Microsoft Word - 1_2014.3_Guttentag_FINAL GUTTENTAG – FINAL ON REQUIRING PUBLIC COMPANIES TO DISCLOSE POLITICAL SPENDING Michael D. Guttentag* Mandatory disclosure is a central feature of securities regulation in the United States, yet there is little agreement about how to determine precisely what public companies should be required to disclose. This lack of consensus explains much of the disagreement about whether the Securities and Exchange Commission should require public companies to disclose political spending. To resolve the political spending disclosure debate I therefore begin by considering the more general question of how to evaluate any proposed mandatory disclosure requirement. I show why the presumption should be against adding a new disclosure requirement, and then identify the kinds of evidence that should be sufficient to overcome this presumption. Applying this new analytic framework to the political spending disclosure debate—and basing this analysis in part on previously unpublished empirical findings—shows that public companies should not be required to disclose political spending. I. Introduction ............................................................... 594 II. Current Arguments for Requiring Political Spending Disclosure and Their Shortcomings .......... 599 A. Analogy to Information Already Required to be Disclosed .............................................................. 607 B. Problems with the Argument by Analogy ........... 610 * Professor of Law and John T. Gurash Fellow of Corporate Law and Business, Loyola Law School, Los Angeles. I would like to thank the participants in the Loyola Law School summer scholarship colloquium, as well as James Angel, Stephen Bainbridge, Carlos Berdejo, Victor Fleischer, Sarah C. Haan, Robert J. Jackson, Jr., Justin Levitt, Elizabeth Pollman, Allison Quaglino, Tracy Yue Wang, and Michael Waterstone for very helpful comments on earlier drafts of this article. GUTTENTAG – FINAL 594 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 C. Expressive Investor Protection and Political Spending .............................................................. 616 D. Problems with the Expressive Investor Protection Argument ........................................... 619 E. Market Failures and the Mandatory Disclosure of Political Spending .......................... 627 III. A General Theory of How to Evaluate a Topic- Specific Disclosure Requirement ............................... 630 A. Presumption Against Regulatory Intervention .. 630 B. Market Failure Justifications for Regulatory Intervention ......................................................... 637 1. Positive Externalities and Topic-Specific Disclosure Requirements ............................... 638 2. Reducing Tunneling and Topic-Specific Disclosure Requirements ............................... 642 IV. The General Theory Applied to a Political Spending Disclosure Requirement ............................ 645 A. Positive Externalities and Political Spending Disclosure ............................................................ 646 B. Reducing Tunneling and Political Spending Disclosure ............................................................ 652 V. Conclusion .................................................................. 661 I. INTRODUCTION The arguments for requiring public companies to disclose their political spending are certainly alluring. Political spending is quite similar to much of the information public companies are already required to disclose.1 Access to this information would appear to be a necessary prerequisite for shareholders to exercise their basic rights as the firm’s owners.2 Finally, there are many reasons to suspect that private ordering alone will not lead public companies to adopt socially optimal disclosure policies.3 1 See infra Part II.A. 2 See infra Part II.C. 3 See infra Parts II.E and III.B. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 595 These arguments help to explain why a preeminent group of scholars asked the Securities and Exchange Commission (“SEC”) to develop a rule requiring disclosure of political spending by public companies,4 and why their proposal has received an unprecedented amount of support.5 However, the 4 A committee co-chaired by Lucian Bebchuk, Harvard Law School, and Robert J. Jackson, Jr., Columbia Law School, submitted a rule- making petition to the Securities and Exchange Commission (SEC), requesting that “the Commission develop rules to require public companies to disclose to shareholders the use of corporate resources for political activities.” Letter from Comm. on Disclosure of Corporate Political Spending to Elizabeth M. Murphy, Sec’y, U.S. Sec. & Exch. Comm’n (Aug. 3, 2011) [hereinafter Bebchuk Petition], available at http://perma.cc/P75P-BAF5. The other co-authors of the Bebchuk Petition are: Bernard S. Black, Northwestern University Law School; John C. Coffee, Jr., Columbia Law School; James D. Cox, Duke Law School; Ronald J. Gilson, Stanford Law School and Columbia Law School; Jeffrey N. Gordon, Columbia Law School; Henry Hansmann, Yale Law School; Donald C. Langevoort, Georgetown Law School; and Hillary Sale, Washington University in St. Louis School of Law. 5 The SEC received over 1,113,000 comments in support of such a requirement as of September 13, 2014. U.S. SEC. & EXCH. COMM’N, COMMENTS ON RULEMAKING PETITION: PETITION TO REQUIRE PUBLIC COMPANIES TO DISCLOSE TO SHAREHOLDERS THE USE OF CORPORATE RESOURCES FOR POLITICAL ACTIVITIES, FILE NO. 4-637, available at http://perma.cc/7SPG-Z8KY. See also Nicholas Confessore, S.E.C. Gets Plea: Force Companies to Air Donations, N.Y. TIMES, Apr. 24, 2013, at A1. The SEC decided at the end of 2013 to remove consideration of a political spending disclosure requirement from its 2014 agenda. Dina El Boghdady, SEC Drops Disclosures of Corporate Political Spending from Its Priority List, WASH. POST, Dec. 1, 2013, at A8. The New York Times responded with a lead editorial rebuking the SEC’s decision. Editorial, Keeping Shareholders in the Dark, N.Y. TIMES, Dec. 4, 2013, at A32 (“Good corporate governance requires that companies are transparent about their use of corporate resources [on political spending]. Shareholders know this and have demanded disclosure.”). A group of seventeen U.S. senators wrote a letter to SEC Chair Mary Jo White expressing their disappointment about the apparent non-decision. Letter from Robert J. Menendez, U.S. Sen., et al., to Mary Jo White, Chair, U.S. Sec. & Exch. Comm’n (Jan. 9, 2014), available at http://perma.cc/QXM2-BJUD. See also John Coates, SEC’s Non-Decision Decision on Corporate Political Activity a Policy and Political Mistake, HARV. LAW SCH. FORUM ON CORPORATE GOVERNANCE & FIN. REGULATION (Dec. 13, 2013, 8:51 AM), http://perma.cc/8ABC-RJG4. GUTTENTAG – FINAL 596 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 arguments thus far offered for requiring public companies to disclose political spending are unpersuasive. An argument based on similarities between political spending and information that public companies are currently required to be disclosed ignores similarities between political spending and information not required to be disclosed, and places unjustifiable reliance on the efficacy of existing disclosure requirements.6 The argument that mandatory political spending disclosure is necessary to protect shareholder rights fails to explain why the federal government should grant expressive protection to public company shareholders or to acknowledge the degree to which voluntary disclosures already provide such protection to concerned investors.7 Finally, observing that market failures can distort public company disclosure policies only begins the process of evaluating a topic-specific disclosure proposal.8 To advance the debate about political spending disclosure it is necessary to first answer a more general question: how the efficacy of any topic-specific disclosure requirement should be evaluated. In Part III of this Article, I answer this more general question. The answer makes two claims. The first claim is that the presumption should be against imposing new topic-specific disclosure requirements on public companies because of: (1) the breadth of the materiality disclosure obligation already imposed on these companies, (2) the heterogeneity among firms subject to the mandatory disclosure regime, (3) the increasing ease with which firms can avoid public company disclosure obligations, (4) the efficacy of private ordering, and (5) challenges to removing disclosure requirements once imposed.9 Earlier editorials on the mandatory disclosure of political spending by public companies include: Editorial, Corporate Donations and the S.E.C., N.Y. TIMES, Apr. 25, 2013, at A30; Editorial, Serving Shareholders and Democracy, N.Y. TIMES, Aug. 10, 2011, at A22; and Editorial, The Corporate Disclosure Assault, WALL ST. J., Mar. 19, 2012, at A16. 6 See infra Part II.B. 7 See infra Part II.D. 8 See infra Parts II.E and III. 9 See infra Part III.A. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 597 The second claim about how to evaluate the efficacy of any topic-specific disclosure requirement is that the presumption against creating new topic-specific disclosure requirements should only be set aside if there is evidence that such a requirement will effectively address one of two market failures. The two market failures that justify regulating public company disclosures are those caused by positive externalities and by excessive tunneling.10 Positive externalities can distort public company disclosure policies when these firms are unable to capture all of the benefits their disclosures provide to others. Tunneling, which occurs when insiders take assets from the firm at a discount to their real value, can create incentives for those in control of the firm to implement opaque disclosure policies.11 Disclosure regulation can be an effective tool to address market failures caused by either a positive externalities or an excessive tunneling market failure, but finding evidence of a link between either of these market failures and the systematic underdisclosure of a specific category of information is challenging.12 Positive externalities may lead public companies to systematically disclose less information, but less information does not necessarily make investors worse off. Positive externalities only harm investors when underdisclosure caused by this market failure somehow harms investors for some other reason. Evidence of this kind of combined effect is difficult to discern.13 There are similar challenges in identifying when specific disclosure requirements would provide an effective means to reduce 10 See infra Part III.B. 11 For a discussion of the rationale for using the tunneling terminology, see Michael D. Guttentag, Protection from What? Investor Protection and the JOBS Act, 13 U.C. DAVIS BUS. L.J. 207, 227 (2013) [hereinafter, Guttentag, Investor Protection]. Other terms frequently used to describe similar phenomena are self-dealing and agency costs. See id. at 227 n.101 (providing examples). 12 See infra Part III.B. 13 See infra Part III.B.1. GUTTENTAG – FINAL 598 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 tunneling, because our understanding of the links between more disclosure and less tunneling is highly imperfect.14 In Part IV of this Article, I then evaluate the efficacy of requiring the disclosure of political spending based on the general method for carrying out such an analysis developed in Part III. I begin with a presumption against requiring public companies to disclose political spending, and then consider whether there is evidence of a positive externalities or a tunneling market failure large enough to justify rejecting the presumption. Ultimately, I do not find sufficient evidence to overcome this presumption. Positive externalities might lead public firms to systematically under-disclose political spending, but the trivially small dollar amounts involved provide the better explanation for the non-disclosure of political spending by many firms.15 There is, however, one caveat before dismissing entirely a positive externalities market failure justification for requiring disclosure of political spending. Even though the amounts involved in political spending are trivially small from the perspective of overall firm value, the amount of a firm’s political spending does provide useful information about the future performance of a firm’s securities.16 One might argue that the non-disclosure of value-relevant information is prima facie evidence of a market failure; however, the evidence also shows that political spending is value-relevant because it provides an informative signal about otherwise unobservable firm attributes, not because it causes a future decline in firm value. Mandating the disclosure of an informative signal is not likely to benefit shareholders, even in the event (unlikely here) that the signal was withheld because of positive externalities. 14 See infra Part III.B.2. 15 See infra notes 168–177 and accompanying text. 16 See, e.g., Rajesh K. Aggarwal, Felix Meschke & Tracy Yue Wang, Corporate Political Donations: Investment or Agency?, 14 BUS. & POL., no. 1, Article 3, 1, 36 (2012) (finding that corporate political donations are negatively correlated with future excess returns). GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 599 Persistent underdisclosure by firm insiders to facilitate tunneling is the second market failure that can justify mandating the disclosure of a specific category of information.17 There are indications that political spending may be a form of tunneling by firm insiders. For example, there is a high correlation between the political beliefs supported by the firm and those held by the firm’s CEO, and higher levels of political spending are observed at firms with weaker corporate governance practices in general.18 However, even if public company political spending often benefits firm insiders more than the firm itself, there is little reason to expect that mandatory disclosure would provide a cost-effective means to reduce this form of tunneling. Previously unpublished findings from research by Rajesh Aggarwal and his colleagues that suggest transparency about political spending might reduce the deleterious effects of political spending on future stock returns do not, despite appearances, show that political spending disclosure will deter this form of tunneling.19 Mandatory disclosure does make sense as a central pillar of securities regulation, but the justifications for this central role are more subtle and complex than most realize. If the mandatory disclosure system is to function effectively, it is crucial to understand these justifications and to only turn to mandatory disclosure regulation when the evidence suggests such intervention is likely to be efficacious. The available evidence simply does not justify requiring public companies to disclose their political spending. II. CURRENT ARGUMENTS FOR REQUIRING POLITICAL SPENDING DISCLOSURE AND THEIR SHORTCOMINGS Recent interest in requiring the disclosure of political spending by public companies arose out of the Supreme 17 See infra Part III.B.2. 18 See infra notes 191–196 and accompanying text. 19 See infra notes 209–219 and accompanying text. GUTTENTAG – FINAL 600 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 Court’s Citizens United v. Federal Election Commission decision.20 The Citizens United decision overturned a statute limiting corporate contributions to contests for federal political office, and raised the specter of corporate spending unduly influencing the federal political process.21 But within the Citizens United decision was the germ of an idea about how to limit corporate involvement in federal political contests: the mechanisms of corporate governance. In the majority opinion in Citizens United, Justice Kennedy wrote that “[t]here is, furthermore, little evidence of abuse that cannot be corrected by shareholders ‘through the procedures of corporate democracy.’”22 20 Citizens United v. Fed. Election Comm’n, 558 U.S. 310 (2010). 21 For a review of some of the concerns about the adverse effects of the Citizens United decision, see Bradley A. Smith & Allen Dickerson, The Non-Expert Agency: Using the SEC to Regulate Partisan Politics, 3 HARV. BUS. L. REV. 419, 423–24 (2013). See also President Barack H. Obama, State of the Union Address (Jan. 27, 2010), available at http://perma.cc/RJ5H-W73X (“With all due deference to separation of powers, last week the Supreme Court reversed a century of law [with the Citizens United decision] that I believe will open the floodgates for special interests—including foreign corporations—to spend without limit in our elections. I don’t think American elections should be bankrolled by America’s most powerful interests, or worse, by foreign entities.”). For evidence of increased corporate political spending after the Citizens United decision, see Samuel Issacharoff & Jeremy Peterman, Special Interests after Citizens United: Access, Replacement, and Interest Group Response to Legal Change, 9 ANN. REV. L. & SOC. SCI. 185, 199–201 (2013); Tilman Klumpp, Hugo M. Mialon & Michael A. Williams, The Business of American Democracy: Citizens United, Independent Spending, and Elections (July 2014), available at http://perma.cc/GZ6J-RRYZ. 22 Citizens United, 558 U.S. at 361 (quoting First Nat’l Bank of Bos. v. Belotti, 425 U.S. 765, 794 (1978)). Justice Kennedy’s assertion assumes that the interests of a firm’s shareholders are generally aligned with the interests of the public at large when considering political spending by corporations. This assumption raises issues beyond the scope of this Article, which considers the more limited question of whether making the disclosure of political spending mandatory will benefit public company shareholders. For a discussion of the importance of distinguishing between the interests of public company shareholders and other constituents in information about political spending, see Sarah C. Haan, Voter Primacy, 83 FORDHAM L. REV. (forthcoming 2015). GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 601 Following up on Kennedy’s suggestion about the potential efficacy of these “procedures of corporate democracy” to limit corporate political spending, Lucien Bebchuk and Robert J. Jackson, Jr. wrote “Corporate Political Speech: Who Decides,” a Comment in the Harvard Law Review (“Who Decides”).23 In Who Decides, Bebchuk and Jackson observe that access to information about political spending by public companies is a necessary, but often missing, prerequisite to the application of the corporate governance tools that Kennedy referred to in Citizens United.24 Following up on Who Decides, Bebchuk and Jackson drafted a petition that was submitted by a group of scholars to the SEC asking the SEC to develop a rule requiring disclosure of political spending by public companies (“Bebchuk Petition”).25 The Bebchuk Petition has garnered an unprecedented amount of attention.26 More recently, Bebchuk and Jackson published Shining Light on Corporate Political Spending, which offers a comprehensive analysis of the arguments for and against requiring public companies to disclose political spending (“Shining Light”).27 A detailed consideration of arguments offered by Bebchuk and Jackson for political spending disclosure highlights what 23 Lucien A. Bebchuk & Robert J. Jackson, Jr., Corporate Political Speech: Who Decides?, 124 HARV. L. REV. 83 (2010) [hereinafter Bebchuk & Jackson, Who Decides]. 24 Id. at 104–07. As evidence of the potential magnitude of undisclosed political spending by public companies, Bebchuk and Jackson present evidence of substantial sums that political intermediaries, such as the Chamber of Commerce, raised. Id. at 93–95. Contributions to political intermediaries do not need to be disclosed in any manner, and Bebchuk and Jackson speculate that the Supreme Court’s holding in Citizens United may have increased significantly the amount of such political spending. Id. See also John C. Coates IV, Corporate Politics, Governance, and Value Before and After Citizens United, 9 J. EMPIRICAL LEGAL STUD. 657, 661–63 (2012) (describing the various ways in which public companies can avoid reporting political spending). 25 Bebchuk Petition, supra note 4. 26 See supra note 5 and accompanying text. 27 Lucien A. Bebchuk & Robert J. Jackson, Jr., Shining Light on Corporate Political Spending, 101 GEO. L.J. 923, 925–26 (2013) [hereinafter, Bebchuk & Jackson, Shining Light]. GUTTENTAG – FINAL 602 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 is missing thus far in the debate on making this disclosure mandatory for public companies. I focus on Bebchuk and Jackson’s work for four reasons.28 First, Bebchuk and 28 There are certainly other sources for arguments for requiring public companies to disclose political spending. For example, there is the excellent scholarship by Ciara Torres-Spelliscy. See, e.g., Corporate Governance after Citizens United: Hearing Before the Subcomm. on Capital Mkts., Ins. & Gov’t Sponsored Enters. of the H. Comm. on Fin. Servs., 111th Cong. 192–210 (2010) (written statement of Ciara Torres- Spelliscy, Counsel, Brennan Center for Justice); Ciara Torres-Spelliscy, Hiding Behind the Tax Code, the Dark Election of 2010 and Why Tax- Exempt Entities Should Be Subject to Robust Federal Campaign Finance Disclosure Laws, 16 NEXUS: CHAP. J.L. & POL’Y 59, 62–63 (2011) (discussing disclosure loopholes in federal and state laws); CIARA TORRES- SPELLISCY, BRENNAN CTR. FOR JUSTICE, CORPORATE CAMPAIGN SPENDING: GIVING SHAREHOLDERS A VOICE 4 (2010), http://perma.cc/9R68-S7JY (arguing for disclosure and mandatory shareholder consent for political spending); CIARA TORRES-SPELLISCY, CORPORATE REFORM COAL., THE SEC AND DARK POLITICAL MONEY: AN HISTORICAL ARGUMENT FOR REQUIRING DISCLOSURE 4–5 (June 18, 2013), http://perma.cc/UNA4-GSTA [hereinafter, TORRES-SPELLISCY, DARK POLITICAL MONEY] (reviewing three examples of SEC involvement in matters related to political spending over the past four decades). Also relevant to the debate on requiring the disclosure of political spending are congressional hearings held shortly after the decision in Citizens United and, in particular, the testimony of Professor John Coffee, one of the co-authors of the Bebchuk Petition. See Corporate Governance After Citizens United: Hearing Before the Subcomm. on Capital Mkts., Ins. & Gov’t Sponsored Enters. of the H. Comm. on Fin. Servs., 111th Cong. 5–8 (2010) (statement of John C. Coffee, Jr., Adolf A. Berle Professor of Law, Columbia Law School). The Bebchuk and Jackson arguments for mandatory disclosure of political spending have already generated a large amount of criticism. Among the important criticisms that were submitted in comment letters to the SEC are: Letter from James J. Angel, Assoc. Professor of Fin., Georgetown Univ., McDonough Sch. of Bus., to U.S. Sec. & Exch. Comm’n (Mar. 22, 2013) [hereinafter Angel Letter], available at http://perma.cc/PJB4-PAXW (noting shortcomings in the political spending disclosure proposal, including that: (1) placing disclosure burdens on public companies will further the flight of companies from public securities markets; (2) the proponents of the proposed rule conflate public company and private company political spending; (3) political spending is not important to most investors; (4) political spending is different than executive compensation, which is subject to disclosure obligations; and (5) GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 603 the costs of implementing the proposed rule are likely to exceed the “nonexistent benefits”); Letter from Stephen M. Bainbridge, William D. Warren Distinguished Professor of Law, UCLA Sch. of Law, et al., to Elizabeth M. Murphy, Sec’y, U.S. Sec. & Exch. Comm’n (Mar. 23, 2012) [hereinafter Bainbridge Letter], available at http://perma.cc/E3RR-4J4L (noting shortcomings in the political spending disclosure proposal, including: (1) current political disclosure rules already require disclosure of much political spending by public companies; (2) the proposed rule might lead to “over-reporting [that] would be confusing and misleading to shareholders;” (3) shareholders have consistently rejected proposals for more political spending disclosure by public companies; (4) the proposed rule would chill political speech by public firms; and (5) the proposed rule would unnecessarily draw the SEC into a political debate); Letter from Keith Paul Bishop, Adjunct Professor of Law, Chapman Univ. Law Sch., to Elizabeth M. Murphy, Sec’y, U.S. Sec. & Exch. Comm’n (Aug. 6, 2011), available at http://perma.cc/66YZ-4R47 (arguing that disclosure requirements would unnecessarily burden reporting companies); Letter from 60 Plus Ass’n, et al., to Elizabeth M. Murphy, Sec’y, U.S. Sec. & Exch. Comm’n (Jan. 4, 2013) [hereinafter 60 Plus Association Letter], available at http://perma.cc/6BL2-JX7R (arguing that the SEC cannot legally adopt the proposals in the petition because they exceed the SEC’s authority and violate the First Amendment). The 60 Plus Association Letter was filed on behalf of the U.S. Chamber of Commerce and twenty- eight other similar organizations. Finally, following the publication of Shining Light, a symposium issue of the Harvard Business Law Review published several additional critiques of a political spending mandatory disclosure requirement, including: Paul Atkins, Materiality: A Bedrock Principle Protecting Legitimate Shareholder Interests Against Disguised Political Agendas, 3 HARV. BUS. L. REV. 363 (2013); James R. Copland, Against An SEC- Mandated Rule on Political Spending Disclosure: A Reply to Bebchuk and Jackson, 3 HARV. BUS. L. REV. 381 (2013); Matthew Lepore, A Case for the Status Quo: Voluntary Disclosure, 3 HARV. BUS. L. REV. 413 (2013); Smith & Dickerson, supra note 21; and J.W. Verret, The Securities Exchange Act is a Material Girl, Living in a Material World: A Response to Bebchuk and Jackson’s “Shining Light on Corporate Political Spending,” 3 HARV. BUS. L. REV. 453 (2013). Opponents of the Bebchuk Petition have argued, among other things: (1) that such a requirement is superfluous given existing public company disclosure obligations (see, e.g., Atkins, supra, at 367; Verret, supra, at 464–65); (2) that such a requirement is more likely to harm than benefit public company shareholders (see, e.g., Copland, supra, at 384); and (3) that public company political spending disclosure is not an appropriate topic for SEC consideration (see, e.g., Smith & Dickerson, supra, at 422). But these three critiques do not convincingly address evidence suggesting a reasonable investor would be interested in political GUTTENTAG – FINAL 604 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 Jackson offer a coherent and sustained analysis of arguments for requiring the disclosure of political spending by public companies. Second, Bebchuk and Jackson consider in detail several objections to a mandatory political spending disclosure requirement.29 Third, much of the Bebchuk and Jackson argument for requiring public companies to disclose political spending is grounded in the economic analysis of disclosure regulation. Their approach, therefore, provides a welcome opportunity to situate a rule that would force public companies to disclose political spending within the broader context of scholarship on the economics of mandatory disclosure regulation.30 Finally, the work of Bebchuk and Jackson on this topic is central to efforts to bring this issue to public attention. Bebchuk and Jackson, for example, were not only the principal draftsmen, but also the co-chairmen of the committee that submitted the Bebchuk Petition to the SEC.31 I rely, in turn, on three sources to reconstruct the Bebchuk and Jackson arguments for requiring public companies to disclose political spending.32 The first source is the Who Decides comment.33 Who Decides primarily highlights the extent to which Justice Kennedy’s reliance in his majority opinion in Citizens United on the “procedures of corporate democracy” to limit political spending by public spending information (see infra Part IV.A), or that political spending disclosure could reduce self-serving behavior by firm insiders (see infra Part IV.B). More generally, these attacks suffer from many of the same shortcomings as does the Bebchuk and Jackson analysis because they do not adequately ground their critiques in a more fundamental analysis of the purposes legitimately served by the public company mandatory disclosure regime. 29 Bebchuk & Jackson, Shining Light, supra note 27, at 953–66. 30 See infra Part IV. 31 Bebchuk & Jackson, Shining Light, supra note 27, at 925 n.1. 32 Bebchuk and Jackson also write an ongoing series of posts regarding mandatory disclosure of public company political spending. Posts Tagged ‘Rulemaking Petition on Corporate Political Spending’, HARVARD LAW SCH. FORUM ON CORPORATE GOVERNANCE & FIN. REGULATION, http://perma.cc/H95Q-LKR2 (last visited Nov. 24, 2014). 33 See Bebchuk & Jackson, Who Decides, supra note 23. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 605 companies is contingent on the availability of such mechanisms. Who Decides identifies four new corporate governance mechanisms that might better align corporate spending on political contests with shareholder interests: (1) mandatory shareholder approval of political spending, (2) mandatory independent director approval of political spending, (3) permitting shareholders to determine the extent to which shareholder approval or independent director approval of political spending is required, and (4) requiring public firms to disclose political spending.34 The second source I rely upon to reconstruct the Bebchuk and Jackson arguments is the Bebchuk Petition.35 I look to the Bebchuk Petition because Bebchuk and Jackson were the co-chairmen of the committee and the “principal draftsmen” of the petition.36 The Bebchuk Petition primarily argues that investor interest in public company political spending justifies making disclosure of this information mandatory.37 The third source I rely on to reconstruct the Bebchuk and Jackson arguments is the Shining Light article.38 In Shining Light, Bebchuk and Jackson consider several aspects of the debate on political spending disclosure, including: (1) the SEC’s past practice of enacting disclosure requirements in response to investor interest,39 (2) evidence of substantial investor interest in political spending,40 (3) evidence that public company political spending information is not yet 34 Bebchuk & Jackson, Who Decides, supra note 23, at 117 (“We have examined which rules would best address agency problems and align political speech decisions with the interests of shareholders.”). 35 See Bebchuk Petition, supra note 4. 36 See Bebchuk & Jackson, Shining Light, supra note 27, at 925 n.1. 37 See Bebchuk Petition, supra note 4, at 7, 9–10. The arguments offered in the Bebchuk Petition for requiring political spending disclosures are, perhaps surprisingly, quite limited in scope. There are several possible explanations for this limited scope, including differences of opinion between the committee members, a desire to keep the rule-making proposal short and concise, and an effort to focus on arguments that would be especially appealing to the SEC. 38 See Bebchuk & Jackson, Shining Light, supra note 27. 39 Id. at 939–40. 40 Id. at 940–41. GUTTENTAG – FINAL 606 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 readily available,41 (4) why political spending is sufficiently different from ordinary types of corporate spending to justify its own mandatory disclosure requirement,42 (5) why a voluntary disclosure regime alone will not address the demand for information about public company political spending,43 and (6) recommendations as to how to implement a corporate political spending disclosure requirement.44 Bebchuk and Jackson also identify and respond to ten objections against requiring public companies to disclose political spending.45 For example, they argue that such disclosure would impose only minimal costs on public companies.46 Bebchuk and Jackson may be understood from these sources as advancing three main arguments for requiring public companies to disclose political spending. First, Bebchuk and Jackson argue that similarities between political spending and other types of information public companies are already required to disclose justify requiring the disclosure of political spending (the “argument by analogy”).47 Second, Bebchuk and Jackson claim that the mandatory disclosure of political spending provides shareholders the information they need to protect against investing in firms that support political causes they may find objectionable (the “expressive investor protection argument”).48 Third, Bebchuk and Jackson identify several problems with relying on private ordering to establish an optimal level of political spending disclosure by public 41 Bebchuk & Jackson, Shining Light, supra note 27, at 930–37. 42 Id. at 942–44. 43 Id. at 947–49. 44 Id. at 949–53. 45 Id. at 953–65. 46 Id. at 964. 47 See infra Part II.A. 48 See infra Part II.C. I use the term “expressive investor protection” to describe the distinctive type of investor protection Bebchuk and Jackson claim a political spending disclosure requirement might provide. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 607 companies (the “market failure arguments”).49 Each of these arguments and their limitations are reviewed next. A. Analogy to Information Already Required to be Disclosed The first argument Bebchuk and Jackson offer for requiring public companies to disclose political spending is an argument by analogy. The essence of this argument is that similarities between political spending and other information that the SEC already mandates disclosure of justify requiring the disclosure of political spending. Bebchuk and Jackson emphasize two similarities in particular in advancing this argument by analogy. First, Bebchuk and Jackson highlight the extent to which political spending, as with other information the SEC has previously mandated disclosure of, is an expenditure that can benefit firm insiders as much as or more than shareholders.50 Second, they observe that investor interest in information about political spending by public companies is at least as great, if not greater than, the level of investor interest in other information that the SEC has required public firms to disclose.51 There are, in turn, two categories of information that Bebchuk and Jackson highlight as providing an especially relevant precedent for the mandatory disclosure of 49 See infra Part II.E. 50 Bebchuk and Jackson write in Who Decides that political spending decisions “may be a product not merely of a business judgment regarding the firm’s strategy, but also of the directors’ and executives’ own political preferences and beliefs.” See Bebchuk & Jackson, Who Decides, supra note 23, at 90. Bebchuk and Jackson also observe that such divergences “may be particularly likely” in the context of “rules concerning corporate governance and shareholder rights.” Id. at 91. 51 For more recent evidence of investor interest in this information, see Jason Zweig, Corporate Political Gifts: Count ’Em if You Can, WALL ST. J., Aug. 30–31, 2014, at B1. As a general matter, it is unclear why investor interest alone would be a good metric for determining when information should be subject to a mandatory disclosure obligation. Among other things, such an observation does not address either the costs of such disclosures or why private ordering is inadequate. GUTTENTAG – FINAL 608 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 political spending: information about executive compensation arrangements and information about related party transactions.52 The Bebchuk and Jackson argument by analogy for requiring the disclosure of political spending by public companies first appears in Who Decides.53 In Who Decides, Bebchuk and Jackson treat the benefits of mandating disclosure of political spending by public companies as sufficiently self-evident as to not require detailed consideration. They write that “[e]ffective disclosure should be supported even by those who are fully content with the existing ‘procedures of corporate democracy’ under corporate law rules, as such procedures cannot be expected to have a meaningful impact on political speech decisions if shareholders are uninformed about the corporation’s political spending.”54 The argument by analogy is developed in slightly more detail in the Bebchuk Petition. The Bebchuk Petition provides evidence of a high level of investor interest in political spending by companies55 and reviews past examples of situations in which the SEC relied upon indications of investor interest to determine what public companies should 52 According to Bebchuk and Jackson, where “the interests of directors and executives diverge from those of shareholders with sufficient regularity and magnitude,” the types of “special [disclosure] requirements” associated, for example, with “decisions related to executive compensation” should be applied. Bebchuk & Jackson, Who Decides, supra note 23, at 90; see also id. at 93. 53 Id. at 96–97. 54 Id. at 105. 55 Evidence of investor interest in the Bebchuk Petition includes: (1) polls of shareholders about their level of interest in such spending, (2) shareholder proposals offered in support of political spending disclosure, and (3) indications of interest from large institutional shareholders in information about political spending by public firms. Bebchuk Petition, supra note 4, at 3–6. But see Angel Letter, supra note 28, at 4–5 (observing that some of the data regarding investor interest in political spending by public corporations is indicative of interest in political spending by corporations generally—not just corporations that are public). GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 609 be required to disclose.56 The Bebchuk Petition explains that “the Commission’s disclosure rules have changed in response to increased investor interest in receiving particular types of information about the companies they invest in, changes in disclosure practices, or external events that increase the importance of certain types of information for shareholders.”57 Based on this past practice, the Bebchuk Petition concludes that a high level of investor interest in public company political spending justifies requiring the disclosure of this information.58 In Shining Light, Bebchuk and Jackson continue the argument by analogy for requiring public companies to disclose political spending. Bebchuk and Jackson note that, unlike many other business decisions, political spending is apt to be an area in which the interests of directors and officers systematically diverge from the interests of other shareholders because “corporate political spending may reflect not only directors’ and executives’ business judgment, but also their political preferences.”59 Shining Light also highlights several similarities between political spending 56 Among the past examples reviewed in the Bebchuk Petition are the decisions to mandate additional disclosure related to executive compensation arrangements and to mandate disclosure about director oversight of a firm’s risk-taking activities. See Bebchuk Petition, supra note 4, at 2–3. 57 Id. at 2. 58 Id. The Bebchuk Petition also observes that disclosure of political spending is a necessary prerequisite for the type of shareholder monitoring that the Court suggested in the Citizens United opinion. Id. at 7. However, it is unclear why the Court’s dicta in Citizens United about how corporate political spending might be constrained should be accepted as an accurate description of what the law of corporate governance is or should be with respect to this activity. Furthermore, the Bebchuk Petition cites Who Decides to support the assertion that a political spending disclosure requirement would benefit shareholders, but the discussion in Who Decides did not include analytic support for the claim that political spending should be disclosed. See Bebchuk Petition, supra note 4, at 7 (citing Bebchuk & Jackson, Who Decides, supra note 23, at 97) (but discussion of the efficacy of disclosure actually appears on pages 104 to 107 of Who Decides). 59 See Bebchuk & Jackson, Shining Light, supra note 27, at 942. GUTTENTAG – FINAL 610 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 and other types of expenditures where a conflict of interest exists and disclosure is mandatory.60 B. Problems with the Argument by Analogy The argument by analogy as a justification for mandatory disclosure of political spending is unconvincing. Such an argument relies on two premises: (1) that political spending is analogous to information public companies are already required to disclose, and (2) that current mandatory disclosure requirements provide reliable precedent. Both premises are suspect. The first premise of the Bebchuk and Jackson argument by analogy––that political spending is analogous to information public companies are already required to disclose––is suspect because political spending appears to be more similar in important respects to information that is not currently subject to a mandatory disclosure requirement than to information that is required to be disclosed. For example, Bebchuk and Jackson highlight similarities between political spending and executive compensation.61 Bebchuk and Jackson, however, do not consider how limited executive compensation disclosure obligations are in certain respects. Public companies are required to disclose the compensation provided to, at most, five executives.62 There are many compensation arrangements that may involve significant conflicts of interest, but for which there is no mandatory disclosure obligation currently in place.63 A 60 Bebchuk & Jackson, Shining Light, supra note 27, at 948–49. 61 See supra note 52 and accompanying text. 62 Item 402(a)(3) of Regulation S-K sets out the criteria for determining which executives will be subject to the compensation disclosure obligations. Those executives are typically the firm’s Chief Executive Officer (CEO), Chief Financial Officer (CFO), and the three other most highly compensated firm employees. 17 C.F.R. § 229.402 (2014). See also Smith & Dickerson, Non-Expert Agency, supra note 28, at 448. 63 A study that Bebchuk and Jesse Fried cite in their book on executive compensation shows that compensation arrangements for employees other than the top five executives may result in significant GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 611 convincing argument for mandatory disclosure of political spending based on its similarity to executive compensation arrangements should show that political spending is not only similar to executive compensation generally, but also that political spending is more comparable to the compensation received by the firm’s top five executives than to the compensation received by all of the firm’s other employees.64 Considering similarities between political spending and related party transactions not subject to a mandatory disclosure obligation also complicates the argument by analogy for requiring political spending disclosure. While it is true that any transaction between the firm and someone defined as a related party with a value in excess of $120,000 is subject to a mandatory disclosure obligation, not all relationships that might involve related parties are included within the definition of a related party for the purposes of this disclosure obligation.65 For example, a transaction between the firm and the step-child of a director with a value in excess of $120,000 must be disclosed, whereas a similar extraction of value from the firm by firm insiders. Bebchuk and Fried discuss a study finding that “where the CEO is highly paid (relative to similarly situated CEOs), executives at all four levels below the CEO also tend to be overpaid.” LUCIAN BEBCHUK & JESSE FRIED, PAY WITHOUT PERFORMANCE: THE UNFULFILLED PROMISE OF EXECUTIVE COMPENSATION 64 (2004) (citing James B. Wade, Charles A. O’Reilly, III & Timothy G. Pollock, Overpaid CEOs and Underpaid Managers: Fairness and Executive Compensation, 17 ORG. SCI. 527 (2006)). 64 One might imagine how such an argument could proceed. Presumably one would focus on the need for disclosure when personal benefits are received directly by those in control of the firm, as is likely the case for compensation paid to the firm’s top five executives. Comparatively, disclosure may be less helpful when the benefits are provided to those in control of the firm indirectly. It is not self-evident, however, where political spending falls in the spectrum between direct personal benefit and indirect personal benefit, because political spending by the firm does not directly put more cash in the pockets of those in control of the firm. 65 The criteria for determining when a conflict of interest transaction must be disclosed is set out in Item 404 of Regulation S-K, and the criteria for determining what constitutes a related party transaction is set out in Instructions to Item 404(a)(1)(b)(ii). 17 C.F.R. § 229.404 (2014). GUTTENTAG – FINAL 612 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 transaction between the firm and the cousin of a director does not need to be disclosed.66 Which of these two relationships is more analogous to political contributions again presents a challenging question that Bebchuk and Jackson do not address.67 Another problem with relying on apparent similarities between political spending and other types of information already requiring disclosure is that similarities can be feigned. Bebchuk and Jackson argue that the large number of political spending shareholder proposals and the unprecedented number of comments received by the SEC on the Bebchuk Petition indicate a high degree of investor interest in this information.68 However, the costs of submitting a shareholder proposal or providing a comment on a rule-making proposal are quite low. For example, the seventy-one shareholder political spending proposals made in the 2012 proxy season that Bebchuk and Jackson report upon could have been made by shareholders holding less than $150,000 of the relevant companies’ stock.69 Similarly, the vast majority of the comments received by the SEC on the political spending disclosure proposal consist of form letters.70 Why should similarities to information that public 66 See 17 C.F.R. § 229.404 (2014). 67 I would suspect that this kind of question is simply too difficult to answer conclusively, which is why I prefer to evaluate the proposal to require public companies to disclose political spending in the context of the more fundamental purposes served by mandating firm disclosure policies. See infra Parts III and IV. 68 Bebchuk & Jackson, Shining Light, supra note 27, at 937–41. 69 Id. at 938. A shareholder must hold at least $2,000 worth of a company’s stock during the relevant period to submit a shareholder proposal to the firm. 17 C.F.R. § 240.14a-8(b)(1) (2012). The ability to generate a large number of such proposals with limited resources is more than just a theoretical possibility. See Steven Davidoff Solomon, Grappling with the Cost of Corporate Gadflies, N.Y. TIMES, Aug. 20, 2014, at B1 (discussing how three individuals who owned a relatively small number of shares accounted for seventy percent of all individual-sponsored proposals among Fortune 250 companies that year). 70 For the various comments, including those that appear to be form letters, on the Bebchuk Petition, see U.S. SEC. & EXCH. COMM’N, COMMENTS ON RULEMAKING PETITION: PETITION TO REQUIRE PUBLIC COMPANIES TO GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 613 companies are required to disclose be given more weight than similarities to information not required to be disclosed, particularly when such similarities can be feigned? These are questions raised by the argument by analogy, which Bebchuk and Jackson do not and, perhaps, cannot fully address.71 The second premise of the argument by analogy is that current mandatory disclosure requirements provide a useful precedent for establishing new disclosure requirements. There are several reasons to be hesitant in relying on the precedent of SEC determinations as to what information should be subject to a mandatory disclosure requirement. Both scholars and courts have raised concerns about whether the SEC has established the content of past disclosure requirements in an optimal manner. The scholarly critique of the SEC mandatory disclosure regime has been ongoing for several decades.72 Academics have, for example, argued that DISCLOSE TO SHAREHOLDERS THE USE OF CORPORATE RESOURCES FOR POLITICAL ACTIVITIES, FILE NO. 4-637, available at http://perma.cc/7SPG- Z8KY. 71 See supra note 67. 72 The debate about the efficacy of public company disclosure regulation can be divided into three epochs. The first epoch was largely shaped by econometric studies in the 1960s and 1970s, which failed to find any measureable benefits from the enactment of either the Securities Act, see Gregg A. Jarrell, The Economic Effects of Federal Regulation of the Market for New Security Issues, 24 J.L. & ECON. 613, 615, 666–69 (1981); Carol J. Simon, The Effect of the 1933 Securities Act on Investor Information and the Performance of New Issues, 79 AM. ECON. REV. 295, 295 (1989); or the Exchange Act, see George J. Benston, Required Disclosure and the Stock Market: An Evaluation of the Securities Exchange Act of 1934, 63 AM. ECON. REV. 132, 152 (1973). Several important articles from this epoch appeared in a Virginia Law Review issue dedicated to the Securities Act and the Exchange Act fifty years after their passage, including John C. Coffee, Jr., Market Failure and the Economic Case for a Mandatory Disclosure System, 70 VA. L. REV. 717 (1984), and Frank H. Easterbrook & Daniel R. Fischel, Mandatory Disclosure and the Protection of Investors, 70 VA. L. REV. 669 (1984). A second epoch primarily revolved around the publication of one seminal article. In 1995, the publication of Paul G. Mahoney, Mandatory Disclosure as a Solution to Agency Problems, 62 U. CHI. L. REV. 1047 (1995) reminded legal academics that disclosure rules can be used to both GUTTENTAG – FINAL 614 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 the SEC consistently fails to use the best available evidence when engaging in rulemaking because of behavioral biases that limit the efficacy with which the SEC carries out its mission.73 Other critiques of the SEC rulemaking process note that the decision-making process within the SEC is skewed because the SEC Commissioners have to navigate enhance share price accuracy and deter untoward behavior. In his article, Mahoney nicely illustrates the long history of reliance on disclosure rules as a way to deter untoward behavior that preceded federalization of securities regulation in the United States by several centuries. Another article published in the same year and reaching conclusions similar to Mahoney’s was Edmund W. Kitch, The Theory and Practice of Securities Disclosure, 61 BROOK. L. REV. 763 (1995). The third epoch of articles on the efficiency of the regulation of public company disclosure policies appeared starting in 1998. These articles addressed the potential costs and benefits of allowing issuers to select the regime under which their securities would be regulated, much as firms in the United States select the state law that governs their internal affairs. Among the important articles from this epoch are: Stephen J. Choi & Andrew Guzman, Portable Reciprocity: Rethinking the International Reach of Securities Regulation, 71 S. CAL. L. REV. 903 (1998); Merritt B. Fox, The Issuer Choice Debate, 2 THEORETICAL INQUIRIES L. 563 (2001); Merritt B. Fox, Retaining Mandatory Securities Disclosure: Why Issuer Choice Is Not Investor Empowerment, 85 VA. L. REV. 1335 (1999); Joseph A. Franco, Why Antifraud Prohibitions Are Not Enough: The Significance of Opportunism, Candor and Signaling in the Economic Case for Mandatory Securities Disclosure, 2002 COLUM. BUS. L. REV. 223 (2002); Michael D. Guttentag, An Argument for Imposing Disclosure Requirements on Public Companies, 32 FLA. ST. U. L. REV. 123 (2004); Edward Rock, Securities Regulation as Lobster Trap: A Credible Commitment Theory of Mandatory Disclosure, 23 CARDOZO L. REV. 675 (2002); Roberta Romano, Empowering Investors: A Market Approach to Securities Regulation, 107 YALE L.J. 2359 (1998); and Roberta Romano, The Need for Competition in International Securities Regulation, 2 THEORETICAL INQUIRIES L. 387 (2001). 73 Stephen J. Choi & A.C. Pritchard, Behavioral Economics and the SEC, 56 STAN. L. REV. 1, 23 (2003) (noting that the SEC “prefers to remain above the grubbiness of empirical data, preferring to ground its policy prescriptions in ‘investor confidence’”); see also Donald C. Langevoort, Taming the Animal Spirits of the Stock Markets: A Behavioral Approach to Securities Regulation, 97 NW. U. L. REV. 135, 173 (2002) (noting the absence of empirical support for policies implemented by the SEC). GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 615 through a challenging political environment.74 One knowledgeable observer, Professor Joan Heminway, supports the “complex, dissonant characterization of the SEC [] as open-minded and reformist, yet also conservative, hierarchical, and slow-moving.”75 Heminway notes that another factor that might limit the efficacy of the SEC is constraints on resources.76 Federal courts have likewise criticized the SEC for failing to carry out a thorough enough review of costs and benefits in the rule-making process. In Business Roundtable v. SEC, for example, the District of Columbia Court of Appeals overturned implementation of SEC rules designed to provide shareholders greater access to the firm’s proxy statement, holding that the SEC had failed to sufficiently evaluate the economic consequences of the proposed rule.77 Uncertainty about the efficacy of executive compensation disclosure requirements provides another example of why current SEC disclosure requirements might not provide a useful precedent. Despite their growing scope, the efficacy of the mandatory disclosure of executive compensation arrangements remains a matter of ongoing dispute. For example, Benjamin Hermalin and Michael Weisbach 74 See, e.g., Floyd Norris, Independent Agencies, Sometimes in Name Only, N.Y. TIMES, Aug. 9, 2013, at B1 (“But while [an] agency may be formally independent, individual commissioners are widely expected to be anything but independent.”). 75 Joan M. Heminway, Desire, Conservatism, Underfunding, Congressional Meddling, and Study Fatigue: Ingredients for Ongoing Reform at the Securities and Exchange Commission?, 81 U. CIN. L. REV. 443, 446 (2012). 76 See id. 77 Business Roundtable v. SEC, 647 F.3d 1144, 1148–49 (D.C. Cir. 2011). However, many commentators have, in turn, criticized the District of Columbia Court of Appeals’ critique of the SEC’s methods as unsound. See, e.g., James D. Cox & Benjamin J.C. Baucom, The Emperor Has No Clothes: Confronting the D.C. Circuit’s Usurpation of SEC Rulemaking Authority, 90 TEX. L. REV. 1811, 1813 (2012); Bruce Kraus & Connor Raso, Rational Boundaries for SEC Cost-Benefit Analysis, 30 YALE J. ON REG. 289, 293 (2013); Cass R. Sunstein & Adrian Vermeule, Libertarian Administrative Law, U. CHI. L. REV. (forthcoming). This Article does not take any position on whether this case was rightly or wrongly decided. GUTTENTAG – FINAL 616 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 developed an analytic model that shows conditions under which “better disclosure regimes can also aggravate agency problems and related costs, including executive compensation.”78 Other disclosure provisions related to executive compensation, such as the requirement that firms compare the CEO’s pay with that of the median employee at the firm, have likewise been criticized for their expense and lack of efficacy.79 Ralph Waldo Emerson observed that a “foolish consistency is the hobgoblin of little minds.”80 It may be possible to rescue the argument by analogy by addressing the various concerns with Bebchuk and Jackson’s reliance on this approach noted here. The more promising approach, however, is to develop a method to systematically evaluate evidence related to any topic-specific disclosure requirement and apply such a method to the evaluation of a political spending disclosure requirement rather than rely on analogies.81 C. Expressive Investor Protection and Political Spending Bebchuk and Jackson’s second main argument for mandating the disclosure of political spending by public companies is that such disclosure would provide shareholders with the information they need to protect 78 Benjamin E. Hermalin & Michael S. Weisbach, Information Disclosure and Corporate Governance, 67 J. FIN. 195, 195 (2012). 79 Steven M. Davidoff, A Simple Solution That Made a Hard Problem More Difficult, N.Y. TIMES, Aug. 28, 2013, at B7. But see Editorial, Exposing the Pay Gap, N.Y. TIMES, Sept. 25, 2013, at A28 (arguing that exposing the pay gap between CEOs and the median employee “will be enormously helpful.”). 80 RALPH WALDO EMERSON, THE SELECTED WRITINGS OF RALPH WALDO EMERSON 138 (Brooks Atkinson ed., 1992). 81 See infra Parts III and IV. Bebchuk and Jackson probably recognize limitations in relying solely on an argument by analogy to justify requiring the disclosure of political spending by public companies, which is why their discussion also includes more fundamental justifications for imposing mandatory disclosure requirements. See infra Part II.E. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 617 themselves from investing in firms supporting political causes they find objectionable. This second argument might be described in terms of expressive investor protection,82 and emphasizes the expressive, rather than economic, benefits political spending disclosure can provide to the firm’s shareholders. This expressive investor protection argument for political spending disclosure first appears in Who Decides. There, Bebchuk and Jackson observe that “shareholders may attach expressive significance to corporate political speech that goes far beyond the amount spent.”83 The potential expressive significance of corporate political spending is also discussed in Shining Light where Bebchuk and Jackson highlight both the “unique expressive significance for shareholders” of this type of spending and that some shareholders may have “a strong interest in not being associated with political speech they oppose.”84 As they do with their argument by analogy, Bebchuk and Jackson provide a precedent in securities regulation policy- making for granting special weight to activities that might affect the expressive interests of a firm’s shareholders. The precedent they cite is the shareholder proposal process.85 The SEC put in place a formal system in 1942 that allowed firm shareholders to place a brief proposal in the firm’s annual proxy statement for shareholder vote.86 To be included in the 82 See supra note 48. 83 Bebchuk & Jackson, Who Decides, supra note 23, at 96. See also Bebchuk & Jackson, Shining Light, supra note 27, at 927 (“[B]ecause of the expressive significance of political spending, shareholders may attach greater importance, beyond the amounts spent, to political spending that deviates from their preferences.”). 84 Bebchuk & Jackson, Shining Light, supra note 27, at 943. 85 Bebchuk & Jackson, Who Decides, supra note 23, at 96; Bebchuk & Jackson, Shining Light, supra note 27, at 938. 86 For a review of the history of the shareholder proposal process, see Maya Mueller, The Shareholder Proposal Rule: Cracker Barrel, Institutional Investors, and the 1998 Amendments, 28 STETSON L. REV. 451, 456–67 (1998). For a recent critique of this process, see Leo E. Strine, Jr., Can We Do Better by Ordinary Investors? A Pragmatic Reaction to the GUTTENTAG – FINAL 618 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 firm’s annual proxy statement, both the shareholder and the proposal need to meet various criteria.87 Bebchuk and Jackson correctly observe that the criteria used to evaluate whether a firm is required to include a particular proposal in the firm’s proxy materials include expressive investor protection concerns. More specifically, Rule 14a-8(i)(5) of the Exchange Act allows firms to exclude shareholder proposals from the firm’s proxy statement if the issues raised by the proposal are not material to shareholders; however, this basis for exclusion of a shareholder proposal is not available if the proposal involves firm operations that are “otherwise significantly related to the company’s business.”88 In Lovenheim v. Iroquois Brands Ltd., a District Court interpreted the “otherwise significantly related” language as requiring a firm to include a proposal involving the method of preparing fois gras, even though the financial value of the business to the firm was minimal.89 As John Coffee and Hillary Sale explain in their securities regulation textbook, shareholder proposals can “survive exclusion due to social and political reasons.”90 Bebchuk and Jackson conclude that expressive investor protection concerns similarly justify requiring the disclosure of political spending by public firms regardless of the dollar amounts involved. Dueling Ideological Mythologists of Corporate Law, 114 COLUM. L. REV. 449, 488–91 (2014). 87 Criteria for exclusion by the company are set out in 17 C.F.R. § 240.14a-8 (2014). See also STEPHEN J. CHOI & A.C. PRITCHARD, SECURITIES REGULATION: CASES AND ANALYSIS 677–91 (3d ed. 2012); JOHN C. COFFEE, JR. & HILLARY A. SALE, SECURITIES REGULATION: CASES AND MATERIALS 1248–76 (12th ed. 2012); Bebchuk & Jackson, Shining Light, supra note 27, at 943–44. 88 17 C.F.R. § 240.14a-8(i)(5) (2014). 89 Lovenheim v. Iroquois Brands, Ltd., 618 F. Supp. 554, 561 (1985). 90 COFFEE & SALE, supra note 87, at 1253. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 619 D. Problems with the Expressive Investor Protection Argument The investor expressive protection argument for requiring the disclosure of political spending, however, is, at best, incomplete. Bebchuk and Jackson take it as a given that investor expressive concerns are a legitimate basis for determining the content of mandatory disclosure requirements. But it is not obvious why purchases of public company securities, which are in essence private contracts between the investor and the firm, should entitle purchasers to have their expressive concerns addressed by federal regulation.91 This particular form of investor protection— investor expressive protection—does not, for example, have a historical foundation in the federal securities statutes’ goal of protecting investors from various harms.92 Three additional observations further suggest why the public company mandatory disclosure regime is probably not the best tool for addressing expressive investor protection concerns, such as investor expressive concerns about political spending.93 91 I thank Sarah Haan for bringing this to my attention. 92 Guttentag, Investor Protection, supra note 10 (identifying efforts to protect investors from fraud, an unlevel informational playing field, the extraction of private benefits from the firm by firm insiders, and investors’ propensity to make unwise investment decisions as the harms from which securities regulations in the United States were historically designed to protect investors); TORRES-SPELLISCY, DARK POLITICAL MONEY, supra note 28, at 20–21. 93 An additional observation about Bebchuk and Jackson’s expressive investor protection argument for requiring the disclosure of political spending by public companies is that this aspect of their argument is not buttressed by noting similarities between political spending, executive compensation arrangements, and related party transactions. At least historically, the justifications for requiring the disclosure of executive compensation and related party transactions do not depend on how such disclosures might provide expressive investor protection, but rather on the deterrent value of such disclosure requirements. See infra Part IV.B. But Jackson has argued that “corporate spending on politics has unique, expressive significance that goes far beyond the bottom line of the corporation. That’s why, for example, executive compensation is required GUTTENTAG – FINAL 620 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 The first observation is that information unraveling should address most concerns about expressive investor protection without the need for regulatory intervention. Comparative efficacy of private ordering is an issue that inevitably arises when considering imposing mandatory obligations. In the absence of market failures, private ordering would presumably provide a more nuanced way to determine what information firms commit to disclose. One argument that might be offered about why relying on voluntary disclosure to address expressive investor protection concerns is doomed to fail is that firms will not voluntarily disclose questionable behavior.94 However, a general theory about information disclosure suggests that non-disclosure in the absence of mandatory disclosure regulation is not inevitable. The theory is known by its unraveling result.95 This theory describes how voluntary disclosures occur as firms that are high quality (as measured by whatever the relevant criteria are) choose to reveal their quality in order to distinguish themselves from lower quality firms. Individually rational disclosure decisions can thus lead to all information about firm quality being voluntarily revealed (“unraveling”). The unraveling process can solve many expressive investor protection concerns without the need for regulatory intervention. If investors find certain kinds of firm behavior troubling, firms can voluntarily and credibly commit either not to engage in the expressively problematic behavior or to disclose their actions when they do engage in this type of behavior. By holding a portfolio consisting of only firms that commit to the desired behavior or disclosure policies, an investor could be assured that her expressive concerns are addressed. Moreover, the financial cost of following this to be disclosed.” PBS NewsHour: SEC Considering New Rule for Political Contributions by Public Companies, (PBS television broadcast May 6, 2013), available at http://perma.cc/R285-7BJN. 94 Bebchuk & Jackson, Shining Light, supra note 27, at 947–49. 95 For a survey of research related to unraveling, see David Dranove & Ginger Z. Jin, Quality Disclosure and Certification: Theory and Practice, 48 J. ECON. LITERATURE 935, 941–45 (2010). GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 621 investment strategy is likely to be minimal, because the foregone opportunity cost of investing in any one firm or group of firms is quite small.96 Indeed, Bebchuk and Jackson 96 From a financial perspective, the economic benefits of diversification can be achieved even with an investment in a relatively small number of firms. Meir Statman, How Many Stocks Make a Diversified Portfolio?, 22 J. FIN. & QUANTITATIVE ANAL. 353, 353 (1987) (discussing the minimum diversification required for economic benefit). Moreover, if enough investors share similar expressive concerns, two market forces are likely to further reduce the cost of making investments that address such concerns. First, financial intermediaries may have a sufficient incentive to create a fund that consists only of the securities of firms that agree not to engage in the troubling behavior. See, e.g., Paul Sullivan, When Buying Stock in Gluttony Is a Good Investment, N.Y. TIMES, Feb. 1, 2014, at B5. The article describes an online company, Motif, which offers investors theme-based collections of securities to invest in at a relatively low cost. Another organization, US SIF, offers a list of mutual funds which are designed to serve various expressive interests of investors. See About Us, US SIF, http://perma.cc/R23C-CC8H (last visited Nov. 23, 2014). The creation of special purpose funds is likely to lower the costs for investors interested in assuring that their expressive interests are addressed. The second market force that might lower the cost for investors to invest in firms that address their expressive preferences is that firms will have an incentive to make such a commitment in order to access additional sources of capital. Professor Robert Sitkoff nicely described more than a decade ago how such a voluntary mechanism was already addressing many various expressive investor concerns, including political spending. Robert H. Sitkoff, Corporate Political Speech, Political Extortion, and the Competition for Corporate Charters, 69 U. CHI. L. REV. 1103, 1119 (2002) (“Take the assumption that in the absence of a statutory prohibition no firm would swear off political activity. If there are resources held by prospective investors who are skittish about funding political speech (perhaps because they are a church or a university) but otherwise would be happy to invest in stock, then some firms would simply insert ‘no politics’ clauses into their corporate charters, or mutual fund companies would create a ‘no politics’ fund, to tap into this source of capital. Consider the proliferation of ‘social responsibility’ funds. These funds assure investors that their money will not be invested in corporations engaged in certain specific forms of behavior, such as the sale of alcohol or tobacco, military contracting, abortion-related services, and so on. There is a fund for everyone: just as the Meyers Pride Value Fund avoids companies that lack stated policies against discrimination on the basis of sexual GUTTENTAG – FINAL 622 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 report that a number of public companies have already committed to disclosing their political spending.97 Following an investment strategy that avoids firms engaging in objectionable political spending is already a real option for concerned investors. It is well-recognized that unraveling may not solve all public company disclosure market failures,98 but Bebchuk and Jackson do not explain why unraveling would fail to provide expressive investor protection in the context of political spending by public firms. Even partial unraveling should, and probably already does, provide investors enough options to ensure that they can find investment opportunities that both address their expressive concerns and provide market returns without any regulatory intervention. The second observation suggesting why mandatory disclosure is unlikely to be the best tool for addressing expressive investor protection relates to a line-drawing challenge. It is certainly true that line-drawing challenges are ubiquitous and not reason enough to avoid regulatory intervention, but without a theory about why expressive investor concerns deserve a regulatory response such line- drawing challenges would likely prove insurmountable. There are many actions firms might take that could offend some of the firm’s shareholders. Robert Sitkoff lists just a few of the ways a firm’s business operations may touch on issues related to expressive investor protection concerns, including: the sale of alcohol or tobacco, providing goods or services for military purposes, the production of firearms, providing abortion-related services, and employee policies including the absence of a stated policy against discrimination on the basis of sexual orientation, or, on the other side of the political spectrum, the granting of domestic partner benefits.99 Bebchuk and Jackson’s analysis provides orientation, the Timothy Plan funds avoid companies that provide domestic partner benefits.”). 97 Bebchuk & Jackson, Shining Light, supra note 27, at 946–48. 98 See infra Part III.B. 99 Sitkoff, supra note 96, at 1119. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 623 no guidance for determining when the potential expressive significance to investors of the firm’s activities should trigger a mandatory disclosure obligation, and mandatory disclosure of any behavior that touches on investor expressive concerns seems unwieldy. The third observation suggesting why mandatory disclosure is unlikely to be the best tool for addressing expressive investor protection relates to the mismatch between the investors who might have such concerns and differences between public firms and private firm. Firms are public or private depending on considerations such as where the firm’s securities are traded or how many “beneficial” shareholders the firm has.100 Bebchuk and Jackson’s proposal is that only public firms, which are already required to periodically disclose financial and operating information, also be required to disclose political spending. Limiting a political spending disclosure requirement to public firms is both unjustifiable and potentially counterproductive if limiting these disclosure obligations to public companies is based on protecting investors’ expressive concerns. There are good reasons to be equally, if not more troubled by the expressive concerns of investors in private companies. Differences between public firms and private firms often have little to do with differences between the investors who are the ultimate owners of the firm.101 As a result, 100 Michael D. Guttentag, Patching a Hole in the JOBS Act: How and Why to Rewrite the Rules that Require Firms to Make Periodic Disclosures, 88 IND. L.J. 151, 152–53, 157 (2013) [hereinafter Guttentag, Patching a Hole]. 101 One important example of similar beneficiaries for public and private firms is provided by the widespread ownership of businesses by private equity firms. Private equity firms, which act as intermediaries between institutional investors and businesses of various kinds, grew in popularity starting in the 1980s. By 2008 there had “been a large movement of public companies into private ownership through leveraged acquisitions by private equity firms.” Ronald J. Gilson & Charles K. Whitehead, Deconstructing Equity: Public Ownership, Agency Costs, and Complete Capital Markets, 108 COLUM. L. REV. 231, 233 (2008). One estimate is that private equity firms now “own and operate trillions of GUTTENTAG – FINAL 624 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 shareholders in private firms in the absence of disclosure are equally, if not more, unlikely to have the information necessary to avoid investing in businesses whose activities they find objectionable. This comparison is important because companies have a large and growing degree of control over whether to become public.102 Particularly after changes enacted as part of the JOBS Act of 2012, it is now quite palatable for even very large firms to remain outside of the public company reporting regime.103 To the extent a disclosure requirement dollars’ worth of American businesses.” Robert J. Jackson, Jr., Private Equity and Executive Compensation, 160 UCLA L. REV. 638, 642 (2013). The ultimate financial beneficiaries of private equity firms are quite often the same investors, who, if they owned shares directly, might have the types of expressive investor protection concerns a political spending disclosure requirement is designed to address. Investments in private equity firms are “dominated by public pension funds, with CalPERS (California Public Employees’ Retirement System), CalSTERS (California State Teachers’ Retirement System), PSERS (Pennsylvania Public School Employees’ Retirement System), and the Washington State Investment Board occupying the top four slots.” Steven N. Kaplan & Per Strömberg, Leveraged Buyouts and Private Equity, 23 J. ECON. PERSP. 121, 124 (2009). The beneficiaries of these institutional investors are those who have a stake in the financial health of public pension funds. Anne Tucker, Retirement Revolution: Unmitigated Risks in the Defined Contribution Society, 51 HOUS. L. REV. 153, 153–54 (2013). It is true that, even if non- public firms have access to political spending information, indirect investors might not be able to use the traditional tools of corporate governance to affect undesirable corporate behavior. On the other hand, the presence of intermediary owners may, at least in some circumstance, help overcome collective action problems and facilitate the ability of the ultimate owners to address their expressive concerns. 102 As but one example of a firm choosing to remain private when, in the past, it might have gone public, see Michael J. de la Merced, SurveyMonkey Raises Nearly $800 Million While Staying Private, N.Y. TIMES DEALBOOK (Jan. 17, 2013, 11:43 AM), http://perma.cc/BUS6-8GGU. 103 Several aspects of the JOBS Act made staying private easier and more palatable. First, the JOBS Act substantially increased the threshold level of shareholders of record that trigger mandatory compliance with federal periodic disclosure requirements, Guttentag, Patching a Hole, supra note 100, at 153. A second way in which the JOBS Act has reduced the incentives for firms to comply with federal public reporting disclosure obligations is by making it easier for firms to raise funds privately. Title II GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 625 is especially burdensome for companies involved in business that touch on expressive concerns, and that disclosure requirement is only imposed on public companies, the economic effect of imposing a public company disclosure requirement may simply be to create a subsidy that rewards firms in the affected industry that avoid or exit the public disclosure regime.104 The ownership of firearm manufacturers in the United States illustrates that the consequences of the mismatch between whether a firm is public and the expressive concerns of a firm’s ultimate owners is not simply a matter of theoretical curiosity. The largest firms engaged in the manufacture of firearms in the United States offer investors two very different ways in which to address expressive investor protection concerns. On the one hand, some firms engaged in the manufacture of firearms, including Smith & Wesson and Sturm, Ruger & Company, are public companies. Investors can relatively easily avoid investing in these firms’ securities because of their expressive concerns, if they so choose.105 Another major manufacturer of firearms in the United States, the Freedom Group, is not a public company, but is owned by a private equity firm, Cerberus Capital of the JOBS Act required the SEC to remove the “prohibition against general solicitation or general advertising” when firms raise funds pursuant to Rule 506 of Regulation D, so long as only accredited investors purchase the securities offered. Jumpstart Our Business Startups Act, 2 Pub. L. No. 112-106, § 201(a)(1), 126 Stat. 306, 313–314 (2012). The rules implementing this Section of the JOBS Act took effect on September 23, 2013. See Press Release, U.S. Sec. & Exch. Comm’n, SEC Approves JOBS Act Requirement to Lift General Solicitation Ban (July 10, 2013), http://perma.cc/VH6E-N2BB. For a discussion of the legislative history surrounding the adoption of this section of the JOBS Act, see Guttentag, Investor Protection, supra note 11, at 245–47. For a discussion of the use of private markets to trade securities, see Elizabeth Pollman, Information Issues on Wall Street 2.0, 161 U. PA. L. REV. 179, 179 (2012). 104 See Angel Letter, supra note 28, at 1–4. 105 There are, for example, specialized funds, which commit not to hold any securities issued by a firm that manufactures firearms. Jeff Sommer, The Guns Hiding in Your Portfolio, N.Y. TIMES, Jan. 20, 2013, at MB1. GUTTENTAG – FINAL 626 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 Management.106 For those who find profiting from gun manufacturing unpalatable, there is little they can do about the fact that they are invested, albeit indirectly, in the Cerberus Capital Management funds that own the Freedom Group. Investors that have an investment in Cerberus Capital Management, and, therefore, profit from the sale of products such as the Bushmaster rifle include the University of California and CalSTRS.107 Moreover, the intermediation by private equity firms between businesses and investors, which is likely to limit the ability of investors to act on their expressive investor protection concerns, is not costless.108 Neither the argument by analogy nor the expressive investor protection argument justify imposing a disclosure requirement on public company political spending. Political spending is analogous both to information that is subject to a mandatory disclosure requirement and information that is not subject to such a disclosure obligation. There is also a legitimate issue as to how much precedential weight should be given to past disclosure requirements. As for the expressive investor protection argument for mandating political spending disclosure, there is not an obvious rationale for using the mandatory disclosure regime to address these concerns. Moreover, given the amount of voluntary disclosure regarding these activities already 106 Natasha Singer, The Big Shot: How an Unknown Company Became the Gun Industry’s Giant, N.Y. TIMES, Nov. 27, 2011, at BU1. 107 Greg Roumeliotis, Cerberus Raises $2.61 Billion Private Equity Fund: Sources, REUTERS, May 1, 2013, available at http://perma.cc/B5G4- P7Y8; Ricardo Duran, CalSTRS Releases Statement on Investments in Firearms Manufacturers, CALSTRS.COM (Dec. 18, 2012), http://perma.cc/ZRV5-92H2 (describing the end of CalSTRS investment in Cerberus). Shortly after Adam Lanza killed twenty-six people on December 14, 2012 using guns manufactured by the Freedom Group, apparently because of the ensuing publicity, Cerberus attempted to sell its stake in the Freedom Group. Peter Lattman, Dropping Its Guns: In Unusual Move, Cerberus Is to Sell Firearms Company, N.Y. TIMES, Dec. 19, 2012, at B1; Michael J. de la Merced, Cerberus Regrouping in Bid to Sell Gun Maker, N.Y. TIMES, Dec. 10, 2013, at B1. 108 See, e.g., Victor Fleischer, Two and Twenty: Taxing Partnership Profits in Private Equity Funds, 83 N.Y.U. L. REV. 1, 1–2 (2008). GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 627 occurring and the ease with which firms can avoid the public company reporting regime entirely, imposing disclosure requirements based on expressive concerns is probably unnecessary and potentially even counter-productive. E. Market Failures and the Mandatory Disclosure of Political Spending Bebchuk and Jackson do not rely solely on the argument by analogy and the expressive investor protection argument to justify requiring the disclosure of political spending by public firms. They also consider market failure justifications for mandatory disclosure regulation. Two instances in which Bebchuk and Jackson turn to market failure arguments are especially noteworthy. The first instance in which Bebchuk and Jackson offer a more fundamental economic justification for requiring political spending disclosure by public companies comes when they consider why private ordering may not lead to an optimal amount of political spending disclosure. Bebchuk and Jackson offer four reasons related to disclosure market failures to explain why private ordering could fail in the context of political spending disclosure. First, Bebchuk and Jackson observe that firms that do not disclose political spending may nevertheless be the low-cost provider of this information.109 In this regard, Bebchuk and Jackson observe 109 Bebchuk & Jackson, Shining Light, supra note 27, at 935 (“Investors in public companies should not have to bear the costs of assembling this information when the corporation, which already has the information, can easily provide it to shareholders.”). See also Bebchuk & Jackson, Who Decides, supra note 23, at 106 (“The corporation, rather than individual investors, is in the best position to put together the needed information in a cost-effective way.”). This Bebchuk and Jackson argument harkens back to observations made in Frank H. Easterbrook & Daniel R. Fischel, Mandatory Disclosure and the Protection of Investors, 70 VA. L. REV. 669, 684 (1984) (“For most information about a firm, the firm itself can create and distribute the knowledge at less cost than the shareholders.”). Easterbrook and Fischel observe that the argument that the firm may be the low-cost provider of firm-specific information and other economic arguments for mandatory disclosure requirements “have a common problem: they do not link the GUTTENTAG – FINAL 628 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 that “the quality of [political spending] information that large public companies have so far provided to investors through voluntary disclosure policies is generally low.”110 However, it is unclear why it might not simply be the case that low quality disclosures are the optimal output of an efficient private ordering process. Second, Bebchuk and Jackson observe that private ordering might result in a “lack of uniformity [which] makes comparison among companies costly for investors.”111 Third, Bebchuk and Jackson express concern that relying on voluntary disclosure of political spending is problematic, because of the collective action problems shareholders face generally.112 To support this claim, Bebchuk and Jackson observe that “most public companies currently do not disclose any information at all about political spending,”113 but do not present evidence that, but for collective action problems, firms would disclose this information.114 Fourth, Bebchuk and Jackson observe that, even if most firms did comply with a voluntary disclosure regime, there might remain a few recalcitrant firms that refuse to disclose their political spending, and that the “companies that do decline to disclose might be benefit of disclosure and the benefit of mandatory disclosure. If disclosure is worthwhile to investors, the firm can profit by providing it.” Id. at 682. 110 Bebchuk & Jackson, Shining Light, supra note 27, at 947. 111 Id. at 948. This is, again, a familiar argument for a mandatory disclosure regime, but, as others have noted, there are less intrusive ways of establishing disclosure practices that facilitate investor comparison between companies. Easterbrook & Fischel, supra note 109, at 685–87 (“Mandatory disclosure rules promulgated by the government are one means to achieve standardization, but it does not follow that mandatory disclosure is necessary. Markets frequently devise ingenious solutions to problems of information.”). Paul Mahoney has considered in detail how securities exchanges have in the past, and could in future, continue to play these kinds of coordination roles without regulatory intervention. Paul G. Mahoney, The Exchange as Regulator, 83 VA. L. REV. 1453 (1997). 112 Bebchuk & Jackson, Shining Light, supra note 27, at 948. 113 Id. 114 Such evidence might be provided, for example, by looking at the disclosure practices of firms where there are sufficiently few shareholders that collective action problems are unlikely to present a bar to adopting otherwise efficient disclosure practices. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 629 disproportionately likely to engage in political spending that is inconsistent with shareholder interests.”115 Bebchuk and Jackson are correct to observe that there are often limitations to private ordering in the context of public company disclosures, but situating such claims more precisely in the context of market failure arguments for mandatory disclosure regulation would make this aspect of their analysis more compelling.116 The second instance in which Bebchuk and Jackson offer a more fundamental economic justification for requiring political spending disclosure by public companies comes by implication from their emphasis on similarities between political spending and executive compensation. Presumably Bebchuk and Jackson are relying on justifications for the mandatory disclosure of executive compensation arrangements. Bebchuk is, for example, the co-author with Jesse Fried of the highly influential book on market failures associated with executive compensation at public firms, Pay Without Performance: The Unfulfilled Promise of Executive Compensation.117 In Pay Without Performance, Bebchuk and Fried argue that disclosure can limit excessive executive compensation, because “[e]verything else being equal, we believe, increased transparency should operate to constrain pay.”118 A similar argument about the prophylactic benefits of the mandatory disclosure of political spending is presumably made when drawing a parallel between political spending and executive compensation. Again, Bebchuk and Jackson are right to reach beyond an argument by analogy or expressive investor protection to justify requiring public companies to disclose political 115 Bebchuk & Jackson, Shining Light, supra note 27, at 949 (the recalcitrant firms would likely be the firms that would have to “reveal spending that shareholders would find objectionable.”). 116 See infra Part III and Part IV. 117 BEBCHUK & FRIED, supra note 63 at vii. 118 Id. at 72. See also id. at 192 (“The main aim of requiring disclosure of executive compensation is not to enable accurate pricing of the firm’s securities. Rather, this disclosure is primarily intended to provide some check on arrangements that are too favorable to executives.”). GUTTENTAG – FINAL 630 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 spending. However, the Bebchuk and Jackson analysis could be improved by situating their argument in the context of the broader scholarly debate on why firms should be required to adopt certain disclosure policies. Such an analysis is provided next. III. A GENERAL THEORY OF HOW TO EVALUATE A TOPIC-SPECIFIC DISCLOSURE REQUIREMENT This Section extends research on public company disclosure regulation to develop a new method to determine when information regarding a specific topic should be subject to a mandatory disclosure requirement. The method I propose addresses two questions. The first question addressed is where the burden of proof should be placed when evaluating a topic-specific disclosure proposal. The second question addressed is what evidence should be sufficient to overcome such a presumption. A. Presumption Against Regulatory Intervention The first question I address regarding a topic-specific disclosure proposal, such as the proposal to require disclosure of political spending, is where the burden of proof should be placed. One might argue that the presumption should be in favor of imposing a disclosure requirement, as long as there is some evidence that disclosure of the information would be of interest to a reasonable investor and that the costs of mandatory disclosure of the information in question would be minimal.119 There are, however, five reasons to start with a strong presumption against mandating the disclosure of information regarding a specific topic. These five reasons are: (1) the breadth of the 119 One advantage of moving forward with mandating disclosure is that doing so would likely generate information, which could be useful in the future in improving this and other disclosure interventions. Cf. Michael Abramowicz et al., Randomizing Law, 159 U. PA. L. REV. 929, 929–30 (2011) (arguing that, while much can be learned from experimenting with legal changes, randomized trials produce better results). GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 631 materiality disclosure obligation already imposed on all public companies,120 (2) the heterogeneity among firms subject to the public company mandatory disclosure regime,121 (3) the ease with which firms can avoid public company reporting,122 (4) the availability of private ordering to establish firm-specific disclosure obligations,123 and (5) challenges to removing disclosure requirements once imposed.124 Each of these reasons is considered in greater detail next. First, the breadth of the materiality disclosure obligation already imposed on public companies provides at least some justification for arguing that any topic-specific disclosure requirement is superfluous. Both the Securities Act and the Exchange Act contain provisions requiring public companies, when making disclosures, to always disclose “such further material information . . . as may be necessary to make the required statements . . . not misleading.”125 The Supreme Court in Basic v. Levinson held that the term material as used in the federal securities statutes should be interpreted based on “the significance the reasonable investor would place on the withheld or misrepresented information.”126 120 See infra notes 127–132 and accompanying text. 121 See infra notes 133–137 and accompanying text. 122 See infra notes 136–137 and accompanying text. 123 See infra note 139 and accompanying text. 124 See infra notes 141–142 and accompanying text. 125 17 C.F.R. § 230.408a (2014) (Securities Act Filings); 17 C.F.R. §240.12b-20 (2014) (Exchange Act Filings). This requirement echoes President Franklin D. Roosevelt’s call that “no essentially important element attending the [issuance of securities] shall be concealed from the buying public.” JOEL SELIGMAN, THE TRANSFORMATION OF WALL STREET: A HISTORY OF THE SECURITIES AND EXCHANGE COMMISSION AND MODERN CORPORATE FINANCE 53 (3d ed. 2003) (quoting 73 CONG. REC. 937, 954 (1933)). See also CHOI & PRITCHARD, supra note 87, at 48–98; COFFEE & SALE, supra note 87, at 943–77. But see Bebchuk & Jackson, Who Decides, supra note 23, at 87 (“[T]here is under current corporate law . . . no mandatory disclosure to investors [of corporate political speech decisions].”). 126 Basic Inc. v. Levinson, 485 U.S. 224, 240 (1988). GUTTENTAG – FINAL 632 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 Thus, public companies already appear to be required to disclose all information of interest to a reasonable investor in many circumstances. The breadth of the materiality disclosure obligation might, in fact, appear to obviate the need for any topic- specific disclosure requirements.127 However, the broad reach of the materiality disclosure obligation does not fully dispense with the need for topic-specific disclosure requirements for two reasons. First, there is a fair degree of ambiguity and uncertainty in practice as to how to apply the materiality standard.128 One manifestation of this ambiguity and uncertainty is that the application of the materiality standard by courts tends to vary by jurisdiction along many dimensions.129 The use of topic-specific disclosure 127 See, e.g., Atkins, supra note 28, at 364; J.W. Verret, supra note 28, at 453. 128 To address this particular concern and as an alternative to introducing topic-specific disclosure requirements, the SEC could also provide guidance as to what specific types of information are material. The SEC has provided such guidance in the past, as in the case of the financial impact of climate change, without altering the rule that all material information should be disclosed. Commission Guidance Regarding Disclosure Related to Climate Change, Exchange Act Release Nos. 33- 9106, 34-61469, 75 Fed. Reg. 6290, 6291, 6296 (Feb. 8, 2010). 129 See, e.g., Dale A. Oesterle, The Overused and Under-Defined Notion of “Material” in Securities Law, 14 U. PA. J. BUS. L. 167, 167 (2011) (describing the application of the term materiality by federal courts as “quixotic at best”). There is, for example, uncertainty about whether the reasonable investor should be defined based on the attributes of individual or institutional investors, and whether common behavioral biases should be included in the characterization of the reasonable investor. Barbara Black, Behavioral Economics and Investor Protection: Reasonable Investors, Efficient Markets, 44 LOY. U. CHI. L.J. 1493, 1493 (2013); see also Michael D. Guttentag, An Argument for Imposing Disclosure Requirements on Public Companies, 32 FLA. ST. U. L. REV. 123, 170 n.185 (2004) [hereinafter Guttentag, Imposing Disclosure Requirements] (discussing the difficulties with the “reasonable investor” standard); Elizabeth Pollman, A Corporate Right to Privacy, 99 MINN. L. REV. (forthcoming Apr. 2014) (discussing how the materiality standard is vague and it is therefore unclear whether it would require disclosure of executive health problems). GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 633 requirements may provide an efficient and reliable way to ensure firms make appropriate public disclosures.130 The second reason that the broad reach of the materiality disclosure obligation does not fully dispense with the need for topic-specific disclosure requirements is that there is an entire class of disclosure requirements that are valuable to investors for reasons other than because they generate information a reasonable investor would find useful in valuing the company’s securities. Many disclosure requirements are adopted because they alter behavior, not because they are likely to lead to the disclosure of value- relevant information. The use of disclosure rules to alter behavior, in fact, has an older pedigree than the use of disclosure rules as a way to ensure investors receive information relevant to a firm’s value.131 These two limitations to the application of the general materiality disclosure obligation, ambiguity in applying such a general standard, and the usefulness of disclosure rules as a way to deter untoward behavior, explain why the federal public disclosure regime has historically combined the materiality disclosure obligation with a number of topic- 130 The choice between the application of general principles and specific rules in determining what information public companies should be required to disclose touches on broader questions about the trade-offs between these two approaches to lawmaking. See, e.g., Louis Kaplow, Rules Versus Standards: An Economic Analysis, 42 DUKE L.J. 557, 557 (1992). 131 Louis Brandeis highlighted the prophylactic benefits of disclosure rules in his 1913 essay, What Publicity Can Do. Louis D. Brandeis, What Publicity Can Do, HARPER’S WEEKLY, Dec. 20, 1913, reprinted in LOUIS D. BRANDEIS, OTHER PEOPLE’S MONEY AND HOW THE BANKERS USE IT 92 (1914) (“Sunlight is said to be the best of disinfectants; electric light the most efficient policeman.”). The prototypical disclosure rule relied upon specifically to deter untoward behavior is a requirement that a firm disclose related party transactions (Regulation S-K, Item 404). Such a rule is most effective if it results in non-disclosure, because of its deterrent effect. See also Mahoney, supra note 72. GUTTENTAG – FINAL 634 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 specific disclosure obligations, rather than rely solely on the materiality standard.132 A second reason that the initial presumption should be against imposing topic-specific disclosure obligations on public companies is that there is a large amount of heterogeneity among public firms in the United States. There are over 5,000 firms subject to the mandatory federal public company disclosure regime in the United States,133 and these firms come in many shapes and sizes. To consider just one dimension, the market capitalization of Apple Inc. in August of 2014 was over $600 billion,134 while many public companies have market capitalizations under $10 million. This means the multiple between the market capitalization of the smallest and largest firms subject to roughly the same mandatory disclosure obligations is on the order of at least 6,000 to 1, a dramatic range. This heterogeneity inevitably challenges the ability of regulators to create topic-specific disclosure requirements that work well for all public companies.135 A third reason that the presumption should be against imposing topic-specific disclosure obligations on public 132 The advisability of including topic-specific disclosure requirements was one of the few areas of disagreement between the original drafters of the Securities Act. Joel Seligman writes in his history of the SEC: In particular, Landis fought with Cohen about the wisdom of including a detailed schedule of data to be disclosed by each firm before it sold securities to the public. . . . By April 21, [1933,] Landis and Cohen had completed a revised draft. The hard-won schedule of items to be disclosed remained and had been somewhat expanded. SELIGMAN, supra note 125, at 64–65. 133 Dan Strumpf, In Public Markets, Echoes of a Boom, WALL ST. J., Feb. 6, 2014, at C1. 134 Apple Market Cap, YCHARTS, available at http://perma.cc/9BKA- AYSD (last visited Nov. 24, 2014). Apple Inc.’s market capitalization was in excess of $500 billion at the end of the day on May 13, 2014; Guttentag, Patching a Hole, supra note 100, at 200–05. 135 This is probably true even when disclosure regulations vary by firm size to some degree. For a review of size-based disclosure obligations in federal securities regulation, see CHOI & PRITCHARD, supra note 87, at 409–10; COFFEE & SALE, supra note 87, at 151–53. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 635 companies has to do with one of the potential ramifications of increasing the disclosure burden on public companies in general. It is not the case that firms have no choice but to comply with any topic-specific disclosure requirement imposed upon them. There may be firms that view a particular disclosure requirement as too costly or otherwise too burdensome to comply with, and that, therefore, choose to exit or avoid the public company disclosure regime rather than comply. With the passage of the JOBS Act in 2012, it is not especially difficult for many firms to avoid a topic-specific disclosure requirement they dislike by avoiding the public company disclosure regime entirely.136 Moreover, a firm’s choice to avoid or exit from public disclosure obligations in response to new disclosure requirements may have undesirable repercussions. The imposition of costly disclosure requirements could, for example, reward financial intermediaries that help firms avoid public securities markets, rather than provide any benefit to the firms’ ultimate investors.137 The fourth reason that the presumption should be against imposing topic-specific disclosure obligations on public companies has to do with the availability of private ordering.138 There are likely many firms that would benefit from a commitment to disclose specific categories of information. In fact, many public firms already provide investors information in addition to that required by federal periodic reporting requirements. Perhaps including these kinds of supplemental disclosures in the mandatory disclosure regime would make a firm’s commitment to such ongoing disclosure more credible.139 However, firms probably 136 See supra notes 105–111 and accompanying text. 137 In his comment letter to the SEC on the Bebchuk Proposal, Professor James Angel identified avoidance of public markets as one possible response to a mandatory public company political spending disclosure obligation. Angel Letter, supra note 28, at 1–4. See also Victor Fleischer, Regulatory Arbitrage, 89 TEX. L. REV. 227, 229 (2010). 138 See supra notes 96–100 and accompanying text. 139 Rock, supra note 72, at 676, 685 (arguing that “an important but largely unappreciated function of the U.S. mandatory disclosure regime is GUTTENTAG – FINAL 636 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 already would anticipate retribution if they choose to reverse past disclosure practices without a legitimate justification for doing so. The choice to rely on private ordering to determine individual firm disclosure commitments is likely superior to regulatory intervention, unless there is evidence of some significant market failure.140 A fifth reason that the presumption should be against imposing topic-specific disclosure obligations on public companies has to do with challenges to rolling back disclosure obligations once they are imposed. The extent of disclosure obligations imposed on public companies is steadily increasing.141 One explanation for this phenomenon is that it is even more difficult to eliminate a disclosure obligation once imposed than it is to impose one in the first place.142 Ideally, it would be feasible to experiment with the introduction of new disclosure obligations and then remove them if the resulting evidence showed that the costs of the disclosure obligation exceeded the benefits. However, our disclosure system is not yet that facile, and a disclosure obligation once imposed is likely to remain in place for no other reason than administrative inertia. A nimble public company disclosure regime would be one in which disclosure obligations are regularly imposed and removed in an ever-improving process that systematically the extent to which it permits issuers to make a credible commitment to a level and permanence of disclosure”). 140 See infra Part IV.B. for a detailed consideration of market failure justifications for disclosure regulation. 141 See, e.g., Angel Letter, supra note 28, at 3–4. 142 One possible solution to this difficulty would be to include sunset provisions in disclosure requirements. Despite many potential benefits, sunset provisions do not appear to have garnered widespread acceptance in this context. See, e.g., Roberta Romano, Regulating in the Dark, in REGULATORY BREAKDOWN: THE CRISES OF CONFIDENCE IN U.S. REGULATION 96–98, 101 (discussing opposition to sunsetting and possible reasons why it has not been more widely adopted). But see, e.g., Keith F. Higgins, Div. of Corp. Fin. Dir., U.S. Sec. & Exch. Comm’n, Disclosure Effectiveness, Remarks before the American Bar Association Business Law Section Spring Meeting (Apr. 11, 2014), available at http://perma.cc/Z8Q5-UT38 (arguing for removing disclosure when it is “immaterial or outdated even if it was included in a prior filing”). GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 637 lowers capital costs for public firms. Our disclosure system is not such a system, and for the five reasons listed above, one implication of the current state of the mandatory disclosure regulation in the United States is that we should start with a presumption against imposing new topic-specific disclosure obligations on public companies. B. Market Failure Justifications for Regulatory Intervention The second question to be addressed is what evidence should be sufficient to overcome the presumption against making a new topic-specific disclosure mandatory. Scholarship on public company disclosure regulation provides a useful starting point to answer this question. Whether and to what extent the disclosure policies of public companies should be regulated is the subject of a lively and ongoing academic debate.143 The debate on the efficacy of public company disclosure regulation brings clarity to the question of if or when a new topic-specific disclosure requirement might be beneficial. More specifically, this scholarship identifies only two market failures that might be sufficiently large to justify regulatory intervention into public company disclosure practices. The first market failure involves positive externalities. Without regulatory intervention, public companies may systematically under-disclose, because of difficulties they inevitably face in capturing all of the benefits their disclosures provide to outsiders.144 The second market failure involves the possibility of tunneling by firm insiders. Without regulatory intervention, firm insiders may establish disclosure policies that provide insiders with more opportunities for personal benefit, but diminish firm value.145 I next consider how a better understanding of each of these market failures can guide an evaluation of the likely efficacy of a topic-specific disclosure requirement. 143 See supra note 72. 144 See supra Part III.B.1. 145 See supra Part III.B.2. GUTTENTAG – FINAL 638 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 1. Positive Externalities and Topic-Specific Disclosure Requirements One market failure that justifies regulating public company disclosure policies arises out of the benefits these disclosures can provide to those outside of the firm. More specifically, without regulatory intervention, public firms may disclose less than a socially optimal amount of information, because they are not able to capture the full value their disclosures provide to: (1) competitors, (2) other firms with securities traded on the same venue as the disclosing firm, or (3) the economy at large.146 There are several reasons to suspect that regulating public company disclosure policies can provide a corrective for distortions in public company disclosure policies caused by these positive externalities.147 However, there are also significant challenges in linking a market failure caused by positive externalities to the systematic under-disclosure of information regarding a specific topic. In considering how a positive externalities market failure might justify requiring that information regarding a specific topic be disclosed, it is helpful to consider in turn each of the three potential beneficiaries of such disclosures: (1) competitors, (2) other firms with securities traded on the same venue as the disclosing firm, and (3) the economy at large. With respect to the first group of potential beneficiaries from these disclosures, firms’ competitors, it is worth noting that whether positive interfirm externalities ever justify regulatory intervention is a controversial topic.148 146 Another positive spillover effect, potentially not captured by a firm making public disclosures, comes from the value the disclosed information provides to investors generally. I choose to exclude this potential positive spillover effect for reasons I have elaborated on elsewhere. Guttentag, Patching a Hole, supra note 100, at 193. 147 Guttentag, Imposing Disclosure Requirements, supra note 129, 182–84. 148 See, e.g., Merritt B. Fox, The Issuer Choice Debate, supra note 72, at 570; Merritt B. Fox, Retaining Mandatory Securities Disclosure: Why Issuer Choice Is Not Investor Empowerment, supra note 72, at 1345–46; Guttentag, Imposing Disclosure Requirements, supra note 129, at 128; GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 639 I believe the available evidence shows that: (1) positive interfirm externalities do significantly distort public firm disclosure policies, and (2) regulatory intervention can provide a welfare-enhancing corrective for these distortions.149 However, the implications of positive interfirm externalities for the advisability of adopting a requirement to disclose information regarding a specific topic are quite difficult to discern for several reasons. First, it is hard to determine when specific information would have been disclosed, but for a firm’s inability to realize the benefits such disclosures provide to competitors.150 Moreover, even if information is withheld because of positive interfirm externalities, the disclosure of less information does not inevitably make investors worse off. Withholding information because of positive interfirm externalities is only harmful when the information involved is both competitively disadvantaging to disclose and, at the same time, if Roberta Romano, Empowering Investors: A Market Approach to Securities Regulation, supra note 72, at 2373–75; Roberta Romano, The Need for Competition in International Securities Regulation, supra note 72, at 387– 88, 420. 149 Guttentag, Imposing Disclosure Requirements, supra note 129, at 149–53. Gains from regulatory interventions intended to provide corrective for positive interfirm externalities generally are, however, difficult to measure. See Guttentag, Patching a Hole, supra note 100, at 189–90. 150 One practical solution is to look to disclosure policies when there are no positive interfirm externalities present. Guttentag, Imposing Disclosure Requirements, supra note 129, at 172–74 (relying on this approach). However, determining whether information regarding a specific topic is customarily disclosed in the absence of positive interfirm externalities can be challenging. A second practical solution is to assume that all information of interest to investors would be disclosed, but for the presence of positive interfirm externalities. This solution has the advantage of establishing a minimum standard for raising concerns about the possibility of distortions caused by interfirm externalities, but relies on a tenuous link between investor interest and a positive interfirm market failure. GUTTENTAG – FINAL 640 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 disclosed, would increase social welfare.151 Identifying specific types of information that have both of these attributes is difficult, and so linking positive interfirm externalities to a recommendation to require disclosure of a particular category of information is challenging.152 Other firms with securities traded on the same venue are the second group that firms making disclosures may not realize are benefiting from these disclosures. Brian Bushee and Christian Leuz studied the impact of a rule enacted in 1999 that required all firms with shares traded on the Over- the-Counter Bulletin Board (OTCBB) to either improve the quality of their disclosures or leave the OTCBB.153 Bushee and Leuz found that the securities of firms traded on the OTCBB that had already been complying with federal disclosure requirements “experience[d] positive stock returns and permanent increases in liquidity” when this rule was implemented.154 This finding provides evidence of the positive spillover effects from public company disclosures that can flow to other firms with securities traded on the same venue. However, our understanding of precisely how enhanced disclosures by some firms benefits other firms with securities traded on the same venue is quite limited. The positive spill- over effects observed by Bushee and Leuz occurred in a situation where the disclosure policies of several thousand firms were suddenly shifted from a voluntary disclosure regime to a disclosure regime where the full panoply of federal disclosure requirements was imposed.155 It is unclear how the spill-over benefits observed from such a large-scale 151 Michael D. Guttentag, Accuracy Enhancement, Agency Costs, and Disclosure Regulation, 3 REV. L. & ECON. 611, 624–25, 627–29 (2007) [hereinafter Guttentag, Accuracy Enhancement] (providing an analytic model formalizing this observation). 152 Id. at 627–29. 153 Brian J. Bushee & Christian Leuz, Economic Consequences of SEC Disclosure Regulation: Evidence from the OTC Bulletin Board, 39 J. ACCT. & ECON. 233, 234–35 (2005). 154 Id. at 233–34. 155 Id. at 234–35. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 641 change could be used to determine when requiring the disclosure of information regarding a specific topic would also provide a spill-over benefit to other firms with securities traded on the same venue. The third positive externality that may justify requiring more disclosure by public companies comes from benefits to the economy at large that may result when firms are more transparent. Marcel Kahan in “Securities Laws and the Social Cost of ‘Inaccurate’ Stock Prices” identifies several pathways through which more accurate share prices can provide benefits to the real economy.156 In a more limited context, Robert Bartlett has shown how public disclosures may help in monitoring systemically important firms.157 However, except perhaps in the special case studied by Bartlett, drawing a link between the mandatory disclosure of information regarding a specific topic and the economy-wide benefits public disclosures can provide would seem to be exceedingly difficult. Even if investors and securities markets have benefited from the imposition of disclosure requirements that address distortions caused by positive externalities, our ability to draw insight from this observation to reach conclusions about whether information regarding a specific topic should be subject to a mandatory disclosure requirement is quite limited. Various approaches could be tried to overcome these obstacles. One approach would be to identify those types of information that would both benefit competitors and prove useful to the firm’s investors, if disclosed. An alternative approach is to look to disclosure policies where externalities do not distort disclosure policies, such as in private 156 Marcel Kahan, Securities Laws and the Social Costs of “Inaccurate” Stock Prices, 41 DUKE L.J. 977, 1005–42 (1992). See also Guttentag, Patching a Hole, supra note 100, at 183. There is related evidence that fraudulent financial reporting, which is admittedly an extreme version of inaccurate financial reporting, can distort real economic investment policy. Anne Beatty et al., The Spillover Effect of Fraudulent Financial Reporting on Peer Firms’ Investments, 55 J. ACCT. & ECON. 183, 185 (2013). 157 See, e.g., Robert P. Bartlett, III, Making Banks Transparent, 65 VAND. L. REV. 293, 295–96 (2012). GUTTENTAG – FINAL 642 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 transactions, for guidance as to when information regarding a specific topic should be required to be disclosed. A third approach would be to assume that any information that might be of interest to a reasonable investor would be disclosed, but for distortions caused by positive externalities. This third approach is almost certainly an over-inclusive way to address distortions caused by positive externalities. 2. Reducing Tunneling and Topic-Specific Disclosure Requirements The second type of market failure that justifies imposing disclosure requirements on public companies arises out of insiders’ desire to adopt disclosure policies that maximize their own welfare rather than firm value.158 There is evidence that without mandatory disclosure regulation insiders will adopt disclosure policies that facilitate tunneling at the expense of maximizing firm value. Michael Greenstone and colleagues (“Greenstone et al.”) studied the impact of an amendment to the Exchange Act enacted in 1964 (“1964 Amendment”), which forced several thousand firms to comply with public company disclosure requirements for the first time.159 Greenstone et al. found that the affected firms’ value went up both when the legislation was announced and during the period between the announcement of the legislation and the legislation going into effect. This measurable benefit is best explained as 158 For a model illustrating how this type of self-serving insider behavior is sustainable in efficient public securities markets, see Lucian A. Bebchuk & Zvika Neeman, Investor Protection and Interest Group Politics, 23 REV. FIN. STUD. 1089, 1089–90 (2010). 159 Michael Greenstone, Paul Oyer & Annette Vissing-Jorgensen, Mandated Disclosure, Stock Returns, and the 1964 Securities Act Amendments, 121 Q.J. ECON. 399, 399 (2006). Similar findings based on a smaller sample of the affected securities are reported in Allen Ferrell, Mandatory Disclosure and Stock Returns: Evidence from the Over-the- Counter Market, 36 J. LEGAL STUD. 213, 213–16 (2007). For a more detailed discussion, see Guttentag, Patching a Hole, supra note 100, at 184–88. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 643 evidence that imposing disclosure requirements on public companies can efficiently reduce tunneling.160 But it is difficult to know when specific disclosure requirements will benefit shareholders by reducing tunneling opportunities. This difficulty arises because there are several ways in which disclosure requirements might reduce tunneling, and it is unclear which ways are most important. Disclosure requirements might prevent tunneling through their deterrent effect, a possibility that harkens back to Brandeis’ metaphor of disclosure rules acting like a street light.161 There is, for example, a longstanding precedent of relying on the disclosure of conflict-of-interest transactions to reduce tunneling and deter untoward behavior,162 but evidence of the extent to which disclosure rules can effectively deter untoward behavior in this way is ambiguous. One psychology experiment found that the mere presence of a disclosure requirement, even if the information required to be disclosed was trivial, could significantly reduce morally questionable behavior, while another psychology experiment found that mandatory disclosure of a conflict of interest can have a disinhibitory effect.163 A second way in which disclosure requirements can affect the level of tunneling by firm insiders is through the 160 Firms can voluntarily elect in various ways to be subject to the same disclosure requirements that were forcibly imposed upon them by the 1964 Amendment. Rock, supra note 72, at 678–82. 161 See Brandeis, supra note 131 and accompanying text. 162 See, e.g., Mahoney, supra note 72, at 1048–49. 163 Daylian M. Cain, George Loewenstein & Don A. Moore, The Dirt on Coming Clean: Perverse Effects of Disclosing Conflicts of Interest, 34 J. LEGAL STUD. 1, 5–6 (2005) (finding a disinhibitory effect from conflict of interest mandatory disclosure); Michael D. Guttentag, Christine L. Porath & Samuel N. Fraidin, Brandeis’ Policeman: Results from a Laboratory Experiment on How to Prevent Corporate Fraud, 5 J. EMPIRICAL LEGAL STUD. 239, 239 (2008) (finding requiring disclosure of even trivial information deterred untoward behavior). See also James P. Naughton, Mandatory Disclosure and Managerial Discretion: Evidence from SFAS132R 3 (June 5, 2014) (unpublished manuscript), http://perma.cc/EJ7D-TVWP (finding that increased mandatory disclosure reduced managerial discretion in recognizing the costs of pension plans). GUTTENTAG – FINAL 644 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 mechanisms of corporate governance. Disclosure requirements can insure that the information necessary to effectively use the levers of control is available to shareholders. As Gerard Hertig, Reinier Kraakman, and Edward Rock observe, “informed shareholders can better exercise their decision and appointment rights.”164 A third way in which disclosure policies can affect the level of tunneling by firm insiders is through their effects on the accuracy of the firm’s share price. By enhancing share price accuracy, disclosures can limit opportunities for firm insiders to profit from exclusive access to firm information.165 Correctly identifying and understanding the ways in which disclosure rules reduce tunneling is crucial when evaluating the likelihood a topic-specific disclosure requirement will reduce tunneling in a cost-effective manner. For example, if disclosure rules reduce tunneling because of their deterrent effect, then disclosure rules are more likely to reduce tunneling when the information subject to the disclosure obligation would be embarrassing if revealed. Alternatively, if disclosure rules reduce tunneling because of their link to corporate governance, then the most important information to require disclosure of is information that can empower shareholders. Finally, if disclosure rules reduce tunneling through more accurate share pricing, then the information most likely to be effective in reducing tunneling will be those disclosures that enhance share price accuracy. There is little evidence that any one of these pathways is clearly dominant. Insight into whether a topic-specific 164 Gerard Hertig et al., Issuers and Investor Protection, in THE ANATOMY OF CORPORATE LAW: A COMPARATIVE AND FUNCTIONAL APPROACH 280 (Reinier Kraakman et al. eds., 2nd ed. 2009). The link between disclosure and corporate governance appears to have been of central importance to Bebchuk and Jackson when they first offered arguments for mandatory disclosure of political spending information. Bebchuk & Jackson, Who Decides, supra note 23. 165 See, e.g., David Aboody & Baruch Lev, Information Asymmetry, R&D, and Insider Gains, 55 J. FIN. 2747, 2748–49 (2000); Hollis A. Skaife, David Veenman & Daniel Wangerin, Internal Control over Financial Reporting and Managerial Rent Extraction: Evidence from the Profitability of Insider Trading, 55 J. ACCT. & ECON. 91, 91–92 (2013). GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 645 disclosure requirement is likely to provide a cost-effective way to reduce tunneling needs to be sought on a case-by-case basis. The debate on the mandatory disclosure of political spending has focused primarily on questions such as the extent of investor interest in the topic under consideration or how similar the information is to information firms are already required to disclose.166 These are not the best questions to ask when evaluating whether to require disclosure of information regarding a specific topic. Rather, the evaluation of any proposed topic-specific disclosure rule is best considered in light of the two market failures that mandatory disclosure requirements imposed on public companies have effectively addressed in the past: (1) positive externalities, and (2) excessive tunneling. Each of these market failures illuminates a new way to determine what information firms should be required to disclose. In implementing disclosure requirements as a corrective for positive externalities, we can look to whether the information involved is both competitively disadvantaging and at the same time, if disclosed, would increase social welfare. To provide a corrective for insider capture of the firm’s disclosure policies, the types of disclosure rules that would be most effective are those that can cost-effectively: (1) deter untoward behavior, (2) empower shareholders, or (3) prevent firm insiders from maintaining an informational advantage over those outside the firm. I consider next whether the evidence suggests that the mandatory disclosure of political spending meets any of the criteria linked to disclosure market failures. IV. THE GENERAL THEORY APPLIED TO A POLITICAL SPENDING DISCLOSURE REQUIREMENT In this Section I start from the presumption that political spending should not be subject to a mandatory disclosure obligation, and then consider whether there is evidence that 166 See supra Part II.A. GUTTENTAG – FINAL 646 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 requiring political spending disclosure would provide enough of a corrective for positive externalities or excessive tunneling to overcome this presumption. A. Positive Externalities and Political Spending Disclosure The benefits that disclosure requirements can provide as a corrective for positive externalities are unlikely to justify requiring disclosure of political spending. It is doubtful that a firm’s choice about whether to disclose its political spending is affected by positive externalities. Many public firms do not disclose their political spending, but there is a much more plausible explanation, other than distortions caused by positive externalities, for this non-disclosure of political spending.167 For almost all, if not all, public firms, the amounts involved in political spending are so small that they are trivial from the perspective of overall firm value. The dollar amounts involved in political spending by public firms may appear large. For example, publicly- reporting corporate Political Action Committees (“PACs”) spent almost $350 million in 2012.168 Moreover, not all public company political spending can be observed.169 As one indicator of the magnitude of undisclosed political spending by public corporations, Bebchuk and Jackson note that “spending by just eight of the most active intermediaries between public companies and politics exceeded $1.5 billion 167 Bebchuk and Jackson report that some, but not all, public firms already do voluntarily disclose political spending information. Bebchuk & Jackson, Shining Light, supra note 27, at 934. 168 Press Release, Federal Election Commission, FEC Summarizes Campaign Activity of the 2011-2012 Election Cycle (Apr. 19, 2013), http://perma.cc/7AQ6-FY4Q. But see Eduardo Porter, Business Losing Clout in a G.O.P. Moving Right, N.Y. TIMES, Sept. 4, 2013, at B1 (noting that “companies openly spent about $75 million from their treasuries on federal elections in 2012) (citing Adam Bonica, Avenues of Influence: On the Political Expenditures of Corporations and Their Directors and Executives 1 (June 20, 2014) (unpublished manuscript), available at http://perma.cc/E38U-UYSX). Moreover, Bonica estimates that the majority of this $75 million does not come from public companies. Id. at 1. 169 See Bebchuk & Jackson, Who Decides, supra note 23, at 104–07. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 647 between 2005 and 2010.”170 About this amount, Bebchuk and Jackson observe, “[This] is hardly a trivial sum.”171 However, from the perspective of the enterprise value of public firms, the amounts involved in political spending are, in fact, trivial.172 The market capitalization of public firms in the United States exceeded $18 trillion at the end of 2012.173 In comparison, the value of all known expenditures on federal election campaigns in the 2012 election cycle ($7 billion) is a minor rounding error, less than four-one hundredths of a percentage point (0.04%).174 Moreover, the value of all known expenditures on federal election campaigns in 2012 certainly overstates the amount of political spending on federal elections by public firms for two reasons. First, this estimate assumes that all federal campaign financing is provided exclusively by public companies, which is obviously not the case. A more plausible, although still quite generous, upper bound on political spending by public companies during this election cycle comes from the work of political scientist Adam Bonica. Bonica estimated that private and public companies combined spent no more than $400 million in 2012.175 This amount represents less than two one-thousandths of a percentage point (0.002%) of public firm market 170 Bebchuk & Jackson, Shining Light, supra note 27, at 931, 956. 171 Id. at 956. 172 See also Copland, supra note 28, at 385–89. 173 Data: Market capitalization of listed companies (current US$), WORLD BANK, http://perma.cc/5JQ2-KR43 (last visited Nov. 23, 2014). 174 Press Release, Federal Election Commission, FEC Summarizes Campaign Activity of the 2011-2012 Election Cycle (Apr. 19, 2013), http://perma.cc/7AQ6-FY4Q (estimating expenditures on all federal election campaigns in the 2012 election cycle). 175 Bonica, supra note 168, at 1, estimates that even with the “heroic assumption that all ‘dark money’ funneled through non-disclosing 501(c) non-profit organizations originated from corporate treasuries,” the maximum contributions to federal election campaigns during the 2012 cycle from both private companies and public companies did not exceed $400 million. GUTTENTAG – FINAL 648 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 capitalization.176 Second, this estimate assumes that all public companies fund political spending equally, but the evidence suggests that political spending is carried out disproportionately by larger firms.177 If larger firms are disproportionately more likely to engage in significant political spending, then, on average, the relative impact on firm value is likely to be even less than that estimated above. It may be true, as Bebchuk and Jackson observe, that: (1) only some public companies voluntarily provide detailed information about the total amount of their political spending,178 (2) much public company political spending is not required to be disclosed in any forum,179 and (3) other information about public company political spending is not readily accessible.180 But even with this limited information, the evidence shows that the magnitude of political spending must be trivial from the perspective of total firm value for most, if not all, public firms. Therefore, its non-disclosure by many public firms does not appear to be the result of 176 An argument might be made that the amounts that corporations spent on lobbying provide a better measure of the influence of corporations in politics, and that these amounts are substantially larger than the amounts contributed to federal political campaigns. Bonica, for example, reports that corporations spent over $5 billion on lobbying in 2010. Bonica, supra note 168, at 8. However, even if we consider the amount spent on lobbying relative to public firm market capitalization, the amounts involved are still relatively trivial (less than three-one hundredths of a percentage point (0.03%)). The relatively small magnitude of political spending when compared to private sector economic activity generally in the United States prompted one group of scholars to write an article asking why there is so little political spending in the United States. Stephen Ansolabehere, John M. de Figueiredo & James M. Snyder, Jr., Why Is There So Little Money in U.S. Politics?, 17 J. ECON. PERSP. 105 (2003). Bonica, supra note 168, at 2, reaches a similar conclusion. 177 Aggarwal et al., supra note 16, at 12–13. 178 Bebchuk & Jackson, Shining Light, supra note 27, at 934. 179 Id. at 934–35 (reviewing the amount of contributions made to political intermediaries that have voluntarily chosen to disclose these contributions). See also Coates, supra note 24, at 661–63. 180 Bebchuk & Jackson, Shining Light, supra note 27, at 935–36. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 649 distortions in disclosure policies caused by positive externalities. However, there is one caveat before concluding that the disclosure of political spending is not a plausible candidate for concerns about distortions caused by positive externalities. Even though the dollar amounts involved in political spending are small, political spending information does appear to provide valuable information about the future performance of the firm’s securities. Aggarwal et al. study the stock returns of firms for which political spending information is publicly available during the period from 1991 to 2004, and find that higher levels of political spending are associated with negative future stock returns.181 This finding might be interpreted as providing a justification for requiring the disclosure of political spending, because this finding might suggest that political spending is the type of information firms would choose to disclose, but for the effects of a market failure. But a deeper analysis of the link between the value- relevance of political spending uncovered by Aggarwal et al. and the justifications for mandatory disclosure requirements is necessary to fully understand the relevance of findings such as those by Aggarwal et al. to the debate on the mandatory disclosure of political spending. How can the amounts involved in political spending be so trivial from the perspective of total firm value, and yet provide investors useful information about future stock performance? The answer to this puzzle is that these expenditures, although trivially small, must provide an informative signal about some other otherwise unobservable firm attributes that do affect future firm value. 181 Aggarwal et al., supra note 16. More specifically, the Aggarwal et al. study analyzes the relationships between future stock returns and public company political donations made in the form of soft money contributions and 527 Committee donations. Id. at 6. Aggarwal et al. also find that in a regression that controls for other publicly available information about corporate governance, the “negative relation between donations and returns is exacerbated.” Id. at 25. GUTTENTAG – FINAL 650 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 The magnitude of the decline in firm value associated with political spending reported by Aggarwal et al. reveals that political spending provides an informative signal, rather than directly causing, future declines in firm value.182 Aggarwal et al. report that for the median firm making a political contribution in their sample, a $10,000 political contribution is correlated with a $1.33 million decline in firm value.183 This means that each $1 spent on politics is correlated with a $130 loss in value. This huge multiple (over 100 times) makes clear that political spending is an informative signal about, rather than the actual cause of, the subsequent decline in firm value.184 182 Another example of how information about trivially small expenditures can provide an informative signal about firm value comes from David Yermack’s research on the personal use of corporate jets by public company CEOs. David Yermack, Flights of Fancy: Corporate Jets, CEO Perquisites, and Inferior Shareholder Returns, 80 J. FIN. ECON. 211 (2006). Yermack finds that increased personal use of corporate jets by these CEOs is correlated with significantly lower future stock returns for that firm. Id. at 213. This future decline in firm value is many, many multiples of the actual cost of the flights involved. Id. Therefore, rather than identifying a direct cause of future financial loss, Yermack’s analysis on the personal use of corporate jets provides an informative signal about other firm attributes that are associated with lower future stock returns. Yermack also finds that a firm’s stock price drops when personal use of corporate jets by the CEO is first disclosed. Id. Presumably, some investors realize the value-relevance of information about other attributes of the firm contained in the decision of the CEO to use corporate jets for personal trips. 183 Aggarwal et al., supra note 16, at 15. 184 Coates also rejects direct causation as providing a plausible explanation of the relationship between political spending and a future decline in firm value. Coates, supra note 24, at 659. However, Coates hypothesizes that there may be an indirect but causal link between more political spending and the decline in value, noting that “the costs of politics could extend far beyond direct costs to include opportunity costs of manager time, distraction and confusion for middle managers and employees, the risks of consumer backlash, and the risks that politically contingent operational investments turn sour.” Id. at 689. I am not convinced that spending $1 on politics could impose an average cost $130 on the firm, even through these indirect channels. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 651 The evidence that political spending is an informative signal about, rather than the cause of, future share price declines, significantly weakens an argument that mandatory disclosure of this information would provide a helpful corrective for positive externalities.185 Even if we accept the unlikely proposition that political spending information is withheld because of distortions caused by positive externalities, the benefits of corrective mandatory disclosure would probably be minimal in this context. The effects of mandating the disclosure of an informative signal will be different than the effects of mandating the disclosure of the underlying cause of a decline in firm value. Requiring the disclosure of an informative signal is more likely to alter the behavior producing the signal than to affect the behavior that is reducing the firm’s value. Looking to the analogy of personal use of corporate jets by public company CEOs is instructive, because such jet use is also a signal of future stock price underperformance.186 It would seem implausible that a ban on personal use of corporate jets (a more severe intervention than a disclosure requirement) would prevent the substantial decline in value correlated with this type of jet use, because corporate jet use was not the cause of the decline in firm value in the first place.187 Similarly, banning or requiring disclosure of political spending is unlikely to alter the underlying dynamics about which it provides an informative signal. One justification for the regulation of public company disclosure policies is that such regulation might be helpful in correcting for distortions in disclosure policies caused by 185 The related question of whether an informative signal of firm value should be subject to the materiality disclosure requirement raises an interesting question that, to the best of my knowledge, has not been previously considered elsewhere. Technically, an informative signal would seem to meet the U.S. Supreme Court’s definition of materiality as set out in Basic Inc. v. Levinson, 485 U.S. 224, 240 (1988), because it is information that would be of interest to a reasonable investor. This is a topic worthy of further consideration, but is beyond the scope of this Article. 186 See Yermack, supra note 182, at 211. 187 See id. at 225. GUTTENTAG – FINAL 652 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 positive externalities. The potential beneficiaries of public company disclosures whose interests may not be fully realized by firms when selecting disclosure policies are: (1) competitors, (2) other firms with securities traded on the same venue as the disclosing firm, and (3) the economy at large.188 The evidence does not suggest that more political spending information would be disclosed by public firms, but for distortions caused by unrealized benefits such disclosures would provide to any of these third parties. Therefore, the available evidence suggests that positive externalities do not justify mandating the disclosure of political spending by public companies. B. Reducing Tunneling and Political Spending Disclosure Public company disclosure regulation can also benefit shareholders by reducing tunneling. Without regulatory intervention public firms may not adopt a sufficient number of disclosure rules to maximize firm value.189 The possibility of reducing tunneling via political spending disclosure would appear to offer a more plausible justification than positive externalities for requiring this disclosure for two reasons. First, under-disclosure to facilitate tunneling appears to be a fairly common phenomenon.190 Second, and more specifically, evidence suggests that political spending often constitutes a form of tunneling.191 Two pieces of evidence, in particular, suggest that political spending is often carried out for the benefit of firm insiders rather than for the benefit of the firm as a whole. First, there is a high correlation between the political beliefs supported by the firm and those held by the firm’s CEO.192 This correlation suggests that CEOs may be spending firm resources to support political causes aligned with their own 188 See supra Part III.B.1. 189 See supra notes 158 and 159 and accompanying text. 190 See supra Part III.B.2. 191 See, e.g., Aggarwal et al., supra note 16, at 37. 192 Aggarwal et al., supra note 16, at 11. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 653 personal political beliefs, rather than because these expenditures are in the best interest of the firm. There are, however, explanations other than self-serving political spending that might explain this correlation. For example, CEOs may come to personally embrace the political beliefs that serve the best interests of their firms, or CEOs may choose to work at firms whose political interests are aligned with their own personal political beliefs. A second piece of evidence suggesting that political spending is not solely done to benefit firm shareholders is that there is a statistically significant correlation between higher levels of political spending and both sub-par future stock returns and firm attributes associated with weak corporate governance.193 The correlation between political spending and sub-par future stock returns might even naively be taken as providing evidence that wasteful political spending reduces firm value.194 The correlation between political spending and weak corporate governance similarly might suggest that much political spending is a form of tunneling, because the level of such spending is elevated when the mechanisms of corporate governance are less constraining on self-serving behavior generally. There are, however, reasons to be cautious in inferring causation from 193 The study by Aggarwal et al., for example, reports several indicia of such a correlation. First, Aggarwal et al. find that firms with high levels of political spending subsequently generated lower returns to the firm’s shareholders. Aggarwal et al., supra note 16, at 13–22. Second, Aggarwal et al. find that measures of poor corporate governance, including when a CEO is also Chairman of the Board, when the CEO receives abnormally high compensation, when there is less insider ownership of the firm, and when the firm has fewer block or institutional shareholders, are associated with higher amounts of political spending. Id. at 22–25. Aggarwal et al. conclude that their findings “support the view that lack of transparency allows donations to function as a form of private benefits for managers.” Id. at 37. Coates carries out a similar empirical research project, and his findings are similar. See Coates, supra note 24, at 658. Coates also finds that higher levels of political spending are correlated with various indicia of weak corporate governance and greater personal use of corporate jets. Coates, supra note 24, at 675–78. 194 But see supra notes 183 to 187 and accompanying text. GUTTENTAG – FINAL 654 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 these correlations between higher levels of political spending, sub-par future stock returns, and weaker corporate governance, in addition to the general admonition against inferring causation from correlation.195 A major factor arguing against there being a causal link between tunneling via political spending and a decline in firm value is the very small dollar amounts involved in political spending as compared to the observable effects on firm value. Recall that each dollar spent on politics is correlated with a $130 future decline in firm value.196 Because the statistical significance of political spending evidently comes from its role as an informative signal, correlations between political spending, future stock returns, and corporate governance are more likely to be the result of altering the nature of the signaling effect, rather than evidence of a causal relationship. But, more importantly, the relevant question is not whether political spending sometimes constitutes tunneling, but rather whether mandatory disclosure of political spending is likely to be a cost-effective means of reducing political spending that does not benefit the firm. There are three pieces of evidence that suggest mandating disclosure of political spending would not provide a cost-effective means to reduce this form of tunneling, each discussed more fully below. First, the dollar amounts involved in political spending are so trivially small that it seems unlikely that any disclosure requirement related to such spending could be justified solely based on the benefits from reducing tunneling. Second, the related British experience suggests such requirements will not have a significant deterrent effect. Third, previously unpublished findings from the Aggarwal et al. research are not supportive, despite appearances, of a causal link between more transparency and less wasteful political spending. 195 See, e.g., ROBERT M. LAWLESS, JENNIFER K. ROBBENNOLT & THOMAS S. ULEN, EMPIRICAL METHODS IN LAW 290 (1st ed. 2010). 196 Aggarwal et al., supra note 16, at 15. See also supra notes 182 to 184 and accompanying text. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 655 The first reason to doubt that a political spending disclosure requirement will deter self-serving political spending in a cost-effective manner is that, even in the best case, the potential deterrent benefits these disclosures might provide are going to be very small.197 For example, the average annual political spending by the firms studied by Aggarwal et al. was $44,000.198 Another estimate of the amounts of political spending per public company can be derived by considering the work of Bonica discussed above.199 Bonica conservatively estimated that the combined political spending by both public and private firms in the 2012 election cycle was not more than $400 million.200 If we also conservatively assume this was all spent only by public firms, then, since there are approximately 5,000 public firms in the United States, the aggregate amount of political spending by public firms would be $80,000 per firm.201 Even if the entire amount of a range between $2,000 and $80,000 per firm was wasteful and entirely eliminated by a political spending disclosure requirement, these savings would probably not be sufficient to offset a reasonable estimate of the costs of mandating disclosure of political spending by all public companies.202 197 See supra notes 168–177 and accompanying text. 198 Aggarwal et al., supra note 16, at 7 (this amount only includes the donations made in the form of soft money contributions and 527 Committee donations listed in Panel A, Table 1, because PAC Donations and Individual Money Donations are not expenditures funded by the firm, id. at 5). 199 Bonica, supra note 168. 200 Id. at 1. 201 See Strumpf, supra note 133 and accompanying text. 202 Bebchuk and Jackson hypothesize that the costs of requiring disclosure of political spending would be minimal. Bebchuk & Jackson, Shining Light, supra note 27, at 964–65. They do not provide specific cost estimates, however, and past experience suggests there are reasons to be skeptical of minimal cost estimates of topic-specific disclosure requirements. See Davidoff, supra note 79 (discussing the unanticipated costs of a requirement that public firms disclose the ratio between CEO compensation and average worker compensation). GUTTENTAG – FINAL 656 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 A second reason the evidence does not support a conclusion that mandatory disclosure of political spending will deter wasteful political spending in an efficient manner comes from a study of the effects of changes in the regulation of political spending by public companies in the United Kingdom. Ciara Torres-Spelliscy and Kathy Fogel analyzed political spending disclosure rules enacted in 2000 in the United Kingdom (“UK 2000 Amendments”).203 These rules required both shareholder preapproval and ongoing disclosure of political spending by public companies. Torres- Spelliscy and Fogel did not find evidence of a significant impact on the level of political spending by public companies in the United Kingdom as a result of this change in the law. Torres-Spelliscy and Fogel instead found that “corporate political spending by publicly traded companies remained relatively stable by aggregate during the 1993-2010 period.”204 In considering the extent to which an inference can be drawn from the British experience as to what might happen in the United States if a disclosure rule were imposed on public company political spending, two countervailing factors are noteworthy.205 On the one hand, in the period following the adoption of the UK 2000 Amendments, political contributions by private firms in the United Kingdom increased dramatically, suggesting that, perhaps, but for the UK 2000 Amendments, British public company political 203 Ciara Torres-Spelliscy & Kathy Fogel, Shareholder-Authorized Corporate Political Spending in the United Kingdom, 46 U.S.F. L. Rev. 525, 526–27 (2011). 204 Id. at 558. See also Saumya Prabhat & David M. Primo, Risky Business? Corporate Political Spending, Shareholder Approval, and Stock Volatility (Sept. 1, 2014) (unpublished manuscript), http://perma.cc/B78N- TUK4 (finding that the UK 2000 Amendments increased the stock volatility of politically active firms). 205 There are, of course, many additional challenges in drawing an inference from the failure of a mandatory disclosure requirement to alter political spending by public companies in the United Kingdom. As Torres- Spelliscy and Fogel observe, there are major differences between the British and United States political and corporate governance systems. Torres-Spelliscy & Fogel, supra note 203, at 536–42. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 657 spending might have increased more significantly during the period under study.206 On the other hand, the UK 2000 Amendments implemented two different hurdles to continued political spending by public companies. In addition to putting in place a disclosure requirement, the UK 2000 Amendments also required that all public company political spending above a minimal amount be pre-approved by shareholders.207 This additional hurdle suggests that imposing both the requirements of the UK 2000 Amendments would be more likely to have a deterrent effect than imposing a disclosure requirement alone, which is the proposal under consideration in the United States. A third reason to doubt that mandatory disclosure of political spending reduces tunneling is the lack of empirical evidence, despite appearances, of a causal link between more disclosure and less wasteful political spending. The one piece of evidence most suggestive of a causal link between more political spending disclosure and less tunneling comes from the Aggarwal et al. study.208 In a footnote, Aggarwal et al. write that: “[i]n unreported results, we have also examined the effect of greater disclosure (transparency) of political donations. We find that firms with greater disclosure of political donations have less negative donation-return relations, suggesting that greater transparency attenuates the negative effects of donations.”209 This footnote suggests that for firms that are more transparent about political spending, the political spending disclosures that do occur are less likely to be associated with sub-par future stock returns. This footnote could reasonably be interpreted as providing enough evidence of a causal relationship between more disclosure and less tunneling to justify the mandatory disclosure of political spending.210 206 Torres-Spelliscy & Fogel, supra note 203, at 561. 207 Id. at 545. 208 Aggarwal et al., supra note 16, at 25 n.14. 209 Id. 210 To be clear, Aggarwal et al. are certainly aware of the general and specific details discussed below that mandate caution in concluding their findings provide direct evidence of causation. GUTTENTAG – FINAL 658 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 Aggarwal et al. were kind enough to allow me to review and publish the data underlying the finding summarized in their footnote.211 My understanding of their analysis is as follows.212 First, Aggarwal et al. compiled information about the extent to which the firms they studied were transparent about this kind of spending in 2010.213 Aggarwal et al. then use this 2010 political spending transparency information to sort the firms whose political spending and subsequent stock returns they study (even though they studied firms during the period from 1991 through 2004). Aggarwal et al. find that within the subset of firms for which they were able to collect transparency estimates, the extent to which political spending was predictive of reduced future returns declined as transparency increased. Specifically, Aggarwal et al. find that in a regression with excess returns over a twelve-month period as the dependent variable, the coefficient for the interaction term between amounts of donations and transparency is positive at a statistically significant level.214 211 Letter from Tracy Yue Wang to author (June 4, 2013) (on file with author). 212 I, of course, assume full responsibility to the extent my description does not fairly represent their findings. 213 “[W]e use the 2010 Baruch Index of Corporate Political Disclosure, which assigns a score from 0 to 100 to each of the S&P 100 companies about the transparency of their corporate political spending and political activities. Of course, the index is constructed based on the information in 2010, which is outside our sample period. However, if the transparency of corporate political spending is negatively correlated with the degree of agency problems, then such transparency could be quite persistent over time. We create a variable called ‘Transparency’ based on the information in this index. ‘Transparency’=4 if 81<=index value<=100, =3 if 61<=index value<=80, =2 if 41<=index value<=60, =1 if 21<=index value<=40, and =0 if 0<=index value<=20.” Letter from Tracy Yue Wang to author, supra note 211. 214 The direct effect of donation on excess return (the -0.185* coefficient) is negative. The interaction effect on excess return between donation and transparency (the 0.064* coefficient) is positive at a statistically significant level. The full regression results are as follows: Dependent Variable Excess Return (S&P 100 Companies) GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 659 This means that for firms that were more transparent about political spending as measured in 2010, the correlation between more political spending and lower future returns was reversed to statistically significant degree.215 This finding could be interpreted as showing that requiring disclosure of political spending would have precisely the effect proponents of mandatory disclosure of political spending desire, namely reducing self-serving political spending. There are, however, three reasons for caution in inferring causation from the Aggarwal et al. findings of a moderating effect of political spending transparency on the link they otherwise find between political spending and a downturn in future returns. First, the most plausible explanation of the finding of a correlation between political spending and lower future returns that Aggarwal et al. report on is that political spending provides an informative signal, rather than actually causing declines in future stock prices. This is true because the dollars spent on political spending are so small, even as compared to the decline in value they predict.216 If political spending is associated with future stock price performance because it provides an informative signal, then Log(Donation) -0.185* (0.104) Transparency 0.007 (0.013) Log(Donation) x Transparency 0.064* (0.036) Constant 0.147** (0.062) Year Dummies Included # of Observations 927 R-squared 0.03 See Letter from Tracy Yue Wang to author, supra note 211. 215 “The direct effect of donation on excess return is still negative. However, it is less negative for firms that are more transparent in disclosing their corporate political spending and political activities.” Letter from Tracy Yue Wang to author, supra note 211. 216 See supra notes 183–184 and accompanying text. GUTTENTAG – FINAL 660 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 variables that moderate this relationship are generating these moderating effects through changes in signaling effects, rather than by actually changing behavior within the firm. In other words, the moderating effect observed by Aggarwal et al. is almost certainly the result of a refinement in the quality of the political spending signal, rather than because more transparency about political spending alters behavior. The second reason for caution before concluding from the Aggarwal et al. findings that there is a causal link between greater political spending transparency and less wasteful political spending comes from another finding (or lack thereof) in this regression. Aggarwal et al. do not find evidence of a statistically significant relationship between the extent of a firm’s transparency about political spending and that firm’s future stock performance among the firms for which political spending transparency data is available.217 It is incorrect to conclude based on the Aggarwal et al. interaction effect findings that a causal link exists between transparency and less wasteful political spending given the absence of a statistically significant correlation between additional transparency in political spending and lower future returns in this subsample.218 The third reason for caution before concluding from the Aggarwal et al. finding that there is a causal link between greater transparency and less wasteful political spending is that there is a more obvious explanation for the moderating effect they observe. Increased transparency should make information about a firm’s political spending more readily accessible, which means current spending is less likely to 217 The direct effect of transparency (the 0.007 coefficient) is not statistically significant. 218 Tracy Yue Wang noted the potential relevance of this coefficient in her correspondence as follows: “if transparency is negatively correlated with the degree of agency problems in the firm, then the political donation from more transparent firms may be less likely to due to agency motives, although the fact that the coefficient on transparency itself is insignificant weakens this argument.” Letter from Tracy Yue Wang to author, supra note 211. GUTTENTAG – FINAL No. 3:593] ON POLITICAL SPENDING DISCLOSURE 661 provide a valuable signal about losses at some point in the future.219 This explanation for the moderating effect of transparency on the relationship between political spending and future losses is, again, not related to changes in the actual amount of political spending. In summary, there are at least three reasons that the previously unpublished data underlying the Aggarwal et al. findings should probably not be interpreted as providing evidence of a causal a link between more transparency and less wasteful political spending. The idea that requiring public companies to disclose political spending will lead managers to engage in less wasteful political spending is intuitive and appealing. However, the available evidence does not suggest that requiring political spending disclosure will have this effect. Thus, there is no compelling evidence that either of the market failures that can provide a sound basis for regulating public company disclosure policies, positive externalities or the deterrence of tunneling, justifies the mandatory disclosure of political spending. In the absence of such evidence, the presumption against imposing new topic- specific disclosure requirements on public companies should prevail. V. CONCLUSION For those concerned about too much corporate money flowing into federal political contests, a rule requiring public companies to disclose political spending would seem to offer a reasonable, albeit imperfect, solution. The use of disclosure requirements is a common feature of the securities regulation regime in the United States. Disclosure rules provide an appealing way to avoid prohibiting firms from 219 Tracy Yue Wang also noted the potential relevance of this coefficient in her correspondence as follows: “There are two possible interpretations of this result. . . . Second, transparency means that the consequences of political donations are already priced and thus less likely to be reflected in abnormal returns.” Letter from Tracy Yue Wang to author, supra note 211. GUTTENTAG – FINAL 662 COLUMBIA BUSINESS LAW REVIEW [Vol. 2014 engaging in certain activities while, at the same time, ensuring that a firm’s investors are aware of important firm activities. Perhaps this is why such a preeminent group of scholars requested that the SEC begin the rulemaking process with respect to a political spending disclosure rule, and why this request has received a truly unprecedented amount of support. However, a careful review of the evidence, including previously unpublished empirical findings, shows that even those who favor increased disclosure of corporate spending in political contests generally should be hesitant to support a rule that would require only public companies to disclose political spending. Such a rule is unlikely to address the market failures that legitimately justify regulatory intervention into public company disclosure practices. In fact, to impose this disclosure requirement only on public firms could easily prove counter-productive by encouraging companies to exit the public company reporting regime. Looking at the proposal to require the disclosure of political spending by public companies in the proper context shows that this simple and appealing idea is neither necessary nor in the best interest of public companies or their shareholders.