and chief executive officer Mark Hurd told the committee that Baskins was to blame for the spying fiasco. They accused her of rendering bad legal advice and claimed that she knew about and permitted the use of “pretexting”—using false pretenses to obtain certain board members’ personal information from telephone companies.3 Baskins’s ordeal comes at a watershed moment for general counsel of public 1. Ben W. Heineman, Jr., Caught in the Middle, CORP. COUNS., Apr. 2007, at 84, 89 (arguing that inside counsel do and must play the role of “guardian of the corporation’s integrity and reputation”). 2. Sue Reisinger, Saw No Evil, CORP. COUNS., Jan. 2007, at 68. 3. Id. Electronic copy available at: http://ssrn.com/abstract=969536 2008] GATEKEEPERS INSIDE OUT 413 corporations. In a little more than a year, twelve general counsel have stepped down amid allegations of options-backdating at their companies,4 with at least one general counsel pleading guilty to criminal conspiracy.5 Not only have these general counsel lost their jobs but some face civil or criminal indictment and shareholder derivative actions that name them individually as defendants.6 Although it remains to be seen whether the general counsel under suspicion are scapegoats or masterminds, the catalogue of scandals in the past year has certainly soiled the public’s perception of inside counsel. These scandals have lent support to the general consensus that inside lawyers are too “captured” to exercise the independent judgment that is the hallmark of professionalism.7 In the securities context, these scandals have cast doubt about inside counsel’s ability to fulfill their role as “gatekeepers,” private intermediaries who can prevent harm to the securities markets by disrupting the misconduct of their client representatives.8 John Coffee exemplifies the conventional wisdom: While the outside attorney has been increasingly relegated to a specialist’s role and is seldom sought for statesman-like advice, the in-house general counsel seems even less suited to play a gatekeeping role . . . . [T]he in-house counsel is less an independent professional–indeed he is far more exposed to pressure and reprisals than even the outside audit partner.9 True, inside lawyers may well be “more exposed to pressure and reprisals” than outside lawyers or outside auditors; however, this does not necessarily make inside counsel less suited to the gatekeeping role than outside intermediaries. The truth is much more complex. Inside counsel are superior to outside counsel in crucial respects, and outside counsel face their own impediments, which differ in kind (and not just degree) from those encountered on the inside. In short, the 4. D.M. Osborne, The Great Plague: Stock option backdating scandals have taken down a record number of GCs, CORP. COUNS., Jan. 2007, 16. 5. See, e.g., Exit-of-the-Month Club, CORP. COUNS., Jan. 2007, at 17 (reporting Comverse Technology, Inc.’s former general counsel’s guilty plea to criminal charges relating to an options manipulation scheme); Judith Burns, Ex-Counsel At Comverse Settles Lawsuit, WALL ST. J., Jan. 11, 2007, at A11 (reporting settlement of SEC civil lawsuit for $3 million); Jessie Seyfer, The SEC Land a Big One, CORP. COUNS., June 2007, at 16 (reporting the SEC’s filing of civil fraud suit against former Apple Inc. general counsel for backdating stock options); Sue Reisinger, Caught in Backdating’s Web, CORP. COUNS., Feb. 2008 (summarizing back-dating scandals implicating general counsel). 6. Osborne, supra note 4, at 16. 7. See, e.g., Deborah A. DeMott, The Discrete Roles of General Counsel, 74 FORDHAM L. REV. 955, 967 (2005) [hereinafter DeMott, Discrete Roles] (noting that the “conventional skepticism” about in-house lawyers “focuses on the exclusivity of their relationship with a single client [their employer]”). 8. This definition of “gatekeepers” is modified and adapted from that set forth in Reinier H. Kraakman, Gatekeepers: The Anatomy of a Third-Party Enforcement Strategy, 2 J.L. ECON. & ORG. 53 (1986) [hereinafter Kraakman, Gatekeepers] (describing “gatekeeper liability” as “liability imposed on private parties who are able to disrupt misconduct by withholding their cooperation from wrongdoers”). See infra Part I.A. for a discussion on Kraakman’s framework for gatekeeper liability. 9. JOHN C. COFFEE, JR., GATEKEEPERS: THE PROFESSIONS AND CORPORATE GOVERNANCE 195 (2006) [hereinafter COFFEE, GATEKEEPERS]. 414 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 difference between the gatekeeping efficacy of inside and outside counsel is caricatured and exaggerated. Further, a careful comparison between inside and outside counsel’s potential for gatekeeping reveals deeper insights into gatekeeping theory and its practical implementation. Or so I shall argue. In Part I, I briefly summarize classical gatekeeping theory,10 introduced into the legal literature in 1986 by Reinier Kraakman. Going beyond simple restatement, I identify three key elements of a gatekeeper enforcement regime and translate that theory into a set of tractable performance criteria (“Four Quadrants of Gatekeeping”) by which we can measure the gatekeeping potential of any professional. These criteria arise from two simple observations. First, in order to be successful, gatekeepers must not only be willing to disrupt misconduct but also be capable of doing so. Second, gatekeepers must not only be prepared to interdict misconduct but also monitor to detect such happenings in the first place. Intersecting these observations, we see that potential gatekeepers can be evaluated by their: (1) willingness to interdict, (2) willingness to monitor, (3) capacity to monitor, and (4) capacity to interdict. In Parts II through V, I employ these criteria to elicit a textured, institutional comparison between inside and outside lawyers as gatekeepers. In doing so, I make a methodological intervention by using social psychology11 and sociology, in addition to traditional rational choice theory, to analyze a more comprehensive set of factors that influence lawyer behavior in the face of client wrongdoing. In this way, this project is consistent with a call for behavioral realism,12 which has recently gained traction within legal analysis. In my conclusion, I explore how my new synthesis of gatekeeping theory and the insights gained from my analysis suggest pathways for reform and, more broadly, call for an inquiry into some of our long-held assumptions about gatekeeping. 10. By “classical” or “traditional” gatekeeping theory, I refer to a particular triad of articles authored by Kraakman and Ronald Gilson in the mid-1980s. These are: Kraakman, Gatekeepers, supra note 8; Reinier H. Kraakman, Corporate Liability Strategies and the Costs of Legal Controls, 93 YALE L.J. 857, 888-96 (1984) [hereinafter Kraakman, Corporate Liability]; and Ronald J. Gilson & Reinier H. Kraakman, The Mechanisms of Market Efficiency, 70 VA. L. REV. 549 (1984) [hereinafter Gilson & Kraakman, Market Efficiency]. Theoretical contributions and clarifications have been made by Ronald J. Gilson, The Devolution of the Legal Profession: A Demand Side Perspective, 49 MD. L. REV. 869 (1990) [hereinafter Gilson, Devolution]; COFFEE, GATEKEEPERS, supra note 9; Stephen Choi, Market Lessons for Gatekeepers, 92 NW. U. L. REV. 916 (1998); Assaf Hamdani, Gatekeeper Liability, 77 S. CAL. L. REV. 53 (2003); Frank Partnoy, Barbarians at the Gatekeepers?: A Proposal for a Modified Strict Liability Regime, 79 WASH. U. L.Q. 491 (2001). 11. This article and prior work are also heavily influenced by scholars in the field of social cognition. Social cognition is the intersection between the traditionally separate fields of social psychology and cognitive psychology. For an overview of social cognition, see SUSAN T. FISKE AND SHELLEY E. TAYLOR, SOCIAL COGNITION (McGraw-Hill, 2d ed. 1991). 12. See infra Part I.B.2. 2008] GATEKEEPERS INSIDE OUT 415 I. FRAMEWORK FOR ANALYSIS A. CLASSICAL GATEKEEPING THEORY Reinier Kraakman described gatekeeper liability as a genre of third party liability used by the government to “supplement efforts to deter primary wrongdoers directly by enlisting their associates and market contacts as de facto ‘cops on the beat.’”13 These de facto cops are in fact “gatekeepers,” originally defined as “private parties who are able to disrupt misconduct by withholding their cooperation from wrongdoers.”14 A review of the gatekeeping literature suggests that a gatekeeper enforcement regime must have at least three key elements: (1) a gatekeeper—someone “who can and will prevent misconduct reliably,”15 (2) a gate—some service which the wrongdoer needs to accomplish his goal,16 and (3) a law enforcement mechanism—an enforceable duty—that obligates private parties to take some action aimed at averting misconduct when detected.17 In the capital markets context, gatekeeping scholars have traditionally conceived the gatekeeper as an outside professional services firm which has a contractual relationship with the primary enforcement target (the client). The gate has traditionally been that firm’s specialized certification (e.g., legal opinion from a law firm, audit letter from an accounting firm) needed to consummate the client’s securities transactions. And the specific mechanism has traditionally been the gatekeeper’s professional duty to withhold services when it finds that it cannot vouch for the veracity of its client. Thus, by withholding the firm’s certification, the gatekeeper warns the market and shuts the gate, effectively foreclosing the issuer’s access to the capital markets.18 Under the original formulation, gatekeeper liability was “designed to enlist the support of outside participants [of] the firm when controlling managers commit offenses, that is, when the firm’s internal monitors have failed.”19 The logic 13. Kraakman, Gatekeepers, supra note 8, at 53. 14. Id. (defining “gatekeeper liability” as “liability imposed on private parties who are able to disrupt misconduct by withholding their cooperation from wrongdoers”); Gilson, Devolution, supra, note 10, at 883 (defining gatekeepers as “someone in a position to decline when his service will be misused”). 15. See Kraakman, Gatekeepers, supra note 8, at 61 (“[G]atekeepers who can and will prevent misconduct reliably, regardless of the preferences and market alternatives of wrongdoers.”). 16. Gilson, Devolution, supra note 10, at 883. 17. See Kraakman, Gatekeepers, supra note 8, at 57 (“[A]n enforceable duty—that allows private parties to avert misconduct when they detect it.”) (emphasis added). 18. COFFEE, GATEKEEPERS, supra note 9, at 2. Of course, if the client successfully shops for another certification, the fraudulent transaction may close and the market may not be warned. See infra Part V (discussing how gatekeepers may be circumvented by issuers). 19. Kraakman, Corporate Liability, supra note 10, at 890 (emphasis added). However, Kraakman acknowledged the possiblity of using intra-firm monitors as gatekeepers. See Kraakman, Gatekeepers, supra note 8, at 69 (observing gatekeeper regimes that enlist intrafirm monitors, e.g., those that police disclosure for pollution discharge permits and public issues of securities). 416 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 behind this preference for outsiders was that “in the usual case, [outsiders] likely have less to gain and more to lose from firm delicts than inside managers.”20 Thus, the ideal gatekeeper would be an “outsider with a career and assets beyond the firm,”21 such as auditors, investment bankers, securities analysts, securities attorneys, and, as more recently posited, credit rating agencies.22 But theory has crashed into some inconvenient truths. The vivid failure of a Big Five accounting firm, Arthur Andersen, to prevent the ongoing frauds of Enron and WorldCom, raised the question whether outsiders, in fact, make reliable gatekeepers. This is not to say that insiders had a better track record in these particular fiascos; however, it no longer seems quixotic to reexamine our basic assumptions about gatekeeping inside and out. B. TWO COMPLICATIONS Kraakman declared that a successful gatekeeping strategy requires gatekeepers “who can and will prevent misconduct reliably, regardless of the preferences and market alternatives of wrongdoers.”23 This simple sentence hides two complications worthy of close examination. 1. DISTINGUISHING “MONITORING” First, focus on the word “prevent.” In order to prevent or interdict24 managerial misconduct, it goes without saying that the gatekeeper must find out about it in the first place. In other words, prevention presupposes that the gatekeeper 20. Kraakman, Corporate Liability, supra note 10, at 891. 21. Id. 22. The one prominent exception to the view that outsiders are the only appropriate gatekeeping candidates is Gilson, Devolution, supra note 10, at 915 (arguing that, since the market power necessary for private gatekeeping has moved in-house, so, too, must the gatekeeping function). 23. Kraakman, Gatekeepers, supra note 8, at 61 (setting forth the third of four criteria used to evaluate a gatekeeper enforcement strategy). The other three criteria are: serious misconduct that practicable penalties cannot deter; missing or inadequate private gatekeeping incentives; and gatekeepers whom legal rules can induce to detect misconduct at reasonable cost. Id. Offering a slightly different formulation, Ronald Gilson notes that not only must there be a gatekeeper and a gate, but (1) the “gatekeeper must be able to detect with some precision when his service will be misused,” and (2) “supply and demand conditions in the market for the service that functions as a gate must support an enforcement regime: the gatekeeper must be willing to play that role and consumers of that service must be willing to accept it.” Gilson, Devolution, supra note 10, at 883. 24. While I adopt Kraakman’s verb, “interdict,” to describe what is expected of the effective gatekeeper, I intend to include any and all steps taken by the gatekeeper (including reporting information up-the-ladder to the board) for the purpose of averting or thwarting misconduct, short of the revelation of client confidences outside of the corporate client (which would implicate an alternative “whistleblower” enforcement regime). See Kraakman, Gatekeepers, supra note 8, at 56 (distinguishing the whistleblower enforcement strategy that would “compel third parties to disclose misconduct directly to potential victims or enforcement officials”). While the specific design of gatekeeping duties is beyond the scope of this article, one could imagine that a typology of duty design, based on potential impact, would be illuminating. 2008] GATEKEEPERS INSIDE OUT 417 monitor to detect potential malfeasance.25 Granted, monitoring and interdiction cannot always be sharply distinguished because they sit along a detection-to-response continuum.26 When a gatekeeper has some reasonable suspicion (but not quite probable cause) about some misbehavior, its further actions can be framed as either deeper monitoring or incipient interdiction.27 That said, in many contexts, the distinction will be plain and analytically useful. There are at least two reasons why. First, by distinguishing these two attributes of good gatekeepers, we can better sort among potential actors that can play that role.28 After all, certain actors may be quite ready and willing to interdict but have no ability to monitor, and vice versa. As I demonstrate below, even if outside law firms will be more willing to interdict, inside counsel have an overwhelming advantage in their ability to monitor. Second, this distinction makes us mindful of how solutions intended to incentivize the gatekeeper to interdict (e.g., harsh sanctions imposed on the gatekeeper who has “knowledge” of client wrongdoing but nonetheless fails to 25. See Kraakman, Gatekeepers, supra note 8, at 86 (suggesting that a gatekeeper should have “the capacity to interdict misconduct with some reliability and the ability to detect it at a relatively low cost”) (emphasis added). 26. Depending on the type of gatekeeper regime, the distinction between monitoring and interdiction may be more or less material. As Kraakman notes, in one genre of gatekeeper regimes known as “bouncer” regimes, for example, a tavernkeeper’s duty to exclude minors, the distinction between monitoring and interdiction may not be material. The tavernkeeper merely has a duty to check ID and to refuse to serve alcohol for those who fail to present the proper ID. Thus, the duty to monitor and interdict are highly focused, mechanical and fairly coincident. See Kraakman, Gatekeepers, supra note 8, at 79. “Bouncer” regimes are a subset of gatekeeper regimes where misconduct is easily disrupted by revoking access to a particular good or service. By contrast, in “chaperone” regimes, the gatekeeper can detect and disrupt misconduct in a more complex, longer-term (often contractual) relationship with enforcement targets. Thus, more complex monitoring duties are appropriate for chaperone regimes. See Kraakman, Gatekeepers, supra note 8, at 63. 27. To understand why monitoring and interdiction cannot always be sharply distinguished, consider the following three scenarios: (1) Will the gatekeeper—even in the absence of red flags—voluntarily assume an active monitoring role to keep apprised of ongoing facts or will she do the bare minimum and passively “see no evil, hear no evil”? (2) If red flags appear, will the gatekeeper probe for further facts or will she look the other way? (3) Finally, once the gatekeeper knows of actual or anticipated misconduct, will she interdict—either by saying “no” to the wrongdoer, reporting misconduct up-the-ladder, or withholding the critical support from a recalcitrant client? The first question above clearly goes to the gatekeeper’s willingness to monitor. The last question clearly relates to her willingness to interdict. The middle question lies somewhere in between; after red flags first appear (and the gatekeeper’s “spider sense” starts to tingle) but before the gatekeeper definitively knows a law has been broken. In this grey zone, a lawyer’s further actions can be framed as either deeper monitoring or incipient interdiction. And in situations characterized by complexity and ambiguity, self-serving rationalizations of unethical conduct are likely to take root. See infra note 46. 28. See, e.g., Arthur B. Laby, Differentiating Gatekeepers, 1 BROOK. J. CORP. FIN. & COM. L. 119, 120 (2006) (criticizing the rational actor model for failing to distinguish among gatekeepers, “who are likely to respond differently to incentives,” and proposing to evaluate “dependent” and “independent” gatekeepers from a behavioral viewpoint). 418 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 interdict) may perversely disincentivize monitoring.29 A gatekeeper who potentially faces the hard decision of either losing one’s job (being fired for resisting the wrongdoer) or losing one’s professional license (for failing to resist the wrongdoer) will prefer to cloister herself within her office and hear no evil. 2. MODELING “WILLINGNESS” Second, notice that Kraakman distinguished between “can” and “will.” Although Kraakman did not make much of that distinction, I contend that willingness and capacity are also independent attributes worthy of separate analysis. This is so because certain actors may be very willing but quite unable, and vice versa. As I argue below, outside law firms may arguably be more willing to interdict misconduct but they may be deeply unable to do so, because they are never quite aware of when it takes place. Further, I want to make a methodological intervention in our understanding of “willingness.” Gatekeeping theory has largely adopted “rational choice theory” (“RCT”) or “expected utility theory” as its model of human behavior.30 RCT presents an idealized model of decisionmaking built on the foundational assumption that man is a rational maximizer of his expected utilities.31 As a result, gatekeeping theory has focused almost exclusively on the expected gains of a gatekeeper’s corruption or acquiescence (e.g., continued fees from wrongdoer) weighed against its expected costs (probability of detection multiplied by sanctions levied, e.g., reputational harm, civil liability, incarceration). If we are interested in measuring a lawyer’s willingness to monitor, to interdict—indeed to do anything, including eating a ham sandwich—we should simply measure the incentives she faces, in costs and benefits. But in the 1990s, behavioral-law-and-economics scholars began challenging RCT’s foundational assumption that people make outcome-maximizing decisions.32 Drawing on the “heuristics and biases” literature from cognitive psychology and behavioral economics, they called for a more empirically accurate model of human-decisionmaking.33 I am fully on board with the call for 29. See infra Part III. 30. See Russell B. Korobkin & Thomas S. Ulen, Law and Behavioral Science: Removing the Rationality Assumption from Law and Economics, 88 CAL. L. REV. 1051, 1061-1066 (2000) (classifying and describing the different versions of rational choice theory, including the “expected utility” version). Gatekeeping theory has thus far relied heavily on RCT, and its descendants–finance theory and institutional economics. Herbert Simon observes that institutional economics “retains the centrality of markets and exchanges. All phenomena are to be explained by translating them into . . . market transactions based upon negotiated contracts.” H. A. SIMON, MODELS OF MY LIFE, 26 (1991). 31. See, e.g., RICHARD A. POSNER, ECONOMIC ANALYSIS OF LAW 2-6 (1st ed. 1973). 32. For a survey, see Donald C. Langevoort, Behavioral Theories of Judgment and Decision Making in Legal Scholarship: A Literature Review, 51 VAND. L. REV. 1499 (1998). 33. See, e.g., BEHAVIORAL LAW AND ECONOMICS (Cass Sunstein ed., 2000); Korobkin & Ulen, supra note 30 (critiquing rational choice theory and advocating “law and behavioral science”). 2008] GATEKEEPERS INSIDE OUT 419 greater empirical accuracy in modeling decisionmaking, which in this context is ethical decisionmaking.34 But behavioral law and economics, notwithstanding protestations to the contrary, privileges “economics” and undervalues scientific findings from other fields. In practice, behavioral law and economics seems to identify a laundry list of heuristics and biases that deviate from perfect rationality, which still remains the default assumption. Accordingly, I am more interested in what has recently been called a “behavioral realist” approach,35 which calls forth on the law to adopt the most accurate model of human decisionmaking and behavior provided by the mind sciences, or if it does not, to explain in transparent terms why not.36 This approach does not privilege one branch of scientific knowledge, such as economics, over another in word or deed. It can, for instance, easily encompass important insights from modern social psychology: that the situation is a better predictor of human behavior than an individual’s personal characteristics or views.37 It is not shy to extend beyond heuristics and biases that compromise rationality but to consider less apparent yet “pervasive, fundamental and arational cognitive processes”38 that RCT, even with modifications, declines to consider. In the context of lawyer gatekeeping, these powerful but subtler and, for the most part, arational cognitive processes include alignment, obedience, and conformity pressures that are automatically activated by specific situational factors in the lawyer’s environment, such as the organizational setting and the lawyer’s position in the organizational hierarchy.39 In “role” terms that I have 34. A good example of a behavioral-law-and-economics approach to ethical decisionmaking by lawyers is Richard W. Painter, Lawyers’Rules, Auditors’Rules and the Psychology of Concealment, 84 MINN. L. REV. 1399 (2000) (using prospect theory to explain lawyers’ unethical actions to conceal fraud). See also Jeffrey J. Rachlinski, Gains, Losses and the Psychology of Litigation, 70 S. CAL. L. REV. 113 (1996) for one of the earliest applications of prospect theory to lawyers’ behavior (in the litigation context). 35. Jerry Kang & Mahzarin R. Banaji, Fair Measures: A Behavioral Realist Revision of “Affirmative Action,” 94 CAL. L. REV. 1063, 1064-65 (2006) (describing “behavioral realism”). See Symposium on Behavioral Realism, 94 CAL. L. REV. 45 et seq. (2006). Other scholars have similarly introduced different nomenclatures. See, e.g., John Hanson & David Yosifon, The Situation: An Introduction to the Situational Character, Critical Realism, Power Economics, and Deep Capture, 152 U. PA. L. REV. 129 (2003) (introducing “critical realism”); Korobkin & Ulen, supra note 30 (advocating “law and behavioral science”). 36. See Kristin A. Lane, Jerry Kang, & Mahzarin R. Banaji, Implicit Social Cognition and the Law, 3 ANN. REV. L. & SOC. SCI. 19.1, 19.14-19.15 (2007); Kang & Banaji, supra note 35, at 1065 (noting that behavioral realism is not a set of “first-order normative commitments or policy preferences” but a second-order commitment against hypocrisy and self-deception: “If science reveals that the law is failing to do so because it is predicated on erroneous models of human behavior, then the law must transparently account for the gap instead of ignoring its existence”). 37. See, e.g., Hanson & Yosifon, supra note 35 (introducing “situationism”); Sung Hui Kim, The Banality of Fraud: Re-Situating the Inside Counsel as Gatekeeper, 74 FORDHAM L. REV. 983, 992-97 (2005). By “situation,” I mean features in the person’s environment. For situational factors, see infra note 39. 38. Jerry Kang, Trojan Horses of Race, 118 HARV. L. REV. 1489, 1494 at n.21 (2005) (advocating a “behavioral realist” approach to law, drawing on the traditions of legal realism and behavioral science). 39. Situational factors include, among others, the “physical setting, roles, rules, uniforms, symbols, and group consensus.” See Kim, supra note 37, at 995. 420 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 described elsewhere and below,40 we should be mindful of the roles that outside and inside counsel play as mere employee, team player, and faithful agent. Further, numerous psychological and economics experiments reveal how we automatically engage in biased and motivated reasoning to self-serving ends.41 Thus when the situation pits self-interest against one’s moral responsibilities, we often fail to perceive that as a moment of ethical choice and instead rationalize our self-interested behavior. That individuals often fail to classify situations as “ethical” is consistent with sociological findings that law firm partners deny the moral dimensions of their work,42 inside counsel report experiencing little ethical conflict in their jobs,43 and experienced business managers do not frame situations in moral terms.44 In sum, when thinking about “willingness,” we should not imagine a gatekeeper experiencing an overt ethical dilemma, pondering the decision to interdict. Instead, a more behaviorally realistic model understands the prevailing psychological forces as akin to gravitational pulls–pervasive, powerful, and yet, fundamentally invisible and unnoticed. This insight has two implications. First, explicit moral commandments to be better gatekeepers without a concurrent shift in the ethical norms of lawyering45 will have limited effect. Second, given our robust ability to unconsciously and selectively process information in a selfinterested manner, especially in situations characterized by complexity and 40. See infra Part II.B.; Kim, supra note 37, at 1001-26. 41. Social psychologists refer to the mechanism by which self-interest (through the “self-serving bias” or “egocentric bias”) is automatically processed by the human brain as “motivated reasoning,” the selective information processing that has been identified and supported by the empirical literature. For a discussion on the self-serving bias, motivated reasoning and the dual processing model of cognition, see Kim, supra note 37, at 1026-34. See also Ziva Kunda, The Case for Motivated Reasoning, 108 PSYCHOL. BULL. 480 (1990) for a review of evidence that supports the existence of motivated reasoning. 42. See, e.g., Mark Suchman, Working Without A Net: The Sociology of Legal Ethics in Corporate Litigation, 67 FORDHAM L. REV. 837, 844-45 (1998) (noting that large firm associates “readily acknowledged the moral dimensions of their work, but often collapsed these into pragmatic concerns” and that large firm “partners . . . tended to deny the moral dimensions of their work entirely, and to reduce most issues to either ethical rules or pragmatic strategies”); Kimberly Kirkland, Ethics in Large Law Firms: The Principle of Pragmatism, 632 U. MEM. L. REV. 632, 726 (concluding that “Partners do not see moral questions . . . because partners do not measure their conduct against internal or fixed principles; their habit of mind is to glean expectations, to read situations, to collapse the distinction between appearance and substance, and to equate etiquette with ethics”). 43. Hugh P. Gunz & Sally P. Gunz, The Lawyer’s Response to Organizational Professional Conflict: An Empirical Study of the Ethical Decision Making of In-House Counsel, 39 AM. BUS. L.J. 241, 250, 264 (2002) (noting that the degree of “organizational-professional conflict” in studies of inside counsel is low). In the Gunz study, “organizational-professional conflict” was related to the frequency with which counsel reported encountering ethical dilemmas in their professional lives. Id. at 275. 44. Milton C. Regan, Jr., Moral Intuitions and Organizational Culture, 51 ST. LOUIS U. L.J. 941, 965 (2007) (summarizing research that shows how managers employ non-moral prototypes as their default perceptual framework); ROBERT JACKALL, MORAL MAZES: THE WORLD OF CORPORATE MANAGERS 6 (1988). 45. See Part II.B.3 (“Faithful Agent”) for a discussion of prevailing lawyers’ role ideologies that may contribute to pressures to align one’s views to those of your de facto clients. 2008] GATEKEEPERS INSIDE OUT 421 ambiguity,46 clear and specific prescriptions will be more effective in encouraging gatekeeping than vague standards.47 C. FOUR QUADRANTS OF GATEKEEPING To repeat, the first complication distinguished monitoring from interdiction. The second complication emphasized the significance of distinguishing willingness from capacity, especially if we allow ourselves to incorporate not only economics but social psychology and sociology in our modeling of “willingness.” By intersecting these insights, we see that potential gatekeepers can be understood and evaluated by their: (1) willingness to interdict, (2) willingness to monitor, (3) capacity to monitor, and (4) capacity to interdict. Stated another way, these four factors are necessary components to effective gatekeeping. I have numbered the Quadrants in clockwise fashion to correspond to the order that I address them in this article. Interdict Monitor Willingness I. Willingness to Interdict II. Willingness to Monitor Capacity IV. Capacity to Interdict III. Capacity to Monitor Quadrant I (Willingness to Interdict). Suppose that a gatekeeper discovers facts suggesting illegality. Will she then take the next action to interdict misconduct? Using Kraakman’s cops-on-the-beat metaphor, once the cop actually sees a crime occurring, will she do something to stop it? If she’s two months away from retirement with a pension and a violent melee has just erupted, will she intervene without backup, or will she look the other way? If she sees a health code violation at the doughnut shop that treats her daily to hot coffee and crullers, will she be inclined to issue a warning or notify the local health agency?48 Quadrant II (Willingness to Monitor). A gatekeeper closes the gate when she knows that there has been misconduct. But how does the gatekeeper come to know? In part, the gatekeeper must be willing to expend resources to be vigilant. When red flags appear, the gatekeeper must be willing to probe further those 46. See Kim, supra note 37, at 1029-31 (noting how complex and ambiguous environments serve as a fertile breeding ground for motivated reasoning) and 1048-52 (applying such insights to criticize regulations enacted under SOX § 307 and the 2003 amendments to the ABA’s Model Rules of Professional Conduct). 47. See, e.g., Painter, supra note 34, at 1406-10 (describing how the American Bar Association has persistently declined—for many years—to revise and clarify its Model Rule 1.13 which vaguely commanded that the lawyer must act in the “best interests of the organizational client” when confronted with crime or fraud committed by an organizational agent). 48. The two aforementioned hypothetical scenarios (imminent retirement and doughnut shop freebies) raise “duty of care” and “duty of loyalty” issues, respectively. The first scenario raises the question of whether the cop’s incentive structure encourages shirking while the second scenario raises the question of whether the cop has conflicted loyalties. 422 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 suspicious circumstances. To continue the policing metaphor, it’s not enough that the cop has a gun strapped to her hip and is willing to fire it when she sees a mugging. She must actually do some patrolling–walk around the four square blocks that constitute her beat. She must talk to the people who live there, follow up on suspicious shadows down long alleys–even when it’s cold outside and a warm doughnut shop beckons. Quadrant III (Capacity to Monitor). Resolute willingness may in the end matter little if fundamental capacity is lacking. For example, earnest vigilance will fail to uncover misconduct if the gatekeeper has no access to the facts. Imagine not a cop on the beat, but a police officer using videocameras from a remote surveillance site. No matter how sincere and duty-bound, the technology constrains what that officer can discover at a distance. If there is no audio, gunshots cannot be heard. If the camera has blindspots, corners will remain in the shadows. If the magnification and resolution are poor, no matter how much the officer strains, she simply won’t be able to see. Quadrant IV (Capacity to Interdict). Finally, even assuming that a gatekeeper is capable and willing to monitor and perfectly eager to close the gate, she is only as effective as her capacity to interdict. One reason why this capacity may be lacking is if the wrongdoing client has access to multiple gates. If one gate shuts, ten more may swing wide open. Suppose the cop on the beat also moonlights as private security and is hired to help transport some item in the trunk across state lines—“no questions asked.” A guard who asks too many questions (what am I transporting? to whom? why?) won’t be hired again because another contractor who just happens to be less curious will be waiting in the wings. The capacity to interdict ultimately turns on the ease with which the wrongdoer can circumvent any particular gatekeeper. By unpacking two complications and cross-tabulating them, I have remapped Kraakman’s gatekeeping theory and identified four criteria by which potential gatekeepers may be evaluated. In analyzing “willingness,” my approach is to be behaviorally realistic and not constrained by stylized models that emphasize explicit, overt, expected utility calculations. By systematically running both inside and outside counsel through this analytic mill, we come to unexpected conclusions about the relative strengths and weaknesses of inside and outside counsel for the gatekeeping role. II. WILLINGNESS TO INTERDICT The analytic mill starts here with Quadrant I. In this Part, my principal objective is to revise prior understandings of “willingness” by being more behaviorally realistic. Although behavioral realism is critical of the stylized simplifications of rational choice theory (“RCT”), the insights from RCT and behavioral realism are often mutually illuminating, not mutually exclusive. In other words, we learn a great deal from examining both individual economic 2008] GATEKEEPERS INSIDE OUT 423 incentives—even if we accept that entirely different cognitive mechanisms (than what RCT posits) are at work—and the pervasive situational forces unknowingly experienced.49 In this vein, this Part identifies five important factors that go to the willingness of a lawyer to interdict. Some of these factors are commonly attributed to RCT (and thus I will refer to them as “rational choice factors”); others sound more in behavioral realism, where I draw heavily from social psychology and sociology (“behavioral realist factors”). I apply these five factors to elicit comparative strengths and weaknesses of inside and outside counsel. A. RATIONAL CHOICE FACTORS Gatekeeping theory operates on the assumption that for the gatekeeper to act, the expected benefits should outweigh the expected costs. The most salient benefit of closing the gate is avoiding the costs associated with acquiescing in misconduct. Thus, one can safely infer that a gatekeeper is likely to interdict client wrongdoing if two RCT conditions are satisfied: the gatekeeper (1) stands to lose much if its acquiescence in misbehavior is detected (the “high cost of complicity”)50 and (2) stands to lose little by resisting the wrongdoer’s bribes (the “low cost of resistance”).51 Put another way, a gatekeeper will interdict if (1) the cost of complicity is high and (2) the cost of resistance is low. 1. HIGH COST OF COMPLICITY a. Loss of Reputational Capital Gatekeepers will interdict misconduct if they fear that failing to do so might result in an irreparable loss to their longstanding reputations. In the securities context, market gatekeepers such as investment banking and accounting firms act as reputational intermediaries whom issuers pay to vouch for their disclosures to the market. By certifying the issuer’s public statements to the market, these intermediaries attest that they have evaluated the issuer and are prepared to stake their reputation on the accuracy of the issuer’s statements.52 They effectively 49. See supra Part I.B.2 (describing my behavioral realist revision of gatekeeping theory’s modeling of a gatekeeper’s willingness to interdict and monitor and how my revision, which fully considers arational cognitive processes that are unleashed by situational forces, differs from the account posited by RCT). 50. See, e.g., Kraakman, Gatekeepers, supra note 8, at 70 (noting that professionals who have made large investments in their licenses and reputations make attractive gatekeepers because “they stand to lose too much if their corruption is detected”). 51. Id. at 71 (noting that “diversified gatekeepers, who have proportionately less at stake in relationships with particular clients or customers, are less likely to receive threats (or corrupt offers) that are large enough to offset their expected costs of corruption”); Kraakman, Corporate Liability, supra note 10, at 891 (noting that outsiders have different incentives from those of inside managers: “they are likely to have less to gain and more to lose from firm delicts than inside managers”). 52. Gilson & Kraakman, Market Efficiency, supra note 10, at 620. 424 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 pledge their reputation to the issuer, which in turn increases investor confidence.53 Because reputation is hard to gain but easy to lose, these intermediaries are viewed as ideal market gatekeepers because they already “face powerful private incentives to prevent misconduct.”54 Of course, this reputation story presupposes that market gatekeepers have already accumulated enough reputational capital to deter them from squandering it away by vouching for a wayward issuer.55 The only way to accrue such reputation is by acting as a repeat player in many securities transactions for many issuers over many years.56 Large and long-lived accounting and investment banking firms are thus seen as the gatekeepers of choice. Likewise, elite law firms (not individual attorneys) that have accrued reputations for trustworthiness and independence can serve this function.57 In sharp contrast, inside counsel are viewed as lacking the threshold amount of reputational capital necessary to discipline their behavior.58 As Coffee states: [I]n-house counsel is seldom a reputational intermediary (as law and accounting firms that serve multiple clients are) because the in-house counsel cannot easily develop reputational capital that is personal and independent from the corporate client.59 This is a persuasive, logical story. But it is overstated. First, the empirical data raise questions about the importance of reputational capital to effective gatekeeping.60 To pick a salient counterexample, why would 53. Id. 54. Kraakman, Gatekeepers, supra note 8, at 62. 55. Jonathan Macey & Hillary A. Sale, Observations on the Role of Commodification, Independence, and Governance in the Accounting Industry, 48 VILL. L. REV. 1167, 1173 (2003). 56. COFFEE, GATEKEEPERS, supra note 9, at 2. Coffee goes on to note: Because the gatekeeper is inherently an agent of its principal, its expected fee or commission is likely to be far less than the gain that the principal itself expects to make from the transaction. As a result, because the gatekeeper/agent expects less profit than its principal does, it can be more easily deterred than its principal. See id. at 5. 57. See Karl S. Okamoto, Reputation and the Value of Lawyers, 74 OR. L. REV. 15, 19 (1995); Ronald J. Gilson & Robert H. Mnookin, Sharing Among the Human Capitalists: An Economic Inquiry into the Corporate Law Firm and How Partners Split Profits, 37 STAN. L. REV. 313, 368 (1985) (“At the core of the firm’s ability to pledge its reputation on behalf of its client is a perception by other parties that the firm is ‘independent’—that the firm would not put its reputation on the line were the client’s statements not true.”). 58. Okamoto, supra note 57, at 19, 21, 33 (noting inside counsel’s relative inability to serve as a reputational intermediary for their clients). 59. COFFEE, GATEKEEPERS, supra note 9, at 195. 60. To be sure, classical gatekeeping theory maintained a more nuanced understanding of reputation than I have thus far portrayed (for the sake of clarity and brevity). Kraakman has observed that outsiders often lack the ability to distinguish among firms, so they apply average reputations from a broad market segment. See Kraakman, Gatekeepers, supra note 8, at 97-98. 2008] GATEKEEPERS INSIDE OUT 425 Arthur Andersen have squandered its legendary reputation61 by signing off on Enron’s shady accounting?62 Perhaps Anderson was an outlier, but an empirical study covering 1,000 large public firms from 1997 to 2001 yielded no evidence that the quality of public company audits performed by Andersen was any worse than the other large accounting firms.63 Also, according to surveys of companies who hire outside counsel, their reputational value is not declared to be important for most transactions.64 Second, reputational information markets may not work as perfectly as imagined. Public reports of ethical or legal breaches may not even link the malfeasant individual to her firm,65 or may not be significantly correlated with the seriousness of the infraction committed or the degree of harm inflicted. For example, an exhaustive empirical study of securities class action settlements from 1989 to 1994 observed that although underwriters settled twice as many securities class action lawsuits as accountants and also paid more per case, there was little to no press coverage on underwriters, in contrast to the voluminous publicity devoted to accountants.66 Third, the information that does flow out may be artfully managed by sophisticated public relations campaigns that introduce just enough ambiguity to stanch serious reputational losses.67 Like public companies, law firms can settle cases with regulators or plaintiffs while denying guilt. They can take advantage of plausible deniability by characterizing individual wrongdoers as rogue attorneys punished for their ultra vires actions while the firm publicly renews its 61. See Ken Brown & Ianthe Jeanne Dugan, Sad Account: Andersen’s Fall From Grace Is a Tale of Greed and Miscues—Pushed to Boost Revenue, Auditors Acted as Sellers, Warred With Consultants—‘Three Pebbles and a Boulder’, WALL ST. J., June 7, 2002, at A1 (citing an Andersen Heritage Center display devoted to press clippings memorializing former firm leader Leonard Spacek’s accusing Bethlehem Steel of overstating profits in 1964 by more than 60% and lambasting the SEC for failing to crack down on fraudulent accounting). 62. See Robert A. Prentice, The Inevitability of a Strong SEC, 91 CORNELL L. REV. 775, 780-799 (2006) (citing evidence that reputation insufficiently constrains gatekeeper behavior); Hillary A. Sale, Gatekeepers, Disclosure and Issuer Choice, 81 WASH. U. L.Q. 403, 407 (2003) (critiquing reliance on reputational constraints). 63. See Theodore Eisenberg & Jonathan R. Macey, Was Arthur Andersen Different? An Empirical Examination of Major Accounting Firm Audits of Large Clients, 1 J. EMPIRICAL LEGAL STUD. 263, 264-65 (2004) (noting that its analysis “yields no evidence that accounting profession problems that lead to [financial] restatements were unique to Andersen” and suggesting that “accounting profession problems are industrywide and not linked to any particular firm”). 64. Steven L. Schwarcz, Explaining the Value of Transactional Lawyering, 12 STAN. J.L. BUS. & FIN. 486, 502-03 (2007) [hereinafter Schwarcz, Transactional Lawyering] (finding only weak support for the proposition that transactional lawyers add value by acting as reputational intermediaries). 65. Ted Schneyer, Professional Discipline for Law Firms?, 77 CORNELL L. REV. 1, 34 (1991) [hereinafter Schneyer, Professional Discipline]. 66. Steven P. Marino & Renee D. Marino, An Empirical Study of Recent Securities Class Action Settlements Involving Accountants, Attorneys or Underwriters, 22 SEC. REG. L. J. 115, 174 (1994). 67. Donald C. Langevoort, Where Were the Lawyers? A Behavioral Inquiry into Lawyers’ Responsiblity for Clients’ Fraud, 46 VAND. L. REV. 75, 112 (1993) [hereinafter Langevoort, Where Were the Lawyers?]. 426 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 commitment to ethics.68 Finally, due to intense competition, law firms may be more worried about cultivating reputations for client devotion than client independence.69 Independence can mean, when one’s ethical duties call, having to deviate from the clients’ express requests, something that high paying clients, who feel entitled to every possible advantage, do not especially value.70 Of course, none of this is to say that inside lawyers have comparatively more reputation than large outside law firms to lose: they do not. My only point is to caution against an overeager acceptance of the reputation story. b. Private Civil Liability In addition to loss of valuable reputation, incentives to interdict can arise from the threat of private civil liability.71 If outside law firms were held liable for facilitating securities fraud, surely that would incentivize them to interdict by raising the costs of complicity. Interestingly, recent legislative and judicial developments have dramatically decreased the likelihood of such liability.72 First, in 1994, the U.S. Supreme Court decided in Central Bank of Denver v. First Interstate Bank of Denver against private liability for secondary actors— those who aided and abetted a primary violation under the antifraud provisions of the Securities Exchange Act of 1934.73 This decision greatly inhibited private plaintiffs from suing law and accounting firms, who otherwise would be vicariously liable for their agents’ facilitation of a primary violation.74 Because of this decision, the “legal profession remained far more insulated than the accounting profession, which still faced primary liability for its certification of 68. Donald C. Langevoort, Monitoring: The Behavioral Economics of Corporate Compliance with Law, 2002 COLUM. BUS. L. REV. 71, 102 (discussing companies’ strategies); Schneyer, Professional Discipline, supra note 65, at 34. 69. Milton C. Regan, Jr., Professional Reputation: Looking for the Good Lawyer, 39 S. TEX. L. REV. 549, 558-60 (1998) [hereinafter Regan, Professional Reputation]. 70. Id. at 560. As noted by Milton Regan, “With increased competition, potentially mobile clients correctly perceive that they have leverage to press lawyers not just to do their bidding, but to do it exactly as they would have it done and without regard for the legal or ethical niceties that typically give pause to members of the guild.” Id. at 561. 71. Of course, all threats of legal liability have attendant reputational consequences, making it impossible to segregate completely the “loss of reputational capital” from the threat of “private civil liability.” 72. For a comprehensive list of developments that have reduced the legal threat to gatekeepers, see COFFEE, GATEKEEPERS, supra note 9, at 60-61. 73. Central Bank of Denver, N.A. v. First Interstate Bank of Denver, N.A., 511 U.S. 164, 191 (1994) (holding “that a private plaintiff may not maintain an aiding and abetting suit under § 10(b)” of the Securities Exchange Act of 1934). 74. To be clear, Central Bank of Denver did not abolish the vicarious liability of firms for their agents’ violations. See Robert A. Prentice, Conceiving the Inconceivable and Judicially Implementing the Preposterous: The Premature Demise of Respondeat Superior Liability Under Section 10(b), 58 OHIO ST. L.J. 1325 (1997) (arguing in favor of maintaining respondeat superior liability for Section 10(b) actions). 2008] GATEKEEPERS INSIDE OUT 427 the issuer’s financial statements.”75 Second, a year later, the Private Securities Litigation Reform Act of 1995 (“PSLRA”)76 enacted a number of measures that collectively eroded plaintiffs’ ability to pursue claims against law firms.77 These developments effectively capped a general trend toward reducing the liability of law and accounting firms.78 According to an SEC study completed in 1997,79 out of a total of 105 securities class actions filed in 1996, not a single lawsuit named a law firm as defendant.80 In short, if civil liability for the large law firm was ever a significant threat, it certainly is not now. Moreover, recent judicial attempts to expand the primary 75. COFFEE, GATEKEEPERS, supra note 9, at 216. Auditors were still potentially liable as primary participants based on their public certifications. 76. Private Securities Litigation Reform Act of 1995, Pub. L. No. 104-67, 109 Stat. 737, 758 codified as amended at 15 U.S.C. § 78u-4 [hereinafter PSLRA]. 77. Plaintiffs-side securities litigators blame PSLRA’s adoption of a particular judgment reduction method (for Securities Exchange Act actions) as severely hampering plaintiffs’ ability to maintain actions against law firms. See PSLRA, supra note 76, § 21D(g)(7)(B), 15 U.S.C. § 78u-4(g)(7)(B). The judgment reduction provision complicates settling multi-party cases and can lead to plaintiffs’ opting to dismiss a co-defendant (such as a law firm) rather than risk the verdict against the non-settling defendants being reduced by the amount that corresponds to the percentage of responsibility of a settling co-defendant and thus potentially leading to a denial of full recovery for the plaintiffs. If the law firm is not named as co-defendant or is otherwise dismissed, the final judgment will not be reduced by the law firm’s percentage of fault (since the law firm will not be deemed a settling co-defendant), in which case the non-settling co-defendants remain liable for the entire verdict. Correspondence from William S. Lerach of Lerach Coughlin Stoia Geller Rudman & Robbins LLP to author dated Aug. 8, 2007 (on file with author) [hereinafter Lerach Correspondence]. See also Marc I. Steinberg & Christopher D. Olive, Contribution and Proportionate Liability under the Federal Securities Laws in Multidefendant Securities Litigation After the Private Securities Litigation Reform Act of 1995, 50 SMU L. REV. 337, 378 (1996) (noting that, under PSLRA’s judgment reduction method, the “risk of an inadequate settlement falls on the plaintiffs, resulting in the potential denial of full compensation”). For other reasons why secondary actors (law and accounting firms) enjoyed a reduced threat of legal liability in the 1990s, see COFFEE, GATEKEEPERS, supra note 9, at 60-61 (citing PSLRA’s replacement of joint-andseveral liability with proportionate liability, heightened pleading standards, restriction of the sweep of the RICO statute, adoption of safe harbor for forward-looking information as having a hindering effect on private litigation). However, it seems that PSLRA’s proportionate liability provisions in particular had much less an impact on law firms than on other secondary actors. See OFFICE OF GEN. COUNSEL, U.S. SEC. & EXCH. COMM’N, REPORT TO THE PRESIDENT AND THE CONGRESS ON THE FIRST YEAR OF PRACTICE UNDER THE PRIVATE SECURITIES LITIGATION REFORM ACT OF 1995 (1997), § IV. D (“Effect of the Act on Secondary Defendants”), available at http://www.sec.gov/news/studies/lreform.txt [hereinafter PRACTICE UNDER THE PSLRA]; Marino & Marino, supra note 66, at 163 (concluding that law firms do not act as deep pockets when an issuer is bankrupt and were never much hurt by the joint-and-several liability aspects of securities laws in the first place). 78. Even prior to Central Bank of Denver and PSLRA, law firms may have already been facing a declining threat of legal liability. See PRACTICE UNDER THE PSLRA, supra note 77, at § IV.D (reporting results of the National Economic Research Associates study that from 1991 to June 1996, law firms were defendants in only seven cases and reporting that plaintiffs’ bar attributed part of the decline in lawsuits to their inability to obtain timely discovery—that may reveal misconduct by secondary actors—within the one-to-three year statute of limitations imposed by Lampf, Pleva, Lipkind, Prupis & Petigrow v. Gilbertson, 501 U.S. 351, 359-61 (1991)). 79. See PRACTICE UNDER THE PSLRA, supra note 77, at § IV.D. 80. COFFEE, GATEKEEPERS, supra note 9, at 61. See PRACTICE UNDER THE PSLRA, supra note 77, at § IV.D (“Our review of complaints in the 105 class actions filed under the Act reveals that accounting firms have been named in six cases, corporate counsel in no cases, and underwriters in 19 cases.”). 428 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 liability of secondary actors, such as law and accounting firms,81 have been rebuffed by the U.S. Supreme Court.82 It is thus unlikely that the central course of Central Bank of Denver and PSLRA will be changed. By contrast, what can be said about potential civil liability of inside counsel? This question can be addressed on two levels—entity and individual liability. The inside counsel is an employee of the company. To the extent that securities class actions deter public companies from engaging in securities fraud,83 it follows that the employees through which the company “acts” are also deterred from securities fraud. Of course, this trickle down theory of deterrence, from entity to complying employee, has been hotly disputed.84 Consider just the single complication of moral hazard induced by insurance,85 which covers almost all of the costs of corporate and individual liability (including defense and settlement costs) arising out of securities lawsuits.86 There is, however, one area under the Securities Act of 1933 (the “Securities Act”) where the in-house general counsel should be directly incentivized because of potential individual liability. If the general counsel signs her company’s registration statement in her capacity as a senior executive officer of the company, she is personally liable for misstatements under Section 11 of the 81. In the past several years, lower courts have adopted various theories to allow private lawsuits against secondary actors including law firms and lawyers. See, e.g., Steve L. Schwarcz, Financial Information Failure and Lawyer Responsibility, 31 J. CORP. L. 1097, 1099-100 (2006); John C. Coffee, Jr., Gatekeeper Failure and Reform: The Challenge of Fashioning Relevant Reforms, 84 B.U. L. REV. 301, 338, 340 (2004) (observing a ‘judicial shift . . . toward imposing greater liability on gatekeepers‘ in financial frauds and showing empirical data which indicate that ‘the risk for gatekeepers is real and growing‘). 82. See Stoneridge Inv. Partners, LLC v. Scientific-Atlanta, Inc. 128 S. Ct. 761 (2008) (reaffirming that the implied private right of action for securities fraud under Section 10(b) of the Securities Exchange Act of 1934 does not extend to aiders and abettors and rejecting plaintiffs’ “scheme liability” theory). 83. John C. Coffee, Reforming the Securities Class Action: An Essay on Deterrence and Its Implementation, 106 COLUM. L. REV. 1534, 1543, 1548 (2006) [hereinafter Coffee, Securities Class Action]. 84. See, e.g., Jennifer H. Arlen & William J. Carney, Vicarious Liability for Fraud on Securities Markets: Theory and Evidence, 1992 U. ILL. L. REV. 691, 704-17 (arguing against entity liability—and in favor of agent liability—because firms are not superior to courts and private plaintiffs in identifying and sanctioning wrongdoers ex post, firms do not have superior incentives to screen ex ante for honest agents, and firms do not have superior incentives to monitor current employees); Jennifer Arlen, The Potentially Perverse Effects of Corporate Criminal Liability, 23 J. LEGAL STUD. 833 (1994) (arguing that strict vicarious criminal liability may increase corporate crime by incentivizing companies to reduce corporate enforcement expenditures—that would otherwise facilitate crime detection—in an effort to decrease the company’s expected criminal liability); Deborah A. DeMott, Organizational Incentives to Care About the Law, 60 LAW & CONTEMP. PROBS. 39 (1997) (defending vicarious liability as a reflection of the principal’s right and presumed ability to define the agent’s incentive structure). 85. Tom Baker & Sean J. Griffith, The Missing Monitor in Corporate Governance: The Directors’ & Officers’ Liability Insurer, 95 GEO. L.J. 1795, 1798, 1808 (2007) (reporting that insurers generally do not offer loss prevention services and do not otherwise monitor the corporate governance of their corporate insureds). According to the authors, executive and risk manager agency costs and the fact that insurance insulates companies and their managers from the financial impact of liability together suggest that these insurance policies create a moral hazard for corporate officers. Id. at 31, 47-51. 86. Id. at 1804-05. Directors’ and officers’ liability insurance (D&O insurance) protects the assets of the corporation as well as its directors and officers. See id. at 1797. 2008] GATEKEEPERS INSIDE OUT 429 Securities Act,87 subject to a “due diligence” defense.88 By contrast, outside law firms are never subject to Section 11 liability.89 In sum, if we expect private civil liability to raise the costs of complicity, that expectation is not met by current law. And as between outside law firms and inside counsel, if there is any marginal difference in potential liability, the inside counsel may bear a slightly higher cost. c. SEC Enforcement Even if there is little chance of private civil liability, what about the threat of SEC enforcement? Even in the aftermath of Central Bank of Denver, the SEC retained its ability to sanction secondary participants for knowingly aiding and abetting a securities fraud.90 The SEC can also sanction lawyers and law firms as primary violators or for “causing” a securities fraud under § 15(c)(4) of the Securities Exchange Act of 193491 and has a number of civil and administrative tools at its disposal.92 In a post-Sarbanes-Oxley93 world, do outside law firms risk being charged by the SEC? And if so, how does that probability compare to that faced by inside counsel? So far, while the SEC has stepped up enforcement actions against individual inside and outside lawyers,94 the SEC has generally refrained from charging large law firms. When it has gone after outside lawyers, it has targeted sole 87. 48 Stat. 74, 82 (codified as amended at 15 U.S.C. § 77k (1982)). 88. As summarized by Kraakman, “All § 11 participants must be able to prove that ‘after reasonable investigation,’ they had ‘reasonable ground’ to believe those portions of the registration statement not made under the authority of an expert . . . [A]ll participants (other than experts) must not have had reasonable ground to disbelieve the portions of the registration statement prepared by the experts (§11 (b)(3)).” Kraakman, Gatekeepers, supra note 8, at 86. 89. Kraakman, Gatekeepers, supra note 8, at 82 (noting that “virtually all key participants in the preparation of the [registration] statement except attorneys are potentially iable under Section 11”). Of course, attorneys remain subject to private malpractice liability vis-a-vis their clients. 90. In 1995, PSLRA clarified that the SEC (but not private parties) may bring enforcement actions and administrative proceedings against aiders and abettors of securities fraud, so long as the SEC could prove that such persons “knowingly” did so. See COFFEE, GATEKEEPERS, supra note 9, at 215; JAMES D. COX ET AL., SECURITIES REGULATION—CASES AND MATERIALS 734-35 (4th ed. 2004) (discussing SEC’s authority to prosecute secondary actors as aiders and abettors). 91. THOMAS LEE HAZEN, 3 LAW SEC. REG. § 9.8[2] (5th ed. 2006). 92. The SEC may bring civil injunction actions in the federal district courts under § 20(b) of the Securities Act of 1933 and § 21(d) of the Securities Exchange Act of 1934, and administrative proceedings under Rule 102(e) of the SEC’s Rules of Practice, and administrative cease-and-desist proceedings. Lewis D. Lowenfels, Alan R. Bromberg, & Michael J. Sullivan, Attorneys as Gatekeepers: SEC Actions Against Lawyers in the Age of Sarbanes-Oxley, 37 U. TOL. L. REV. 877 (2006). In addition to civil fines, possible SEC sanctions include censure, temporary suspension, a cease and desist order, and permanent disbarments from practice before the SEC. 17 C.F.R. § 201.102(e)(1) (2002). 93. Sarbanes-Oxley Act of 2002. Pub. L. No. 107-204, 116 Stat. 745 (codified in scattered sections of 11, 15, 18, 28 and 29 U.S.C. (Supp. III 2003)). 94. See, e.g., Lowenfels et al, supra note 92, at 929 (noting that the “sheer number of SEC actions against lawyers after the enactment of SOX has increased dramatically”). In a preliminary study that I have performed, based only on a review of SEC litigation releases from January 2002 to April 2007, of approximately 82 430 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 practitioners or individual lawyers from small firms.95 So why has the SEC been reluctant to bring cases against large law firms?96 One commentator has suggested that SEC attorneys, who often come from prestigious law firms, are leery of bringing cases against the very kinds of firms they hope to join in the future.97 Or perhaps the SEC is continuing to ration its limited (although recently expanded) financial resources by avoiding costly prosecutions of large law firms.98 Unless we are to believe that large law firms simply do not commit securities infractions,99 SEC inaction translates into weak deterrence of large law firms.100 To summarize so far: large law firms were identified as the legal gatekeepers of choice because they would suffer a high cost of complicity: the potential loss of their reputation and legal liability (in the form of private civil liability and SEC enforcement). But my analysis casts some doubt on that thesis in light of modern developments, especially the judicial erosion of private aiding-and-abetting liability, which had historically operated as one of the primary deterrents for law firms. 2. LOW COST OF RESISTANCE On the other side of the balance sheet of rational choice factors is the cost of resistance. To be effective, a gatekeeper should also stand to lose little from publicized SEC actions naming attorneys as defendants, a little more than half of the cases implicate inside counsel. Research on file with author. 95. Michael A. Perino, SEC Enforcement of Attorney Up-The-Ladder Reporting Rules: An Analysis of Institutional Constraints, Norms and Biases, 49 VILL. L. REV. 851, 860 (2004). See also Sec. & Exch. Comm’n, Study and Report on Violations by Securities Professionals (2003), available at http://www.sec.gov/news/studies/ sox703report.pdf. This Report lists 13 enforcement actions—from 1998 to 2001—taken against “accounting firms” but none listed against “law firms.” Id. app. at 10. By contrast, this Report notes that the SEC took action against 49 (individual) lawyers during 1998-2001; however, it is unclear from the compilation whether the alleged infractions arose out of the lawyer’s rendering of legal services or conduct unrelated to lawyering (for example, insider trading infractions). It is noteworthy that of the 1,596 securities professionals charged, only 13 of them were charged solely as aiders and abettors. Id. at 10. 96. See, e.g., Perino, supra note 98, at 858-65 (describing the institutional incentives and constraints that explain SEC’s reluctance to bring disciplinary proceedings against lawyers and law firms, including the fact that the SEC is a lawyer-dominated agency sharing cultural norms of lawyers generally, implicit recognition of the hindsight bias, cognitive conservatism, risk-aversion in the selection of cases, and self-serving “revolving-door” careerist motivations). 97. Id. at 863-64. 98. See, e.g., COFFEE, GATEKEEPERS, supra note 9, at 155 (citing the SEC’s frozen budget during the 1990s and its other enforcement priorities as explaining its past reluctance to charge a Big Six accounting firm); Perino, supra note 95, at 853-57 (describing the SEC’s budget and personnel constraints). 99. See David J. Beck, The Legal Profession at the Crossroads: Who Will Write the Future Rules Governing the Conduct of Lawyers Representing Public Corporations?, 34 ST. MARY’S L.J. 873, 905 (2003) (reporting that SEC staff members have remarked that lawyers failed their gatekeeping duties because, in part, the SEC has declined to use its disciplinary powers, preferring to leave the work to state disciplinary committees). 100. See Erling Eide, Economics of Criminal Behavior, IN ENCYCLOPEDIA OF LAW & ECONOMICS 345, 359 (Boudewijn Bouckaert & Gerrit De Geest eds., 2000) (reviewing studies showing empirical support that the probability of punishment has a deterrent impact, while the effect of the severity of sanctions is less conclusive). 2008] GATEKEEPERS INSIDE OUT 431 resisting the wrongdoing client. The cost of resistance is typically the gatekeeper’s loss of ordinary patronage from its wrongdoing client. So, gatekeepers who rely heavily on single clients are quite susceptible to clients’ threats to take their business elsewhere.101 Conversely, gatekeepers with more diversified client bases have proportionately less at stake in preserving particular client relationships.102 This simple logic underlies the common observation that for law and accounting firms, a “diversified client base fosters professional independence.”103 And since large size often serves as a reasonably proxy for larger assets and a wider portfolio of clients,104 larger law firms should theoretically be superior gatekeepers to stand-alone individuals, such as inside counsel. But, again, recent developments have complicated the conventional wisdom. How could Arthur Andersen, an accounting firm with 2,300 audit clients generating $9.34 billion in revenues fail to gatekeep when its client Enron represented less than 1% of its total annual revenue?105 The answer may lie in the fact that large firms still suffer from principal-agent problems. The nominal gatekeeper is the outside firm, but the “functional gatekeepers” are “small teams, directed by one or more partners, which actually conduct audits, prepare legal opinions, or otherwise facilitate client transactions.”106 That small team (as agent) may act against the interests of the firm (principal), and some such story is typically used to explain the Andersen debacle.107 If we examine the incentives of the functional gatekeeper, the conventional wisdom about the vast superiority of large firms becomes tough to defend. Recently published sociological research on 787 members of the Chicago bar108 suggests that lawyers at larger firms are more dependent on any particular client than lawyers at small firms.109 The study reported that sole practitioners and lawyers at the smallest law firms had large numbers of clients and “spent a 101. Kraakman, Gatekeepers, supra note 8, at 71. As noted by Kraakman, the threat to withdraw ordinary patronage is powerful because it is “easily valued, easily arranged (it may be wholly implicit) and, most important, difficult to detect in the event of a later investigation.” Id. at 71. 102. Id. 103. Id. 104. Id. at 72. 105. Arthur Andersen, Andersen Revenues by Geographic Area (2002), http:// www.andersen.com/website.nsf/ content/MediaCenterFacts& Figures?OpenDocument (reporting 2001 figures). 106. Kraakman, Gatekeepers, supra note 8, at 72. 107. Standard explanations as to why Arthur Andersen caved into Enron management’s frauds emphasize the role of agency costs (noting the malfeasance of David Duncan, lead audit partner assigned to Enron) and the erosion of internal quality controls occasioned by the change in partner compensation practices that overvalued the consulting business while commodifying the auditing business. See, e.g., COFFEE, GATEKEEPERS, supra note 9, at 147-56. 108. JOHN P. HEINZ ET AL., URBAN LAWYERS: THE NEW SOCIAL STRUCTURE OF THE BAR, 19 (2005). 109. Id. at 115 (“[T]he degree of dependence upon any particular client is likely to be greater in the large firms.”). This study’s findings are consistent with findings of a separate study on four large Chicago firms during 1979-80. See ROBERT L. NELSON, PARTNERS WITH POWER: THE SOCIAL TRANSFORMATION OF THE LARGE LAW FIRM 251 (1988). 432 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 smaller share of their time on the client for whom they did the most work,” while lawyers in larger firms “had fewer clients and billed significantly more of their time to their leading client.”110 Interestingly, this study also found that sole practitioners and lawyers in the smallest firms were also more likely than their large firm counterparts to “refuse a potential client, or work assignment, not because of a formal conflict of interest but because of . . . personal values.”111 In addition, modern partner compensation practices no longer insulate individual attorneys from catastrophic client losses. Until the 1980s, law firm partners shared the firm’s profits according to a lock-step seniority system: partners admitted in the same class were promoted and compensated alike.112 Because partners shared income across all clients, any single client’s threat to fire the firm carried less weight.113 Today, however, all but a few law firms have replaced lock-step compensation114 with “eat-what-you-kill” schemes which divide the firm’s profits based on each partner’s direct contributions rather than seniority.115 Since a partner’s welfare is now based almost entirely on her own individual efforts to generate revenue from her own client base, the potential threat of client 110. HEINZ ET AL., supra note 108, at 115. The most time-consuming client for sole practitioners and the smallest law firms (defined as having fewer than ten lawyers) typically received less than a quarter of the lawyers’ effort (an average of 22 percent and 24 percent, respectively), but lawyers in larger firms (defined as firms with 100 or more lawyers) had allocated, on average, from 36 percent to 39 percent to a single client. 111. Id. at 118, 119. The study reports that nearly two-fifths of solo practitioners and of lawyers in the smallest firms reported that they had refused a client or work assignment at least once, but only one-fifth of lawyers in the large law firms and only 9 percent of inside counsel had reported that they had done so. As possible explanation, the authors point out that “the clients of solo practitioners and of lawyers in small firms are, on the whole, not likely to be steady customers” and thus the business consequences of refusing work from these clients are less severe than refusing work from “repeat-players.” Id. To be clear, I do not argue from the survey data that inside lawyers are more independent than lawyers at large firms. Like lawyers in large law firms, inside and government lawyers reported rarely refusing a work assignment because of personal values. As explained by the study’s authors, however, the reasons are different from that of large firm lawyers: “For the most part, [inside and government lawyers] know what kind of clients they will have when they take the job . . . . If they discover later that they do not like the clients, their principal alternative is to move to another job.” Id. at 119. 112. Gilson & Mnookin, supra note 57, at 343. 113. Kraakman, Gatekeepers, supra note 8, at 72. 114. See William D. Henderson, An Empirical Study of Single-Tier Versus Two-Tier Partnerships in the Am Law 200, 84 N.C. L. REV. 1691, 1700 (2006) (positing that lock-step partnerships can only exist in single-tier partnerships, which are tracked by The American Lawyer, and noting that 79% of the Am Law 200 law firms are two-tier partnerships); NELSON, supra note 109, at 65. 115. Gilson & Mnookin, supra note 57, at 346 (describing the marginal product approach which rewards factors that purport to measure productivity). These schemes tend to give great weight to business origination (client generation) and fees collected by the individual partner. See, e.g., Law Firms Still Pay for What Is Most Valued—Origination & Billables, COMPENSATION & BENEFITS FOR L. OFF., July 2006 (noting that business origination and personal fees collected were still the two most important factors—even rated more highly than client service—in determining partner compensation in law firms according to Altman Weil’s 2006 Survey of Compensation Systems in Private Law Firms); HEINZ ET AL., supra note 108, at 120 (noting that lawyers in all types of law firms were likely to assign great importance to business origination but that there was more unanimity for corporate law firms). 2008] GATEKEEPERS INSIDE OUT 433 defection incentivizes her to accede to client demands out of a desire to keep the client.116 Also, the client’s increased leverage over the partner is exacerbated by the looming possibility that she might be dismissed from the firm for failing to generate the threshold amount of revenue. Following the trend in the accounting industry117 beginning in the early 1990s, unproductive partners in law firms have been summarily fired.118 Short of termination, partners can be de-equitized for failing to bring in business.119 In sum, the firm’s diversified client base gives cold comfort to the individual partner—the functional gatekeeper—whose only large client is about to fire him for resistance. There is one potential rejoinder to this story, one that emphasizes the role of “peer review”120 in disciplining functional gatekeepers and steeling them against the high costs of resisting wayward clients. Theoretically, disinterested partners have strong market incentives (enhanced by joint legal liability and profitsharing) to check up on one another’s work in an effort to maintain quality and adherence to professional mandates.121 But the same eat-what-you-kill system that raises the costs of resistance for specific lawyers converts disinterested law partners into uninterested law partners. After all, eat-what-you-kill means that their finances are not much affected by what their fellow partners do. In addition, since the 1990s, many and perhaps most large U.S. law firms—including many elite firms122—have re-organized as limited liability partnerships (LLPs).123 In LLPs, a partner’s liability is generally limited to debts 116. Geoffrey Miller, From Club to Market: The Evolving Role of Business Lawyers, 74 FORDHAM L. REV. 1105, 1124 (2005) [hereinafter Miller, From Club to Market]. 117. Ianthe Jeanne Dugan, Depreciated: Did You Hear the One About the Accountant? It’s Not Very Funny—How Decade of Greed Undid The Proud Respectability Of a Very Old Profession—Prospects for Andersen Dim, WALL ST. J., Mar. 14, 2002, at A1 (describing how auditor C. Anthony Rider was fired by Ernst & Young after 25 years of service because he missed his billings quota); COFFEE, GATEKEEPERS, supra note 9, at 28 (describing how Arthur Andersen fired an estimated 10 percent of its U.S. partners in 1992). 118. See, e.g., David Margolick, Pink Slips for Law Firm Partners As Tradition Bows to Tough Times, N.Y. TIMES, Dec. 24, 1990, at 11; Donna Gill & Nancy D. Holt, Law Firms Grapple with Partner Firings, CHICAGO LAWYER, May 1992, at 1; Emily Barker, Winston and Strawn Gets Ruthless: Profits Override Personal Ties at a Firm Where Market Dynamics Dictate the Top-Down Management, and That’s Just the Way the Partners Like It, AM. LAW., June 1993, at 70. The terminations occurred in reaction to the recession of the U.S. economy which squeezed profits at law firms. 119. “De-equitizing” is the practice of revoking the “equity partner” status of a partner and converting him to a non-equity partner. See Kirkland, supra note 42, at 678; Nathan Koppel, ‘Partnership Is No Longer A Tenured Position’: More Law Firms Thin Ranks of partners to Boost Profits, Attract, Keep High Earners, WALL ST. J., July 6, 2007, at B1 (discussing wholesale de-equitization at law firms). 120. Inside the law firm, the supervisory function is delegated to the partners, who are simultaneously principals and agents of the partnership—either through a centralized committee of partners or individually on an ongoing, decentralized and informal basis, all of which I refer to broadly as “peer review.” 121. See Kraakman, Gatekeepers, supra note 8, at 72. 122. Kimberly D. Krawiec, Organizational Form as Status and Signal, 40 WAKE FOREST L. REV. 977, 980, 987 (2005) (noting that many elite New York law firms decided to convert to LLPs after the Arthur Andersen bankruptcy but also noting that 13% of all New York law firms remain general partnerships). 123. Robert W. Hillman, Organizational Choices of Professional Service Firms: An Empirical Study, 58 BUS. LAW. 1387, 1401 (2003) (noting that the LLP is emerging as the associational form of choice for larger law 434 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 arising out of her own conduct or the conduct of someone under her direct supervision.124 As Susan Saab Fortney argues, this reduction in the risk of personal liability for the conduct of one’s colleagues has “largely remove[d] a significant economic incentive to monitor the conduct of other firm participants.”125 Further, to the extent that direct supervision (or monitoring) of another potentially triggers liability, the law creates the perverse incentive not to review the work of one’s peers.126 Finally, most law firms have inadequate structural controls to ensure peerreview, which is troubling because lawyers are increasingly practicing in large, multi-jurisdictional firms where structural controls provide the only feasible means of peer-review.127 In one survey of 191 law firms conducted in 1995, a majority of firms reported not requiring committee or second principal review of legal opinions128 and not employing formal procedures for reviewing the work of partners.129 While the recent trend of law firms’ hiring in-house compliance specialists seems promising,130 those firms are likely to devote more resources to firms; 48% of the approximately 750 law firms with 50 or more lawyers are organized as LLPs; and the preference for LLP status strengthens as firm size grows). Within ten years of the first LLP statute in 1991, every state had adopted legislation allowing firms to convert to LLPs. Id. at 1394. 124. Scott Baker & Kimberly D. Krawiec, The Economics of Limited Liability: An Empirical Study of New York Law Firms, 2005 U. ILL. L. REV. 107, 109 (2005). 125. Susan Saab Fortney, Seeking Shelter in the Minefield of Unintended Consequences—The Traps of Limited Liability Law Firms, 54 WASH. & LEE L. REV. 717, 733 (1997) [hereinafter Fortney, Seeking Shelter]. There is, however, considerable debate about whether firms’ elections into LLP status is the cause or merely the effect of slack peer review. See Regan, Professional Reputation, supra note 69, at 551-54 for an excellent account of the debate; Baker & Krawiec, supra note 124, at 148 (arguing that other factors, e.g., exponential growth in law firm size, geographical dispersion, specialization, are likely to be root causes of weak monitoring and suggesting that conversion to LLP status merely reflects firms’ acknowledgment of the lack of capacity to monitor). Baker & Krawiec also note that interviews with insurers indicate that most insurance companies do not believe unlimited liability causes law firms to render higher quality legal services. Also, notwithstanding the death of vicarious personal liability, the firm itself remains potentially liable, putting the firm’s assets (and the partners’ profits) at risk, which should theoretically preserve some incentives to peer review. 126. Fortney, Seeking Shelter, supra note 125, at 734 (arguing that shirking supervisory responsibilites is a more attractive option than potentially subjecting oneself to the risk of personal liability for consulting other attorneys who end up violating professional rules). 127. Elizabeth Chambliss & David B. Wilkins, A New Framework for Law Firm Discipline, 16 GEO. J. LEGAL ETHICS 335, 342 (2003) [hereinafter Chambliss & Wilkins, Law Firm Discipline] (citing to evidence that firms have inadequate structural controls). 128. Susan Saab Fortney, Soul for Sale: An Empirical Study of Associate Satisfaction, Law Firm Culture, and the Effects of Billable Hour Requirements, 69 UMKC L. REV. 239, 286 (2000) [hereinafter Fortney, Empirical Study] (noting that only 20% of firm respondents reported utilizing committee approval of all written opinions, 35% of firms require written approval by two principals, and 59% of firms require a “standard approach” to legal opinion letters). 129. Id. at 290 (noting that 34% of firms reported designating a principal or committee to evaluate the manner in which principals handle client matters and only 10% formally review the work of principals—other than regarding compensation decisions). 130. See Elizabeth Chambliss & David B. Wilkins, The Emerging Role of Ethics Advisors, General Counsel, and Other Compliance Specialists in Large Law Firms, 44 ARIZ. L. REV. 559, 565 (2002); Chambliss & Wilkins, Law Firm Discipline, supra note 127, at 349 (noting that the “mere delegation of special responsibility 2008] GATEKEEPERS INSIDE OUT 435 adopt structural controls that benefit client interests (such as formal billing guidelines or conflicts-of-interests detection procedures) rather than third party interests, including shareholder or public interests131 (such as peer review of legal opinions delivered in connection with transactions).132 In sum, peer review cannot effectively check a functional gatekeeper’s vulnerability to a large client’s threat to withdraw patronage. To be clear, the fact that the functional gatekeepers in outside law firms are dependent on a handful of clients does not change the fact that inside counsel are dependent on a single client, their employer. In prior work, I described the formidable economic pressures faced by inside counsel: First, inside counsel are necessarily economically dependent on a single client. If they get fired,133 they lose their entire income, their insurance, their basic livelihood. If pensions or stock options have not vested, then enormous sums of money can be forfeited as well. Even worse, if they get fired for whistleblowing, they may get blacklisted—without recourse under the law to sue for retaliatory discharge. Second, even if getting fired is not likely, [the] inside [lawyer] feels unremitting pressure to justify herself and her department as a corporate cost center. In today’s competitive and profit-oriented environment, no position feels completely secure, and the case that an adequate return on firm investment is being achieved must always be made. The best way to do so is to facilitate, not interfere with, corporate transactions favored by management. Third, pleasing management can produce bigger bonuses, which have become a growing fraction of the total salary. As Patricia Hewlin has argued, the “degree to which organizations base rewards on subjective appraisals of performance will influence the degree to which employees experience dilemmas of choosing how or how not to behave when their values differ from those [to an in-house compliance specialist] provides an incentive for the responsible partner to invest in the role” and noting the development of a professional network of in-house compliance specialists). 131. Although shareholders are a constituent of the corporation, state and federal statutory and case law generally treat shareholders as third parties and not clients. See Beck, supra note 99, at 884-90 (noting the judicial tendency to characterize shareholders as nonclients and the judicial reluctance to impose liability on attorneys when the attorney was working in his capacity as an attorney). 132. Chambliss & Wilkins, Law Firm Discipline, supra note 127, at 342 (noting that ethical issues other than detection of conflicts of interest appear to get even less attention). Law firms’ structural controls are designed to limit the law firm’s overall liability. Thus, structural controls will ultimately be driven by predictions about the likelihood and magnitude of judicial findings of liability. And, so far, the judiciary has been reluctant to charge law firms, especially large law firms, with liability. See Beck, supra note 99, at 884, 893 (noting that, even prior to Central Bank, courts traditionally have been reluctant to impose civil liability on lawyers acting in their capacity as lawyers, as reflected in judicial opinions). 133. “Getting fired” has become a much more salient outcome in the era of “rank and yank” system of performance management which was popularized by Jack Welch, former CEO of General Electric. Every year employees of a company would be evaluated and ranked by their supervisors and 10% would be fired at each evaluation. In fact, Enron had used a “rank and yank” system. See generally BETHANY MCLEAN & PETER ELKIND, THE SMARTEST GUYS IN THE ROOM: THE AMAZING RISE AND SCANDALOUS FALL OF ENRON (2003). 436 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 of the organization.” As subjectivity increases, the general counsel will more likely feel pressured to conceal her personal or ethical values that differ from organizational values. In addition to routine annual “bonuses,” there may be other opportunities to be rewarded for pleasing management. For example, Kristina Mordaunt, Enron’s inside counsel, not only facilitated some of the more controversial transactions, but also greatly profited from her personal investments in one of Enron’s special purpose entities. Finally, if the inside lawyer owns stock in the firm, [a practice which had become common in the 1990s], then it may be in her self-interest not to do anything that might send the stock price into a tailspin. As shareholder, the [lawyer has a competing interest] to maximize the value of [her] equity investment [which may conflict with her] fiduciary duty to the corporation. One could argue that equity-based compensation provides incentives to be honest, since fraud should be discovered in the long run. But it is not clear that this is even true in the long run. And in the short term, it may well be in [her perceived] self-interest to misrepresent the company’s financial condition in public filings–for example, by postponing the disclosure of an adverse financial development until the next quarter (after she quickly liquidates her investment) or understating adverse items that a reasonably prudent securities lawyer would more fully disclose.134 Because inside counsel are economically at the mercy of their sole client, the costs of resisting the wrongdoer will be quite high. That said, as discussed above, the overwhelming advantage previously credited to the large outside law firm has been substantially eroded. *** In two rational choice factors—the high cost of complicity and the low cost of resistance—classical gatekeeping theory gave an overwhelming lead to outside counsel, especially large established law firms with high levels of reputational capital and diversified clienteles. But changes in substantive law have lowered the cost of complicity for law firms, which in turn reduced their competitive advantage over inside counsel. While law firms and their lawyers may still enjoy a lower cost of resistance than inside counsel, the shift to eat-what-you-kill compensation has eaten away at that previous sharp advantage. Furthermore, modern trends in firm compensation and partner liability have reduced the law firm’s incentives to control principal-agent problems, making peer review much less likely as a matter of theory and observation. B. BEHAVIORAL REALIST FACTORS As noted above, in order to understand the willingness of gatekeepers to interdict misbehavior, it may not be most illuminating to consider costs and 134. Kim, supra note 37, at 1005-07. 2008] GATEKEEPERS INSIDE OUT 437 benefits consciously calculated based on the model of the rational actor. Indeed, the efficacy of sanctions in constraining behavior may be seriously eroded by various psychological factors.135 Instead, as social psychology teaches us, we may be inclined toward certain behaviors through cognitive processes guided by the situation and the roles we inhabit in those situations. If so, we need to analyze seriously what roles lawyers generally play since these roles might clash with any imputed role of gatekeeper. Further, that clash may be more or less significant for inside or outside lawyers. In prior work, I identified three significant roles experienced by inside counsel: (1) mere employee, (2) team player, and (3) faithful agent.136 Although all lawyers are faithful agents of their clients and, to varying degrees, team players, only inside counsel experience the first role of mere employee vis-à-vis the company. Further, that role severely undermines inside counsel’s ability to be effective gatekeepers. I present that analysis first. 1. MERE EMPLOYEE Inside counsel are mere employees subject to obedience pressures. To varying degrees, we all have strong tendencies to obey authority figures out of a sense of duty.137 Psychologist Stanley Milgram conducted a series of famous “electric shock” experiments at Yale University in the early 1960s, which showed just how enormously powerful this tendency is.138 Milgram argued that in modern society the child is socialized to obey not just mom and dad, but impersonal, legitimate authority figures, e.g., schoolteachers, police officers, bosses,139 who are 135. See, e.g, Kim, supra note 37, at 1033-34 (noting the tendency to ignore abstract victims; the tendency to be more responsive to definite, immediate consequences over uncertain, delayed ones; the tendency to discount sins of omission; and the tendency to engage in small ethical lapses and gradually escalate the infraction); Langevoort, Where Were the Lawyers?, supra note 67, at 101-06 (describing how cognitive conservatism, motivated reasoning, pressures to conform to prevailing norms, and diffusion of responsibility may result—in the lawyer’s mind—in the preservation of positive schema about their clients and diminished cognitive capacity to perceive red flags). 136. Much of the following description of the situation of outside and inside counsel is adapted from prior work focusing on inside counsel. See Kim, supra note 37, at 1001-26. 137. HERBERT C. KELMAN & V. LEE HAMILTON, CRIMES OF OBEDIENCE 78 (1989). 138. Milgram’s research showed that “ordinary people responding to an advertisement for casual labor could be led, by a cunning but not coercive set of instructions, to deliver what they believed to be painful electric shocks to a protesting victim who was receiving punishments for wrong answers in a putative learning experiment.” Lee Ross & Donna Shestowsky, Contemporary Psychology’s Challenges to Legal Theory and Practice, 97 NW. U. L. REV. 1081, 1097 (2003). For a detailed discussion of Milgram’s obedience experiments, see Kim, supra note 37, at 992-97 and 1001-04. 139. A.G. MILLER, THE OBEDIENCE EXPERIMENTS: A CASE STUDY OF CONTROVERSY IN THE SOCIAL SCIENCES, 223 (1986). Milgram argued that obedience derived from the perceived legitimacy of the experimenter and his commands, although others have argued that the perceived “greater expertise” of the experimenter was the driving force behind subjects’ obedience. But the “trusting the expert” explanation fails adequately to account for subjects’ behavior in one experimental variation where the experimenter gives his orders from another room and, thus, cannot see what level of shock the teacher is actually administering. In this version, compliance drops considerably. See Stanley Milgram, Some Conditions of Obedience and Disobedience to Authority, 18 HUM. 438 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 perceived as having the right to issue commands.140 Dress, diploma, title, insignia, and institutional settings are often sufficient earmarks of “legitimate” status.141 In bureaucratic settings, the authority holder is generally seen as being entitled to make demands and the subordinate is obligated to accede to them by virtue of their respective positions in the organizational hierarchy.142 Once an authority holder is perceived as legitimate, the subordinate will look upon him as a source of valid commands, and the commitments and obligations inherent in the subordinate’s role will automatically be activated.143 In the face of a legitimate command, the subordinate’s focus becomes not on what she wants to do among all available options but, rather, on what is required of her to do.144 In other words, her personal preferences take a back seat to her role obligations. She effectively interprets the meaning of the situation as the authority-holder defines it—as if she has no choice in the matter, unless she is able to marshal the psychological resources to redefine the situation, her relationship to the authority-holder, and the broader social context.145 And once a subordinate accepts the authority-holder’s “definition of the situation,” an act that would normally be viewed as heinous—such as administering lethal electric shocks to a protesting victim as punishment for errors of memory—may seem fully warranted in the situation.146 Accordingly, inside counsel, as employees of the firm, are inclined to take orders and accept the “definition of the situation” from their superiors. These superiors happen to be a cohort of non-lawyer senior managers vested with the authority to speak on behalf of the organization and entrusted to give direction to inside counsel on corporate matters. They create the reality for inside counsel: REL. 65-66 (1965); David J. Luban, The Ethics of Wrongful Obedience, in ETHICS IN PRACTICE: LAWYERS’ ROLES, RESPONSIBILITIES, AND REGULATION 94, 101 (Deborah L. Rhode ed., 2000) (the experimenter’s “superior knowledge is no different than if he were standing directly behind the teacher”). For a survey of the different aspects of the nature of authority, see Thomas Blass, The Milgram Paradigm After 35 Years: Some Things We Now Know About Obedience to Authority, in OBEDIENCE TO AUTHORITY: CURRENT PERSPECTIVES ON THE MILGRAM PARADIGM 35, 41-44 (Thomas Blass ed., 2000). 140. Blass, supra note 139, at 38-41 (noting Milgram’s position on the nature of the authority). 141. MILLER, supra note 139, at 223. 142. KELMAN & HAMILTON, supra note 137, at 89. 143. Id. at 90. 144. Id. at 90. The subordinate essentially enters into a situation where her personal preferences become much less relevant predictors of her behavior. Id. at 89. 145. Although there are instances of unquestioning obedience, in reality, most authority situations are less black-and-white and personal preferences do play some role, albeit a limited one. Disobedience typically occurs when there are “competing definitions of the situation anchored in alternative claims in legitimacy.” KELMAN & HAMILTON, supra note 137, at 90, 96, 97, 100. See Kim, supra note 37, at 1004 (clarifying and qualifying the nature of obedience pressures). That said, Milgram’s experiments are often cited to show how truly difficult it is for people to redefine the situation, instead succumbing to the situation as defined by the authority holder. As David Luban notes, “Milgram demonstrates that each of us ought to believe three things about ourselves: that we disapprove of destructive obedience, that we think we would never engage in it, and, more likely than not, that we are wrong to think we would never engage in it.” Luban, supra note 139, at 97. 146. STANLEY MILGRAM, OBEDIENCE TO AUTHORITY: AN EXPERIMENTAL VIEW 145 (1974). 2008] GATEKEEPERS INSIDE OUT 439 they define objectives, identify specific responsibilities for inside lawyers and, ultimately, determine whether an inside lawyer’s performance is acceptable.147 Adopting management’s definition of the situation means accepting management’s framing of the inside lawyer’s role and responsibilities. This framing may provide that compliance responsibilities be segmented. Although inside counsel’s duties include a prominent role in corporate compliance, it is not uncommon for business managers to jealously guard the right to decide whether to comply with the law,148 which is seen as the ultimate risk management decision. For inside counsel to challenge management’s commands or self-professed authority to make such decisions would be perceived as a violation of inside counsel’s role commitments and an embarrassing disruption of a well-defined social script. Obedience in the corporate context will be substantial, so we should not be surprised by the banal tendency to listen to superiors. Also, given the stark financial self-interest,149 which is threatened by asking too many questions or being viewed as obstructionist, we should not be surprised that inside counsel often succumb to the situation. 2. TEAM PLAYER The situational characteristics of employee status emphasize the consequences of vertical hierarchy. But the inside lawyer is also a team player, a loyal member of the corporate club. As such, horizontal social pressures further press lawyers to countenance unethical behavior. As team players, inside counsel are subject to considerable conformity pressures that arise from personal relationships with colleagues and peers.150 Half a century ago, Solomon Asch conducted a series of famous experiments that showed that a group norm could influence behavior, even in the face of clear, 147. Although the general counsel of most corporations is considered a member of senior management, she invariably reports to another member of senior management, usually the CEO, a co-agent but de facto boss. See, e.g., Stephen M. Bainbridge, The Tournament at the Intersection of Business and Legal Ethics, 1 U. ST. THOMAS L.J. 909, 920 (2004) [hereinafter Bainbridge, Tournament] (observing that, although often formally appointed by the board, the general counsel’s tenure normally depends on her relationship with the CEO). For surveys on inside counsel’s reporting relationships, see infra note 198. Most inside counsel view non-lawyer business managers as their de facto clients. Kim, supra note 37, at 1008-09. 148. See Robert L. Nelson & Laura Beth Nielsen, Cops, Counsel, and Entrepreneurs: Constructing the Role of Inside Counsel in Large Corporations, 34 LAW & SOC’Y REV. 457, 473 (2000) (noting that half of the lawyers in the sample acknowledge that, most of the time, the businesspeople in the company make the final determination regarding whether to assume a legal risk). 149. Indeed, the stark financial interest for inside counsel to accede to managment’s demands may be potent enough to induce obedience even if inside counsel personally views such demands as illegitimate. See supra Part II.A.2.; Kim, supra note 37, at 1005-08 (discussing the financial interest of inside counsel). For understanding how financial self-interest is processed via motivated reasoning, see Kim, supra note 37, at 1026-34. 150. “Conformity” is “a change in belief or behavior in response to real or imagined group pressure when there is no direct request to comply with the group nor any reasons given to justify the behavior change.” PHILIP G. ZIMBARDO & MICHAEL R. LEIPPE, THE PSYCHOLOGY OF ATTITUDE CHANGE AND SOCIAL INFLUENCE 55 (1991). 440 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 objective standards.151 Group consensus is a powerful facet of the situation, and the observed tendency to reach agreement with the group is a “dynamic requirement of the situation.”152 Asch’s experiments revealed an important motivation behind conformity pressures. Even those subjects who consistently remained independent from group consensus expressed fear of being stigmatized for dissenting from the group.153 This fear of stigmatization is a normal by-product of group identity– however the group is defined. Countless psychological studies have demonstrated that “mere categorization” of people into different nominal groups, even in the absence of any close relationship among group members, can elicit tendencies to associate positive attributes toward in-group members (in-group favoritism) and associate negative attributes toward out-group members (outgroup derogation or discrimination).154 This tendency is so robust that even the most arbitrary and inconsequential group classifications can have real behavioral consequences.155 Since the inside lawyer is a member of a group-the company’s employees-she is pressured to conform her behavior to that of other group members. To interdict fellow group members would set the inside lawyer apart from the group, risking stigmatization and exclusion. Accordingly, organizational behavior studies report widespread employee silence about ethical issues in the workplace,156 confirming strong team player pressures to look the other way. The fear of being labeled 151. For a detailed discussion of the Asch experiments, see Kim, supra note 37, at 1019-22; SOLOMON E. ASCH, SOCIAL PSYCHOLOGY 483 (1952) [hereinafter ASCH, SOCIAL PSYCHOLOGY] (noting that “In different ways and with different strengths the situations studied showed a ‘pull toward the group’; changes of estimation went overwhelmingly in the majority direction.”); ZIMBARDO & LEIPPE, supra note 150, at 57. 152. ASCH, SOCIAL PSYCHOLOGY, supra note 151, at 484. 153. Asch concluded that one of the underlying motivations for subjects’ conforming behavior was the avoidance of group disapproval: “subjects expressed fear of conspicuousness, of public exposure of personal defects, and of group disapproval; they felt the loneliness of their situation.” The most typical reason for errors given was “the painfulness of standing alone against the majority.” Some subjects were “dominated by an imperious desire not to appear different, apparently out of fear of revealing a general and undefined defect.” Solomon E. Asch, Studies of Independence and Conformity, 70 PSYCHOL. MONOGRAPHS: GEN. & APPLIED 1, 70 (1956). The term ‘stigma avoidance’ has been coined to describe ‘efforts organization members take to avoid character blemishes associated with exposed wrongdoings, or displaying objection to wrongdoings they witness at work.’ Patricia Faison Hewlin & Ashleigh Shelby Rosette, Stigma Avoidance: A Precursor to Workplace Discrimination, Presentation at the Academy of Management Annual Meeting 8 (2004) (manuscript on file with author) (discussing stigma avoidance in the context of workplace discrimination as a motivating factor for organizational silence). The definition of ‘stigma‘ that I adopt is from sociologist Erving Goffman who defined ‘stigma’ as an attribute of a person that is deeply discrediting, reduced him or her ‘in our minds from a whole and usual person to a tainted, discounted one.’ See ERVING GOFFMAN, STIGMA: NOTES ON THE MANAGEMENT OF SPOILED IDENTITY 3 (1963). 154. See LEE ROSS & RICHARD E. NISBETT, THE PERSON AND THE SITUATION 40 (1991). 155. The view that “we” are somehow better and more deserving than “they” is a “rather basic aspect of social perception.” Id. For a recent discussion of Social Identity Theory in the context of race relations, see Kang, Trojan Horses of Race, supra note 38, at 1533. 156. For a summary of empirical findings of widespread employee silence within organizations, see Kim, supra note 37, at 1024-26. 2008] GATEKEEPERS INSIDE OUT 441 negatively as a “troublemaker,” “complainer,” or “tattletale”157 is a strong motivation not to interdict. Moreover, to the extent that the group is highly cohesive and characterized by strong civility and cooperation norms, achieving consensus will be valued at the expense of high-quality decision-making.158 Sobering accounts of inside lawyers who sued their employers under a claim of retaliatory discharge reveal that some inside lawyers are ostensibly fired for not being “team players.”159 An inside lawyer who goes over the head of her boss and makes a report under SOX to the Board of Directors is likely to be perceived as not being a team player. More likely than not, she will find it easier to stay silent rather than risk the consequences of stigmatization. To the extent that outside counsel identify themselves as members of the client’s in-group, they, too, will be subject to team player pressures to facilitate the client’s objectives, not to obstruct them. They will be influenced by the motivations and biases of a fairly small number of client representatives with whom they have direct contact160 and will often not even be exposed to dissenting voices from within the corporate client. Of course, it is plausible that outside counsel will experience countervailing team player pressures from their law firms or their local bar association to be ethical and independent. But, as my discussion above on eat-what-you-kill compensation systems suggests,161 and as Stephen Bainbridge has argued, partners who win the law firm’s promotion-topartner “tournament” are likely to overlook management wrongdoing.162 3. FAITHFUL AGENT The above two factors—the role of mere employee generating obedience, and 157. Frances J. Milliken et al., An Exploratory Study of Employee Silence: Issues that Employees Don’t Communicate Upward and Why, 40 J. MGMT. STUD. 1453, 1463 (2003). 158. Marleen A. O’Connor, The Enron Board: The Perils of Groupthink, 71 U. CIN. L. REV. 1233, 1238 (2003); see also Michael B. Dorff, The Group Dynamics Theory of Executive Compensation, 28 CARDOZO L. REV. 2025 (2007) (applying groupthink and other theories to executive compensation); IRVING L. JANIS, GROUPTHINK: PSYCHOLOGICAL STUDIES OF POLICY DECISIONS AND FIASCOES (2d ed. 1982). 159. See, e.g., Willy v. Coastal States Mgmt., 939 S.W.2d 193, 197 (Tex. App. 1996) (noting that although the discharged attorney was a “capable environmental attorney,” “his inability to get along with others . . . led to his termination”); GTE Prods. Corp. v. Stewart, 653 N.E.2d 161, 164 (Mass. 1995) (noting that the discharged attorney was purportedly fired for being “confrontational” and that [the attorney] “was going to have to learn to get along with [the manager] or his future with GTE would be at risk”). 160. Donald Langevoort notes that lawyers can become so much a part of the client team that they share the motivations and biases that affect corporate managers, especially if they are involved in the decision making to take a course of action. Donald C. Langevoort, The Epistemology of Corporate-Securities Lawyering: Beliefs, Biases and Organizational Behavior, 63 BROOK. L. REV. 629, 647 (1997) [hereinafter Langevoort, Epistemology]. 161. See supra Part II.A.2. 162. See Bainbridge, Tournament, supra note 147, at 918 (noting that “winning the tournament requires developing a set of skills and attitudes aimed squarely at keeping clients happy,” which in turn “develops lawyers with strong incentives to overlook management wrongdoing”). 442 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 the role of team player generating conformity—applied mainly to inside counsel (although outside counsel also suffer from team player pressures). Further, both factors cut strongly against inside counsel as successful gatekeepers. By contrast, both inside and outside counsel are full-fledged faithful agents of their clients and are thus both subject to powerful alignment pressures.163 On comparison, whether inside or outside attorneys better bear these pressures turns out to be a complicated question. Like all agents, the lawyer is accountable164 to her client,165 which means that she expects to be called on to justify her behavior to her client. This expectation, in turn, creates pressures to align her views with those of her client. One study performed on 139 professional auditors employed full-time by one of the Big Four U.S. accounting firms indicates that accountability can produce a chameleonlike shift in both public behavior and private beliefs in the client’s direction.166 Since agents are, by definition, accountable to their principals, lawyers (as agents) are more likely to adopt positions that are aligned with their clients. But the lawyer is no ordinary agent. The characteristic activities of lawyering may contribute to the strength of a lawyer’s accountability to her client. Gerald Postema has explained that, unlike the physician or auto mechanic, the lawyer “act[s] in the place of the client, [which] require[s] the direct involvement of the lawyer’s moral faculties—i.e., his capacities to deliberate, reason, argue, and act in the public arena.”167 As the client’s agent, the lawyer speaks on behalf of the client and enters into relationships with others in the name of the client,168 becoming–at the invitation of the client–an extension of the legal and moral personality of the client.169 Consistently acting out behavior may have farreaching psychological consequences. Cognitive dissonance theory predicts that our internal attitudes are likely to shift toward the position implied by our actions.170 Another key variable that may affect the strength of an agent’s accountability to her principal is her understanding of the nature of her role as attorney and the 163. See Kim, supra note 37, at 1008-11 for a discussion of alignment pressures. 164. “Accountability” has been defined as the “implicit or explicit expectation that one may be called on to justify one’s beliefs, feelings, and actions to others.” J.S. Lerner & P.R. Tetlock, Accounting for the Effects of Accountability, 125 J. PERSONALITY & SOC. PSYCHOL. 255 (1999). 165. Formally, the client is the organization and not any of its constituents (e.g., board, managers, shareholders). Of course, the organization is entirely fictitious and can only act through the flesh and blood of its constituents. Therefore, in practice, lawyers suspend this fiction and interact with senior management as their de facto client or principal, when in fact they are mere co-agents. See Kim, supra note 37, at 1008. 166. Id. at 1009-11 (summarizing the results of the auditing vignettes study); Don A. Moore et al., Conflict of Interest and the Unconscious Intrusion of Bias (Harvard Negotiations, Orgs., and Mkts. Unit, Working Paper No. 02-40, 2002) (manuscript on file with author). 167. See Gerald J. Postema, Moral Responsibility in Professional Ethics, 55 N.Y.U. L. REV. 63, 76 (1980). 168. Id. at 77. 169. Id. 170. See ZIVA KUNDA, SOCIAL COGNITION: MAKING SENSE OF PEOPLE, 216-20, 533 (1999). 2008] GATEKEEPERS INSIDE OUT 443 ideological or normative commitments that such a role entails–role ideology.171 Since humans are social animals who are highly attuned to group norms, role ideologies derived from group norms present nontrivial ex ante situational influences that constrain action by defining the outer limits of socially acceptable behavior. Role ideologies also serve to legitimate any post hoc rationalizations of unethical behavior, by framing an ethical problem in a way that dismisses the moral component or by making unethical conduct appear more palatable. Modern legal culture makes available various role ideologies off-the-shelf, ranging from those openly hostile to law (such as the libertarian-antinomian view that “law is the enemy”) to those deferential to the law (such as the lawyer-statesman or officer-of-the-court view).172 But the dominant role ideology for transactional lawyers seems to be a middle-of-the-road, agnostic view that law is a neutral constraint, and the lawyer’s role is that of an amoral risk-assessor.173 On this view, the lawyer’s accountability is not divided in any sense between justice and the client.174 Accordingly, accountability to the client remains strong. The lawyer’s role becomes that of a deprofessionalized technician who exploits her mastery of the rules to lower tariffs as much as possible to the narrow advantage of her client.175 Comparison between Inside and Outside. This discussion of faithful agent pressures has been, so far, about lawyers vis-à-vis their clients. What about comparing the pressures experienced by inside as opposed to outside counsel? In my view, inside counsel (more so than outside counsel) experience two sets of alignment pressures within their faithful agent role because, on the inside, the distinction between the de facto and de jure client is more salient. Formally, the de jure client is the organization and not its constituents,176 such as the board of directors, the employees, or even the shareholders. Of course, since the organization is entirely fictitious, the company’s duly authorized senior manag- 171. The term “role ideology” is borrowed from “gender-role ideology” which refers to prescriptive beliefs about roles and behaviors for women and men, with the dimensions being defined by “traditional” and “feminist” poles. 172. For a more detailed discussion of the various role ideologies discussed by scholars, see Kim, supra note 37, at 1013-15. 173. See Robert W. Gordon, A New Role for Lawyers?: The Corporate Counselor After Enron, 35 CONN. L. REV. 1185, 1192 (2003) (describing dominant conceptions of lawyers’ role). 174. For example, the agnostic view of (regulatory) law is reflected in the comments of Professor (Judge) Easterbrook and Professor Daniel Fischel regarding corporate law compliance: “Managers have no general obligation to avoid violating regulatory laws, when violations are profitable to the firm . . . We put to one side laws concerning violence or other acts thought to be malum in se.” Frank H. Easterbrook & Daniel R. Fischel, Antitrust Suits by Targets of Tender Offers, 80 MICH. L. REV. 115, 1168 n.36 (1982) (citations omitted). For a summary and critique of this agnostic view of law, see Cynthia A. Williams, Corporate Compliance with the Law in the Era of Efficiency, 76 N.C. L. REV. 1265 (1998). See also Anthony V. Alfieri, The Fall of Legal Ethics and the Rise of Risk Management, 94 GEO. L.J. 1909, 1933-1940 (2006) (discussing risk-management norms). 175. Gordon, supra note 173, at 1192. 176. Geoffrey C. Hazard, Ethical Dilemmas of Corporate Counsel, 46 EMORY L.J. 1011, 1013 (1997). 444 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 ers177 are treated as de facto clients, although they are technically co-agents with all lawyers representing the organization.178 But inside counsel are also typically part of the company’s senior management team. Therefore, the CEO and other senior managers who are perceived as the de facto client are simultaneously recognized by inside counsel as arguably co-equal delegatees of corporate power, specialized according to corporate function.179 This twin perspective—that senior managers are both clients and co-agents— sensitizes inside counsel to the distinction between de facto and de jure clients. This understanding is reinforced by common practices. For example, inside counsel are often restricted from representing individual corporate employees while outside counsel are free to do so (absent a legal conflict of interest).180 Inside counsel are also cautioned in continuing legal education to give senior managers the “corporate Miranda warning,” which puts the manager on notice that counsel owes a duty of loyalty to the company alone (and not to any manager) and thus cannot promise confidentiality or loyalty to any individual manager.181 Further, as purchasers of outside legal services, inside lawyers largely act as the de facto client to the outside law firm. This puts inside counsel in the position and habit of defining corporate objectives and making certain risk decisions on behalf of the corporation. Such regular exercises of corporate authority create room for inside counsel to develop distinct views about what the best interests of the corporation are, and some of those views may conflict with those of senior managers. To the extent that inside counsel feel that their views better reflect the corporation’s best interests, this will harbor conflict about what they perceive their fiduciary duties require them to do. In sharp contrast, outside counsel rarely experience much tension within their 177. For inside counsel, the “senior managers” who function as de facto clients are the senior members of other departments within the corporation who are delegated the authority to give instruction to inside counsel on specific projects. For outside counsel, the “senior managers” who serve as the de facto clients are typically the inside lawyers. 178. See Kim, supra note 37, at 1008-09. William Simon notes that although lawyers know that “in principle the corporation is not the same thing as its management,” they have “no clear conception of what else it could be,” so the tendency is to “instinctively fall back on views that conflate the organization and its personnel.” William H. Simon, After Confidentiality: Rethinking the Professional Responsibilities of the Business Lawyer, 75 FORDHAM L. REV. 1453, 1454 (2006). Indeed, the treatment of senior managers of the corporation as principals is basically endorsed by the American Bar Association’s Model Rules of Professional Conduct. See Kim, supra note 37, at 1044-48 (critiquing Model Rules as contributing to the conflation of managers with the corporate client). 179. See, e.g., Robert Eli Rosen, “We’re All Consultants Now”: How Change in Client Organizational Strategies Influences Change in the Organization of Corporate Legal Services, 44 ARIZ. L. REV. 637, 670 [hereinafter Rosen, We’re All Consultants] (noting that for redesigned corporations in the information industry where multidisciplinary project teams are employed, “lawyers meet as equals with, for example, engineers and accountants”). 180. See LISA G. LERMAN & PHILIP G. SCHRAG, ETHICAL PROBLEMS IN THE PRACTICE OF LAW 317 (2005). 181. John K. Villa, When and How to Issue Corporate Miranda Warnings, 24 NO. 8 ACC DOCKET 76 (2006). 2008] GATEKEEPERS INSIDE OUT 445 “faithful agent” role about who the real client is. This is because, as I explain further below, the outside counsel is so starved of relevant information.182 Being at the end of multiple chains of authority and receiving only sanitized information filtered through inside counsel for discrete, sharply circumscribed projects, outside counsel will not presume that it has any privileged access to or special insight about the de jure principal and its true best interests. What are the consequences of this complicated agency picture? If lawyers generally are bad gatekeepers because they play the role of faithful agents, it may be that inside counsel are in a slightly better situation than outside counsel. Inside counsel, in effect, recognize and feel the conflicting currents of serving two masters, de facto and de jure. They may better perceive that their duty to the de jure client (the organization) may at times require them to take a stand against the de facto client (the senior managers). Some recent sociological research is consistent with this portrayal. In 2000, interviews with 86 informants from approximately 46 Fortune 1000 companies reported that 17 percent of inside counsel identified their primary concern as policing the conduct of their co-agents.183 In 2003, the leading in-house bar association performed a survey of 1,216 inside counsel, which reported that a significant percentage of inside counsel want an expanded role in preventing and reporting fraud committed by corporate employees.184 This is notwithstanding their being faithful agents by virtue of being lawyers. And, given law firms’ increasing “customer service” orientation,185 it seems unlikely that outside lawyers would identify with these gatekeeping objectives at the same frequency. *** With respect to the behavioral realist factors, outside counsel seem to have a distinct gatekeeping advantage because they are not mere employees subject to obedience pressures, as inside counsel are. However, some outside counsel remain subject to team-player pressures to conform, albeit to a lesser extent than inside counsel. As for alignment pressures arising from inside and outside 182. See infra Part IV. 183. Nelson & Nielsen, supra note 148, at 460, 461, 463, 468 (defining the ideal type (classification) of “cops” as those corporate counsel who are “primarily concerned with policing the conduct of their business clients” and noting that 17 percent of inside counsel in their sample approximated the role of “cop” within their organizations). 184. Chad R. Brown, In-House Counsel Responsibilities in the Post-Enron Environment, 21 NO. 5 ACCA DOCKET 92 (2003) [hereinafter Brown, In-House Counsel]. To summarize a few significant findings from this survey, 78 percent of respondents felt that the general counsel or other inside attorneys should report misconduct to appropriate corporate officials; 71 percent felt that the law should be clearly defined and reporting illegal behavior made mandatory, regardless of the attorney-client privilege, to ensure the well-being of the corporate client; 35 percent felt that disclosure of confidential communications (to report misconduct to the SEC) should be mandatory while 46 percent felt that it should be permissive. Id. at 97, 98. 185. See Rosen, We’re All Consultants, supra note 179, at 671-72 (describing the re-orientation of law firms during the 1990s, as exemplified by slogans such as “Clients First,” “Committed to You and Your Achievements,” “Legal Solutions Tailor Made”). 446 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 counsel’s common status as faithful agents, the analysis is indeed complicated. It may well be that inside counsel are more adept at dissociating the identities of their de facto clients (senior managers) from their de jure client (the organization). If we add these results to our analysis of the rational choice factors (the high cost of complicity and the low cost of resistance) discussed in Part II.A, we get conflicted support for the conventional wisdom. On net, it does appear that outside counsel may be more willing than inside counsel to interdict client misbehavior. That said, the magnitude of that difference is smaller than commonly assumed. III. WILLINGNESS TO MONITOR Before a gatekeeper can respond, she must, of course, know to respond. How does the gatekeeper come to know? In part, the gatekeeper must be willing to expend the time, energy and money to maintain a general vigilance at all times. When red flags appear, the gatekeeper must be willing to further probe those suspicious circumstances. Quadrant II interrogates the gatekeeper’s willingness to monitor. Generally, all the factors that I have already discussed analyzing the willingness to interdict also affect the willingness to monitor. Still, there is one key difference worth examining—the fundamental tension between the incentives that promote interdiction and the incentives that promote vigilance. A gatekeeper who potentially faces liability for having knowledge of wrongdoing but failing to stop it may avoid expending the resources ex ante to gain such knowledge. In short, the problem is the familiar one of “willful blindness.” Recognizing this tension, the law has devised both positive inducements (“carrots”) and sanctions (“sticks”) to directly encourage companies to monitor. As an example of carrots, the 1991 federal Organizational Sentencing Guidelines afford companies a fine reduction for effective compliance programs in place at the time of the offense or the prompt self-reporting of offenses to enforcement authorities.186 As an example of sticks, the Delaware Chancery Court case In re 186. H. Lowell Brown, The Corporate Director’s Compliance Oversight Responsibility in the Post Caremark Era, 26 DEL. J. CORP. L. 1, 85-88 (2001) (describing incentives for corporate compliance efforts). The Guidelines’ approach of using carrots to supplement the sticks of heavy criminal sanctions has since been emulated by judicial, legislative and regulatory bodies in a wide variety of civil and criminal contexts. See, e.g., Kimberly D. Krawiec, Organizational Misconduct: Beyond the Principal-Agent Model, 32 FLA. ST. U. L. REV. 571, 585-91 (2005) (summarizing various incentives for companies to adopt internal compliance structures created by the Environmental Protection Agency, the Department of Health and Human Services, states’ attorneys general, judicial decisions, securities law provisions (Section 15(b)(4)(E) of the Securities Exchange Act), Delaware corporate law, and equal employment opportunity law (especially the law governing workplace harassment)); Richard S. Gruner, General Counsel in an Era of Compliance Programs and Corporate Self-Policing, 46 EMORY L.J. 1113, 1126-34 (1997) (summarizing various incentives for companies to adopt internal compliance structures or self-report offenses, including those created by federal amnesty programs, the Environmental Protection Agency’s civil penalty standards, formal Department of Justice prosecution 2008] GATEKEEPERS INSIDE OUT 447 Caremark International Inc. Derivative Litigation187 imposed a duty to monitor on companies’ boards of directors, backed by the threat of personal civil liability.188 While Caremark did not explicitly require directors to implement compliance programs to monitor their companies, many boards have proceeded to do so as a precaution. Given the above carrots and sticks designed to encourage monitoring, companies have invested resources to adopt corporate compliance programs. Much of this work fell on the lap of inside counsel, who have been key players in the design, implementation and, to varying degrees, continued oversight of these monitoring programs.189 In fact, many inside counsel have ongoing monitoring responsibilities,190 with a significant minority identifying “policing” as their primary responsibility.191 Inside counsel often make it a point to continually educate themselves about best practices in compliance.192 Given that monitoring is often clearly part of their job, we should expect inside counsel to be willing. By contrast, given the prevalence of eat-what-you-kill compensation systems that tie the partner’s compensation to business origination and fees collected,193 outside counsel are unlikely to expend resources on monitoring the client’s ongoing affairs unless they can charge for it. And due to cost considerations, companies are typically unwilling to retain outside counsel for general monitoring duties, preferring instead to use law firms for discrete budgeted services in a task-specific and ad hoc manner.194 While outside counsel are retained to design corporate compliance programs and perform specialized audits, ongoing monitor- standards, informal prosecutorial charging decisions, the Occupational Safety and Health Administration Star Program standards, and federal standards for government contractors). 187. 698 A.2d 959 (Del. Ch. 1996). 188. The sustained failure to observe the duty may lead to director liability under Delaware’s fiduciary duty law. Id. at 970 (holding that directors of Delaware corporations may incur personal liability for losses arising out of employee misconduct if the directors fail to “attempt in good faith to assure that a corporate information and reporting system [aimed at detecting misconduct] . . . exists” and is adequate for that task). 189. Gruner, supra note 186, at 1141, 1142; Abram Chayes & Antonia H. Chayes, 37 STAN. L. REV. 277, 284, 285 (1985) (noting that that corporate legal departments report that “programmatic prevention” requires a substantial portion of inside counsel’s time). 190. Id. at 287-88 (noting that corporate legal departments bear primary, though not exclusive, responsibility for compliance efforts and describing inside counsel’s continuing monitoring responsibilities, including performing audits). 191. Nelson & Nielsen, supra note 148, at 460, 461, 463, 468 (defining the ideal type (classification) of “cops” as those corporate counsel who are “primarily concerned with policing the conduct of their business clients” and noting that 17 percent of inside counsel in their sample approximated the role of “cop” within their organizations). 192. The annual meeting of the leading in-house bar association, The Association of Corporate Counsel, regularly includes sessions on corporate compliance. See, e.g., Program Schedule for ACC’s 2007 Annual Meeting, Compliance Readiness ⫺ Avoiding Surprises, http://am.acc.com/conference_program.cfm (last visited Mar. 18, 2008). 193. See supra Part II.A.2. 194. See infra Part IV.A. 448 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 ing of compliance remains largely within the province of inside counsel.195 *** Because of legal developments that have spurred companies to expend resources on corporate compliance programs with responsibilities that often fall on inside counsel, inside counsel should have an edge over outside counsel in their willingness to monitor. Stepping back one level to summarize roughly, we see that outside counsel are probably more willing to interdict, while inside counsel may be more willing to monitor. But who is the more capable monitor? IV. CAPACITY TO MONITOR It is not enough to be willing; the gatekeeper must also be able. Quadrant III interrogates this capacity, as applied to monitoring.196 In other words, can the gatekeeper access the facts necessary to uncover misconduct?197 It makes no sense to impose elaborate gatekeeping duties on a party who simply does not have access. Surprisingly, no one has seriously analyzed this factor in the gatekeeping literature. A. FORMAL COMMUNICATION CHANNELS Access is a function of both the gatekeeper’s formal and informal communication channels. As defined here, formal communication channels are those information flows arising out of the company’s need to operate through delegated power. These information flows may be vertical (e.g., relationship between subordinate inside lawyer and general counsel) or horizontal (e.g., relationship between inside counsel and an employee in the marketing department). These formal information channels are planned, organized, routinized, and suggested by the company’s organizational chart. Inside counsel’s formal communication channels are incident to the various functions they serve. In performing three distinguishable functions—internal advising, compliance, and external liaising—inside counsel are well-positioned to access critical information that may reveal company misconduct. Internal advising. The general counsel is typically positioned as an officer and member of the senior executive team. The majority of general counsel report 195. Chayes & Chayes, supra note 189, at 284 (“While some law firms have been quite aggressive in establishing compliance programs for clients and in offering legal audits in specialized corporate activities, programmatic prevention has been, and by its nature is likely to remain, a creature of internal counsel.”). 196. Reinier Kraakman describes this as “the ability to detect [misconduct] at a relatively low cost.” Kraakman, Gatekeepers, supra note 8, at 86. 197. The “selection of gatekeepers and the design of their duties . . . must be tailored to the realities of access to information.” Kraakman, Corporate Liability, supra note 10, at 896 (emphasis added). 2008] GATEKEEPERS INSIDE OUT 449 directly to the CEO198 and, under normal circumstances, furnish legal advice to the board of directors, the CEO, and other senior executives on major transactions and litigation.199 In fact, most significant day-to-day and long-term strategic company actions raise complicated legal issues. Accordingly, the general counsel must regularly provide critical legal advice, which in turn supposes that she is regularly exposed to and informed about critical business facts. In practice, of course, inside counsel may be far less “in the loop” than the formal organizational chart suggests. They may have more limited access to the CEO and the board,200 as well as limited review of financial and accounting information.201 Although most companies have centralized legal departments,202 others with geographically dispersed operations tend to have more decentralized legal departments, with on-site lawyers reporting directly to the top manager of the operating unit and only indirectly to the corporation’s general counsel.203 In those cases, while on-site lawyers have easy access to facts about their own business units, the flow of information upward to the general counsel at company headquarters may be impeded. 198. A survey conducted by ACC in 2001 reported that 61.4% of general counsels reported to the CEO, 15.3% reported to the president, and 12.7% reported to another executive. Approximately 7% of respondents indicated that the general counsel reported to the CFO. Susan Hackett, Inside Out: An Examination of Demographic Trends in the In-House Profession, 44 ARIZ. L. REV. 609, 612 (2002) (reporting survey results and noting that general counsels are fighting the trend to transfer reporting to the CFO). According to another survey of 167 mostly public large companies, 74% of corporate general counsel report to the CEO, with the remaining reporting to the president, chairman, chief financial officer or someone else. Ashby Jones, Silicon Valley’s Outsiders: In-House Lawyers? WALL ST. J., Oct. 2, 2006, at B3. 199. Carl D. Liggio, The Changing Role of Corporate Counsel, 46 EMORY L. J. 1201, 1208 (1997). 200. Brown, In-House Counsel, supra note 184, at 96 (reporting that in-house counsel believe that greater access to the CEO and Board of Directors on their part will reduce fraud and that 44 percent of in-house counsel surveyed believed that better access to the board of directors was needed to ensure the well-being of their company). 201. In a survey of inside counsel performed in 2003, a significant percentage of inside counsel reported that they are too often left “out of the loop” on financial and accounting issues. Brown, In-House Counsel, supra note 184, at 97, 98 (reporting that 49% of in-house counsel polled felt that in-house counsel were generally kept informed but were still kept “out of the loop” on important developments as they related to financial and accounting issues, while 39% felt that in-house counsel were kept well-informed). 202. DeMott, Discrete Roles, supra note 7, at 970 (noting that most companies have centralized legal departments); Hackett, supra note 198, at 612 (noting that 81.2% of respondents indicated that the structure of their legal staff within their organization is centralized or mostly centralized). 203. See, e.g., DeMott, Discrete Roles, supra note 7, at 970. See also Milton C. Regan, Jr., Teaching Enron, 74 FORDHAM L. REV. 1139, 1154 (2005) [hereinafter Regan, Teaching Enron] (describing Enron’s decentralized structure where the general counsel of a business unit reported to the head of a business unit as well as James Derrick, Enron’s General Counsel); Compliance Readiness—General Counsel’s Expanded Role, METRO. CORP. COUNS., Sept. 2006, at 1 (describing WorldCom’s legal department as geographically dispersed and unclear lines of authority with at least one lawyer reporting directly to the CEO and not the General Counsel). An example of relatively decentralized in-house legal department is General Electric’s, where the general counsel of a particular business unit has a solid reporting line to the leader of the business unit and a dotted line to the General Counsel of General Electric (corporate). See Ben W. Heineman, Jr., Imagination at Work, AM. LAW., Apr. 2006, at 73, 74. 450 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 Compliance. As already discussed, inside counsel increasingly manage and monitor a company’s legal compliance: it is squarely part of their job. To ensure that employees carry out their corporate duties in ways that do not expose the company to unreasonable risks of criminal or civil liability, inside counsel’s duties have formally expanded to include training employees about potential liability (e.g., training about antitrust or sexual harassment liability), planning and design of corporate compliance programs (e.g., creating standard operating procedures and protocols for dealing with consumer product complaints), and monitoring ongoing compliance practices (i.e., receiving periodic reports and pursuing actions which pose the threat of corporate liability).204 In this capacity, inside counsel receive important information about a company’s non-compliance with rules and regulations. Even if a separate department handles compliance, inside counsel are typically consulted for legal advice. Anecdotal evidence from reported cases in which inside counsel have sued their employers for retaliatory discharge reveals that inside counsel do learn of unlawful conduct while monitoring their corporate compliance programs.205 External liaising. Finally, inside counsel act as the company’s external liaison in dealings with third parties.206 The general counsel is often the company’s corporate secretary, entrusted with the official records of the company and the authority to sign certain contracts, as well as its agent for service of process. More important, she and her in-house legal staff are the purchasing agents for the company’s outside legal services. In this capacity, inside counsel are armed with the authority to decide whether to retain outside counsel or handle an issue internally.207 Once outside counsel’s services are acquired, inside counsel act as the communicative interface between company employees and the outside firm. They often act as the gateway through which all information must pass. Outside counsel’s access. While inside counsel have access to multiple information flows mandated by their formal job duties, outside counsel’s access is far more restricted and fragmented. Access to facts can be understood in both quantity and quality terms. As a matter of quantity, most corporations now do much of their routine legal work in-house and farm out only large, especially complex or specialized matters.208 In the past decade, law firms’ share of 204. Gruner, supra note 186, at 1141-42. Antitrust law enforcement officers often remark that inside counsel do more than anyone else to ensure compliance with federal antitrust laws. I thank Warren Grimes for this comment. 205. See, e.g., Considine v. Compass, 551 S.E.2d 179, 181 (N.C. Ct. App. 2001) (plaintiff inside counsel discovered unlawful conduct of employer “in carrying out his duties regarding the compliance program”). 206. DeMott, Discrete Roles, supra note 7, at 970. 207. Robert Eli Rosen, The Inside Counsel Movement, Professional Judgment and Organizational Representation, 64 IND. L.J. 479, 484 (1989) [hereinafter Rosen, Inside Counsel Movement]. 208. George P. Baker & Rachel Parkin, The Changing Structure of the Legal Services Industry and the Careers of Lawyers, 84 N.C. L. REV. 1635, 1655 (2006). Litigation is the area in which inside counsel seek help from outside counsel most often. 2004 ASS’N OF CORP. COUNS. CENSUS OF IN-HOUSE COUNS., EXECUTIVE SUMMARY 2 (2004). 2008] GATEKEEPERS INSIDE OUT 451 companies’ transactional work appears to be steadily declining.209 Moreover, companies now use multiple firms, in a task-specific and ad hoc manner.210 That means that the work assigned—and the curious facts seen—by any single outside firm is even more limited.211 In terms of quality of information, the outside firm typically does not have access to the raw, unfiltered stream of first-hand reports from employees, managers, or witnesses.212 Rather, reports are often pre-processed and organized by their clients’ inside lawyers,213 who for all intents and purposes are the outside firm’s real “client.” Indeed, the facts are sometimes framed as mere “hypotheticals,”214 about which the outside firm is asked to render a cost-effective legal opinion. Although cost considerations primarily drive this practice, inside counsel may seek opinions on stylized facts to garner useful ammunition in their own attempts to steer the company’s course.215 For transactions, such as mergers and acquisitions, outside counsel’s involvement often occurs late in the game. By the time outside counsel are brought in, most of the key strategizing has 209. Steven L. Schwarcz, To Make or Buy: In-House Lawyering and Value Creation, 33 J. CORP. L. (forthcoming 2008) (manuscript dated Aug. 29, 2007, at 2, on file with author) [hereinafter Schwarcz, To Make or Buy]. See id. at 4-8 for methodological limitations and disclaimers, including low response rate and potential bias. 210. As Marc Galanter and Thomas Palay noted, there has been a fundamental shift in the nature of the relationship between law firms and clients, from “comprehensive and enduring retainer relationships toward less exclusive and more task-specific ad hoc engagements.” MARC GALANTER & THOMAS PALAY, TOURNAMENT OF LAWYERS: THE TRANSFORMATION OF THE BIG LAW FIRM, 50 (1991). Inside counsel routinely shop for the most capable (and cost-effective) lawyer for each project, occasionally holding “beauty contests” where law firms are forced to compete for the right to do the client’s legal work. David Wilkins, Do Clients Have Ethical Obligations to Lawyers? Some Lessons from the Diversity Wars, 11 GEO. J. LEGAL ETHICS 855, 884 (1998). Inside counsel have even auctioned off legal work over the Internet. Joel Henning, Bar Associations, Law Firms and other Medieval Guilds, 32 No. 1 Litigation 17, 19 (2005). 211. DeMott, Discrete Roles, supra note 7, at 971. See infra Part V (discussing the practice of multiple contracting, which seriously hinders a gatekeeper’s capacity to interdict). 212. See also Richard E. Moberley, Sarbanes-Oxley’s Structural Model to Encourage Corporate Whistleblowers, 2006 B.Y.U. L. REV. 1107, 1114-15 (noting the flow-of-information problems that external monitors encounter). 213. For instance, in response to the famous Sherron Watkins’ memo warning that Enron might implode in a “wave of accounting scandals,” General Counsel James Derrick had expressly limited the “preliminary” investigation by instructing outside law firm Vinson & Elkins (“V&E”) not to second guess Arthur Andersen’s treatment of underlying accounting issues, not to perform a detailed analysis of the transactions in question, and not to conduct a discovery-style investigation. It is thus unsurprising that V&E concluded to Enron’s management that no further scrutiny into Watkins’ allegations was necessary. Letter from Max Hendrick, III, Vinson & Elkins, to James V. Derrick, Jr., Executive Vice President & General Counsel, Enron Corp., Preliminary Investigation of Allegations of an Anonymous Employee, 2001 WL 1764266, Oct. 15, 2001 (“[T]he facts disclosed through our preliminary investigation do not, in our judgment, warrant a further widespread investigation by independent counsel and auditors.”). The report ackowledged, however, that “there is a serious risk of adverse publicity and litigation.” Id. 214. See Rosen, Inside Counsel Movement, supra note 207, at 485. See also ABA Standing Comm. on Ethics and Prof’l Responsibility, Formal Op. 346 (1982). 215. See Rosen, We’re All Consultants, supra note 179, at 649 (noting how managers will employ independent outside firms in their political battles within the company). 452 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 happened, the major deal points have been agreed upon, the client has already (although not irrevocably) committed to the transaction proceeding (which triggers alignment pressures),216 and managers turn their attention to “squeez[ing] the best possible deal from the other side.”217 The belated involvement by outside counsel constrains their access to key facts and opportunities to remonstrate with the client. Securities law developments. Finally, developments in the securities field have significantly foreclosed access by law firms to company facts. In the early 1980s, the SEC enacted deregulatory reforms that expedited the securities offering process. These reforms permitted the use of both “integrated disclosure”218 and “shelf offerings,”219 which enabled certain issuers to access the capital markets for certain types of transactions in a compressed time frame. These new procedures effectively suppressed the demand for law firms’ investigatory services traditionally used by underwriters to verify the accuracy of the issuer’s disclosures in connection with transactions registered under the Securities Act of 1933.220 As a result, apart from initial public offerings and certain registrations ineligible for shelf offerings, underwriters dispensed with the full-blown due diligence221 that law firms performed in the exercise of their gatekeeping function. Thus, the use of the law firm “as a reputational intermediary [became] a waning dynamic in corporate law practice.”222 A recent survey of companies who hire law firms confirms the declining use of law firms for their reputational value.223 In sum, as internal adviser, compliance overseer, and external liaison, inside counsel are well-positioned to access critical information that may reveal company misconduct. That information will be of high volume and, more importantly, less processed and impression managed than the information that outside counsel will be privy to. In addition, regulatory reforms that have expedited the securities offering process have further diminished outside 216. See supra Part II.B.3 (alignment pressures arising out of the faithful agent role). 217. Rosen, We’re All Consultants, supra note 179, at 652-53 (observing M&A best practices for redesigned companies in the information industry). 218. Integrated disclosure “permitted established issuers to use a streamlined prospectus that simply listed previously filed Exchange Act reports . . . which were thereby deemed to have been ‘incorporated by reference.’” COFFEE, GATEKEEPERS, supra note 9, at 205. 219. Shelf offerings permitted the registration of securities for sale for up to two years later. Id. at 206. 220. Id. at 202-03 (tracing the development of the “due diligence” investigation) and 206 (describing that the shelf registration process generated competitive pressures for underwriters to accept the prospect of increased Section 11 liability as a cost of doing business). 221. Id. at 206. See also Stephen J. Choi, James D. Cox & G. Mitu Gulati, Elite Lawyers, Shelf Transactions and the Failure of Due Diligence? An Empirical Inquiry (Sept. 13, 2007) (unpushlished manuscript, on file with author) (studying the quality of due diligence in shelf and non-shelf offerings). 222. Okamoto, supra note 57, at 19. 223. Schwarcz, To Make or Buy, supra note 209, at 34-37 (reporting that, except for transactions requiring certain types of third party legal opinions and especially complex matters which are handled by super-elite law firms, outside law firms are rarely selected for their reputational value). 2008] GATEKEEPERS INSIDE OUT 453 counsel’s access to facts that might unearth wrongdoing. B. INFORMAL COMMUNICATION CHANNELS If the gatekeeping literature underexamined capacity to monitor, it has entirely ignored the fact that access to information turns also on informal channels of communication. In contrast to formal exchanges, informal communication channels are less purposeful, less constrained, and less organized. They arise from accidental, everyday social interactions among employees of the company who share the same physical space. Because the inside lawyer is physically proximate to those who hold information about misconduct, she may learn of misdeeds during her trips to the local “water cooler.”224 Inside counsel’s access to back-channel information means that inside counsel may learn of misconduct even if such information never arises formally while executing various assigned functions. This is likely to happen where knowledge of misbehavior is widespread throughout the company. News accounts of the frauds at Enron and WorldCom confirm that many employees knew of their companies’ questionable accounting practices, although such knowledge was shielded from the public eye.225 Cynthia Cooper, the famed whistle-blower of WorldCom, learned of WorldCom’s accounting shenanigans not through her formal job responsibilities but informally from a colleague in the company’s wireless division.226 Anecdotal evidence from reported cases in which inside counsel have sued their employers for retaliatory discharge support the impor- 224. The “water cooler” is a metaphor that refers to the places and occasions for informal information interchange that occur in all employment settings: the company cafeteria, the car pool, the informal exchanges preceding company committee meetings, the evening get-togethers during out-of-town trips, and even the company picnics. These are opportunities for exchange of back-channel information, office gossip, rumors and portents of future corporate undertakings that have not yet been announced officially. Hazard, supra note 176, at 1018. 225. See, e.g., Jodie Morse & Amanda Bower, The Party Crasher, TIME (Dec. 30, 2002-Jan. 6, 2003), 53, 56 (reporting Sherron Watkins’ exclaiming “how many people at Enron stayed silent”); Rebecca Blumenstein & Susan Pulliam, Leading the News: WorldCom Fraud Was Widespread—Ebbers, Many Executives Conspired to Falsify Results in Late 1990s, Probes Find, WALL ST. J., June 10, 2003, at A3 (reporting that “while dozens of people knew about the fraud, it remained hidden from public view because employees were afraid to speak out”); Richard A. Oppel, Jr., Enron Official Says Many Knew About Shaky Company Finances, N.Y. TIMES, Feb. 15, 2002, at A1 (noting there was a “culture of intimidation at Enron where there was widespread knowledge of the company’s shaky finances”); Yochi J. Dreazen & Deborah Solomon, WorldCom Aide Conceded Flaws – Controller Said Company Was Forced to Disguise Expenses, Ignore Warnings, WALL ST. J., July 16, 2002, at A3 (reporting that lawmakers released documents in an attempt to bolster their case that WorldCom’s accounting irregularities aroused the suspicion of more employees than the company’s management had disclosed). See also Moberley, supra note 215, at 1116-17 (arguing that employees have an information advantage over traditional corporate monitors because they have greater knowledge of the inner workings of a large company and citing examples from recent scandals). 226. Amanda Ripley, The Night Detective, TIME, Dec. 30, 2002, at 45, 46 (describing how Cooper learned of the “rotten accounting” and noting that Cooper’s internal audit department handled operational audits, which set 454 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 tance of “accidental” encounters occasioned by physical proximity.227 Indeed, the “back channel” may be the most critical source of information. In contrast, outside counsel have almost no access to back-channel information and, as explained, are entirely dependent on inside counsel for information delivered formally for specific delimited and budgeted services. One might wonder whether friendships between outside lawyers and employees of the client (including inside lawyers) might develop, which could generate frequent casual information exchanges. Although certainly possible, such friendships are now less common, since. with the advent of the geographical dispersion of law firms,228 many outside lawyers are not even in the same city as the client representatives with whom they interact. Geographical dispersion also impedes outside lawyers’ ability to gather facts about the same client from their colleagues within the same firm. For example, it is now commonplace that within the very same law firm the lobbyist is in Washington, D.C., the franchise expert is in Chicago, and the banking specialist is in New York. None of these three are regularly gossiping, story sharing, and chit-chatting about their shared client.229 Admittedly, some corporate legal departments are also geographically dispersed, but in those cases, lawyers are nested within the relevant business unit and thus have even closer access to the relevant facts.230 Finally, given the high churn rate of associates and the increased lateral mobility of partners at law firms, any institutional knowledge that law firms happen to acquire quickly dissipates.231 In sum, inside counsel have an overwhelming advantage in their capacity to monitor via informal communication channels. company budget standards and evaluated performance, while financial audits were the province of Arthur Andersen). 227. See, e.g., Balla v. Gambro, 584 N.E.2d 104, 106 (Ill. 1991) (noting that plaintiff inside counsel was not informed by the president of his decision to import and distribute defective kidney dialyzers in contravention of FDA regulations but became aware of it through other company employees); Considine v. Compass, 551 S.E.2d 179, 181 (N.C. Ct. App. 2001) (reporting that when plaintiff inside counsel returned to his office, he found “the general counsel rifling through his desk in search of documents which would show the unlawful conduct of Defendant”); Alexander v. Tandem Staffing Solutions, Inc., 881 So. 2d 607, 608 (Fla. Dist. Ct. App. 2004) (reporting that two employees informed the plaintiff general counsel that a company officer was accessing sex-related websites on the internet; the general counsel initiated an investigation and was subsequently fired). 228. The well-documented growth of large law firms has been accompanied by large increases in the number of multi-office firms. Baker & Parkin, supra note 208, at 1662 (noting that an increase in the total number of law offices over time with a decrease in the number of law firms implies that the number of branches per firm must have increased on average). Neal Solomon, Economic Principles Drive Mergers among U.S. Firms Most Recent Law Firm Growth Has Occurred Due to Branching via M&A, NAT’L L. J., Sept. 26, 2005, at S2 (noting that much of the recent growth in large law firms is due to “branching”—carried out mainly through firm mergers). 229. Moreover, given the billing pressures, lawyers are incentivized to maximize their billable time when they are in the office, which may discourage non-billable, casual conversations. 230. In decentralized legal departments, however, information flowing to the general counsel may be impeded. See supra note 203 and accompanying text. 231. Schwarcz, To Make or Buy, supra note 209, at 26. 2008] GATEKEEPERS INSIDE OUT 455 C. OBJECTION: KNOWING THE LAW Skeptics may concede that inside counsel have an obvious edge in getting the facts but retort that outside counsel know the law better. They might argue that outside lawyers have higher-quality expertise because of their vast experience from handling many different transactions. While inside counsel were once perceived to be less competent than outside counsel,232 that gap has substantially eroded, as inside counsel are gaining in their reputations for handling complex legal matters.233 In the past decade, more transactional work has moved in-house, with one survey showing that approximately 68% of transactions are currently handled by inside lawyers.234 Skeptics may also argue that only outside counsel possess the requisite specialized expertise gained from advising a wide array of clients in different contexts to determine whether any law has potentially been broken.235 Even if expert specialization helps to detect fraud, it’s not that clear how much of a specialization advantage outside attorneys actually enjoy. While law firms have increasingly specialized in response to competition,236 inside counsel have followed suit. Corporate legal departments increasingly organize themselves according to substantive practice areas of law.237 In fact, the number of inside counsel specializing in securities law has more than doubled from 1975 to 1995.238 Also, reports suggest that, even for a fraud as unusually complex as Enron’s,239 inside counsel of Enron had no trouble fully comprehending the 232. Carl D. Liggio, The Changing Role of Corporate Counsel, 46 EMORY L.J. 1201, 1202-1203 (1997) (describing the decline in reputation of corporate counsel from the 1940s to the mid-1970s). 233. Schwarcz, To Make or Buy, supra note 209, at 20. 234. Id. at 2, 84 app. A at H.1. 235. For example, Deborah A. DeMott has posed the question of whether the lack of specialized expertise may have played some role in the inability of James Derrick, former general counsel of Enron, to have stopped the fraud at Enron. Derrick was a litigator by training, which may explain the “limited and episodic character of his involvement in transactional questions” which were at the heart of the Enron scandal. DeMott, Discrete Roles, supra note 7, at 978. 236. See HEINZ ET AL., supra note 108, at 37-38 (reporting that specialization has increased substantially from 1975 to 1995). This study found that the number of practitioners who professed to work in only one field increased from 23 percent in 1975 to 33 percent in 1995. By 1995, large law firms performed more than half of all securities practice. Id. at 101, 103 (noting that in 1995, even though firms with 100 or more lawyers employed only a quarter of the Chicago bar, they did more than 51% of the securities practice, as compared to 34% in 1975). 237. See id. at 129 (noting increased specialization among inside counsel and that the predominant organizing principle was substantive area of law). 238. The percentage of inside counsel who specialize in securities law more than doubled from 1975 to 1995, while the percentage of inside counsel specializing in business tax, corporate litigation and patent law has decreased. See id. at 102 (inside lawyers specializing in securities law increased from 12% in 1975 to 27% 1995). 239. It may be that the Enron fraud is not representative of frauds generally. As one investigator commented, “Every other white-collar case in history is arithmetic,” while “Enron is calculus.” Jeffrey Toobin, End Run at Enron, NEW YORKER, Oct. 27, 2003, at 48, 50. The complexity of Enron’s fraud is probably attributable to its high degree of reliance on derivative instruments. See Regan, Teaching Enron, supra note 203, at 1140, 1143-44 456 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 critical legal and accounting issues at hand.240 Moreover, this “knowing the law” objection assumes that it is hard to know when fraud is going on—that specialized expertise is really necessary. But crime detection typically doesn’t require an expert criminologist. In most cases, an ordinary cop with a good dose of “spider sense” will do. Similarly, fraud detection typically doesn’t require years of derivatives and structured finance experience.241 What is necessary for fraud detection is a developed situational sense that is only possible with understanding the client’s entire situation and not just some discrete facet of it.242 In this way, inside counsel are better situated. By virtue of inside counsel’s access to both formal and informal communication channels, inside counsel have the “capacity to synthesize, to integrate from a single point of view all the [relevant] considerations”243 necessary to understanding what is really going on. Unlike the outside lawyer, who is only privy to a limited range of facts that have been carefully sanitized and selected, the inside lawyer can perceive all the underlying motivations behind the actions of her colleagues, not just the ones that the manager insists are relevant to the narrow issue at hand. *** Compared to outside counsel, inside counsel have greater access to company facts about managerial misbehavior through both formal and informal channels. A recent survey of inside and outside lawyers confirms this view.244 This special access enables inside counsel to understand the client’s entire situation, which gives them the situational sense needed to detect fraud. This is not to say that the intelligence gathered by inside counsel will be perfect; indeed, much of it will be (describing how Enron began as a gas pipeline company with high reliance on physical assets into something more akin to an investment bank, utilizing derivatives–complex financial instruments whose prices were based on the underlying price of gas). 240. In an internal e-mail, describing a meeting with an Andersen employee in which these issues were raised, one V&E attorney wrote: I think that I am blamed by some of the inside Enron attorneys . . . for drawing this distinction [between a true issuance and true sale opinion] to AA’s attention, as it could jeopardize Enron’s FAS 125 transactions. The Enron theory is, apparently, that relations with AA must be carefully managed and that AA is a sophisticated organization that can read opinions and draw their own conclusion. Final Report of Neal Batson, Court-Appointed Examiner, app. C, at 47, n.169, In re Enron Corp., No. 01-16034 (Bankr. S.D.N.Y. Nov. 4, 2003). 241. Enron may indeed be the exception to this general rule. See supra note 239 for why the Enron fraud may not be representative of most frauds. But see Lawrence A. Cunningham, Sharing Accounting’s Burden: Business Lawyers in Enron’s Dark Shadows, 57 BUS. LAW. 1421, 1422-23 (2002) (urging business lawyers to become educated in accounting issues because “prevailing professional cultures create a crack between law and accounting that resolute fraud artists exploit”). 242. ANTHONY T. KRONMAN, THE LOST LAWYER: FAILING IDEAS OF THE LEGAL PROFESSION 289 (1993). 243. Id. at 289 (discussing the advantages of a generalist over a specialist). 244. Schwarcz, To Make or Buy, supra note 209, at 28 (reporting a consensus view that inside counsel are better suited to observe managerial misbehavior). 2008] GATEKEEPERS INSIDE OUT 457 hearsay, innuendo, and office gossip.245 However, given the more frequent informal encounters between inside counsel and other employees, exchanges with inside counsel will be less managed and sanitized than exchanges with outside counsel. Moreover, inside counsel will be exposed to dissenting viewpoints from within the company while outside counsel typically hear only one carefully managed voice. Inside counsel are therefore better able to judge the reliability of any single piece of information than outside counsel. If we step back still one more level on the comparison, we see that inside counsel enjoy a greater capacity to monitor (getting the facts) and perhaps even are more willing to do so, but that outside counsel remain more willing to interdict once they know a law has been broken V. CAPACITY TO INTERDICT Finally, what about the possibility of circumvention? Recall my “cop moonlighting as private security guard” example. A guard who is hired to help transport an undisclosed item across state lines—“no questions asked”—may lose business if she openly and freely displays a concern about the law. No doubt, a less vigilant contractor will be waiting in the wings to take her place. Quadrant IV discusses a particular gatekeeper’s capacity to interdict, which ultimately turns on the ease with which the wrongdoer can circumvent that gatekeeper.246 A gatekeeper will have a high capacity to interdict if she is indispensable to the 245. I do not want to overstate the case for inside counsel’s superior access to and knowledge of the facts. As noted by Donald Langevoort, much of the information gathered by lawyers remains largely derivative in nature—either from documentary evidence, e.g., reading files and reports, or from conversations with people with first-hand knowledge of the situation. Accordingly, all lawyers will initially adopt schemas, “mental roadmaps that people use to make sense of a situation,” that are based at least in part on these employees’ oral or written reports. Langevoort, Epistemology, supra note 160, at 636. For a perspective on outside lawyers’ ethical behavior in the litigation context, see Andrew M. Perlman, Unethical Obedience by Subordinate Lawyers: Lessons from Social Psychology, 36 HOFSTRA L. REV. (forthcoming 2008). 246. Here, my framework differs from that of Kraakman who sought to analyze three problems that gatekeeper regimes face in disrupting misconduct: (i) illicit markets, (ii) individual corruption, and (iii) multiple contracting. Kraakman, Gatekeepers, supra note 8, at 66. (“Illicit markets” is defined as “markets in proscribed goods or services.” Id. at 66.) My framework intentionally omits a detailed analysis of the problem of illicit markets (that the wrongdoer will evade interdiction by hiring lawyers who specialize in providing false legal opinions) because illicit markets do not pose a great regulatory obstacle for numerous reasons. See, e.g., id. at 66 (noting that would-be wrongdoers “cannot participate in too many scams without risking detection and dissipating their value to potential clients”); id. at 68 (noting that it is unlikely that lawyers will advertise illicit services); Peter J. Henning, The Sarbanes-Oxley Act of 2002: Sarbanes-Oxley Act 307 and Corporate Counsel: Who Better to Prevent Corporate Crime?, 8 BUFF. CRIM. L. REV. 323, 328 (2004) (“One can take some comfort in the fact that it is unlikely a group of lawyers will advertise themselves as ‘Counsel of Last Resort . . . .’”); Kraakman, Gatekeepers, supra note 8, at 68 (noting that to the extent that regulatory authorities gain leaked information about these “offense specialists,” it is relatively easy for them to detect and penalize); id. at 67 (noting that effective gatekeeping can coexist with illicit markets as long as access to these markets is limited). My framework also differs from Kraakman’s framework of gatekeeper problems in one other way. I do not discuss “individual corruption,” because analysis of that problem is subsumed into my discussion of “willingness to interdict” in Part II. Instead, I focus here on two circumvention scenarios that I think are the 458 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 wrongdoing client. One way that this might be so is if she can claim a monopoly on some specialized service that the client needs to commit its fraud. But lawyers are more or less fungible and, due to increased competition in the market for legal services, more so now than ever before. Thus, a lawyer’s capacity to interdict will depend less on a particular gatekeeper’s individual traits, such as her legal acumen or specialized expertise, and turn more on the extent to which the client’s efforts to circumvent her will signal wrongdoing to others. Wrongdoing clients can circumvent their gatekeepers by strategically contracting with numerous gatekeepers, either serially or simultaneously.247 With serial multiple contracting, the client may terminate one outside law firm and then search for a more compliant substitute. But the firing of any law firm is a red flag and the subsequent search may trigger information leaks that signal wrongdoing to law enforcement or other gatekeepers.248 Moreover, the law can intervene by imposing additional costs on the client’s ability to switch law firms.249 Take, for example, the SEC’s requirement that public companies disclose the termination of their auditing firm, which the SEC has not required with respect to law firms serving as outside securities compliance counsel.250 Such ancillary enforcement measures can have bite: in case of auditor terminations, the disclosure sends a negative signal to the market, which typically responds with a market penalty.251 But such hypothetical ancillary enforcement measures will be less effective if, as a regular practice, the client strategically contracts with other gatekeepers simultaneously by spreading work around rather than assigning it to a single lawyer or law firm.252 This practice not only cabins the information that each lawyer is exposed to, thereby constraining her capacity to monitor, but also blurs the information signal associated with any single lawyer’s or law firm’s termination. Firing one law firm among a dozen on routine grounds such as excessive fees or poor responsiveness signals nothing. Inside counsel can likewise be circumvented, again through serial or simulta- greatest threats to lawyer gatekeeper regimes: multiple contracting with members of the legal profession and multiple contracting with players outside of the legal profession. 247. Kraakman, Gatekeepers, supra note 8, at 73. 248. Id. at 73 (suspecting that reputable law firms are probably dismissed for resisting managerial practices and that the practice of serial multiple contracting poses the risk of signaling wrongdoing to outsiders). 249. The notion of “switching costs” is discussed in Gilson, Devolution, supra note 10, at 897 et seq. 250. See Jeffrey N. Gordon, What Enron Means for the Management and Control of the Modern Business Corporation: Some Initial Reflections, 69 U. CHI. L. REV. 1233, 1237 n.11 (2002) (noting that auditor firing is a “high visibility sanction” that may ultimately cause more harm to officers and directors than to the accountants). 251. See James A. Yardley et al., Supplier Behavior in the U.S. Audit Market, 11 J. ACCT. LITERATURE 151 (1992) (noting that four studies report significant negative market reactions to auditor switches following disclosure of disagreements with auditors and/or receipt of a qualified opinion). However, it may be that the practice of publicizing terminations may have the unintended consequence of normalizing an otherwise disdainful practice if, for example, a large number of companies switch counsel. Common knowledge of the prevalence of this practice may work to reduce the stigma associated with switching gatekeepers. 252. Fred C. Zacharias, Coercing Clients: Can Lawyer Gatekeeper Rules Work?, 47 B.C. L. REV. 455, 468 (2006). 2008] GATEKEEPERS INSIDE OUT 459 neous multiple contracting. But if the client terminates a non-compliant general counsel and then replaces her with a pliable one, enforcement agencies could catch wind of the serial terminations. Indeed, in today’s sensitized corporate environment, the unexplained termination of a general counsel of a large public corporation is likely to raise a brow with enforcement authorities and perhaps even the markets.253 Moreover, as with outside counsel, the law can intervene to raise the client’s switching costs, by, e.g., requiring public disclosure of general counsel terminations or providing inside counsel the clear right to sue their employers for retaliatory discharge under federal statutes.254 With respect to simultaneous multiple contracting, the client might simply circumvent inside counsel by directly soliciting the services of outside counsel,255 or—if the company has a large corporate department–it might cherry-pick among the various inside lawyers, some of whom may have reputations for being more “service-oriented” or better “team players” than others.256 But the inside lawyer, through her superior capacity to monitor via informal information channels, will often learn of such gaming and thus may be tougher to circumvent. Here, for the inside counsel, her greater capacity to interdict relies, in part, on her superior capacity to monitor. Put another way, because she can detect clumsy attempts at circumvention, she can raise the costs of the client’s attempt to do so. Of course, another possible circumvention scenario is that the client will successfully evade interdiction by legal gatekeepers by resorting to the legal profession’s most serious competitors—multi-disciplinary practices that provide legal services (MDPs)—to close their deals or generate their public disclosures.257 Without opining on the imminence or likelihood of this development,258 I think that, due to cost considerations associated with outsourcing, MDPs are more likely to displace outside counsel than inside counsel. In addition, for all the 253. See DeMott, Discrete Roles, supra note 7, at 968 (noting that the “General counsel’s withdrawal . . . may send a louder signal to audiences both internal and external to the corporation”). 254. Sarbanes-Oxley sections 806 and 1107 are the civil and criminal anti-retaliation (also called “whistle-blower protection”) provisions. See Kim, supra note 37, at 1064-71 (discussing the right to sue for retaliatory discharge for inside counsel). 255. See, e.g., Ashby Jones, Silicon Valley’s Outsiders: In-House Lawyers?, WALL ST. J., Oct. 2, 2006 at B3 (noting that in the Silicon Valley it is not uncommon for in-house lawyers to play second fiddle to outside lawyers). 256. Nelson & Nielsen, supra note 148, at 470 (reporting the practice of “forum shopping” within legal departments to circumvent inside lawyers who play the role of “cops”). 257. For a description of the various types of MDPs, see Bryant G. Garth & Carole Silver, The MDP Challenge in the Context of Globalization, 52 CASE W. RES. L. REV. 903 (2002). 258. Compare Schwarcz, Transactional Lawyering, supra note 64, at 30 (reporting that few respondents in the survey felt that the work of a transactional lawyer could be performed by non-lawyers to any significant extent), with Ronald J. Gilson, Value Creation by Business Lawyers: Legal Skills and Asset Pricing, 94 YALE L.J. 239, 301 (1984) (arguing that transactional lawyers’ value lies not in the inherently legal character of their work and predicting the possibility that the transactional lawyer’s role could be reduced from transaction cost engineer to draftsman due to competition from other professionals, such as investment bankers and accountants). 460 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 same reasons stated above, inside counsel will be tougher to circumvent than outside counsel. Suffice it to say that while the displacement of outside counsel by MDPs will be bad for the legal profession and its (disputed) hegemony259 in advising on corporate matters, it is less clear whether it would bad for law enforcement or society generally.260 VI. CONCLUSION Is the hallmark of a lawyer—independent judgment—blurred because the lawyer serves as inside counsel? Or is inside counsel better able to bring independent judgment to a corporation’s problems, even perhaps to go beyond the law to activate the corporate conscience? Brian D. Forrow, former general counsel, Allied Chemical Corporation261 The conventional wisdom, which draws almost exclusively from RCT and heavily relies on the role of market gatekeepers and reputational constraints, contends that outside lawyers are overwhelmingly superior for the gatekeeping function. But a more careful analysis that distinguishes capacity from willingness, monitoring from interdiction, conscious cost benefit analysis from pervasive situational forces when applied to a more up-to-date empirical snapshot of law firms reveals a far more complicated picture. In short, inside lawyers have an overwhelming advantage, through both formal and informal communication channels, in getting the facts. Also, inside lawyers may well be tougher to circumvent. In contrast, outside lawyers generally have limited capacity to monitor and will have difficulty in getting the facts that might serve as the critical red flags. This is not to say inside counsel are the clear 259. It may very well be that the legal profession has already lost its commanding position as primary adviser to corporations to investment bankers or perhaps even management consultants. For a recent history of management consulting, see CHRISTOPHER D. MCKENNA, THE WORLD’S NEWEST PROFESSION (2006). 260. As noted by eminent legal ethics scholar Deborah Rhode, “[L]awyers have been implicated in almost all of the major health, safety, and financial scandals of recent decades.” Deborah L. Rhode, Moral Counseling, 75 FORDHAM L. REV. 1317, 1320 (2006). Thus, it is not clear that lawyers as a class necessarily make better gatekeepers than other professionals, some of whom have stricter gatekeeper duties than lawyers (for example, accountants have Section 10A “reporting out” obligations, investment bankers and accountants are subject to liability under Section 11 of the Securities Act of 1933). And, at least in one narrow sense, lawyers may be worse candidates. To the extent that any other non-lawyer professional is used to do the work traditionally done by lawyers, such professional cannot claim the attorney-client privilege as lawyers may. Ironically, a greater reliance on non-lawyers for corporate transactions may actually facilitate, not hinder, law enforcement efforts due to the lack of the attorney-client privilege. For an insightful analysis of the disputes surrounding the attorney-client privilege in the context of corporate malfeasance, see, for example, Michael L. Seigel, Corporate America Fights Back: The Battle Over Waiver of the Attorney-Client Privilege, 49 B.C. L. REV. 1 (2008). 261. Brian D. Forrow, The Corporate Law Department Lawyer: Counsel to the Entity, 34 BUS. LAW. 1797 (1979). 2008] GATEKEEPERS INSIDE OUT 461 winners or even that they will, on net, be effective at gatekeeping. My analysis also reveals that inside counsel may be seriously compromised in their willingness to interdict (once misconduct is suspected) by virtue of their multiple roles as mere employee, team player and faithful agent. In this regard, outside counsel’s willingness to interdict may be better, although—as my analysis reveals—recent developments are certain to have diminished that. These gained insights cast doubt on the effectiveness of reforms that chiefly target behavior falling under Quadrant I (the willingness to interdict), which has been the traditional domain of gatekeeping theorists. Any contemplated reform relying on cross-checking gatekeepers should carefully consider each gatekeeper’s strengths and weaknesses with respect to each of the Four Quadrants of Gatekeeping. Policy makers may consider pursuing either or both of the following strategies: (1) they can capitalize on the existing strengths of each class of gatekeeper or (2) they can compensate for such gatekeeper’s weaknesses. As an example of the “capitalizing-on-strengths” strategy, policy makers may want to exploit inside counsel’s superior monitoring capacity by imposing more explicit and routine monitoring duties in the specific service of gatekeeping. As an example of the “compensating-for-weaknesses” strategy, since outside counsel are hindered in their ability to gather facts, policy makers may consider measures to ensure outside counsel’s access to information. This can be done by either formally carving into the public reporting process meaningful gatekeeping occasions on which outside counsel is required to solicit facts262 or incentivizing companies to retain outside counsel on a more ongoing basis, in the role of standing counsel for the independent directors.263 To compensate for inside counsel’s weaknesses, policy makers may consider what, if anything, should be done to neutralize the formidable obedience pressures264 by, for example, providing inside counsel a clear right to sue for retaliatory discharge under federal law. Alternatively, companies may consider modifying the best practices relating to inside counsel’s reporting relationships to more closely align inside counsel to the corporate constituent that best represents the corporation’s interests.265 262. See, e.g., John C. Coffee, Can Lawyers Wear Blinders? Gatekeepers and Third-Party Opinions, 84 TEX. L. REV. 59, 66-67 (2005) (proposing that law firms be required to certify for securities transactions (via a legal opinion addressed to third parties) the absence of any material misrepresentation or omission and give affirmative assurance that in their judgment the transaction has a legitimate business purpose, thus creating an affirmative duty of inquiry into the transaction’s purpose). 263. See, e.g., Geoffrey C. Hazard, Jr. & Edward B. Rock, A New Player in the Boardroom: The Emergence of the Independent Directors’ Counsel, 59 BUS. LAW. 1389 (2004) (predicting that the current regulatory environment will create the demand for counsel to the independent directors to be engaged on a standing basis). 264. See supra Part II.B.1 (discussing obedience pressures arising out of inside counsel’s role as “mere employee”). 265. See Kim, supra note 37, at 1053-75 (proposing a number of reforms that may counteract some of inside counsel’s obedience pressures, including guaranteeing protection to employed lawyers under SOX’s civil and criminal anti-retaliation provisions, sections 806 and 1107, and having the general counsel report to the 462 THE GEORGETOWN JOURNAL OF LEGAL ETHICS [Vol. 21:411 To be clear, my analysis does not presume that legal intervention is desireable,266 as this article’s primary purpose is diagnostic rather than programmatic. Of course, any programmatic reform must carefully consider the unintended consequences of any particular measure,267 when comparing the costs and benefits of that measure. For example, former Chief Justice of the Delaware Supreme Court E. Norman Veasey argues that the proposal of retaining counsel for independent directors on a standing basis will not pass the cost-benefit analysis and that the “heavy lifting” in corporate governance should generally remain with the general counsel and her in-house staff.268 More broadly, my analysis urges reconsideration of gatekeeping theory’s overwhelming reliance on outsiders generally—the market gatekeepers who were originally posited as being ideal. My research suggests that it is no longer the case that outside market gatekeeping firms are the self-evident choice for the company’s gatekeeping function. Accordingly, this article calls for an inquiry into the other two essential elements of a gatekeeper enforcement regime: (1) the “gate”—some service that the wrongdoer needs to accomplish his goal, and (2) the law enforcement “mechanism”—a legally enforceable duty imposed on the gatekeeper in furtherance of averting misconduct. If we shift our focus away from market gatekeepers, we see that the gate need not be a discrete service traditionally provided by market gatekeepers, such as a third-party legal opinion, and the specific mechanism need not be the gatekeeper’s traditional duty to withhold services, which theoretically provided the potential signal to the independent board or subcommittee of the board in an effort to better reflect general counsel’s fiduciary duties). Also, to the extent that inside counsel embedded in a business unit report directly to the managerial head of the business unit, rather than the general counsel, best practices should be revised to ensure that inside counsel have a direct report to the company-wide general counsel, as a way of facilitating the reporting of negative information about senior managers up to the company-wide general counsel. 266. See Jeffrey J. Rachlinski, The Uncertain Psychological Case for Paternalism, 97 NW. U. L. REV. 1165, 1168 (2003) (arguing that the “psychological case for paternalism . . . must rest on a relative assessment of the cognitive costs of improved decision against the costs of supplanting individual choice”). 267. See, e.g., supra Part III (noting that solutions intended to enhance the gatekeeper’s willingness to interdict—for example, harsh sanctions imposed on the gatekeeper who has “knowledge” but nonetheless failed to interdict—may perversely disincentivize monitoring); Kraakman, Gatekeepers, supra note 10, at 77 (noting that “costly gatekeeping duties might also induce regulatory targets to alter their contracting behavior in ways that erode the enforcement value of gatekeeping and may even increase the incidence of misconduct”); Donald C. Langevoort, The Human Nature of Corporate Boards: Law, Norms, and the Unintended Consequences of Independence and Accountability, 89 GEO. L.J. 797, 823-825 (2001) (suggesting that increasing lawyers’ leverage over clients may exacerbate lawyers’ tendency to overstate legal risk and thus gain at the unwitting client’s expense); Lawrence A. Cunningham, Carrots for Vetogates: Incentive Systems to Promote Capital Market Gatekeeper Effectiveness, 92 MINN. L. REV. (forthcoming 2007) (manuscript at 26, on file with author) (criticizing a sticks-based approach to gatekeeper enforcement as encouraging gatekeepers’ self-interested campaigns to limit or eliminate the scope of their undertakings). 268. E. Norman Veasey, Separate and Continuing Counsel for Independent Directors: An Idea Whose Time Has Not Come as a General Practice, 59 BUS. LAW. 1413, 1414, 1418 (2004). Veasey also argues that the “board must have a voice in the selection and retention of the general counsel and must have complete and unfettered access to the legal advice and legal services of the general counsel.” Id. 2008] GATEKEEPERS INSIDE OUT 463 securities market.269 My inquiry argues in favor of broadening the parameters for conceiving the gatekeeping function and calls for boards of directors to more thoughtfully assess their alternatives in their decisions to appoint gatekeepers. Instead of buying them, perhaps they should be made.270 In fact, they may already be embedded within the firm, in the form of inside counsel, awaiting the opportunity to activate the corporate conscience. 269. Under Kraakman’s framework, the gatekeeper’s duty to interdict would, at minimum, require the gatekeeper to withhold its services once it detects that its services will be misused. But Kraakman’s duty to withhold services is grounded in gatekeeping theory’s overwhelming reliance on market gatekeepers, whose duties are specifically tailored to how the market gatekeeper might facilitate the market’s evaluation of the client. See supra Part II.A.1.a (discussing market gatekeepers). Since my framework can be applied to evaluate the potential of any gatekeeper—market or public gatekeepers, it is not constrained by the traditional duty to withhold services, which was originally intended for market gatekeepers. Accordingly, it is not necessary to adopt Kraakman’s specific duty to withhold services. In fact, Sarbanes-Oxley § 307 merely imposes the duty to report evidence of a law violation “up-the-ladder.” 270. The decision whether to rely on inside versus outside counsel (or both) for the gatekeeping function is just a “subset of the broader question of vertical integration: whether a company should make needed products in-house, or whether it should buy them (the classic ‘make-or-buy’ decision).” Schwarcz, To Make or Buy, supra note 209, at 4.