Emory Law Journal

Litigation Discovery and Corporate Governance: The Missing Story About the “Genius of American Corporate Law”
Érica Gorga,
Michael Halberstam Érica Gorga, Ph.D., is Associate Research Scholar in Law and the John R. Raben/Sullivan & Cromwell Executive Director for the Yale Law School Center for the Study of Corporate Law, Professor of Law at Fundação Getulio Vargas School of Law at São Paulo (FGV Direito SP), erica.gorga@fgv.br.Michael Halberstam, J.D., Ph.D., is Associate Professor S.U.N.Y. Law School University at Buffalo, mhalbers@buffalo.edu.The authors wish to thank Guyora Binder, Margaret Blair, Michael Boucai, Kate Brown, Jens Dammann, Matthew Dimick, Brian Fitzpatrick, Merritt Fox, Mira Ganor, Jeffrey Gordon, Zohar Goshen, Peter Gruber, Bert Huang, Robert Jackson Jr., Dan Kramer, Susan Mangold, Curtis Milhaupt, Donna Nagy, Katharina Pistor, Jayanth K. Krishnan, Morgan Ricks, Roberta Romano, Judge Ruth Ronen, Amanda Rose, Jack Schlegel, Jeffrey Schoenblum, Suzanna Sherry, Michael Simkovic, Matt Steilen, Rick Su, Randall Thomas, Kenju Watanabe, Bert Westbrook, James Wootin, and the participants of the 23rd Annual Meeting of the American Law and Economics Association (ALEA) at Vanderbilt University, the 29th Annual Conference of the European Association of Law and Economics (EALE) at Stockholm University, the 16th Annual Conference of the International Society for New Institutional Economics (ISNIE) at University of Southern California Gould School of Law, the 2012 Meetings of the Canadian Law and Economics Association, the Columbia Law School Fundacao Getulio Vargas Global Alliance Workshop held in 2012 at Columbia Law School, the Center for Law and Global Affairs Workshop on International and Transnational Law held at Sandra Day O’Connor College of Law, Arizona State University, Faculty Workshops held at Vanderbilt Law School and SUNY Buffalo Law School, for their insightful thoughts and comments on this project. Michael Halberstam is grateful to Columbia Law School, its Law and Economic Studies Program, and especially to Jeffrey Gordon and Dean Shizer for their generous support of this research. Érica Gorga is especially grateful to Roberta Romano for critiques and suggestions and for the Yale Law School Center for the Study of Corporate Law for research support during the finalization of this project. Finally, we thank Joel Langdon and William Bradbury, at the Emory Law Journal, for all the work they put into editing this Article.

Abstract

Strikingly absent from the entire corporate governance and corporate litigation debate is a unique feature of American civil procedure that deserves special attention: the modern civil discovery regime. This Article attempts to fill this gap. We argue that modern discovery—first established by the Federal Rules of Civil Procedure in 1938—has had a profound impact on the evolution of shareholder litigation, corporate governance, and the culture of corporate disclosure in the United States.

This Article shows that (1) litigation discovery, and its threat, have driven and structured the process of corporate shareholder litigation; (2) the information generated by discovery has stimulated the development of case law defining shareholder rights and managerial duties; (3) the episodic legal demands for detailed corporate internal information (and the threat of discovery) have induced incremental improvements in corporate governance practices, including more exacting decision procedures, internal monitoring, recordkeeping, and disclosure; (4) highly developed, continuously evolving discovery practices have established templates for independent corporate internal investigations by boards and regulators; and (5) discovery has given regulators steady insight into changing corporate internal practices and patterns of wrongdoing to which regulators have responded with broad legal and regulatory changes. This Article concludes that litigation discovery serves, inter alia, as a form of ex post disclosure, which complements and enforces ex ante disclosure under the federal securities laws.

These observations have important normative implications for legal transplants and the enforcement debate. Among other things, this Article cautions against legal transplants of U.S.-style securities disclosure, aggregate litigation mechanisms, and other enforcement mechanisms without considering appropriate tools for investigating corporate internal wrongdoing ex post, and points to problems in the empirical literature on U.S. shareholder litigation outcomes. It also questions current proposed reforms to the federal discovery rules.

“As stated by the Supreme Court in Hickman v. Taylor, ‘Mutual knowledge of all the relevant facts gathered by both parties is essential to proper litigation.’ Such, however, is not the basis of procedural policy in any other country in the world.”

Geoffrey C. Hazard, Jr. 1Geoffrey C. Hazard, Jr., From Whom No Secrets Are Hid, 76 Tex. L. Rev. 1665, 1665 (1998) (footnotes omitted) (quoting Hickman v. Taylor, 329 U.S. 495, 507 (1947)).

Introduction

Professor Romano’s seminal work, The Genius of American Corporate Law, inspired an international debate on corporate governance and influenced the research agendas of law, economics, and finance scholars for over two decades. Romano’s book offered an explanation for the curious American exceptionalism in financial markets development based on the counterintuitive theory that the competition among states for incorporations generated a “race to the top” in state legal regimes governing corporate internal affairs. 2Roberta Romano, The Genius of American Corporate Law 14–15, 118 (1993). The dominant view was, and in many areas of law still is, that regulatory competition among jurisdictions creates a “race to the bottom” in regulatory standards. The “race to the bottom” thesis in corporate law was advanced by Adolph A. Berle, Jr. and Gardiner C. Means in The Modern Corporation and Private Property (1932), and was adopted by Justice Louis Brandeis in Louis K. Liggett Co. v. Lee, 288 U.S. 517, 557–60 (1933) (Brandeis, J., dissenting). She famously argued that the creation of a market for legal rules allowed firms to select corporation codes with better corporate governance devices to mitigate agency costs. 3Romano, supra note 2, at 1–2. The genius of American corporate law, according to Professor Romano, relied on the consequences of federalism for the evolution of American corporate and securities laws while much less successful institutional frameworks were implemented in other countries. 4Id.

Ever since Romano first articulated her theory of the genius of American corporate law, researchers have attempted to analyze, identify, and compare across national legal regimes and economies precisely what legal variables have been responsible for resolving (or failing to resolve, as it were) the agency-cost problems between shareholders and managers inherent in the structure of large public firms. This comparative enterprise has been highly consequential in that the legal variables so identified have, in turn, been deemed “preconditions” to the highly developed capital markets in the United States that other nations across Europe, Asia, and Latin America ought to emulate. 5See Curtis J. Milhaupt & Katharina Pistor, Law and Capitalism: What Corporate Crises Reveal About Legal Systems and Economic Development Around the World 17–20 (2008); Bernard Black et al., Corporate Governance in Korea at the Millennium: Enhancing International Competitiveness, 26 J. Corp. L. 537, 539–40 (2001) (noting Asian recognition of the contribution of corporate law to economic prosperity); 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, 1234 (2002) (arguing that “the core beliefs and practices that have underpinned the academic analysis of corporate law and governance . . . amount to an interlocking set of institutions that constitute ‘shareholder capitalism,’ American-style that [U.S. academics] have been aggressively promoting throughout the world”).

By now the corporate governance debate has, with mixed success, explained national differences as depending variously on corporate ownership structures, 6See Rafael La Porta et al., Corporate Ownership Around the World, 54 J. Fin. 471, 471–73, 511 (1999). minority shareholder protections, 7See Rafael La Porta et al., Investor Protection and Corporate Governance, 58 J. Fin. Econ. 3, 4 (2000) (“[T]he protection of shareholders and creditors by the legal system is central to understanding the patterns of corporate finance in different countries.”). stock exchange rules, 8See John C. Coffee, Jr., The Rise of Dispersed Ownership: The Roles of Law and the State in the Separation of Ownership and Control, 111 Yale L.J. 1, 8–9, 76 (2001). mandatory disclosure regimes, and liability standards, 9See Rafael La Porta et al., What Works in Securities Laws?, 61 J. Fin. 1, 2, 27–28 (2006). as well as much broader explanations, including differences in social and cultural norms, 10See John C. Coffee, Jr., Do Norms Matter? A Cross-Country Evaluation, 149 U. Pa. L. Rev. 2151, 2151–52, 2171–75 (2001); Mark J. Roe, Can Culture Constrain the Economic Model of Corporate Law?, 69 U. Chi. L. Rev. 1251, 1262–64 (2002). political histories, 11See Mark J. Roe, Political Determinants of Corporate Governance: Political Context, Corporate Impact 1–5, 49–51, 201–04 (2003); Mark J. Roe, Political Preconditions to Separating Ownership from Corporate Control, 53 Stan. L. Rev. 539, 542–43 (2000). historic legal origins of national legal systems, 12See Edward L. Glaeser & Andrei Shleifer, Legal Origins, 117 Q.J. Econ. 1193, 1194 (2002); Rafael La Porta et al., Law and Finance, 106 J. Pol. Econ. 1113, 1116 (1998); Rafael La Porta et al., Legal Determinants of External Finance, 52 J. Fin. 1131, 1131–33, 1149 (1997). Contra Holger Spamann, The “Antidirector Rights Index” Revisited, 23 Rev. Fin. Stud. 467, 467–68, 477 (2010); Holger Spamann, On the Insignificance and/or Endogeneity of La Porta et al.’s ‘Antidirector Rights Index’ Under Consistent Coding 69–71 (European Corporate Governance Inst., Law Working Paper No. 67, 2006), available at http://ssrn.com/abstract=894301. and, more recently, varying levels of “intensity” in the enforcement of corporate and securities laws. 13See, e.g., John C. Coffee, Jr., Law and the Market: The Impact of Enforcement, 156 U. Pa. L. Rev. 229, 232–33 (2007); Howell E. Jackson & Mark J. Roe, Public and Private Enforcement of Securities Laws: Resource-Based Evidence, 93 J. Fin. Econ. 207, 208–10, 230–33 (2009); La Porta et al., supra note 9, at 2–3; see also John Armour et al., Private Enforcement of Corporate Law: An Empirical Comparison of the United Kingdom and the United States, 6 J. Empirical Legal Stud. 687, 688–90 (2009); Mathias M. Siems, Private Enforcement of Directors’ Duties: Derivative Actions as a Global Phenomenon, in Collective Actions: Enhancing Access to Justice and Reconciling Multilayer Interests? 93, 93–96 (Stefan Wrbka et al. eds., 2012).

The enforcement literature has studied differences in levels of public and private enforcement. 14See Coffee, supra note 13, at 266–67 (explaining that “private enforcement” refers to the enforcement of corporate and securities statutes and laws by plaintiffs who are private parties); Jackson & Roe, supra note 13, at 237; La Porta et al., supra note 9, at 20–21, 27–28. A civil action commenced against an officer of a corporation by the Securities and Exchange Commission is an example of public enforcement. See La Porta et al., supra note 9, at 2–3. It has also revisited various aspects of derivative and class action litigation from a more comparative perspective with the express intent of paying closer attention to procedure. Professor Bernard Black and coauthors, for example, surveyed procedural rules that affect the adjudication of directorial and managerial liability claims in the United States, Russia, France, Germany, Austria, Korea, and the United Kingdom. 15Bernard Black et al., Legal Liability of Directors and Company Officials Part 2: Court Procedures, Indemnification and Insurance, and Administrative and Criminal Liability (Report to the Russian Securities Agency), 2008 Colum. Bus. L. Rev. 1, 2–4; see also Armour et al., supra note 13, at 689 (providing a quantitative comparison of private enforcement lawsuit outcomes in the United States and United Kingdom). But Black and coauthors mostly confine their survey to analyzing who can file a claim, what the available filing mechanisms are, and how attorneys’ fees and litigation costs are allocated among the litigants. 16See Black et al., supra note 15, at 1–13. In a separate study on private enforcement, Professor Coffee argues that U.S. aggregate litigation rules establish a “unique” system of “entrepreneurial litigation.” 17John C. Coffee, Jr., Litigation Governance: Taking Accountability Seriously, 110 Colum. L. Rev. 288, 291–92 (2010) (internal quotation marks omitted). In particular, the “intensity” of such enforcement by “private attorneys general” is generally considered to be a distinctive feature of the U.S. system. 18Id. at 344–45. But Coffee, like the rest of the literature, confines his analysis to a relatively narrow discussion of class action rules and the incentives created in the class action context by the American rules on fee shifting and contingency fees. 19See id. at 292–93. Professor Merritt Fox attributes the limited development toward class action litigation in Korea, Sweden, France, Spain, Germany, Norway, and the other countries to “some mix of the absence of contingent fees for plaintiffs lawyers, the existence of a ‘loser pays’ rule concerning the victor’s legal fees, an ‘opt in’ rather than an ‘opt out’ structure of class action, and the lack of explicit reference to securities fraud as being among the claims that can be brought collectively.” Merritt B. Fox, Civil Liability and Mandatory Disclosure, 109 Colum. L. Rev. 237, 251 (2009).

Similarly, academic and jurisprudential debates in the corporate law literature about exporting U.S.-style aggregate litigation to European and other countries have concentrated on such rules as shareholder standing requirements, fee shifting, contingency fees, and class action rules. But the enforcement debate has ignored how private litigants might obtain information about corporate internal wrongdoing in civil law systems in Europe, Asia, or Latin America. 20See, e.g., The Law and Economics of Class Actions in Europe: Lessons from America (Jürgen G. Backhaus, Alberto Cassone & Giovanni B. Ramello eds., 2012) (referencing discovery sporadically, but never discussing or comparing civil law judicial fact investigation with party-on-party discovery in the United States nor noting the complete absence of U.S.-style discovery in all other jurisdictions); Coffee, supra note 13, at 288 (noting that “the contemporary debate in Europe centers on whether certain elements of the U.S. model—namely, opt-out class actions, contingent fees, and the ‘American rule’ on fee shifting—must be adopted in order to assure access to justice,” and offering “an alternative ‘nonentrepreneurial model’ for aggregate litigation that is consistent with European traditions”).

The extensive (twenty-year) debate about what legal variables are conducive to private and public enforcement, and therefore to corporate governance and capital markets development, has ignored the distinctive features of U.S. fact investigation—even where civil procedure has been a focus. 21 See Simeon Djankov et al., Courts, 118 Q.J. Econ. 453, 454–55 (2003). But see Holger Spamann, Legal Origin, Civil Procedure, and the Quality of Contract Enforcement, 166 J. Institutional & Theoretical Econ. 149, 149–50, 162 (2010) (examining the claim that civil procedure is less efficient in civil law countries than in common law countries and shedding doubt on the claim). One notable exception is the inclusion of a variable measuring rights of “access to evidence” in the “anti-self-dealing index” by financial economists Simeon Djankov et al. in The Law and Economics of Self-Dealing, 88 J. Fin. Econ. 430, 434 (2008). Their index considers whether plaintiffs can request the court to appoint an inspector to examine a company’s affairs, whether plaintiffs must identify specific documents before receiving any discovery, whether they can question defendants and nonparties in court, and to what extent they can do so without prior court approval. Id. at 434, 437. Surprisingly, however, the legal literature has failed to pick up on and analyze these variables. Strikingly absent from this entire literature is any analysis of a truly exceptional feature of the U.S. legal landscape: The United States is unique in how it structures the process of litigation and, in particular, the process of information and evidence acquisition by parties in private shareholder litigation. 22Stephen N. Subrin, Discovery in Global Perspective: Are We Nuts? 52 DePaul L. Rev. 299, 301–03 (2002); see Geoffrey C. Hazard, Jr., Discovery and the Role of the Judge in Civil Law Jurisdictions, 73 Notre Dame L. Rev. 1017, 1017 (1998) [hereinafter Hazard, Discovery and the Role of the Judge]; Geoffrey C. Hazard, Jr., Discovery Vices and Trans-substantive Virtues in the Federal Rules of Civil Procedure, 137 U. Pa. L. Rev. 2237, 2237–39 (1989) (discussing the broad access to documents provided by the Federal Rules of Civil Procedure).

There is no equivalent to U.S. discovery in civil law jurisdictions, 23See generally, R.R. Verkerk, Fact-Finding in Civil Litigation: A Comparative Perspective 142 (2010) (providing an in-depth treatment of the historical evolution of principles and practices of fact investigation in different legal systems); Oscar G. Chase, American “Exceptionalism” and Comparative Procedure, 50 Am. J. Comp. L. 277, 292–96 (2002); John H. Langbein, The German Advantage in Civil Procedure, 52 U. Chi. L. Rev. 823, 824 (1985) (explaining that “by assigning judges rather than lawyers to investigate the facts, the Germans avoid the most troublesome aspects of [American] practice”); Richard L. Marcus, Putting American Procedural Exceptionalism into a Globalized Context, 53 Am. J. Comp. L. 709, 732 (2005) (book review); Subrin, supra note 22, at 301–02 (explaining that “two of the biggest differences between civil law countries and the United States with respect to pretrial discovery are the centrality of the judge in civilian civil litigation and the continuity of the proceedings”); Yasukei Taniguchi, The 1996 Code of Civil Procedure of Japan—A Procedure for the Coming Century?, 45 Am. J. Comp. L. 767, 772–76 (1997) (describing very limited changes to fact investigation in Japan’s new civil code). See generally Oscar G. Chase et al., Civil Litigation in Comparative Context 207–40 (2007) (comparing the discovery mechanisms in England, the United States, Germany, and Japan). and common law jurisdictions also differ dramatically from the U.S. model. 24See Neil Andrews, English Civil Procedure: Fundamentals of the New Civil Justice System 595, 613–21 (2003) (discussing limited situations in which judges are likely to grant pre-action disclosure orders); Kuo-Chang Huang, Introducing Discovery into Civil Law 231 (2003) (“Unlike discovery under the FRCP, there is no general scope of discovery uniformly applicable to all discovery devices under the English system.”); id. at 235; Paul Matthews & Hodge M. Malek, Disclosure 27–29 (4th ed. 2012) (discussing the British system); Michael Zander, Cases and Materials on the English Legal System 86–88 (10th ed. 2007). See generally Adrian A.S. Zuckerman, Civil Procedure (2003) (treating the English system). The lasting influence of such fundamental principles as nemo tenetur edere contra se, the principle of immediacy, the principle of party presentation, and the code provisions governing evidence acquisition and presentation in civil law countries, bar direct party-on-party discovery entirely. 25See Huang, supra note 24, at 21–27; Peter L. Murray & Rolf Stürner, German Civil Justice 275–78 (2004); Verkerk, supra note 23, at 16, 30, 248, 264–65. And the “inquisitorial” design of civil law proceedings centralizes all fact gathering and review in the hands of the judge, who has neither the authority nor the resources to conduct the kind of factual investigation that is standard in U.S. litigation. 26The term “inquisitorial” is here used as a short-hand for the American reader. Civil law proceduralists do not view their adjudication system as “inquisitorial” but as “adversarial,” and the neutrality of the judge is a fundamental principle of modern civil law. See Verkerk, supra note 23, at 280–81. Like translations of legal vocabulary more generally, the word “adversarial” also has different meanings in different legal systems and must be understood in connection with the other principles that inflect it, and in the context of the practices that inform and are informed by the concept. Note that, in civil law systems, appeals court judges make their own evidentiary findings and may seek additional evidence. See, e.g., Murray & Stürner, supra note 25, at 16. But this does not markedly increase judicial resources available for fact-finding, if at all.

This oversight in the corporate law literature is remarkable in that this literature has paid so much attention to the disclosure regime established by the 1933 Securities Act and the 1934 Securities Exchange Act, even as it has neglected the modern discovery rules that armed litigants with vastly expanded investigatory powers for obtaining information about corporate internal mismanagement, for the same reasons at roughly the same time. The U.S. rules of civil procedure and litigation discovery, in particular, are intricately related (both historically and functionally) to the basic principles that inform the U.S. style of regulation.

Modern discovery rules refashioned civil procedure, creating a long phase of adversarial, direct party-on-party fact investigation that has become the focus of U.S. litigation. 27See Fed. R. Civ. P. 26–37 & advisory committee’s notes (describing these developments). During litigation discovery, parties wade through millions of documents, take dozens (and, in large cases, even hundreds) of witness depositions, subpoena third parties for information and testimony, hire experts and produce expert reports—and do all of this before trial with almost no involvement by a judge. 28See infra notes 460–62 and accompanying text, for a discussion of the theory that the federal rules have turned trial judges into “managerial judges.”

The “discovery revolution” in private litigation was part of a much broader transformation of economic governance during the 1930s. 29See, e.g., Governance of the American Economy 3–5 (John L. Campbell, J. Rogers Hollingsworth & Leon N. Lindberg eds., 1991) (explaining how and why transformations in governance occurred in different industries). This transformation was animated by a “scientific approach” that deployed information-pushing regimes in contexts as diverse as financial markets regulation and private litigation for the purpose of obtaining the data that government officials, markets, and private parties needed to make rational, fact-based decisions. 30For a discussion of legal realism’s influence on securities regulation and civil procedure during the New Deal, see infra notes 105–19 and accompanying text. Just as securities disclosure by issuers would generate information for markets and regulators, modern litigation discovery would generate information for litigants and courts. In both cases, increased efficiencies were a major goal.

But the corporate-law scholarship has largely failed to recognize litigation discovery as a distinctive factor shaping the evolution of U.S. corporate and securities law, even as scholars in other legal fields have long understood that substantive “developments in areas such as products liability, employment discrimination, and consumer protection have been the result at least partly of broad-ranging discovery provisions.” 31Jack H. Friedenthal, A Divided Supreme Court Adopts Discovery Amendments to the Federal Rules of Civil Procedure, 69 Calif. L. Rev. 806, 818 (1981); see also Geoffrey C. Hazard, Jr., Conference on the Federal Rules of Civil Procedure: Report 3 (1995) (reporting on the proceedings of the Conference on the Federal Rules of Civil Procedure, convened in Dallas, Texas, on March 30–31, 1995); Richard L. Marcus, Discovery Containment Redux, 39 B.C. L. Rev. 747, 749 (1998) (“[B]road discovery put pressure on the substantive law to expand . . . .”).

The discovery revolution has been a bête noire for the business community ever since. 32See, e.g., Franklin S. Wood, Chamber of Commerce of the State of N.Y., Survey and Report Regarding Stockholders’ Derivative Suits 61–66 (1944) (discussing the cost and nuisance of discovery). A 2010 white paper on discovery by the American Chamber of Commerce, for example, observed that “discovery has become the focus of litigation, rather than a mere step in the adjudication process . . . [and that] the effort and expense associated with electronic discovery are so excessive that settlement is often the most fiscally prudent course—regardless of the merits of the case.” 33John H. Beisner, U.S. Chamber Inst. for Legal Reform, The Centre Cannot Hold: The Need for Effective Reform of the U.S. Civil Discovery Process 2 (May 2010), available at http://ilr.iwssites.com/uploads/sites/1/ilr_discovery_2010_0.pdf. On this basis, business leaders and especially corporate defendants have, for decades, complained that the costs and burdens of discovery are too high. 34See, e.g., Paul D. Carrington, Politics and Civil Procedure Rulemaking: Reflections on Experience, 60 Duke L.J. 597, 606–09 (2010) (detailing the efforts of business leaders to effect reform); Danya Shocair Reda, The Cost-and-Delay Narrative in Civil Justice Reform: Its Fallacies and Functions, 90 Or. L. Rev. 1085, 1123–24 (2012) (“[P]ro-business and insurance organizations helped to underwrite surveys that were then relied upon by . . . the CJRA [Civil Justice Reform Act of 1990] legislation . . .‘[i]t is no secret that the anti-discovery pressure [in the 1998 proposed discovery rules] has come from defendants, especially defendants in product liability, securities, and antitrust cases.’” (quoting Elizabeth G. Thornburg, Giving the “Haves” a Little More: Considering the 1998 Discovery Proposals, 52 SMU L. Rev. 229, 243 (1999))).

The costs and abuses of discovery were also a key focus of the debates concerning the Private Securities Litigation Reform Act of 1995 (PSLRA). 35Pub. L. No. 104-67, 109 Stat. 737 (1995) (codified as amended in scattered sections of 15 U.S.C.). Congress enacted the PSLRA in response to industry complaints (especially from Silicon Valley) 36See, e.g., Timothy K. Roake & Gordon K. Davidson, The Private Securities Litigation Reform Act of 1995, Corp. & Sec. L. Update (Fenwick & West LLP, Mountain View, Cal.), Jan. 1996, at 1, available at http://www.fenwick.com/fenwickdocuments/corp_sec_01-00-96.pdf (“Abusive securities litigation has been an unfortunate fact of life in the Silicon Valley for years, and the high tech community lobbied long, hard and well for [this] reform. . . . against ‘strike suits.’”). that class actions under the SEC’s Rule 10b-5 antifraud regulation 3717 C.F.R. § 240.10b-5 (2012) (prohibiting “any untrue statement of a material fact . . . in connection with the purchase or sale of any security”). led to countless “strike suits” that extracted settlements from defendant corporations regardless of the merits of such claims by threatening them with costly discovery and potentially ruinous damages. 38See Hillary A. Sale, Heightened Pleading and Discovery Stays: An Analysis of the Effect of the PSLRA’s Internal-Information Standard on ’33 and ’34 Act Claims, 76 Wash. U. L.Q. 537, 552–53 (1998) (reviewing the arguments behind the reform). In Novak v. Kasaks, 216 F.3d 300 (2d Cir. 2000), the Second Circuit stated that “[l]egislators were apparently motivated in large part by a perceived need to deter strike suits wherein opportunistic private plaintiffs file securities fraud claims of dubious merit in order to exact large settlement recoveries,” id. at 306. This view was also endorsed by the academic literature. See generally Janet Cooper Alexander, Do the Merits Matter? A Study of Settlements in Securities Class Actions, 43 Stan. L. Rev. 497, 499–500 (1991) (describing how the structural characteristics of securities class actions, rather than substantive merits, determine the amount of settlements). Discovery was exclusively treated as an expense or burden that had to be contained. 39See H.R. Rep. No. 104-369, at 31, 37 (1995) (Conf. Rep.). See generally Craig C. Martin & Matthew H. Metcalf, The Fiduciary Duties of Institutional Investors in Securities Litigation, 56 Bus. Law. 1381, 1382–83 (2001) (outlining the PSLRA’s stated purposes and extensive legislative history). The view prevailed, leading Congress to heighten pleading requirements for securities class actions under 10b-5 and to impose a discovery stay during the pendency of any motions to dismiss. 40See Securities Act of 1933, 15 U.S.C. § 77z-1(b)(1) (2012); Securities Exchange Act of 1934, id. § 78u-4(b)(3)(A)–(B). Both measures made it more difficult for plaintiffs to obtain discovery. 41See, e.g., Stephen J. Choi, Do the Merits Matter Less after the Private Securities Litigation Reform Act?, 23 J.L. Econ. & Org. 598, 599 (2007) (“Without the ability to engage in discovery, plaintiffs’ attorneys face a higher cost in determining the presence of specific misleading statements and omissions and the materiality of such misstatements and omissions.”); Sale, supra note 38, at 538.

Legislators, practitioners, and scholars have for the most part accepted the narrative of disproportionately high discovery costs and focused on diminishing discovery costs. 42See, e.g., Jason M. Rosenthal, Staying Discovery in Federal Securities Lawsuits, Prac. Litigator, Nov. 2002, at 7 (admitting the high cost of discovery). See generally John H. Beisner, Discovering a Better Way: The Need for Effective Civil Litigation Reform, 60 Duke L.J. 547, 571–75 (2010) (pointing out flaws in arguments by scholars attempting to refute the assertion that discovery is expensive and prone to abuse); Earl C. Dudley, Jr., Discovery Abuse Revisited: Some Specific Proposals to Amend the Federal Rules of Civil Procedure, 26 U.S.F. L. Rev. 189, 190 (1992) (arguing that the “managerial approach” to discovery is not cost-effective); Frank H. Easterbrook, Discovery as Abuse, 69 B.U. L. Rev. 635, 636 (1989) (noting how the high costs of discovery shape litigants’ decisions). But see Sale, supra note 38, at 538, 540 (arguing that the PSLRA reform was “overinclusive” in that meritorious claims would be dismissed and proposing a solution based on judicially managed discovery); Lynn A. Stout, Type I Error, Type II Error, and the Private Securities Litigation Reform Act, 38 Ariz. L. Rev. 711, 713–15 (1996) (discussing potential underdeterrence problems). They have largely neglected careful analysis of discovery’s information production and scrutiny mechanism for the governance of corporations and securities markets, which is the topic of this Article. Even if we grant that discovery costs are high, especially in large corporate cases, such costs must be assessed in relative terms, not only with regard to the stakes at issue in a particular lawsuit but with regard to discovery’s benefits to the efficiency of the capital markets system. This Article therefore focuses on the aggregate benefits of the discovery cost–benefit equation. It analyzes the positive externalities of discovery.

Along the way, this Article also speaks to important debates in the corporate law literature. For example, the literature has questioned the continued significance of shareholder derivative suits and fiduciary duty litigation in the United States. 43See generally, Robert W. Hamilton & Jonathan R. Macey, Cases and Materials on Corporations 734 (10th ed. 2007) (suggesting that “virtually all derivative litigation filed today is without merit”); Kenneth B. Davis, Jr., The Forgotten Derivative Suit, 61 Vand. L. Rev. 387, 388 (2008); Robert B. Thompson & Randall S. Thomas, The New Look of Shareholder Litigation: Acquisition-Oriented Class Actions, 57 Vand. L. Rev. 133, 135 (2004). The current view—in spite of some holdouts—is that shareholder derivative suits no longer discipline directors and officers, as evidenced by the fact that such suits almost never impose liability on outside directors, 44See Bernard Black, Brian Cheffins & Michael Klausner, Liability Risk for Outside Directors: A Cross-Border Analysis, 11 Eur. Fin. Mgmt. 153, 155 (2005) [hereinafter Black et al., Liability Risk for Outside Directors]; Bernard Black, Brian Cheffins & Michael Klausner, Outside Director Liability, 58 Stan. L. Rev. 1055, 1059 (2006) [hereinafter Black et al., Outside Director Liability]; Brian R. Cheffins & Bernard S. Black, Outside Director Liability Across Countries, 84 Tex. L. Rev. 1385, 1387 (2006). most cases settle (thus protecting directors at the expense of shareholders), 45See Armour et al., supra note 13, at 703 (2009). and the twin hurdles of the demand requirement and the business judgment rule mean that most cases do not even go to discovery in the first place. 46See Kenneth B. Davis, Jr., Structural Bias, Special Litigation Committees, and the Vagaries of Director Independence, 90 Iowa L. Rev. 1305, 1331 (2005) (discussing the analytical link between demand requirements and business judgment considerations); Robert B. Thompson & Randall S. Thomas, The Public and Private Faces of Derivative Lawsuits, 57 Vand. L. Rev. 1747, 1759 (2004) (“Even if a suit survives the demand hurdle, the board of directors can still form a special independent litigation committee that can dismiss the case.”). But see Minor Myers, The Decisions of the Corporate Special Litigation Committees: An Empirical Investigation, 84 Ind. L.J. 1309, 1311 (2009) (finding that special litigation committees do not always dismiss derivative litigation). This literature has, perhaps understandably, focused primarily on liability and litigation outcomes in its assessment of the usefulness of shareholder litigation. However, in doing so, it never considers to what extent salutary internal investigations are triggered by shareholder complaints of corporate internal wrongdoing, by the threat of litigation discovery, and by defensive discovery—regardless of the litigation outcome. Moreover, it misses the very connection between discovery and corporate internal investigations, which have become critical features of U.S.-style corporate governance. 47See infra Part III.A.5. Such investigations make use of the tools, and track the practices of litigation discovery. 48See infra Part III.A.4–5.

In this Article we show, inter alia, that private enforcement in the United States depends on the power that litigation discovery affords private attorneys general 49The term “private attorney general” was coined by Judge Frank in Associated Industries of New York State, Inc. v. Ickes, 134 F.2d 694, 704 (2d Cir. 1943), and refers to a private party who brings a lawsuit considered to be in the public interest. See David Shub, Note, Private Attorneys General, Prevailing Parties, and Public Benefit: Attorney’s Fees Awards for Civil Rights Plaintiffs, 42 Duke L.J. 706, 708 (1992). to investigate, uncover, and reconstruct the facts and circumstances of corporate mismanagement in great detail. We argue that the process of discovery itself disciplines management. 50See infra Part III.A.45.

The thesis of this Article is that civil procedure, and in particular the investigation of corporate internal wrongdoing by outsiders authorized and encouraged by the federal discovery rules, has had a substantial influence on the evolution of corporate governance in the United States. In this Article, we argue that litigation discovery is a critical legal variable in the development of U.S. corporate and securities laws, and, more broadly, of the U.S. culture of corporate disclosure.

Litigation discovery, and the threat of litigation discovery, has driven and structured the process of corporate shareholder litigation. 51In this Article, “shareholder litigation” includes all types of shareholder litigation at the state and federal level, including shareholder direct actions (such as breach of duty in the merger context, or requests under title 8, section 220 of the Delaware Code), shareholder derivative actions, and shareholder securities class actions. We focus primarily on shareholder derivative actions under state law and shareholder securities class actions under federal law as examples. While shareholder derivative actions may, of course, be brought in federal courts, the Securities Litigation Uniform Standards Act of 1998 (SLUSA) precludes federal securities class actions from being litigated in state courts. See Pub. L. No. 105-353, 112 Stat. 3227 (1998) (codified in scattered sections of 15 U.S.C.). The results of litigation discovery have led to the development of the case law defining shareholder rights and managerial duties. The threat of discovery and the episodic legal demands for detailed corporate internal information have induced incremental improvements in corporate governance practices, including more exacting decision procedures, internal monitoring, recordkeeping, and securities disclosure. Defensive discovery exposes decision-making by individual managers or directors to the scrutiny of the entire board and outside litigation counsel, all of whom have either fiduciary duties or professional obligations to the company. 52See infra Part III.A.4. Moreover, the highly developed, continuously evolving litigation discovery practices have established templates for independent corporate internal investigations by boards and regulators. 53See Paul Lomas & Daniel Kramer, Introduction to Corporate Internal Investigations: An International Guide 1, 3 (Paul Lomas & Daniel J. Kramer eds., 2008) (“The US has been in the forefront of the trends that have led corporations to conduct internal investigations and to cooperate with governmental inquiries, but the procedures in the US are constantly evolving themselves.”).

Discovery has also given regulators steady insight into changing corporate governance practices to which they have responded with regulatory changes. The information revealed through discovery of corporate internal wrongdoing in response to scandals, such as Enron and WorldCom, stock options backdating, executive pay, and subprime mortgage lending, has driven broad legal, statutory, and regulatory reforms. 54See infra note 233 and accompanying text.

This Article concludes that litigation discovery serves as a form of ex post disclosure that complements and enforces ex ante disclosure under the federal securities laws. Securities disclosure proceeds ex ante, is standardized, and is necessarily limited and summary in nature. 55See Frank B. Cross & Robert A. Prentice, The Economic Value of Securities Regulation, 28 Cardozo L. Rev. 333, 363–64 (2006) (discussing the virtues of mandatory disclosures). In contrast, discovery occurs ex post and generates additional, extensive, specific disclosures not mandated by the regulator ex ante. Consequently, litigation discovery promotes and influences the U.S. culture of public company disclosure.

This Article is structured as follows. Part I presents the basic insight of the paper. It argues that discovery is the “elephant in the boardroom.” By treating the Disney shareholder litigation as an example, it illustrates how discovery disciplines corporate management.

Part II describes the U.S. information-pushing regime implemented by the modern discovery rules. It recalls the New Deal origins of modern discovery, describes the unprecedented scope of the rules, and considers the significance of devolving judicial authority for fact gathering onto private parties.

Part III, the core of the Article, examines a range of institutional consequences that flow from the rules and practices of discovery: It describes how discovery (a) generates information and disciplines management at every stage of shareholder litigation (even prior to the motion to dismiss); (b) has established templates for independent corporate internal investigations; (c) induces incremental improvements in corporate governance practices, including more exacting decision procedures, internal monitoring, recordkeeping, and controls; (d) stimulates and shapes the development of case law and judicial precedent; (e) reveals information of corporate internal wrongdoing that drives substantive corporate law, securities laws, and regulatory reforms; and (f) concludes that discovery serves as a form of ex post disclosure, which complements and enforces ex ante disclosure under the federal securities laws.

Part IV then addresses specific objections to our thesis that corporate and civil procedure scholars have advanced.

Part V offers several normative conclusions concerning policy proposals regarding legal transplants, criticisms of the empirical literature on shareholder actions, and reforms of the current discovery regime. This Part cautions against legal transplants of U.S.-style securities disclosure, aggregate litigation mechanisms, or other aspects of enforcement without considering appropriate tools for investigating corporate internal wrongdoing ex post. Here we also return to the enforcement debate and offer some suggestions on disaggregating data on litigation outcomes by focusing instead on what fact investigations have taken place at what phase in the adjudication process. Finally, we consider the implications of our thesis for the current proposed changes to the federal discovery rules.

I. The Role of Discovery in U.S. Corporate Governance

A. The Elephant in the Board Room

All companies guard their knowledge resources, including information about their internal operations, business practices, and decision procedures. 56See generally Érica Gorga & Michael Halberstam, Knowledge Inputs, Legal Institutions, and Firm Structure: Towards a Knowledge-Based Theory of the Firm, 101 Nw. U. L. Rev. 1123, 1127 (2007) (arguing that firms must produce knowledge, disperse it within the firm, but also “prevent its transfer outside the firm”). Public companies only disclose what they must. 57See Tom C.W. Lin, Undressing the CEO: Disclosing Private, Material Matters of Public Company Executives, 11 U. Pa. J. Bus. L. 383, 385, 394–95 (2009) (discussing the growing prominence and wealth of corporate CEOs and the need for a corresponding increase in disclosure); Michael R. Siebecker, Trust & Transparency: Promoting Efficient Corporate Disclosure Through Fiduciary-Based Discourse, 87 Wash. U. L. Rev. 115, 117 (2009) (“[D]espite a host of federal and state statutes mandating disclosure of various corporate practices, corporations seem reluctant to disclose fully what consumers and investors want to know . . . .”).

Yet in the United States, the CEO of a company must be prepared to allow an army of litigation attorneys to subject every aspect of her company’s operations to intense scrutiny. Every document, electronic record, and communication among employees, managers, executives, and board members may have to be revealed during litigation discovery—on short notice. 58See John S. Beckerman, Confronting Civil Discovery’s Fatal Flaws, 84 Minn. L. Rev. 505, 525 (2000) (“Proficient advocates maximize strategic advantage during the investigation and discovery phase of pretrial litigation by obtaining as much relevant and useful information, admissions and impeachment material as possible from and about the adversary, for potential use at trial . . . .”). The scope of such investigations into the corporation’s internal affairs can be extensive, covering several years of operations and including the interrogation of employees who have long since left the company. 59See William W. Schwarzer, Mistakes Lawyers Make in Discovery, Litigation, Winter 1981, at 31, 31 (“Deposition notices drop like autumn leaves.”); see also, e.g., Sperano v. Invacare Corp., No. 03-CV-6157, 2006 WL 3524483, at *1–2 (W.D.N.Y. Dec. 6, 2006) (allowing deposition of the defendant’s former employee, even though she suffered from a rare brain disease). A CEO can never be sure that even highly confidential conversations with her most trusted advisors will be safe from forced disclosure. Discovery reaches personal e-mail accounts, home computers, and PDA devices. 60Hopson v. Mayor of Balt., 232 F.R.D. 228, 245 (D. Md. 2005) (describing the scope of electronic records as including “e-mail, voice mail, archived data, back-up or disaster recovery data, laptops, personal computers, PDA’s, deleted data”). And even if the production of confidential information is forestalled by achieving a settlement at an early stage in the litigation, the collection and review of relevant information by a corporate defendant’s outside litigation counsel for the purpose of preparing for settlement will generally expose a manager’s communications and conduct to the corporation’s general counsel and other gatekeepers, whose legal duties and loyalties all run to the corporation. 61See infra notes 296–308 and accompanying text.

Once underway, the discovery process threatens to expose to public scrutiny 62Exposure need not be to the general public in order to effectively discipline. As we discuss below, defensive discovery alone, or even a prediscovery corporate internal investigation by a special litigation committee, both of which operate in the shadow of the discovery rules, can lead to employment actions, improve compliance monitoring, and deter aggressive reporting practices. anything from minor lapses in judgment to serious wrongdoing that could lead to separate litigation or criminal charges unconnected with the pending litigation. 63See, e.g., Press Release, U.S. Dep’t of Justice, Former Computer Associates Executives Indicted on Securities Fraud, Obstruction Charges (Sept. 22, 2004), http://www.justice.gov/opa/pr/2004/September/04_crm_642.htm (stating that the financial restatement and the realization of the magnitude of the fraud by Computer Associates executives resulted from corporate internal investigation). Discrepancies between public statements and corporate internal information publicized by discovery will raise the specter of securities law and corporate fiduciary duty violations. And even if the threat of personal liability is remote, 64See Black et al., Outside Director Liability, supra note 44, at 1099. the reputational consequences of embarrassing revelations could be severe. 65See Eliezer M. Fich & Anil Shivdasani, Financial Fraud, Director Reputation, and Shareholder Wealth, 86 J. Fin. Econ. 306, 335 (2007) (showing that “fraud allegations have significant reputational consequences for outside directors”); Jeffrey N. Gordon, The Rise of Independent Directors in the United States, 1950–2005: Of Shareholder Value and Stock Market Prices, 59 Stan. L. Rev. 1465, 1489 (2007) (concluding that “in light of the Delaware courts’ reluctance to impose monetary liability on directors, the most significant independence-enhancing effect of litigation is probably through improving the operation of the reputation market rather than through the threat of monetary sanctions”); Bruce Haslem, Aimee Hoffmann Smith & Irena Hutton, How Much Do Corporate Defendants Really Lose? A New Verdict on the Reputation Loss Induced by Corporate Litigation 7 (July 8, 2013) (unpublished manuscript), http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2290821.

Discovery materials are deemed public by default. 66Unless there is a confidentiality agreement between the parties, or a protective order is obtained by the court, a party is generally free to pass on documents or deposition transcripts to third parties who are not involved in the litigation. Because discovery materials are not filed with the court, see Fed. R. Civ. P. 5, it is not always recognized that they are not protected from publication, even at an early stage in the litigation process. Some academics who have attended to the issue have deplored the “public by default” standard. But in so doing, their focus is on the rights and burdens of the parties and other narrow procedural issues without considering the broader impact that publication and its threat have on the regulation of securities markets and corporate governance more generally. See, e.g., Richard L. Marcus, A Modest Proposal: Recognizing (At Last) That the Federal Rules Do Not Declare That Discovery Is Presumptively Public, 81 Chi.-Kent L. Rev. 331, 331 (2006) (“[T]he real problem is that the overbroad notion that all information turned over in discovery (no matter what it contains, or what the court or even the parties, know about the material involved) is presumptively open to the public, even if no party wants to permit public access to the information.”). Of course, most information disclosed during discovery will not become public. Discovery often produces millions of documents, which are not published, and dozens of deposition transcripts, which may become available only in part. But, there are many examples of the publication of discovery materials, including very large datasets, 67One need only search YouTube to locate hundreds of excerpts of videotaped depositions that may embarrass defendant corporate managers. See, e.g., CenterJD, Enron Banks—Will President Bush Support Them?, YouTube (Jul. 30, 2007), http://www.youtube.com/watch?v=9zxAJO7owy8 (showing deposition of Enron’s Chief Financial Officer, Andrew Fastow); see also infra note 226. and of plaintiffs in different litigations sharing discovery materials. 68See Citizens First Nat’l Bank of Princeton v. Cincinnati Ins. Co., 178 F.3d 943, 946 (7th Cir. 1999) (observing that “[m]ost cases endorse a presumption of public access to discovery materials” and citing numerous such cases); Okla. Hosp. Ass’n v. Okla. Publ’g Co., 748 F.2d 1421, 1424 (10th Cir. 1984) (“[P]arties to litigation have a constitutionally protected right to disseminate information obtained by them through the discovery process absent a valid protective order . . . .”); see also Manual for Complex Litigation (Fourth) § 11.432, at 67–68 (2004) (endorsing the sharing of information obtained from discovery for litigation purposes). Moreover, private parties have incentives to select the most troubling documents and the most relevant information and use them to make their cases. In important cases, the public may obtain information if any party leaks it to the specialized media. The media will be eager to generate news. And parties have incentives to reveal information in egregious cases when the generation of public outrage supports their case.

Transparency generated by discovery is not measured by the quantity of information that is published, but by its quality. The most important content, by a qualitative standard, will generally inform the litigation process and therefore become available, if not to the public, at least to lawyers, gatekeepers, judicial authorities, and regulators.

Confidentiality agreements among the parties offer limited protection against disclosure of discovery materials. 69Note that the deposition transcript of Neil Cotty, Bank of America’s Chief Accounting Officer, was marked “confidential” but was nonetheless filed with the court and testimony was cited in the press. See Continued Videotaped Deposition of Neil Andrew Cotty, In re Bank of Am. Corp., Sec., Derivative, & ERISA Litig., 757 F. Supp. 2d 260 (S.D.N.Y. 2010) (No. 09 MD 2058 (PKC)), ECF No. 694-6. Even where sealed documents are obtained and leaked, as in the Zyprexa case, judges craft their injunctions narrowly and have not issued general injunctions on republication of already leaked documents. See, e.g., In re Zyprexa Injunction, 474 F. Supp. 2d 385, 395, 397 (E.D.N.Y. 2007). Even where parties sign a confidentiality agreement, these discovery materials may nonetheless become available to lawyers, legal experts, gatekeepers, 70See John C. Coffee Jr., Gatekeepers: The Professions and Corporate Governance 2 (2006) (“[G]atekeeper . . . [means] an agent who acts as a reputational intermediary to assure investors as to the quality of the ‘signal’ sent by the corporate issuer.”). Examples of gatekeepers include auditors, investment bankers, securities analysts, and attorneys. Id. at 2, 192. or regulators, 71See infra Part III.A.5 (discussing information sharing by special litigation committees with the SEC). and thereby inform legal proceedings indirectly. The publication of discovery documents depends on the stage the suit reaches. The further the litigation goes, the more information is likely to become available. Close to trial, more information is revealed. But that does not mean information obtained during discovery will not be revealed at earlier stages. Cases dismissed, settled, or resolved at earlier stages tend to generate less information as compared with cases that are resolved at trial. However, as we show in Part III, the U.S. discovery system generates information flows at every stage of the litigation process.

The main point is that the U.S. civil procedure system, through litigation discovery, generates more high-quality information about corporate wrongdoing than any other legal system in the world. In this way, the U.S. civil procedure system is exceptional.

Litigation discovery, and its threat, disciplines the entire corporate hierarchy. Not personal liability, as has often been assumed, but discovery itself encourages truth telling, constrains officers and directors from breaching their fiduciary duties, and mandates great care in the dutiful management of corporate internal affairs. Litigation discovery is the proverbial “elephant in the room.”

Case law shows how consequential information obtained through discovery can be.

B. The Example of the Disney Company Shareholder Derivative Litigation

The shareholder derivative litigation against the board of the Walt Disney Company serves as a prime example of how litigation discovery exposes corporate governance failures, disciplines management, and focuses public attention on corporate governance issues—in this case, on the excesses of executive pay—regardless as to whether directors face personal liability for mismanagement or misconduct.

The basic facts are well known. Disney’s board awarded Michael Ovitz, its departing president, a severance package of approximately $140 million after just fourteen months in office, in accordance with an early termination provision in Ovitz’s employment contract. 72See Brehm v. Eisner, 746 A.2d 244, 249, 252–53 (Del. 2000). In this derivative action based on waste, complaining shareholders challenged not merely the board’s initial approval of the excessive downside protection in the original employment contract, but also the board’s approval of Disney Chairman Michael Eisner’s decision to terminate Ovitz without cause. 73In re Walt Disney Co. Derivative Litig. (Disney I), 731 A.2d 342, 353 (Del. Ch. 1998), aff’d in part, rev’d in part sub nom. Brehm v. Eisner, 746 A.2d 244 (Del. 2000).

The litigation lasted more than eight years and resulted in five decisions, three by the Delaware Court of Chancery and two by the Delaware Supreme Court. 74In re Walt Disney Co. Derivative Litig. (Disney IV), 906 A.2d 27 (Del. 2006); Brehm, 746 A.2d 244; In re Walt Disney Co. Derivative Litig. (Disney III), 907 A.2d 693 (Del. Ch. 2005), aff’d, 906 A.2d 27 (Del. 2006); In re Walt Disney Co. Derivative Litig. (Disney II), 825 A.2d 275 (Del. Ch. 2003); Disney I, 731 A.2d 342. After initially dismissing the case for failure to satisfy the demand requirement, 75Disney I, 731 A.2d at 380. the Delaware Chancery ultimately ordered defendants to answer the complaint and granted the plaintiffs discovery. 76Disney II, 825 A.2d at 291. We note that Delaware rules of discovery track the discovery rules of the Federal Rules of Civil Procedure (FRCP). 77See infra Part II.D.

After extensive discovery, the case went to trial. The trial lasted 37 days (from October 20, 2004, to January 19, 2005), included 24 witness examinations, and generated 9,360 pages of transcripts. 78Disney III, 907 A.2d at 697. Ultimately, the Delaware Supreme Court found in favor of the directors, which some have described as a “complete victory.” 79See, e.g., Jonathan Macey, Delaware: Home of the World’s Most Expensive Raincoat, 33 Hofstra L. Rev. 1131, 1131 (2005). But the court explicitly found that their conduct “fell significantly short of the best practices of ideal corporate governance,” that Eisner “enthroned himself as the omnipotent and infallible monarch of his personal Magic Kingdom,” and that he “stacked his . . . board of directors with friends and other acquaintances who, though not necessarily beholden to him in a legal sense, were certainly more willing to accede to his wishes.” 80Disney III, 907 A.2d at 697, 760, 763. For public reaction, see, for example, Steven Taub, Disney Wins Shareholder Lawsuit, CFO (Aug. 10, 2005), http://ww2.cfo.com/risk-compliance/2005/08/disney-wins-shareholder-lawsuit/ (reporting that the court “did not exonerate Eisner and the Disney board outright”). Michael Eisner resigned as Disney’s CEO the following month and from Disney’s board shortly thereafter. 81See Bloomberg News, Eisner Makes Clean Break, Resigns from Board of Directors, Chi. Trib. (Oct. 7, 2005), http://articles.chicagotribune.com/2005-10-07/business/0510070136_1_robert-iger-michael-eisner-disney (reporting that Eisner resigned as Disney’s CEO on September 30, 2005).

For attorneys trained in civil law jurisdictions, the extraordinary detail of the factual recitations and the extent to which they render transparent highly sensitive, confidential internal decision-making by the company’s officers and directors are bewildering. “The Court . . . reviewed thousands of pages of deposition transcripts and 1,033 trial exhibits that filled more than twenty-two 3½-inch binders.” 82Disney III, 907 A.2d at 697. In an eighty-two-page opinion on its decision after trial, the Chancery spent nearly forty-seven pages reciting the facts of the case, another twenty-one pages analyzing the facts in light of the applicable law, and only eleven pages reciting the law. 83See id. at 696–97. One reason for trial courts to include such lengthy recitations of fact is, of course, to avoid getting overturned on appeal, as the standard of review for facts is clear error, but a de novo standard applies to judicial review of matters of law. See Fed. R. Civ. P. 52(a)(6); United States v. U.S. Gypsum Co., 333 U.S. 364, 399 (1948); see also Steven Alan Childress, Standards of Review Primer: Federal Civil Appeals, 229 F.R.D. 267, 270 (2005).In European civil law systems, opinions are shorter and include significantly less recitations of facts. See, e.g., Mitchel de S.-O.-L’E. Lasser, Judicial Deliberations: A Comparative Analysis of Judicial Transparency and Legitimacy 104–112 (2004) (discussing the brevity of European Court of Justice opinions); Gillian K. Hadfield, The Quality of Law: Judicial Incentives, Legal Human Capital and the Evolution of Law 28 (Univ. S. Cal. Ctr. in Law, Econ. & Org., Research Paper No. C07-3, 2007), available at http://ssrn.com/abstract=967494 (“Common law regimes, for example, tend to produce extensive, publicly available judicial opinions laying out factual findings and legal reasoning; civil code regimes tend to produce detailed academic commentary, published alongside short, relatively opaque, legal opinions with extensive judicial analysis sometimes confined to documents distributed only within the judiciary.” (citing Lasser, supra)); Annette Marfording & Ann Eyland, Civil Litigation in New South Wales: Empirical and Analytical Comparisons with Germany, 500–01 (Univ. of N.S.W. Faculty of Law Research Series, Paper No. 28, 2010), available at http://ssrn.com/abstract=1641554 (comparing page length of decisions in New South Wales and Germany); see also, Zivilprozessordnung [ZPO] [Code of Civil Procedure] § 313(2) (Ger.) (providing that, in the reasons for judgment, the plaintiff’s claim and the parties’ respective factual allegations are to be presented only as far as essential, and otherwise the court is to simply refer to the parties’ pleadings and the transcript).

The Court’s factual recitations were, of course, entirely dependent on the extensive discovery that plaintiffs had conducted during the pretrial period over many years. 84U.S. judges do not collect facts, identify witnesses, or control the preparation of evidence for trial. See infra notes 148–53 and accompanying text. Moreover, they reflect only the tip of the iceberg. The opinion discusses only the most relevant evidence, that was admissible and which the attorneys culled from mountains of discovery for presentation at trial. 85The scope of discovery extends to “relevant” information and other information “reasonably calculated to lead to the discovery of admissible evidence.” Fed. R. Civ. P. 26(b)(1). The Disney case illustrates how discovery by plaintiffs in U.S. shareholder derivative actions succeeds in reconstructing the “who, what, when, where, and why” of corporate internal decision-making. 86See infra note 254 and accompanying text.

We note here in particular the kinds of facts to which plaintiffs gained access as a result of discovery. Plaintiffs’ discovery substantially covered Ovitz’s business arrangements since the founding of Creative Artist Agency in the 1970s. 87Disney III, 907 A.2d at 700. Plaintiffs dissected Ovitz’s career, business partnerships, leadership positions, and opportunity costs, including confidential internal personnel and management issues of the partnership that he founded, and the details and structure of a completely separate failed deal to install him as CEO of the Music Corporation of America (formerly MCA, now Universal Studios) just prior to his appointment as president of the Walt Disney Company. 88Id. at 700–02. Extensive pretrial collection, review, analysis, and synthesis of evidence from testimony, business records, correspondence, public records, and news accounts, from the parties as well as entirely uninvolved third parties, allowed plaintiffs to offer the fact-finder a blow-by-blow account of the process of hiring and firing Ovitz. 89Id. at 699–740. The investigation uncovered what each and every board member knew, when they knew it, and what they did or did not do to inform themselves. 90See, e.g., id. at 724–40. The Court recited in detail what information was kept from the board entirely, what private and business relationships Eisner maintained with Ovitz and with every board member, the express or implied motivations of Eisner’s conduct at each stage of the process to hire and fire Ovitz, and much more. 91See id. at 702–40.

Discovery revealed that Eisner and Irwin Russell largely controlled the hiring process. Russell, who was the head of the compensation committee, and also Eisner’s personal counsel, was thus determined to be a nonindependent director. 92Disney I, 731 A.2d 342, 360 (Del. Ch. 1998), aff’d in part, rev’d in part sub nom. Brehm v. Eisner, 746 A.2d 244 (Del. 2000). Discovery revealed that the “first instance where a board member other than Russell or Eisner was brought into the Ovitz negotiation process” was when the financial terms of Ovitz’s employment agreement had already been worked out. 93Disney III, 907 A.2d at 704 & n.40. Discovery revealed that the analysis of Ovitz’s proposed compensation agreement by the executive compensation consultant, Graef Crystal, was “never circulated to any board member other than Eisner.” 94Id. at 705–06. The analysis—now available to the entire world—voiced serious concerns the board should have seen before it approved Ovitz’s contract. In it, Crystal emphasized that he “was philosophically opposed to a pay package that would give Ovitz the best of both worlds—i.e., low risk and high return.” 95Id. at 705. It also took the discovery process for the board to learn that it had not been properly informed of the opposition to Ovitz’s hiring by two of the most important Disney executives, Sanford Litvack (Disney’s General Counsel) and Stephen Bollenbach (Disney’s CFO). 96The record states, “In executive session, the board was informed of the reporting structure that Eisner and Ovitz agreed to, but no discussion of the discontent Litvack or Bollenbach expressed at Eisner’s home was recounted.” Id. at 710. Again, it was discovery that turned up all this information, not the court, and not the press.

The Court used the wealth of information about Disney’s executive pay practices to castigate Eisner, announce stricter fiduciary duty standards on board decision-making, and identify best practices that corporate management should adopt to shield themselves from similar claims going forward:

Eisner’s actions in connection with Ovitz’s hiring should not serve as a model for fellow executives and fiduciaries to follow. His lapses were many. He failed to keep the board as informed as he should have. He stretched the outer boundaries of his authority as CEO by acting without specific board direction or involvement. He prematurely issued a press release that placed significant pressure on the board to accept Ovitz and approve his compensation package in accordance with the press release. To my mind, these actions fall far short of what shareholders expect and demand from those entrusted with a fiduciary position. Eisner’s failure to better involve the board in the process of Ovitz’s hiring, usurping that role for himself, although not in violation of law, does not comport with how fiduciaries of Delaware corporations are expected to act. 97Id. at 762–63 (footnote omitted).

What was remarkable about the litigation for U.S. observers, as Professor Macey and others have noted, is that the outcome of the case—no liability for any defendant—was widely predicted, but the Delaware Courts nonetheless permitted the litigation to proceed for so long, at considerable cost to the company and its shareholders. 98See Macey, supra note 79, at 1132–33. Explanations proffered in the corporate law literature included the Delaware Chancery’s attempt to maintain its relevancy in light of the increasing federalization of corporate and securities laws, and in particular Congress’s passage of the Sarbanes-Oxley legislation in 2002; the signaling function of the opinion; the capture of Delaware’s legislature and judiciary by the interests of the corporate bar; and the reputational sanction the court imposed on the Disney Board and its CEO as a result. 99See id. at 1132, 1134–36. But the critical importance of the discovery regime for the court’s ability to accurately describe the facts in such granular detail, to deliver the reputational sanction to Eisner, and to signal to market participants that they could be exposed to the same kind of scrutiny, was never considered. The reason for this oversight is that we have come to take this kind of discovery for granted.

Litigation discovery, and the threat of litigation discovery, defines corporate shareholder litigation in the United States. As we show in Parts III and IV, the impact of shareholder litigation is intimately bound up with how modern discovery (1) affords litigants unprecedented tools to investigate corporate internal wrongdoing, (2) significantly decentralizes and leverages judicial investigative powers, (3) empowers parties with the greatest incentives to uncover wrongdoing, and (4) structures the litigation process more generally.

In agency-cost terms, the availability of litigation discovery by shareholders addresses an information asymmetry problem. 100See generally George A. Akerlof, The Market for “Lemons”: Quality Uncertainty and the Market Mechanism, 84 Q.J. Econ. 488, 489 (1970) (explaining the information asymmetry problem in the context of used cars). Shareholders can use litigation discovery to get information about the details of corporate self-dealing transactions or suspicious deals, which would otherwise be inaccessible to them. 101See Lucian Arye Bebchuk, Litigation and Settlement Under Imperfect Information, 15 Rand J. Econ. 404, 413 (1984) (“[T]he discovery requirement will likely reduce the informational asymmetry between the parties . . . .”). Knowing that this opportunity is available, and that shareholders can use the information once they obtain it, 102Third-generation transparency work suggests that transparency and disclosure are effective if interested parties can act on the information. See Archon Fung et al., Full Disclosure: The Perils and Promise of Transparency 173–74 (2007). makes managers more likely to avoid such conduct. By allowing monitoring, ex post discovery reduces the private benefits of control associated with running public corporations. The impact of discovery goes far beyond the litigation process in that it promotes an overall culture of transparency in U.S. corporate governance.

II. The U.S. Information-Forcing Regime

While most scholars and practitioners are generally aware of the atypical character of U.S. discovery, few fully appreciate the profound disparities between pretrial fact investigation in the U.S. courts and fact investigation in virtually every other court system. 103See Marcus, supra note 23, at 709. (“American proceduralists have not been comparativists.”) American proceduralists instead believe “it is best and sufficient to attend only to American topics.” Id. at 710. Simply stated, the U.S. “system of pretrial discovery is unique”; no other country has anything like it. 104Hazard, Discovery and the Role of the Judge, supra note 22, at 1018.

A. New Deal Origins of Modern Discovery

The significance of modern civil discovery for corporate governance is not an unintended consequence of the rules. The modern rules originated with the same legal-realist philosophy that informed FDR’s New Deal legislation and helped revolutionize the governance of U.S. markets in the 1930s. 105See generally David Marcus, The Federal Rules of Civil Procedure and Legal Realism as a Jurisprudence of Law Reform, 44 Ga. L. Rev. 433, 438 (2010) (showing how the positive potential of legal realism translated into reform); Stephen N. Subrin, How Equity Conquered Common Law: The Federal Rules of Civil Procedure in Historical Perspective, 135 U. Pa. L. Rev. 909, 944–48 (1987) (documenting the Federal Rules’ heavy reliance on equity procedures, which mirrored New Deal principles). See also Judith Resnik, Failing Faith: Adjudicatory Procedure in Decline, 53 U. Chi. L. Rev. 494, 502–05 (1986). The goal of legal realists was nothing short of a modern revolution in governance. They sought to place governance on a scientific footing, turn over the details of public administration to experts, and thereby create a modern state that would promote social welfare. 106See Marcus, supra note 105, at 468–69. Unfettered competition would be tempered by social cooperation. Scientific decision-making by experts based on objective data would replace haphazard judgments by lay politicians and judges. 107See id. Centralization would overcome the unsurveyable thicket of state and local governance regimes and ensure uniform, rational policies based on empirical evidence.

The Rules Enabling Act, which authorized the judiciary to develop and promulgate new rules of civil procedure for Article III courts, was passed in 1934 108Pub. L. No. 73-415, 48 Stat. 1064 (1934) (codified as amended at 28 U.S.C. § 2072 (2012)). The Rules Enabling Act gave the Supreme Court the power to make rules of procedure for federal courts as long as they did not “abridge, enlarge, or modify any substantive right.” 28 U.S.C. § 2072(b). along with the second wave of New Deal legislation that included the Securities Act of 1933 109Pub. L. No. 73-22, tit. I, 48 Stat. 74 (1933) (codified as amended at 15 U.S.C. §§ 77a–77aa (2012)). and the Securities Exchange Act of 1934. 110Pub. L. No. 73-291, 48 Stat. 881 (1934) (codified as amended at 15 U.S.C. §§ 78a–78pp). The rules were adopted in 1937 and became effective in 1938. 111See generally Stephen N. Subrin, Fishing Expeditions Allowed: The Historical Background of the 1938 Federal Discovery Rules, 39 B.C. L. Rev. 691, 729 (1998) (providing a historical survey of discovery reforms and their driving rationale). Charles Clark, then Dean of Yale Law School, the hotbed of legal realist thinking at the time, became the Reporter to the Judicial Conference’s Advisory Committee on Civil Rules (“Committee” or “Advisory Committee”) responsible for drafting the new rules. 112See generally, American Bar Association, Rules of Civil Procedure for the District Courts of the United States with Notes and Proceedings of the Institute on Federal Rules (William W. Dawson ed., 1938) [hereinafter IFR Proceedings] (providing a transcript of the proceedings of the ABA Institute on Federal Rules in Cleveland, Ohio, on July 21–23, 1938, at which Dean Clark was a lecturer). At the same time, another Yale Professor, William Douglas (later appointed to the Supreme Court) brought the same legal realist agenda to implementing the new securities disclosure regime at the Securities Exchange Commission, which was created by the 1934 Securities Exchange Act. 113See C. Paul Rogers III, The Antitrust Legacy of Justice William O. Douglas, 56 Clev. St. L. Rev. 895, 897, 899–903 (2008).

The Rules Enabling Act authorized the promulgation of new uniform federal rules of procedure, 114Pub. L. No. 73-415, 48 Stat. 1064 (1934) (codified at 28 U.S.C. § 2072 (2012)). whereas, previously, federal district courts followed the procedural rules of the state in which the district was located. 115See Subrin, supra note 111, at 693–729. The arcane and esoteric nineteenth century writ system was abolished. 116See Marcus, supra note 105, at 476 (describing the writ system). The outcome of litigation would no longer depend principally upon the skill of adversaries to manipulate the complicated pleading rules and other technicalities. 117See Conley v. Gibson, 355 U.S. 41, 48 (1957) (“The Federal Rules reject the approach that pleading is a game of skill in which one misstep by counsel may be decisive to the outcome . . . .”), abrogated on other grounds by Bell Atl. Corp. v. Twombly, 550 U.S. 544 (2007).

The new rules created a single cause of action, permitted the pleader to state multiple, alternative, and even mutually inconsistent claims, and joined bills in equity with actions at law. Pleading became a simple affair that, in simple negligence cases, barely required the assistance of counsel. 118See, e.g., Fed. R. Civ. P. 8; Fed. R. Civ. P. Form 20. All that was needed was an ordinary-language complaint (no magic words required) that gave the defendant notice of the claims and stated the grounds for relief. 119Fed. R. Civ. P. 8(a)(2); Swierkiewicz v. Sorema N.A., 534 U.S. 506, 508 (2002). The goal was the resolution of disputes on their merits and in the interests of justice. The rule-makers explicitly rejected dismissals based on technical defects in the pleadings before the plaintiff was given the chance to develop facts in support of his claims. 120See supra note 117; cf. Subrin, supra note 111, at 693.

Instead of deciding cases at the pleading stage or at trial, the new rules placed great emphasis on broad fact investigation by the parties themselves during an extended discovery period that would unveil the facts and circumstances of the dispute to both parties. 121See Subrin, supra note 111, at 703. Extensive discovery was to afford all the parties to a case access to all the facts on either side before trial, thus allowing them to rationally assess the value of their case and encouraging settlements. 122See id. at 716.

B. The New Tools of Modern Discovery

The new discovery rules gave private litigants an entirely new panoply of tools to obtain information, expanding considerably on discovery mechanisms that first emerged in courts of equity and were gradually introduced in different forms in cases at common law in about half of the states during the late nineteenth and early twentieth centuries. 123See George Ragland, Jr., Discovery Before Trial 19–24 (1932) (surveying state discovery practices); Subrin, supra note 105, at 922–25. Combining the most liberal discovery rules to be found within the several states (and in common law jurisdictions abroad) and reshaping them systematically, the reformers changed the nature of civil adjudication. 124See Subrin, supra note 111, at 718–19. Although the earlier Federal Equity Rules had contained some related provisions for depositions and the subpoena duces tecum that authorized the procurement of documents from third parties, these tools were essentially limited to situations in which witnesses were unavailable for examination in open court and required a showing of good cause. See id. at 699.

The reformers recognized the unprecedented and experimental character of the new rules. Their approach “complemented the legal realist movement[’s]” emphasis on “the importance of amassing all of the facts before deciding social policy or a case.” 125Id. at 739–40. The proposed modern discovery tools were analogized to the latest scientific tools of observation. According to George Ragland, one of the main proponents of modern discovery, a “lawyer who does not use discovery procedure” is like “a physician who treats a serious case without first using the X-ray.” 126Ragland, supra note 123, at 251 (internal quotation mark omitted). Discovery would place civil adjudication on a more rational, scientific foundation. Edson Sunderland, who was responsible for drafting the modern discovery rules, described the central role of discovery as follows:

It is probable that no procedural process offers greater opportunities for increasing the efficiency of the administration of justice than that of discovery before trial. Much of the delay in the preparation of a case, most of the lost effort in the course of the trial, and a large part of the uncertainty in the outcome, result from the want of information on the part of litigants and their counsel as to the real nature of the respective claims and the facts upon which they rest. 127Edson R. Sunderland, Foreword to Ragland, supra note 123, at iii.

According to Professor Hazard, the new rules of civil procedure led to a new role for civil claims as “an integral part of law enforcement in this country.” 128Hazard, supra note 31, at 3. “[T]he scope of discovery,” he concluded, has “determine[d] the scope of effective law enforcement in many fields regulated by law.” 129Id.

The most striking new tool was the deposition. 130See Fed. R. Civ. P. 26. Prior to the implementation of the FRCP, many states, including Delaware, did not permit the taking of oral depositions of witnesses before trial. 131See Del. Rev. Stat. ch. 121, § 49 (1915) (“The deposition must be taken in writing . . . . [N]either party shall be present at the taking the [sic] deposition, and no question shall be put but those sent by the justice.”); see Ragland, supra note 123, at 283–84. Even states that allowed oral depositions generally permitted the practice only for the purpose of preserving evidence. 132See Ragland, supra note 123, at 22–23. In courts at equity, depositions were traditionally permitted only as written questions. 133See Subrin, supra note 105, at 918–19. Interrogatories as written questions, however, were widely understood to be a “very ineffective” method of inquiry. 134See, e.g., IFR Proceedings, supra note 112, at 285. As Sunderland explained,

To draw up a series of questions and present them all at once to be answered, is certainly far less searching than to present questions one at a time, framing each successive question on the basis of prior answers given. Answers usually suggest lines of further inquiry which often lead to the most important disclosures. This is, of course, the chief reason for the effectiveness of the oral cross-examination. By submitting a complete set of interrogatories prepared in advance, the parties seeking discovery entirely lose this enormous advantage in eliciting the truth. 135Id. at 279–80.

The Delaware Court of Chancery’s 1868 rules of practice did allow “taking the testimony of witnesses on behalf of such party within the State . . . upon oral examination and subject to cross-examination and re-examination by the parties or their solicitors, before an examiner or examiners, to be appointed by the Chancellor.” 136Del. Ch. Ct. R. 40 (1868). But depositions were permitted only of supporting witnesses. Witnesses to be called by the opponent could not be deposed before trial, and party opponents were not required to provide testimony detrimental to their own case. 137Supra note 131. For the European perspective, compare Verkerk, supra note 23, at 246 (“[A] common restriction [of the scope of party interrogation in European systems] is that the law recognizes privileges that grant parties the right to withhold information.”). Such privileges go far beyond attorney client privileges but include the right to refuse to bear witness against oneself. The principal purpose of depositions had been to preserve evidence. 138See Ragland, supra note 123, at 19–24. By the 1920s, a few states tolerated the use of depositions as tools of discovery. 139See, e.g., Cal. Civ. Proc. Code § 2021 (1921); Ragland, supra note 123, at 279. But the oral deposition of parties before trial for the purpose of developing evidence remained highly contentious. 140See Ragland, supra note 123, at 126.

In a radical departure from prior practice, the FRCP incorporated deposition rules of the broadest scope. 141See IFR Proceedings, supra note 112, at 278–80. In the 1938 Rules, Rule 26 on depositions was the most important discovery rule, and the discovery section was entitled “Depositions and Discovery.” See id. at 57. Neither leave of court nor the presence of a judge or commissioner was required. 142Fed. R. Civ. P. 26(a) (1938). Contrast this with civil law rules that require both. Verkerk, supra note 23, at 193 (“A personal appearance of the parties is a court hearing . . . .”). “[O]n the continent the party seeking discovery has to convince the court that there is a good reason to order the discovery.” Id. at 247. Both supporting and hostile witnesses could now be deposed, including party opponents and their employees. 143Fed. R. Civ. P. 26 (1938). Parties were required to appear on simple notice. 144Id. Third parties could be subpoenaed directly without leave of court. 145Fed. R. Civ. P. 30(a) (1938). Attorneys could directly examine the witnesses. 146Fed. R. Civ. P. 26(b) (1938). Contrast this with civil law rules making the judge the principal examiner. Verkerk, supra note 23, at 247–48; see also Langbein, supra note 23, at 826–30.

Under the current rules, examination proceeds as at trial. 147Fed. R. Civ. P. 30(c). But unlike at trial, cross-examination during deposition is not limited to matters raised on direct. Both parties can cross-examine a witness. And examiners can ask questions going far beyond what can be asked at trial. Witnesses must answer any question “relevant to the subject matter [now “claims and defenses”] involved in the action,” including questions seeking inadmissible evidence, such as hearsay. 148Fed. R. Civ. P. 26(b), 30(c). The deposition is recorded verbatim, allowing the examiner to pin down the witness. 149See 2 James Wm. Moore & Joseph Friedman, Moore’s Federal Practice Under the New Federal Rules § 26.01, at 2443 (1938) (“A party or witness whose deposition has been taken at an early stage in the litigation cannot, at a later date, readily manufacture testimony in contradiction to his deposition.”). The testimony can be used as evidence at summary judgment, at any hearing, or at trial. 150Fed. R. Civ. P. 32(a). Other deposition testimony is also admissible, so long as it is otherwise admissible under the Federal Rules of Evidence. See, e.g., Fiber Sys. Int’l, Inc. v. Roehrs, 470 F.3d 1150, 1160 (5th Cir. 2006). Testimony can be taken from witnesses who are not called at trial. 151See Fed. R. Civ. P. 27(a)(4) (“[A] deposition to perpetuate testimony may be used . . . in any later-filed district-court action involving the same subject matter . . . .”). In civil law jurisdictions, such as Germany, employees have a contractual duty to answer questions from an employer, e.g., in connection with a company internal investigation, but former employees or third parties have no obligation to answer questions outside of court. See Hanna Blanz et al., Investigations in Germany, in Corporate Internal Investigations: An International Guide, supra note 53, at 287, paras. 6.168–.170, at 332, paras. 6.180–.186, at 335–36. Examinations can last several days.

The rules specifically contemplate the need to obtain information from large organizations without knowing who in the organization has the information sought. 152See Fed. R. Civ. P. 30(b)(1). A party can issue a deposition notice to a witness whose “name is unknown” by “provid[ing] a general description sufficient to identify the person or the particular class or group to which the person belongs.” Id. This is not possible under civil law rules, because taking the examination of a witness under oath is a judicial prerogative. Moreover, there is no obligation to answer questions by an attorney absent a specific contractual relationship, such as that between employer and employee. See, e.g., Blanz et al., supra note 151, paras. 6.180–.186, at 335–36. A special provision requires “a public or private corporation” to “designate one or more officers, directors, or managing agents, or designate other persons who consent to testify on its behalf” to “testify about information known or reasonably available to the organization” on topics designated by the requesting party. 153Fed. R. Civ. P. 30(b)(6).

The significance of this discovery tool cannot be overstated. The power to force a party opponent to sit for deposition under oath, answer virtually every question put to him by opposing counsel, confront him with documentary evidence, and have his testimony recorded verbatim, was unprecedented. It shocked at least one member of the original Advisory Committee, as the following transcript shows:

[Sen. George Wharton Pepper]. . . . [T]his sort of power given to a plaintiff is simply going to be used as a means of ruining the reputation of responsible people. You bring a suit against a man, without any ground whatever—the president of some important company, the president of a utilities company or a bank or something. You take his deposition, have the reporters present, and grill him in the most unfair way, intimating that he is a burglar or murderer, or this, that, and the other. He has no redress, and the next morning the papers have a whole lot of front-page stuff. . . .

[Chairman Mitchell]. It is too much like some of these Senate committees you used to sit on. (Laughter)

[Sen. Pepper]. Exactly; and that is where I got a taste of the kind of lawlessness that ruins people’s reputations without the opportunity ever to redress the harm that is done. 154Subrin, supra note 111, at 721 (quoting Proceedings of the Meeting of the Advisory Committee on Rules for Civil Procedure of the Supreme Court of the United States (Feb. 22, 1936), in Records of the U.S. Judicial Conference: Committees on Rules of Practice and Procedures, 1935–1988, at CI-209-59 to CI-209-60).

The Committee thus recognized that the new deposition rules would give plaintiffs much the same authority as a congressional committee. It had taken precisely this type of authority to investigate and uncover allegedly coercive, deceptive, and collusive practices among corporate and financial elites in the period before the rise of private enforcement through civil litigation. 155In 1913, the Congressional Pujo Committee investigated the money trusts, concluding that in almost “all great corporations with numerous and widely scattered stockholders[,] . . . management is virtually self-perpetuating and is able through the power of patronage, the indifference of stockholders[,] and other influences to control a majority of the stock.” H.R. Rep. No. 62-1593, at 147 (1913). See generally Louis D. Brandeis, Other People’s Money and How the Bankers Use It 2–3 (1914) (relying heavily on the Pujo Committee’s findings). These and other confrontations in the 1920s informed the debate about corporate governance problems arising from the separation of ownership and control. See, e.g., Adolf A. Berle, Jr. & Gardiner C. Means, The Modern Corporation and Private Property 86–88, 124 (1932). But see Harwell Wells, The Birth of Corporate Governance, 33 Seattle U. L. Rev. 1247, 1248 (2010) (suggesting that Berle and Means’s work “was not a radical break from earlier thought but the end product of several decades in which [scholars] tried to understand both the governance problems of the new modern corporations and what impact those corporate governance problems had on the nation’s public”). There has been some debate as to whether the findings of the Pujo Committee and of subsequent attacks on the money “trusts” by the Committee on Banking and Currency (1932–1934), also known as the Pecora Hearings, were misguided. See Guolin Jiang, Paul G. Mahoney & Jianping Mei, Market Manipulation: A Comprehensive Study of Stock Pools, 77 J. Fin. Econ. 147, 153–54 (2005) (finding informed trading and no manipulation in an empirical analysis of stock pools in the 1920s).

Document discovery was, until 1946, subject to a narrow scope permitting discovery of documents “which constitute or contain evidence material to any matter involved in the action,” 156Compare Fed. R. Civ. P. 34 (1938) (amended 1946) (emphasis added), with id. (1946) (amended 1970) (permitting discovery of documents “which constitute or contain evidence relating to any of the matters within the scope” (emphasis added)). and, until 1970, available only on motion based on a showing of “good cause.” 157Compare Fed. R. Civ. P. 34 (1946) (amended 1970), with id. (1970) (amended 1980) (removing the requirement of a showing of good cause and allowing that “[a]ny party may serve any other party a request . . . within the scope of [the rules]”). Both limitations were removed in 1970, dramatically expanding document discovery; the current rules allow unlimited document requests covering any “nonprivileged matter that is relevant to any party’s claim or defense.” 158Fed. R. Civ. P. 26(b); see also Fed. R. Civ. P. 34 (allowing document discovery within the scope of Fed. R. Civ. P. 26(b)); Thompson v. Dept. of Hous. & Urban Dev., 199 F.R.D. 168, 172 (D. Md. 2001) (explaining that while the scope of discovery became narrower after “subject matter” in Rule 26(b) was changed to “claims and defenses,” debating the difference between the two is “equivalent to debating the number of angels that can dance on the head of a pin”). See generally 6 James Wm. Moore et al., Moore’s Federal Practice § 26.43 (3d ed. 2013) (discussing the scope of discovery after the 2010 amendments). Relevance—as opposed to some property or other specific right to a document, as under European civil law 159Murray & Stürner supra note 25, at 277 (“[Prior to the recent 2001 reforms of the German code of civil procedure,] there was no right to production if the request was based solely on the document’s relevance to a decisive issue in the case. . . . If a party had a substantive right to possession of a document in the hands of a third party, that party could request that the proceeding be stayed to permit him to exercise his right to possession . . . .”); Verkerk, supra note 23, at 31 (discussing “common document” principle). —now defines the scope of discovery. 1608 Charles Alan Wright et al., Federal Practice and Procedure § 2008 (3d ed. 2010). Significantly, “relevance” is defined very broadly as “‘any matter that bears on, or that reasonably could lead to other matter that could bear on, any issue that is or may be in the case.’” 161Hager v. Graham, 267 F.R.D. 486, 491 (N.D. W. Va. 2010) (quoting Hickman v. Taylor, 329 U.S. 495, 501 (1947)) (internal quotation marks omitted); cf. Oil, Chem. & Atomic Workers Local Union No. 6-418 v. NLRB, 711 F.2d 348, 359–60 (D.C. Cir. 1983).

This broad conception of relevance extends to all of discovery, including party-on-party interrogatories, document requests, witness examinations, requests for admissions, expert discovery, and third-party subpoenas. The combination of these tools, which can, for the most part, be used in any order, enable U.S. attorneys to design extremely effective fact investigations. 162These tools are not available in this form in civil law adjudication. See infra notes 524–37. Of special significance, perhaps not entirely recognized by the original 1938 rules, is that the extraordinary authority to force the other party to produce all of its relevant documents makes depositions much more effective. It is by confronting a witness with her own documents on the record, under oath, and with the help of cross-examination that an examiner can extract important admissions. Moreover, the use of interrogatories and the opportunity to depose record-keepers early during discovery provide all-important information about the existence and location of documents. 163In civil law countries, parties have no such means to identify discoverable information. See infra notes 537–39 and accompanying text.

The Federal Rules of Civil Procedure’s broad scope of discovery has had an especially far-reaching impact on document discovery in the age of e-mail and electronically stored information (ESI), beginning in the late 1980s. 164See James N. Dertouzos et al., The Legal and Economic Implications of Electronic Discovery: Options for Future Research 1 (2008) (describing the vast amount of discoverable data available in an electronic world). The 2013 rules do not place any specific limits on the kind or quantity of information that may be obtained via document requests. 165See Fed. R. Civ. P. 34. In practice, this means that blanket requests for “all documents”—including all electronic records—will be propounded, and that a responding party may, for example, need to produce all electronic records and communications that reference the names of certain employees within relevant time periods. As is well known to U.S. observers, this routinely leads to the disclosure of millions of documents and e-mails by corporate defendants in large cases. While this statistic is vaguely appreciated outside the United States, the full extent and logistical requirements of litigation discovery in a U.S. securities class action or shareholder derivative action are still mind-numbing for attorneys trained in a civil law jurisdiction. 166Subrin, supra note 22, at 307, 311.

C. The Devolution of Judicial Authority onto Private Parties

A critical feature of the modern rules as they eventually evolved by the 1970s is the “self-executing” nature of discovery, dramatically reducing the court’s involvement in fact investigation. 167See Hazard, Discovery and the Role of the Judge, supra note 22, at 1018. Discovery is said to be “self-executing,” and experienced litigators know not to bother the judge with motions to compel or protective orders, under Fed. R. Civ. P. 26(b)(2)(B) and 26(c), respectively, unless necessary. But see infra notes 460–66 and accompanying text (discussing the thesis that “case management” rules have produced “managerial judges”). The FRCP allow parties to engage in the exchange of information with very little supervision by the judge. 168See, e.g., Fed. R. Civ. P. 26(a)(1) (requiring parties to provide preliminary information before discovery requests); Fed R. Civ. P. 26(b) (allowing discovery without court interference for relevant matters); Fed. R. Civ. P. 30(a) (allowing parties to orally depose any person without leave of court); Fed. R. Civ. P. 31 (allowing parties to depose any person through written depositions without leave of court). In a departure from prior practice, and contrary to civil law jurisdictions, 169In civil law jurisdictions, parties must petition the court for an order to obtain evidence. See, e.g., Murray & Stürner, supra note 25, at 242 (explaining that the court may order production but is required to “assess the relevance of any such document under the standards applicable to proof in general when issuing such orders”); see also Huang, supra note 24, at 39–42. the burden of showing relevance is on the party resisting discovery, thus avoiding the involvement of the judge altogether until a dispute arises. 170See Fed. R. Civ. P. 26(b)(2)(B), (c); see also, Marcus, supra note 31, at 748 (explaining that prior to the 1970 amendments, Rule 34 document discovery was available only on a motion based on a showing of “good cause”). The party resisting discovery must challenge the request by specific objection, which, in the first instance, is returned to the requesting party in the form of a “response.” 171Fed. R. Civ. P. 34(b)(2). Before seeking an order to compel or a protective order from a judge, the parties must first attempt to resolve their dispute. 172Fed. R. Civ. P. 26(c), 37.

Discovery practice is principally a sustained negotiation between the parties. When one party loses an unreasonable discovery motion, that party must pay the costs of the opposing party, which further discourages involving judges. 173Fed. R. Civ. P. 37(a)(5)(A). Objections based on lack of relevance are disfavored. Orders of protection in document discovery are granted sparingly. 174Hazard, Discovery and the Role of the Judge, supra note 22, at 1018. The results of discovery—interrogatory responses, documents, deposition transcripts, and other information—are not filed with the court, unless provided as evidence on a substantive motion; nor does the court review a majority of the testimony or documents produced during discovery. This shows how decentralized the process of fact investigation is.

The structure of the U.S. litigation process as a whole thus gives litigants control over exploring, collecting, organizing, and presenting information in civil litigation that is peculiar when viewed from the European civil law perspective. The authority to conduct lengthy and probing investigations, and the extrajudicial resources that private parties bring to bear on this process, gives the term “private attorney general”—a term frequently used, but not defined, in the enforcement debate—its full meaning. 175See John C. Coffee, Jr., Rescuing the Private Attorney General: Why the Model of the Lawyer as Bounty Hunter Is Not Working, 42 Md. L. Rev. 215, 215–16 (1983).

U.S. discovery devolves judicial and prosecutorial authority to investigate wrongdoing onto the private parties. 176See id. at 218. This means that the parties need not rely on the initiative or the presence of a judge; they control and perform this time-intensive phase of adjudication on their own. 177See Marcus, supra note 66, at 334 (“Nowhere in this country, nor anywhere in the world, had there previously been such a broad or unconfined opportunity for private litigants to use the power of the state to compel disclosure of information by others.”). This reduces the burden on judges and prosecutors, whose resource constraints are most acutely felt in discovery; expedites the process; and enables more in-depth investigation by parties with the interests, expertise, and resources suited to pursue them. In this way, the U.S. system transfers the investigative authority of the state, and the costs of investigation, to private parties.

As we discuss in the next section, federal rules of discovery influenced practice in state courts, like the Delaware Court of Chancery, which are important venues for corporate litigation.

D. The Influence of Federal Rules of Discovery on State Civil Procedure: The Case of Delaware

The Supreme Court’s decision in Erie Railroad Co. v. Tompkins held that federal courts apply state and federal substantive law, but federal procedural rules. 178304 U.S. 64, 78 (1938). State courts, where most shareholder derivative actions are litigated, apply state procedural rules. Do state procedural rules afford the same kind of probing discovery as the federal rules? The short answer is yes.

While there are differences between federal and state court rules, which can easily trip up attorneys not familiar with local practices, as of 1986, most U.S. jurisdictions had adopted the principles, if not the precise wording, of the liberal federal discovery rules. 179John B. Oakley & Arthur F. Coon, The Federal Rules in State Courts: A Survey of State Court Systems of Civil Procedure, 61 Wash. L. Rev. 1367, 1377–78 (1986). In their 1986 study, comparing state and federal rules, Oakley and Coon found that twenty-two states, plus the District of Columbia, could be classified as having procedural systems that were true replicas of the FRCP. These are Alabama, Alaska, Arizona, Colorado, District of Columbia, Hawaii, Indiana, Kentucky, Maine, Massachusetts, Minnesota, Montana, New Mexico, North Dakota, Ohio, Rhode Island, South Dakota, Tennessee, Utah, Vermont, Washington, West Virginia, and Wyoming. Id. at 1377. Further, ten states, including Arkansas, Delaware, Georgia, Idaho, Kansas, Mississippi, Nevada, North Carolina, Oklahoma, and South Carolina largely replicated the federal rules, except for slight variations or codifications that had nothing to do with the nature or conduct of discovery. Id. at 1378. Oakley and Coon concluded that in a majority of states there was “‘but one procedure for state and federal courts.’” Id. at 1427 (quoting 1 W. Barron & A. Holtzoff, Federal Practice and Procedure § 9, at 44–45 (Wright ed. 1960)). The FRCP were adopted in toto with some minor differences in at least thirty-five states, including Delaware. 180See id. at 1378. But most jurisdictions, including the ones that follow their own rules, like California, New Jersey, and New York, do not significantly differ from the federal rules with regard to the scope of discovery or the available tools and procedures. 181See John B. Oakley, A Fresh Look at the Federal Rules in State Courts, 3 Nev. L.J. 354, 355 (2002/2003) (finding that states failed to keep up with the federal rules changes). Oakley concludes that “significant” amendments were frequently rejected, id. at 359, 383, but a careful examination of the differences he describes between state and federal rules shows that these differences, while important to recognize and apply with precision for attorneys litigating in these courts, are minor when compared to the fundamental differences between the U.S. system and civil law jurisdictions. See infra notes 528–39 and accompanying text.

While there are fewer “exact replica” states today, this is because states have not felt the need to keep up with the many rule changes that the federal rules have undergone in 1993, 2000, 2006, 2007, 2010, and 2013. 182See Oakley, supra note 180, at 355, 383 (“Even among states that fifteen years ago could be counted as substantially conforming to the federal model of procedure, recent significant amendments have been more frequently rejected or ignored than adopted.”); see also Fed. R. Civ. P. historical note at xi–xii. These rule changes, however, did not expand discovery; if anything, they attempted to expedite discovery and limit discovery excesses by, inter alia, requiring parties to agree on a discovery plan early in the case, 183Fed. R. Civ. P. 26(f). insisting on judicial involvement in approving such a plan, 184Fed. R. Civ. P. 26(f)(2). ever so slightly limiting the definition of relevancy in the text of the rules, 185Supra note 158; see also Sanyo Laser Prods., Inc. v. Arista Records, Inc., 214 F.R.D. 496, 500 (S.D. Ind. 2003). and adding rules to address the special issues of e-discovery. 186Fed. R. Civ. P. 26(b)(2)(B). But the basic version of the FRCP prevails in all states. 187Thomas D. Rowe, Jr., Suzanna Sherry & Jay Tidmarsh, Civil Procedure 18 (3d ed. 2012).

This is true of Delaware. Delaware’s discovery rules track those in the FRCP by number and in content. 188Oakley, supra note 179, at 379–81 (“Delaware has separate but substantially identical rules of civil procedure for each of its three principal systems of courts: the Court of Chancery, the Superior Court, and the inferior civil court of non-equitable jurisdiction, the Court of Common Pleas.”). Compare, e.g., Del. Ch. Ct. R. 11 (sanctions), Del. Ch. Ct. R. 12 (motion to dismiss), Del. Ch. Ct. R. 26 (discovery generally, scope, privileges, expert witnesses, timing, limitations), Del. Ch. Ct. R. 30 (oral depositions), Del. Ch. Ct. R. 31 (written depositions), Del. Ch. Ct. R. 32 (use of deposition testimony), Del. Ch. Ct. R. 33 (interrogatories), Del. Ch. Ct. R. 34 (production of documents and things), Del. Ch. Ct. R. 35 (physical and mental examinations of persons), Del. Ch. Ct. R. 36 (requests for admissions), Del. Ch. Ct. R. 37 (discovery disputes and sanctions), and Del. Ch. Ct. R. 45 (subpoena power), with Fed. R. Civ. P. 11, 12, 26, 30–37, 45. But they are, in certain respects, more permissive than the ones presently in effect in federal court. Thus, for example, Delaware Chancery Court Rule 26 is almost identical to the version of Federal Rule 26 from the 1970s, when the philosophy of liberal discovery was at its apex. 189Oakley, supra note 179, at 380 (“Delaware Court of Chancery Rule 26 is a virtually exact copy of the 1970 version of Federal Rule 26, and does not incorporate any part of later amendments.”). The same is true for other jurisdictions. Id. at 362, 365–66, 369–70, 375. And Delaware Chancery Court Rule 26(b) still retains the broader “subject matter discovery” language that was excised from the Federal Rules in 2010 to slightly limit discovery to “the claims and defenses” of the parties, unless the judge finds there is good cause for obtaining additional subject matter discovery. 190Compare Del. Ch. Ct. R. 26(b)(1) (“Parties may obtain discovery regarding any matter, not privileged, which is relevant to the subject matter involved in the pending action, whether it relates to the claim or defense of the party seeking discovery or to the claim or defense of any other party, including the existence, description, nature, custody, condition and location of any documents, electronically stored information, or tangible things and the identity and location of persons having knowledge of any discoverable matter. It is not ground for objection that the information sought will be inadmissible at the trial if the information sought appears reasonably calculated to lead to the discovery of admissible evidence.” (emphasis added)), with Fed. R. Civ. P. 26(b)(1) (allowing for discovery of any nonprivileged matter relevant to a party’s claim or defense but disposing of the “subject matter” language). Similarly, the chancery rules have no presumptive limits on the number of depositions or the number of interrogatories, as the Federal Rules now do. 191Compare Del. Ch. Ct. R. 30(a), 33(a), with Fed. R. Civ. P. 30(a)(2)(A)(i), 33(a)(1). But the Delaware Chancery amended its electronic discovery rules in 2013 to bring them in line with similar amendments to the FRCP. 192Leo E. Strine, Jr., Delaware Court of Chancery Announces Rule Changes and New Discovery Guidelines, Metropolitan Corp. Couns., May 2013, at 35, 35, available at http://www.metrocorpcounsel.com/pdf/2013/May/35.pdf.

In short, the nature and scope of discovery in civil litigation is very similar in federal and state courts. The tools of discovery deployed in federal courts (including interrogatories, depositions, the authority of attorneys to issue subpoenas, and the right to obtain all nonprivileged information, including ESI) are also available in state courts. Discovery is broad in scope, conducted by the parties themselves in accordance with the rules, and performed with minimal involvement by a judge. As Delaware shows, this does not mean that state rules are exact replicas of the federal rules, but that they operate in substantially the same way and produce substantially the same results. And, if anything, discovery in federal courts can sometimes be less permissive than in state courts.

III. Consequences of Discovery for U.S. Corporate Governance Institutions

This Part describes in detail how litigation discovery has shaped U.S. corporate governance institutions.

Section A shows how the practices of offensive and defensive discovery generate and disseminate company internal information relevant to the determination of management misconduct at every stage of shareholder litigation, even prior to the beginning of formal discovery. It argues that the process of discovery generates positive information externalities and disciplines management ex post and ex ante.

Section B briefly discusses how the rules and practices of discovery have shaped mechanisms of internal corporate governance that protect against mismanagement and wrongdoing.

Section C describes changes in substantive law that have been influenced by discovery. We first examine developments of fiduciary duty doctrine, such as the “proceduralization” of the boardroom in the wake of Smith v. Van Gorkom. 193488 A.2d 858 (Del. 1985), overruled in part on other grounds by Gantler v. Stephens, 965. A.2d 695 (Del. 2009). Second, we describe the gradual transformation of shareholder information rights into vehicles for prelitigation discovery. Third, we consider the impact of information obtained through litigation discovery on changes in securities regulation.

Section D concludes that litigation discovery complements federal securities disclosure and serves as a form of ex post disclosure.

This Part shows that, even if one concedes that discovery is not cost-effective in every individual case (and abusive in some), discovery’s systemic benefits are more significant than heretofore understood.

A. Shareholder Litigation

Corporate law scholars have acknowledged the role of procedural rules in mitigating the risk of false positives (Type I Errors) and false negatives (Type II Errors) in the enforcement of corporate and securities laws. 194Stout, supra note 42, at 711–12; see also Stephen J. Choi & Robert B. Thompson, Securities Litigation and Its Lawyers: Changes During the First Decade After the PSLRA, 106 Colum. L. Rev. 1489, 1500 (2006) (discussing “empirical findings on the merits of post-PSLRA claims”); Sherrie R. Savett, The Merits Matter Most and Observations on a Changing Landscape Under the Private Securities Litigation Reform Act of 1995, 39 Ariz. L. Rev. 525, 531 (1997) (observing that the PSLRA “produces great delay in getting the case moving to the merits”); Thompson & Thomas, supra note 46, at 1758 (2004) (describing efforts to deter “strike” or “nuisance” suits). Commenting on the burdens imposed on securities class action plaintiffs by the Private Securities Litigation Reform Act of 1995, Lynn A. Stout writes that “an example of a . . . false positive would be a judicial finding that a defendant had fraudulently misrepresented something, when in fact no fraud occurred.” 195Stout, supra note 42, at 711. In contrast, a “false negative occurs when a court trying to decide whether the defendant has committed fraud mistakenly finds there has been no fraud, even though fraud actually occurred.” 196Id. False positives lead to overdeterrence, whereas false negatives lead to underdeterrence. The risk of false positives could be affected by legislation that heightened or lowered burdens of proof, demanded or forgave heightened pleading standards for enforcers, required that bond be posted by plaintiffs for costs incurred as a result of litigation, adopted “loser pays” fee-shifting rules, or employed sanctions to deter marginal lawsuits. 197See Amanda M. Rose, The Multienforcer Approach to Securities Fraud Deterrence: A Critical Analysis, 158 U. Pa. L. Rev. 2173, 2192 (2010). The PSLRA used all of these methods to deter strike suits, or, in Stout’s terminology, avoid Type I errors. See Stout, supra note 42, at 711. On the other hand, Type II errors would increase where “meritorious suits [are] thrown out of court because plaintiffs, without discovery, cannot offer sufficient evidence of fraud.” 198Stout, supra note 42, at 712.

But neither Stout nor others 199See, e.g., Rose, supra note 197, at 2192. have appreciated that the availability of discovery per se in the United States (as compared to its absence in other countries) attacks the problem of false positives and false negatives at its core, because discovery reveals company internal information that bears directly on the merits of the parties’ claims and defenses. 200Indeed, the elimination of Type I and Type II errors was foremost in the minds of the drafters of the Federal Rules. In an extended discussion of New York practice, for example, George Ragland notes “the ineffectiveness of restrictions upon the scope of the discovery as an aid in arriving at the truth.” Ragland, supra note 123, at 130. In particular, he cites the New York limitation that “the defendant can have no discovery except on his affirmative defenses.” Id. at 132. Broad discovery reduces both false positives and false negatives by increasing accuracy, because it affords access to critical information. The better the information, the greater the accuracy, and the less likely a judge is to conclude from a single rotten apple that every apple in the barrel has worms.

In re Caremark International Inc. Derivative Litigation, 201698 A.2d 959 (Del. Ch. 1996). which we discuss below, provides a good example of how discovery avoids Type I errors. In Caremark, the board was exonerated in a settlement that took place after discovery established that the board had not engaged in systematic oversight failures, in spite of $250 million in fines and payments the corporation had to pay to public and private parties because employees of the company criminally violated federal and state health care laws. Chancellor Allen could make these findings in the civil case only because of extensive civil discovery into what the board knew and what systems of monitoring and reporting the company had in place. 202Id. at 960–61, 971 (“Concerning the possibility that the Caremark directors knew of violations of law, none of the documents submitted for review, nor any of the deposition transcripts appear to provide evidence of it.”).

Litigation discovery is integral to private enforcement and defines shareholder litigation in concrete and specific ways. Shareholder derivative actions are subject to different rules that intricately calibrate under what circumstances shareholders will be given the opportunity to obtain discovery against directors or officers of a corporate defendant. Specialized judicial doctrines and procedural rules have emerged in this context. The business judgment rule, 203See, e.g., Smith v. Van Gorkom, 488 A.2d 858, 872 (Del. 1985) (explaining that the business judgment rule “is a presumption that in making a business decision, the directors of a corporation acted on an informed basis, in good faith and in the honest belief that the action taken was in the best interests of the company” (quoting Aronson v. Lewis, 473 A.2d 805, 812 (Del. 1984), overruled on other grounds by Brehm v. Eisner, 746 A.2d 244 (Del. 2000)) (internal quotation marks omitted)), overruled in part on other grounds by Gantler v. Stephens, 965. A.2d 695 (Del. 2009); Bayer v. Beran, 49 N.Y.S.2d 2, 6 (Sup. Ct. 1944). Absent fraud, illegality, or self-dealing, the business judgment rule ordinarily bars courts from reviewing the decisions of corporate management for breach of fiduciary duty. Van Gorkom, 488 A.2d at 872–73. the demand requirement, 204See Aronson, 473 A.2d 805, 811–12; see also Fed. R. Civ. P. 23.1 (setting pleading requirements for shareholder derivative actions). The demand requirement forces a potential plaintiff to first make a demand on the company’s board of directors to remedy the alleged misconduct, or to allege why such demand would be futile. See Charles R.T. O’Kelley & Robert B. Thompson, Corporations and Other Business Associations: Cases and Materials 369 (5th ed. 2006). and the rule of Auerbach v. Bennett 205393 N.E.2d 994, 999–1000 (N.Y. 1979). Delaware followed a substantially similar rule in Zapata Corp. v. Maldonado, 430 A.2d 779, 786 (Del. 1981). —authorizing dismissal based on a corporate defendant’s nonadversarial self-investigation by a special litigation committee (SLC) of the board 206See Robert Charles Clark, Corporate Law § 15.2.3, at 646–47 (1986). —all considerably raise the threshold for obtaining discovery against a corporate defendant.

The hurdles to obtaining discovery that plaintiffs face have developed over time, mostly in response to a new wave of shareholder derivative actions during the 1970s. 207See Choi & Thompson, supra note 194, at 1492–93. Perhaps not incidentally, the 1970s amendments to the FRCP represented the highpoint of liberal discovery, generating backlash by the business community. 208Marcus, supra note 31, at 748.

Subsequent developments in securities litigation resulted in the passage of the Private Securities Litigation Reform Act of 1995. 209Pub. L. No. 104-67, 109 Stat. 737 (1995) (codified as amended in scattered sections of 15 U.S.C.). The PSLRA raised the pleading requirements for securities fraud actions and instituted other measures, such as a mandatory stay of discovery during the pendency of any motions to dismiss, in response to a wave of securities class actions during the late 1980s and early 1990s. 210H.R. Rep. No. 104-369, at 31–32 (1995) (Conf. Rep.), reprinted in 1995 U.S.C.C.A.N. 730, 731. See generally Eugene Zelensky, Recent Legislation, New Bully on the Class Action Block—Analysis of Restrictions on Securities Class Actions Imposed by the Private Securities Litigation Reform Act of 1995, 73 Notre Dame L. Rev. 1135, 1135–37 & n.19 (1998) (discussing legislative history). When plaintiff-side attorneys responded by filing actions in state court, Congress passed the Securities Litigation Uniform Standards Act of 1998 (SLUSA), which preempted most state securities litigation. 211Pub. L. No. 105-353, §§ 2(2), (5), 112 Stat. 3227, 3227 (1998) (codified as amended in scattered sections of 15 U.S.C.). SLUSA required that all federal securities actions be litigated in federal courts, to reinforce the PSLRA’s restrictions on discovery. 212Id. § 101(a), 112 Stat. at 3227–33. State cases filed in the aftermath of the PSLRA often had nearly identical claims to those brought by the same law firm in federal court. 213See H.R. Rep. No. 105-803, at 14 (1998) (Conf. Rep.). SEC Chairman Arthur Levitt pointed out that “[i]t is reasonable to assume that these cases were filed primarily to get discovery for use in the Federal action.” 214Securities Litigation Abuses: Hearing Before the Subcomm. on Sec. of the S. Comm. on Banking, Hous., & Urban Affairs, 105th Cong. 49 (1997) (statement of Arthur Levitt, Chairman, SEC); see also Randall S. Thomas & Kenneth J. Martin, Using State Inspection Statutes for Discovery in Federal Securities Fraud Actions, 77 B.U. L. Rev. 69, 71 (1997). The passage of SLUSA prevented plaintiffs from avoiding the discovery stay in federal court (because state courts would apply their own procedural rules).

Significantly for this Article’s thesis, the political battles over securities class actions in the 1990s were about whether plaintiffs would get past the pleading stage and obtain discovery. While the corporate defense lobby successfully argued that the high discovery costs effectively allowed plaintiff-side firms to blackmail defendants into settling cases without merit, this is only one side of the story. As the plaintiff-side lobby argued in the debate about the PSLRA and SLUSA, discovery is what allows plaintiffs to expose management misconduct and mismanagement.

In sum, the special hurdles to obtaining discovery have evolved in response to the special threat of discovery for corporate defendants in probes of corporate internal wrongdoing.

The threat of discovery emanates from multiple factors. Discovery is a costly procedural mechanism per se. 215First, the parties must complete all discovery before trial. Therefore, they gather as much information as possible. See generally Fed. R. Civ. P. 26 (a)(3)(B) (timing); Fed. R. Civ. P. 26(b)(1) (scope of discovery). See supra note 42 on the costs of discovery. Moreover, unlike civil law jurisdictions where the loser pays, U.S. procedure requires the parties to pay their own costs. 216Hensley v. Eckerhart, 461 U.S. 424, 429 (1983) (affirming the “American Rule”). Thus, even if a defendant corporation succeeds on the merits in a shareholder action, it is saddled with its own litigation costs—which largely flow from discovery. In most shareholder actions, the corporate defendant’s discovery costs will be higher than the plaintiff’s discovery costs, because the corporation is the party being investigated. It is the defendant corporation that must search for, review, and produce almost all of the documents and witnesses.

But the threat of discovery is not limited to the high litigation costs incurred by corporate defendants. Quite apart from the costs, discovery increases chances that management will face liability for wrongdoing—whether related or unrelated to the claims set forth in any given complaint. Even if the threat of personal liability is remote, 217See Black et al., Outside Director Liability, supra note 44, at 1059; Cheffins & Black, supra note 44, at 1465. discovery may uncover mistakes and misjudgments, some of which may have to be publicly acknowledged and corrected. 218M. Todd Henderson, Impact of the Rakoff Ruling: Was the Judge’s Scuttling of the SEC/BofA Settlement Legally Pointless or Incredibly Important—or Both?, Wall St. Law., Nov. 2009, at 1, 6 (“[A] suit generates not only legal costs but also negative publicity and the potential that even more damning information will be revealed during discovery or the trial.”). Such mistakes may result in direct and indirect financial penalties. Executives may be forced to forgo expected compensation. 219Id. at 6 (noting that Bank of America’s CEO, Kenneth Lewis, was forced to “leav[e] the bank, after reportedly being forced to give up many millions of dollars in compensation as a result of the suit and the handling of the merger with Merrill”). Note that this was nearly a year before the court ruled partially against defendants’ motion to dismiss in the plaintiff shareholders’ action. See In re Bank of Am. Corp. Sec., Derivative, & ERISA Litig., 757 F. Supp. 2d 260, 269–70 (S.D.N.Y. 2010); Tami Luhby, Bank of America CEO Ken Lewis to Retire, CNNMoney (Sept. 30, 2009, 6:53 PM), http://money.cnn.com/2009/09/30/news/companies/bank_of_america_ken_lewis_resigns/index.htm. And because executives often hold significant quantities of stock options they are also vulnerable to any decline in their company’s stock price that may result from revelations of mismanagement. Discovery, moreover, threatens management with reputational costs, diminishing the value of an executive’s most important asset. 220See Chelsea C. Liu, Corporate Litigation, Corporate Governance Restructuring, and Executive Compensation 221–81 (Jan. 2013) (unpublished Ph.D. dissertation, University of Adelaide), http://digital.library.adelaide.edu.au/dspace/bitstream/2440/81971/3/02whole.pdf (reviewing literature on executive reputational damages resulting from litigation and providing empirical evidence). Finally, management also faces opportunity costs as a result of discovery.

The rules governing shareholder litigation are thus structured by determinations about when to allow, and when to deny, access to discovery. Special features that characterize contemporary shareholder litigation in the United States—and that are generally taken for granted by U.S. scholars—emerged and developed because of the unique role that discovery plays in U.S. civil procedure. These characteristics do not exist in other countries, because discovery does not exist.

Some have objected to our thesis on the central importance of discovery by arguing that discovery materials, like corporate internal documents, e-mails, and witness testimony, generally do not become available to nonlitigants. Where discovery materials are subject to confidentiality agreements, only the litigating parties have access to these materials. And the public only gains access to information obtained through discovery where such information is disclosed during hearings in open court—principally at trial. But cases are almost always settled before trial. Therefore, discovery cannot generate the kind of information externalities or disciplinary results that we suggest.

These issues are addressed in subsections 1 through 5 below.

In response, we first note that some shareholder derivative actions, like Disney 221Disney II, 825 A.2d 275 (Del. Ch. 2003). and Van Gorkom, 222Smith v. Van Gorkom, 488 A.2d 858 (Del. 1985), overruled in part by Gantler v. Stephens, 965. A.2d 695 (Del. 2009). do go to trial, as do some securities class actions, like the Apollo Group Securities litigation, 223See In re Apollo Grp. Inc. Sec. Litig., No. CV-04-2147-PHX-JAT, 2008 WL 410625, at *1 (D. Ariz. Feb. 13, 2008). —although this is admittedly rare. 224See Kevin LaCroix, Plaintiffs Prevail in Mixed Jury Verdict in Household International Securities Fraud Trial, D&O Diary (May 7, 2009), http://www.dandodiary.com/2009/05/articles/securities-litigation/plaintiffs-prevail-in-mixed-jury-verdict-in-household-international-securities-fraud-trial/ (“According to data compiled by the Securities Litigation Watch, [as of 2009] only 21 cases have gone to trial since the PSLRA was enacted in 1995.” (citation omitted)). We address such cases in subsection 1.

Discovery, however, is presumptively public before and after trial. 225Supra notes 67–68. Plaintiffs are not required to enter confidentiality agreements that cover documents and testimony produced during discovery. Moreover, confidentiality agreements have their limitations. There are many examples of discovery materials becoming public before trial, including materials that are marked “confidential.” 226See, e.g., Karen Gullo, Goldman, Merrill E-Mails Show Naked Shorting, Filing Says, Bloomberg (May 16, 2012, 5:34 PM), http://www.bloomberg.com/news/2012-05-15/goldman-merrill-e-mails-show-naked-shorting-filing-says.html. See also, for example, Legacy Tobacco Documents Library, U.C.S.F., http://legacy.library.ucsf.edu (last visited May 28, 2014), which contains “more than 14 million documents (80+ million pages) created by major tobacco companies related to their advertising, manufacturing, marketing, sales, and scientific research activities.” A simple search turns up more than two million documents, which (based on spot checks) are Bates-stamped, and some of which are marked “Confidential: Tobacco Litigation.” Two situations in which this occurs are on summary judgment, where all materials that are relied upon as evidence must be disclosed, 227For example, in the Bank of America case, Neil Cotty’s deposition transcript was marked “confidential,” but was nonetheless filed, and available online, as an exhibit to the plaintiffs’ motion for summary judgment. See Continued Videotaped Deposition of Neil Andrew Cotty, supra note 69. and during settlement of derivative litigation and class actions, which require supporting discovery. We discuss these scenarios in subsections 2 and 3.

Finally, even if discovery materials are not filed with the court, explored during settlement hearings, or otherwise made available to the general public, discovery nonetheless generates detailed information about potential corporate internal wrongdoing during defensive discovery and during prediscovery internal investigations. In both cases, such information is shared with gatekeepers, and typically with regulators and parties to any settlement. We discuss these scenarios in subsections 4 and 5.

1. Cases That Go to Trial

Critics of shareholder litigation have focused on discovery costs, 228See sources supra note 42. especially where, as in the Disney shareholder derivative action, directors avoid personal liability. 229Stephen M. Bainbridge, Corporation Law and Economics § 8.5, at 403−04 (2002) (“In almost all cases, the legal fees collected by plaintiff counsel exceeded the monetary payment to shareholders. . . . A radical solution would be elimination of derivative litigation.”). But see Thompson & Thomas, supra note 46, at 1172−73. Recent empirical studies show that managers of public companies are rarely subject to personal liability, and liability for outside directors is almost nonexistent. 230See Black et al., Liability Risk for Outside Directors, supra note 44; Black et al., Outside Director Liability, supra note 44; Cheffins & Black, supra note 44.

But our discussion of the Disney litigation shows that, independent of any remedy, shareholder actions reveal and publicize substantial information about corporate internal practices, controls, judgments, and failures, including information about culpable wrongdoing and wrongdoing that falls just short of liability. The knowledge and information produced by discovery about what really happened and how executives run their companies is an informational public good that private enforcement generates for markets, institutional investors, regulators, courts, gatekeepers, self-regulatory organizations, and the general public. 231See Gorga & Halberstam, supra note 56, at 1168−69 (discussing the public goods characteristics of knowledge).

The public good of information generated by shareholder litigation is illustrated by the Disney example. The Disney court articulated new standards of fiduciary duty in board decision-making, relying on the facts revealed by discovery, which became part of a much broader corporate governance discussion about executive pay practices. 232Supra Part I.B. Discovery in this case, and many other cases, such as Enron, Worldcom, and Tyco, informed the broader executive pay discussion, which ultimately led the SEC to require that public companies include a compensation, disclosure, and analysis (CDA) section in their periodic disclosures. 233See Executive Compensation and Related Person Disclosure, Securities Act Release No. 8732A, Exchange Act Release No. 54302A, Investment Company Act Release No. 27444A, 71 Fed. Reg. 53,158 (Sept. 8, 2006); see also Jeffrey N. Gordon, Executive Compensation: If There’s a Problem, What’s the Remedy? The Case for “Compensation Discussion and Analysis, 30 J. Corp. L. 675, 692 (2005) (affirming that “[t]he Disney litigation, revived in the post-Enron environment, has become, regardless of outcome, an extended morality tale on the board’s responsibility to monitor executive compensation” (footnote omitted)). As Thompson and Thomas have stated,

Public company suits continue to be filed and to make new law. The impact of decisions in derivative cases like Caremark, Disney, and Oracle goes well beyond the outcome of the cases themselves. These decisions changed the rules for future legal practice by allowing well-motivated legal counselors to get their clients to accept better conduct and procedures. 234Thompson & Thomas, supra note 46, at 1749 (footnotes omitted). For an example of such advice on best practices after Disney, see Kevin M. LaCroix, Board Decision-Making Practices After the Disney Decision, OakBridge InSights (OakBridge Ins. Servs., Bloomfield, Conn.), Apr. 2006, available at http://www.rtspecialty.com/rtproexec/insights/Issue1DisneyArticle.pdf. Oakbridge is a leading directors and officers liability (D&O) insurer.

Just as securities regulation is influenced by litigation discovery, shareholder litigation is also frequently driven by the actions of the securities regulator. Administrative agencies play a crucial role in investigating and requiring information about alleged corporate wrongdoing. This role results in an important interaction between shareholder private litigation and SEC investigations. Plaintiffs’ firms will often file shareholder actions against public companies after the SEC announces an investigation. 235Kevin LaCroix, First the Regulatory Investigation, Then the Securities Suit, D&O Diary (Dec. 9, 2013), http://www.dandodiary.com/2013/12/articles/securities-litigation/first-the-regulatory-investigation-then-the-securities-suit/. In this respect, discovery is not the only mechanism available for ex post information gathering and may be induced by other regulatory actions.

But investigations and enforcement actions by the SEC have serious shortcomings with regard to information production. SEC investigations are generally confidential, and those confronted with an SEC enforcement action will provide the SEC with information, including documents and testimony, subject to a confidential treatment request. Also, unlike a civil complaint, an SEC notice of formal investigation is confidential. 236See Ralph C. Ferrara & Philip S. Khinda, SEC Enforcement Proceedings: Strategic Considerations for When the Agency Comes Calling, 51 Admin. L. Rev. 1143, 1151 (1999); see also Donna M. Nagy, Richard W. Painter & Margaret V. Sachs, Securities Litigation and Enforcement: Cases and Materials 668 (3d ed. 2012); cf. Ferrara & Khinda, supra, at 1149 (“[C]ounsel should request that [during the course of an informal inquiry a witness examination] be done via an interview rather than through testimony, i.e., not under oath and without the use of a court reporter. Interviews are generally preferable whenever parallel proceedings or litigation exists or is likely, as such discussions with the staff will not generate witness transcripts (as would be the case with formal testimony) that may be discoverable . . . .”). Discovery, by contrast, promotes an important informational flow to the judiciary, to the administrative regulatory authorities, and to the public.

Moreover, the SEC cannot pursue all cases that it should. There are many reasons why this is so. The SEC does not have nearly enough staff or budgetary resources to investigate every allegation of wrongdoing. 237See, e.g., U.S. Gen. Accounting Office, GAO-09-358, Securities and Exchange Commission: Greater Attention Needed to Enhance Communication and Utilization of Resources in Enforcement Division 24 (2009) (finding that “resource challenges . . . [have delayed cases], reduc[ed] the number of cases that can be brought, and potentially undermin[ed] the quality of cases”); U.S. Gen. Accounting Office, GAO-02-302, SEC Operations: Increased Workload Creates Challenges 11–19 (2002) (finding that workload growth and limited resources have forced the SEC to be selective in its enforcement, have delayed investigations and case closings, and raise general concerns about SEC's enforcement; the SEC cannot pursue every case that it should); Joshua Gallu, SEC Enforcement Story Doesn't Add Up for 2011, Bloomberg (March 2, 2012, 12:00 AM), http://www.bloomberg.com/news/2012-03-02/sec-accounting-of-record-enforcement-year-in-2011-doesn-t-add-up.html (claiming that SEC filed fewer enforcement actions in 2011 than in 2009). One response is to dramatically increase the resources of the SEC. See Rose, supra note 197, at 2209–10. But the SEC’s congressional appropriations have already tripled since 2002. And there are political limits to further increases. Nagy et al., supra note 236, at 665. Furthermore, according to the theory of regulatory capture put forward by George Stigler, the SEC may not have the right incentives to do so. 238See generally George J. Stigler, The Theory of Economic Regulation, 2 Bell J. Econ. & Mgmt. Sci. 3 (1971) (discussing a theory for regulatory capture according to a supply and demand model for regulation). The SEC staff may suffer from selection or cognitive biases, or make mistakes in evaluating cases, which may interfere with efficient oversight. 239See generally Christine Jolls, Behavioral Law and Economics 11−18 (Nat’l Bureau of Econ. Research, Working Paper No. 12879, 2007), available at http://www.nber.org/papers/w12879 (discussing biases and judgment errors by rational actors). For example, the SEC may under-monitor smaller firms or foreign issuers, 240See Natalya Shnitser, Note, A Free Pass for Foreign Firms? An Assessment of SEC and Private Enforcement Against Foreign Issuers, 119 Yale L.J. 1638, 1645–46 (2010); see also Érica Gorga, Is the U.S. Law Enforcement Stronger? The Case of Securities Fraud by Brazilian Corporations and Lessons for the Private and Public Enforcement 34 (Feb. 5, 2014) (unpublished manuscript) (on file with authors). or choose to focus on cases with greater repercussions on Wall Street. The SEC may focus on certain types of claims, in which wrongdoing is easier to observe. Finally, the SEC may simply not have the authority to take action in cases in which there are regulatory gaps, when the supposed wrongdoing falls under the overlapping authority of state law or a foreign jurisdiction.

All of these administrative issues, biases, and hindsight may prevent the regulator from making the best or most efficient decisions from a market perspective. In situations where the regulator has failed to pursue enforcement actions, the system must rely on the work of private attorneys general to oversee market participants and investigate misconduct. As the Disney litigation and many other cases show, often a complaint by shareholders is first brought by private plaintiffs. 241See supra Part I.B.

As described in our discussion of the Disney litigation, the very process of discovery disciplines management, aside from the generation of informational public goods. Becoming the object of the kind of intense adversarial scrutiny that the Disney management endured is burdensome and disciplinary per se. Discovery forces managers to answer questions they do not want to answer; it challenges their power and authority in a public setting; it requires them to reveal their business secrets; and they face contempt and possible criminal charges if they engage in misrepresentations. The process can result in employment and reputational consequences. 242See Rose, supra note 197, at 2191. Discovery may trigger a duty to file a Form 8-K with the SEC or restate earnings (as was typical in the stock options backdating cases), because once obtained, whether routinely or through litigation discovery, material information must be disclosed. Such probes also raise the specter of criminal liability and other sanctions by regulators. 243See Paul J. Cohen et al., Investigations in the United States, in Corporate Internal Investigations: An International Guide, supra note 53, at 53, paras. 3.13–.20, at 59–62, paras. 3.134–.142, at 112–15. Finally, there are opportunity costs associated with the process for all executives affected by discovery.

The threat of discovery creates positive externalities, because it not only affects parties involved in the particular litigation, but signals to management’s peers at other corporations that they are likely to be subject to similar scrutiny 244Richard H. McAdams, An Attitudinal Theory of Expressive Law, 79 Or. L. Rev. 339, 340 (2000) (“[L]aw changes behavior by signaling the underlying attitudes of a community or society. . . . [T]he information signaled by legislation and other law affects [people’s] behavior.”). at some point during their tenure. Thus, they must exercise extreme care in the discharge of their disclosure and oversight responsibilities to make sure they are not personally damaged in this event.

Cases informed by extensive fact investigation through discovery that go to trial result in a decision on the merits. This will shape case law development and legal change, as we discuss in section C below. But what about cases that do not go to trial? These include most shareholder litigation. Critics contend these cases do not have the benefit of exposing information to the public because discovery is not revealed at trial.

In response, we describe how cases on summary judgment also publicize information, as do cases settled prior to summary judgment. Even cases settled prior to the motion to dismiss, as we will show, often generate substantial information on corporate internal practices and wrongdoing. An important point that has been overlooked is that defensive discovery produces at least as much information as adversarial discovery produces, and that information is shared with gatekeepers and regulators even prior to the motion to dismiss. The same is true for corporate internal investigations.

Below we examine what information is revealed at each stage in the litigation process.

2. Cases Resolved at Summary Judgment

Cases that go through discovery and are resolved upon the filing of a motion for summary judgment will typically produce a substantial portion of the information that the parties would have presented at trial. Motions for summary judgment are decided based on evidence submitted to the court by the parties. 245Fed. R. Civ. P. 56. Motion papers include statements of fact that identify and explain why, based on all the evidence presented to the court on the motion, there is no need to go to trial. 246Id. Admissible evidence is attached to the motion in the form of exhibits. 247Id. In shareholder actions, plaintiffs may present the court with thousands of pages of compromising documents, excerpts from deposition transcripts, responses to interrogatories and requests for admission, and other admissible evidence. This information is public, because it can be obtained from the court’s clerk. For more than a decade now, the federal trial courts have required electronic filing on PACER, making this information readily available online, and increasingly entire dockets are becoming available on Bloomberg and other legal information services. 248PACER stands for “Public Access to Court Electronic Records,” which is “an electronic public access service that allows users to obtain case and docket information from federal appellate, district and bankruptcy courts, and the PACER Case Locator via the Internet. PACER is provided by the federal Judiciary in keeping with its commitment to providing public access to court information via a centralized service.” PACER, http://www.pacer.gov (last visited May 28, 2014). Attorneys who practice in federal court, legal academics, law students, librarians, and journalists working for a major news organization have full access to documents filed with the federal courts in every case across the country. Proprietary databases, such as Bloomberg Law, will also include content provided by PACER.

The 2012 settlement of the Bank of America (BofA) securities class action litigation concerning BofA’s 2008 acquisition of Merrill Lynch illustrates how extensive the information is at this stage. 249See In re Bank of Am. Corp., Sec., Derivative, & ERISA Litig., 757 F. Supp. 2d 260, 274−84 (S.D.N.Y. 2010). This case consolidated securities fraud, derivative, and ERISA claims against Bank of America for misstatements and omissions during the acquisition of Merrill Lynch following the 2008 financial meltdown. The full docket, including all pleadings, motions, court decisions, and exhibits filed with the court, are readily available, for a per-page fee, on PACER. Plaintiffs claimed, inter alia, that BofA had failed to disclose more than $15 billion in late-2008 Merrill Lynch losses, as well as billions of dollars set aside to pay bonuses to Merrill Lynch executives, in the proxy statements to the merger agreement. 250Id. at 278−79. Shortly before the parties reached a settlement on June 20, 2012, plaintiffs and defendants each moved for summary judgment. 251Motion for Summary Judgment, In re Bank of Am. Corp., 757 F. Supp. 2d 260 (No. 09-MD-2058-PKC); Order, In re Bank of Am. Corp., 757 F. Supp. 2d 260 (No. 09-MD-2058-PKC), ECF No. 620 (giving notice that the court had been advised of an executed settlement agreement on June 20, 2012). In the month of June alone, the parties attached at least 600 exhibits—including thousands of pages of deposition transcripts, documents, e-mails, confidential memos, and other information, to support their motions. 252See Docket Items 577−659, Civil Docket, In re Bank of Am. Corp., 757 F. Supp. 2d 260 (No. 09 MD 2058 (PKC)). Again, all of these exhibits, and thousands of others, are readily available online to anyone with a PACER account.

The exhibits paint a picture of top executives at BofA pointing fingers at one another and their attorneys as to who was responsible for negotiating, overseeing, and monitoring key terms of the merger and for disclosing relevant material information in their SEC filings. 253See, e.g., Continued Videotaped Deposition of Neil Andrew Cotty, supra note 69, at 14 (claiming he did not understand that Merrill Lynch was selling substantial assets to cut their balance sheet days before the deal was approved by the shareholders by stating, “My only comment earlier . . . that it would have come from either Joe or [BofA CEO] Ken [Lewis], but I don’t recall a discussion on the balance sheet at December 3rd”). Collectively, the exhibits allow the same kind of step-by-step reconstruction of the entire negotiation and decision-making process of BofA’s acquisition of Merrill Lynch, including the part that individual managers and their inside and outside advisors played in the events as they unfolded. In an article reviewing the plaintiffs’ filings, including deposition testimony by top executives at BofA, the New York Times wrote the following:

What Bank of America’s top executives, including its chief executive then, , knew about Merrill’s vast mortgage losses and when they knew it emerged in court documents filed Sunday evening in a shareholder lawsuit being heard in Federal District Court in Manhattan.

. . . .

The filing in the shareholder suit included sworn testimony from Mr. Lewis in which he concedes that before Bank of America stockholders voted to approve the deal he had received loss estimates relating to the Merrill deal that were far greater than reflected in the figures that had appeared in the proxy documents filed with regulators. 254Gretchen Morgenson, Merrill Losses Were Withheld Before Merger, N.Y. Times, June 4, 2012, at A1.

The settlement was negotiated even as the compromising discovery materials were being filed with the court. The $2.43 billion settlement, in which BofA admitted to no wrongdoing, was finalized within weeks of the summary judgment filings. 255Jessica Silver-Greenberg & Susanne Craig, Bank of America Settles Suit Over Merrill for $2.43 Billion, DealBook (Sept. 28, 2012, 9:19 PM), http://dealbook.nytimes.com/2012/09/28/bank-of-america-to-pay-2-43-billion-to-settle-class-action-over-merrill-deal/?_php=true&_type=blogs&_r=0.

The Bank of America case illustrates how settlement prior to trial does not necessarily affect the publication of information revealed by discovery. Top management did not escape scrutiny in extensive depositions. Encroaching on BofA’s attorney–client privilege, the court even permitted the deposition of BofA’s longtime outside counsel at Wachtell Lipton, one of the most powerful corporate law firms in the United States. 256While depositions of attorneys are disfavored, defendants who claim to have relied on advice of counsel may waive their attorney–client privileges in this regard, thus opening the door to testimony by their counsel. See, e.g., Arthur Andersen LLP v. United States, 544 U.S. 696, 699 (2005); Rhone-Poulenc Rorer Inc. v. Home Indem. Co., 32 F.3d 851, 863 (3d Cir. 1994); Joy v. North, 692 F.2d 880, 893 (2d Cir. 1982); In re Bank of Am. Corp., 757 F. Supp. 2d at 279; In re ML-Lee Acquisition Fund II, L.P., 859 F. Supp. 765, 766 (D. Del. 1994).

This case also illustrates how the very information produced by discovery influences the parties’ choices about going to trial. Settlement rates are likely to increase where parties have full access to the facts and information before trial, as intended by the rule-makers. 257In an economic model of litigation and settlement decisions under imperfect information, Bebchuk argues that discovery requirements “increase the probability of settlement.” Bebchuk, supra note 101, at 413. In a model with one-sided discovery, Sobel shows that mandatory discovery rules reduce the probability of trials. Joel Sobel, An Analysis of Discovery Rules, Law & Contemp. Probs., Winter 1989, at 133, 133; see also Robert D. Cooter & Daniel L. Rubinfeld, An Economic Model of Legal Discovery, 23 J. Legal Stud. 435, 436 (1994) (“[D]iscovery increases settlements and decreases trials by organizing the voluntary exchange of information.”). In civil law jurisdictions, which do not allow broad discovery, settlement rates are lower. Huang reports an interesting development after the introduction of civil discovery in Taiwan in a 2000 legal reform. Kuo-Chang Huang, Does Discovery Promote Settlement? An Empirical Answer, 6 J. Empirical Legal Stud. 241 (2009). He shows that settlement rates for civil cases consistently increased over time in all district courts following the adoption of the discovery regime. Id. at 257−59. Alison Frankel, for example, wrote for Thompson Reuters that the “settlement reflects the nuanced understanding of Bank of America’s failure to disclose billions of dollars in escalating Merrill Lynch losses that shareholders’ counsel gained through dozens of depositions and millions of pages of discovery.” 258Alison Frankel, How BofA Was Forced to Settle $2.43 Bln Merrill Class Action, Reuters (Oct. 1, 2012), http://blogs.reuters.com/alison-frankel/2012/10/01/how-bofa-was-forced-to-settle-2-43-bln-merrill-class-action/.

Settlements before trial thus introduce a selection bias into the empirical debate about outcomes in shareholder litigation because the very cases that could lead to liability are likely to be settled; in other words, the same cases in which discovery has turned up evidence that would encourage the defendants to settle.

All class action and derivative settlements must be approved by the court. As the Bank of America case showed, judges will sometimes insist on additional discovery before approving a settlement. 259See, e.g., Peter J. Henning, Behind Rakoff’s Rejection of Citigroup Settlement, DealBook (Nov. 28, 2011, 5:14 PM), http://dealbook.nytimes.com/2011/11/28/behind-judge-rakoffs-rejection-of-s-e-c-citigroup-settlement/. The more damning the information revealed by discovery (or internal investigations, discussed below), the more likely a case is to settle before trial. Defendants have the incentive to settle because of the risk of an adverse litigation outcome, the publicity that a trial brings, and because the directors’ and officers’ liability insurance will cover settlement costs. The lack of liability, including personal liability, thus does not necessarily show that shareholder litigation fails to discipline corporate wrongdoing—as most of the literature concludes. Information about alleged wrongdoing is revealed and will be assessed by market participants, with the potential for generating serious reputation damages. And settlement values will also reflect the seriousness of the alleged wrongdoing.

We show in the next subsection how cases resolved during discovery also reveal substantial information about mismanagement or management misconduct.

3. Cases Settled Prior to Summary Judgment

In cases settled prior to summary judgment, discovery materials are not usually filed with the court unless required to support a motion. 260Fed. R. Civ. P. 5(d)(1) (“[D]isclosures under Rule 26(a)(1) or (2) and the following discovery requests and responses must not be filed until they are used in the proceeding or the court orders filing . . . .”). But discovery materials are nonetheless available under various circumstances. In federal securities class actions, the parties may file discovery materials with the court in connection with the motion for class certification. 261Until the Supreme Court held that plaintiffs were not required to prove materiality (an essential element in securities fraud claims) on a motion for class certification, plaintiffs conducted discovery and filed relevant admissible evidence with the court on this issue. See Amgen Inc. v. Conn. Ret. Plans & Trust Funds, 133 S. Ct. 1184, 1191, 1202 (2013). In shareholder derivative actions, plaintiffs may support their complaint with evidence obtained by means of a shareholder information request. 262See Del. Code Ann. tit. 8, § 220 (West Supp. 2014). And judges must approve settlements in shareholder actions, which requires a hearing and a “finding that it is fair, reasonable, and adequate.” 263Fed. R. Civ. P. 23(e)(2). The Advisory Committee Notes to the 2003 Amendments state that “[t]he court must make findings that support the conclusion that the settlement is fair, reasonable, and adequate. The findings must be set out in sufficient detail to explain to class members and the appellate court the factors that bear on applying the standard.” Fed. R. Civ. P. 23 advisory committee’s note (2003 Amendments); see also Fed. R. Civ. P. 23.1 (governing derivative actions); Donald F. Parsons, Jr. & Jason S. Tyler, Docket Dividends: Growth in Shareholder Litigation Leads to Refinements in Chancery Procedures, 70 Wash. & Lee L. Rev. 473, 497 (2013) (citing Barkan v. Amsted Indus., Inc., 567 A.2d 1279, 1283 (Del. 1989)) (“[T]he Court of Chancery vigilantly examines the merits of every settlement of a representative action, whether objected to or not, to determine whether, in the exercise of the reviewing court’s independent business judgment, the settlement is in the shareholders’ interest.”). In so doing, courts may request the full settlement agreement from the parties and review what discovery revealed. This occurred in the Caremark case, Delaware’s most important oversight case. 264In re Caremark Int’l Inc. Derivative Litig., 698 A.2d 959, 961 (Del. Ch. 1996). The court may also require additional discovery before approving a settlement.

Chancellor Allen’s opinion in Caremark was issued on a motion to approve a settlement in a consolidated derivative action that was negotiated prior to summary judgment. 265Id. at 960. Nonetheless, the court engaged in a lengthy examination of the factual record. 266See id. at 961–66. In Caremark, plaintiffs alleged that Caremark’s directors had breached their fiduciary duty of care, for failure to oversee employees who violated federal and state laws governing health care providers, resulting in a criminal indictment of the company and a $250 million fine. 267Id. at 960–61. In reviewing the standard for approving settlements in shareholder derivative actions, Chancellor Allen held that

[a] motion of this type requires the court to assess the strengths and weaknesses of the claims asserted in light of the discovery record and to evaluate the fairness and adequacy of the consideration offered to the corporation in exchange for the release of all claims made or arising from the facts alleged. . . . In this effort the court does not determine contested facts, but evaluates the claims and defenses on the discovery record to achieve a sense of the relative strengths of the parties’ positions. 268Id. at 961 (emphasis added).

Allen proceeded to review the facts that emerged in discovery, concluding that

in light of the discovery record, . . . there is a very low probability that it would be determined that the directors of Caremark breached any duty to appropriately monitor and supervise the enterprise. Indeed the record tends to show an active consideration by Caremark management and its Board of the Caremark structures and programs that ultimately led to the company’s indictment and to the large financial losses incurred in the settlement of those claims. It does not tend to show knowing or intentional violation of law. 269Id.

Discovery revealed detailed information about the day-to-day operations of the firm, Caremark’s management structure, fees paid to physicians for services to Medicare and Medicaid, efforts made to ensure compliance with company policies and contracts, legal advice by inside and outside counsel, uncertainty about relevant legal interpretations, the scope of ongoing government investigations, an internal audit plan, policies governing management supervision, and employee training to ensure legal compliance and adherence to company ethics rules. 270Id. at 962–63. Adversarial discovery and factual development were key to Chancellor Allen’s decision that “[t]he Board appears to have been informed about [these] and other efforts to assure compliance with the law.” 271Id. at 963.

The Caremark settlement included several corporate governance improvements, including (a) changes in compensation structure, commissions, and fees paid variously to employees, agents, physicians and health care providers; (b) mandatory review of material changes in government health care regulations and their implications for Caremark on a semi-annual basis by the full Board; (c) written disclosure of the financial relationship between Caremark and health care providers to patients; (d) the establishment of a four-member board committee on compliance and ethics (with at least two independent directors) to meet at least four times a year to carry out these policies, monitor compliance, and report back to the full board; and (e) direct reporting by corporate officers to the committee, which was charged with reviewing and approving contracts and new contract forms with the assistance of outside counsel. 272Id. at 966.

In sum, although the case settled before summary judgment, Caremark did publicize facts obtained through discovery. 273Id. at 961–66. Chancellor Allen acknowledged that “none of the documents submitted for review, nor any of the deposition transcripts appear to provide evidence of [the possibility that the Caremark directors knew of violations of law].” Id. at 971. More generally, the Caremark decision represented an important development in corporate case law. While the pronouncements of the court were technically mere dicta, the court identified a new legal standard that only “sustained or systematic failure . . . to exercise oversight—such as an utter failure to attempt to assure a reasonable information and reporting system exits [sic]—will establish the lack of good faith that is a necessary condition to liability.” 274Id. at 971. This “Caremark standard” was subsequently applied in board oversight cases. 275See Thompson & Thomas, supra note 46, at 1749 (pointing out the importance of Caremark beyond the outcome of the case).

One standard criticism of settlements in shareholder derivative and class actions is that they do not accurately reflect the merits of the case. 276Alexander, supra note 38, at 499; Roberta Romano, The Shareholder Suit: Litigation Without Foundation?, 7 J.L. Econ. & Org. 55, 55–56 (1991); see, e.g., Kevin M. LaCroix, More Options Backdating Settlements and Other Web Notes, D&O Diary (Sept. 8, 2007, 3:44 PM), http://dandodiary.blogspot.com/2007/09/more-options-backdating-settlements-and.html (“[F]or the cases that are settling, companies are agreeing to adopt some mild corporate therapeutics, and paying some negotiated amount supposedly corresponding to the amount of the plaintiffs’ attorneys’ fees. The sole benefit to the shareholders on whose behalf the plaintiffs . . . ostensibly proceeded is the ostensible benefit of the corporate therapeutics. I am sure there are skilled advocates who have persuaded themselves, at least, that this process represents something more than a highly stylized form of larceny.”). A judge may approve a settlement that awards attorneys’ fees to plaintiff’s counsel only if the parties can show a benefit to the corporation. 277See Mark J. Loewenstein, Shareholder Derivative Litigation and Corporate Governance, 24 Del. J. Corp. L. 1, 2 (1999) (discussing the “substantial benefit” rule). This benefit need not be pecuniary, but may consist in corporate governance improvements. 278See id. at 1–3.

But the critics maintain that such corporate governance improvements are typically devised by the defendant corporation’s counsel after a settlement has been reached and are designed to minimize structural changes to the defendant corporation’s management routines. 279See LaCroix, supra note 276. To the extent that such changes are meaningful, they would have been undertaken anyway, but instead they are attributed to the litigation so that the plaintiffs’ attorneys can get their fees. 280See, e.g., Daniel Fisher, Lawyer Fights $10 Million Fee for “Cosmetic” Johnson and Johnson Settlement, Forbes (Sept. 4, 2012, 6:40 PM), http://www.forbes.com/sites/danielfisher/2012/09/04/lawyer-fights-10-million-fee-for-cosmetic-johnson-johnson-settlement/. Such “corporate therapeutics” are thus largely seen as window dressing, which adds no value, but merely serves to ensure judicial acceptance of the settlement. 281See sources cited supra note 276. On this view, lawyers maximize their returns and the whole settlement effort constitutes nothing more than a rent-seeking device. Settlement negotiations are not affected by discovery, and the judge merely rubber-stamps the settlement. According to this rationale, parties and the litigation itself do not benefit from discovery, because litigation results have been somehow fixed ex ante.

There are many examples one could cite to make this point. A general observer might find the $2.75 million fee award that Barnes & Noble paid plaintiff’s counsel (in connection with the settlement of a shareholder derivative action for stock options backdating) hard to justify as a fair price for relatively minor changes in internal controls (considering that, in addition, the company spent millions of dollars on its own litigations costs). 282See Barnes & Noble, Inc., Annual Report (Form 10-K), 20 (Apr. 2, 2008). However, Barnes & Noble’s 2006 annual report shows that more was accomplished. It includes extensive findings of management, oversight, and accounting failures and reports the repayment of $2 million to the company and further repricing, saving another $2.64 million. 283Barnes & Noble, Inc., Annual Report (Form 10-K), exhibit 13.1, at F-6 to F-7 (Apr. 4, 2007). If one reviews Barnes & Noble’s own press release of the findings of the special litigation committee, it is clear that more was accomplished than the adoption of superficial “corporate therapeutics” and that the SLC’s report was hardly a whitewash. 284See Press Release, Barnes & Noble, Inc., Barnes & Noble, Inc. Announces Findings of Special Committee Review of the Company’s Stock Options Practices and Adoption of Remedial Recommendations (Apr. 4, 2007), available at http://www.barnesandnobleinc.com/press_releases/2007_april_4_stock_options_practices.html (detailing wrongdoing that internal corporate investigation uncovered with regard to stock options backdating). Note that these findings constitute only a summary of the SLC’s full report, which was never published or filed with the court.

Stock option awards can end up costing a company like Barnes & Noble tens or even hundreds of millions of dollars. And an empirical study shows significant reputational penalties for compensation committee members of firms involved in backdating. 285See Yonca Ertimur, Fabrizio Ferri & David A. Maber, Reputation Penalties for Poor Monitoring of Executive Pay: Evidence from Option Backdating, 104 J. Fin. Econ. 118, 120 (2012). The negative view of “corporate therapeutics” thus exaggerates and oversimplifies. It neglects that important information concerning the diagnosis of the misconduct may have been revealed by the discovery process. 286Cf., e.g., Affidavit of Jeffrey N. Gordon in Support of Plaintiffs’ Motion for Preliminary Approval of Derivative Action Settlement at 1, In re Pfizer Inc. S’holder Derivative Litig., 722 F. Supp. 2d 453 (S.D.N.Y. 2010) (No. 09-CV-7822), 2010 WL 5068906 (“[T]he Reforms embodied in the Proposed Settlement will significantly strengthen Board oversight of Pfizer’s compliance with the FDA’s drug marketing regime and related compliance mandates and will produce other improvements to internal compliance and accountability.”).

Assessing the quality of structural settlements is an empirical issue we do not pursue here, but we note that these agreements may include important changes in executive compensation, changes in board composition, and restrictions on self-dealing transactions. 287Romano, supra note 276, at 63 (commenting that some corporate governance changes resulting from settlements did reflect important reforms, such as changing the board’s majority). Moreover, the nature and content of the changes themselves respond to information obtained during discovery. While attorneys’ fees can explain the efforts to include corporate governance changes in settlements, they do not explain the nature of these changes. Settlement of representative litigation is conditioned on confirmatory discovery. 288See Parsons & Tyler, supra note 263, at 504 (“[I]f [the plaintiff] discovered evidence before the final settlement hearing that suggested its claims were stronger than [it] had realized, it could rescind the settlement and litigate post-closing damages.”); see also In re Celera Corp. S’holder Litig., No. 6304-VCP, 2012 WL 1020471, at *8 (Del. Ch. Mar. 23, 2012) (detailing the various steps the court must take in the approval process), aff’d in part, rev’d in part, 59 A.3d 418 (Del. 2012). In Caremark, for example, the board agreed to some very concrete changes that addressed the failure of the board to become aware of the compliance problems within the company. 289In re Caremark Int’l Inc. Derivative Litig., 698 A.2d 959, 966 (Del. Ch. 1996). Moreover, Chancellor Allen also reduced the fees to be awarded to the plaintiffs’ attorneys. 290Id. at 972. In a recent case, the Delaware Court of Chancery examined the therapeutic benefit achieved in light of the numerous cases evaluating supplemental disclosures and reduced the attorneys’ fees from the $500,000 requested to $300,000. See Order, In re Inspire Pharm. Inc. S’holders Litig., No. 6378-VCP, (Del. Ch. Jan. 30, 2012), 2012 WL 275115; Parsons & Tyler, supra note , at 496. All this leads us to conclude that judicial approval of settlements is not as meaningless as is often suggested.

Recent studies refute the view that, in settling class actions, the merits do not matter. 291James D. Cox, Randall S. Thomas & Lynn Bai, There Are Plaintiffs and . . . There Are Plaintiffs: An Empirical Analysis of Securities Class Action Settlements, 61 Vand. L. Rev. 355, 384–85 (2008). But even assuming that the information revealed by discovery is deficient in certain cases, or lawyers fail to propose useful corporate governance improvements, the general thesis that discovery promotes corporate governance can still be sustained in a Kaldor–Hicks framework. Discovery may generate significant positive externalities in a few cases, or small positive externalities in a large number of cases, which may compensate for these flaws. In other words, instead of looking only at the aggregate costs of discovery, even imperfect outcomes may produce positive effects that result in compound aggregate benefits to corporate governance, whether at the firm level or systemwide.

To be sure, we do not deny that parties are driven by economic incentives and self-interest in prosecuting a civil law suit. Rent-seeking will certainly drive outcomes in some cases. 292Exemplary of this are the cases of William Lerach and Melvin Weiss, top plaintiffs lawyers involved in egregious securities class actions scandals. See Jonathan D. Glater, Class-Action Lawyer Given a 30-Month Prison Term for Hiding Kickbacks, N.Y. Times, June 3, 2008, at C3; Michael Parrish, Leading Class-Action Lawyer Is Sentenced to Two Years in Kickback Scheme, N.Y. Times, Feb. 12, 2008, at C3. We do not attempt to provide a comprehensive solution to this problem here. One normative prescription that follows, however, is that judges should take the social utility of discovery into account when approving settlements and awarding attorneys’ fees. Judges should monitor settlements to maximize the value of the information revealed by discovery. 293See generally Hillary A. Sale, Judges Who Settle, 89 Wash. U. L. Rev. 377 (2011) (analyzing the role of judges as gatekeepers and monitors of settlements).

4. Cases That Go Through “Some” Discovery

Some cases do not go through the entire process of discovery. But even if discovery is not completed, its disciplining function is much more extensive than often recognized. The disciplining function of discovery is typically associated with the ability of a plaintiff to obtain information from a defendant corporation, that is, with offensive discovery. 294See Bebchuk, supra note 101, at 413; cf. Cooter & Rubinfeld, supra note 257 (providing an economic analysis of discovery). But the disciplining function of defensive discovery is rarely, if ever, considered.

Party-on-party discovery involves both offensive and defensive discovery. Every party to a lawsuit may serve and must respond to discovery requests during the discovery period approved by the court. The task of serving discovery requests is called “offensive discovery.” It entails obtaining information from the other side by, inter alia, serving document requests and taking depositions. In a typical shareholder derivative action, the plaintiffs are the ones who will conduct most of the offensive discovery, because their interest is to investigate the company. 295Corporate defendants will conduct some offensive discovery on jurisdictional requirements and issues like standing to sue under Fed. R. Civ. P. 23.1. Defendants in securities class actions will seek discovery in connection with a plaintiff’s motion for class certification and to assess damages.

The task of responding to discovery requests is called “defensive discovery.” This involves providing information to the other side by producing documents or witnesses. Defensive discovery is a task that all parties to the litigation must complete. It typically begins with the collection of information from the client even before discovery begins. There is a disciplining effect to having litigation counsel perform the internal investigation necessary to prepare the case and conduct defensive discovery.

Consider concretely what occurs when a large public corporation defends itself and its executives against a shareholder action. A good faith discovery effort (and standard practice) requires that litigation counsel to a corporate defendant must collect, restore, sort, and review a much larger quantity of electronic information and documents than are eventually disclosed. 296See, e.g., Kathleen B. Havener, Defensive Strategies in Discovery: A Refresher, A.B.A. (Aug. 30, 2012), http://apps.americanbar.org/litigation/committees/pretrial/email/summer2012/summer2012-0812-defen sive-strategies-discovery-refresher.html (“Ask your client to send you copies of everything remotely related to your claims or defenses and to your opponent’s document requests. After you have reviewed those documents, select those that are responsive, note documents that are not clearly responsive but that may have some significance to your claims or defenses, and remove and log any privileged documents. Then follow up with your client to confirm that nothing has been missed.”). This includes highly confidential documents subject to attorney–client privilege, many marginally relevant documents, as well as documents that will eventually not be produced. 297See id. This process often begins before discovery and means that no one in the corporate hierarchy—including the company’s executives and directors—can expect that their communications or documents will be sequestered from the litigation discovery machinery and kept private.

Public companies typically hire outside litigation counsel for shareholder litigation—especially in cases alleging management misconduct or securities fraud. The technology and complexity of ESI discovery make it very hard, if not impossible, for the executive suite to manage or limit what information is turned over to outside counsel once an investigation gets underway. 298See Cohen et al., supra note 243, para. 3.132, at 112 (describing extensive e-discovery). Review by outside counsel will necessarily occur before any discovery is disclosed to parties in litigation or the SEC. And it will need to be done, even if discovery is ultimately avoided altogether because the case is settled or the regulator decides not to proceed with an investigation. Corporate internal practices and operations will therefore be subject to intense, granular review by sophisticated outside professionals who have independent professional and reputational concerns. 299See generally id.

As already noted, these professionals include gatekeepers—such as outside litigation counsel, accountants, auditors, and financial institutions with independent legal or fiduciary obligations—as well as other outside professionals, such as financial experts, consultants, and expert witnesses. All these professionals are thus exposed to detailed information and learn from their experience, improving their expertise and organizational knowledge, which they can later apply in their practice for other clients. 300See generally Gorga & Halberstam, supra note 56 (discussing the process of generating organizational knowledge and individual knowledge). Gatekeepers, in particular, play an important role in counseling management. 301See John C. Coffee, Jr., Understanding Enron: “It’s About the Gatekeepers, Stupid, 57 Bus. Law. 1403, 1405 (2002) (“[G]atekeepers are reputational intermediaries who provide verification and certification services to investors.”). Their exposure to this information therefore has consequences for management not only in present, but also future, litigation. Here again, we note that the discovery regime—and how it works in practice—generates positive informational externalities for corporate governance.

Outside counsel represents the company, not the individuals within the company. 302See Upjohn Co. v. United States, 449 U.S. 383, 389–90 (1981). Covering up management misconduct violates an attorney’s duty to his or her client 303See, e.g., Cohen v. Telsey, No. 09-2033 (DRD), 2009 WL 3747059, at *17 (D.N.J. Nov. 2, 2009). —and can lead to a wide range of sanctions, including ethics charges brought against an entire firm, fines, civil malpractice claims, and criminal charges. The Sarbanes-Oxley legislation (Sarbox) provides for sanctions on gatekeepers who participate in a cover-up. Sarbox requires attorneys and accountants to report “evidence of a material violation” of the securities laws up the “ladder” to management, and if necessary to the board of directors, when counsel becomes aware of such evidence—even before, and regardless as to whether, the company is required to disclose such information during litigation discovery. 304Sarbanes-Oxley Act of 2002 § 307, 15 U.S.C. § 7245 (2012); 17 C.F.R. § 205.3(b) (2013). The regulators are, of course, well aware of the practices of defensive discovery and corporate internal investigations. Just as New York Mayor Edward Koch’s parking signage warned drivers “Don’t Even Think of Parking Here,” 305Sewell Chan, At the Sign-Making Shop, No Standing, No Idling, N.Y. Times, Aug. 14, 2005, at 30. Sarbox signals to outside counsel and audit firms how seriously regulators will take it if a law firm conspires with management to conceal evidence of corporate internal wrongdoing from the board or from the regulator. 306See, e.g., Stephen M. Cutler, Dir., Div. of Enforcement, SEC, The Themes of Sarbanes-Oxley as Reflected in the Commission’s Enforcement Program, Address at the UCLA School of Law (Sept. 20, 2004), text available at http://www.sec.gov/news/speech/spch092004smc.htm; see also Paul B. Murphy & Lucian E. Dervan, Watching Your Step: Avoiding the Pitfalls and Perils of Corporate Internal Investigations, ALAS Loss Prevention J., Summer 2005, at 2, 6. Independently of Sarbox, the institutions, techniques, professional rules, and relationships that have emerged in connection with the practice of modern litigation discovery have made it very hard to hide corporate internal wrongdoing. 307For a historical account of the developments of internal investigations, see Arthur F. Mathews, Internal Corporate Investigations, 45 Ohio St. L.J. 655 (1984); and Jim Hubbell, Note, Discovery of Internal Corporate Investigations, 32 Stan. L. Rev. 1163 (1980).

Even if none of the documents, e-mail, testimony, or other information obtained through discovery are published or turned over to regulators or adversaries in litigation, any serious response to allegations of wrongdoing, whether by regulators, plaintiffs’ attorneys, or, increasingly, originating internally with a company’s employees, means that a company will at the very least be subject to such internal review. 308See, e.g., Robert Trigaux, Backdating Inquiry Ends, SEC Tells Jabil Circuit, Tampa Bay Times (Nov. 26, 2008, 9:27 AM), http://www.tampabay.com/blogs/venturebiz/content/backdating-inquiry-ends-sec-tells-jabil-circuit/2093484. Jabil’s Form 10-K for the fiscal year ending August 31, 2008, reads as follows:In response to shareholder derivative actions that also were filed in connection with these certain grants, an independent Special Committee of our Board of Directors (the “Special Committee”) was appointed to review the allegations in such actions. . . . The Special Committee concluded that the evidence does not support a finding of intentional manipulation of stock option grant pricing by any member of management. In addition, the Special Committee concluded that it was not in our best interests to pursue the derivative actions. The Special Committee identified certain factors related to our controls surrounding the process of accounting for option grants that contributed to the accounting errors that led to the restatement of our consolidated operations for certain of our previous fiscal years (as further described in the Explanatory Note immediately preceding Part I of our Annual Report on Form 10-K for the fiscal year ended August 31, 2006 and discussed below). Pursuant to the state court’s approval on April 7, 2008, and the federal court’s approval on April 25, 2008, of our proposed settlement of the derivative actions, these actions are no longer pending.Jabil Circuit, Inc., Annual Report (Form 10-K), 14–15 (Oct. 29, 2008), available at http://www.sec.gov/Archives/edgar/data/898293/000119312508219096/d10k.htm. Such cases include derivative actions settled by special litigation committees, which we cover in the next subsection. Defensive discovery necessarily anticipates and responds to offensive discovery, and is thus shaped by discovery’s evolving rules and practices. When companies face adversarial discovery under the federal rules, they must necessarily investigate themselves, not merely comply with discovery requests. In anticipation of discovery, and even prior to any motion to dismiss, a defendant company must thoroughly investigate the factual basis of the charges. To proceed with a proper defense, and also to avoid future claims of securities disclosure violations, the litigation defense must develop a theory of the case and establish facts to support its own defenses in the litigation.

Defensive discovery practices contribute to compliance and the development of corporate governance standards, producing positive externalities, even if adversarial discovery is ultimately not fully pursued. The disciplinary function of defensive discovery helps us understand why corporate internal investigations play such an important role in U.S. contemporary corporate governance. We develop the role of internal investigations in the next subsection.

5. Cases That Are Dismissed or Settled Prior to the Motion to Dismiss Based on a Corporate Internal Investigation by a Special Litigation Committee

Since the mid-1970s, corporate internal investigations by special committees have become a powerful procedural mechanism that defendant corporations have at their disposal to respond to shareholder derivative actions, securities violations, and other charges, including charges of criminal wrongdoing. 309See Davis, supra note 43, at 390–405. See generally Corporate Internal Investigations: An International Guide, supra note 53 (giving a comprehensive account of corporate internal investigations in major jurisdictions, including the United States).

SLCs originated in the 1970s in connection with the questionable political payment cases. 310See Davis, supra note 43, at 390. In response to revelations by the Watergate special prosecutor that many companies had made undisclosed illegal campaign contributions and funneled money to foreign politicians to secure contracts, the SEC brought enforcement actions against several companies and their executives. 311Id. at 393. Because so many companies were implicated in these probes—ultimately almost 400 companies 312Id. at 395. —the SEC could not investigate every instance. 313Id. at 395–96. (“Unable to pursue each instance of questionable payments on its own, the SEC had leveraged its resources by pressuring corporations to investigate themselves.”). The SEC thus applied a new form of ancillary relief in these cases. The companies were required to “create a special review committee composed exclusively of independent directors. Typically using independent counsel and accountants, the committee was to investigate the irregularities alleged in the SEC’s complaint and submit a public report to the board of directors, which was responsible for reviewing and implementing the committee’s recommendations.” 314Id. at 395.Gall v. Exxon Corp. 315418 F. Supp. 508 (S.D.N.Y. 1976). was the first case in which a federal court implemented the same remedy in a civil case. 316See Davis, supra note 43, at 395–98.

The SLC procedure in derivative litigation thus implemented the same type of equitable remedy that courts allowed the SEC to fashion in its civil enforcement actions and administrative negotiations with issuers. But this does not mean that modern litigation discovery was irrelevant to this type of enforcement, or that the SEC has been the sole or primary force behind the U.S. corporate governance culture of transparency.

First, it is not coincidental that self-investigations by special board committees became an enforcement tool in the 1970s, precisely at the time that the federal discovery regime had reached its apex. SLCs were a response to the liberal grant of discovery in shareholder derivative actions and served, at least in part, to limit the costs and burdens of litigation discovery for corporate defendants. 317See id. at 400. Second, the SEC’s use of self-investigation was a response to its limited resources as the regulator. Moreover, the “reasonableness” of corporate internal investigations has been measured by the standard of full disclosure that animates the federal rules and the tools of discovery that have been developed in private litigation. Corporate internal investigations thus have continued to evolve together with the tools of offensive and defensive discovery. With the advent of e-discovery, for example, the methods, software, and standards of ESI collection have become standard in corporate internal investigations. 318See Brad Mixner, Reducing Discovery Challenges Through Innovative Use of Technology in Cross-Border Litigation, Inside Couns. (March 6, 2014), http://www.insidecounsel.com/2014/03/06/reducing-discovery-challenges-through-innovative-u (“Since its initial beginnings, e-discovery has been transformed drastically, now having an exciting array of technology options for managing data which is exponentially different in variety, scale, source and complexity. This is further complicated by today’s trend toward the globalization of many industries, the proliferation of multi-national companies and the complexities of cross border litigation.”). For countless new e-discovery platforms and technologies, see, for example, eDiscoveryTimes, http://ediscoverytimes.com (last visited May 29, 2014). Put differently, without private litigants devoting enormous resources to the process and demanding to collect ESI in accordance with the discovery rules, e-discovery would not have developed into the full-fledged assault on corporate internal secrecy that it has.

In response to a shareholder derivative action, a board of directors may appoint a special litigation committee composed of independent directors to investigate the allegations in the complaint. The investigation is performed prior to any decision on a motion to dismiss and prior to any discovery. 319Myers, supra note 46, at 1313. This includes a motion to dismiss for failure to make a demand. SLCs therefore allow defendants to avoid the risk of losing on the issue of demand futility. Upon application, a judge will grant a stay of discovery until the SLC has completed its internal investigation and produced a report (this usually takes eight or ten months). 320See Douglas M. Branson, The Rule That Isn’t a Rule—The Business Judgment Rule, 36 Val. U. L. Rev. 631, 648 (2002) (describing the formation and nature of the SLC). The SLC can then move to dismiss the complaint, if it concludes that the pursuit of litigation against any fiduciaries is not in the “best interests of the corporation.” 321See Richard C. Brown, Shareholder Derivative Litigation and the Special Litigation Committee, 43 U. Pitt. L. Rev. 601, 620–21 (1982). Shareholders may theoretically also sue third parties on behalf of the corporation, but we are not interested in such cases here. If the judge is satisfied that the SLC’s members are disinterested and independent, and that the SLC’s investigation was “reasonable” under the circumstances and carried out in “good faith,” 322Plaintiffs may seek “limited discovery” solely on whether the SLC’s members are independent and the investigation was reasonable. New York law requires that judges apply the business judgment rule to the SLC’s recommendation. 323See, e.g., Auerbach v. Bennett, 408 N.Y.S.2d 83 (App. Div. 1978), aff’d as modified, 393 N.E.2d 994 (N.Y. 1979). The Supreme Court accepted this type of approach in Burks v. Lasker, 441 U.S. 471 (1979), reversing Lasker v. Burks, 567 F.2d 1208 (2d Cir. 1978). Some states have codified the SLC procedure. See, e.g., Wis. Stat. § 180.0744 (2012); Einhorn v. Culea, 2000 WI 65, 235 Wis. 2d 646, 612 N.W.2d 78; see also Ann M. Scarlett, Confusion and Unpredictability in Shareholder Derivative Litigation: The Delaware Courts’ Response to Recent Corporate Scandals, 60 Fla. L. Rev. 589, 596–99 (2008). Delaware’s approach is somewhat less deferential, affording a judge discretion to weigh the evidence and substitute his or her business judgment for the business judgment of the SLC. 324See infra note 325. The dismissal has preclusive effect.

Doctrinally, two arguments are advanced to justify judicial deference to an SLC’s recommendation, even in cases where the SLC investigation has uncovered corporate internal wrongdoing: (1) the decision whether to pursue litigation in the name of the company is a matter that lies within the business judgment of the board, 325See Auerbach, 393 N.E.2d 994, 1000–03 (explaining that the plaintiff bears the burden of rebutting the business judgment rule presumption with respect to the SLC’s decision, and the judicial inquiry is limited to the disinterestedness and independence of the SLC). Delaware’s less deferential approach in Zapata Corp. v. Maldonado is as follows:First the Court should inquire into the independence and good faith of the committee and the bases supporting its conclusions. Limited discovery may be ordered to facilitate such inquiries. The corporation should have the burden of proving independence, good faith and a reasonable investigation, rather than presuming independence, good faith and reasonableness. 430 A.2d 779, 788 (Del. 1981) (footnote omitted). The Court may then, in its discretion, apply “its own independent business judgment, [in determining] whether the motion should be granted.” Id. at 789. In other words, step two permits a consideration of the substance of the report. and (2) shareholders should exhaust corporate internal remedies before a judge steps in. 326See Davis, supra note 43, at 398.

Special committees have, for the most part, been viewed negatively by academics and corporate governance proponents. 327See, e.g., Davis, supra note 46, at 1306 (citing literature). The prevailing view is that SLCs invariably recommend dismissal, 328Franklin A. Gevurtz, Corporation Law § 4.3.4, at 434 (2d ed. 2010) (“Special litigation committees usually have concluded that derivative suits which the committees looked into were not in the corporation’s best interest.”); see also 9 Louis Loss, Joel Seligman & Troy Paredes, Securities Regulation § 11(C)(4)(b) n.340, at 420 (4th ed. 2013); Jonathan R. Macey, Corporate Governance: Promises Kept, Promises Broken 65–66 (2008); George W. Dent, Jr., The Power of Directors to Terminate Shareholder Litigation: The Death of the Derivative Suit?, 75 Nw. U. L. Rev. 96, 109 & n.70 (1980) (surveying SLC dismissals). But see Myers, supra note 46, at 1311. which courts grant “in the vast majority of cases.” 329Scarlett, supra note 323, at 598–99 (quoting Lisa M. Fairfax, Spare the Rod, Spoil the Director? Revitalizing Directors’ Fiduciary Duty Through Legal Liability, 42 Hous. L. Rev. 393, 409 (2005)) (internal quotation marks omitted). The SLC mechanism has been thought of as yet another procedural hurdle to shield directors from exposure and liability for misconduct. 330See Myers, supra note 46, at 1314–16; see also Brown, supra note 321, at 620–21 (collecting reported cases in support of the conclusion that “[s]pecial litigation committees, after completing their investigations, have so far always determined that the derivative action was not in the best interests of the corporation and have voted to terminate the suit”); Davis, supra note 46, at 1357 (“[J]udicial deference to the dismissal recommendations of SLCs is the product of a broader recognition of the diminished role of the derivative suit within the portfolio of devices available to hold directors and officers to account.”); Dent, supra note 328, at 109 & n.70 (collecting reported cases on demand in support of the conclusion that in the SLC context “almost invariably, the directors charged with the decision decide to oppose the suit”).

A recent empirical study by Professor Minor Myers, however, shows that (1) SLCs decide to pursue or settle claims much more frequently than heretofore recognized, 331Myers, supra note 46, at 1320. and (2) most shareholder claims subject to SLC review end up settling and are not dismissed. 332Id. at 1327 tbl.6. Myers also observed that “SLC decisions to settle . . . were usually made longer after the filing of claims than either decisions to pursue [the litigation] or dismiss.” 333Id. at 1331. The time to settlement has also been lengthened in securities class actions after the PSLRA. Choi & Thompson, supra note 194, at 1498. He speculates that this might be attributable to the time needed to negotiate the settlement. 334Myers, supra note 46, at 1331. Once the SLC decided to settle, however, the cases were resolved quickly. 335Id. (“[O]nce the SLC decided to settle, the cases were resolved almost immediately.”).

Myers’s empirical findings are consistent with the view that SLCs spend significant time conducting corporate internal investigations before they return to the negotiating table and negotiate a settlement with plaintiffs’ counsel. This makes sense, because an SLC’s leverage in settlement discussions depends on actually conducting an adequate internal investigation into the specific allegations in the complaint. Only a thorough investigation can present a credible threat of dismissal and generate the information that helps the parties come to a better assessment of the value of the case. In other words, the SLC investigation produces information not otherwise available to the decision-makers and provides a basis for deciding whether to pursue or settle a claim.

SLC investigations push fact investigation, which usually takes place during discovery, into the prediscovery phase. In so doing, an SLC incorporates all of the disciplining practices of defensive discovery.

The investigation is conducted by independent board members who are not to take direction from the company or its management. The board members must hire an outside law firm, which, preferably, has had no prior dealings with the company. 336The SEC guidelines state the same criteria regarding cooperation for crediting the results of a corporate internal investigation as in the derivative context. Exchange Act Release No. 44,969, 76 SEC Docket 220 (Oct. 23, 2001) (asking whether an outside counsel or auditor had “done other work for the company” and whether “management previously engaged such counsel” in considering whether “the company commit[ted] to learn the truth, fully and expeditiously”). The reputation of the law firm itself is important, given that the decision to dismiss considers the quality of the internal investigation. The outside law firm’s client is the SLC. Communications between the SLC and the law firm are thus privileged. Anything that the SLC shares with the board is potentially discoverable by plaintiffs. 337See, e.g., Ryan v. Gifford, No. 2213-CC, 2007 WL 4259557, at *2–3 (Del. Ch. Nov. 30, 2007). Sharing information with the board before the conclusion of the investigation may compromise the independence of the investigation in the eyes of the court.

In assessing an SLC’s conclusion that maintaining the suit is not in the best interests of the company, a court will conduct two separate inquiries. The first focuses on the independence of the SLC’s members. 338This requires an inquiry into the relationship of the SLC members to the company and its management, as well as an inquiry into any relationship between the company and the professionals the SLC has hired to conduct the investigation. See Zapata Corp. v. Maldonado, 430 A.2d 779, 788–79 (Del. 1981). The second focuses on the thoroughness of the investigation. The SLC has the burden of proving that it engaged in a “reasonable investigation” that was “independent” and conducted in “good faith.” 339Id. at 788.

At this stage, plaintiffs may seek “‘limited discovery’ . . . to facilitate the inquiries of the trial court into the independence and good faith of the Committee and the reasonableness of its investigation and conclusions.” 340Kaplan v. Wyatt, 484 A.2d 501, 510 (Del. Ch. 1984); see also Kindt v. Lund, No. Civ.A. 17751, 2001 WL 1671438, at *1 (Del. Ch. Dec. 14, 2001). A court should not dismiss if either “the committee is not independent or has not shown reasonable bases for its conclusions, or, if the Court is not satisfied for other reasons relating to the process, including but not limited to the good faith of the committee.” 341Zapata Corp., 430 A.2d at 789.

A credible internal investigation must “explore all relevant facts and sources of information that bear on the central allegations in the complaint.” 342London v. Tyrrell, No. 3321-CC, 2010 WL 877528, at *17 (Del. Ch. Mar. 11, 2010) (“If the SLC fails to investigate facts or sources of information that cut at the heart of plaintiffs’ complaint this will usually give rise to a material question about the reasonableness and good faith of the SLC’s investigation.”). The scope of the investigation must encompass “all theories of recovery asserted in the plaintiffs’ complaint.” 343Id. (“To conduct a good faith investigation of reasonable scope, the SLC must investigate all theories of recovery asserted in the plaintiffs’ complaint.”). An SLC may be required to go beyond the allegations in the complaint. Courts will thus closely examine the SLC’s investigative procedures and methods. 344See generally Zapata Corp., 430 A.2d 779 (holding that the board’s delegation of its managerial authority over litigation to a special committee was effectual). The SLC must convince a judge, or the plaintiffs during settlement negotiations, that discovery is unlikely to produce important additional information about the nature or extent of corporate internal wrongdoing. In ruling on the motion, a judge will want to be assured that the plaintiffs are unlikely to come up with anything substantial that is not already in the SLC report, if the case went to adversarial discovery. 345See id. at 788–89.

In practice, this means that an SLC will use the same methods of investigation that the defendant corporation’s outside litigation counsel would have used, if the case went to discovery. The law firm conducting the investigation must engage in credible electronic discovery, including the restoration of backup tapes, and the collection and review of all relevant documents and e-mail communications of the parties involved in the alleged wrongdoing. 346See Cohen et al., supra note 243, para. 3.177, at 124 (“For the investigative counsel—and the company on whose behalf the investigation is undertaken—the ability to say, at the conclusion of the investigation, that all known relevant documents have been collected and reviewed is critical to the credibility of the investigation.”). Outside counsel is expected to interview all relevant witnesses, including senior management, 347Alice A. Seebach, Special Litigation Committees: A Practitioner’s Guide, 24 Loy. L.A. L. Rev. 1, 28–30 (1990) (“A broad investigation appears thorough. Courts accepting special committees’ recommendations to dismiss often comment on the number of witnesses interviewed. For example, one special committee that considered claims that a CEO misused a corporation interviewed 140 witnesses ‘throughout the world.’” (quoting Kaplan v. Wyatt , 484 A.2d 501, 511 (Del. Ch. 1984), aff’d 499 A.2d 1184 (Del. 1985))); see also Cohen et al., supra note 243, paras. 3.166–.211, at 121–35. and question the witnesses about potentially compromising documents as necessary. 348See, e.g., London, 2010 WL 877528, at *21 (finding SLC investigation not reasonable, because, inter alia, “Tyrrell’s own emails suggest that he believed these higher internal forecasts where [sic] achievable, in direct contradiction to the testimony he provided the SLC, but the SLC does not appear to have questioned him thoroughly about these emails”). Depending on the nature of the case, outside counsel may be required to hire an independent forensic accounting team to pursue an in-depth investigation of accounting procedures.

The SLC process thus mimics the procedures of defensive discovery. It pushes witness interrogation, document discovery, expert testimony, and the exchange of facts between the parties into the prediscovery period, the same way that the modern discovery rules pushed trial testimony and the exchange of facts at trial into the pretrial period. Instead of preparing a case for trial, however, an SLC produces a report detailing its findings. Reports are lengthy and attach witness accounts, documentary proof, and any expert findings on the basis of which a judge will render an opinion. 349See Seebach, supra note 347, at 25 & n.171. If the SLC report is filed with the court, it is served on the plaintiffs and becomes the basis for the defendant’s motion to dismiss.

The overly cynical view of SLCs that is characteristic of much of the literature thus overlooks the significant informational and disciplinary functions performed by SLC investigations. 350As Myers’s empirical study finds, the “claim about SLC behavior—that, even if they do not dismiss claims, they may nevertheless take it easy on defendants by failing to pursue claims diligently or by settling claims for nothing— . . . finds no support in the data.” Myers, supra note 46, at 1311.

It is often attractive for the parties to settle without the SLC actually having to file a motion to dismiss with the court. 351See Dent, supra note 236, 99 n.17 (“The purpose of requiring a demand on the board has been stated variously as . . . giving the board the opportunity to settle the dispute without litigation, thus promoting judicial economy . . . .” (citing Winter v. Farmers Educ. & Coop. Union, 107 N.W.2d 226, 233 (Minn. 1961))). Instead of filing a motion to dismiss, an SLC can share its final report with the plaintiffs subject to a confidentiality agreement. The advantage for plaintiffs is to obtain a settlement without additional investment in discovery, or the risk of an adverse ruling by the court. The advantage for the board is to avoid filing, and thereby publicizing, the SLC report, which typically will include at least some unpleasant information about internal practices. 352Apart from the reputational damage, facts discovered in the internal investigation can be used as a basis for additional litigation, including criminal indictments. See Cohen et al., supra note 243, para. 3.173, at 123.

What the SLC procedure accomplishes, even if the parties decide to settle prior to the motion to dismiss, and the SLC report remains confidential, is that the corporation thoroughly investigates itself. Outside counsel knows that, “[a]t a minimum, the [SLC] investigation should address the actual allegations of wrongdoing” as well as “other possible violations that are uncovered during the course of the investigation.353Id. para. 3.168, at 121–22 (emphasis added). What this means is that once the investigation is initiated, the SLC essentially takes on the responsibility of a regulator and conducts the fact investigation that a regulator would otherwise conduct. 354See Davis, supra note 43, at 392–95 (describing how “specific credit for developing the SLC itself goes to the Securities and Exchange Commission’s . . . Enforcement Division,” which originated the process to deal with a wave of disclosure violations in the 1970s by affording issuers more flexible relief and conserving its own resources). It is not enough for an SLC to narrowly consider whether the factual allegations in the complaint are accurate, or whether plaintiffs can support their claims. The SLC is charged with broadly investigating the allegations in the complaint, but then pursuing other “possible violations” that are uncovered incidentally during the course of the investigation, regardless as to whether such violations are related to the claims in the complaint. A failure to acknowledge “red flags” during an SLC investigation could result in securities disclosure violations by the company and ethics charges against the law firm. The decision to undertake such an investigation is therefore a consequential decision.

As for remediation, the settlement or dismissal of a case after an internal investigation does not necessarily mean that there are no consequences for officers and directors. 355Cohen et al., supra note 243, para. 3.171, at 122. A proper internal investigation should be broad enough to include consideration of appropriate remedial actions. 356As did the Barnes & Noble SLC. See Press Release, Barnes & Noble, supra note 284; see also Cohen et al., supra note 243, para. 3.169, at 122. While SLC investigations may not result in judicial sanction, 357See Myers, supra note 46, at 1332 (“[T]he SLC appears to function as a form of alternative dispute resolution.”). they often enough result in financial recoveries by the company, employment actions, changes in leadership, and specific corporate governance improvements. 358See id. If any material misstatements in the financial disclosures of the company are discovered that lead to earnings restatements—not an uncommon situation—then a Form 8-K must be filed with the SEC, and shareholders and market watchers will expect the company to explain what went wrong and what oversight mechanisms will be improved to avoid such problems in the future. 359See, e.g., Press Release, Barnes & Noble, supra note 284. In the Barnes & Noble stock options backdating case, and other backdating cases, executive stock options were repriced, 360Id. and often earnings had to be restated. This information becomes available to markets and regulators. 361Indeed, even if an SLC report is not filed with the court, remains confidential, and is only shared with plaintiffs’ counsel subject to a confidentiality agreement, it is often shared with the SEC in an effort to negotiate and assuage the regulator—subject to a “confidential treatment request.” See Ferrara & Khinda, supra note 236, at 1165–67.

Again, dismissal or settlement does not mean that internal wrongdoing is covered up. If unconvincing in scope, the findings of the internal investigation run the risk of being discounted, regardless as to whether the audience for the final report is a judge, the SEC, or plaintiffs’ counsel.

Derivative actions filed in state court are often filed parallel to securities class actions in federal courts. 362See Parsons & Tyler, supra note 263, at 517 (“[A]s has become common today, multiple plaintiffs file multiple complaints in multiple jurisdictions . . . .”); Thomas & Martin, supra note 214, at 72. Defendants seek a stay of the proceedings in federal court, while they proceed with their SLC investigation in a state derivative action. The cases are typically subject to a joint settlement, or are consolidated, as in the Bank of America case. 363See supra Part III.A.2. Cases dismissed because the plaintiffs failed to satisfy the heightened pleading requirements under the PSLRA, or the demand requirement in derivative actions, may raise concerns about underdeterrence, but we do not discuss these cases here. SLC investigations were, for example, common in the stock options backdating cases, in which a corporation might face inquiries by the SEC, one or more state court derivative actions, and one or more federal securities fraud actions. 364See, for example, Roth v. Reyes, No. C 06-02786 CRB, 2007 WL 2470122, (N.D. Cal. Aug. 27, 2007), for derivative litigation; and Smajlaj v. Brocade Communications Systems, Inc., No. C 05-02042 CRB, 2007 WL 2457534 (N.D. Cal. Aug. 27, 2007) for securities class action. In such cases, the final SLC report would sometimes be shared with the SEC and the plaintiffs at the same time to achieve a simultaneous resolution in all three forums.

Corporate internal investigations have increasingly become a standard tool of corporate governance. Corporations may launch internal investigations in response to any type of lawsuit or inquiry by the government, 365See Murphy & Dervan, supra note 306, at 2. but also in connection with problems that are brought to management’s attention internally by a company’s own employees or auditors. 366Id. As a result of the obligations imposed on public corporations by the Sarbanes-Oxley Act of 2002, such “internally-sparked investigations have increased dramatically in the post-Enron business climate, so much so that ‘internal investigation attorneys are becoming a dreaded necessity for a growing number of public companies.’” 367Id. (quoting Leigh Jones, Call for Internal Probes Growing, Nat’l L.J., Nov. 22, 2004, at 1). The literature has underestimated the increasing importance of corporate internal investigations for the development of U.S. and global corporate governance, 368Lomas & Kramer, supra note 53, at 1–2. in large part, because it has not appreciated the impact of defensive discovery.

B. Discovery Has Shaped Internal Corporate Governance

The rules and practices of litigation discovery have had an important influence on the development of new internal corporate governance procedures and compliance systems. The requirement to produce information during discovery forces corporations to maintain, store, and monitor their in-house communications and business documentation ex ante. Corporations must develop records management policies that make the timely search and collection of documents feasible. 369See, e.g., Litigation Preparedness, E-Discovery Basics (Gibson, Dunn & Crutcher LLP, L.A., Cal.), May 31, 2011, http://www.gibsondunn.com/publications/Documents/E-DiscoveryBasics3-LitigationPreparedness.pdf. They must develop contingency plans to suspend auto-delete functions and to prevent the recycling of backup tapes in response to litigation holds. 370Id. The failure to implement firm-wide information governance can be costly.

An entire industry of consultants and service providers has sprung up to assess and improve the litigation preparedness of large corporations. Consultants who advise on improving electronic data management for litigation preparedness are often the same ones who have developed sophisticated software for ESI discovery during litigation. 371See, e.g., D4 eDiscovery, http://www.d4discovery.com/services-solutions/ (last visited May 29, 2014) (offering comprehensive discovery management across platforms, including discovery best practices and solutions for social media); Managed Services, Xerox Litig. Servs., http://www.xerox-xls.com/ediscovery/managed-services.html?PHPSESSID=4804a414c2ed6c54cf38202f8619841e (last visited May 29, 2014) (offering “expertise in all phases of discovery” to law firms and corporations); see also Top Discovery Software Products, Capterra, http://www.capterra.com/electronic-discovery-software (last visited May 29, 2014). The routines and practices of ESI discovery are thus setting standards for information governance and transparency in everyday operations.

These responses to the demands of litigation discovery contribute to the effectiveness of internal monitoring. In the age of ESI, corporations must engage in “proactive” information governance to become litigation ready. 372See Peter Pepiton II, Becoming Litigation Ready Through Proactive Information Governance 1 (2008) This includes the ability to implement adequate litigation holds, the ability to produce requested documents within a reasonable timeframe, and the ability to do so efficiently and cost-effectively. 373See id. at 5. See Sheri Qualters, 30% of Companies Still Lack Policies for Preserving Evidence for Discovery, Nat’l L.J., Dec. 9, 2008 (“[H]aving no legal hold policy is a significant risk factor for companies.” (emphasis removed) (internal quotation mark removed)). Corporations may be subject to a range of sanctions if they fail to implement litigation holds. 374See, e.g., Sekisui Am. Corp. v. Hart, 945 F. Supp. 2d 494, 509 (S.D.N.Y. 2013) (sanctioning the plaintiff for failing to preserve electronic information on its servers).

Not only is information about organizational flows systematized in this way, but organizational knowledge may thereby also be assessed by professionals and improved. 375See Gorga & Halberstam, supra note 56, at 1127.

C. How Discovery Informs Legal Change

Corporate law scholars have never identified discovery as a key driver of legal change, even as scholars in other legal fields have recognized that substantive developments in other law areas have been driven by “broad-ranging discovery provisions.” 376Marcus, supra note 31, at 749 (quoting Jack H. Friedenthal, A Divided Supreme Court Adopts Discovery Amendments to the Federal Rules of Civil Procedure, 69 Calif. L. Rev. 806, 818 (1981)); see also Geraldine Szott Moohr, The Balance Among Corporate Criminal Liability, Private Civil Suits, and Regulatory Enforcement, 46 Am. Crim. L. Rev. 1459, 1472 (2009). Nonetheless, the discovery regime has influenced case law on fiduciary duties and the substantive development of corporate law in many areas. In this section, we look at just two examples.

1. The Development of Fiduciary Duty Doctrines

As already observed in our discussions of Disney and Caremark, critical developments in the judicial interpretation of fiduciary duties have depended on information that would not have become available to courts but for probing discovery. Moreover, the fiduciary duty standards that they have articulated assume and depend upon the availability of discovery for their meaningful enforcement. The seminal case of Van Gorkom can also be read this way. 377Smith v. Van Gorkom, 488 A.2d 858 (Del. 1985), overruled in part on other grounds by Gantler v. Stephens, 965. A.2d 695 (Del. 2009).

In Van Gorkom, shareholders sued the directors of Trans Union, Inc., for breach of their fiduciary duty of care in connection with the sale of the company to the Chicago investor Jay Pritzker. 378Id. at 863–64. The Court found that the defendants acted “reckless[ly]”—and awarded substantial damages—because of their failure to adequately inform themselves and ask the right questions, and for approving the sale too quickly. 379Id. at 871. The court held that the business judgment rule barred a review of the substance of such board decisions, but not of the adequacy of the decision-making process. 380Id. at 872.Van Gorkom thus resulted in what some have called the “proceduralization” of the boardroom. 381See, e.g., Charles M. Elson & Robert B. Thompson, Van Gorkom’s Legacy: The Limits of Judicially Enforced Constraints and the Promise of Proprietary Incentives, 96 Nw. U. L. Rev. 579, 584 (2002) (arguing that Van Gorkom’s legacy is the “rise of elaborate decision-making procedures involving lengthy meetings [and] voluminous documentation . . . that today accompany board decisions”); Lynn A. Stout, In Praise of Procedure: An Economic and Behavioral Defense of Smith v. Van Gorkom and the Business Judgment Rule, 96 Nw. U. L. Rev. 675, 676–77 (2002) (defending the business judgment rule’s emphasis on process after Van Gorkom).

The vast literature on Van Gorkom has never considered how critical discovery was for arriving at this result. 382See generally Bainbridge, supra note 229, at 393–405 (discussing the fate of shareholder derivative suits). As in the Disney and Caremark cases, the decision in Van Gorkom depended on the court’s ability to reconstruct the events leading up to the merger in rich detail. 383To be precise, the proceedings before the Delaware Chancery generated the extensive factual record that the Delaware Supreme Court ultimately relied on in issuing its opinion. See Van Gorkom, 488 A.2d at 864–70. The reconstruction included information about the understandings, opinions, and conclusions of all those involved in the deal as the deal evolved, and enabled the ability to identify who knew what, when they knew it, who communicated with whom, what they said, what information they did or did not obtain, and how assessments of the deal varied considerably as between the various actors involved. 384See id.

The record revealed that Van Gorkom did not consult with the board or his top executives during his negotiation with Pritzker, but only with Trans Union’s controller. 385Id. at 866. He proposed the price of $55 per share to Pritzker at a social event without seeking a bid or negotiating the price. 386Id. Depositions revealed that Van Gorkom chose this price based on an earlier determination by Trans Union’s CFO, Romans, that a management-sponsored leveraged buyout could be financed at between $50 to $60 per share, instead of an assessment of the value of the company. 387Id. at 865, 867 (“According to Romans: They did not ‘come up’ with a price for the Company. They merely ‘ran the numbers’ at $50 a share and at $60 a share with the ‘rough form’ of their cash figures at the time. Their ‘figures indicated that $50 would be very easy to do but $60 would be very difficult to do under those figures.’”). This evidence emerged during Romans’s depositions. But Van Gorkom never even saw the actual report prepared by Romans, nor did he ask Romans to discuss his valuation of the company (at up to $65 per share) or his concerns about the “lock-up” agreement (that interfered with a true market test) with the board. 388Id. at 867. And the board never heard about internal dissent to the terms of the sale among senior management that discovery revealed.

The evidence showed that “the Board accepted without scrutiny Van Gorkom’s representation as to the fairness of the $55 price per share for sale of the Company—a subject that the Board had never previously considered,” 389Id. at 877. and failed to inform itself “as to the intrinsic value of the Company.” 390Id. at 874; see also id. at 866 (“Apart from the Company’s historic stock market price, and Van Gorkom’s long association with Trans Union, the record is devoid of any competent evidence that $55 represented the per share intrinsic value of the Company.” (footnote omitted)). Neither Van Gorkom nor any other director read the agreement prior to its signing. 391Id. at 869. The proposed merger agreement was approved at a board meeting that lasted two hours. 392Id.

Just as in the Disney litigation, the evidence in the record—although presented at trial—was generated during extensive and probing discovery. 393See, e.g., id. at 864 n.1 (“[F]ollowing extensive discovery, the Trial Court denied the plaintiff’s motion for preliminary injunction . . . .”). Only party-driven depositions, supported by document discovery, could have elicited the kinds of admissions and conflicting accounts and impressions on the part of the various actors involved in the deal.

To appreciate this point, it is critical to recognize that this type of information could not have been obtained in a civil suit by shareholders in a civil law jurisdiction because of the lack of a discovery process. In Germany, for example, the only case that engages in the close examination of a board’s decision-making process in the sale of a company is the Mannesmann case. 394Bundesgerichtshof [BGH] [Federal Court of Justice] Dec. 21, 2005, Neue Juristische Wochenschrift [NJW] 522, 2006 (Ger.); see also, Franklin A. Gevurtz, Disney in a Comparative Light, 55 Am. J. Comp. L. 453, 459–62 (2007) (discussing the case in English). In that case, management was charged with paying Mannesmann’s CEO an improper £10 million bonus for the successful sale of the company. 395Gevurtz, supra note 394, at 461. But Mannesmann, which was hailed as the first fiduciary duty case in Germany, was, in fact, a criminal prosecution that relied on the investigative powers of a criminal prosecutor. 396Id. at 455.

In contrast, the U.S. system incentivizes parties and trial courts to focus on the facts. 397In order to avoid de novo review of the law on appeal, trial courts have an incentive to heavily emphasize the facts, which are subject to a “clear error” standard on appeal. Not only did the Van Gorkom court explicitly acknowledge its heavy dependence on the factual record produced by discovery, it emphasized the importance of entire record review in such fiduciary duty cases going forward. 398Van Gorkom, 488 A.2d at 871. This result means that the “proceduralization” of the boardroom and the Court’s interpretation of the business judgment rule in Van Gorkom only make sense if private litigants are in a position to discover what procedures were actually followed.

In this way, litigation discovery has repeatedly underwritten substantive corporate fiduciary duty law. One can reread the seminal Delaware fiduciary duty cases, including the previously discussed Van Gorkom, Caremark, and Disney cases, as well as Stone v. Ritter, 399911 A.2d 362 (Del. 2006).In re Citigroup Inc. Shareholder Derivative Litigation, 400964 A.2d 106 (Del. Ch. 2009). and others in this light, focusing on how judicial standards have been shaped by the practice and results of litigation discovery over time. 401See Parsons & Tyler, supra note 263, at 479 (discussing the “importance of the Delaware Court of Chancery’s procedural rules to Delaware’s substantive law of corporations,” in the most recent spate of shareholder challenges to mergers and acquisitions). Parsons and Tyler discuss the use of motions to expedite discovery to sort out meritorious from unmeritorious cases. Id. at 499–500.

2. The Development of DGCL Section 220 Case Law

Another good example of how discovery has shaped substantive law and shareholder rights is the changing standards governing shareholder information requests under section 220 of the Delaware General Corporation Law (DGCL). 402Del. Code Ann. tit. 8, § 220 (West Supp. 2014). While originally different in scope from discovery itself, recent case law shows how shareholder information requests have changed substantially from very limited, pinpointed inquiries to much broader, discovery-like requests for the production of entire categories of documents, including ESI, and that courts are enforcing such requests. 403Stephen A. Radin, The New Stage of Corporate Governance Litigation: Section 220 Demands—Reprise, 28 Cardozo L. Rev. 1287, 1412 (2006). In part, broader shareholder information requests can be seen as a response to the increasing difficulty of overcoming the business judgment rule in shareholder litigation. In part, this development might be seen as a correction of the increasing difficulty of obtaining discovery in securities class actions after the PSLRA. In a 1997 article in the Boston University Law Review, Professors Randall Thomas and Kenneth Martin thus argued for the use of state inspection statutes to obtain discovery in federal securities fraud actions. 404Thomas & Martin, supra note 214, at 70.

In Rales v. Blasband 405634 A.2d 927 (Del. 1993). and Grimes v. Donald, 406673 A.2d 1207 (Del. 1996). the Delaware Supreme Court noted that, “[s]urprisingly, little use has been made of section 220 as an information-gathering tool in the derivative context,” to engage in a prefiling investigation of corporate mismanagement. 407Id. at 1216 n.11 (quoting Rales, 634 A.2d at 934–35 n.10). In Security First Corp. v. U.S. Die Casting & Development Co. 408687 A.2d 563 (Del. 1997). and the subsequent case law,

The court emphasized that books and records inspections permitted under Section 220 and discovery permitted in the context of a shareholder derivative action are “not the same and should not be confused.” Section 220 books and records requests, the court stated, must be “circumscribed with rifled precision,” while discovery in other contexts “may often be broader” in scope. 409Radin, supra note 403, at 1294–95 (footnote omitted) (quoting Sec. First Corp., 687 A.2d at 570).

But in Brehm v. Eisner, for example, the Delaware Court dismissed a shareholder action against the Disney Corporation for waste in the hiring and firing of Ovitz, but ordered the Chancery to allow plaintiffs leave to amend after they made use of the “tools at hand”—that is, a valid shareholder information request under DGCL section 220—“to develop the necessary facts for pleading purposes.” 410746 A.2d 244, 266–67 (Del. 2000) (internal quotation marks omitted).

Delaware books and records demands under section 220 also play an increasing role in compensation and corporate governance disputes outside the derivative litigation context. 411See John F. Grossbauer & Charles T. Williams, III, The Increasing Role of Delaware Books and Records Demands in Compensation and Governance Disputes, Potter Anderson Corroon LLP (Nov. 1, 2004), http://www.potteranderson.com/publication/the-increasing-role-of-delaware-books-and-records-deman ds-in-compensation-and-governance-disputes (describing the use of section 220 in the context of executive compensation and proxy contests). Litigators have identified a “new frontier in Section 220 . . . as a means of gathering information for other purposes,” specifically in matters surrounding executive compensation and proxy contests. 412Id.

The court has persisted in drawing a distinction between the “full panoply of discovery rights” 413Radin, supra note 403, at 1339. and the “rifled precision” of section 220 demands. 414Brehm, 746 A.2d at 266. However, while the scope of a section 220 request is limited to a “proper purpose,” and the Delaware Supreme Court has insisted that the requesting shareholder must “make specific and discrete identification, with rifled precision, of the documents sought,” 415Id. we note that section 220 requests increasingly include broad categories of documents, corporate internal communications, and ESI.

The Delaware Supreme Court’s ruling in the Saito v. McKesson HBOC, Inc. derivative action contributed to this development. 416806 A.2d 113 (Del. 2002). In McKesson, the Court of Chancery granted McKesson’s motion to dismiss a shareholder lawsuit challenging a stock-for-stock merger between McKesson Corporation and HBO & Company after the revelation of certain accounting irregularities. 417Id. at 115. Saito, the only plaintiff, filed suit to enforce his demand for inspection when the corporation rejected his information request. 418Id. The Delaware Chancery held that, while Saito stated a “proper purpose” for his demand, the scope of inspection “only extended to potential wrongdoing after the date on which Saito acquired his McKesson stock.” 419Id. at 116. The Delaware Supreme Court reversed, substantially expanding the scope of inspection to time frames well before Saito acquired his shares, ruling that section 327 of the DGCL did not “defin[e] the temporal scope of a stockholder’s inspection rights under § 220.” 420Id. at 117. The court also reasoned that shareholders could make use of information gathered via section 220 request for purposes other than filing shareholder actions. 421See id. at 116.

The court thus expanded the scope of documents a corporation is required to make available in response to a books and records demand in two respects. Interpreting DGCL section 220 broadly, following the civil procedure principle that the scope of litigation discovery should be interpreted broadly, the court appears to have made “relevance” the touchstone of its analysis, instead of the temporal limitations under DGCL section 327, which it arguably could have relied upon to justify a much narrower scope for the shareholder information request. Furthermore, the limitation of the scope of inspection to a qualifying “proper purpose” was in practice considerably expanded beyond what the “rifle shot precision” doctrine would allow. 422See Radin, supra note 403, at 1292 n.33.

The language of the information requests that the Delaware Chancery Court allowed on remand in McKesson shows precisely how broad and discovery-like such section 220 requests have become. 423See McKesson Corp. v. Saito, 818 A.2d 970 (Del. Ch. 2003) (citing Saito, 806 A.2d 113). (See the appendix for the McKesson information request.)

Scholars have now proposed that Delaware courts explicitly include access to ESI. Their reasoning is that in interpreting section 220, the Chancery should look to the discovery provisions in the Federal Rules of Civil Procedure. 424Francis G.X. Pileggi, Kevin F. Brady & Jill Agro, Inspecting Corporate “Books and Records” in a Digital World: The Role of Electronically Stored Information, 37 Del. J. Corp. L. 163, 171–72 (2012) (arguing for a “readily accessible” standard for the restoration of back-up tapes in response to a “books and records” request, instead of the more demanding “reasonably accessible” standard under Fed. R. Civ. P. 26 (citing Zubulake v. UBS Warburg LLC, 220 F.R.D. 212, 218 (S.D.N.Y. 2003))).

We understand these developments to be a response to the retreat from discovery in shareholder derivative actions and in securities fraud class actions. Because information gained through shareholder inspection can be used to develop facts to support both types of claims, the court appears to be giving back with one hand what the Court and Congress took away with the other. It is worth asking whether expanding the tools and the scope of section 220 requests is preferable to allowing discovery proper in the first place. One might favor expanded section 220 requests over full discovery because they allow judges control in shaping a plaintiff's investigatory demands, while nonetheless still leaving the initial formulation of the request and the collection, processing, and review of the information obtained to the parties themselves. This approach appears to strike a balance between U.S. decentralized fact investigation and the European concentration of fact investigation in the hands of the judge.

3. Regulatory Changes and Federal Law Reform

Similarly, discovery influences changes in administrative and federal regulation. The new rules for gatekeepers, adopted by the Sarbanes-Oxley legislation in light of corporate scandals, are intricately calibrated to take advantage of private enforcement and, very specifically, of the access that gatekeepers have to company internal information through corporate internal investigations and through defensive discovery.

New regulations, such as the “Compensation Disclosure and Analysis” section in public filings of company financial statements and “say on pay” provisions in the Dodd-Frank Wall Street Reform and Consumer Protection Act, have been influenced, at least in part, by public outrage about executive compensation practices disclosed as a result of shareholder litigation involving such companies as Tyco, Enron, WorldCom, Global Crossing, and Disney, 425See, e.g., Alan Palmiter & Frank Partnoy, Corporations: A Contemporary Approach 681–85 (2010); Coffee, supra note 301, at 1404 n.6; Gordon, supra note 233, at 686 (“Part of what fuels the sense that the executive compensation-setting process is seriously flawed are the high-profile cases of exceptionally large payouts or mega-stock option grants, particularly where the firm’s subsequent performance is subpar, if not disastrous.”). as well as the hundreds of stock options cases that went forward in 2006 and 2007. 426See David I. Walker, Unpacking Backdating: Economic Analysis and Observations on the Stock Option Scandal, 87 B.U. L. Rev. 561, 563 (2007).

Many of the stock options cases were settled, but enough went to discovery to generate considerable public outrage. 427See, e.g., M.P. Narayanan, Cindy A. Schipani & H. Nejat Seyhun, The Economic Impact of Backdating of Executive Stock Options, 105 U. Mich. L. Rev. 1597, 1599, 1638 (2007); Eric Dash, Report Estimates the Costs of a Stock Options Scandal, N.Y. Times, Sept. 6, 2006, at C4; Mark Maremont & Charles Forelle, Bosses’ Pay: How Stock Options Became Part of the Problem, Wall St. J. (Dec. 27, 2006, 12:01 AM), http://online.wsj.com/news/articles/SB116718927302760228. Even corporations whose cases were settled or dismissed were often subject to a corporate internal investigation that generated sufficient public information, either by disclosing the SLC report or by including details of an SLC’s findings in the notice of settlement, and the settlement agreement to shareholders, to document widespread negligence or malfeasance on the part of compensation committees, CEOs, and CFOs of public companies. 428See Perfect Payday: Options Scorecard, Wall St. J. (Sept. 4, 2007), http://online.wsj.com/public/resources/documents/info-optionsscore06-full.html (documenting the 2006–2007 stock options backdating probes, revelations, and outcomes on a company for company basis).

D. Litigation Discovery Complements and Enforces Securities Disclosure—Discovery as Ex Post Disclosure

Our previous discussion about the role of offensive discovery, defensive discovery, and discovery-related practices for generating information about corporate wrongdoing leads us to conclude that discovery is a form of ex post disclosure. This type of ex post disclosure is unique to the U.S. system and virtually nonexistent in every other country. This view of discovery as supplementing disclosure has not been previously explored.

Litigation discovery complements and enforces the securities disclosure regime. It constitutes an ex post form of particularized disclosure that enforces and complements the standardized ex ante disclosures required by the SEC pursuant to SEC regulations, the 1933 Securities Act, and the 1934 Securities Exchange Act. The first obvious thing to recognize is that litigation discovery allows private actors to explore and generate information about corporate internal wrongdoing that is not already known to the markets (or other public and private actors), via a mechanism entirely distinct from regulatory disclosure.

Disclosure under the securities laws proceeds ex ante in accordance with disclosure requirements that have been regulated ex ante in a standardized form applicable to all companies. 429See generally Alan R. Palmiter, Toward Disclosure Choice in Securities Offerings, 1999 Colum. Bus. L. Rev. 1, 17.Ex ante disclosure is necessarily limited and summary in nature: it is made at regular time intervals known in advance by the issuer, or generated by known events, and the issuer determines how to interpret the underlying facts. Once the formal disclosures requirements have been met, they are generally not challenged ex ante by the regulator.

In contrast, litigation discovery occurs ex post and generates additional disclosures that were not required at the time a company completed its standardized ex ante regulatory disclosure. It is particularized and extensive. And it enables the exploration of information that goes far beyond the disclosure of “material information” that issuers must make. Private parties may obtain additional disclosure in situations that were not specifically contemplated by the regulator, including specific transactions that a company was not required to disclose, but which security holders now want to scrutinize. This can supplement information to back up more general disclosures in an issuer’s regulatory filings (like the original records of stock options grants to document the timing and market value of certain stock options awards), or concerning particular individuals and related parties involved in particular transactions (as countless cases from Van Gorkom to Disney show). 430See supra Part III.A.

Litigation discovery thus fills gaps and loopholes in periodic, mandatory public company disclosure. Moreover, it allows plaintiffs to challenge an issuer’s interpretation of particular facts or developments, including the issuer’s interpretation of “materiality.” 431See supra note 41.

While regulators and prosecutors also have the ability to engage in such probes, the fact that private litigants also do—and with some frequency—changes the rules of the game. 432See Coffee, supra note 13, at 245 (“In the United States, public enforcement of law is supplemented by a vigorous, arguably even hyperactive, system of private enforcement. Relying on class actions and an entrepreneurial plaintiffs’ bar motivated by contingent fees, the U.S. system of private ‘enforcement by bounty hunter’ appears in fact to exact greater annual aggregate sanctions than do its public enforcers. This system has no true functional analogue anywhere else in the world.”). Regulators in the United States do not have the resources to provide the current level of verification and enforcement of securities disclosure without the substantial resources dedicated by private parties. 433See, e.g., Jason Krause, Government Agencies Look Within to Solve E-Discovery Woes, Law Tech. News, Aug. 1, 2011, at 56 (“[T]here is a misperception that government agencies have unlimited resources to shoulder the burden of electronic data discovery costs in big cases like [the antitrust investigation of Google] . . . . Government litigators often manage EDD without proper software.”); Melanie Rodier, Lack of SEC Resources Could Be Biggest Hurdle for Volcker Rule, Wall St. & Tech. (Jan. 21, 2011), http://wallstreetandtech.com/articles/229100012/. Shareholders and the plaintiffs’ bar are widely recognized to have very different motivations from regulators and prosecutors for filing shareholder actions. 434See, for example, the dramatic difference between the SEC’s $150 million settlement with Bank of America regarding its acquisition of Merrill Lynch, compared with the $2.43 billion settlement shareholders ultimately achieved. Silver-Greenberg & Craig, supra note 255. Moreover, materials and information obtained through discovery are public by default. 435See supra note 65 and accompanying text. A plaintiff may thus share information that he or she has obtained with others, including the press and regulators. Management cannot expect the same kind of “confidential treatment” from private plaintiffs that it often enjoys when dealing with the regulator. 436See SEC Enforcement Manual § 4.3.1 (2013), available at http://www.sec.gov/divisions/enforce/enforcementmanual.pdf (describing confidential treatment requests).

Modern discovery is what enables private attorneys general to pursue securities violations or fraud claims. For example, the ability of private plaintiffs to enforce strict liability for failure to disclose “material information” in a registration statement would be meaningless without the ability to engage in broad discovery—even if all fee arrangements, aggregate litigation rules, and standing requirements stayed the same. Similarly, the private right of action for securities fraud claims under section 10(b) of the Securities Exchange Act would mean little if plaintiffs did not have the ability to obtain detailed information from issuers. Such information is critical for testing not just whether particular developments or events were disclosed, but when they were disclosed, what the underlying facts were, how management interpreted these facts when it became aware of the information, and ultimately whether there was an intent to defraud. To reconstruct the kind of detailed factual accounts and timelines about who knew what, when they knew it, what they did about it, and why, requires substantial factual investigation. The very least one can say on behalf of the private exploration of facts concerning corporate internal wrongdoing is that it has the potential to generate additional information that may affect the price of a firm’s securities and its market valuation. 437There is empirical evidence on the effect of filings on share prices. See Gregg A. Jarrell & Annette B. Poulsen, Stock Trading Before the Announcement of Tender Offers: Insider Trading or Market Anticipation?, 2 J.L. Econ. & Org. 225, 228–29 (1989); Gregg A. Jarrell, James A. Brickley & Jeffry M. Netter, The Market for Corporate Control: The Empirical Evidence Since 1980, J. Econ. Persp., Winter 1988, at 49, 52.

Discovery interacts with the securities disclosure regime in other ways. The additional information about corporate internal practices that comes to light by way of corporate internal investigations and litigation discovery becomes available to the regulator. Discovery thus serves as an additional source of intelligence for the regulator about constantly evolving internal corporate governance practices and strategies that might be used to avoid full compliance with the regulatory regime. The SEC relies on corporate internal investigations and private securities litigation, because it cannot investigate every case.

IV. Objections and Replies

Numerous objections have been raised to our thesis that litigation discovery plays an integral role in the U.S. corporate governance regime.

Discovery, it is argued, is increasingly more difficult to obtain, particularly in shareholder suits. Moreover, even if discovery is available, the rules and practices of discovery have created perverse incentives that may well undermine truth seeking more than they promote it.

First, it has been argued that federal and state laws have increasingly curtailed the ability of plaintiffs to obtain litigation discovery in all types of shareholder suits. 438See Jessica Erickson, Corporate Misconduct and the Perfect Storm of Shareholder Litigation, 84 Notre Dame L. Rev. 75, 93–94 (2008) (recognizing that federal laws have dramatically affected plaintiffs’ access to information in securities litigation, limiting many to only self-obtained information). Critics point to the heightened pleading requirements under the PSLRA, that have considerably raised the threshold for obtaining discovery in securities class actions, and the discovery stay pending decision on a motion to dismiss. 43915 U.S.C. § 78u-4(b)(1)–(3) (2012) (including scienter and other specific pleading requirements). Choi and Thompson argued that heightened pleading and other elements of the PSLRA “all work to increase the costs for plaintiffs’ attorneys seeking to bring a class action” and have proven to be a difficult hurdle without access to witnesses or documents. See Choi & Thompson, supra note 194, at 1529; see also 15 U.S.C. §§ 77z-1(b)(1), 78u-4(b)(3)(B) (providing a mandatory discovery stay during the pendency of motion to dismiss). Further there is an extensive literature on shareholder derivative actions that strongly suggests their disciplining function is limited. Such actions face significant procedural hurdles, rarely lead to success on the merits, and are frequently either dismissed or settled before a case gets to discovery. 440See supra notes 206–14 and accompanying text.

Civil procedure scholars have also objected to our thesis. They have insisted that the modern discovery regime did not live up to the expectations of the drafters in that adversarial discovery offers litigants both motive and substantial opportunity to conceal damaging information during discovery. 441See, e.g., Wayne D. Brazil, The Adversary Character of Civil Discovery: A Critique and Proposals for Change, 31 Vand. L. Rev. 1295, 1330 (1978). The most frequently heard complaint about discovery is that witnesses are so extensively prepped before depositions that their testimony often becomes worthless. 442See id. at 1330–31. Civil procedure scholars have also noted that the litigation process has been reformed in a way that calls into question the account we have given of the judge in the U.S. litigation process. 443See Steven S. Gensler, Judicial Case Management: Caught in the Crossfire, 60 Duke L.J. 669, 705 (2010) (recognizing that any reform that would give judges “more case-management power” must be designed so as to respond to critics); see also Judith Resnik, Managerial Judges, 96 Harv. L. Rev. 374, 378 (1982). These critics claim that the “case management revolution” of the 1990s significantly expanded the role and authority of judges in the discovery process. 444Steven S. Gensler & Lee H. Rosenthal, Managing Summary Judgment, 43 Loy. U. Chi. L.J. 517, 520 (2012); see also Gensler, supra note 443, at 680–81. Moreover, cumulative developments in civil procedure over the past several decades, such as the imposition of sanctions regimes in the 1980s (FRCP Rules 11 and 37) and the heightening of pleading requirements in such cases as Bell Atlantic v. Twombly and Ashcroft v. Iqbal during the past decade, have arguably all sought to curb the original drafters’ enthusiasm for discovery, and have reduced the likelihood that plaintiffs will obtain discovery as well as the scope and breadth of that discovery.

Finally, it has been argued that the threat of litigation discovery may well have made it more difficult to obtain evidence of corporate internal wrongdoing because firms have responded strategically to its threat. For example, firms have put elaborate document management and “retention” policies in place that destroy sensitive documents, e-mails, and other information on a regular basis. 445See David Ferris, Setting Retention Policy for Electronic Information 2 (2011), available at http://www.ironmountain.com/Knowledge-Center/Reference-Library/View-by-Document-Type/White-Papers-Briefs/S/Setting-Retention-Policy-for-Electronic-Information.aspx (“Producing information in response to an eDiscovery request can be extremely time-consuming and costly. If the material has been deleted, the costs of production are obviously reduced. Following a clear, communicated retention policy adds defensibility to the deletion of electronic information.”). On this view, the threat of exposure has resulted in perverse incentives to eliminate important records, to avoid creating them in the first place, and may result in forcing decision-making underground. “Most executives are smart enough not to say incriminating stuff in emails these days,” as a columnist for The American Lawyer states in an article reviewing a jury’s decision to acquit former Citigroup employee Brian Stoker of negligence in a suit brought by the SEC. 446Susan Beck, Susan Beck’s Summary Judgment: Stoker Jury Foreman Explains How Verdict Didn’t Absolve Citi, Am. Law. Litig. Daily (Aug. 1, 2012, 12:00 AM), http://www.litigationdaily.com/id=1202565746363/Susan-Beck’s-Summary-Judgment:-Stoker-Jury-Foreman-Explains-How-Verdict-Didn’t-Absolve-Citi?slreturn=20140105010532.

In light of these criticisms, how can it be maintained that litigation discovery is so important for corporate governance? In the following sections, we discuss and address each of these objections.

A. The Special Hurdles to Discovery in Shareholder Actions

In answering objections based on the hurdles of discovery in shareholder actions, it is important to recognize that legal regimes are neither static nor do they typically reflect the neutral implementation of an ideal policy blueprint. Legal rules are constantly evolving and their development is a political process that reflects the pressures that dominant interests bring to bear on the policymaking process at various junctures. 447Milhaupt & Pistor, supra note 5, at 40. Given the necessarily political process of negotiating legal rules for investor protection, it is therefore not surprising that the balance between robust litigation discovery and its burdens on corporate defendants and their officers and directors have been contested for decades. 448Most academics agree there are problems but defend securities class actions as a necessary mechanism for shareholder monitoring and general deterrence. See, e.g., Coffee, supra note 13, at 245; Jill E. Fisch, Confronting the Circularity Problem in Private Securities Litigation, 2009 Wis. L. Rev. 333, 345–48; Merritt B. Fox, Why Civil Liability for Disclosure Violations When Issuers Do Not Trade?, 2009 Wis. L. Rev. 297, 310–18; Lawrence E. Mitchell, The “Innocent Shareholder”: An Essay on Compensation and Deterrence in Securities Class-Action Lawsuits, 2009 Wis. L. Rev. 243, 243. These academics are largely responding to new criticism that today’s securities class actions yield little compensation or deterrence. See John C. Coffee, Jr., Reforming the Securities Class Action: An Essay on Deterrence and Its Implementation, 106 Colum. L. Rev. 1534, 1534 (2006). As Professors Stephen Choi and Robert Thompson note, “The PSLRA was not the first time policymakers had sought to curb the abuses of representative litigation brought to police management behavior within corporations.” 449Choi & Thompson, supra note 194, at 1491.

The empirical literature on the impact of the PSLRA reforms has been voluminous but hardly conclusive. Important for our purposes is whether the PSLRA significantly compromised the integral role of litigation discovery in policing corporate management. Summarizing the empirical evidence available in 2006, ten years after passage of the PSLRA, Choi and Thompson reported that the number of pre-PSLRA filings and post-PSLRA filings remained stable. 450Id. at 1496. Higher value claims were more likely to be filed after the PSLRA. 451Id. at 1497–98. And empirical studies reported an increase in the time between filing and settlement in post-PSLRA cases. 452Id. at 1498. Almost all securities class actions are either settled or dismissed. 453Stephen J. Choi, Karen K. Nelson & A.C. Pritchard, The Screening Effect of the Private Securities Litigation Reform Act, 6 J. Empirical L. Stud. 35, 35–36 (2009) (“Trials . . . are almost unheard of in this area . . . .”). However, there is evidence that a greater percentage of cases are now dismissed or settled prior to the motion to dismiss, raising concerns about underdeterrence. 454Id. at 35 (“There is evidence . . . that pre-PSLRA nonnuisance claims would be less likely to be filed under the PSLRA regime. The latter result, which we refer to as the screening effect, is particularly pronounced for claims lacking hard evidence of securities fraud or abnormal insider trading.”). This raises concerns about the balance that has been struck by the PSLRA. In light of the mandatory discovery stay, this development suggests plaintiffs have less opportunity to obtain discovery. 455See id. at 38. Actions claiming fraud must generally satisfy the heightened pleading requirements under Fed. R. Civ. P. 8(a), requiring plaintiffs to plead the circumstances surrounding the fraud with specificity—but not intent. The PSLRA additionally requires that plaintiffs allege facts that support a “strong inference” of scienter. 15 U.S.C. § 78u-4(b)(2) (2012). Most securities fraud complaints are now dismissed before the plaintiffs get any discovery. Our account of the discovery process therefore suggests that the PSLRA was problematic.

This development does not mean that discovery is no longer important to “outcomes.” For our purposes, “outcomes” does not refer to whether plaintiffs win or lose, or how and at what stage a case is ultimately disposed of. Rather, “outcomes” refers to the information production, disciplining effects, and positive externalities that flow from strategic behavior within an enforcement scheme structured by litigation discovery. Only by engaging in more fine-grained analyses of how the contest between plaintiffs seeking discovery in shareholder litigation, and defendants seeking to avoid it, plays out, and how the law in practice intricately configures this contest, can we understand the implications of the discovery system. This is what we have done in Part III. Whereas discovery in securities class actions is very hard to obtain, such actions are often filed together with parallel state court actions that allege fiduciary duty violations based on the same facts and circumstances. 456“[T]he intertwining of 10b-5 violations and breaches of fiduciary duty . . . pervades most derivative actions against directors where a transfer of securities is involved . . . .” Kurt M. Swenson, Remedies for Private Parties Under Rule 10b-5, 10 B.C. Indus. & Com. L. Rev. 337, 355–56 (1969) (discussing Mutual Shares Corp. v. Genesco, Inc., 384 F.2d 540 (2d Cir. 1967)). This is also why SLUSA was passed in the wake of the PSLRA. Plaintiffs, however, have been able to adjust by bringing claims in the form of shareholder derivative actions in state court. Settlements in federal securities class actions may thus be informed by discovery in state actions, or by information produced by special litigation committee investigations on the part of defendants in a state derivative action, with all the positive consequences of such SLC investigations.

Nonetheless, judges have the flexibility to allow securities fraud actions to go forward, as the wave of stock options backdating cases showed in 2006 and 2007. 457See, e.g., Perfect Payday: Options Scorecard, supra note 428. The threat of discovery is therefore ever present even in securities litigation. As the Wall Street Journal’s Options Scorecard shows, and as already discussed in Part III, internal investigations also generated earnings restatements, information for the SEC (which would typically obtain any SLC report subject to a confidential treatment agreement), corporate governance improvements, and other information for markets, even when they were subject to confidentiality and never filed with the court. 458Id. Some cases went to discovery, and in some cases, the Justice Department and state authorities launched criminal investigations. 459Id.

In response to the objection that shareholder derivative actions are frequently dismissed or settled prior to discovery, we again refer to our discussion in Part III on the information-producing impact and disciplining effects of internal investigations in such cases.

B. The FRCP’s Retreat from Liberal Discovery Since the 1970s

In spite of the “case management revolution” of the 1990s and other developments, the judicial role in discovery remains very limited when compared to civil law adjudication. To be sure, “case management” requirements and authority under FRCP 16 and 37 have increased since the 1980s, generating an entire literature on “managerial judges.” 460Resnik, supra note 443, at 378; see also, Richard L. Marcus, Reining in the American Litigator: The New Role of American Judges, 27 Hastings Int’l & Comp. L. Rev. 3, 17–20 (2003); Resnik, supra note 105, at 497–98. Although the new rules give judges significant authority to set deadlines and timetables for discovery, case management rules have not changed the basic structure of litigation. Parties conduct discovery adversarially. The tools remain the same, and the scope of discovery is still very broad. Although judges are now required to issue a scheduling order, there is no evidence that this interferes with thorough discovery. 461Moreover, deadlines are not infrequently extended. These changes represent little more than the proverbial fly on the elephant’s back. Comparatists who take this literature as somehow reflecting a dramatic change in U.S. discovery rules and practices have mistaken superficial similarities in separate discourses as functional equivalents. 462See infra notes 524–39 and accompanying text. Civil law countries are introducing small changes in light of their increasing contact with the U.S. system. See Huang, supra note 24, at 160 (arguing for conversion of procedural rules). But see, e.g., Verkerk, supra note 23, at 370 (“A remarkable conclusion is that the differences between the fact-finding arrangements in Continental and American jurisdiction are larger today than they were a century ago.”).

The case Acuna v. Brown & Root Inc. illustrates how far a judge might go under the case management rules to restructure the ordinary course of pretrial litigation and deny plaintiffs discovery. 463See 200 F.3d 335, 340–41 (5th Cir. 2000). In Acuna, the court required plaintiffs to submit expert affidavits to establish with particularity the illnesses complained of by 600 plaintiffs, who had allegedly suffered from defendants’ uranium mining and processing operations, before allowing plaintiffs any discovery at all. 464Id. at 337–38. When the plaintiffs’ attorneys failed to comply with the scheduling order for the expert reports, the judge dismissed the claims. 465Id. at 338. But Acuna is a controversial case that illustrates the limits of judicial case management in a situation where the class action mechanism was pushed to its limits. 466See id. The litigation involved a large plaintiff class, with each plaintiff suffering from different symptoms, occurring during different time frames, in different places, purportedly caused in different ways by one or more of a large class of diverse defendant companies engaged in a variety of different activities at different times, and in different places. Id. at 340. To be sure, in large mass-tort actions, case management has become more specialized. But these changes do not alter the fundamental structure of shareholder litigation.

Twombly and Iqbal have been influential and reflect the Supreme Court’s concern that discovery imposes substantial burdens on defendants that are not always justified. A WestlawNext database search in the thirty months after Iqbal was decided shows that federal courts issued 27,500 reported opinions citing both Twombly and Iqbal. “Twombly and Iqbal now rank among the Court’s most-cited decisions,” 467Rowe, Sherry & Tidmarsh, supra note 187, at 64. and the academic commentary on this line of cases is voluminous. But a study by the Federal Judicial Center on reported and unreported cases in twenty-three federal districts from 2006 to 2010 showed that, while the rate of filing motions to dismiss for failure to state a claim increased over this period, the rate at which such motions were granted without leave to amend did not show a statistically significant increase in any area other than cases challenging mortgage loans. 468Id. And this was likely the result of a tripling of such cases from 2006 to 2010. 469See Joe S. Cecil et al., Fed. Judicial Ctr., Motions to Dismiss for Failure to State a Claim After Iqbal: Report to the Judicial Conference Advisory Committee on Civil Rules, at vii, 12 (2011). The empirical evidence therefore fails to show that discovery has been restricted by these cases in shareholder litigation.

Moreover, even if there has been a retreat in discovery as a result of Twombly and Iqbal, the retreat is marginal compared to the fundamental difference between the U.S. discovery regime and the lack thereof in civil law and other common law jurisdictions. To the extent that judges exert more control over whether a case goes to discovery, this does not detract from our more general claim that discovery serves as an ex post mechanism for disclosure. If judges are able to weed out meritless cases prior to discovery, that only benefits shareholders.

C. Dysfunctions of Discovery

There is considerable disagreement about the value and efficiency of discovery. Critics have claimed that the adversarial system has distorted the drafters’ original intent and turned discovery into a game of hide and seek. At the same time, discovery has been severely criticized for being far too costly, intrusive, and wasteful. 470See supra note 42.

In response, it should first be noted that no serious observer maintains the general claim that discovery fails to afford litigants the information they need to prosecute their cases. Even the most ardent critics of discovery do not claim that discovery is ineffective. To the contrary, their claims are that discovery is used too frequently and extensively, burdens litigants with costly discovery requests, and encourages “open-ended fishing expeditions” that cause excessive delays in the disposition of cases, clogging the judicial system and overwhelming the judiciary. 471See Beisner, supra note 33, at 1–2 (internal quotation marks omitted). Worse, discovery, they argue, is frequently abused and coerces litigants into unjust settlements simply to avoid its tremendous costs, which have skyrocketed since the advent of e-discovery. 472Id.; see also Final Report on the Joint Project of the American College of Trial Lawyers Task Force on Discovery and the Institute for the Advancement of the American Legal System 9 (2009), available at http://www.actl.com/AM/Template.cfm?Section=Home&template=/CM/ContentDisplay.cfm&ContentID=4053; Henry S. Noyes, Good Cause Is Bad Medicine for the New E-Discovery Rules, 21 Harv. J.L. & Tech. 49, 67–68 (2007) (“[E]-discovery is more time-consuming, more burdensome, and more costly than conventional discovery.”); Martin H. Redish, Electronic Discovery and the Litigation Matrix, 51 Duke L.J. 561, 592 (2001). While e-discovery certainly poses many new challenges for litigants, these claims are controversial and do not seem to be borne out by the empirical evidence. 473 See infra notes 491–95. Moreover, such claims are not new and predate the advent of e-discovery. 474See, e.g., William H. Erickson, The Pound Conference Recommendations: A Blueprint for the Justice System in the Twenty-First Century, 76 F.R.D. 277, 288 (1978) (observing that “[w]ild fishing expeditions . . . seem to be the norm,” and lamenting the “[u]necessary intrusions into the privacy of the individual, high costs to the litigants, and correspondingly unfair use of the discovery process as a lever toward settlement” that had come to characterize the American legal system).

Critics have argued that the rules and practices of discovery have created perverse incentives that undermine truth-seeking more than they promote it. Civil procedure scholars have pointed out that the modern discovery regime did not live up to the expectations of the drafters in that adversarial discovery offers litigants both motive and substantial opportunity to conceal damaging information during discovery. 475Subrin, supra note 111, at 706–07, 728. A frequently heard complaint is that witnesses are so extensively coached before depositions that their testimony often becomes worthless. 476See, e.g., Ronald J. Allen, Idealization and Caricature in Comparative Scholarship, 82 Nw. U. L. Rev. 785, 786 (1988) (discussing Langbein’s criticism of depositions and citing contrary evidence).

With regard to witness preparations, it is useful to recall that the authority under the modern discovery rules to cross-examine a witness under oath during deposition is undoubtedly the most powerful tool in an attorney’s investigative arsenal—in spite of the frequent criticism that deponents are all too well prepared for testimony by opposing counsel. Critics rarely consider that the thorough preparation of the examiner is a very important distinguishing feature of U.S. witness examination. The discovery rules allow the parties to determine the order in which they use the discovery tools and notice depositions. A common strategy is to document and lock in as many facts as possible via interrogatories, document discovery, and depositions of other knowledgeable but less important witnesses, so that the examiner has a thorough understanding of the case before taking the final depositions of key witnesses. 477Another strategy is to take a first deposition of key witnesses early in the case to lock them into positions before defensive discovery allows them to assess how far they can bend the truth without their testimony being later challenged and contradicted by documents and other witness testimony, but to reserve the right to reexamine the witness again at a later date. A plaintiff’s attorney in a shareholder action is likely to save depositions of the most important witnesses—typically executives and other named defendants—for last. 478By contrast, in civil law systems, witness examinations occur issue by issue in an order entirely determined by the judge. A judge likely would take the most important witnesses first, given that the judge is always looking for dispositive issues. See Langbein, supra note 23, at 830, 846.

During a deposition, the examiner may ask open and leading questions, confront the witness with his or her own prior statements, and refresh a witness’s memory with the help of copious documents and e-mails previously disclosed during document discovery—which the examiner reviews, organizes, and strategically deploys during questioning. ESI is particularly devastating in that e-mails are produced in strings of correspondence—permitting the reconstruction of who said what to whom, who was in the loop, who was later apprised of a conversation on what date and at which time, and who was excluded from subsequent conversations. ESI documents may have to be produced with metadata, containing information on whose hard drive they came from, when they were created and revised, and supplying all drafts and revisions with time-stamps and author identification together with the original document—making it very hard for a witness to credibly deny knowledge of specific transactions, communications, routines, and practices that obviously bear the witness’s electronic signature. 479See generally Philip J. Favro, A New Frontier in Electronic Discovery: Preserving and Obtaining Metadata, 13 B.U. J. Sci. & Tech. L. 1 (2007) (discussing the hazards of metadata and recent developments in the case law requiring production of metadata).

If an examiner is well prepared, depositions tend to be very productive. The only relatively safe strategy for resisting is to state “I do not recall” to every question. While this does occur—as it did, for example, in the deposition of Arthur Andersen witnesses in the Enron cases—it is extremely embarrassing to the witness. Although this strategy may be open to witnesses who have no professional reputation left to preserve and whose primary concern is to avoid incriminating themselves, it tends not to be available to executives for reputational, psychological, and institutional reasons. In any case, it is the rare witness who can sustain this type of testimony during the hours of humiliation that such behavior will inevitably draw from an examiner intent on obtaining information from an important witness. 480See, for example, the deposition of Timothy J. Mayopoulos, BofA’s associate general counsel in the Merrill Lynch case. See supra note 249 and accompanying text. And it may prompt the examiner to call the judge to threaten the witness with contempt, if the witness is worth the trouble. While an examiner may not get the precise admission she wants from a well-coached party-opponent, who might thereby be able to avoid liability, depositions are generally productive and extraordinarily uncomfortable for anyone who has something to hide. 481See Marcus, supra note 31, at 750 (“One of the striking things about American discovery is that parties often do reveal damaging information, and the importance of that discovered material in litigation cannot be overstated.”); see also H. Christopher Boehning & Daniel J. Toal, More Than a Security Risk: Director E-mails in Discovery, N.Y.L.J., Apr. 3, 2012 (discussing the vulnerability to discovery of electronic communications among directors).

Sophisticated document retention policies that regularly purge ESI also do not concern us. While theoretically the threat of discovery could incentivize the destruction of evidence, this is not plausible for several reasons. First, firm internal decisions and decision-making processes must be documented at every level of the firm hierarchy for business reasons and for purposes of regulatory compliance. Gaps in such documentation can severely compromise a firm’s business, its position in negotiations with regulators, and its ability to mount a defense in civil litigation. 482See generally Lee H. Rosenthal, A Few Thoughts on Electronic Discovery After December 1, 2006, 116 Yale L.J. Pocket Part 167 (2006), http://www.thepocketpart.org/images/pdfs/82.pdf (discussing issues likely to arise under new discovery laws and potential challenges for lawyers, litigants, and judges). If management expects to enjoy the protection of the business judgment rule, it must keep detailed records, as Van Gorkom’s “proceduralization of the boardroom” dictates in fiduciary duty cases. 483See Brehm v. Eisner, 746 A.2d 244, 259 (Del. 2000) (“[I]n making business decisions, directors must consider all material information reasonably available, and . . . the directors’ process is actionable only if grossly negligent.”). Decision-makers cannot anticipate ex ante which documents to destroy or to retain in the face of potential future legal challenges. Therefore, a point can be made that they are more likely to retain documents than destroy them.

Another reason document retention policies are not a concern is that the e-discovery regimes and practices strongly discourage routine destruction or mishandling of documents. The demands of e-discovery make meticulous document management protocols, including the firm-wide standardization of file types, software platforms, and IT systems critical. 484See generally Richard L. Marcus, Confronting the Future: Coping with Discovery of Electronic Material, Law & Contemp. Probs., Spring/Summer 2001, at 253; Richard Marcus, Only Yesterday: Reflections on Rulemaking Responses to E-Discovery, 73 Fordham L. Rev. 1 (2004). The cost of defending even a handful of depositions can become astronomical if a firm has failed to properly maintain the data. And judges will not necessarily protect firms from the consequences of such mismanagement where such problems could have been readily foreseen. Courts are proposing the use of counsel with expertise in computer systems and internal experts to assist with the challenges presented by e-discovery. 485See, e.g., The Chief Judge’s Task Force on Commercial Litig. in the 21st Century, Report and Recommendations to the Chief Judge of the State of New York 19 (2012), available at http://www.nycourts.gov/courts/comdiv/PDFs/ChiefJudgesTaskForceOnCommercialLitigationInThe21stpdf.pdf. Of course, it is always possible for management to go out of its way to destroy records for certain transactions. But, if anything, such activities have become far more difficult in an era of e-discovery, where metadata that trace all sorts of user activity in multiple ways are routinely subject to discovery. 486See James H.A. Pooley & David M. Shaw, Finding Out What’s There: Technical and Legal Aspects of Discovery, 4 Tex. Intell. Prop. L.J. 57, 64 (1995) (“There are sophisticated computer programs that can retrieve e-mail messages and computer files from the guts of the computer system long after those messages and files were thought to have been forever deleted. . . . [T]he mere fact that files were destroyed, perhaps systematically, could prove to be more valuable information than the actual files retrieved, especially if such fact is used to prove a cover-up, for example.”). Finally, failure to produce evidence and document destruction can lead to civil and criminal sanctions. 487See, e.g., Sekisui Am. Corp. v. Hart, 945 F. Supp. 2d 494, 509–10 (S.D.N.Y. 2013) (imposing sanctions in the form of monetary damages and an adverse inference jury instruction where a party destroyed electronic documents); Pension Comm. of the Univ. of Montreal Pension Plan v. Banc of Am. Sec., LLC, 685 F. Supp. 2d 456, 469–72 (S.D.N.Y. 2010) (discussing cases involving sanctions for nonproduction of evidence), abrogated by Chin v. Port Auth. of N.Y & N.J., 685 F.3d 135 (2d Cir. 2012). As a result, corporations are spending more money on recordkeeping systems. This is what legal counsel and, most recently, the Dodd-Frank Act require—to keep detailed records of all transactions. 488See generally Dodd-Frank Wall Street Reform and Consumer Protection Act, Pub. L. No. 111-203, 124 Stat. 1376 (2010) (codified in scattered sections of 12 and 15 U.S.C.).

D. The Costs of Discovery

The usual approach to discovery in the business community and academic scholarship is to point out the excesses and costs imposed by discovery. 489See supra note 42. A report by the U.S. Chamber of Commerce notes that up to 90% of litigation-related costs are incurred during discovery. 490Beisner, supra note , at 2 & n.7 (citing Thomas E. Willging et al., Fed. Judicial Ctr., Discovery and Disclosure Practice, Problems, and Proposals for Change: A Case-based National Survey of Counsel in Closed Federal Civil Cases 15 tbl. 4 (1997), available at https://bulk.resource.org/courts.gov/fjc/discovry.pdf; see also H.R. Rep. No. 104-369, at 37 (1995) (Conf. Rep.) (“[D]iscovery costs account for roughly 80% of total litigation costs in securities fraud cases” (quoting the general counsel of an investment bank)); Judicial Conference Adopts Rules Changes, Confronts Projected Budget Shortfalls, Third Branch (Admin. Office U.S. Cts., D.C.), Oct. 1999, at 2–3 (“Discovery represents 50 percent of the litigation costs in the average case and up to 90 percent of the litigation costs in cases in which it is actively used.”), available at http://www.uscourts.gov/News/TheThirdBranch/99-10-01/Judicial_Conference_Adopts_Rules_Changes_Confronts_Projected_Budget_Shortfalls.aspx). But note that the 90% number is highly misleading. The report cited in the Chamber of Commerce report stated that, in about 15% of cases, there was no discovery, that discovery costs on average constituted 3% of the value of a dispute, and that in a small number of large cases, 90% of litigation-related costs are incurred during discovery. Willging et al., supra, at 15 tbl. 4, 17 tbl. 6. That 90%, however, notably does not consider the value of the claim or dispute. Thus, even if discovery costs constituted 90% of the litigation costs, these costs quite likely reflect a very small percentage (i.e., in the single digits) of the value of the dispute.

However, empirical literature suggests that discovery abuse, and impositional discovery in particular, is not as significant a problem as is generally assumed. 491Separate empirical studies by the Federal Judicial Center in 1996 and the RAND Corporation in 1993 found that only a small minority of cases presented significant discovery problems, but that, among other findings, 54% of attorneys thought that the expense of discovery was about right in relation to the stakes of the case, 15% thought the expenses too high, 20% thought them too low, and the rest had no opinion. Rowe, Sherry & Tidmarsh, supra note 187, at 118 (reviewing empirical literature). Surveying the recent literature, Professor Danya Shocair Reda concludes that “[e]mpirical studies have repeatedly failed to document exorbitant costs or widespread discovery abuse; they have generally found the volume of discovery, the costs of discovery, and the total costs of litigation to be less than expected.” 492Reda, supra note 34, at 1111; see also Lawyers for Civil Justice et al., Litigation Cost Survey of Major Companies, Presentation at the Duke Law School 2010 Conference on Civil Litigation 2 (May 10, 2010), http://www.uscourts.gov/uscourts/RulesAndPolicies/rules/Duke%20Materials/Library/Litigation%20Cost%20Survey%20of%20Major%20Companies.pdf (“[F]ew empirical studies document the costs and utility of discovery.”). The robust finding of every independent empirical study on litigation costs is that the amount at stake in the litigation is “the greatest determinant of cost in civil litigation.” 493Reda, supra note 34, at 1104. A 2009 Federal Judicial Center study reported that for defendants the median ratio of discovery costs to stakes was 3.3%, 494Emery G. Lee III & Thomas E. Willging, Fed. Judicial Ctr., Preliminary Report to the Judicial Conference Advisory Committee on Civil Rules 42 (2009), available at http://www.fjc.gov/public/pdf.nsf/lookup/dissurv1.pdf/$file/dissurv1.pdf. a finding consistent with prior studies. 495Willging et al., supra note 490, at 17 tbl. 6 (1997); see also Linda S. Mullenix, The Pervasive Myth of Pervasive Discovery Abuse: The Sequel, 39 B.C. L. Rev. 683, 684 (1998) (“[D]iscovery data have remained surprisingly constant over time.”). Therefore, discovery costs should not be discussed in abstract and generic formulations based on absolute and aggregate numbers, but should be assessed relative to the stakes at issue in judicial cases, in a way that specifies and disaggregates the costs in relation to litigation outcomes.

The corporate law literature considers the costs of discovery in shareholder actions primarily from the perspective of the outcome that shareholders obtain from a particular lawsuit, but the benefits of discovery are not considered in the balance. 496See supra notes 44–46 and accompanying text. Indeed, this is also the law governing the judicial approval of settlements. The literature has focused on the negative externalities of discovery in encouraging and rewarding meritless strike suits. It has also focused on the misallocation of costs, in that ultimately the shareholders bear the costs of management wrongdoing in successful lawsuits. In other words, it has considered the negative externalities of discovery, but not its positive externalities. Moreover, it has focused on the micro level—and done so inadequately 497See supra Part III.A. —but ignored the macro level.

The only argument in the corporate law literature we are aware of that considers the private costs of discovery at the micro level in relation to its social benefits at the macro level is by Professor Lynn Stout. Stout provided a cost–benefit analysis of the additional burdens the PSLRA imposes on plaintiffs. 498Stout, supra note 42, at 712–13. She compared the losses from defending against meritless strike suits with the potential effects of discouraging legitimate suits. 499Id. at 713–14. By raising pleading requirements and preventing plaintiffs from obtaining discovery, investors are also discouraged from pursuing legitimate fraud claims. 500Stout notes that corporate losses from defending against meritless strike suits were described at the time as being in the hundreds of millions of dollars annually. But if one considers the possibility of a mere 1% discount applied by investors to shares of public companies, based on investor concern of underdeterrence (or Type II errors), the combined losses to the securities markets could surmount $100 billion. Id. See also supra notes 194–202 and accompanying text for a discussion of Type I and Type II errors. In other words, she concludes that the social costs of (perceived) underdeterrence far outweigh the private costs that the PSLRA was intended to avoid. 501Stout, supra note 42, at 713–14.

But even Stout’s argument considers only one of the positive externalities of discovery—the impact on the securities markets in connection with Type II errors. This Article has set forth a host of concrete positive externalities at the micro and macro level.

First, at the micro level, considering the private costs and benefits of every individual lawsuit separately, the literature generally does not properly value the benefits of corporate governance improvements in cases that are settled—and it entirely misses the disciplining (and sometimes remedial) benefits of defensive discovery that occur after the motion to dismiss, as well as the benefits of corporate internal investigations that occur prior to any motion to dismiss. These positive externalities are all direct results of evolving discovery rules and practices.

Still at the micro level, the literature generally does not take into account the positive externalities of discovery at the firm level after a particular shareholder action: deterrence of misconduct based on the experience of probing internal investigations by outside gatekeepers, improved internal controls and monitoring systems, corporate governance enhancements at the board level, and improved data management and litigation-hold protocols developed in connection with litigation preparedness.

These private benefits, of course, are not cost free. It can be argued, for example, that corporations face significant private costs because of the threat of discovery in that they must implement these changes, establish litigation preparedness procedures, bear the costs of litigation discovery, establish more sophisticated document management systems, etc. But even if these private costs are large, they must be considered in connection with the social benefits at the macro level.

At the macro level, the positive externalities include continuous information production for markets, regulators, and courts about constantly evolving corporate internal practices; the resulting developments of substantive law (both case law and statutory law); efficiency gains by courts and regulators, which heavily rely on private actors to generate the information critical for enforcement; deterrence based on the threat of discovery; the development of best practices of internal and external corporate governance; lower costs of capital (if Stout is right); and a general corporate governance culture of transparency.

Here, and in our discussion in Part III, we have identified concrete variables that should affect any calculus of the private/social costs and benefits of discovery. The efficiency of discovery must therefore be assessed in a Kaldor–Hicks framework, which evaluates whether discovery’s aggregate social benefits surpass the private costs associated with it. 502Supra note 291 and accompanying text.

V. Policy Implications for Comparative Corporate Governance

The account of litigation discovery’s integral role in U.S. corporate governance generates important normative prescriptions. We briefly discuss them here and set up a research agenda for future study.

A. The Enforcement Debate

A substantial literature has emerged on the importance of private enforcement of corporate and securities laws. In particular, the level of private enforcement has been identified as a significant aspect of American exceptionalism. The enforcement debate, however, is neither homogenous nor precise about what is meant by enforcement or how it is measured. Different studies use different concepts and proxies for studying levels of enforcement, including funds expended, 503See, e.g., Jackson & Roe, supra note 13. director and officer liability, 504See supra Part III.A; see also supra note 44. incentives such as fee structures, 505See, e.g., Coffee, supra note 17, at 292. number of filings, the availability of certain causes of action and case outcomes. 506See Armour et al., supra note 13, at 688–89.

Surprisingly, no studies so far have paid attention to the broad role that American fact investigation exerts on private enforcement. The detailed account of the impact of discovery at each stage of the litigation process enables us to complement the analysis that some empirical studies on private enforcement make, and to propose to further disaggregate the variables used in such studies.

In a comparison of private enforcement in the United Kingdom and the United States, Professor John Armour and coauthors concluded that private enforcement is much more intensive in the United States than in the United Kingdom, where private enforcement is virtually nonexistent. 507Id. at 690. But their empirical data lead them to conclude that “private enforcement of corporate law is less central to strong securities markets than might be anticipated.” 508Id. at 687.

Armour and coauthors base their conclusions on both an analysis of available causes of action and empirical research on filings and outcomes of lawsuits. 509Id. at 689. They report that lawsuits against directors of public companies alleging breach of duty are nearly nonexistent in the United Kingdom, where, “absent exceptional circumstances, direct suits are not available, and derivative suits have also been extremely hard to sustain.” 510Id. at 695. Their empirical results show that the United Kingdom has almost no formal private enforcement of corporate substantive law against directors of publicly traded companies. 511Id. at 690. They attribute the absence of formal private enforcement to procedural rules. 512Id. at 721. Like other scholars, they focus on “rules governing class actions, contingency fees, and who pays the winner’s legal expenses.” 513Id.

In contrast, shareholders in the United States are able to commence and sustain litigation (both direct and derivative) in many circumstances where directors may have breached their duties. 514Id. at 695. While directors are much more likely to be sued under U.S. corporate law than their British counterparts, the authors remark that out-of-pocket liability is rare. 515Id. at 722. They present their main results in the following table. 516Id. at 708 tbl. 6.

Table 6

As Armour and coauthors’ Table 6 shows, their empirical study of U.S. private enforcement focuses on cases filed and outcomes. 517See id.; see also id. at 707 (“[O]ur primary concern is with director liability.”). Their analysis shows that “[t]he most common outcome was for the defense to succeed on a motion to dismiss or for summary judgment.” 518Id. at 708. They report that 52% (184 out of 355) of damages cases were resolved on pretrial motions (in favor of defendants). 519Id. The problem with their approach is that they aggregate the results of dismissals before discovery with dispositions of summary judgment, which in most cases occurs after discovery in a case has been completed. In so doing, the authors overlook the substantial threat that discovery represents for individual defendants. Discovery in these cases is not merely expensive; it subjects the actions of the directors and officers, as well as the behavior of all company employees, to a level of scrutiny that is virtually nonexistent in any other country.

The authors do not specify what percentage of the remaining 48% (171) of the cases that were not disposed of on pretrial motion received discovery. 520Id. at 709. Looking at their Table 6, one can infer that all but 48 cases, that is, the 14% that were settled, received some discovery. 521See id. at 708 tbl. 6. But the authors do not disaggregate settlements before and after discovery, which means that this number may include cases that went through discovery. Even if a disproportionate number of the settled cases in Armour and coauthors’ data set are settled prior to discovery, this does not necessarily mean there was no extensive fact investigation into the challenged practices, as our discussion on the special litigation committees showed.

The failure to disaggregate the data and to consider what consequence the procedural posture has for the “intensity of private enforcement” variable to which the study ultimately intended to speak, may have led Armour and coauthors to underestimate the importance of private litigation. Private litigation might significantly discipline management, because of the effectiveness of discovery to render corporate internal practices transparent, thereby influencing settlement rates. As noted in Part III, the cases in which discovery turns up evidence that are likely to lead to a defendant’s liability are also the ones that are most likely to settle. This introduces a selection bias. It is therefore important to distinguish between cases resolved with and without discovery, even if plaintiffs lose on the merits of the claim.

In this and other ways, the empirical literature on private enforcement needs to be revisited in light of our observations.

B. Exporting U.S. Corporate Governance and Securities Laws

The comparative corporate governance debate has been highly consequential by encouraging other nations across Europe, Asia, and Latin America to emulate the U.S. corporate and securities laws. 522See supra notes 2–13 and accompanying text. But the role of the discovery regime in fashioning the laws, institutions, and practices of corporate governance, compared with the significant barriers to private fact investigation in other countries, has broad implications for legal transplants. 523See, e.g., Scott Dodson & James M. Klebba, Global Civil Procedure Trends in the Twenty-First Century, 34 B.C. Int’l & Comp. L. Rev. 1, 1–2 (2011) (“In a number of areas, civil procedure practices in the United States differ significantly from the rest of the world, particularly the civil law world. Notable differences include the use of civil juries, the prevalence of partisan experts paid for directly by the litigants, the existence and extent of party-controlled pre-trial discovery, standards for second instance (or appellate) review, notions of finality in litigation (res judicata), the use of class actions and other forms of aggregate litigation, pleading requirements, and the role and status of the judge. Together, these individual attributes have given rise to a holistic assessment that U.S. civil procedure is highly exceptionalist when compared to the civil law systems in the rest of the world.” (footnotes omitted)).

Civil law systems have no pretrial discovery. 524See Hazard, Discovery and the Role of the Judge, supra note 22, at 1022; Langbein, supra note 23, at 824. In part, this is a function of the very different roles of civil law judges and attorneys, reflecting substantial differences in the design of the litigation process.

The U.S. pretrial period presupposes a distinction between at least two stages of litigation: a concentrated trial at which evidence is presented, in many cases to a jury, and a period before the trial during which the advocates prepare for trial and conduct their fact investigation. By contrast, civil law adjudication “proceeds according to an entirely different logic.” 525Hazard, Discovery and the Role of the Judge, supra note 22, at 1021. It follows an inquisitorial model, in which the judge drives the entire process, and there is no distinction between trial and pretrial proceedings. 526Id. at 1022; see also Langbein, supra note 23, at 826. Adjudication is akin to a series of business meetings at which the judge investigates and considers only the evidence relevant to the one or two issues that she believes might decide the case. The judge only seeks additional evidence if the case cannot be disposed of on the issues presently before the court. 527See Langbein, supra note 23, at 830. It is the judge who explores, sifts through, and controls the investigation and presentation of evidence, all of which must take place during a hearing. 528Hazard, Discovery and the Role of the Judge, supra note 22, at 1022; Langbein, supra note 23, at 829 (“The process merges the investigatory function of our pretrial discovery and the evidence-presenting function of our trial.”). The judge decides what evidence parties may present at what time, how the evidence is to be interpreted, and whether any issue warrants further investigation. 529See Langbein, supra note 23, at 831.

Solely for practical reasons, this concentration of fact-finding in the hands of the judge makes it impossible to replicate U.S.-style discovery. A civil law judge (and his handful of clerks) cannot equal the work of teams of privately financed U.S. attorneys, deposing dozens of witnesses and reviewing millions of documents in a complex corporate case. Moreover, they do not have the same pecuniary incentives to engage in the type of fact exploration that is routine in U.S. complex litigation. 530Huang, supra note 24, at 40 (“One important cause of incomplete information for fact-finding [in civil law systems] is the misalignment of authority, burden, and incentives.”). Commentators have responded to the incentives argument by pointing to the quality of a country’s judiciary. See, e.g., Langbein, supra note 23, at 848. But Langbein explicitly excluded “Big Case[s]” from his analysis, and Germany’s system may be hard to replicate. Id. at 858.

But more fundamental civil law principles of due process and litigants’ rights preclude U.S.-style discovery independently. Judges lack the authority and the tools to conduct U.S.-style fact investigation.

Civil law jurisdictions have traditionally exempted party defendants from producing information that can be used against them. This right against “self-incrimination” in civil cases is a due process principle (the nemo tenetur edere contra se principle) deeply embedded in the civil law legal culture. 531See Verkerk supra note 23, at 16. Although procedural codes in civil law systems have rejected the nemo tenetur principle in its rigid nineteenth-century form, and civil codes now generally allow parties to seek some evidence from their opponents—albeit always via a specific request to the judge 532See id. at 66–73. —the basic privilege of a party not to disclose evidence against its own interest or to open itself up to searching discovery, remains in place in Germany 533Id. at 68. and other countries, 534See, e.g., Lei No. 5.869, de 11 de Janeiro de 1973, Código de Processo Civil [C.P.C.]: art. 363 de 11.1.1973 (Braz.). and its influence still significantly limits the scope of judicial fact investigation in other jurisdictions. Thus, direct party-on-party discovery is not available in any civil law country, parties cannot be called as witnesses in many civil law countries, and courts are constrained in what information a court can demand from parties. 535See, e.g., id.; see also Huang, supra note 24, at 40.

In many civil law systems, a plaintiff is required to identify each document she seeks from a party-opponent with specificity in the complaint at the beginning of the litigation and justify on what basis she is entitled to it. 536“Relevance” has only recently appeared as a factor, and some “legal right” to the document that can override the privacy and property interests of the owner must be shown. Compare the discarded doctrine of Boyd v. United States, 116 U.S. 616, 622 (1886) (holding that “a search and seizure [was] equivalent [to] a compulsory production of a man’s private papers” and that the search was “an ‘unreasonable search and seizure’ within the meaning of the Fourth Amendment”). The German Gerichstordnung, for example, requires the requesting party to “identify the document requested with reasonable specificity, describe its relevance to some fact in issue, and set forth the basis for the belief that it is in the possession or control of the opponent.” 537Murray & Stürner, supra note 25, at 277 (translating Zivilprozessordnung [ZPO] §§ 421 ff., 427). Under the Italian-influenced Brazilian Code, the party must specify with particularity “the document or the thing” that is being sought, explain how “the document” is relevant to the demands of proof in the case, and state the circumstances based on which the party believes that the requested “document” exists. 538Lei No. 5.869, de 11 de Janeiro de 1973, Código de Processo Civil [C.P.C.]: art. 356 de 11.1.1973 (Braz.) (emphasis added). The emphasis on “document” in the singular in Brazil’s code reflects expectations about the scope of discovery. Civil law attorneys look at a standard U.S.-style discovery request that includes words like “all” or “any” and find them not merely curious, but laughable. 539Conversation with a government attorney in the Office of General Counsel for the Mayor of the City of São Paulo, in São Paulo, Braz. (Jan. 20, 2012).

In the following section, we consider two problems that arise from the lack of pretrial discovery in civil law countries in relation to transplants of U.S.-style regulatory disclosure and litigation procedures. As we discuss, these transplants will not be effective unless a transplant of the underlying discovery regime that supports them is also implemented.

1. How Well Does Disclosure Work Without Litigation Discovery?

In his article, The Legal and Institutional Preconditions for Strong Securities Markets, Professor Bernard Black identifies information asymmetry between issuers and investors as the key problem that countries face in reproducing the kinds of institutional networks that support strong securities markets in the United States. 540Bernard S. Black, The Legal and Institutional Preconditions for Strong Securities Markets, 48 UCLA L. Rev. 781, 786 (2001). In the quest to develop their financial markets, many countries have, accordingly, looked to the U.S. mandatory disclosure regime and its implementation by the SEC as a model for addressing the information asymmetry problem. 541See, e.g., Theodor Baums, Changing Patterns of Corporate Disclosure in Continental Europe: The Example of Germany (European Corporate Governance Inst., Working Paper No. 04, 2002), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=345020. But, as Black emphasizes, “[d]elivering information to investors is easy, but delivering credible information is hard. Insiders have an incentive to exaggerate the issuer’s performance and prospects, and investors can’t directly verify the information that the issuer provides.” 542Black, supra note 540, at 786. In fact, as we have shown, they can, but ex post through litigation discovery, an institution that Black does not mention specifically.

Brazil serves as an example where the credibility of disclosure has suffered from lack of verification. While Brazil’s Securities Exchange Commission has disclosure requirements for related-party transactions, executive remuneration, and shareholders’ agreements, 543See Érica Gorga, Changing the Paradigm of Stock Ownership from Concentrated Towards Dispersed Ownership? Evidence from Brazil and Consequences for Emerging Countries, 29 Nw. J. Int’l L. & Bus. 439, 474 (2009) (noting disclosure inconsistencies regarding the number of shares bound by shareholders agreements, and the validity of these agreements on the Brazilian Securities Exchange Commission (CVM) website). evidence suggests that the information companies actually disclose is incomplete and does not reflect reality. 544See id. The simple explanation for this is that there is no effective mechanism for verifying the accuracy of disclosures ex post.

Lacking litigation discovery, private parties in Brazil have restricted means of verifying company disclosures. 545For evidence that litigation constrains managers from making opportunistic reporting choices, see Justin Hopkins, Private Enforcement of Securities Laws and Financial Reporting Choices 5 (2012) (unpublished manuscript), available at http://ssrn.com/abstract=1872068. Because Brazil lacks adversarial discovery, the kind of detailed information about such transactions necessary to successfully prosecute such cases is never revealed, even though shareholders are entitled to mount legal challenges to related-party transactions in the courts. The absence of the kind of “discovery culture” that informs U.S. private and regulatory enforcement means that neither regulators nor private interests even consider what type of disclosure or information might shed light on the legality of such transactions, or how such information might be obtained. 546See supra notes 524–30 and accompanying text. Plaintiffs’ attorneys neither have the tools, nor do they scrutinize such transactions. 547See supra notes 524–30 and accompanying text. Similarly, regulators are limited by civil enforcement rules and resource limitations. 548See supra notes 524–30 and accompanying text. Lacking the ability for ex post discovery, the value of ex ante disclosure in Brazil is therefore compromised.

More generally, it follows that implementation of standard substantive corporate and securities laws may fail to produce expected corporate governance improvement in civil law jurisdictions, not merely because of the lack of private enforcement mechanisms, such as class actions, but because of all the other consequences that follow from the lack of litigation discovery. 549See J. Maria Glover, The Structural Role of Private Enforcement Mechanisms in Public Law, 53 Wm. & Mary L. Rev. 1137, 1140 (2012) (arguing for the centrality of private enforcement to the American regulatory system, but failing to consider issues of information production or litigation discovery).

As a strategy of developing markets, introducing regulatory disclosures was an important part of eliminating information asymmetries. But to make regulatory disclosure more effective, enforcement mechanisms must be made available to obtain additional information ex post to assess the quality of disclosure. In some jurisdictions like Brazil, it is unlikely that regulators will have the resources to investigate material misrepresentations and omissions in issuer disclosures. Changes in the Brazilian code of civil procedure that would grant plaintiff shareholders greater authority to investigate probable corporate wrongdoing—even if only by means of a special shareholder information request proceeding that allows for more probing, discovery-like investigations—would likely be much more fruitful for policing the accuracy of securities disclosures. Large Brazilian issuers who cross-list their securities on U.S. markets are already subject to ex post scrutiny by litigation discovery and discovery-related mechanisms. 550See Gorga, supra note 543, at 486–87. In this way, U.S. discovery practices also serve to discipline a limited number of U.S.-listed foreign issuers.

2. Exporting Aggregate or Representative Litigation Mechanisms to Europe

In the private enforcement literature and in separate debates on securities class actions, 551See, e.g., Coffee, supra note 448 (proposing a variety of ways in which securities class action penalties would fall less on the corporation and more on those who are truly culpable); Samuel Issacharoff & Geoffrey P. Miller, Will Aggregate Litigation Come to Europe?, 62 Vand. L. Rev. 179 (2009) (discussing various examples of securities class actions). derivative actions, 552See, e.g., Siems, supra note 13 (exploring the availability of derivative actions across countries). and civil procedure reform, 553See, e.g., Christopher Hodges, The Reform of Class and Representative Actions in European Legal Systems: A New Framework for Collective Redress in Europe (2008); Richard A. Nagareda, Aggregate Litigation Across the Atlantic and the Future of American Exceptionalism, 62 Vand. L. Rev. 1, 19–37 (2009) (providing comparative analysis of recent developments in Europe regarding aggregate litigation). scholars have discussed the merits of adopting U.S.-style aggregate litigation mechanisms in Europe and elsewhere. 554See, e.g., The Law and Economics of Class Actions in Europe: Lessons from America (Jürgen G. Backhaus, Alberto Cassone & Giovanni B. Ramello eds., 2012); Issacharoff & Miller, supra note 551, at 179 (“Once decried as the perversity of rapacious Americans, class actions are now the focus of significant reform efforts in many European countries and even at the level of the European Union.”). These literatures typically consider how far Europe might go in emulating the “unique” system of U.S. “entrepreneurial litigation.” 555See Coffee, supra note 17, at 291–93. While they consider standing requirements, rules on attorneys’ fees, the dynamics and agency costs of attorney-driven aggregate litigation, and the costs of such lawsuits, no one has focused on the obvious question: what mechanisms are available to private litigants in civil court to investigate the facts and circumstances of corporate internal wrongdoing?

Several European countries, including Denmark (2007), England and Wales (1999), Finland (2007), France (1990s), Germany (2005), Italy (2007), the Netherlands (2005), Norway (2008), and Sweden (2002), have adopted some aggregate litigation mechanisms. 556See Nagareda, supra note 553, at 21–25. But it is unclear to what extent these mechanisms have been successfully employed to prosecute management misconduct.

The German Capital Markets Test Case Act (KapMus), passed in 2004, presents just one example of how the lack of discovery thwarts private enforcement even where certain aggregate mechanisms are introduced. 557Kapitalanleger-Musterverfahrensgesetz [KapMuG] [Capital Market Investors’ Model Proceeding Act], Aug. 16, 2005, Bundesgesetzblatt, Teil I [BGBl. I] 2437, as amended, Gesetz [G], Oct. 19, 2012, BGBl I 2182 (Ger.), translated at http://www.bmj.bund.de/kapmug.

The Deutsche Telekom litigation was the test case for this new “aggregate litigation” mechanism in Germany. 558Nils-Viktor Sorge, Historic Lawsuit: Angry Shareholders Take on Deutsche Telekom in Court Showdown, Spiegel Online Int’l (Apr. 7, 2008, 2:44 PM), http://www.spiegel.de/international/business/historic-lawsuit-angry-shareholders-take-on-deutsche-telekom-in-court-showdown-a-545827.html. The KapMus has been called “Lex Telekom,” because it was passed to provide an enforcement mechanism for outraged German investors. See Axel Halfmeier, Recht und Kapitalmarkt: Von der Lex Telekom zum Instrument des Rechtsschutzes, Börsen-Zeitung, Nov. 8, 2010, available at https://www.boersen-zeitung.de/index.php?li=1&artid=2010152040. Around 17,000 shareholders had sued the German telephone and communications giant, Deutsche Telekom (DT), for misstatements in its May 2000 offering prospectus concerning its ill-fated July 2000 acquisition of the U.S. company VoiceStream. 559Sorge, supra note 558.

In Deutsche Telekom, the German judges investigated and considered the same questions at issue in the parallel securities class action, filed against DT in 2002 in the U.S. District Court for the Southern District of New York. 560See In re Deutsche Telekom AG Sec. Litig., 229 F. Supp. 2d 277, 278 (S.D.N.Y. 2002). After two weeks of oral proceedings, the German judge at the Oberlandesgericht in Frankfurt credited DT CEO Ron Sommer’s testimony that preliminary, confidential negotiations for the acquisition of VoiceStream (now T-Mobile) in April of 2000 were interrupted, and that actual plans for the acquisition only took shape after the May 2000 offering prospectus was issued and after the June 16, 2000, filing date. 561See OLG Frankfurt a.M.: Musterentscheid im KapMuG-Verfahren gegen die Deutsche Telekom, Ebner Stolz (May 16, 2012), http://www.ebnerstolz.de/de/olg-frankfurt-a-m-musterentscheid-im-kapmug-verfahren-gegen-die-deutsche-telekom-15136.html (“Nach den Angaben der Zeugen, zu denen auch die ehemaligen Vorstandsvorsitzenden der Telekom Dr. Ron Sommer und Kai-Uwe Ricke gehörten, war das Geschäft erst Ende Juli 2000 und somit deutlich nach der Erstnotiz am 19.6.2000 abschließend und entscheidungsreif verhandelt gewesen.”). In the settlement agreement of the U.S. securities class action, plaintiffs’ claim that “the Prospectus was materially false and misleading because despite the fact . . . that Deutsche Telekom was engaged in serious and advanced negotiations to acquire VoiceStream, the Prospectus merely disclosed that Deutsche Telekom was ‘actively considering and discussing a number of potential acquisition transactions.’” Notice of Proposed Settlement of Class Action, Motion for Attorneys’ Fees and Settlement Fairness Hearing at 4, In re Deutsche Telekom AG Sec. Litig., No. 00-CV-9475 (S.D.N.Y. 2002), available at http://securities.stanford.edu/filings-documents/1016/USD00/200544_r02n_00CV9475.pdf. The court also held that the prospectus contained no material misstatements regarding DT’s real estate valuations. 562OLG Frankfurt a.M.: Musterentscheid im KapMuG-Verfahren gegen die Deutsche Telekom, supra note 561.

But the lack of adequate discovery made it impossible for the plaintiffs to question management’s dubious account of the timeline of events leading up to VoiceStream’s acquisition. Andreas Tilp, the German plaintiffs’ attorney, told the Economist that “[c]ompared with America we are at a great disadvantage.” 563Bad Connection, Economist, Apr. 12–18, 2008, at 72, 72 (internal quotation marks omitted), available at http://www.economist.com/node/11021139. The Economist reported that, “[m]ost aggravating for Mr Tilp is his inability to secure documents, such as a Bonn prosecutor’s report that he believes concludes there was balance-sheet fraud, and another report from the Federal Audit Court, which was pivotal in the American settlement.” 564 Id.

Especially noteworthy for the thesis that vigorous discovery is critical for investigating corporate internal wrongdoing is that the German Court and the plaintiffs’ attorneys relied, at least in part, on discovery obtained in the U.S. securities class action. The German Court required DT to produce transcripts of depositions taken during discovery in the U.S. securities class action, as well as some e-mails and other documents produced in New York. 565Markus Zydra, Wende im Telekom-Prozess, Suddeutsche.de (May 17, 2010, 9:39 PM), http://www.sueddeutsche.de/wirtschaft/gericht-fordert-us-unterlagen-wende-im-telekom-prozess-1.574782. The plaintiffs knew to ask for the e-mails and documents only because they had become aware of them as a result of following the New York proceedings and contacted the U.S. attorneys involved. Initially, DT heavily redacted the U.S. deposition transcripts, blacking out over 50% of the testimony given in the U.S. depositions. But knowing that there was more, and that the transcripts were in the possession of the defendants’ attorneys, the German court then wanted to see it all. 566Id.

Recognizing the significant advantage of conducting their fact investigations under the U.S. discovery regime, attorneys for the German investors had already applied in January 2003 to the presiding judge in the New York case, U.S. District Judge Sidney Stein, for permission to obtain discovery pursuant to 28 U.S.C. § 1782, which provides discovery assistance for foreign proceedings. 567In re Application of Schmitz, 259 F. Supp. 2d 294 (2003), aff’d sub nom Schmitz v. Bernstein Liebhard & Lifshitz, LLP, 376 F.3d 79 (2d Cir. 2004). The Court denied the application, based on the vigorous objections by German prosecutors, the German Federal Justice Department, and German civil procedure scholars, and a meek clarification by the court presiding over the DT case in Frankfurt, that its “willingness to consider such documents ‘was indeed no declaration that the Court supported—in opposition to other authorities of the Federal Republic of Germany—the production of such documents.’” 568Schmitz, 376 F.3d at 82.

The Deutsche Telekom case thus proceeded on the basis of limited information. While the New York case settled in 2005 for $120 million after discovery had been completed, the German court found in favor of the defendants. 569See Stipulation and Agreement of Settlement at 11, In re Deutsche Telekom AG Sec. Litig., 229 F. Supp. 2d 277 (S.D.N.Y. 2002) (No. 00-CV-9475 (SHS)), available at http://securities.stanford.edu/filings-documents/1016/USD00/2005128_r04s_00CV9475.pdf. The parties settled for $120 million without DT admitting any wrongdoing. Id. at 4. While other factors may have played a role in explaining this outcome, the lack of effective mechanisms for fact-finding in the case is striking.

Other cases that have been pursued on both sides of the Atlantic, like the case against the Italian agri-business Parmalat, confirm the German experience. 570See, e.g., Guido Ferrarini & Paolo Giudici, Financial Scandals and the Role of Private Enforcement: The Parmalat Case, in After Enron: Improving Corporate Law and Modernising Securities Regulation in Europe and the US 159, 201 (John Armour & Joseph A. McCahery eds., 2006) (“Given the lack of efficient discovery rules, investor action against mass wrongdoing is virtually impossible in Italy as it is in the rest of Europe, unless information is gathered by public authorities.”); Paolo Giudici, Representative Litigation in Italian Capital Markets: Italian Derivative Suits and (If Ever) Securities Class Actions, 6 Eur. Company & Fin. L. Rev. 246, 254 (2009) (“Italian law does not grant any inspection right to shareholders of public companies.”).

Only two cases have been decided under KapMus since it was passed in 2004: the DaimlerChrysler case and the LBB Fonds 13 case. 571Axel Halfmeier et al., Evaluation des Kapitalanleger-Musterverfahrensgesetzes: Forschungsvorhaben im Auftrag des Bundesministeriums der Justiz 50 (2009), available at http://www.bmj.de/SharedDocs/Downloads/DE/pdfs/Abschlussbericht_KapMuG_Frankfurt%20School_2009.pdf?__blob=publicationFile; see also Daniela Kuhr, Sammelklagen Floppen in Deutschland, Suddeutsche.de (Apr. 24, 2012, 10:07 AM), http://www.sueddeutsche.de/wirtschaft/telekom-prozess-und-seine-folgen-sammelklagen-floppen-in-deutschland-1.1340101. The respective courts found in favor of the defendants in DaimlerChrysler, but identified some material misstatements in the LLB Fonds 13 prospectus. 572Halfmeier et al., supra note 571, at 50. Significantly, a 2009 study, commissioned by the German Justice Department, assessing the impact of the KapMus noted that in both cases, the only evidence taken by the courts was witness testimony. 573Id. at 3. Based on the limited success plaintiff investors have had bringing KapMus cases, observers have concluded that the law has been “a flop.” 574See Kuhr, supra note 571.

As the example of the German securities “class action” shows, the lack of discovery is a serious impediment to the effective implementation of aggregate-style litigation. Neither securities class actions nor shareholder derivative actions are likely to produce significant disciplining effects in systems that do not also implement appropriate discovery rules.

We thus caution against transplanting aggregate litigation mechanisms without a careful analysis of the discovery rules that exist in foreign jurisdictions. On the other hand, implementing discovery in a civil law regime is also not an easy task. Japan provides one prominent example of this attempt. After an extensive academic debate, Japan’s new Code of Civil Procedure of 1996 implemented significant but modest changes to Japan’s fact-finding regime. 575See Huang, supra note 24, at 157. These changes in procedure took years to negotiate and involved extensive academic and public debate. 576See id. Nonetheless, they remain very limited. 577See id. at 157–216 (describing the Japanese discovery regime).

C. Reforming Discovery in the United States

Our account of the benefits of discovery in Part III has a direct bearing on the persistent and often successful efforts to cut back on discovery under the Federal Rules of Civil Procedure. In 2013, the Federal Judicial Conference proposed amendments to the Federal Rules that promise to introduce yet another round of significant limitations on discovery. 578Comm. on Rules of Practice and Procedure of the Judicial Conference of the U.S., Preliminary Draft of the Proposed Amendments to the Federal Rules of Bankruptcy and Civil Procedure (Aug. 15, 2013) [hereinafter Proposed Rule Changes], available at http://www.uscourts.gov/uscourts/rules/preliminary-draft-proposed-amendments.pdf. The extensive public debate and commentary reflects a widespread understanding that “the proposed amendments have the potential to radically impact the legal system.” 579See Adam Cohen, Commenting on the Comments to the Proposed Amendments to the Federal Rules of Civil Procedure, Inside Couns. (Feb. 18, 2014), http://www.insidecounsel.com/2014/02/18/commenting-on-the-comments-to-the-proposed-amendme. The most controversial changes include a new emphasis on “proportionality” in defining the scope of discovery 580Proposed Rule Changes, supra note 578, at 264–69. and a new sanctions standard for e-discovery violations. 581 Id. at 270–75. The proposed amendments also lower the presumptive limits on depositions and interrogatories and, for the first time, introduce limits on the number of requests for admissions.

The proposed amendments to Rule 26(b) would limit the scope of discovery to discovery that is “proportional to the needs of the case, considering the amount in controversy, the importance of the issues at stake in the action, the parties’ resources, the importance of the discovery in resolving the issues, and whether the burden or expense of the proposed discovery outweighs its likely benefit.” 582Id. at 289. The change would allow respondents to withhold discovery materials based on an objection of proportionality, where previously objecting parties had to seek an order of protection from the judge. Judge Shira Scheindlin, a leader in e-discovery, warned that the new rule may be interpreted to shift the burden of proving proportionality to the requesting party. 583Comment Letter on Proposed Rules from Shira A. Scheindlin, U.S. Dist. Judge, to Comm. on Rules of Practice and Procedure 3 (Jan. 13, 2014), available at http://blogs.reuters.com/alison-frankel/files/2014/02/ruleschanges-scheindlin.pdf. She argued that the rules unnecessarily cut back on the scope of discovery and undermine the sanctions available for punishing parties that destroy electronic documents. 584Id. at 3–5.

It is clear from the comments that all sides agree that these changes to the discovery rules will primarily affect large, complex cases. The U.S. Chamber of Commerce and the corporate defense bar favor the new proposed rules because they are expected to curb the high cost of e-discovery. 585See Alison Frankel, Two Judges Question Proposed Discovery Limits in Federal Rules, Reuters (Feb. 5, 2014), http://blogs.reuters.com/alison-frankel/2014/02/05/two-judges-question-proposed-discovery-limits-in-federal-rules/. On the other side, public interest groups, plaintiffs’ attorneys, as well as leading civil procedure scholars and judges, are concerned that, together with the presumptive limits on discovery, these changes to the scope of discovery will unfairly disadvantage plaintiffs suing large organizations, including corporations and governments. 586Id.; see also Comment Letter from Shira A. Scheindlin, supra note 583, at 6 (voicing concern that proposed rules unnecessarily cut back on the scope of discovery). This debate is thus no different from previous debates about the costs of discovery we have discussed in Part IV.D, and no new evidence is adduced to support the conclusion that discovery costs are excessive. 587The claims by the defense bar also fail to take account of the likelihood that e-discovery costs will begin to drop sharply, as document review by machine intelligence is gaining acceptance with the courts and document retention is becoming cheaper as a result of improved technologies.

The proposed changes raise red flags, given our account of discovery for the development of the U.S. corporate governance system. The changes provide negative incentives for corporate defendants. Under the new proposed rules, corporations may have incentives to destroy documents and weaken internal governance controls. Conflicts about proportionality are likely to lead to increased involvement by the judge at the outset of discovery, thus diminishing the decentralization of fact investigation characteristic of the U.S. system.

In spite of Article III’s “case or controversy” provision, it has long been acknowledged that representative litigation, such as shareholder suits, involve courts in shaping policy and tailoring remedies for future compliance. 588See, e.g., Abram Chayes, The Role of the Judge in Public Law Litigation, 89 Harv. L. Rev. 1281, 1281 (1976). The SLC process recognizes this by requiring that SLCs pursue a thorough investigation of misconduct allegations. Corporate internal investigations are not limited by proportionality to the particular damage caused by a particular breach of fiduciary duty.

Moreover, measuring the “proportionality” of discovery narrowly by reference to the value of plaintiff shareholders’ claims against a corporate defendant—as the proposed new rules appear to emphasize—underestimates the value of discovery. It fails to appreciate the value of exposing inadequate compliance or accounting controls for the company over time; the positive information externalities that discovery generates for markets, regulators, and gatekeepers; the disciplining effects of the process of discovery; and, last but not least, the threat to management that the details of their daily business and management decisions may be subject to the kind of intense scrutiny that we have described in Part I.

While there may be good reasons for allowing judges discretion to limit discovery, it is hard to see why the scope of discovery should increasingly be limited presumptively by the rules. This Article questions the retreat from discovery that has occurred in shareholder litigation since the mid-1990s. While we leave proposals regarding discovery reforms for another opportunity, we note that discovery’s costs might be reduced without giving up on its benefits for corporate governance. For example, one can envision cost-sharing rules among plaintiffs for allowing discovery conducted in one case to be used in another case. It may also be worth considering whether different discovery rules should apply to different types of claims. 589See, e.g., David Marcus, The Past, Present, and Future of Trans-substantivity in Federal Civil Procedure, 59 DePaul L. Rev. 371 (2010) (discussing whether discovery rules should be trans-substantive).

Conclusion

The literature has largely ignored the role of fact investigation and discovery for corporate governance. But litigation discovery is the “elephant in the boardroom.” The U.S. information-pushing regime implemented by modern discovery rules is unique in the authority, scope, and tools of fact investigation that it affords private litigants. As we have shown, the institutional consequences that flow from the rules and practices of discovery are complex and multifaceted.

Discovery generates information and disciplines management at every stage of shareholder litigation (even prior to the motion to dismiss). Publication at trial, or at evidentiary hearings, of information obtained through offensive discovery exposes corporate internal practices, controls, judgments, and failures—regardless as to whether liability is ultimately imposed on defendants. This informs judges, regulators, and the public about the modus operandi of corporate governance failures.

Even in cases that are settled, the benefits of litigation discovery are considerable. In cases resolved on summary judgment or settled after a summary judgment motion is filed, the facts obtained through discovery are nonetheless publicly available, as the parties must file admissible evidence in support of their claims and defenses at the summary judgment stage. 590Of course, not all issues must be presented at summary judgment. But this does not vitiate the argument. Similar positive externalities thus accrue as in the case of evidence presented at trial. Even if parties settle prior to summary judgment, litigation discovery often informs judges, regulators, and the public about corporate internal practices, because judges must approve a settlement and the parties must provide evidence that the settlement is just. While such evidence is sometimes obtained through discovery after the settlement has been negotiated, this does not vitiate the usefulness of the information publicized at settlement hearings. 591Moreover, judges have not always approved settlements based on the evidence provided and have sometimes demanded more information. See, for example, Judge Rakoff’s controversial decision to refuse approval of the settlement between the SEC and Bank of America in the Merrill Lynch case. Supra note 259 and accompanying text.

More importantly perhaps, defensive discovery plays a critical role in rendering corporate practices transparent, disciplining management, and encouraging the development and adoption of best practices. Discovery can have an important corporate governance function even in cases dismissed or settled prior to the motion to dismiss. Its practices have established templates for independent corporate internal investigations.

Discovery has thus induced incremental improvements in corporate governance practices and controls. It has influenced and shaped the development of case law and corporate and securities statutory reforms. The information revealed through discovery serves regulators and market participants. Discovery also serves as an ex post form of particularized disclosure that enforces and complements the standardized ex ante disclosures required by the SEC. All these effects combined promote the overall U.S. corporate governance culture of transparency—which has also affected corporate governance practices globally.

This Article sets forth policy proposals and implications of these insights for the empirical literature on shareholder actions, and it begins to outline a research program for addressing such complex issues as corporate law and civil procedure reforms. The Article has important implications for the legal transplant of U.S.-style securities disclosure, aggregate litigation mechanisms, and other aspects of enforcement, that fail to consider appropriate tools for investigating corporate internal wrongdoing ex post. And it has implications for civil procedure reform in the United States. Our account of the role and effects of discovery in shareholder litigation suggests that further research is necessary on how the rules of civil procedure could be improved to better tackle problems of corporate litigation and governance.

Appendix

Information Request, Saito v. McKesson HBOC, Inc. 592Complaint Pursuant to 8 Del. C. § 220 for Access to Corporate Books and Records at 2–5, Saito v. McKesson HBOC, Inc., No. 18553-NC (Del. Ch. Dec. 14, 2000), 2000 WL 34549543.

A. All sales agreements and documents related to sales agreements (including documents, which have been referred to as “side letter,” [sic] containing terms and conditions of any agreements not included within the agreement itself) between HBOC or the Company and [33 specified] entities[.]

B. All documents evidencing communications between the following persons and entities or representatives thereof regarding: (i) “side letters”; (ii) backdating of sales contracts; (iii) sales contingent upon terms not contained within the main sales contract; or (iv) accounting practices at HBOC, McKesson, McKesson HBOC or HBOC as the post-merger subsidiary of the Company (“HBOC sub”):

1. HBOC or HBOC sub and any entities identified in Request A(l)-(33) above;

2. Arthur Andersen and any entities identified in Request A(l)-(33) above;

3. Bear Stearns and any entities identified in Request A(1)-(33)

4. Arthur Andersen and Bear Stearns;

5. Arthur Andersen and Deloitte & Touche;

6. Bear Stearns and HBOC or HBOC sub;

7. Arthur Andersen and HBOC or HBOC sub; and

8. Deloitte & Touche and HBOC or HBOC sub.

C. All documents concerning Arthur Andersen’s pre-merger review of the internal controls of HBOC.

D. All documents concerning verification by Arthur Andersen of any representations about revenues and/or expenses made by HBOC management during the due diligence process in preparation for or in connection with HBOC’s merger with McKesson.

E. All documents relating to or reflecting the verification by McKesson, its employees, directors or advisors that the revenues and/or expenses reported in the financial statements of HBOC that were incorporated by reference in the joint proxy statement, issued in connection with the merger between McKesson and HBOC, were properly and accurately reported.

F. All documents reflecting discussions among or communications with any members of the Board of Directors of HBOC, McKesson, or McKesson HBOC concerning (i) reports dated April 14, 1997 and August 19, 1998 by the Center for Financial Research and Analysis (CFRA); (ii) any other published public analysis of the accounting practices of HBOC prior to and following the HBOC’s merger with McKesson; (iii) any public response to such published reports by any employee of HBOC; or (iv) any HBOC, McKesson or McKesson HBOC shareholder reaction to such published reports.

G. All documents relating to or reflecting communications among or between members of HBOC management and/or HBOC’s Board of Directors concerning HBOC’s purpose in pursuing a merger with McKesson or any other entity between January 1, 1997 and October 17, 1998.

H. All documents concerning the decision to change the structure of the merger of HBOC and McKesson from a “merger of equals” to an acquisition of HBOC by McKesson.

I. All documents, including board minutes or other documents evidencing any communications with or among HBOC’s or McKesson HBOC’s Board of Directors related to the termination or resignation of the following individuals: Jay Gilbertson, Albert Bergonzi, Charles W. McCall, Mark A. Pulido, David Held, Jay Lapine, Michael Smeraski and Dominick DeRosa.

J. All documents concerning SEC, Department of Justice or any other governmental agency investigation of any individual listed in Request No. 9 [sic] above, or any other employee of HBOC, McKesson HBOC or HBO sub between January 1, 1997 and the present.

K. All documents reflecting discussions among or communications with McKesson HBOC’s management or members of its Board of Directors about whether to initiate litigation against any past or present employee of HBOC, McKesson HBOC or HBOC sub or against any advisor to or auditor of any of the foregoing entities in connection with the restatements of HBOC’s and/or the Company’s financial results for fiscal year 1997, 1998 and 1999.

Footnotes

Érica Gorga, Ph.D., is Associate Research Scholar in Law and the John R. Raben/Sullivan & Cromwell Executive Director for the Yale Law School Center for the Study of Corporate Law, Professor of Law at Fundação Getulio Vargas School of Law at São Paulo (FGV Direito SP), erica.gorga@fgv.br.Michael Halberstam, J.D., Ph.D., is Associate Professor S.U.N.Y. Law School University at Buffalo, mhalbers@buffalo.edu.The authors wish to thank Guyora Binder, Margaret Blair, Michael Boucai, Kate Brown, Jens Dammann, Matthew Dimick, Brian Fitzpatrick, Merritt Fox, Mira Ganor, Jeffrey Gordon, Zohar Goshen, Peter Gruber, Bert Huang, Robert Jackson Jr., Dan Kramer, Susan Mangold, Curtis Milhaupt, Donna Nagy, Katharina Pistor, Jayanth K. Krishnan, Morgan Ricks, Roberta Romano, Judge Ruth Ronen, Amanda Rose, Jack Schlegel, Jeffrey Schoenblum, Suzanna Sherry, Michael Simkovic, Matt Steilen, Rick Su, Randall Thomas, Kenju Watanabe, Bert Westbrook, James Wootin, and the participants of the 23rd Annual Meeting of the American Law and Economics Association (ALEA) at Vanderbilt University, the 29th Annual Conference of the European Association of Law and Economics (EALE) at Stockholm University, the 16th Annual Conference of the International Society for New Institutional Economics (ISNIE) at University of Southern California Gould School of Law, the 2012 Meetings of the Canadian Law and Economics Association, the Columbia Law School Fundacao Getulio Vargas Global Alliance Workshop held in 2012 at Columbia Law School, the Center for Law and Global Affairs Workshop on International and Transnational Law held at Sandra Day O’Connor College of Law, Arizona State University, Faculty Workshops held at Vanderbilt Law School and SUNY Buffalo Law School, for their insightful thoughts and comments on this project. Michael Halberstam is grateful to Columbia Law School, its Law and Economic Studies Program, and especially to Jeffrey Gordon and Dean Shizer for their generous support of this research. Érica Gorga is especially grateful to Roberta Romano for critiques and suggestions and for the Yale Law School Center for the Study of Corporate Law for research support during the finalization of this project. Finally, we thank Joel Langdon and William Bradbury, at the Emory Law Journal, for all the work they put into editing this Article.

1Geoffrey C. Hazard, Jr., From Whom No Secrets Are Hid, 76 Tex. L. Rev. 1665, 1665 (1998) (footnotes omitted) (quoting Hickman v. Taylor, 329 U.S. 495, 507 (1947)).

2Roberta Romano, The Genius of American Corporate Law 14–15, 118 (1993). The dominant view was, and in many areas of law still is, that regulatory competition among jurisdictions creates a “race to the bottom” in regulatory standards. The “race to the bottom” thesis in corporate law was advanced by Adolph A. Berle, Jr. and Gardiner C. Means in The Modern Corporation and Private Property (1932), and was adopted by Justice Louis Brandeis in Louis K. Liggett Co. v. Lee, 288 U.S. 517, 557–60 (1933) (Brandeis, J., dissenting).

3Romano, supra note 2, at 1–2.

4Id.

5See Curtis J. Milhaupt & Katharina Pistor, Law and Capitalism: What Corporate Crises Reveal About Legal Systems and Economic Development Around the World 17–20 (2008); Bernard Black et al., Corporate Governance in Korea at the Millennium: Enhancing International Competitiveness, 26 J. Corp. L. 537, 539–40 (2001) (noting Asian recognition of the contribution of corporate law to economic prosperity); 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, 1234 (2002) (arguing that “the core beliefs and practices that have underpinned the academic analysis of corporate law and governance . . . amount to an interlocking set of institutions that constitute ‘shareholder capitalism,’ American-style that [U.S. academics] have been aggressively promoting throughout the world”).

6See Rafael La Porta et al., Corporate Ownership Around the World, 54 J. Fin. 471, 471–73, 511 (1999).

7See Rafael La Porta et al., Investor Protection and Corporate Governance, 58 J. Fin. Econ. 3, 4 (2000) (“[T]he protection of shareholders and creditors by the legal system is central to understanding the patterns of corporate finance in different countries.”).

8See John C. Coffee, Jr., The Rise of Dispersed Ownership: The Roles of Law and the State in the Separation of Ownership and Control, 111 Yale L.J. 1, 8–9, 76 (2001).

9See Rafael La Porta et al., What Works in Securities Laws?, 61 J. Fin. 1, 2, 27–28 (2006).

10See John C. Coffee, Jr., Do Norms Matter? A Cross-Country Evaluation, 149 U. Pa. L. Rev. 2151, 2151–52, 2171–75 (2001); Mark J. Roe, Can Culture Constrain the Economic Model of Corporate Law?, 69 U. Chi. L. Rev. 1251, 1262–64 (2002).

11See Mark J. Roe, Political Determinants of Corporate Governance: Political Context, Corporate Impact 1–5, 49–51, 201–04 (2003); Mark J. Roe, Political Preconditions to Separating Ownership from Corporate Control, 53 Stan. L. Rev. 539, 542–43 (2000).

12See Edward L. Glaeser & Andrei Shleifer, Legal Origins, 117 Q.J. Econ. 1193, 1194 (2002); Rafael La Porta et al., Law and Finance, 106 J. Pol. Econ. 1113, 1116 (1998); Rafael La Porta et al., Legal Determinants of External Finance, 52 J. Fin. 1131, 1131–33, 1149 (1997). Contra Holger Spamann, The “Antidirector Rights Index” Revisited, 23 Rev. Fin. Stud. 467, 467–68, 477 (2010); Holger Spamann, On the Insignificance and/or Endogeneity of La Porta et al.’s ‘Antidirector Rights Index’ Under Consistent Coding 69–71 (European Corporate Governance Inst., Law Working Paper No. 67, 2006), available at http://ssrn.com/abstract=894301.

13See, e.g., John C. Coffee, Jr., Law and the Market: The Impact of Enforcement, 156 U. Pa. L. Rev. 229, 232–33 (2007); Howell E. Jackson & Mark J. Roe, Public and Private Enforcement of Securities Laws: Resource-Based Evidence, 93 J. Fin. Econ. 207, 208–10, 230–33 (2009); La Porta et al., supra note 9, at 2–3; see also John Armour et al., Private Enforcement of Corporate Law: An Empirical Comparison of the United Kingdom and the United States, 6 J. Empirical Legal Stud. 687, 688–90 (2009); Mathias M. Siems, Private Enforcement of Directors’ Duties: Derivative Actions as a Global Phenomenon, in Collective Actions: Enhancing Access to Justice and Reconciling Multilayer Interests? 93, 93–96 (Stefan Wrbka et al. eds., 2012).

14See Coffee, supra note 13, at 266–67 (explaining that “private enforcement” refers to the enforcement of corporate and securities statutes and laws by plaintiffs who are private parties); Jackson & Roe, supra note 13, at 237; La Porta et al., supra note 9, at 20–21, 27–28. A civil action commenced against an officer of a corporation by the Securities and Exchange Commission is an example of public enforcement. See La Porta et al., supra note 9, at 2–3.

15Bernard Black et al., Legal Liability of Directors and Company Officials Part 2: Court Procedures, Indemnification and Insurance, and Administrative and Criminal Liability (Report to the Russian Securities Agency), 2008 Colum. Bus. L. Rev. 1, 2–4; see also Armour et al., supra note 13, at 689 (providing a quantitative comparison of private enforcement lawsuit outcomes in the United States and United Kingdom).

16See Black et al., supra note 15, at 1–13.

17John C. Coffee, Jr., Litigation Governance: Taking Accountability Seriously, 110 Colum. L. Rev. 288, 291–92 (2010) (internal quotation marks omitted).

18Id. at 344–45.

19See id. at 292–93. Professor Merritt Fox attributes the limited development toward class action litigation in Korea, Sweden, France, Spain, Germany, Norway, and the other countries to “some mix of the absence of contingent fees for plaintiffs lawyers, the existence of a ‘loser pays’ rule concerning the victor’s legal fees, an ‘opt in’ rather than an ‘opt out’ structure of class action, and the lack of explicit reference to securities fraud as being among the claims that can be brought collectively.” Merritt B. Fox, Civil Liability and Mandatory Disclosure, 109 Colum. L. Rev. 237, 251 (2009).

20See, e.g., The Law and Economics of Class Actions in Europe: Lessons from America (Jürgen G. Backhaus, Alberto Cassone & Giovanni B. Ramello eds., 2012) (referencing discovery sporadically, but never discussing or comparing civil law judicial fact investigation with party-on-party discovery in the United States nor noting the complete absence of U.S.-style discovery in all other jurisdictions); Coffee, supra note 13, at 288 (noting that “the contemporary debate in Europe centers on whether certain elements of the U.S. model—namely, opt-out class actions, contingent fees, and the ‘American rule’ on fee shifting—must be adopted in order to assure access to justice,” and offering “an alternative ‘nonentrepreneurial model’ for aggregate litigation that is consistent with European traditions”).

21 See Simeon Djankov et al., Courts, 118 Q.J. Econ. 453, 454–55 (2003). But see Holger Spamann, Legal Origin, Civil Procedure, and the Quality of Contract Enforcement, 166 J. Institutional & Theoretical Econ. 149, 149–50, 162 (2010) (examining the claim that civil procedure is less efficient in civil law countries than in common law countries and shedding doubt on the claim). One notable exception is the inclusion of a variable measuring rights of “access to evidence” in the “anti-self-dealing index” by financial economists Simeon Djankov et al. in The Law and Economics of Self-Dealing, 88 J. Fin. Econ. 430, 434 (2008). Their index considers whether plaintiffs can request the court to appoint an inspector to examine a company’s affairs, whether plaintiffs must identify specific documents before receiving any discovery, whether they can question defendants and nonparties in court, and to what extent they can do so without prior court approval. Id. at 434, 437. Surprisingly, however, the legal literature has failed to pick up on and analyze these variables.

22Stephen N. Subrin, Discovery in Global Perspective: Are We Nuts? 52 DePaul L. Rev. 299, 301–03 (2002); see Geoffrey C. Hazard, Jr., Discovery and the Role of the Judge in Civil Law Jurisdictions, 73 Notre Dame L. Rev. 1017, 1017 (1998) [hereinafter Hazard, Discovery and the Role of the Judge]; Geoffrey C. Hazard, Jr., Discovery Vices and Trans-substantive Virtues in the Federal Rules of Civil Procedure, 137 U. Pa. L. Rev. 2237, 2237–39 (1989) (discussing the broad access to documents provided by the Federal Rules of Civil Procedure).

23See generally, R.R. Verkerk, Fact-Finding in Civil Litigation: A Comparative Perspective 142 (2010) (providing an in-depth treatment of the historical evolution of principles and practices of fact investigation in different legal systems); Oscar G. Chase, American “Exceptionalism” and Comparative Procedure, 50 Am. J. Comp. L. 277, 292–96 (2002); John H. Langbein, The German Advantage in Civil Procedure, 52 U. Chi. L. Rev. 823, 824 (1985) (explaining that “by assigning judges rather than lawyers to investigate the facts, the Germans avoid the most troublesome aspects of [American] practice”); Richard L. Marcus, Putting American Procedural Exceptionalism into a Globalized Context, 53 Am. J. Comp. L. 709, 732 (2005) (book review); Subrin, supra note 22, at 301–02 (explaining that “two of the biggest differences between civil law countries and the United States with respect to pretrial discovery are the centrality of the judge in civilian civil litigation and the continuity of the proceedings”); Yasukei Taniguchi, The 1996 Code of Civil Procedure of Japan—A Procedure for the Coming Century?, 45 Am. J. Comp. L. 767, 772–76 (1997) (describing very limited changes to fact investigation in Japan’s new civil code). See generally Oscar G. Chase et al., Civil Litigation in Comparative Context 207–40 (2007) (comparing the discovery mechanisms in England, the United States, Germany, and Japan).

24See Neil Andrews, English Civil Procedure: Fundamentals of the New Civil Justice System 595, 613–21 (2003) (discussing limited situations in which judges are likely to grant pre-action disclosure orders); Kuo-Chang Huang, Introducing Discovery into Civil Law 231 (2003) (“Unlike discovery under the FRCP, there is no general scope of discovery uniformly applicable to all discovery devices under the English system.”); id. at 235; Paul Matthews & Hodge M. Malek, Disclosure 27–29 (4th ed. 2012) (discussing the British system); Michael Zander, Cases and Materials on the English Legal System 86–88 (10th ed. 2007). See generally Adrian A.S. Zuckerman, Civil Procedure (2003) (treating the English system).

25See Huang, supra note 24, at 21–27; Peter L. Murray & Rolf Stürner, German Civil Justice 275–78 (2004); Verkerk, supra note 23, at 16, 30, 248, 264–65.

26The term “inquisitorial” is here used as a short-hand for the American reader. Civil law proceduralists do not view their adjudication system as “inquisitorial” but as “adversarial,” and the neutrality of the judge is a fundamental principle of modern civil law. See Verkerk, supra note 23, at 280–81. Like translations of legal vocabulary more generally, the word “adversarial” also has different meanings in different legal systems and must be understood in connection with the other principles that inflect it, and in the context of the practices that inform and are informed by the concept. Note that, in civil law systems, appeals court judges make their own evidentiary findings and may seek additional evidence. See, e.g., Murray & Stürner, supra note 25, at 16. But this does not markedly increase judicial resources available for fact-finding, if at all.

27See Fed. R. Civ. P. 26–37 & advisory committee’s notes (describing these developments).

28See infra notes 460–62 and accompanying text, for a discussion of the theory that the federal rules have turned trial judges into “managerial judges.”

29See, e.g., Governance of the American Economy 3–5 (John L. Campbell, J. Rogers Hollingsworth & Leon N. Lindberg eds., 1991) (explaining how and why transformations in governance occurred in different industries).

30For a discussion of legal realism’s influence on securities regulation and civil procedure during the New Deal, see infra notes 105–19 and accompanying text.

31Jack H. Friedenthal, A Divided Supreme Court Adopts Discovery Amendments to the Federal Rules of Civil Procedure, 69 Calif. L. Rev. 806, 818 (1981); see also Geoffrey C. Hazard, Jr., Conference on the Federal Rules of Civil Procedure: Report 3 (1995) (reporting on the proceedings of the Conference on the Federal Rules of Civil Procedure, convened in Dallas, Texas, on March 30–31, 1995); Richard L. Marcus, Discovery Containment Redux, 39 B.C. L. Rev. 747, 749 (1998) (“[B]road discovery put pressure on the substantive law to expand . . . .”).

32See, e.g., Franklin S. Wood, Chamber of Commerce of the State of N.Y., Survey and Report Regarding Stockholders’ Derivative Suits 61–66 (1944) (discussing the cost and nuisance of discovery).

33John H. Beisner, U.S. Chamber Inst. for Legal Reform, The Centre Cannot Hold: The Need for Effective Reform of the U.S. Civil Discovery Process 2 (May 2010), available at http://ilr.iwssites.com/uploads/sites/1/ilr_discovery_2010_0.pdf.

34See, e.g., Paul D. Carrington, Politics and Civil Procedure Rulemaking: Reflections on Experience, 60 Duke L.J. 597, 606–09 (2010) (detailing the efforts of business leaders to effect reform); Danya Shocair Reda, The Cost-and-Delay Narrative in Civil Justice Reform: Its Fallacies and Functions, 90 Or. L. Rev. 1085, 1123–24 (2012) (“[P]ro-business and insurance organizations helped to underwrite surveys that were then relied upon by . . . the CJRA [Civil Justice Reform Act of 1990] legislation . . .‘[i]t is no secret that the anti-discovery pressure [in the 1998 proposed discovery rules] has come from defendants, especially defendants in product liability, securities, and antitrust cases.’” (quoting Elizabeth G. Thornburg, Giving the “Haves” a Little More: Considering the 1998 Discovery Proposals, 52 SMU L. Rev. 229, 243 (1999))).

35Pub. L. No. 104-67, 109 Stat. 737 (1995) (codified as amended in scattered sections of 15 U.S.C.).

36See, e.g., Timothy K. Roake & Gordon K. Davidson, The Private Securities Litigation Reform Act of 1995, Corp. & Sec. L. Update (Fenwick & West LLP, Mountain View, Cal.), Jan. 1996, at 1, available at http://www.fenwick.com/fenwickdocuments/corp_sec_01-00-96.pdf (“Abusive securities litigation has been an unfortunate fact of life in the Silicon Valley for years, and the high tech community lobbied long, hard and well for [this] reform. . . . against ‘strike suits.’”).

3717 C.F.R. § 240.10b-5 (2012) (prohibiting “any untrue statement of a material fact . . . in connection with the purchase or sale of any security”).

38See Hillary A. Sale, Heightened Pleading and Discovery Stays: An Analysis of the Effect of the PSLRA’s Internal-Information Standard on ’33 and ’34 Act Claims, 76 Wash. U. L.Q. 537, 552–53 (1998) (reviewing the arguments behind the reform). In Novak v. Kasaks, 216 F.3d 300 (2d Cir. 2000), the Second Circuit stated that “[l]egislators were apparently motivated in large part by a perceived need to deter strike suits wherein opportunistic private plaintiffs file securities fraud claims of dubious merit in order to exact large settlement recoveries,” id. at 306. This view was also endorsed by the academic literature. See generally Janet Cooper Alexander, Do the Merits Matter? A Study of Settlements in Securities Class Actions, 43 Stan. L. Rev. 497, 499–500 (1991) (describing how the structural characteristics of securities class actions, rather than substantive merits, determine the amount of settlements).

39See H.R. Rep. No. 104-369, at 31, 37 (1995) (Conf. Rep.). See generally Craig C. Martin & Matthew H. Metcalf, The Fiduciary Duties of Institutional Investors in Securities Litigation, 56 Bus. Law. 1381, 1382–83 (2001) (outlining the PSLRA’s stated purposes and extensive legislative history).

40See Securities Act of 1933, 15 U.S.C. § 77z-1(b)(1) (2012); Securities Exchange Act of 1934, id. § 78u-4(b)(3)(A)–(B).

41See, e.g., Stephen J. Choi, Do the Merits Matter Less after the Private Securities Litigation Reform Act?, 23 J.L. Econ. & Org. 598, 599 (2007) (“Without the ability to engage in discovery, plaintiffs’ attorneys face a higher cost in determining the presence of specific misleading statements and omissions and the materiality of such misstatements and omissions.”); Sale, supra note 38, at 538.

42See, e.g., Jason M. Rosenthal, Staying Discovery in Federal Securities Lawsuits, Prac. Litigator, Nov. 2002, at 7 (admitting the high cost of discovery). See generally John H. Beisner, Discovering a Better Way: The Need for Effective Civil Litigation Reform, 60 Duke L.J. 547, 571–75 (2010) (pointing out flaws in arguments by scholars attempting to refute the assertion that discovery is expensive and prone to abuse); Earl C. Dudley, Jr., Discovery Abuse Revisited: Some Specific Proposals to Amend the Federal Rules of Civil Procedure, 26 U.S.F. L. Rev. 189, 190 (1992) (arguing that the “managerial approach” to discovery is not cost-effective); Frank H. Easterbrook, Discovery as Abuse, 69 B.U. L. Rev. 635, 636 (1989) (noting how the high costs of discovery shape litigants’ decisions). But see Sale, supra note 38, at 538, 540 (arguing that the PSLRA reform was “overinclusive” in that meritorious claims would be dismissed and proposing a solution based on judicially managed discovery); Lynn A. Stout, Type I Error, Type II Error, and the Private Securities Litigation Reform Act, 38 Ariz. L. Rev. 711, 713–15 (1996) (discussing potential underdeterrence problems).

43See generally, Robert W. Hamilton & Jonathan R. Macey, Cases and Materials on Corporations 734 (10th ed. 2007) (suggesting that “virtually all derivative litigation filed today is without merit”); Kenneth B. Davis, Jr., The Forgotten Derivative Suit, 61 Vand. L. Rev. 387, 388 (2008); Robert B. Thompson & Randall S. Thomas, The New Look of Shareholder Litigation: Acquisition-Oriented Class Actions, 57 Vand. L. Rev. 133, 135 (2004).

44See Bernard Black, Brian Cheffins & Michael Klausner, Liability Risk for Outside Directors: A Cross-Border Analysis, 11 Eur. Fin. Mgmt. 153, 155 (2005) [hereinafter Black et al., Liability Risk for Outside Directors]; Bernard Black, Brian Cheffins & Michael Klausner, Outside Director Liability, 58 Stan. L. Rev. 1055, 1059 (2006) [hereinafter Black et al., Outside Director Liability]; Brian R. Cheffins & Bernard S. Black, Outside Director Liability Across Countries, 84 Tex. L. Rev. 1385, 1387 (2006).

45See Armour et al., supra note 13, at 703 (2009).

46See Kenneth B. Davis, Jr., Structural Bias, Special Litigation Committees, and the Vagaries of Director Independence, 90 Iowa L. Rev. 1305, 1331 (2005) (discussing the analytical link between demand requirements and business judgment considerations); Robert B. Thompson & Randall S. Thomas, The Public and Private Faces of Derivative Lawsuits, 57 Vand. L. Rev. 1747, 1759 (2004) (“Even if a suit survives the demand hurdle, the board of directors can still form a special independent litigation committee that can dismiss the case.”). But see Minor Myers, The Decisions of the Corporate Special Litigation Committees: An Empirical Investigation, 84 Ind. L.J. 1309, 1311 (2009) (finding that special litigation committees do not always dismiss derivative litigation).

47See infra Part III.A.5.

48See infra Part III.A.4–5.

49The term “private attorney general” was coined by Judge Frank in Associated Industries of New York State, Inc. v. Ickes, 134 F.2d 694, 704 (2d Cir. 1943), and refers to a private party who brings a lawsuit considered to be in the public interest. See David Shub, Note, Private Attorneys General, Prevailing Parties, and Public Benefit: Attorney’s Fees Awards for Civil Rights Plaintiffs, 42 Duke L.J. 706, 708 (1992).

50See infra Part III.A.45.

51In this Article, “shareholder litigation” includes all types of shareholder litigation at the state and federal level, including shareholder direct actions (such as breach of duty in the merger context, or requests under title 8, section 220 of the Delaware Code), shareholder derivative actions, and shareholder securities class actions. We focus primarily on shareholder derivative actions under state law and shareholder securities class actions under federal law as examples. While shareholder derivative actions may, of course, be brought in federal courts, the Securities Litigation Uniform Standards Act of 1998 (SLUSA) precludes federal securities class actions from being litigated in state courts. See Pub. L. No. 105-353, 112 Stat. 3227 (1998) (codified in scattered sections of 15 U.S.C.).

52See infra Part III.A.4.

53See Paul Lomas & Daniel Kramer, Introduction to Corporate Internal Investigations: An International Guide 1, 3 (Paul Lomas & Daniel J. Kramer eds., 2008) (“The US has been in the forefront of the trends that have led corporations to conduct internal investigations and to cooperate with governmental inquiries, but the procedures in the US are constantly evolving themselves.”).

54See infra note 233 and accompanying text.

55See Frank B. Cross & Robert A. Prentice, The Economic Value of Securities Regulation, 28 Cardozo L. Rev. 333, 363–64 (2006) (discussing the virtues of mandatory disclosures).

56See generally Érica Gorga & Michael Halberstam, Knowledge Inputs, Legal Institutions, and Firm Structure: Towards a Knowledge-Based Theory of the Firm, 101 Nw. U. L. Rev. 1123, 1127 (2007) (arguing that firms must produce knowledge, disperse it within the firm, but also “prevent its transfer outside the firm”).

57See Tom C.W. Lin, Undressing the CEO: Disclosing Private, Material Matters of Public Company Executives, 11 U. Pa. J. Bus. L. 383, 385, 394–95 (2009) (discussing the growing prominence and wealth of corporate CEOs and the need for a corresponding increase in disclosure); Michael R. Siebecker, Trust & Transparency: Promoting Efficient Corporate Disclosure Through Fiduciary-Based Discourse, 87 Wash. U. L. Rev. 115, 117 (2009) (“[D]espite a host of federal and state statutes mandating disclosure of various corporate practices, corporations seem reluctant to disclose fully what consumers and investors want to know . . . .”).

58See John S. Beckerman, Confronting Civil Discovery’s Fatal Flaws, 84 Minn. L. Rev. 505, 525 (2000) (“Proficient advocates maximize strategic advantage during the investigation and discovery phase of pretrial litigation by obtaining as much relevant and useful information, admissions and impeachment material as possible from and about the adversary, for potential use at trial . . . .”).

59See William W. Schwarzer, Mistakes Lawyers Make in Discovery, Litigation, Winter 1981, at 31, 31 (“Deposition notices drop like autumn leaves.”); see also, e.g., Sperano v. Invacare Corp., No. 03-CV-6157, 2006 WL 3524483, at *1–2 (W.D.N.Y. Dec. 6, 2006) (allowing deposition of the defendant’s former employee, even though she suffered from a rare brain disease).

60Hopson v. Mayor of Balt., 232 F.R.D. 228, 245 (D. Md. 2005) (describing the scope of electronic records as including “e-mail, voice mail, archived data, back-up or disaster recovery data, laptops, personal computers, PDA’s, deleted data”).

61See infra notes 296–308 and accompanying text.

62Exposure need not be to the general public in order to effectively discipline. As we discuss below, defensive discovery alone, or even a prediscovery corporate internal investigation by a special litigation committee, both of which operate in the shadow of the discovery rules, can lead to employment actions, improve compliance monitoring, and deter aggressive reporting practices.

63See, e.g., Press Release, U.S. Dep’t of Justice, Former Computer Associates Executives Indicted on Securities Fraud, Obstruction Charges (Sept. 22, 2004), http://www.justice.gov/opa/pr/2004/September/04_crm_642.htm (stating that the financial restatement and the realization of the magnitude of the fraud by Computer Associates executives resulted from corporate internal investigation).

64See Black et al., Outside Director Liability, supra note 44, at 1099.

65See Eliezer M. Fich & Anil Shivdasani, Financial Fraud, Director Reputation, and Shareholder Wealth, 86 J. Fin. Econ. 306, 335 (2007) (showing that “fraud allegations have significant reputational consequences for outside directors”); Jeffrey N. Gordon, The Rise of Independent Directors in the United States, 1950–2005: Of Shareholder Value and Stock Market Prices, 59 Stan. L. Rev. 1465, 1489 (2007) (concluding that “in light of the Delaware courts’ reluctance to impose monetary liability on directors, the most significant independence-enhancing effect of litigation is probably through improving the operation of the reputation market rather than through the threat of monetary sanctions”); Bruce Haslem, Aimee Hoffmann Smith & Irena Hutton, How Much Do Corporate Defendants Really Lose? A New Verdict on the Reputation Loss Induced by Corporate Litigation 7 (July 8, 2013) (unpublished manuscript), http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2290821.

66Unless there is a confidentiality agreement between the parties, or a protective order is obtained by the court, a party is generally free to pass on documents or deposition transcripts to third parties who are not involved in the litigation. Because discovery materials are not filed with the court, see Fed. R. Civ. P. 5, it is not always recognized that they are not protected from publication, even at an early stage in the litigation process. Some academics who have attended to the issue have deplored the “public by default” standard. But in so doing, their focus is on the rights and burdens of the parties and other narrow procedural issues without considering the broader impact that publication and its threat have on the regulation of securities markets and corporate governance more generally. See, e.g., Richard L. Marcus, A Modest Proposal: Recognizing (At Last) That the Federal Rules Do Not Declare That Discovery Is Presumptively Public, 81 Chi.-Kent L. Rev. 331, 331 (2006) (“[T]he real problem is that the overbroad notion that all information turned over in discovery (no matter what it contains, or what the court or even the parties, know about the material involved) is presumptively open to the public, even if no party wants to permit public access to the information.”).

67One need only search YouTube to locate hundreds of excerpts of videotaped depositions that may embarrass defendant corporate managers. See, e.g., CenterJD, Enron Banks—Will President Bush Support Them?, YouTube (Jul. 30, 2007), http://www.youtube.com/watch?v=9zxAJO7owy8 (showing deposition of Enron’s Chief Financial Officer, Andrew Fastow); see also infra note 226.

68See Citizens First Nat’l Bank of Princeton v. Cincinnati Ins. Co., 178 F.3d 943, 946 (7th Cir. 1999) (observing that “[m]ost cases endorse a presumption of public access to discovery materials” and citing numerous such cases); Okla. Hosp. Ass’n v. Okla. Publ’g Co., 748 F.2d 1421, 1424 (10th Cir. 1984) (“[P]arties to litigation have a constitutionally protected right to disseminate information obtained by them through the discovery process absent a valid protective order . . . .”); see also Manual for Complex Litigation (Fourth) § 11.432, at 67–68 (2004) (endorsing the sharing of information obtained from discovery for litigation purposes).

69Note that the deposition transcript of Neil Cotty, Bank of America’s Chief Accounting Officer, was marked “confidential” but was nonetheless filed with the court and testimony was cited in the press. See Continued Videotaped Deposition of Neil Andrew Cotty, In re Bank of Am. Corp., Sec., Derivative, & ERISA Litig., 757 F. Supp. 2d 260 (S.D.N.Y. 2010) (No. 09 MD 2058 (PKC)), ECF No. 694-6. Even where sealed documents are obtained and leaked, as in the Zyprexa case, judges craft their injunctions narrowly and have not issued general injunctions on republication of already leaked documents. See, e.g., In re Zyprexa Injunction, 474 F. Supp. 2d 385, 395, 397 (E.D.N.Y. 2007).

70See John C. Coffee Jr., Gatekeepers: The Professions and Corporate Governance 2 (2006) (“[G]atekeeper . . . [means] an agent who acts as a reputational intermediary to assure investors as to the quality of the ‘signal’ sent by the corporate issuer.”). Examples of gatekeepers include auditors, investment bankers, securities analysts, and attorneys. Id. at 2, 192.

71See infra Part III.A.5 (discussing information sharing by special litigation committees with the SEC).

72See Brehm v. Eisner, 746 A.2d 244, 249, 252–53 (Del. 2000).

73In re Walt Disney Co. Derivative Litig. (Disney I), 731 A.2d 342, 353 (Del. Ch. 1998), aff’d in part, rev’d in part sub nom. Brehm v. Eisner, 746 A.2d 244 (Del. 2000).

74In re Walt Disney Co. Derivative Litig. (Disney IV), 906 A.2d 27 (Del. 2006); Brehm, 746 A.2d 244; In re Walt Disney Co. Derivative Litig. (Disney III), 907 A.2d 693 (Del. Ch. 2005), aff’d, 906 A.2d 27 (Del. 2006); In re Walt Disney Co. Derivative Litig. (Disney II), 825 A.2d 275 (Del. Ch. 2003); Disney I, 731 A.2d 342.

75Disney I, 731 A.2d at 380.

76Disney II, 825 A.2d at 291.

77See infra Part II.D.

78Disney III, 907 A.2d at 697.

79See, e.g., Jonathan Macey, Delaware: Home of the World’s Most Expensive Raincoat, 33 Hofstra L. Rev. 1131, 1131 (2005).

80Disney III, 907 A.2d at 697, 760, 763. For public reaction, see, for example, Steven Taub, Disney Wins Shareholder Lawsuit, CFO (Aug. 10, 2005), http://ww2.cfo.com/risk-compliance/2005/08/disney-wins-shareholder-lawsuit/ (reporting that the court “did not exonerate Eisner and the Disney board outright”).

81See Bloomberg News, Eisner Makes Clean Break, Resigns from Board of Directors, Chi. Trib. (Oct. 7, 2005), http://articles.chicagotribune.com/2005-10-07/business/0510070136_1_robert-iger-michael-eisner-disney (reporting that Eisner resigned as Disney’s CEO on September 30, 2005).

82Disney III, 907 A.2d at 697.

83See id. at 696–97. One reason for trial courts to include such lengthy recitations of fact is, of course, to avoid getting overturned on appeal, as the standard of review for facts is clear error, but a de novo standard applies to judicial review of matters of law. See Fed. R. Civ. P. 52(a)(6); United States v. U.S. Gypsum Co., 333 U.S. 364, 399 (1948); see also Steven Alan Childress, Standards of Review Primer: Federal Civil Appeals, 229 F.R.D. 267, 270 (2005).In European civil law systems, opinions are shorter and include significantly less recitations of facts. See, e.g., Mitchel de S.-O.-L’E. Lasser, Judicial Deliberations: A Comparative Analysis of Judicial Transparency and Legitimacy 104–112 (2004) (discussing the brevity of European Court of Justice opinions); Gillian K. Hadfield, The Quality of Law: Judicial Incentives, Legal Human Capital and the Evolution of Law 28 (Univ. S. Cal. Ctr. in Law, Econ. & Org., Research Paper No. C07-3, 2007), available at http://ssrn.com/abstract=967494 (“Common law regimes, for example, tend to produce extensive, publicly available judicial opinions laying out factual findings and legal reasoning; civil code regimes tend to produce detailed academic commentary, published alongside short, relatively opaque, legal opinions with extensive judicial analysis sometimes confined to documents distributed only within the judiciary.” (citing Lasser, supra)); Annette Marfording & Ann Eyland, Civil Litigation in New South Wales: Empirical and Analytical Comparisons with Germany, 500–01 (Univ. of N.S.W. Faculty of Law Research Series, Paper No. 28, 2010), available at http://ssrn.com/abstract=1641554 (comparing page length of decisions in New South Wales and Germany); see also, Zivilprozessordnung [ZPO] [Code of Civil Procedure] § 313(2) (Ger.) (providing that, in the reasons for judgment, the plaintiff’s claim and the parties’ respective factual allegations are to be presented only as far as essential, and otherwise the court is to simply refer to the parties’ pleadings and the transcript).

84U.S. judges do not collect facts, identify witnesses, or control the preparation of evidence for trial. See infra notes 148–53 and accompanying text.

85The scope of discovery extends to “relevant” information and other information “reasonably calculated to lead to the discovery of admissible evidence.” Fed. R. Civ. P. 26(b)(1).

86See infra note 254 and accompanying text.

87Disney III, 907 A.2d at 700.

88Id. at 700–02.

89Id. at 699–740.

90See, e.g., id. at 724–40.

91See id. at 702–40.

92Disney I, 731 A.2d 342, 360 (Del. Ch. 1998), aff’d in part, rev’d in part sub nom. Brehm v. Eisner, 746 A.2d 244 (Del. 2000).

93Disney III, 907 A.2d at 704 & n.40.

94Id. at 705–06.

95Id. at 705.

96The record states, “In executive session, the board was informed of the reporting structure that Eisner and Ovitz agreed to, but no discussion of the discontent Litvack or Bollenbach expressed at Eisner’s home was recounted.” Id. at 710.

97Id. at 762–63 (footnote omitted).

98See Macey, supra note 79, at 1132–33.

99See id. at 1132, 1134–36.

100See generally George A. Akerlof, The Market for “Lemons”: Quality Uncertainty and the Market Mechanism, 84 Q.J. Econ. 488, 489 (1970) (explaining the information asymmetry problem in the context of used cars).

101See Lucian Arye Bebchuk, Litigation and Settlement Under Imperfect Information, 15 Rand J. Econ. 404, 413 (1984) (“[T]he discovery requirement will likely reduce the informational asymmetry between the parties . . . .”).

102Third-generation transparency work suggests that transparency and disclosure are effective if interested parties can act on the information. See Archon Fung et al., Full Disclosure: The Perils and Promise of Transparency 173–74 (2007).

103See Marcus, supra note 23, at 709. (“American proceduralists have not been comparativists.”) American proceduralists instead believe “it is best and sufficient to attend only to American topics.” Id. at 710.

104Hazard, Discovery and the Role of the Judge, supra note 22, at 1018.

105See generally David Marcus, The Federal Rules of Civil Procedure and Legal Realism as a Jurisprudence of Law Reform, 44 Ga. L. Rev. 433, 438 (2010) (showing how the positive potential of legal realism translated into reform); Stephen N. Subrin, How Equity Conquered Common Law: The Federal Rules of Civil Procedure in Historical Perspective, 135 U. Pa. L. Rev. 909, 944–48 (1987) (documenting the Federal Rules’ heavy reliance on equity procedures, which mirrored New Deal principles). See also Judith Resnik, Failing Faith: Adjudicatory Procedure in Decline, 53 U. Chi. L. Rev. 494, 502–05 (1986).

106See Marcus, supra note 105, at 468–69.

107See id.

108Pub. L. No. 73-415, 48 Stat. 1064 (1934) (codified as amended at 28 U.S.C. § 2072 (2012)). The Rules Enabling Act gave the Supreme Court the power to make rules of procedure for federal courts as long as they did not “abridge, enlarge, or modify any substantive right.” 28 U.S.C. § 2072(b).

109Pub. L. No. 73-22, tit. I, 48 Stat. 74 (1933) (codified as amended at 15 U.S.C. §§ 77a–77aa (2012)).

110Pub. L. No. 73-291, 48 Stat. 881 (1934) (codified as amended at 15 U.S.C. §§ 78a–78pp).

111See generally Stephen N. Subrin, Fishing Expeditions Allowed: The Historical Background of the 1938 Federal Discovery Rules, 39 B.C. L. Rev. 691, 729 (1998) (providing a historical survey of discovery reforms and their driving rationale).

112See generally, American Bar Association, Rules of Civil Procedure for the District Courts of the United States with Notes and Proceedings of the Institute on Federal Rules (William W. Dawson ed., 1938) [hereinafter IFR Proceedings] (providing a transcript of the proceedings of the ABA Institute on Federal Rules in Cleveland, Ohio, on July 21–23, 1938, at which Dean Clark was a lecturer).

113See C. Paul Rogers III, The Antitrust Legacy of Justice William O. Douglas, 56 Clev. St. L. Rev. 895, 897, 899–903 (2008).

114Pub. L. No. 73-415, 48 Stat. 1064 (1934) (codified at 28 U.S.C. § 2072 (2012)).

115See Subrin, supra note 111, at 693–729.

116See Marcus, supra note 105, at 476 (describing the writ system).

117See Conley v. Gibson, 355 U.S. 41, 48 (1957) (“The Federal Rules reject the approach that pleading is a game of skill in which one misstep by counsel may be decisive to the outcome . . . .”), abrogated on other grounds by Bell Atl. Corp. v. Twombly, 550 U.S. 544 (2007).

118See, e.g., Fed. R. Civ. P. 8; Fed. R. Civ. P. Form 20.

119Fed. R. Civ. P. 8(a)(2); Swierkiewicz v. Sorema N.A., 534 U.S. 506, 508 (2002).

120See supra note 117; cf. Subrin, supra note 111, at 693.

121See Subrin, supra note 111, at 703.

122See id. at 716.

123See George Ragland, Jr., Discovery Before Trial 19–24 (1932) (surveying state discovery practices); Subrin, supra note 105, at 922–25.

124See Subrin, supra note 111, at 718–19. Although the earlier Federal Equity Rules had contained some related provisions for depositions and the subpoena duces tecum that authorized the procurement of documents from third parties, these tools were essentially limited to situations in which witnesses were unavailable for examination in open court and required a showing of good cause. See id. at 699.

125Id. at 739–40.

126Ragland, supra note 123, at 251 (internal quotation mark omitted).

127Edson R. Sunderland, Foreword to Ragland, supra note 123, at iii.

128Hazard, supra note 31, at 3.

129Id.

130See Fed. R. Civ. P. 26.

131See Del. Rev. Stat. ch. 121, § 49 (1915) (“The deposition must be taken in writing . . . . [N]either party shall be present at the taking the [sic] deposition, and no question shall be put but those sent by the justice.”); see Ragland, supra note 123, at 283–84.

132See Ragland, supra note 123, at 22–23.

133See Subrin, supra note 105, at 918–19.

134See, e.g., IFR Proceedings, supra note 112, at 285.

135Id. at 279–80.

136Del. Ch. Ct. R. 40 (1868).

137Supra note 131. For the European perspective, compare Verkerk, supra note 23, at 246 (“[A] common restriction [of the scope of party interrogation in European systems] is that the law recognizes privileges that grant parties the right to withhold information.”). Such privileges go far beyond attorney client privileges but include the right to refuse to bear witness against oneself.

138See Ragland, supra note 123, at 19–24.

139See, e.g., Cal. Civ. Proc. Code § 2021 (1921); Ragland, supra note 123, at 279.

140See Ragland, supra note 123, at 126.

141See IFR Proceedings, supra note 112, at 278–80. In the 1938 Rules, Rule 26 on depositions was the most important discovery rule, and the discovery section was entitled “Depositions and Discovery.” See id. at 57.

142Fed. R. Civ. P. 26(a) (1938). Contrast this with civil law rules that require both. Verkerk, supra note 23, at 193 (“A personal appearance of the parties is a court hearing . . . .”). “[O]n the continent the party seeking discovery has to convince the court that there is a good reason to order the discovery.” Id. at 247.

143Fed. R. Civ. P. 26 (1938).

144Id.

145Fed. R. Civ. P. 30(a) (1938).

146Fed. R. Civ. P. 26(b) (1938). Contrast this with civil law rules making the judge the principal examiner. Verkerk, supra note 23, at 247–48; see also Langbein, supra note 23, at 826–30.

147Fed. R. Civ. P. 30(c).

148Fed. R. Civ. P. 26(b), 30(c).

149See 2 James Wm. Moore & Joseph Friedman, Moore’s Federal Practice Under the New Federal Rules § 26.01, at 2443 (1938) (“A party or witness whose deposition has been taken at an early stage in the litigation cannot, at a later date, readily manufacture testimony in contradiction to his deposition.”).

150Fed. R. Civ. P. 32(a). Other deposition testimony is also admissible, so long as it is otherwise admissible under the Federal Rules of Evidence. See, e.g., Fiber Sys. Int’l, Inc. v. Roehrs, 470 F.3d 1150, 1160 (5th Cir. 2006).

151See Fed. R. Civ. P. 27(a)(4) (“[A] deposition to perpetuate testimony may be used . . . in any later-filed district-court action involving the same subject matter . . . .”). In civil law jurisdictions, such as Germany, employees have a contractual duty to answer questions from an employer, e.g., in connection with a company internal investigation, but former employees or third parties have no obligation to answer questions outside of court. See Hanna Blanz et al., Investigations in Germany, in Corporate Internal Investigations: An International Guide, supra note 53, at 287, paras. 6.168–.170, at 332, paras. 6.180–.186, at 335–36.

152See Fed. R. Civ. P. 30(b)(1). A party can issue a deposition notice to a witness whose “name is unknown” by “provid[ing] a general description sufficient to identify the person or the particular class or group to which the person belongs.” Id. This is not possible under civil law rules, because taking the examination of a witness under oath is a judicial prerogative. Moreover, there is no obligation to answer questions by an attorney absent a specific contractual relationship, such as that between employer and employee. See, e.g., Blanz et al., supra note 151, paras. 6.180–.186, at 335–36.

153Fed. R. Civ. P. 30(b)(6).

154Subrin, supra note 111, at 721 (quoting Proceedings of the Meeting of the Advisory Committee on Rules for Civil Procedure of the Supreme Court of the United States (Feb. 22, 1936), in Records of the U.S. Judicial Conference: Committees on Rules of Practice and Procedures, 1935–1988, at CI-209-59 to CI-209-60).

155In 1913, the Congressional Pujo Committee investigated the money trusts, concluding that in almost “all great corporations with numerous and widely scattered stockholders[,] . . . management is virtually self-perpetuating and is able through the power of patronage, the indifference of stockholders[,] and other influences to control a majority of the stock.” H.R. Rep. No. 62-1593, at 147 (1913). See generally Louis D. Brandeis, Other People’s Money and How the Bankers Use It 2–3 (1914) (relying heavily on the Pujo Committee’s findings). These and other confrontations in the 1920s informed the debate about corporate governance problems arising from the separation of ownership and control. See, e.g., Adolf A. Berle, Jr. & Gardiner C. Means, The Modern Corporation and Private Property 86–88, 124 (1932). But see Harwell Wells, The Birth of Corporate Governance, 33 Seattle U. L. Rev. 1247, 1248 (2010) (suggesting that Berle and Means’s work “was not a radical break from earlier thought but the end product of several decades in which [scholars] tried to understand both the governance problems of the new modern corporations and what impact those corporate governance problems had on the nation’s public”). There has been some debate as to whether the findings of the Pujo Committee and of subsequent attacks on the money “trusts” by the Committee on Banking and Currency (1932–1934), also known as the Pecora Hearings, were misguided. See Guolin Jiang, Paul G. Mahoney & Jianping Mei, Market Manipulation: A Comprehensive Study of Stock Pools, 77 J. Fin. Econ. 147, 153–54 (2005) (finding informed trading and no manipulation in an empirical analysis of stock pools in the 1920s).

156Compare Fed. R. Civ. P. 34 (1938) (amended 1946) (emphasis added), with id. (1946) (amended 1970) (permitting discovery of documents “which constitute or contain evidence relating to any of the matters within the scope” (emphasis added)).

157Compare Fed. R. Civ. P. 34 (1946) (amended 1970), with id. (1970) (amended 1980) (removing the requirement of a showing of good cause and allowing that “[a]ny party may serve any other party a request . . . within the scope of [the rules]”).

158Fed. R. Civ. P. 26(b); see also Fed. R. Civ. P. 34 (allowing document discovery within the scope of Fed. R. Civ. P. 26(b)); Thompson v. Dept. of Hous. & Urban Dev., 199 F.R.D. 168, 172 (D. Md. 2001) (explaining that while the scope of discovery became narrower after “subject matter” in Rule 26(b) was changed to “claims and defenses,” debating the difference between the two is “equivalent to debating the number of angels that can dance on the head of a pin”). See generally 6 James Wm. Moore et al., Moore’s Federal Practice § 26.43 (3d ed. 2013) (discussing the scope of discovery after the 2010 amendments).

159Murray & Stürner supra note 25, at 277 (“[Prior to the recent 2001 reforms of the German code of civil procedure,] there was no right to production if the request was based solely on the document’s relevance to a decisive issue in the case. . . . If a party had a substantive right to possession of a document in the hands of a third party, that party could request that the proceeding be stayed to permit him to exercise his right to possession . . . .”); Verkerk, supra note 23, at 31 (discussing “common document” principle).

1608 Charles Alan Wright et al., Federal Practice and Procedure § 2008 (3d ed. 2010).

161Hager v. Graham, 267 F.R.D. 486, 491 (N.D. W. Va. 2010) (quoting Hickman v. Taylor, 329 U.S. 495, 501 (1947)) (internal quotation marks omitted); cf. Oil, Chem. & Atomic Workers Local Union No. 6-418 v. NLRB, 711 F.2d 348, 359–60 (D.C. Cir. 1983).

162These tools are not available in this form in civil law adjudication. See infra notes 524–37.

163In civil law countries, parties have no such means to identify discoverable information. See infra notes 537–39 and accompanying text.

164See James N. Dertouzos et al., The Legal and Economic Implications of Electronic Discovery: Options for Future Research 1 (2008) (describing the vast amount of discoverable data available in an electronic world).

165See Fed. R. Civ. P. 34.

166Subrin, supra note 22, at 307, 311.

167See Hazard, Discovery and the Role of the Judge, supra note 22, at 1018. Discovery is said to be “self-executing,” and experienced litigators know not to bother the judge with motions to compel or protective orders, under Fed. R. Civ. P. 26(b)(2)(B) and 26(c), respectively, unless necessary. But see infra notes 460–66 and accompanying text (discussing the thesis that “case management” rules have produced “managerial judges”).

168See, e.g., Fed. R. Civ. P. 26(a)(1) (requiring parties to provide preliminary information before discovery requests); Fed R. Civ. P. 26(b) (allowing discovery without court interference for relevant matters); Fed. R. Civ. P. 30(a) (allowing parties to orally depose any person without leave of court); Fed. R. Civ. P. 31 (allowing parties to depose any person through written depositions without leave of court).

169In civil law jurisdictions, parties must petition the court for an order to obtain evidence. See, e.g., Murray & Stürner, supra note 25, at 242 (explaining that the court may order production but is required to “assess the relevance of any such document under the standards applicable to proof in general when issuing such orders”); see also Huang, supra note 24, at 39–42.

170See Fed. R. Civ. P. 26(b)(2)(B), (c); see also, Marcus, supra note 31, at 748 (explaining that prior to the 1970 amendments, Rule 34 document discovery was available only on a motion based on a showing of “good cause”).

171Fed. R. Civ. P. 34(b)(2).

172Fed. R. Civ. P. 26(c), 37.

173Fed. R. Civ. P. 37(a)(5)(A).

174Hazard, Discovery and the Role of the Judge, supra note 22, at 1018.

175See John C. Coffee, Jr., Rescuing the Private Attorney General: Why the Model of the Lawyer as Bounty Hunter Is Not Working, 42 Md. L. Rev. 215, 215–16 (1983).

176See id. at 218.

177See Marcus, supra note 66, at 334 (“Nowhere in this country, nor anywhere in the world, had there previously been such a broad or unconfined opportunity for private litigants to use the power of the state to compel disclosure of information by others.”).

178304 U.S. 64, 78 (1938).

179John B. Oakley & Arthur F. Coon, The Federal Rules in State Courts: A Survey of State Court Systems of Civil Procedure, 61 Wash. L. Rev. 1367, 1377–78 (1986). In their 1986 study, comparing state and federal rules, Oakley and Coon found that twenty-two states, plus the District of Columbia, could be classified as having procedural systems that were true replicas of the FRCP. These are Alabama, Alaska, Arizona, Colorado, District of Columbia, Hawaii, Indiana, Kentucky, Maine, Massachusetts, Minnesota, Montana, New Mexico, North Dakota, Ohio, Rhode Island, South Dakota, Tennessee, Utah, Vermont, Washington, West Virginia, and Wyoming. Id. at 1377. Further, ten states, including Arkansas, Delaware, Georgia, Idaho, Kansas, Mississippi, Nevada, North Carolina, Oklahoma, and South Carolina largely replicated the federal rules, except for slight variations or codifications that had nothing to do with the nature or conduct of discovery. Id. at 1378. Oakley and Coon concluded that in a majority of states there was “‘but one procedure for state and federal courts.’” Id. at 1427 (quoting 1 W. Barron & A. Holtzoff, Federal Practice and Procedure § 9, at 44–45 (Wright ed. 1960)).

180See id. at 1378.

181See John B. Oakley, A Fresh Look at the Federal Rules in State Courts, 3 Nev. L.J. 354, 355 (2002/2003) (finding that states failed to keep up with the federal rules changes). Oakley concludes that “significant” amendments were frequently rejected, id. at 359, 383, but a careful examination of the differences he describes between state and federal rules shows that these differences, while important to recognize and apply with precision for attorneys litigating in these courts, are minor when compared to the fundamental differences between the U.S. system and civil law jurisdictions. See infra notes 528–39 and accompanying text.

182See Oakley, supra note 180, at 355, 383 (“Even among states that fifteen years ago could be counted as substantially conforming to the federal model of procedure, recent significant amendments have been more frequently rejected or ignored than adopted.”); see also Fed. R. Civ. P. historical note at xi–xii.

183Fed. R. Civ. P. 26(f).

184Fed. R. Civ. P. 26(f)(2).

185Supra note 158; see also Sanyo Laser Prods., Inc. v. Arista Records, Inc., 214 F.R.D. 496, 500 (S.D. Ind. 2003).

186Fed. R. Civ. P. 26(b)(2)(B).

187Thomas D. Rowe, Jr., Suzanna Sherry & Jay Tidmarsh, Civil Procedure 18 (3d ed. 2012).

188Oakley, supra note 179, at 379–81 (“Delaware has separate but substantially identical rules of civil procedure for each of its three principal systems of courts: the Court of Chancery, the Superior Court, and the inferior civil court of non-equitable jurisdiction, the Court of Common Pleas.”). Compare, e.g., Del. Ch. Ct. R. 11 (sanctions), Del. Ch. Ct. R. 12 (motion to dismiss), Del. Ch. Ct. R. 26 (discovery generally, scope, privileges, expert witnesses, timing, limitations), Del. Ch. Ct. R. 30 (oral depositions), Del. Ch. Ct. R. 31 (written depositions), Del. Ch. Ct. R. 32 (use of deposition testimony), Del. Ch. Ct. R. 33 (interrogatories), Del. Ch. Ct. R. 34 (production of documents and things), Del. Ch. Ct. R. 35 (physical and mental examinations of persons), Del. Ch. Ct. R. 36 (requests for admissions), Del. Ch. Ct. R. 37 (discovery disputes and sanctions), and Del. Ch. Ct. R. 45 (subpoena power), with Fed. R. Civ. P. 11, 12, 26, 30–37, 45.

189Oakley, supra note 179, at 380 (“Delaware Court of Chancery Rule 26 is a virtually exact copy of the 1970 version of Federal Rule 26, and does not incorporate any part of later amendments.”). The same is true for other jurisdictions. Id. at 362, 365–66, 369–70, 375.

190Compare Del. Ch. Ct. R. 26(b)(1) (“Parties may obtain discovery regarding any matter, not privileged, which is relevant to the subject matter involved in the pending action, whether it relates to the claim or defense of the party seeking discovery or to the claim or defense of any other party, including the existence, description, nature, custody, condition and location of any documents, electronically stored information, or tangible things and the identity and location of persons having knowledge of any discoverable matter. It is not ground for objection that the information sought will be inadmissible at the trial if the information sought appears reasonably calculated to lead to the discovery of admissible evidence.” (emphasis added)), with Fed. R. Civ. P. 26(b)(1) (allowing for discovery of any nonprivileged matter relevant to a party’s claim or defense but disposing of the “subject matter” language).

191Compare Del. Ch. Ct. R. 30(a), 33(a), with Fed. R. Civ. P. 30(a)(2)(A)(i), 33(a)(1).

192Leo E. Strine, Jr., Delaware Court of Chancery Announces Rule Changes and New Discovery Guidelines, Metropolitan Corp. Couns., May 2013, at 35, 35, available at http://www.metrocorpcounsel.com/pdf/2013/May/35.pdf.

193488 A.2d 858 (Del. 1985), overruled in part on other grounds by Gantler v. Stephens, 965. A.2d 695 (Del. 2009).

194Stout, supra note 42, at 711–12; see also Stephen J. Choi & Robert B. Thompson, Securities Litigation and Its Lawyers: Changes During the First Decade After the PSLRA, 106 Colum. L. Rev. 1489, 1500 (2006) (discussing “empirical findings on the merits of post-PSLRA claims”); Sherrie R. Savett, The Merits Matter Most and Observations on a Changing Landscape Under the Private Securities Litigation Reform Act of 1995, 39 Ariz. L. Rev. 525, 531 (1997) (observing that the PSLRA “produces great delay in getting the case moving to the merits”); Thompson & Thomas, supra note 46, at 1758 (2004) (describing efforts to deter “strike” or “nuisance” suits).

195Stout, supra note 42, at 711.

196Id.

197See Amanda M. Rose, The Multienforcer Approach to Securities Fraud Deterrence: A Critical Analysis, 158 U. Pa. L. Rev. 2173, 2192 (2010). The PSLRA used all of these methods to deter strike suits, or, in Stout’s terminology, avoid Type I errors. See Stout, supra note 42, at 711.

198Stout, supra note 42, at 712.

199See, e.g., Rose, supra note 197, at 2192.

200Indeed, the elimination of Type I and Type II errors was foremost in the minds of the drafters of the Federal Rules. In an extended discussion of New York practice, for example, George Ragland notes “the ineffectiveness of restrictions upon the scope of the discovery as an aid in arriving at the truth.” Ragland, supra note 123, at 130. In particular, he cites the New York limitation that “the defendant can have no discovery except on his affirmative defenses.” Id. at 132.

201698 A.2d 959 (Del. Ch. 1996).

202Id. at 960–61, 971 (“Concerning the possibility that the Caremark directors knew of violations of law, none of the documents submitted for review, nor any of the deposition transcripts appear to provide evidence of it.”).

203See, e.g., Smith v. Van Gorkom, 488 A.2d 858, 872 (Del. 1985) (explaining that the business judgment rule “is a presumption that in making a business decision, the directors of a corporation acted on an informed basis, in good faith and in the honest belief that the action taken was in the best interests of the company” (quoting Aronson v. Lewis, 473 A.2d 805, 812 (Del. 1984), overruled on other grounds by Brehm v. Eisner, 746 A.2d 244 (Del. 2000)) (internal quotation marks omitted)), overruled in part on other grounds by Gantler v. Stephens, 965. A.2d 695 (Del. 2009); Bayer v. Beran, 49 N.Y.S.2d 2, 6 (Sup. Ct. 1944). Absent fraud, illegality, or self-dealing, the business judgment rule ordinarily bars courts from reviewing the decisions of corporate management for breach of fiduciary duty. Van Gorkom, 488 A.2d at 872–73.

204See Aronson, 473 A.2d 805, 811–12; see also Fed. R. Civ. P. 23.1 (setting pleading requirements for shareholder derivative actions). The demand requirement forces a potential plaintiff to first make a demand on the company’s board of directors to remedy the alleged misconduct, or to allege why such demand would be futile. See Charles R.T. O’Kelley & Robert B. Thompson, Corporations and Other Business Associations: Cases and Materials 369 (5th ed. 2006).

205393 N.E.2d 994, 999–1000 (N.Y. 1979). Delaware followed a substantially similar rule in Zapata Corp. v. Maldonado, 430 A.2d 779, 786 (Del. 1981).

206See Robert Charles Clark, Corporate Law § 15.2.3, at 646–47 (1986).

207See Choi & Thompson, supra note 194, at 1492–93.

208Marcus, supra note 31, at 748.

209Pub. L. No. 104-67, 109 Stat. 737 (1995) (codified as amended in scattered sections of 15 U.S.C.).

210H.R. Rep. No. 104-369, at 31–32 (1995) (Conf. Rep.), reprinted in 1995 U.S.C.C.A.N. 730, 731. See generally Eugene Zelensky, Recent Legislation, New Bully on the Class Action Block—Analysis of Restrictions on Securities Class Actions Imposed by the Private Securities Litigation Reform Act of 1995, 73 Notre Dame L. Rev. 1135, 1135–37 & n.19 (1998) (discussing legislative history).

211Pub. L. No. 105-353, §§ 2(2), (5), 112 Stat. 3227, 3227 (1998) (codified as amended in scattered sections of 15 U.S.C.).

212Id. § 101(a), 112 Stat. at 3227–33.

213See H.R. Rep. No. 105-803, at 14 (1998) (Conf. Rep.).

214Securities Litigation Abuses: Hearing Before the Subcomm. on Sec. of the S. Comm. on Banking, Hous., & Urban Affairs, 105th Cong. 49 (1997) (statement of Arthur Levitt, Chairman, SEC); see also Randall S. Thomas & Kenneth J. Martin, Using State Inspection Statutes for Discovery in Federal Securities Fraud Actions, 77 B.U. L. Rev. 69, 71 (1997).

215First, the parties must complete all discovery before trial. Therefore, they gather as much information as possible. See generally Fed. R. Civ. P. 26 (a)(3)(B) (timing); Fed. R. Civ. P. 26(b)(1) (scope of discovery). See supra note 42 on the costs of discovery.

216Hensley v. Eckerhart, 461 U.S. 424, 429 (1983) (affirming the “American Rule”).

217See Black et al., Outside Director Liability, supra note 44, at 1059; Cheffins & Black, supra note 44, at 1465.

218M. Todd Henderson, Impact of the Rakoff Ruling: Was the Judge’s Scuttling of the SEC/BofA Settlement Legally Pointless or Incredibly Important—or Both?, Wall St. Law., Nov. 2009, at 1, 6 (“[A] suit generates not only legal costs but also negative publicity and the potential that even more damning information will be revealed during discovery or the trial.”).

219Id. at 6 (noting that Bank of America’s CEO, Kenneth Lewis, was forced to “leav[e] the bank, after reportedly being forced to give up many millions of dollars in compensation as a result of the suit and the handling of the merger with Merrill”). Note that this was nearly a year before the court ruled partially against defendants’ motion to dismiss in the plaintiff shareholders’ action. See In re Bank of Am. Corp. Sec., Derivative, & ERISA Litig., 757 F. Supp. 2d 260, 269–70 (S.D.N.Y. 2010); Tami Luhby, Bank of America CEO Ken Lewis to Retire, CNNMoney (Sept. 30, 2009, 6:53 PM), http://money.cnn.com/2009/09/30/news/companies/bank_of_america_ken_lewis_resigns/index.htm.

220See Chelsea C. Liu, Corporate Litigation, Corporate Governance Restructuring, and Executive Compensation 221–81 (Jan. 2013) (unpublished Ph.D. dissertation, University of Adelaide), http://digital.library.adelaide.edu.au/dspace/bitstream/2440/81971/3/02whole.pdf (reviewing literature on executive reputational damages resulting from litigation and providing empirical evidence).

221Disney II, 825 A.2d 275 (Del. Ch. 2003).

222Smith v. Van Gorkom, 488 A.2d 858 (Del. 1985), overruled in part by Gantler v. Stephens, 965. A.2d 695 (Del. 2009).

223See In re Apollo Grp. Inc. Sec. Litig., No. CV-04-2147-PHX-JAT, 2008 WL 410625, at *1 (D. Ariz. Feb. 13, 2008).

224See Kevin LaCroix, Plaintiffs Prevail in Mixed Jury Verdict in Household International Securities Fraud Trial, D&O Diary (May 7, 2009), http://www.dandodiary.com/2009/05/articles/securities-litigation/plaintiffs-prevail-in-mixed-jury-verdict-in-household-international-securities-fraud-trial/ (“According to data compiled by the Securities Litigation Watch, [as of 2009] only 21 cases have gone to trial since the PSLRA was enacted in 1995.” (citation omitted)).

225Supra notes 67–68.

226See, e.g., Karen Gullo, Goldman, Merrill E-Mails Show Naked Shorting, Filing Says, Bloomberg (May 16, 2012, 5:34 PM), http://www.bloomberg.com/news/2012-05-15/goldman-merrill-e-mails-show-naked-shorting-filing-says.html. See also, for example, Legacy Tobacco Documents Library, U.C.S.F., http://legacy.library.ucsf.edu (last visited May 28, 2014), which contains “more than 14 million documents (80+ million pages) created by major tobacco companies related to their advertising, manufacturing, marketing, sales, and scientific research activities.” A simple search turns up more than two million documents, which (based on spot checks) are Bates-stamped, and some of which are marked “Confidential: Tobacco Litigation.”

227For example, in the Bank of America case, Neil Cotty’s deposition transcript was marked “confidential,” but was nonetheless filed, and available online, as an exhibit to the plaintiffs’ motion for summary judgment. See Continued Videotaped Deposition of Neil Andrew Cotty, supra note 69.

228See sources supra note 42.

229Stephen M. Bainbridge, Corporation Law and Economics § 8.5, at 403−04 (2002) (“In almost all cases, the legal fees collected by plaintiff counsel exceeded the monetary payment to shareholders. . . . A radical solution would be elimination of derivative litigation.”). But see Thompson & Thomas, supra note 46, at 1172−73.

230See Black et al., Liability Risk for Outside Directors, supra note 44; Black et al., Outside Director Liability, supra note 44; Cheffins & Black, supra note 44.

231See Gorga & Halberstam, supra note 56, at 1168−69 (discussing the public goods characteristics of knowledge).

232Supra Part I.B.

233See Executive Compensation and Related Person Disclosure, Securities Act Release No. 8732A, Exchange Act Release No. 54302A, Investment Company Act Release No. 27444A, 71 Fed. Reg. 53,158 (Sept. 8, 2006); see also Jeffrey N. Gordon, Executive Compensation: If There’s a Problem, What’s the Remedy? The Case for “Compensation Discussion and Analysis, 30 J. Corp. L. 675, 692 (2005) (affirming that “[t]he Disney litigation, revived in the post-Enron environment, has become, regardless of outcome, an extended morality tale on the board’s responsibility to monitor executive compensation” (footnote omitted)).

234Thompson & Thomas, supra note 46, at 1749 (footnotes omitted). For an example of such advice on best practices after Disney, see Kevin M. LaCroix, Board Decision-Making Practices After the Disney Decision, OakBridge InSights (OakBridge Ins. Servs., Bloomfield, Conn.), Apr. 2006, available at http://www.rtspecialty.com/rtproexec/insights/Issue1DisneyArticle.pdf. Oakbridge is a leading directors and officers liability (D&O) insurer.

235Kevin LaCroix, First the Regulatory Investigation, Then the Securities Suit, D&O Diary (Dec. 9, 2013), http://www.dandodiary.com/2013/12/articles/securities-litigation/first-the-regulatory-investigation-then-the-securities-suit/.

236See Ralph C. Ferrara & Philip S. Khinda, SEC Enforcement Proceedings: Strategic Considerations for When the Agency Comes Calling, 51 Admin. L. Rev. 1143, 1151 (1999); see also Donna M. Nagy, Richard W. Painter & Margaret V. Sachs, Securities Litigation and Enforcement: Cases and Materials 668 (3d ed. 2012); cf. Ferrara & Khinda, supra, at 1149 (“[C]ounsel should request that [during the course of an informal inquiry a witness examination] be done via an interview rather than through testimony, i.e., not under oath and without the use of a court reporter. Interviews are generally preferable whenever parallel proceedings or litigation exists or is likely, as such discussions with the staff will not generate witness transcripts (as would be the case with formal testimony) that may be discoverable . . . .”).

237See, e.g., U.S. Gen. Accounting Office, GAO-09-358, Securities and Exchange Commission: Greater Attention Needed to Enhance Communication and Utilization of Resources in Enforcement Division 24 (2009) (finding that “resource challenges . . . [have delayed cases], reduc[ed] the number of cases that can be brought, and potentially undermin[ed] the quality of cases”); U.S. Gen. Accounting Office, GAO-02-302, SEC Operations: Increased Workload Creates Challenges 11–19 (2002) (finding that workload growth and limited resources have forced the SEC to be selective in its enforcement, have delayed investigations and case closings, and raise general concerns about SEC's enforcement; the SEC cannot pursue every case that it should); Joshua Gallu, SEC Enforcement Story Doesn't Add Up for 2011, Bloomberg (March 2, 2012, 12:00 AM), http://www.bloomberg.com/news/2012-03-02/sec-accounting-of-record-enforcement-year-in-2011-doesn-t-add-up.html (claiming that SEC filed fewer enforcement actions in 2011 than in 2009). One response is to dramatically increase the resources of the SEC. See Rose, supra note 197, at 2209–10. But the SEC’s congressional appropriations have already tripled since 2002. And there are political limits to further increases. Nagy et al., supra note 236, at 665.

238See generally George J. Stigler, The Theory of Economic Regulation, 2 Bell J. Econ. & Mgmt. Sci. 3 (1971) (discussing a theory for regulatory capture according to a supply and demand model for regulation).

239See generally Christine Jolls, Behavioral Law and Economics 11−18 (Nat’l Bureau of Econ. Research, Working Paper No. 12879, 2007), available at http://www.nber.org/papers/w12879 (discussing biases and judgment errors by rational actors).

240See Natalya Shnitser, Note, A Free Pass for Foreign Firms? An Assessment of SEC and Private Enforcement Against Foreign Issuers, 119 Yale L.J. 1638, 1645–46 (2010); see also Érica Gorga, Is the U.S. Law Enforcement Stronger? The Case of Securities Fraud by Brazilian Corporations and Lessons for the Private and Public Enforcement 34 (Feb. 5, 2014) (unpublished manuscript) (on file with authors).

241See supra Part I.B.

242See Rose, supra note 197, at 2191.

243See Paul J. Cohen et al., Investigations in the United States, in Corporate Internal Investigations: An International Guide, supra note 53, at 53, paras. 3.13–.20, at 59–62, paras. 3.134–.142, at 112–15.

244Richard H. McAdams, An Attitudinal Theory of Expressive Law, 79 Or. L. Rev. 339, 340 (2000) (“[L]aw changes behavior by signaling the underlying attitudes of a community or society. . . . [T]he information signaled by legislation and other law affects [people’s] behavior.”).

245Fed. R. Civ. P. 56.

246Id.

247Id.

248PACER stands for “Public Access to Court Electronic Records,” which is “an electronic public access service that allows users to obtain case and docket information from federal appellate, district and bankruptcy courts, and the PACER Case Locator via the Internet. PACER is provided by the federal Judiciary in keeping with its commitment to providing public access to court information via a centralized service.” PACER, http://www.pacer.gov (last visited May 28, 2014). Attorneys who practice in federal court, legal academics, law students, librarians, and journalists working for a major news organization have full access to documents filed with the federal courts in every case across the country. Proprietary databases, such as Bloomberg Law, will also include content provided by PACER.

249See In re Bank of Am. Corp., Sec., Derivative, & ERISA Litig., 757 F. Supp. 2d 260, 274−84 (S.D.N.Y. 2010). This case consolidated securities fraud, derivative, and ERISA claims against Bank of America for misstatements and omissions during the acquisition of Merrill Lynch following the 2008 financial meltdown. The full docket, including all pleadings, motions, court decisions, and exhibits filed with the court, are readily available, for a per-page fee, on PACER.

250Id. at 278−79.

251Motion for Summary Judgment, In re Bank of Am. Corp., 757 F. Supp. 2d 260 (No. 09-MD-2058-PKC); Order, In re Bank of Am. Corp., 757 F. Supp. 2d 260 (No. 09-MD-2058-PKC), ECF No. 620 (giving notice that the court had been advised of an executed settlement agreement on June 20, 2012).

252See Docket Items 577−659, Civil Docket, In re Bank of Am. Corp., 757 F. Supp. 2d 260 (No. 09 MD 2058 (PKC)).

253See, e.g., Continued Videotaped Deposition of Neil Andrew Cotty, supra note 69, at 14 (claiming he did not understand that Merrill Lynch was selling substantial assets to cut their balance sheet days before the deal was approved by the shareholders by stating, “My only comment earlier . . . that it would have come from either Joe or [BofA CEO] Ken [Lewis], but I don’t recall a discussion on the balance sheet at December 3rd”).

254Gretchen Morgenson, Merrill Losses Were Withheld Before Merger, N.Y. Times, June 4, 2012, at A1.

255Jessica Silver-Greenberg & Susanne Craig, Bank of America Settles Suit Over Merrill for $2.43 Billion, DealBook (Sept. 28, 2012, 9:19 PM), http://dealbook.nytimes.com/2012/09/28/bank-of-america-to-pay-2-43-billion-to-settle-class-action-over-merrill-deal/?_php=true&_type=blogs&_r=0.

256While depositions of attorneys are disfavored, defendants who claim to have relied on advice of counsel may waive their attorney–client privileges in this regard, thus opening the door to testimony by their counsel. See, e.g., Arthur Andersen LLP v. United States, 544 U.S. 696, 699 (2005); Rhone-Poulenc Rorer Inc. v. Home Indem. Co., 32 F.3d 851, 863 (3d Cir. 1994); Joy v. North, 692 F.2d 880, 893 (2d Cir. 1982); In re Bank of Am. Corp., 757 F. Supp. 2d at 279; In re ML-Lee Acquisition Fund II, L.P., 859 F. Supp. 765, 766 (D. Del. 1994).

257In an economic model of litigation and settlement decisions under imperfect information, Bebchuk argues that discovery requirements “increase the probability of settlement.” Bebchuk, supra note 101, at 413. In a model with one-sided discovery, Sobel shows that mandatory discovery rules reduce the probability of trials. Joel Sobel, An Analysis of Discovery Rules, Law & Contemp. Probs., Winter 1989, at 133, 133; see also Robert D. Cooter & Daniel L. Rubinfeld, An Economic Model of Legal Discovery, 23 J. Legal Stud. 435, 436 (1994) (“[D]iscovery increases settlements and decreases trials by organizing the voluntary exchange of information.”). In civil law jurisdictions, which do not allow broad discovery, settlement rates are lower. Huang reports an interesting development after the introduction of civil discovery in Taiwan in a 2000 legal reform. Kuo-Chang Huang, Does Discovery Promote Settlement? An Empirical Answer, 6 J. Empirical Legal Stud. 241 (2009). He shows that settlement rates for civil cases consistently increased over time in all district courts following the adoption of the discovery regime. Id. at 257−59.

258Alison Frankel, How BofA Was Forced to Settle $2.43 Bln Merrill Class Action, Reuters (Oct. 1, 2012), http://blogs.reuters.com/alison-frankel/2012/10/01/how-bofa-was-forced-to-settle-2-43-bln-merrill-class-action/.

259See, e.g., Peter J. Henning, Behind Rakoff’s Rejection of Citigroup Settlement, DealBook (Nov. 28, 2011, 5:14 PM), http://dealbook.nytimes.com/2011/11/28/behind-judge-rakoffs-rejection-of-s-e-c-citigroup-settlement/.

260Fed. R. Civ. P. 5(d)(1) (“[D]isclosures under Rule 26(a)(1) or (2) and the following discovery requests and responses must not be filed until they are used in the proceeding or the court orders filing . . . .”).

261Until the Supreme Court held that plaintiffs were not required to prove materiality (an essential element in securities fraud claims) on a motion for class certification, plaintiffs conducted discovery and filed relevant admissible evidence with the court on this issue. See Amgen Inc. v. Conn. Ret. Plans & Trust Funds, 133 S. Ct. 1184, 1191, 1202 (2013).

262See Del. Code Ann. tit. 8, § 220 (West Supp. 2014).

263Fed. R. Civ. P. 23(e)(2). The Advisory Committee Notes to the 2003 Amendments state that “[t]he court must make findings that support the conclusion that the settlement is fair, reasonable, and adequate. The findings must be set out in sufficient detail to explain to class members and the appellate court the factors that bear on applying the standard.” Fed. R. Civ. P. 23 advisory committee’s note (2003 Amendments); see also Fed. R. Civ. P. 23.1 (governing derivative actions); Donald F. Parsons, Jr. & Jason S. Tyler, Docket Dividends: Growth in Shareholder Litigation Leads to Refinements in Chancery Procedures, 70 Wash. & Lee L. Rev. 473, 497 (2013) (citing Barkan v. Amsted Indus., Inc., 567 A.2d 1279, 1283 (Del. 1989)) (“[T]he Court of Chancery vigilantly examines the merits of every settlement of a representative action, whether objected to or not, to determine whether, in the exercise of the reviewing court’s independent business judgment, the settlement is in the shareholders’ interest.”).

264In re Caremark Int’l Inc. Derivative Litig., 698 A.2d 959, 961 (Del. Ch. 1996).

265Id. at 960.

266See id. at 961–66.

267Id. at 960–61.

268Id. at 961 (emphasis added).

269Id.

270Id. at 962–63.

271Id. at 963.

272Id. at 966.

273Id. at 961–66. Chancellor Allen acknowledged that “none of the documents submitted for review, nor any of the deposition transcripts appear to provide evidence of [the possibility that the Caremark directors knew of violations of law].” Id. at 971.

274Id. at 971.

275See Thompson & Thomas, supra note 46, at 1749 (pointing out the importance of Caremark beyond the outcome of the case).

276Alexander, supra note 38, at 499; Roberta Romano, The Shareholder Suit: Litigation Without Foundation?, 7 J.L. Econ. & Org. 55, 55–56 (1991); see, e.g., Kevin M. LaCroix, More Options Backdating Settlements and Other Web Notes, D&O Diary (Sept. 8, 2007, 3:44 PM), http://dandodiary.blogspot.com/2007/09/more-options-backdating-settlements-and.html (“[F]or the cases that are settling, companies are agreeing to adopt some mild corporate therapeutics, and paying some negotiated amount supposedly corresponding to the amount of the plaintiffs’ attorneys’ fees. The sole benefit to the shareholders on whose behalf the plaintiffs . . . ostensibly proceeded is the ostensible benefit of the corporate therapeutics. I am sure there are skilled advocates who have persuaded themselves, at least, that this process represents something more than a highly stylized form of larceny.”).

277See Mark J. Loewenstein, Shareholder Derivative Litigation and Corporate Governance, 24 Del. J. Corp. L. 1, 2 (1999) (discussing the “substantial benefit” rule).

278See id. at 1–3.

279See LaCroix, supra note 276.

280See, e.g., Daniel Fisher, Lawyer Fights $10 Million Fee for “Cosmetic” Johnson and Johnson Settlement, Forbes (Sept. 4, 2012, 6:40 PM), http://www.forbes.com/sites/danielfisher/2012/09/04/lawyer-fights-10-million-fee-for-cosmetic-johnson-johnson-settlement/.

281See sources cited supra note 276.

282See Barnes & Noble, Inc., Annual Report (Form 10-K), 20 (Apr. 2, 2008).

283Barnes & Noble, Inc., Annual Report (Form 10-K), exhibit 13.1, at F-6 to F-7 (Apr. 4, 2007).

284See Press Release, Barnes & Noble, Inc., Barnes & Noble, Inc. Announces Findings of Special Committee Review of the Company’s Stock Options Practices and Adoption of Remedial Recommendations (Apr. 4, 2007), available at http://www.barnesandnobleinc.com/press_releases/2007_april_4_stock_options_practices.html (detailing wrongdoing that internal corporate investigation uncovered with regard to stock options backdating). Note that these findings constitute only a summary of the SLC’s full report, which was never published or filed with the court.

285See Yonca Ertimur, Fabrizio Ferri & David A. Maber, Reputation Penalties for Poor Monitoring of Executive Pay: Evidence from Option Backdating, 104 J. Fin. Econ. 118, 120 (2012).

286Cf., e.g., Affidavit of Jeffrey N. Gordon in Support of Plaintiffs’ Motion for Preliminary Approval of Derivative Action Settlement at 1, In re Pfizer Inc. S’holder Derivative Litig., 722 F. Supp. 2d 453 (S.D.N.Y. 2010) (No. 09-CV-7822), 2010 WL 5068906 (“[T]he Reforms embodied in the Proposed Settlement will significantly strengthen Board oversight of Pfizer’s compliance with the FDA’s drug marketing regime and related compliance mandates and will produce other improvements to internal compliance and accountability.”).

287Romano, supra note 276, at 63 (commenting that some corporate governance changes resulting from settlements did reflect important reforms, such as changing the board’s majority).

288See Parsons & Tyler, supra note 263, at 504 (“[I]f [the plaintiff] discovered evidence before the final settlement hearing that suggested its claims were stronger than [it] had realized, it could rescind the settlement and litigate post-closing damages.”); see also In re Celera Corp. S’holder Litig., No. 6304-VCP, 2012 WL 1020471, at *8 (Del. Ch. Mar. 23, 2012) (detailing the various steps the court must take in the approval process), aff’d in part, rev’d in part, 59 A.3d 418 (Del. 2012).

289In re Caremark Int’l Inc. Derivative Litig., 698 A.2d 959, 966 (Del. Ch. 1996).

290Id. at 972. In a recent case, the Delaware Court of Chancery examined the therapeutic benefit achieved in light of the numerous cases evaluating supplemental disclosures and reduced the attorneys’ fees from the $500,000 requested to $300,000. See Order, In re Inspire Pharm. Inc. S’holders Litig., No. 6378-VCP, (Del. Ch. Jan. 30, 2012), 2012 WL 275115; Parsons & Tyler, supra note , at 496.

291James D. Cox, Randall S. Thomas & Lynn Bai, There Are Plaintiffs and . . . There Are Plaintiffs: An Empirical Analysis of Securities Class Action Settlements, 61 Vand. L. Rev. 355, 384–85 (2008).

292Exemplary of this are the cases of William Lerach and Melvin Weiss, top plaintiffs lawyers involved in egregious securities class actions scandals. See Jonathan D. Glater, Class-Action Lawyer Given a 30-Month Prison Term for Hiding Kickbacks, N.Y. Times, June 3, 2008, at C3; Michael Parrish, Leading Class-Action Lawyer Is Sentenced to Two Years in Kickback Scheme, N.Y. Times, Feb. 12, 2008, at C3.

293See generally Hillary A. Sale, Judges Who Settle, 89 Wash. U. L. Rev. 377 (2011) (analyzing the role of judges as gatekeepers and monitors of settlements).

294See Bebchuk, supra note 101, at 413; cf. Cooter & Rubinfeld, supra note 257 (providing an economic analysis of discovery).

295Corporate defendants will conduct some offensive discovery on jurisdictional requirements and issues like standing to sue under Fed. R. Civ. P. 23.1. Defendants in securities class actions will seek discovery in connection with a plaintiff’s motion for class certification and to assess damages.

296See, e.g., Kathleen B. Havener, Defensive Strategies in Discovery: A Refresher, A.B.A. (Aug. 30, 2012), http://apps.americanbar.org/litigation/committees/pretrial/email/summer2012/summer2012-0812-defen sive-strategies-discovery-refresher.html (“Ask your client to send you copies of everything remotely related to your claims or defenses and to your opponent’s document requests. After you have reviewed those documents, select those that are responsive, note documents that are not clearly responsive but that may have some significance to your claims or defenses, and remove and log any privileged documents. Then follow up with your client to confirm that nothing has been missed.”).

297See id.

298See Cohen et al., supra note 243, para. 3.132, at 112 (describing extensive e-discovery).

299See generally id.

300See generally Gorga & Halberstam, supra note 56 (discussing the process of generating organizational knowledge and individual knowledge).

301See John C. Coffee, Jr., Understanding Enron: “It’s About the Gatekeepers, Stupid, 57 Bus. Law. 1403, 1405 (2002) (“[G]atekeepers are reputational intermediaries who provide verification and certification services to investors.”).

302See Upjohn Co. v. United States, 449 U.S. 383, 389–90 (1981).

303See, e.g., Cohen v. Telsey, No. 09-2033 (DRD), 2009 WL 3747059, at *17 (D.N.J. Nov. 2, 2009).

304Sarbanes-Oxley Act of 2002 § 307, 15 U.S.C. § 7245 (2012); 17 C.F.R. § 205.3(b) (2013).

305Sewell Chan, At the Sign-Making Shop, No Standing, No Idling, N.Y. Times, Aug. 14, 2005, at 30.

306See, e.g., Stephen M. Cutler, Dir., Div. of Enforcement, SEC, The Themes of Sarbanes-Oxley as Reflected in the Commission’s Enforcement Program, Address at the UCLA School of Law (Sept. 20, 2004), text available at http://www.sec.gov/news/speech/spch092004smc.htm; see also Paul B. Murphy & Lucian E. Dervan, Watching Your Step: Avoiding the Pitfalls and Perils of Corporate Internal Investigations, ALAS Loss Prevention J., Summer 2005, at 2, 6.

307For a historical account of the developments of internal investigations, see Arthur F. Mathews, Internal Corporate Investigations, 45 Ohio St. L.J. 655 (1984); and Jim Hubbell, Note, Discovery of Internal Corporate Investigations, 32 Stan. L. Rev. 1163 (1980).

308See, e.g., Robert Trigaux, Backdating Inquiry Ends, SEC Tells Jabil Circuit, Tampa Bay Times (Nov. 26, 2008, 9:27 AM), http://www.tampabay.com/blogs/venturebiz/content/backdating-inquiry-ends-sec-tells-jabil-circuit/2093484. Jabil’s Form 10-K for the fiscal year ending August 31, 2008, reads as follows:In response to shareholder derivative actions that also were filed in connection with these certain grants, an independent Special Committee of our Board of Directors (the “Special Committee”) was appointed to review the allegations in such actions. . . . The Special Committee concluded that the evidence does not support a finding of intentional manipulation of stock option grant pricing by any member of management. In addition, the Special Committee concluded that it was not in our best interests to pursue the derivative actions. The Special Committee identified certain factors related to our controls surrounding the process of accounting for option grants that contributed to the accounting errors that led to the restatement of our consolidated operations for certain of our previous fiscal years (as further described in the Explanatory Note immediately preceding Part I of our Annual Report on Form 10-K for the fiscal year ended August 31, 2006 and discussed below). Pursuant to the state court’s approval on April 7, 2008, and the federal court’s approval on April 25, 2008, of our proposed settlement of the derivative actions, these actions are no longer pending.Jabil Circuit, Inc., Annual Report (Form 10-K), 14–15 (Oct. 29, 2008), available at http://www.sec.gov/Archives/edgar/data/898293/000119312508219096/d10k.htm.

309See Davis, supra note 43, at 390–405. See generally Corporate Internal Investigations: An International Guide, supra note 53 (giving a comprehensive account of corporate internal investigations in major jurisdictions, including the United States).

310See Davis, supra note 43, at 390.

311Id. at 393.

312Id. at 395.

313Id. at 395–96. (“Unable to pursue each instance of questionable payments on its own, the SEC had leveraged its resources by pressuring corporations to investigate themselves.”).

314Id. at 395.

315418 F. Supp. 508 (S.D.N.Y. 1976).

316See Davis, supra note 43, at 395–98.

317See id. at 400.

318See Brad Mixner, Reducing Discovery Challenges Through Innovative Use of Technology in Cross-Border Litigation, Inside Couns. (March 6, 2014), http://www.insidecounsel.com/2014/03/06/reducing-discovery-challenges-through-innovative-u (“Since its initial beginnings, e-discovery has been transformed drastically, now having an exciting array of technology options for managing data which is exponentially different in variety, scale, source and complexity. This is further complicated by today’s trend toward the globalization of many industries, the proliferation of multi-national companies and the complexities of cross border litigation.”). For countless new e-discovery platforms and technologies, see, for example, eDiscoveryTimes, http://ediscoverytimes.com (last visited May 29, 2014).

319Myers, supra note 46, at 1313. This includes a motion to dismiss for failure to make a demand. SLCs therefore allow defendants to avoid the risk of losing on the issue of demand futility.

320See Douglas M. Branson, The Rule That Isn’t a Rule—The Business Judgment Rule, 36 Val. U. L. Rev. 631, 648 (2002) (describing the formation and nature of the SLC).

321See Richard C. Brown, Shareholder Derivative Litigation and the Special Litigation Committee, 43 U. Pitt. L. Rev. 601, 620–21 (1982). Shareholders may theoretically also sue third parties on behalf of the corporation, but we are not interested in such cases here.

322Plaintiffs may seek “limited discovery” solely on whether the SLC’s members are independent and the investigation was reasonable.

323See, e.g., Auerbach v. Bennett, 408 N.Y.S.2d 83 (App. Div. 1978), aff’d as modified, 393 N.E.2d 994 (N.Y. 1979). The Supreme Court accepted this type of approach in Burks v. Lasker, 441 U.S. 471 (1979), reversing Lasker v. Burks, 567 F.2d 1208 (2d Cir. 1978). Some states have codified the SLC procedure. See, e.g., Wis. Stat. § 180.0744 (2012); Einhorn v. Culea, 2000 WI 65, 235 Wis. 2d 646, 612 N.W.2d 78; see also Ann M. Scarlett, Confusion and Unpredictability in Shareholder Derivative Litigation: The Delaware Courts’ Response to Recent Corporate Scandals, 60 Fla. L. Rev. 589, 596–99 (2008).

324See infra note 325.

325See Auerbach, 393 N.E.2d 994, 1000–03 (explaining that the plaintiff bears the burden of rebutting the business judgment rule presumption with respect to the SLC’s decision, and the judicial inquiry is limited to the disinterestedness and independence of the SLC). Delaware’s less deferential approach in Zapata Corp. v. Maldonado is as follows:First the Court should inquire into the independence and good faith of the committee and the bases supporting its conclusions. Limited discovery may be ordered to facilitate such inquiries. The corporation should have the burden of proving independence, good faith and a reasonable investigation, rather than presuming independence, good faith and reasonableness. 430 A.2d 779, 788 (Del. 1981) (footnote omitted). The Court may then, in its discretion, apply “its own independent business judgment, [in determining] whether the motion should be granted.” Id. at 789. In other words, step two permits a consideration of the substance of the report.

326See Davis, supra note 43, at 398.

327See, e.g., Davis, supra note 46, at 1306 (citing literature).

328Franklin A. Gevurtz, Corporation Law § 4.3.4, at 434 (2d ed. 2010) (“Special litigation committees usually have concluded that derivative suits which the committees looked into were not in the corporation’s best interest.”); see also 9 Louis Loss, Joel Seligman & Troy Paredes, Securities Regulation § 11(C)(4)(b) n.340, at 420 (4th ed. 2013); Jonathan R. Macey, Corporate Governance: Promises Kept, Promises Broken 65–66 (2008); George W. Dent, Jr., The Power of Directors to Terminate Shareholder Litigation: The Death of the Derivative Suit?, 75 Nw. U. L. Rev. 96, 109 & n.70 (1980) (surveying SLC dismissals). But see Myers, supra note 46, at 1311.

329Scarlett, supra note 323, at 598–99 (quoting Lisa M. Fairfax, Spare the Rod, Spoil the Director? Revitalizing Directors’ Fiduciary Duty Through Legal Liability, 42 Hous. L. Rev. 393, 409 (2005)) (internal quotation marks omitted).

330See Myers, supra note 46, at 1314–16; see also Brown, supra note 321, at 620–21 (collecting reported cases in support of the conclusion that “[s]pecial litigation committees, after completing their investigations, have so far always determined that the derivative action was not in the best interests of the corporation and have voted to terminate the suit”); Davis, supra note 46, at 1357 (“[J]udicial deference to the dismissal recommendations of SLCs is the product of a broader recognition of the diminished role of the derivative suit within the portfolio of devices available to hold directors and officers to account.”); Dent, supra note 328, at 109 & n.70 (collecting reported cases on demand in support of the conclusion that in the SLC context “almost invariably, the directors charged with the decision decide to oppose the suit”).

331Myers, supra note 46, at 1320.

332Id. at 1327 tbl.6.

333Id. at 1331. The time to settlement has also been lengthened in securities class actions after the PSLRA. Choi & Thompson, supra note 194, at 1498.

334Myers, supra note 46, at 1331.

335Id. (“[O]nce the SLC decided to settle, the cases were resolved almost immediately.”).

336The SEC guidelines state the same criteria regarding cooperation for crediting the results of a corporate internal investigation as in the derivative context. Exchange Act Release No. 44,969, 76 SEC Docket 220 (Oct. 23, 2001) (asking whether an outside counsel or auditor had “done other work for the company” and whether “management previously engaged such counsel” in considering whether “the company commit[ted] to learn the truth, fully and expeditiously”).

337See, e.g., Ryan v. Gifford, No. 2213-CC, 2007 WL 4259557, at *2–3 (Del. Ch. Nov. 30, 2007).

338This requires an inquiry into the relationship of the SLC members to the company and its management, as well as an inquiry into any relationship between the company and the professionals the SLC has hired to conduct the investigation. See Zapata Corp. v. Maldonado, 430 A.2d 779, 788–79 (Del. 1981).

339Id. at 788.

340Kaplan v. Wyatt, 484 A.2d 501, 510 (Del. Ch. 1984); see also Kindt v. Lund, No. Civ.A. 17751, 2001 WL 1671438, at *1 (Del. Ch. Dec. 14, 2001).

341Zapata Corp., 430 A.2d at 789.

342London v. Tyrrell, No. 3321-CC, 2010 WL 877528, at *17 (Del. Ch. Mar. 11, 2010) (“If the SLC fails to investigate facts or sources of information that cut at the heart of plaintiffs’ complaint this will usually give rise to a material question about the reasonableness and good faith of the SLC’s investigation.”).

343Id. (“To conduct a good faith investigation of reasonable scope, the SLC must investigate all theories of recovery asserted in the plaintiffs’ complaint.”). An SLC may be required to go beyond the allegations in the complaint.

344See generally Zapata Corp., 430 A.2d 779 (holding that the board’s delegation of its managerial authority over litigation to a special committee was effectual).

345See id. at 788–89.

346See Cohen et al., supra note 243, para. 3.177, at 124 (“For the investigative counsel—and the company on whose behalf the investigation is undertaken—the ability to say, at the conclusion of the investigation, that all known relevant documents have been collected and reviewed is critical to the credibility of the investigation.”).

347Alice A. Seebach, Special Litigation Committees: A Practitioner’s Guide, 24 Loy. L.A. L. Rev. 1, 28–30 (1990) (“A broad investigation appears thorough. Courts accepting special committees’ recommendations to dismiss often comment on the number of witnesses interviewed. For example, one special committee that considered claims that a CEO misused a corporation interviewed 140 witnesses ‘throughout the world.’” (quoting Kaplan v. Wyatt , 484 A.2d 501, 511 (Del. Ch. 1984), aff’d 499 A.2d 1184 (Del. 1985))); see also Cohen et al., supra note 243, paras. 3.166–.211, at 121–35.

348See, e.g., London, 2010 WL 877528, at *21 (finding SLC investigation not reasonable, because, inter alia, “Tyrrell’s own emails suggest that he believed these higher internal forecasts where [sic] achievable, in direct contradiction to the testimony he provided the SLC, but the SLC does not appear to have questioned him thoroughly about these emails”).

349See Seebach, supra note 347, at 25 & n.171.

350As Myers’s empirical study finds, the “claim about SLC behavior—that, even if they do not dismiss claims, they may nevertheless take it easy on defendants by failing to pursue claims diligently or by settling claims for nothing— . . . finds no support in the data.” Myers, supra note 46, at 1311.

351See Dent, supra note 236, 99 n.17 (“The purpose of requiring a demand on the board has been stated variously as . . . giving the board the opportunity to settle the dispute without litigation, thus promoting judicial economy . . . .” (citing Winter v. Farmers Educ. & Coop. Union, 107 N.W.2d 226, 233 (Minn. 1961))).

352Apart from the reputational damage, facts discovered in the internal investigation can be used as a basis for additional litigation, including criminal indictments. See Cohen et al., supra note 243, para. 3.173, at 123.

353Id. para. 3.168, at 121–22 (emphasis added).

354See Davis, supra note 43, at 392–95 (describing how “specific credit for developing the SLC itself goes to the Securities and Exchange Commission’s . . . Enforcement Division,” which originated the process to deal with a wave of disclosure violations in the 1970s by affording issuers more flexible relief and conserving its own resources).

355Cohen et al., supra note 243, para. 3.171, at 122.

356As did the Barnes & Noble SLC. See Press Release, Barnes & Noble, supra note 284; see also Cohen et al., supra note 243, para. 3.169, at 122.

357See Myers, supra note 46, at 1332 (“[T]he SLC appears to function as a form of alternative dispute resolution.”).

358See id.

359See, e.g., Press Release, Barnes & Noble, supra note 284.

360Id.

361Indeed, even if an SLC report is not filed with the court, remains confidential, and is only shared with plaintiffs’ counsel subject to a confidentiality agreement, it is often shared with the SEC in an effort to negotiate and assuage the regulator—subject to a “confidential treatment request.” See Ferrara & Khinda, supra note 236, at 1165–67.

362See Parsons & Tyler, supra note 263, at 517 (“[A]s has become common today, multiple plaintiffs file multiple complaints in multiple jurisdictions . . . .”); Thomas & Martin, supra note 214, at 72.

363See supra Part III.A.2. Cases dismissed because the plaintiffs failed to satisfy the heightened pleading requirements under the PSLRA, or the demand requirement in derivative actions, may raise concerns about underdeterrence, but we do not discuss these cases here.

364See, for example, Roth v. Reyes, No. C 06-02786 CRB, 2007 WL 2470122, (N.D. Cal. Aug. 27, 2007), for derivative litigation; and Smajlaj v. Brocade Communications Systems, Inc., No. C 05-02042 CRB, 2007 WL 2457534 (N.D. Cal. Aug. 27, 2007) for securities class action.

365See Murphy & Dervan, supra note 306, at 2.

366Id.

367Id. (quoting Leigh Jones, Call for Internal Probes Growing, Nat’l L.J., Nov. 22, 2004, at 1).

368Lomas & Kramer, supra note 53, at 1–2.

369See, e.g., Litigation Preparedness, E-Discovery Basics (Gibson, Dunn & Crutcher LLP, L.A., Cal.), May 31, 2011, http://www.gibsondunn.com/publications/Documents/E-DiscoveryBasics3-LitigationPreparedness.pdf.

370Id.

371See, e.g., D4 eDiscovery, http://www.d4discovery.com/services-solutions/ (last visited May 29, 2014) (offering comprehensive discovery management across platforms, including discovery best practices and solutions for social media); Managed Services, Xerox Litig. Servs., http://www.xerox-xls.com/ediscovery/managed-services.html?PHPSESSID=4804a414c2ed6c54cf38202f8619841e (last visited May 29, 2014) (offering “expertise in all phases of discovery” to law firms and corporations); see also Top Discovery Software Products, Capterra, http://www.capterra.com/electronic-discovery-software (last visited May 29, 2014).

372See Peter Pepiton II, Becoming Litigation Ready Through Proactive Information Governance 1 (2008)

373See id. at 5. See Sheri Qualters, 30% of Companies Still Lack Policies for Preserving Evidence for Discovery, Nat’l L.J., Dec. 9, 2008 (“[H]aving no legal hold policy is a significant risk factor for companies.” (emphasis removed) (internal quotation mark removed)).

374See, e.g., Sekisui Am. Corp. v. Hart, 945 F. Supp. 2d 494, 509 (S.D.N.Y. 2013) (sanctioning the plaintiff for failing to preserve electronic information on its servers).

375See Gorga & Halberstam, supra note 56, at 1127.

376Marcus, supra note 31, at 749 (quoting Jack H. Friedenthal, A Divided Supreme Court Adopts Discovery Amendments to the Federal Rules of Civil Procedure, 69 Calif. L. Rev. 806, 818 (1981)); see also Geraldine Szott Moohr, The Balance Among Corporate Criminal Liability, Private Civil Suits, and Regulatory Enforcement, 46 Am. Crim. L. Rev. 1459, 1472 (2009).

377Smith v. Van Gorkom, 488 A.2d 858 (Del. 1985), overruled in part on other grounds by Gantler v. Stephens, 965. A.2d 695 (Del. 2009).

378Id. at 863–64.

379Id. at 871.

380Id. at 872.

381See, e.g., Charles M. Elson & Robert B. Thompson, Van Gorkom’s Legacy: The Limits of Judicially Enforced Constraints and the Promise of Proprietary Incentives, 96 Nw. U. L. Rev. 579, 584 (2002) (arguing that Van Gorkom’s legacy is the “rise of elaborate decision-making procedures involving lengthy meetings [and] voluminous documentation . . . that today accompany board decisions”); Lynn A. Stout, In Praise of Procedure: An Economic and Behavioral Defense of Smith v. Van Gorkom and the Business Judgment Rule, 96 Nw. U. L. Rev. 675, 676–77 (2002) (defending the business judgment rule’s emphasis on process after Van Gorkom).

382See generally Bainbridge, supra note 229, at 393–405 (discussing the fate of shareholder derivative suits).

383To be precise, the proceedings before the Delaware Chancery generated the extensive factual record that the Delaware Supreme Court ultimately relied on in issuing its opinion. See Van Gorkom, 488 A.2d at 864–70.

384See id.

385Id. at 866.

386Id.

387Id. at 865, 867 (“According to Romans: They did not ‘come up’ with a price for the Company. They merely ‘ran the numbers’ at $50 a share and at $60 a share with the ‘rough form’ of their cash figures at the time. Their ‘figures indicated that $50 would be very easy to do but $60 would be very difficult to do under those figures.’”). This evidence emerged during Romans’s depositions.

388Id. at 867.

389Id. at 877.

390Id. at 874; see also id. at 866 (“Apart from the Company’s historic stock market price, and Van Gorkom’s long association with Trans Union, the record is devoid of any competent evidence that $55 represented the per share intrinsic value of the Company.” (footnote omitted)).

391Id. at 869.

392Id.

393See, e.g., id. at 864 n.1 (“[F]ollowing extensive discovery, the Trial Court denied the plaintiff’s motion for preliminary injunction . . . .”).

394Bundesgerichtshof [BGH] [Federal Court of Justice] Dec. 21, 2005, Neue Juristische Wochenschrift [NJW] 522, 2006 (Ger.); see also, Franklin A. Gevurtz, Disney in a Comparative Light, 55 Am. J. Comp. L. 453, 459–62 (2007) (discussing the case in English).

395Gevurtz, supra note 394, at 461.

396Id. at 455.

397In order to avoid de novo review of the law on appeal, trial courts have an incentive to heavily emphasize the facts, which are subject to a “clear error” standard on appeal.

398Van Gorkom, 488 A.2d at 871.

399911 A.2d 362 (Del. 2006).

400964 A.2d 106 (Del. Ch. 2009).

401See Parsons & Tyler, supra note 263, at 479 (discussing the “importance of the Delaware Court of Chancery’s procedural rules to Delaware’s substantive law of corporations,” in the most recent spate of shareholder challenges to mergers and acquisitions). Parsons and Tyler discuss the use of motions to expedite discovery to sort out meritorious from unmeritorious cases. Id. at 499–500.

402Del. Code Ann. tit. 8, § 220 (West Supp. 2014).

403Stephen A. Radin, The New Stage of Corporate Governance Litigation: Section 220 Demands—Reprise, 28 Cardozo L. Rev. 1287, 1412 (2006).

404Thomas & Martin, supra note 214, at 70.

405634 A.2d 927 (Del. 1993).

406673 A.2d 1207 (Del. 1996).

407Id. at 1216 n.11 (quoting Rales, 634 A.2d at 934–35 n.10).

408687 A.2d 563 (Del. 1997).

409Radin, supra note 403, at 1294–95 (footnote omitted) (quoting Sec. First Corp., 687 A.2d at 570).

410746 A.2d 244, 266–67 (Del. 2000) (internal quotation marks omitted).

411See John F. Grossbauer & Charles T. Williams, III, The Increasing Role of Delaware Books and Records Demands in Compensation and Governance Disputes, Potter Anderson Corroon LLP (Nov. 1, 2004), http://www.potteranderson.com/publication/the-increasing-role-of-delaware-books-and-records-deman ds-in-compensation-and-governance-disputes (describing the use of section 220 in the context of executive compensation and proxy contests).

412Id.

413Radin, supra note 403, at 1339.

414Brehm, 746 A.2d at 266.

415Id.

416806 A.2d 113 (Del. 2002).

417Id. at 115.

418Id.

419Id. at 116.

420Id. at 117.

421See id. at 116.

422See Radin, supra note 403, at 1292 n.33.

423See McKesson Corp. v. Saito, 818 A.2d 970 (Del. Ch. 2003) (citing Saito, 806 A.2d 113).

424Francis G.X. Pileggi, Kevin F. Brady & Jill Agro, Inspecting Corporate “Books and Records” in a Digital World: The Role of Electronically Stored Information, 37 Del. J. Corp. L. 163, 171–72 (2012) (arguing for a “readily accessible” standard for the restoration of back-up tapes in response to a “books and records” request, instead of the more demanding “reasonably accessible” standard under Fed. R. Civ. P. 26 (citing Zubulake v. UBS Warburg LLC, 220 F.R.D. 212, 218 (S.D.N.Y. 2003))).

425See, e.g., Alan Palmiter & Frank Partnoy, Corporations: A Contemporary Approach 681–85 (2010); Coffee, supra note 301, at 1404 n.6; Gordon, supra note 233, at 686 (“Part of what fuels the sense that the executive compensation-setting process is seriously flawed are the high-profile cases of exceptionally large payouts or mega-stock option grants, particularly where the firm’s subsequent performance is subpar, if not disastrous.”).

426See David I. Walker, Unpacking Backdating: Economic Analysis and Observations on the Stock Option Scandal, 87 B.U. L. Rev. 561, 563 (2007).

427See, e.g., M.P. Narayanan, Cindy A. Schipani & H. Nejat Seyhun, The Economic Impact of Backdating of Executive Stock Options, 105 U. Mich. L. Rev. 1597, 1599, 1638 (2007); Eric Dash, Report Estimates the Costs of a Stock Options Scandal, N.Y. Times, Sept. 6, 2006, at C4; Mark Maremont & Charles Forelle, Bosses’ Pay: How Stock Options Became Part of the Problem, Wall St. J. (Dec. 27, 2006, 12:01 AM), http://online.wsj.com/news/articles/SB116718927302760228.

428See Perfect Payday: Options Scorecard, Wall St. J. (Sept. 4, 2007), http://online.wsj.com/public/resources/documents/info-optionsscore06-full.html (documenting the 2006–2007 stock options backdating probes, revelations, and outcomes on a company for company basis).

429See generally Alan R. Palmiter, Toward Disclosure Choice in Securities Offerings, 1999 Colum. Bus. L. Rev. 1, 17.

430See supra Part III.A.

431See supra note 41.

432See Coffee, supra note 13, at 245 (“In the United States, public enforcement of law is supplemented by a vigorous, arguably even hyperactive, system of private enforcement. Relying on class actions and an entrepreneurial plaintiffs’ bar motivated by contingent fees, the U.S. system of private ‘enforcement by bounty hunter’ appears in fact to exact greater annual aggregate sanctions than do its public enforcers. This system has no true functional analogue anywhere else in the world.”).

433See, e.g., Jason Krause, Government Agencies Look Within to Solve E-Discovery Woes, Law Tech. News, Aug. 1, 2011, at 56 (“[T]here is a misperception that government agencies have unlimited resources to shoulder the burden of electronic data discovery costs in big cases like [the antitrust investigation of Google] . . . . Government litigators often manage EDD without proper software.”); Melanie Rodier, Lack of SEC Resources Could Be Biggest Hurdle for Volcker Rule, Wall St. & Tech. (Jan. 21, 2011), http://wallstreetandtech.com/articles/229100012/.

434See, for example, the dramatic difference between the SEC’s $150 million settlement with Bank of America regarding its acquisition of Merrill Lynch, compared with the $2.43 billion settlement shareholders ultimately achieved. Silver-Greenberg & Craig, supra note 255.

435See supra note 65 and accompanying text.

436See SEC Enforcement Manual § 4.3.1 (2013), available at http://www.sec.gov/divisions/enforce/enforcementmanual.pdf (describing confidential treatment requests).

437There is empirical evidence on the effect of filings on share prices. See Gregg A. Jarrell & Annette B. Poulsen, Stock Trading Before the Announcement of Tender Offers: Insider Trading or Market Anticipation?, 2 J.L. Econ. & Org. 225, 228–29 (1989); Gregg A. Jarrell, James A. Brickley & Jeffry M. Netter, The Market for Corporate Control: The Empirical Evidence Since 1980, J. Econ. Persp., Winter 1988, at 49, 52.

438See Jessica Erickson, Corporate Misconduct and the Perfect Storm of Shareholder Litigation, 84 Notre Dame L. Rev. 75, 93–94 (2008) (recognizing that federal laws have dramatically affected plaintiffs’ access to information in securities litigation, limiting many to only self-obtained information).

43915 U.S.C. § 78u-4(b)(1)–(3) (2012) (including scienter and other specific pleading requirements). Choi and Thompson argued that heightened pleading and other elements of the PSLRA “all work to increase the costs for plaintiffs’ attorneys seeking to bring a class action” and have proven to be a difficult hurdle without access to witnesses or documents. See Choi & Thompson, supra note 194, at 1529; see also 15 U.S.C. §§ 77z-1(b)(1), 78u-4(b)(3)(B) (providing a mandatory discovery stay during the pendency of motion to dismiss).

440See supra notes 206–14 and accompanying text.

441See, e.g., Wayne D. Brazil, The Adversary Character of Civil Discovery: A Critique and Proposals for Change, 31 Vand. L. Rev. 1295, 1330 (1978).

442See id. at 1330–31.

443See Steven S. Gensler, Judicial Case Management: Caught in the Crossfire, 60 Duke L.J. 669, 705 (2010) (recognizing that any reform that would give judges “more case-management power” must be designed so as to respond to critics); see also Judith Resnik, Managerial Judges, 96 Harv. L. Rev. 374, 378 (1982).

444Steven S. Gensler & Lee H. Rosenthal, Managing Summary Judgment, 43 Loy. U. Chi. L.J. 517, 520 (2012); see also Gensler, supra note 443, at 680–81.

445See David Ferris, Setting Retention Policy for Electronic Information 2 (2011), available at http://www.ironmountain.com/Knowledge-Center/Reference-Library/View-by-Document-Type/White-Papers-Briefs/S/Setting-Retention-Policy-for-Electronic-Information.aspx (“Producing information in response to an eDiscovery request can be extremely time-consuming and costly. If the material has been deleted, the costs of production are obviously reduced. Following a clear, communicated retention policy adds defensibility to the deletion of electronic information.”).

446Susan Beck, Susan Beck’s Summary Judgment: Stoker Jury Foreman Explains How Verdict Didn’t Absolve Citi, Am. Law. Litig. Daily (Aug. 1, 2012, 12:00 AM), http://www.litigationdaily.com/id=1202565746363/Susan-Beck’s-Summary-Judgment:-Stoker-Jury-Foreman-Explains-How-Verdict-Didn’t-Absolve-Citi?slreturn=20140105010532.

447Milhaupt & Pistor, supra note 5, at 40.

448Most academics agree there are problems but defend securities class actions as a necessary mechanism for shareholder monitoring and general deterrence. See, e.g., Coffee, supra note 13, at 245; Jill E. Fisch, Confronting the Circularity Problem in Private Securities Litigation, 2009 Wis. L. Rev. 333, 345–48; Merritt B. Fox, Why Civil Liability for Disclosure Violations When Issuers Do Not Trade?, 2009 Wis. L. Rev. 297, 310–18; Lawrence E. Mitchell, The “Innocent Shareholder”: An Essay on Compensation and Deterrence in Securities Class-Action Lawsuits, 2009 Wis. L. Rev. 243, 243. These academics are largely responding to new criticism that today’s securities class actions yield little compensation or deterrence. See John C. Coffee, Jr., Reforming the Securities Class Action: An Essay on Deterrence and Its Implementation, 106 Colum. L. Rev. 1534, 1534 (2006).

449Choi & Thompson, supra note 194, at 1491.

450Id. at 1496.

451Id. at 1497–98.

452Id. at 1498.

453Stephen J. Choi, Karen K. Nelson & A.C. Pritchard, The Screening Effect of the Private Securities Litigation Reform Act, 6 J. Empirical L. Stud. 35, 35–36 (2009) (“Trials . . . are almost unheard of in this area . . . .”).

454Id. at 35 (“There is evidence . . . that pre-PSLRA nonnuisance claims would be less likely to be filed under the PSLRA regime. The latter result, which we refer to as the screening effect, is particularly pronounced for claims lacking hard evidence of securities fraud or abnormal insider trading.”). This raises concerns about the balance that has been struck by the PSLRA.

455See id. at 38. Actions claiming fraud must generally satisfy the heightened pleading requirements under Fed. R. Civ. P. 8(a), requiring plaintiffs to plead the circumstances surrounding the fraud with specificity—but not intent. The PSLRA additionally requires that plaintiffs allege facts that support a “strong inference” of scienter. 15 U.S.C. § 78u-4(b)(2) (2012).

456“[T]he intertwining of 10b-5 violations and breaches of fiduciary duty . . . pervades most derivative actions against directors where a transfer of securities is involved . . . .” Kurt M. Swenson, Remedies for Private Parties Under Rule 10b-5, 10 B.C. Indus. & Com. L. Rev. 337, 355–56 (1969) (discussing Mutual Shares Corp. v. Genesco, Inc., 384 F.2d 540 (2d Cir. 1967)). This is also why SLUSA was passed in the wake of the PSLRA. Plaintiffs, however, have been able to adjust by bringing claims in the form of shareholder derivative actions in state court.

457See, e.g., Perfect Payday: Options Scorecard, supra note 428.

458Id.

459Id.

460Resnik, supra note 443, at 378; see also, Richard L. Marcus, Reining in the American Litigator: The New Role of American Judges, 27 Hastings Int’l & Comp. L. Rev. 3, 17–20 (2003); Resnik, supra note 105, at 497–98.

461Moreover, deadlines are not infrequently extended.

462See infra notes 524–39 and accompanying text. Civil law countries are introducing small changes in light of their increasing contact with the U.S. system. See Huang, supra note 24, at 160 (arguing for conversion of procedural rules). But see, e.g., Verkerk, supra note 23, at 370 (“A remarkable conclusion is that the differences between the fact-finding arrangements in Continental and American jurisdiction are larger today than they were a century ago.”).

463See 200 F.3d 335, 340–41 (5th Cir. 2000).

464Id. at 337–38.

465Id. at 338.

466See id. The litigation involved a large plaintiff class, with each plaintiff suffering from different symptoms, occurring during different time frames, in different places, purportedly caused in different ways by one or more of a large class of diverse defendant companies engaged in a variety of different activities at different times, and in different places. Id. at 340.

467Rowe, Sherry & Tidmarsh, supra note 187, at 64.

468Id.

469See Joe S. Cecil et al., Fed. Judicial Ctr., Motions to Dismiss for Failure to State a Claim After Iqbal: Report to the Judicial Conference Advisory Committee on Civil Rules, at vii, 12 (2011).

470See supra note 42.

471See Beisner, supra note 33, at 1–2 (internal quotation marks omitted).

472Id.; see also Final Report on the Joint Project of the American College of Trial Lawyers Task Force on Discovery and the Institute for the Advancement of the American Legal System 9 (2009), available at http://www.actl.com/AM/Template.cfm?Section=Home&template=/CM/ContentDisplay.cfm&ContentID=4053; Henry S. Noyes, Good Cause Is Bad Medicine for the New E-Discovery Rules, 21 Harv. J.L. & Tech. 49, 67–68 (2007) (“[E]-discovery is more time-consuming, more burdensome, and more costly than conventional discovery.”); Martin H. Redish, Electronic Discovery and the Litigation Matrix, 51 Duke L.J. 561, 592 (2001).

473 See infra notes 491–95.

474See, e.g., William H. Erickson, The Pound Conference Recommendations: A Blueprint for the Justice System in the Twenty-First Century, 76 F.R.D. 277, 288 (1978) (observing that “[w]ild fishing expeditions . . . seem to be the norm,” and lamenting the “[u]necessary intrusions into the privacy of the individual, high costs to the litigants, and correspondingly unfair use of the discovery process as a lever toward settlement” that had come to characterize the American legal system).

475Subrin, supra note 111, at 706–07, 728.

476See, e.g., Ronald J. Allen, Idealization and Caricature in Comparative Scholarship, 82 Nw. U. L. Rev. 785, 786 (1988) (discussing Langbein’s criticism of depositions and citing contrary evidence).

477Another strategy is to take a first deposition of key witnesses early in the case to lock them into positions before defensive discovery allows them to assess how far they can bend the truth without their testimony being later challenged and contradicted by documents and other witness testimony, but to reserve the right to reexamine the witness again at a later date.

478By contrast, in civil law systems, witness examinations occur issue by issue in an order entirely determined by the judge. A judge likely would take the most important witnesses first, given that the judge is always looking for dispositive issues. See Langbein, supra note 23, at 830, 846.

479See generally Philip J. Favro, A New Frontier in Electronic Discovery: Preserving and Obtaining Metadata, 13 B.U. J. Sci. & Tech. L. 1 (2007) (discussing the hazards of metadata and recent developments in the case law requiring production of metadata).

480See, for example, the deposition of Timothy J. Mayopoulos, BofA’s associate general counsel in the Merrill Lynch case. See supra note 249 and accompanying text.

481See Marcus, supra note 31, at 750 (“One of the striking things about American discovery is that parties often do reveal damaging information, and the importance of that discovered material in litigation cannot be overstated.”); see also H. Christopher Boehning & Daniel J. Toal, More Than a Security Risk: Director E-mails in Discovery, N.Y.L.J., Apr. 3, 2012 (discussing the vulnerability to discovery of electronic communications among directors).

482See generally Lee H. Rosenthal, A Few Thoughts on Electronic Discovery After December 1, 2006, 116 Yale L.J. Pocket Part 167 (2006), http://www.thepocketpart.org/images/pdfs/82.pdf (discussing issues likely to arise under new discovery laws and potential challenges for lawyers, litigants, and judges).

483See Brehm v. Eisner, 746 A.2d 244, 259 (Del. 2000) (“[I]n making business decisions, directors must consider all material information reasonably available, and . . . the directors’ process is actionable only if grossly negligent.”).

484See generally Richard L. Marcus, Confronting the Future: Coping with Discovery of Electronic Material, Law & Contemp. Probs., Spring/Summer 2001, at 253; Richard Marcus, Only Yesterday: Reflections on Rulemaking Responses to E-Discovery, 73 Fordham L. Rev. 1 (2004).

485See, e.g., The Chief Judge’s Task Force on Commercial Litig. in the 21st Century, Report and Recommendations to the Chief Judge of the State of New York 19 (2012), available at http://www.nycourts.gov/courts/comdiv/PDFs/ChiefJudgesTaskForceOnCommercialLitigationInThe21stpdf.pdf.

486See James H.A. Pooley & David M. Shaw, Finding Out What’s There: Technical and Legal Aspects of Discovery, 4 Tex. Intell. Prop. L.J. 57, 64 (1995) (“There are sophisticated computer programs that can retrieve e-mail messages and computer files from the guts of the computer system long after those messages and files were thought to have been forever deleted. . . . [T]he mere fact that files were destroyed, perhaps systematically, could prove to be more valuable information than the actual files retrieved, especially if such fact is used to prove a cover-up, for example.”).

487See, e.g., Sekisui Am. Corp. v. Hart, 945 F. Supp. 2d 494, 509–10 (S.D.N.Y. 2013) (imposing sanctions in the form of monetary damages and an adverse inference jury instruction where a party destroyed electronic documents); Pension Comm. of the Univ. of Montreal Pension Plan v. Banc of Am. Sec., LLC, 685 F. Supp. 2d 456, 469–72 (S.D.N.Y. 2010) (discussing cases involving sanctions for nonproduction of evidence), abrogated by Chin v. Port Auth. of N.Y & N.J., 685 F.3d 135 (2d Cir. 2012).

488See generally Dodd-Frank Wall Street Reform and Consumer Protection Act, Pub. L. No. 111-203, 124 Stat. 1376 (2010) (codified in scattered sections of 12 and 15 U.S.C.).

489See supra note 42.

490Beisner, supra note , at 2 & n.7 (citing Thomas E. Willging et al., Fed. Judicial Ctr., Discovery and Disclosure Practice, Problems, and Proposals for Change: A Case-based National Survey of Counsel in Closed Federal Civil Cases 15 tbl. 4 (1997), available at https://bulk.resource.org/courts.gov/fjc/discovry.pdf; see also H.R. Rep. No. 104-369, at 37 (1995) (Conf. Rep.) (“[D]iscovery costs account for roughly 80% of total litigation costs in securities fraud cases” (quoting the general counsel of an investment bank)); Judicial Conference Adopts Rules Changes, Confronts Projected Budget Shortfalls, Third Branch (Admin. Office U.S. Cts., D.C.), Oct. 1999, at 2–3 (“Discovery represents 50 percent of the litigation costs in the average case and up to 90 percent of the litigation costs in cases in which it is actively used.”), available at http://www.uscourts.gov/News/TheThirdBranch/99-10-01/Judicial_Conference_Adopts_Rules_Changes_Confronts_Projected_Budget_Shortfalls.aspx). But note that the 90% number is highly misleading. The report cited in the Chamber of Commerce report stated that, in about 15% of cases, there was no discovery, that discovery costs on average constituted 3% of the value of a dispute, and that in a small number of large cases, 90% of litigation-related costs are incurred during discovery. Willging et al., supra, at 15 tbl. 4, 17 tbl. 6. That 90%, however, notably does not consider the value of the claim or dispute. Thus, even if discovery costs constituted 90% of the litigation costs, these costs quite likely reflect a very small percentage (i.e., in the single digits) of the value of the dispute.

491Separate empirical studies by the Federal Judicial Center in 1996 and the RAND Corporation in 1993 found that only a small minority of cases presented significant discovery problems, but that, among other findings, 54% of attorneys thought that the expense of discovery was about right in relation to the stakes of the case, 15% thought the expenses too high, 20% thought them too low, and the rest had no opinion. Rowe, Sherry & Tidmarsh, supra note 187, at 118 (reviewing empirical literature).

492Reda, supra note 34, at 1111; see also Lawyers for Civil Justice et al., Litigation Cost Survey of Major Companies, Presentation at the Duke Law School 2010 Conference on Civil Litigation 2 (May 10, 2010), http://www.uscourts.gov/uscourts/RulesAndPolicies/rules/Duke%20Materials/Library/Litigation%20Cost%20Survey%20of%20Major%20Companies.pdf (“[F]ew empirical studies document the costs and utility of discovery.”).

493Reda, supra note 34, at 1104.

494Emery G. Lee III & Thomas E. Willging, Fed. Judicial Ctr., Preliminary Report to the Judicial Conference Advisory Committee on Civil Rules 42 (2009), available at http://www.fjc.gov/public/pdf.nsf/lookup/dissurv1.pdf/$file/dissurv1.pdf.

495Willging et al., supra note 490, at 17 tbl. 6 (1997); see also Linda S. Mullenix, The Pervasive Myth of Pervasive Discovery Abuse: The Sequel, 39 B.C. L. Rev. 683, 684 (1998) (“[D]iscovery data have remained surprisingly constant over time.”).

496See supra notes 44–46 and accompanying text. Indeed, this is also the law governing the judicial approval of settlements.

497See supra Part III.A.

498Stout, supra note 42, at 712–13.

499Id. at 713–14.

500Stout notes that corporate losses from defending against meritless strike suits were described at the time as being in the hundreds of millions of dollars annually. But if one considers the possibility of a mere 1% discount applied by investors to shares of public companies, based on investor concern of underdeterrence (or Type II errors), the combined losses to the securities markets could surmount $100 billion. Id. See also supra notes 194–202 and accompanying text for a discussion of Type I and Type II errors.

501Stout, supra note 42, at 713–14.

502Supra note 291 and accompanying text.

503See, e.g., Jackson & Roe, supra note 13.

504See supra Part III.A; see also supra note 44.

505See, e.g., Coffee, supra note 17, at 292.

506See Armour et al., supra note 13, at 688–89.

507Id. at 690.

508Id. at 687.

509Id. at 689.

510Id. at 695.

511Id. at 690.

512Id. at 721.

513Id.

514Id. at 695.

515Id. at 722.

516Id. at 708 tbl. 6.

517See id.; see also id. at 707 (“[O]ur primary concern is with director liability.”).

518Id. at 708.

519Id.

520Id. at 709.

521See id. at 708 tbl. 6.

522See supra notes 2–13 and accompanying text.

523See, e.g., Scott Dodson & James M. Klebba, Global Civil Procedure Trends in the Twenty-First Century, 34 B.C. Int’l & Comp. L. Rev. 1, 1–2 (2011) (“In a number of areas, civil procedure practices in the United States differ significantly from the rest of the world, particularly the civil law world. Notable differences include the use of civil juries, the prevalence of partisan experts paid for directly by the litigants, the existence and extent of party-controlled pre-trial discovery, standards for second instance (or appellate) review, notions of finality in litigation (res judicata), the use of class actions and other forms of aggregate litigation, pleading requirements, and the role and status of the judge. Together, these individual attributes have given rise to a holistic assessment that U.S. civil procedure is highly exceptionalist when compared to the civil law systems in the rest of the world.” (footnotes omitted)).

524See Hazard, Discovery and the Role of the Judge, supra note 22, at 1022; Langbein, supra note 23, at 824.

525Hazard, Discovery and the Role of the Judge, supra note 22, at 1021.

526Id. at 1022; see also Langbein, supra note 23, at 826.

527See Langbein, supra note 23, at 830.

528Hazard, Discovery and the Role of the Judge, supra note 22, at 1022; Langbein, supra note 23, at 829 (“The process merges the investigatory function of our pretrial discovery and the evidence-presenting function of our trial.”).

529See Langbein, supra note 23, at 831.

530Huang, supra note 24, at 40 (“One important cause of incomplete information for fact-finding [in civil law systems] is the misalignment of authority, burden, and incentives.”). Commentators have responded to the incentives argument by pointing to the quality of a country’s judiciary. See, e.g., Langbein, supra note 23, at 848. But Langbein explicitly excluded “Big Case[s]” from his analysis, and Germany’s system may be hard to replicate. Id. at 858.

531See Verkerk supra note 23, at 16.

532See id. at 66–73.

533Id. at 68.

534See, e.g., Lei No. 5.869, de 11 de Janeiro de 1973, Código de Processo Civil [C.P.C.]: art. 363 de 11.1.1973 (Braz.).

535See, e.g., id.; see also Huang, supra note 24, at 40.

536“Relevance” has only recently appeared as a factor, and some “legal right” to the document that can override the privacy and property interests of the owner must be shown. Compare the discarded doctrine of Boyd v. United States, 116 U.S. 616, 622 (1886) (holding that “a search and seizure [was] equivalent [to] a compulsory production of a man’s private papers” and that the search was “an ‘unreasonable search and seizure’ within the meaning of the Fourth Amendment”).

537Murray & Stürner, supra note 25, at 277 (translating Zivilprozessordnung [ZPO] §§ 421 ff., 427).

538Lei No. 5.869, de 11 de Janeiro de 1973, Código de Processo Civil [C.P.C.]: art. 356 de 11.1.1973 (Braz.) (emphasis added).

539Conversation with a government attorney in the Office of General Counsel for the Mayor of the City of São Paulo, in São Paulo, Braz. (Jan. 20, 2012).

540Bernard S. Black, The Legal and Institutional Preconditions for Strong Securities Markets, 48 UCLA L. Rev. 781, 786 (2001).

541See, e.g., Theodor Baums, Changing Patterns of Corporate Disclosure in Continental Europe: The Example of Germany (European Corporate Governance Inst., Working Paper No. 04, 2002), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=345020.

542Black, supra note 540, at 786. In fact, as we have shown, they can, but ex post through litigation discovery, an institution that Black does not mention specifically.

543See Érica Gorga, Changing the Paradigm of Stock Ownership from Concentrated Towards Dispersed Ownership? Evidence from Brazil and Consequences for Emerging Countries, 29 Nw. J. Int’l L. & Bus. 439, 474 (2009) (noting disclosure inconsistencies regarding the number of shares bound by shareholders agreements, and the validity of these agreements on the Brazilian Securities Exchange Commission (CVM) website).

544See id.

545For evidence that litigation constrains managers from making opportunistic reporting choices, see Justin Hopkins, Private Enforcement of Securities Laws and Financial Reporting Choices 5 (2012) (unpublished manuscript), available at http://ssrn.com/abstract=1872068.

546See supra notes 524–30 and accompanying text.

547See supra notes 524–30 and accompanying text.

548See supra notes 524–30 and accompanying text.

549See J. Maria Glover, The Structural Role of Private Enforcement Mechanisms in Public Law, 53 Wm. & Mary L. Rev. 1137, 1140 (2012) (arguing for the centrality of private enforcement to the American regulatory system, but failing to consider issues of information production or litigation discovery).

550See Gorga, supra note 543, at 486–87.

551See, e.g., Coffee, supra note 448 (proposing a variety of ways in which securities class action penalties would fall less on the corporation and more on those who are truly culpable); Samuel Issacharoff & Geoffrey P. Miller, Will Aggregate Litigation Come to Europe?, 62 Vand. L. Rev. 179 (2009) (discussing various examples of securities class actions).

552See, e.g., Siems, supra note 13 (exploring the availability of derivative actions across countries).

553See, e.g., Christopher Hodges, The Reform of Class and Representative Actions in European Legal Systems: A New Framework for Collective Redress in Europe (2008); Richard A. Nagareda, Aggregate Litigation Across the Atlantic and the Future of American Exceptionalism, 62 Vand. L. Rev. 1, 19–37 (2009) (providing comparative analysis of recent developments in Europe regarding aggregate litigation).

554See, e.g., The Law and Economics of Class Actions in Europe: Lessons from America (Jürgen G. Backhaus, Alberto Cassone & Giovanni B. Ramello eds., 2012); Issacharoff & Miller, supra note 551, at 179 (“Once decried as the perversity of rapacious Americans, class actions are now the focus of significant reform efforts in many European countries and even at the level of the European Union.”).

555See Coffee, supra note 17, at 291–93.

556See Nagareda, supra note 553, at 21–25.

557Kapitalanleger-Musterverfahrensgesetz [KapMuG] [Capital Market Investors’ Model Proceeding Act], Aug. 16, 2005, Bundesgesetzblatt, Teil I [BGBl. I] 2437, as amended, Gesetz [G], Oct. 19, 2012, BGBl I 2182 (Ger.), translated at http://www.bmj.bund.de/kapmug.

558Nils-Viktor Sorge, Historic Lawsuit: Angry Shareholders Take on Deutsche Telekom in Court Showdown, Spiegel Online Int’l (Apr. 7, 2008, 2:44 PM), http://www.spiegel.de/international/business/historic-lawsuit-angry-shareholders-take-on-deutsche-telekom-in-court-showdown-a-545827.html. The KapMus has been called “Lex Telekom,” because it was passed to provide an enforcement mechanism for outraged German investors. See Axel Halfmeier, Recht und Kapitalmarkt: Von der Lex Telekom zum Instrument des Rechtsschutzes, Börsen-Zeitung, Nov. 8, 2010, available at https://www.boersen-zeitung.de/index.php?li=1&artid=2010152040.

559Sorge, supra note 558.

560See In re Deutsche Telekom AG Sec. Litig., 229 F. Supp. 2d 277, 278 (S.D.N.Y. 2002).

561See OLG Frankfurt a.M.: Musterentscheid im KapMuG-Verfahren gegen die Deutsche Telekom, Ebner Stolz (May 16, 2012), http://www.ebnerstolz.de/de/olg-frankfurt-a-m-musterentscheid-im-kapmug-verfahren-gegen-die-deutsche-telekom-15136.html (“Nach den Angaben der Zeugen, zu denen auch die ehemaligen Vorstandsvorsitzenden der Telekom Dr. Ron Sommer und Kai-Uwe Ricke gehörten, war das Geschäft erst Ende Juli 2000 und somit deutlich nach der Erstnotiz am 19.6.2000 abschließend und entscheidungsreif verhandelt gewesen.”). In the settlement agreement of the U.S. securities class action, plaintiffs’ claim that “the Prospectus was materially false and misleading because despite the fact . . . that Deutsche Telekom was engaged in serious and advanced negotiations to acquire VoiceStream, the Prospectus merely disclosed that Deutsche Telekom was ‘actively considering and discussing a number of potential acquisition transactions.’” Notice of Proposed Settlement of Class Action, Motion for Attorneys’ Fees and Settlement Fairness Hearing at 4, In re Deutsche Telekom AG Sec. Litig., No. 00-CV-9475 (S.D.N.Y. 2002), available at http://securities.stanford.edu/filings-documents/1016/USD00/200544_r02n_00CV9475.pdf.

562OLG Frankfurt a.M.: Musterentscheid im KapMuG-Verfahren gegen die Deutsche Telekom, supra note 561.

563Bad Connection, Economist, Apr. 12–18, 2008, at 72, 72 (internal quotation marks omitted), available at http://www.economist.com/node/11021139.

564 Id.

565Markus Zydra, Wende im Telekom-Prozess, Suddeutsche.de (May 17, 2010, 9:39 PM), http://www.sueddeutsche.de/wirtschaft/gericht-fordert-us-unterlagen-wende-im-telekom-prozess-1.574782.

566Id.

567In re Application of Schmitz, 259 F. Supp. 2d 294 (2003), aff’d sub nom Schmitz v. Bernstein Liebhard & Lifshitz, LLP, 376 F.3d 79 (2d Cir. 2004).

568Schmitz, 376 F.3d at 82.

569See Stipulation and Agreement of Settlement at 11, In re Deutsche Telekom AG Sec. Litig., 229 F. Supp. 2d 277 (S.D.N.Y. 2002) (No. 00-CV-9475 (SHS)), available at http://securities.stanford.edu/filings-documents/1016/USD00/2005128_r04s_00CV9475.pdf. The parties settled for $120 million without DT admitting any wrongdoing. Id. at 4.

570See, e.g., Guido Ferrarini & Paolo Giudici, Financial Scandals and the Role of Private Enforcement: The Parmalat Case, in After Enron: Improving Corporate Law and Modernising Securities Regulation in Europe and the US 159, 201 (John Armour & Joseph A. McCahery eds., 2006) (“Given the lack of efficient discovery rules, investor action against mass wrongdoing is virtually impossible in Italy as it is in the rest of Europe, unless information is gathered by public authorities.”); Paolo Giudici, Representative Litigation in Italian Capital Markets: Italian Derivative Suits and (If Ever) Securities Class Actions, 6 Eur. Company & Fin. L. Rev. 246, 254 (2009) (“Italian law does not grant any inspection right to shareholders of public companies.”).

571Axel Halfmeier et al., Evaluation des Kapitalanleger-Musterverfahrensgesetzes: Forschungsvorhaben im Auftrag des Bundesministeriums der Justiz 50 (2009), available at http://www.bmj.de/SharedDocs/Downloads/DE/pdfs/Abschlussbericht_KapMuG_Frankfurt%20School_2009.pdf?__blob=publicationFile; see also Daniela Kuhr, Sammelklagen Floppen in Deutschland, Suddeutsche.de (Apr. 24, 2012, 10:07 AM), http://www.sueddeutsche.de/wirtschaft/telekom-prozess-und-seine-folgen-sammelklagen-floppen-in-deutschland-1.1340101.

572Halfmeier et al., supra note 571, at 50.

573Id. at 3.

574See Kuhr, supra note 571.

575See Huang, supra note 24, at 157.

576See id.

577See id. at 157–216 (describing the Japanese discovery regime).

578Comm. on Rules of Practice and Procedure of the Judicial Conference of the U.S., Preliminary Draft of the Proposed Amendments to the Federal Rules of Bankruptcy and Civil Procedure (Aug. 15, 2013) [hereinafter Proposed Rule Changes], available at http://www.uscourts.gov/uscourts/rules/preliminary-draft-proposed-amendments.pdf.

579See Adam Cohen, Commenting on the Comments to the Proposed Amendments to the Federal Rules of Civil Procedure, Inside Couns. (Feb. 18, 2014), http://www.insidecounsel.com/2014/02/18/commenting-on-the-comments-to-the-proposed-amendme.

580Proposed Rule Changes, supra note 578, at 264–69.

581 Id. at 270–75.

582Id. at 289.

583Comment Letter on Proposed Rules from Shira A. Scheindlin, U.S. Dist. Judge, to Comm. on Rules of Practice and Procedure 3 (Jan. 13, 2014), available at http://blogs.reuters.com/alison-frankel/files/2014/02/ruleschanges-scheindlin.pdf.

584Id. at 3–5.

585See Alison Frankel, Two Judges Question Proposed Discovery Limits in Federal Rules, Reuters (Feb. 5, 2014), http://blogs.reuters.com/alison-frankel/2014/02/05/two-judges-question-proposed-discovery-limits-in-federal-rules/.

586Id.; see also Comment Letter from Shira A. Scheindlin, supra note 583, at 6 (voicing concern that proposed rules unnecessarily cut back on the scope of discovery).

587The claims by the defense bar also fail to take account of the likelihood that e-discovery costs will begin to drop sharply, as document review by machine intelligence is gaining acceptance with the courts and document retention is becoming cheaper as a result of improved technologies.

588See, e.g., Abram Chayes, The Role of the Judge in Public Law Litigation, 89 Harv. L. Rev. 1281, 1281 (1976).

589See, e.g., David Marcus, The Past, Present, and Future of Trans-substantivity in Federal Civil Procedure, 59 DePaul L. Rev. 371 (2010) (discussing whether discovery rules should be trans-substantive).

590Of course, not all issues must be presented at summary judgment. But this does not vitiate the argument.

591Moreover, judges have not always approved settlements based on the evidence provided and have sometimes demanded more information. See, for example, Judge Rakoff’s controversial decision to refuse approval of the settlement between the SEC and Bank of America in the Merrill Lynch case. Supra note 259 and accompanying text.

592Complaint Pursuant to 8 Del. C. § 220 for Access to Corporate Books and Records at 2–5, Saito v. McKesson HBOC, Inc., No. 18553-NC (Del. Ch. Dec. 14, 2000), 2000 WL 34549543.