ICI Comment Letter on Competition and Consumer Protection in the 21st Century (pdf)

August 20, 2018 Mr. Donald S. Clark Secretary Federal Trade Commission Office of the Secretary 600 Pennsylvania Avenue NW Suite CC–5610 (Annex C) Washington, DC 20580 Re: Competition and Consumer Protection in the 21st Century Hearings (Project Number P181201) Dear Mr. Clark: The Investment Company Institute (ICI)1 appreciates the opportunity to comment in advance of the Federal Trade Commission’s (FTC or Commission) hearings on competition and consumer protection in the 21st century.2 We welcome the Commission’s decision to solicit views on its near- and long-term law enforcement and policy agenda. ICI takes a keen interest in regulatory and policy developments that could affect the operation of the asset management industry, and thus the ability of Americans to meet their financial goals, such as saving for the purchase of a home, preparing for a secure retirement, or paying for higher education. Our comments address item 6c of the Hearing Notice, which requests comment on “the analysis of acquisitions and holding of a non-controlling ownership interest in competing companies.” This 1 The Investment Company Institute (ICI) is the leading association representing regulated funds globally, including mutual funds, exchange-traded funds (ETFs), closed-end funds, and unit investment trusts (UITs) in the United States, and similar funds offered to investors in jurisdictions worldwide. ICI seeks to encourage adherence to high ethical standards, promote public understanding, and otherwise advance the interests of funds, their shareholders, directors, and advisers. ICI’s members manage total assets of US$22.0 trillion in the United States, serving more than 100 million US shareholders, and US$7.6 trillion in assets in other jurisdictions. ICI carries out its international work through ICI Global, with offices in London, Hong Kong, and Washington, DC. 2 See Federal Trade Commission, Hearings on Competition and Consumer Protection in the 21st Century (June 20, 2018), available at https://www.ftc.gov/policy/advocacy/public-comment-topics-process (Hearing Notice). Mr. Donald S. Clark August 20, 2018 Page 2 of 26 practice, which refers to a type of “common ownership,” is the subject of considerable debate among some in the academic community. The debate focuses on two academic papers that allege common ownership by institutional investors decreases competition and raises consumer prices in concentrated industries, even when all common holdings are small (in percentage terms) and the institutional investor controls none of the commonly- held firms (the “common ownership hypothesis”).3 Although any investor could own stock in competing firms, this research depicts investment advisers that advise regulated funds as major shareholders in concentrated industries.4 Importantly, the proponents of the common ownership hypothesis concede that they have not established a mechanism by which common ownership allegedly reduces competition. More recent papers critiquing this early research describe numerous methodological and theoretical shortcomings of the common ownership hypothesis. Although the debate is now nearly four years old, fundamental questions remain unanswered, including: What is the proper methodology for assessing whether common ownership affects competition? Would advisers or their clients benefit from reduced competition in concentrated industries, as the common ownership hypothesis assumes? And how, and to what degree, would such shareholder incentives (to the extent that they exist across a broad and diverse shareholder base) cause firms in concentrated industries to pursue competitive strategies not intended to maximize own-firm profits? The common ownership hypothesis also has led some academics to propose measures to address the harms that allegedly arise from common ownership. In general terms, these measures would either (i) reduce the ability of investment advisers to acquire or hold shares in competing firms on behalf of their clients, or (ii) compel clients of investment advisers to forfeit certain rights if common ownership, as aggregated at the adviser level, exceeds a de minimis threshold. The papers proposing these measures 3 We recognize that economic literature sometimes uses the term “common ownership” to encompass all situations where a shareholder owns a stake in competing firms, regardless of the size of that stake. Consistent with the Hearing Notice, however, we use “common ownership” more narrowly in this letter to refer only to situations where an investment adviser’s clients hold, in aggregate, non-controlling ownership stakes of competing companies. For example, the common holdings that are the focus of the common ownership hypothesis generally represent less than 10 percent of each competing firm’s shares. See e.g., Edward B. Rock and Daniel L. Rubinfeld, Antitrust for Institutional Investors, New York University School of Law and Economic Research Paper Series, Working Paper No. 17-23 at 41-42 (July 2017), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2998296## (surveying relevant legal authority regarding what constitutes “control” and concluding that ownership stakes under 20 percent generally do not confer control, especially when there is no board representation). 4 Unless otherwise indicated, we use the term “investment advisers” or “advisers” throughout to mean investment advisers registered with the Securities and Exchange Commission (SEC); “funds” or “regulated funds” to include mutual funds, ETFs, closed-end funds, and UITs registered with the SEC; and “pooled investment vehicles” to include funds and any other managed vehicles through which investors seek to gain exposure to stocks, such as private funds (e.g., hedge funds) and common investment trusts. Mr. Donald S. Clark August 20, 2018 Page 3 of 26 assume the validity of the common ownership hypothesis and fail to adequately consider the harms their proposals would inflict on retail investors, companies, and the economy. We strongly disagree with the common ownership hypothesis because it rests on: (1) misunderstandings and misinformation about the asset management industry; (2) incorrect assumptions about the incentives of advisers and their clients; and (3) flawed empirical work. In other words, we believe there is no valid empirical basis to conclude that investment advisers—solely because they manage their clients’ broadly diversified investment portfolios, some of which include shares of competing companies—are causing competitive harm. Consequently, the common ownership hypothesis cannot be the basis for enforcement or a change in competition policy. Put simply, it would be inappropriate for the FTC or other authorities to consider measures that would restrict common ownership by institutional investors, or strip away important shareholder rights (e.g., proxy voting) belonging to the clients of investment advisers. Further, such measures likely would have a disruptive impact on the economy, capital markets, and retail investors (e.g., by upending 401(k) plans’ ability to offer a well-diversified menu of funds to workers). To inform the hearing record, the remainder of this letter describes these issues in greater detail.  Section I describes relevant aspects of asset management practices and regulation to provide a factual baseline—which should anchor any analysis of common ownership. We highlight where and how some of the current academic work fails to accurately understand or describe the operations of investment advisers. ICI is uniquely situated to assist in this process. As the leading association representing regulated funds globally, we have a deep knowledge of the asset management industry, including its regulatory framework and how it functions in practice, and the markets in which it operates.  Section II reviews the academic debate over whether common ownership affects competition. We examine both the preliminary papers asserting that it does and the newer papers challenging the preliminary research, demonstrating clearly that this debate is far from settled.  Section III describes the potential harms associated with proposed measures to address the alleged anticompetitive effects of common ownership. We explain why we agree with commentators, including academics and regulators, who believe that consideration of such measures is inappropriate. I. Analysis of Common Ownership Requires an Accurate Understanding of the Asset Management Industry It is axiomatic that a detailed and comprehensive understanding of an industry should underlie academic study of that industry. In this section, we provide the foundational understanding of the asset Mr. Donald S. Clark August 20, 2018 Page 4 of 26 management industry (with a focus on management of publicly-traded equity securities on behalf of clients) that is lacking in some of the common ownership literature. Specifically, we:  Draw the critical distinction between clients (i.e., the asset owners) and advisers (i.e., the entities that manage client assets in an agency capacity) and describe the nature of the relationship between the adviser and its client;  Explain the variety of client types that an adviser serves and the diversity of investment mandates that the adviser employs on behalf of those clients;  Outline the regulatory requirements that shape advisers’ and regulated funds’ proxy voting activity and provide an overview of how they fulfill their responsibilities; and  Identify the laws and regulation that permit and constrain adviser/portfolio company engagement generally. In Appendix A, we provide an overview of the organizational and regulatory structure that governs advisers and regulated funds. As of July 2018, there were approximately 13,000 registered investment advisers in the US, with over $83 trillion in assets under management.5 As of June 2018, there were approximately 6,300 regulated equity funds in the US, with almost $13.4 trillion in assets under management. Nearly 50 million US households own shares in equity mutual funds.6 Common ownership hypothesis proponents err in that they (i) conflate asset ownership and asset management, essentially treating an adviser and its clients as a monolithic entity with a single and easily discernable interest, and (ii) miscast advisers’ and regulated funds’ engagement activities, in their search for a mechanism by which advisers could adversely influence the competitive strategies of clients’ portfolio companies. As described in Section II below, these defects seriously undermine the validity of the proponents’ conclusions. A. Investment Advisers Manage Client Assets—They Are Not Asset Owners Investment advisers manage assets owned by their clients. An adviser’s “assets under management” are not its assets. Rather, the term refers to all client assets over which the adviser exercises investment discretion. Acting as agent, an adviser manages these assets in accordance with the stated investment 5 See Information About Registered Investment Advisers and Exempt Reporting Advisers, available at https://www.sec.gov/ help/foiadocsinvafoiahtm.html. 6 See 2018 Investment Company Factbook at 143, available at https://www.ici.org/pdf/2018_factbook.pdf. Mr. Donald S. Clark August 20, 2018 Page 5 of 26 objectives, strategies, and policies of each client. Portfolio results, positive or negative, belong to clients, who accept the risk that their investments may lose value. An adviser may serve investors through pooled investment vehicles (in which case the vehicles are the “clients”) or through separately managed accounts (in which case the account owner is the client). Either way, the client is the economic owner of the assets. A pooled investment vehicle is the owner of the shares of its portfolio companies; the vehicle’s investors in turn have pro rata ownership interests in the vehicle. Although an adviser may be the record owner of company shares in which its clients invest, the adviser may exercise certain rights (e.g., proxy voting) on behalf of clients only to the extent authorized by the client. An investment management agreement establishes the terms and conditions of service between the adviser and each client. The adviser and client agree in advance about how the adviser will manage the client’s assets, with the investment management agreement and frequently another ancillary document (e.g., a regulated fund prospectus) setting forth the investment objective, strategies, guidelines, and policies that the adviser will follow.7 More generally, and importantly, the adviser must act consistently with its fiduciary obligations to its clients. As a fiduciary, an adviser owes its clients a duty of loyalty and a duty of care. The Supreme Court has recognized the fiduciary nature of an investment advisory relationship, and found that an adviser’s fiduciary obligations are enforceable under the Investment Advisers Act of 1940 (Advisers Act).8 B. Advisers Provide a Variety of Advisory Services to Clients A single large adviser often maintains hundreds or thousands of separate client relationships. Clients may include, among others, individuals, funds, corporations, pension and profit sharing plans, charitable organizations, other investment advisers, and government entities. Just as its clients are varied in number and type, so too are the strategies that an adviser may employ on behalf of its clients. In the regulated fund industry, for example, an adviser may employ an index-based investment strategy for a fund. The investment objective of that fund is to track the performance of that index as closely as possible—irrespective of whether that index increases or decreases in value. 7 As discussed in Appendix A, the Investment Company Act of 1940 (Investment Company Act) and rules thereunder subject advisers to additional requirements affecting the management and operation of their regulated fund clients, including portfolio management limitations. 8 SEC v. Capital Gains Research Bureau, Inc., 375 U.S. 180, 191 (1963). The SEC has adopted a series of antifraud rules under Section 206(4) of the Advisers Act, including Rule 206(4)-8, which prohibits advisers from defrauding investors and prospective investors in pooled investment vehicles. Mr. Donald S. Clark August 20, 2018 Page 6 of 26 Because minimizing tracking error is of paramount importance, the adviser’s investment discretion generally is limited to determining the most efficient means of doing so. By contrast, an investor choosing an “active” investment strategy generally seeks some investment objective (e.g., capital appreciation or growth) and provides the adviser with the investment discretion to determine the best means of achieving that objective. An adviser employing active strategies therefore has much more latitude in determining the stocks that it will buy, hold, and sell on behalf of its clients to pursue the client’s investment objective—indeed, the skill with which the adviser exercises this discretion is an important part of its value proposition. If such an adviser becomes dissatisfied with a portfolio company for any reason, or believes that better opportunities exist, it generally may dispose of a client’s shares. An adviser’s overall discretion, even when using active strategies, still is subject to agreed-upon guidelines and limitations, and in some cases statutory and regulatory limitations. These limitations may relate to permissible investments, style, and strategy. From this brief overview, three key points emerge. First, while the advisory relationship affords advisers with discretion in the day-to-day management of a client portfolio, the adviser works for the client. Clients may and do fire their advisers, or may hire other advisers to manage some or all of their assets. In the context of regulated funds, investors may redeem some or all of their fund shares or dispose of them on the secondary market (as applicable), and with those proceeds buy shares in other funds or make other investments. Second, regardless of the strategy employed, restrictions placed on the adviser are critically important. The adviser must manage client assets within these agreed-upon parameters. An investor’s overall portfolio may consist of numerous funds and/or accounts. For an investor’s overall portfolio to work in a coordinated and efficient way, it is essential that the constituent parts of the portfolio be managed as expected. For example, US households on average hold six mutual funds, and 26 percent of US households owned seven of more mutual funds in 2017.9 In addition, US households acquired these funds on average from two different sources outside employer-sponsored retirement plans, and 37 percent of US households used three or more sources.10 Two points emerge from this. Not only is it critical to investors that advisers adhere to the terms of the investment agreement, but investors, like advisers, pursue diverse strategies and access advice in a variety of ways. Third, given differing client strategies, an adviser may take different views on the same stock. For instance, clients following an index-based strategy may hold a particular stock because of its inclusion in the index, while clients following an active strategy may sell that same stock or take positions in a competing company. Clients also may prohibit advisers from investing in certain stocks or industries. Organizationally, an adviser may have multiple units and/or portfolio management teams, each with unique expertise and different client relationships to manage. Thus, given this diversity of clients and 9 See Profile of Mutual Fund Shareholders, 2017 at 13, available at https://www.ici.org/pdf/rpt_17_profiles17.pdf. 10 See id. Mr. Donald S. Clark August 20, 2018 Page 7 of 26 strategies, it is incorrect to assume that an adviser simply takes a monolithic “view” of the investment merits of a stock (or an industry), and then applies that view across all client accounts in a rigid top- down way. C. Advisers Vote Client Proxies to Enhance Client Value and Satisfy Legal Obligations The ability to vote proxies on certain matters is one right that corporate shareholders typically possess. These voting matters are determined by the company’s organizational documents (e.g., charter and/or by-laws), applicable state law, and any applicable exchange regulations. In practice, these matters are limited with respect to subject matter and generally do not involve management’s day-to-day prerogatives in running the business. They include routine issues (e.g., election of directors or ratification of independent auditors) and non-routine issues (e.g., changes to the corporation’s governing documents or proposed mergers).11 In general, proxy voting is binary: the shareholder may vote “for” or “against” a proposal (or may “abstain” from voting), but may not express more nuanced views. The SEC has played a critical role in shaping funds’ and advisers’ current proxy voting practices. In its 2003 companion rulemakings, the SEC (i) stated that a fund’s board of directors, acting on the fund’s behalf, has the “right and the obligation” (emphasis added) to vote proxies,12 and (ii) couched advisers’ proxy voting responsibilities in terms of their fiduciary duties.13 11 Over the period from 2010 to 2015, routine proposals, such as the election of directors or ratification of independent auditors, were between 80 and 90 percent of all proposals. See Ryan Bubb and Emiliano Catan, The Party Structure of Mutual Funds (February 14, 2018) at Table 1, available at: https://ssrn.com/abstract=3124039. 12 Disclosure of Proxy Voting Policies and Proxy Voting Records by Registered Management Investment Companies, SEC Release No. IC-25922 (January 31, 2003), available at www.sec.gov/rules/final/33-8188.htm. In discussing the policy rationale for the fund disclosure requirements, the SEC stated, “Proxy voting decisions by funds can play an important role in maximizing the value of the funds’ investments, thereby having an enormous impact on the financial livelihood of millions of Americans. … Finally, requiring greater transparency of proxy voting by funds may encourage funds to become more engaged in corporate governance of issuers held in their portfolios, which may benefit all investors and not just fund shareholders.” 13 Proxy Voting by Investment Advisers, SEC Release No. IA-2106 (Jan. 31, 2003), available at www.sec.gov/rules/final/ia- 2106.htm (“[A]n adviser is a fiduciary that owes each of its clients duties of care and loyalty with respect to all services undertaken on the client's behalf, including proxy voting. The duty of care requires an adviser with proxy voting authority to monitor corporate events and to vote the proxies. To satisfy its duty of loyalty, the adviser must cast the proxy votes in a manner consistent with the best interest of its client and must not subrogate client interests to its own.”). The Department of Labor (DOL) has taken a similarly muscular view of plan fiduciaries’ proxy voting obligations. See generally Interpretive Bulletin Relating to the Exercise of Shareholder Rights and Written Statements of Investment Policy, Including Proxy Voting Policies or Guidelines, 81 Fed. Reg. 95879 (Dec. 29, 2016), at www.gpo.gov/fdsys/pkg/FR-2016-12- 29/pdf/2016-31515.pdf. See also Department of Labor Field Assistance Bulletin 2018-01(Apr. 23, 2018) (reaffirming DOL’s “longstanding view [ ] that plan fiduciaries should engage in traditional and customary proxy voting activities in discharging Mr. Donald S. Clark August 20, 2018 Page 8 of 26 The SEC’s proxy voting framework subjects advisers to substantive and disclosure-oriented requirements.14 Clients may retain, or delegate to an adviser,15 proxy voting responsibility. Advisers with proxy voting authority over client securities must adopt and implement proxy voting policies and procedures reasonably designed to ensure that the adviser votes client proxies in the best interest of the client, and these policies and procedures must address how the adviser will handle material conflicts of interest between the adviser and its clients. Regulated funds are subject to separate proxy voting requirements. A fund must (i) describe in its registration statement the policies and procedures that it uses to determine how to vote proxies relating to its portfolio securities, and (ii) publicly file with the SEC the fund’s records of how it voted proxies relating to its portfolio securities. Regulated funds are unique in this regard—no other type of institutional investor must file with the SEC and publicly disclose how it voted each of its proxies. Regulated funds are also notable in the role that their boards play in proxy voting. A regulated fund’s board of directors is responsible for the voting of proxies relating to the fund’s portfolio securities on behalf of the fund. A fund board typically delegates responsibly to the fund’s adviser. The board also must review and approve the fund’s proxy voting policies and procedures,16 and proxy voting policies and procedures are part of those compliance programs subject to the board review requirements of Investment Company Act Rule 38a-1. Thus, the organizational and regulatory structure of funds contains safeguards that help ensure that advisers will act in funds’ best interest when voting fund proxies. Guided and motivated by these legal obligations, regulated fund boards and fund advisers take their proxy voting responsibilities seriously and devote substantial resources to this function. Because of the large number of portfolio securities that advisory clients may hold, efficient and informed proxy voting is a large undertaking for advisers and regulated funds. One way that fund advisers efficiently handle such a large volume of proxy proposals is through formulation and maintenance of proxy voting guidelines, which they apply to these proposals. Proxy voting guidelines specify how an adviser will vote their fiduciary obligation to prudently manage plan investments”), available at www.dol.gov/sites/default/files/ebsa/employers-and-advisers/guidance/field-assistance-bulletins/2018-01.pdf. 14 See Appendix A for a more detailed discussion of these legal requirements. 15 See, e.g., BlackRock ViewPoint: The Investment Stewardship Ecosystem (July 2018) at 13, available at https://www.blackrock.com/corporate/literature/whitepaper/viewpoint-investment-stewardship-ecosystem-july-2018.pdf (estimating that approximately one-quarter of equity separate account assets managed by BlackRock do not delegate voting authority to BlackRock (representing approximately 9 percent of assets in equity mandates managed by BlackRock), and that across equity holdings managed by BlackRock, approximately 8 percent of assets under management is outsourced to an independent fiduciary to vote). 16 A fund board may adopt separate fund policies and procedures or may adopt or rely on the adviser’s policies and procedures. Mr. Donald S. Clark August 20, 2018 Page 9 of 26 on behalf of its clients (including regulated funds) on various proxy proposals. For instance, the guidelines may specify the circumstances under which the fund adviser generally will vote for or against director nominees, executive compensation plans, or social or environmental issues. Use of guidelines helps ensure consistency in proxy voting and can help protect against potential conflicts of interest.17 Advisers also may hire third parties such as proxy advisory firms to assist in carrying out proxy voting responsibilities.18 Some advisers have individuals or teams dedicated to analyzing proposals and voting proxies on behalf of clients, while others have a more decentralized approach (e.g., individual portfolio managers or others may make voting decisions).19 Irrespective of how an adviser internally manages proxy voting responsibilities, clients ultimately determine the scope of an adviser’s proxy voting responsibilities, and fiduciary duty compels the adviser to vote a client’s shares solely in the client’s best interest when it has voting discretion. Consequently, an adviser with investment discretion may not always vote client shares in a uniform manner. For example:  As noted above, a client may retain proxy voting responsibility, or may share this responsibility with its advisers in some agreed-upon way.  Depending on client preferences, an adviser may adopt and apply different proxy voting guidelines.  Clients’ economic interests may differ. To illustrate, suppose fund A owns shares of the acquiring company in a proposed merger, while fund B owns shares of the target company. If advisory personnel believe the acquiring company would be overpaying for the target company, these funds will vote differently (fund A would vote against, and fund B for, the transaction).  Certain votes present an adviser with difficult judgment calls. Personnel within the adviser (e.g., portfolio managers) may very well reach different conclusions regarding the best means of advancing clients’ interests and vote the proxies for which they have responsibility differently. 17 Of course, application of proxy voting guidelines, no matter how detailed, to proposals will not always yield obvious voting decisions. Some proposals will not have a corresponding guideline that is clearly applicable, and often a fund adviser’s guidelines will require “case-by-case” evaluations of certain complicated or fact-specific proposals (such as merger proposals). 18 See generally IDC and ICI Report on Funds’ Use of Proxy Advisory Firms (January 2015), available at www.ici.org/pdf/pub_15_proxy_advisory_firms.pdf (describing generally (i) the services that proxy advisory firms provide to advisers and funds, and (ii) fund advisers’ due diligence and oversight of proxy advisory firms). 19 An adviser with a more centralized approach to proxy voting still may grant others (e.g., portfolio managers) input or even ultimate discretion in how the adviser votes client proxies. Mr. Donald S. Clark August 20, 2018 Page 10 of 26 Thus, one should not assume that all client shares for which an adviser has investment discretion are voted by the adviser, nor should one assume that the adviser will vote all client shares for which it has voting discretion in a uniform manner. D. Other Forms of Engagement Are Permitted and Circumscribed by Federal Securities Laws Efforts to enhance value for clients can take many forms, including proxy voting and other engagement with portfolio companies. The federal securities laws establish parameters for these other engagements and circumscribe communication and conduct both for investors (and their advisers) and portfolio companies.20 The most relevant of these laws and regulations include the following:  Regulation FD (Fair Disclosure) addresses the selective disclosure of information by publicly traded companies and other issuers; it provides that when an issuer discloses material non-public information (MNPI) to certain individuals or entities—e.g., advisers and funds—the issuer must make public disclosure of that information.21  Federal securities laws generally prohibit insider trading on the basis of MNPI; advisers have affirmative obligations under the Advisers Act to (i) establish, maintain, and enforce written policies and procedures reasonably designed to prevent the misuse of MNPI, and (ii) establish, maintain, and enforce written codes of ethics.22 Further, securities law reporting requirements apply to holders that acquire a significant position in an issuer.23 Section 13(d) of the Securities Exchange Act of 1934 (Exchange Act) requires any person who, after acquiring directly or indirectly the beneficial ownership of an equity security registered under the Exchange Act, is the “beneficial owner” of more than five percent of such class of securities, to file with the SEC certain information on Schedule 13D. Section 13(g), however, permits certain investors to file the less onerous Schedule 13G, provided that the investor acquired the securities with no purpose or effect of changing or influencing the control of the issuer, and not in connection with or as a participant in any transaction having such purpose or effect. The SEC has stated that most solicitations 20 In addition, the federal antitrust laws, which have general application, seek to prohibit anticompetitive behavior and unfair business practices while encouraging competition in the marketplace. Section 1 of the Sherman Act, for example, outlaws “contracts, combinations . . . or conspiracies” in restraint of trade, and could cover cases where a third party acts as a “hub” in a “hub-and-spoke” conspiracy, or a “conduit” of sensitive information to competing firms. 21 See Selective Disclosure and Insider Trading, SEC Release 33-7881 (Aug. 15, 2000), available at www.sec.gov/rules/ final/33-7881.htm. Improper disclosure of confidential information by company officers and/or directors also may constitute a breach of fiduciary duty under state corporate law or common law. 22 See Appendix A for a more detailed summary of these provisions. 23 See Appendix B for a more detailed summary of these provisions. Mr. Donald S. Clark August 20, 2018 Page 11 of 26 regarding social or public interest issues (e.g., environmental policies) would not have the purpose or effect of changing or influencing control of the company (and therefore would not prevent the use of Schedule 13G), and that proposals and soliciting activity related to general corporate governance matters such as executive compensation, director pensions, and confidential voting also normally would not prevent the use of Schedule 13G.24 And the SEC staff has provided similar guidance, identifying forms of engagement subject matter—e.g., executive compensation and social or public interest issues and corporate governance topics—that, without more, generally would not preclude a shareholder from filing on Schedule 13G.25 This SEC guidance has helped shape advisers’ views regarding permissible subject matter for engagement. Based on ICI’s review of filings on EDGAR, during the 12-month period-ended July 31, 2018, on a combined basis the five largest regulated fund advisers (based on equity assets under management) made over 7,000 Schedule 13G filings (and amendments) and only one Schedule 13D filing. II. Research Linking Common Ownership to Decreased Competition Is Hotly Disputed With the preceding information about the asset management industry and regulated funds as a factual baseline, this section provides an overview of the main academic contributions to advancing and disputing the common ownership hypothesis. We start with a description of the initial papers that claim common ownership decreases competition and other papers that search for the mechanisms that cause reduced competition. We then turn to more recent literature that casts doubt on the findings of the preliminary research. We provide some supplemental information about papers on the common ownership hypothesis in Appendix C. A. Preliminary Papers Assert that Common Ownership Decreases Competition and Postulate Potential Mechanisms for This Alleged Anticompetitive Effect The research supporting the common ownership hypothesis has two strands: first, empirical research claiming that common ownership decreases competition; second, research that endeavors to identify the mechanism through which competition is allegedly reduced. Although we treat these strands as distinct, serious consideration of the second strand depends entirely on the validity of the first. 24 Amendments to Beneficial Ownership Reporting Requirements, SEC Release No. 34-39538 (Jan. 12, 1998), available at https://www.sec.gov/rules/final/34-39538.txt. 25 Compliance and Disclosure Interpretations, Exchange Act Sections 13(d) and 13(g) and Regulation 13D-G Beneficial Ownership Reporting, Question 103.11, available at https://www.sec.gov/divisions/corpfin/guidance/reg13d-interp.htm. Mr. Donald S. Clark August 20, 2018 Page 12 of 26 1. Preliminary Empirical Work Claims Common Ownership Decreases Competition The common ownership hypothesis originates primarily from two empirical papers that contend the simultaneous ownership of stock in competing firms in concentrated industries by institutional investors decreases competition, even though the institutional investor controls none of the competing firms. According to an early draft of one paper (the “Airline Paper”), “ticket prices are approximately 3- 5% higher on the average US airline route” as a result of common ownership.26 Another paper (the “Banking Paper”) makes similar claims about the US banking industry, arguing that increases in common ownership of banks causes higher prices and fees paid by consumers.27 The theoretical framework for both papers assumes that a shareholder of competing firms would prefer that these companies take into account the effects of their actions on industry profits, not just firm profits, and that the companies would in fact do so because of this preference. The Airline and Banking Papers also use similar empirical methodologies. They both first identify concentrated industries for which pricing information is readily available publicly. Next, the papers assess the level of common ownership in these industries using a database that relies on publicly available Form 13F filings that large institutional investors make to the SEC.28 The Airline paper then claims to causally link common ownership to higher prices using as the main explanatory variable MHHI delta—the difference between the modified Herfindahl-Hirschman Index (MHHI) and the Herfindahl-Hirschman Index (HHI)—to measure the effect of common ownership.29 The regression 26 See José Azar, Martin C. Schmalz, and Isabel Tecu, Anticompetitive Effects of Common Ownership (January 30, 2015) at 3, available at http://www.utahwfc.org/uploads/2015_10b.pdf. The most recent version of this paper, which claims a 3-7% increase in airline ticket prices, will be published in the Journal of Finance. See infra, note 30. 27 José Azar, Sahil Raina, and Martin Schmalz, Ultimate Ownership and Bank Competition (July 24, 2016), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2710252. 28 Form 13F data are known to have significant and numerous limitations. See Appendix B. See also Anne Anderson and Paul Brockman, Form 13F (Mis)Filings (October 15, 2016), available at https://papers.ssrn.com/sol3/ papers.cfm?abstract_id=2809128 (arguing that, because of the errors in 13F filings, “reliance on 13F filings is unwarranted for both private investors and capital markets researchers.”); Martin C. Schmalz, Common-Ownership Concentration and Corporate Conduct, CESifo Working Paper No. 6908 (May 2018), available at https://papers.ssrn.com/sol3/ papers.cfm?abstract_id=3046829## (acknowledging that relying solely on ownership data reported on 13F filings “can be inadequate” to construct measures of common ownership concentration); and Rock and Rubinfeld, supra note 3 at 13-14, n. 33 (stating that “13Fs are notoriously inaccurate” and incomplete for various reasons, including that they “do not distinguish between the right to sell and the right to vote…”). 29 The HHI is a commonly-accepted measure of market concentration. The HHI is calculated by squaring the market share of each firm competing in the market and then summing the resulting numbers. The MHHI adjusts the HHI to account for the effects of common ownership reflected in cash flow rights and control/voting rights. Since HHI and MHHI depend on market shares, MHHI delta also depends on market shares. See Timothy F. Bresnahan and Steven C. Salop, Quantifying the Competitive Effects of Production Joint Ventures, 4 International Journal of Industrial Organization 155 (1986). Mr. Donald S. Clark August 20, 2018 Page 13 of 26 results in the paper indicate that prices for airline tickets are positively correlated with the MHHI delta. The Banking Paper associates the price of banking services with a generalized Herfindahl-Hirschman Index (GHHI), which is related to MHHI. It presents results that indicate higher prices are positively correlated with larger values of GHHI. The authors, at least in the initial drafts of these papers, interpreted these correlations as proving that common ownership causes consumer prices to be higher than they would be without common ownership. 2. Efforts to Identify a Mechanism Through Which Common Ownership Decreases Competition Have Been Unsuccessful The final version of the Airline Paper concludes by offering thoughts on the mechanisms that might enable a minority, non-controlling shareholder to affect the competitive strategy of its commonly- owned firms. This section is notably tentative, discussing “potential mechanisms…that could implement these results.”30 The paper maintains that higher prices could result from common owners’ “doing nothing” to encourage competition, while also suggesting that the culprit(s) could be common owners’ direct forms of engagement (e.g., “voice,” “incentives,” and the “vote”).31 Under “incentives,” the paper argues that shareholders in competing firms might encourage adoption of executive compensation structures that reward management for competing less aggressively.32 Alternatively (or in addition), the paper hypothesizes, common owners might act to reduce competition by communicating preferred market strategies to corporate managers (“voice”) or by voting against management (e.g., director candidates) as a means of signaling discontent.33 Other papers focus primarily on what the Airline Paper refers to as “incentives” to reduce competition. One paper theorizes that common owners may support executive compensation practices that reward corporate managers based on the success of an industry rather than the success of their firm.34 Another paper in this same vein—and cited in the Airline Paper—claims that executives in industries with greater common ownership generally derive a greater portion of their compensation from industry performance.35 Consequently, the theory goes, these executives ought to have an incentive to focus on 30 Airline Paper, Journal of Finance, Forthcoming, at 42, available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id =2427345. Emphasis added. 31 Id. at 43-52. 32 Id. at 49-51. 33 Id. at 43-44. 34 See Einer Elhauge, Horizontal Shareholding, 129 Harv. L. Rev. 1267 (2016), available at https://www.antitrustinstitute. org/sites/default/files/Elhauge.pdf. 35 See Miguel Anton, Florian Ederer, Mireia Gine, and Martin C. Schmalz, Common Ownership, Competition, and Top Management Incentives, Ross School of Business Paper No. 1328 (June 1, 2018), available at https://ssrn.com/abstract =2802332. Mr. Donald S. Clark August 20, 2018 Page 14 of 26 maximizing industry profits rather than exclusively pursuing strategies that would maximize the value of their own firms.36 Importantly, no academic paper empirically confirms any conjectured mechanism where minority, non- controlling shareholders of competing firms affect these firms’ competitive strategies. B. Other Papers Challenge the Preliminary Research The initial research on common ownership sparked a vigorous response from academics who question the theoretical underpinnings and findings of the Airline and Banking Papers. This section describes methodological critiques of the Airline and Banking Papers, explains newer empirical research that finds common ownership does not have a significant effect on competition in the airline or banking industries, and summarizes other literature that questions assumptions related to incentives and the potential mechanisms that proponents of the common ownership hypothesis suggest could create anticompetitive effects. 1. New Research Questions the Validity of the Airline and Banking Papers’ Methodology Several papers argue that the Airline and Banking Papers fail to prove that common ownership causes higher prices. The difficulty is that the MHHI delta—the key explanatory variable in both the Airline and Banking Papers—depends on market shares, which, in turn, depend on the same factors that generally drive prices.37 Because of this endogeneity, the regression estimates in the Airline and Banking Papers are likely to yield a correlation between MHHI delta and price, even if common ownership has no effect on prices, and this correlation would be positive under plausible circumstances.38 36 See id. 37 See Daniel P. O’Brien and Keith Waehrer, The Competitive Effects of Common Ownership: We Know Less Than We Think, 81 Antitrust Law Journal No. 3 (2017). As noted on the first page of this article, ICI funded this research, in part, but neither ICI nor any of its members provided any substantive input into the analysis in the article. Moreover, Dr. O’Brien publicly raised concerns with the methodology of the Airline Paper in 2015, before he had any business relationship with ICI. See Daniel P. O’Brien, Anticompetitive Effects of Common Ownership: Overview of the Theory, and Review of the Empirical Findings of Azar, et al., Transportation, Energy & Antitrust (Fall 2015). See also Jacob Gramlich and Serafin Grundl, Testing for Competitive Effects of Common Ownership, Finance and Economics Discussion Series 2017-29, Division of Research & Statistics and Monetary Affairs, Board of Governors of the Federal Reserve System, available at https://www.federalreserve.gov/econres/feds/files/2017029pap.pdf (arguing that alternative approaches to analyzing the competitive effects of common ownership would reduce endogeneity concerns associated with using concentration measures); Rock and Rubinfeld, supra note 3 at 19-22. 38 See O’Brien and Waehrer, supra note 37; Patrick J. Dennis, Kristopher Gerardi, and Carola Schenone, Common Ownership Does Not Have Anti-Competitive Effects in the Airline Industry (February 2018), available at Mr. Donald S. Clark August 20, 2018 Page 15 of 26 Consider the following example. Market shares change over time for a variety of reasons, including some that are unrelated to common ownership or price. Airline ticket prices to popular ski destinations may rise during winter because demand increases for tickets to those locations. An airline’s share of the market to a ski destination also might increase during winter if it can expand capacity—by adding flights or flying bigger planes—and competitors cannot. In this case, the higher prices and the increased market share of the more flexible airline are each independently due to the presence of snow and thus, more skiers, and have nothing to do with the extent to which there is common ownership of the airlines. One paper also notes that the Airline and Banking Papers make unjustified assumptions about the influence that minority shareholders have on corporate managers.39 Specifically, the papers supporting the common ownership hypothesis assume that if an institutional investor owns even a small share of a company’s stock, the investor has influence over the company’s business strategy that is proportional to its holdings, i.e., the influence would rise with even small increases in the investor’s ownership share. But there is no empirical evidence to support this assumption. Nevertheless, the assumption of proportionate control likely generates a spurious relationship between price and the MHHI delta. 2. New Papers Offer Improved Empirical Models to Assess Whether Common Ownership Affects Competition Several researchers have developed alternative models to address the methodological shortcomings of the Airline and Banking Papers. These alternative models are described in three papers, two of which address the Airline Paper while the third focuses on the Banking Paper. All three papers conclude that the results in the Airline and Banking Papers do not hold up under scrutiny. The two papers responding to the Airline Paper conclude that there is no evidence that common ownership has raised airline ticket prices and that the results in the Airline Paper are driven by methodological shortcomings.40 To reach this conclusion, the authors of both papers first replicate the Airline Paper’s data set. To verify that their data set is comparable, they apply the same analytical analyses described in the Airline Paper and obtain results nearly identical to those in the Airline Paper. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3063465 (addressing endogeneity concerns and finding no relationship between prices and MHHI delta). 39 See O’Brien and Waehrer, supra note 37 at 752. 40 See Dennis, Gerardi, and Schenone, supra note 38 (“In sharp contrast to the findings [of the Airline Paper], we find no evidence” of a relationship “between ticket prices and common ownership in the airline industry”); Pauline Kennedy, Daniel P. O’Brien, Minjae Song, and Keith Waehrer, The Competitive Effects of Common Ownership: Economic Foundations and Empirical Evidence (July 2017), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3008331 (“Contrary to recent empirical research…we find no evidence that common ownership raises airline prices.”). ICI provided financial support for the Kennedy paper, but neither ICI nor any of its members provided any input into the paper’s analysis or conclusions. Mr. Donald S. Clark August 20, 2018 Page 16 of 26 Verifying the data set of the Airline Paper provides an important robustness check on the results in the new research because it suggests that any discrepancies between the findings of the Airline Paper and the new research result from differences in the models used to measure common ownership’s effect on competition, rather than data discrepancies. After verifying the Airline Paper’s data set, the newer papers apply their own alternative models to measure whether common ownership affects airline ticket prices. Both find it does not. One paper directly estimates the effects of common ownership on airline prices using a structural oligopoly model that allows it to determine whether common ownership is just correlated with higher prices or has in fact caused higher prices.41 The authors find no evidence that common ownership has caused airline ticket prices to rise. A separate paper presents several additional analyses and results that further raise questions about the validity of the Airline Paper’s conclusions.42 In particular, the paper shows that the Airline Paper’s results are highly sensitive to assumptions the authors made in the construction of their data set, and employs an instrumental variables approach to address the endogeneity concern when calculating MHHI. Ultimately, the paper finds no anticompetitive effect from common ownership in the airline industry.43 A third paper by two Federal Reserve economists proposes an alternative approach to analyze whether common ownership has competitive effects in the banking industry. This paper analyzes the weights that firms place on each other’s profits rather than relating measures of industry concentration (e.g., GHHI) to price and finds much more muted effects on competition than the Banking Paper estimated.44 The academic literature critiquing the preliminary research on common ownership makes clear that the Airline and Banking Papers do not prove that common ownership decreases competition. The authors of the Airline Paper have responded to some of these critiques,45 and revised the Airline Paper extensively, potentially to address some of the shortcomings described above. Although the revised 41 See Kennedy, et al, supra note 40. The paper also shows that the theoretical model that underlies the MHHI and MHHI delta leads to an ambiguous relationship between MHHI delta and the price. 42 Dennis, Gerardi, and Schenone, supra note 38. 43 See id. at 24-25. 44 Gramlich and Grundl, supra note 37. 45 See Schmalz, supra note 28; Jose Azar, Martin C. Schmalz, and Isabel Tecu, Reply to: ‘Common Ownership Does Not Have Anti-Competitive Effects in the Airline Industry’ (May 2018), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3168095; Jose Azar, Martin C. Schmalz, and Isabel Tecu, The Competitive Effects of Common Ownership: Economic Foundations and Empirical Evidence: Reply (September 2017), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3044908. Mr. Donald S. Clark August 20, 2018 Page 17 of 26 Airline Paper continues to rely on MHHI delta to demonstrate a correlation between common ownership and higher prices, it does not demonstrate that common ownership has caused higher prices. Instead, the revised Airline Paper includes several new sections designed to persuade readers that an increase in common ownership is the most likely cause of higher prices. 46 However, the academic literature critiquing the preliminary research explains that causal inference is not possible from the analysis in the revised Airline Paper. 3. New Papers Question Assumptions Underlying the Common Ownership Hypothesis The literature putting forth the common ownership hypothesis treats an adviser and its clients as a monolithic entity with a single interest. 47 More recent literature explains why this assumption is incorrect and casts doubt on whether advisers and their clients have the incentives and a mechanism to control the competitive strategy of their portfolio companies. a. Advisers’ Clients Have Different Goals and Could Be Harmed by Soft Competition Among Portfolio Companies As explained in Section I above, an adviser’s clients typically have diverse investment objectives, which the adviser seeks to achieve by employing varying investment strategies. An adviser owes separate contractual and fiduciary duties to each client with which it has a relationship. Each client wants its holdings to appreciate in value; each client generally is indifferent to the performance of the adviser’s other clients or the adviser’s profitability. Academic papers critiquing the common ownership hypothesis emphasize the divergent nature of client interests. The Airline and Banking Papers gloss over this key issue by their use of data from SEC Form 13F filings. That form requires filers to aggregate for reporting purposes certain of their clients’ holdings, even though the economic owners of these holdings—which may be defined benefit pension funds, mutual funds, separately managed accounts, hedge funds, sovereign wealth funds, or others— may have quite different structures and investment objectives.48 Thus, the aggregated reporting structure of Form 13F can mistakenly lead one to assume—as the Airline and Banking Papers do—that 46 See Airline Paper, supra note 30. 47 Airline Paper, supra note 30 at 17 (“With respect to the definition of ‘shareholder,’ we aggregate holdings at the fund family level to match the institutional feature of voting and governance at the family level, as well as fund families' incentives, which—consistent with the incentives of their investors—are determined primarily by the value of their total assets under management.”). 48 See Thomas A. Lambert and Michael E. Sykuta, The Case for Doing Nothing About Institutional Investors’ Common Ownership of Small Stakes in Competing Firms, University of Missouri School of Law Research Paper No. 2018-21 (May 2018), available at https://ssrn.com/abstract=3173787. Mr. Donald S. Clark August 20, 2018 Page 18 of 26 every client of an adviser (and the adviser itself) will want managers of airlines to maximize industry profits, regardless of whether the client holds airline stocks X and Y, or just airline stock X, or just airline stock Y. This is clearly not the case. For example, as one of the critique papers notes, the core of the Airline Paper is the assumption that corporate managers will take into account the holdings of shareholders in competing airlines.49 This, the critiquing paper argues, is a “heroic (and to us unconvincing) assumption for a variety of reasons … [including that] there is substantial heterogeneity among the holdings of the largest shareholders.”50 Another paper provides an example of divergent shareholder interests simply by disaggregating the airline holding data reported by one large adviser on Form 13F to the level of 11 mutual funds managed by the adviser. The disaggregated data show that:  Two funds held only Southwest Airlines, and two held only United Airlines—so those four funds would get their best performance if the managers of Southwest or United endeavor to maximize the profits of their own firms.  One fund held Delta Airlines and United, and three funds held American, Delta, and United— so those funds would benefit from airline managers who took business away from Southwest.  Two funds held all four airlines, but had disproportionately larger stakes in Southwest—so they would get their best performance if Southwest outperformed other airlines.  One fund held all four airlines in roughly equal shares—and hence might benefit from a strategy of emphasizing industry profits. The paper points out that returns to investors in these mutual funds will depend on the performance of the assets held by each fund, and the competitive strategy that would maximize fund performance varies according to fund holdings.51 The heterogeneity in clients’ preferences affects how advisers fulfill their responsibilities to clients, including how they engage with portfolio companies. Because clients have different interests, advisers 49 See Rock and Rubinfeld, supra note 3. 50 Id. at 10. 51 See Lambert and Sykuta, supra note 48 at 25. Another paper notes that aggregating holdings at the adviser level could be responsible for the empirical evidence linking common ownership to decreased competition. See also Douglas H. Ginsburg and Keith Klovers, Common Sense About Common Ownership, George Mason Law & Economics Research Paper No. 18-09 (2018), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3169847. According to this paper, “[a]ttributing stocks owned in common to their individual investors (e.g., retirement account holders) would instead show atomistic share holdings and no significant difference between MHHI and HHI.” Mr. Donald S. Clark August 20, 2018 Page 19 of 26 need not, and do not, feel bound to pursue a uniform strategy of engagement on behalf of all of their clients. If, for example, in the adviser’s view, its clients have clearly diverging interests on a particular matter, the adviser must vote in the manner that is in each client’s interest, even if this means voting some client shares one way and other client shares another. Studies of adviser voting behavior confirm that advisers do indeed vote client shares differently from time to time.52 b. Other Reasons Advisers Are Unlikely to Prefer Soft Competition in Concentrated Industries The common ownership hypothesis implies that all (or most) advisers would prefer soft competition in concentrated industries, such as the airline industry.53 As the previous section demonstrated, this supposition is difficult to square with the fact that advisers have diverse clienteles. But there are other compelling reasons that advisers or their clients would not benefit from soft competition. The proponents call out index funds’ common ownership of airlines as a likely cause of anticompetitive effects.54 Index funds seek to track the performance of an index, generally by holding the stocks in that index in proportion to their weight in the index. Airline stocks make up a small percentage (significantly less than 1 percent) of two major indexes.55 Any hypothetical reduction in competition in the airline industry arguably would hurt index funds (and thus would be contrary to the interests of the fund and its shareholders) for two reasons. First, index funds typically are broadly diversified, so anticompetitive conduct that benefits one industry likely would adversely affect on companies in other industries. If airline ticket prices rise, this should benefit airline stocks, but the other companies in the index almost certainly use airlines for business travel. Rising ticket prices would increase costs for the other (non-airline) companies, which 52 See Ginsburg and Klovers, supra note 51 at 10–11 (”[O]f the shares an investment manager can vote, it does not necessarily vote all of them the same way. Consistent with its fiduciary duty, each fund votes the shares it manages in the best interests of their underlying economic owners. Because these interests may vary among funds, managers of different types of funds may vote their shares differently.”). See also Alon Brav, Wei Jiang, and Tao Li, Picking Friends Before Picking (Proxy) Fights: How Mutual Fund Voting Shapes Proxy Contests, (June 2018), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3101473. According to this paper, there is significant heterogeneity of voting even within fund families. See id. at Table 1 (showing the percentage of funds within the top five and top ten fund families that voted with management in the May 13, 2015 DuPont proxy fight). 53 See e.g., Eric A. Posner and E. Glen Weyl, A Radical Proposal for Improving Capitalism, Barron’s (June 15, 2018), available at https://www.barrons.com/articles/a-radical-proposal-for-improving-capitalism-1529082880 (“Because institutional investors [in context, investment advisers] own the largest stakes of all big firms in virtually every market, they have an incentive to reduce competition among firms…”). 54 See generally id. 55 According to Bloomberg, as of June 29, 2018, airlines comprised 0.43% of the S&P 500 index and 0.63% of the Russell 3000 index. Mr. Donald S. Clark August 20, 2018 Page 20 of 26 may put downward pressure on their stock prices. In addition, if consumers need to spend more for air travel, they likely will curtail spending in other areas, and this reduction in spending would be reflected in the results of many other companies held in the index fund’s portfolio. Thus, whatever gains an index fund might realize from higher airline ticket prices could be more than offset by the headwinds that higher airline prices create for the other stocks in the indexes, which comprise more than 99 percent of the indexes’ market capitalization. Second, other companies in those indexes provide services to the airlines. For example, airlines may buy fuel, food, aircraft and aircraft maintenance services, information technology services, and other goods and services from other companies in the indexes. If airline ticket prices rise, the quantity of air miles traveled could fall, and these other index companies could perform worse. In sum, the empirical work behind the common ownership hypothesis does not account for how reduced competition in one industry might adversely affect the value of client holdings in other industries.56 The Airline and Banking Papers assume that shareholders with diversified portfolios seek to maximize portfolio (rather than individual firm) profits, and that corporate managers may aggregate and internalize these broader shareholder interests when formulating their competitive strategies.57 But, as the example above makes clear, the proponents give no indication of having followed through on the difficult and expansive work that their theoretical assumptions would require. c. Recent Research Also Questions Whether a Viable Mechanism Exists for Common Ownership to Decrease Competition Some papers supporting the common ownership hypothesis postulate that shareholder voting or executive compensation practices might incent corporate managers to reduce competition in concentrated industries, but newer papers cast doubt on this speculation. Consistent with the discussion of voting practices above, one paper finds “no evidence that shareholders vote on competitive strategy” or even that director candidates run on issues related to competitive strategy.58 This paper also explains that shareholder voting plays a limited role in setting executive compensation.59 56 See Lambert and Sykuta, supra note 48 at 20. This paper also argues that proponents cannot assume an adviser would benefit from the maximization of airline industry profits. See id. at 27 (providing an example illustrating why such an assumption is faulty). 57 See Airline Paper, supra note 30 at 7-8. 58 See Rock and Rubinfeld, supra note 3 at 17. 59 This has support as a legal matter. Section 951 of the Dodd-Frank Wall Street Reform and Consumer Protection Act added Section 14A to the Exchange Act, requiring public companies subject to the federal proxy rules to provide their shareholders with an advisory (i.e., non-binding) vote on executive compensation, generally known as "say-on-pay" votes. Mr. Donald S. Clark August 20, 2018 Page 21 of 26 The paper further argues that if the conclusions in the Airline Paper are accepted as true, and if common owners use executive compensation as a mechanism to produce anticompetitive effects, then one would expect airline executives to be compensated according to industry performance. But, this prediction is wrong. In fact, one paper finds that executive compensation structures in the airline industry reward those whose companies outperform their peers.60 This paper also notes that some institutional investors, including investment advisers, encourage companies to adopt compensation policies that reward executives for outperforming industry peers.61 The findings of this paper call into question the hypothesis that institutional investors use executive compensation to decrease competition by rewarding executives based on industry performance. Yet another paper found that as common ownership increases, corporate executives are rewarded more for outperforming peer firms.62 In addition, an FTC Commissioner who recently addressed this issue questioned whether the scholarship linking common ownership to decreased competition relies properly on an intuition that “corporate managers, cognizant that their large institutional shareholders also hold stock in competitors, soften competition to benefit those shareholders.”63 Proponents of the common ownership hypothesis also suggests that common ownership could decrease competition if shareholders lobby firms to compete less aggressively. As described in Section I.D above, the federal antitrust and securities laws establish parameters for these engagement efforts. Moreover, if an adviser plays a role in facilitating anticompetitive activities among portfolio companies—and we are aware of no evidence that this has occurred or is occurring—it and those companies could face liability under Section 1 of the Sherman Act, which prohibits cartels. In this respect, the Airline Paper’s attempt at setting forth persuasive mechanisms is an uncertain grab- bag of contradictory theories (common owners are both “lazy principals” and engaging directly in ways that encourage the softening of competition), citations to news articles, a citation to an executive compensation article that has been challenged by other papers, and anecdotes. In sum, it is provocative but wholly unsupported. 60 See Rock and Rubinfeld, supra note 3 at 24-25 (analyzing 2015 proxy statements from American Airlines and Delta Airlines and concluding that both companies compensate their executives based on relative performance). 61 See id. at 26-27. 62 See Heung Jin Kwon, Executive Compensation Under Common Ownership (November 2016), available at http://fmaconferences.org/Boston/ExecutiveCompensationunderCommonOwnership.pdf. This result contradicts directly the findings of Anton et al., supra note 35. Others have noted that the diametrically opposite results raise doubts about the robustness of the conclusions in both papers. See Rock and Rubinfeld, supra note 3 at 23. 63 Taking Stock: Assessing Common Ownership, Noah Joshua Phillips, Commissioner, Federal Trade Commission, (June 1, 2018), available at https://www.ftc.gov/system/files/documents/public_statements/1382461/phillips_-_taking_stock_6- 1-18_0.pdf. Mr. Donald S. Clark August 20, 2018 Page 22 of 26 III. Policy Measures Based on the Common Ownership Hypothesis Would Harm Investors, Companies, and the Economy, and Are Inappropriate Although the Airline and Banking Papers do not propose solutions for the supposed harms arising from common ownership, other papers do. One paper urges the Commission, the US Department of Justice, state attorneys general, and private plaintiffs to bring cases under Section 7 of the Clayton Act to require institutional investors to divest any common holdings that cause anticompetitive effects.64 To minimize antitrust liability, an institutional investor could refrain from common ownership in concentrated industries, or avoid having any voting influence (e.g., by committing not to vote its stock, or voting in proportion to how non-common shareholders vote, or buying only nonvoting stock across horizontal competitors).65 A second paper agrees that enforcing Section 7 of the Clayton Act would reduce potential anticompetitive effects of common ownership but argues that litigation would be a suboptimal way to address these harms.66 This paper explains that courts might use different standards to determine liability, which would result in “confusion” due to the range of rules, standards, and approaches that might proliferate in a litigation-driven approach to reducing common ownership. To address these shortcomings, the paper urges the Commission and the US Department of Justice to adopt the following enforcement policy: No institutional investor or individual holding shares of more than a single effective firm in an oligopoly may ultimately own more than 1% of the market share unless the entity holding shares is a free-standing index fund that commits to being purely passive.67 To implement this policy, the Commission and the US Department of Justice would need to publish annually, subject to a public comment process, a list of industries that are oligopolies (together with the related company market shares), and institutional investors would receive “at least a month” to “rearrange their holdings to comply with the policy.”68 Compliance would consist of: (1) holding stock (of an unlimited amount) in no more than one firm in an oligopoly; (2) holding stock in more than one firm, but limiting common ownership to no more than a “single effective firm” (i.e., a holding no more 64 See Elhauge, supra note 34. This paper also argues that injured persons could sue under Section 1 of the Sherman Act on the theory that holding shares in competing firms is an agreement that restrains competition. 65 See id. at 1314-1315. 66 Eric A. Posner, Fiona Scott Morton, and E. Glen Weyl, A Proposal to Limit the Anti-Competitive Power of Institutional Investors (March 22, 2017) at 20, available at https://ssrn.com/abstract=2872754. See also Jenny Luna, The Biggest Antitrust Story You’ve Never Heard, Stanford Graduate School of Business (August 6, 2018), available at https://www.gsb.stanford.edu/insights/biggest-antitrust-story-youve-never-heard. 67 Posner, Morton, and Weyl, supra note 66 at 33. 68 Id. Mr. Donald S. Clark August 20, 2018 Page 23 of 26 than HHI/10000 for the industry); or (3) continuing to hold shares in competing firms, but (i) committing not to communicate with top managers or directors of the firms, (ii) relinquishing its voting rights by voting shares in proportion to the way other investors vote, and (iii) trading stocks only in accordance with clear and non-discretionary public rules, such as matching an index as closely as possible.69 Without taking a firm stance, the paper weighs the benefits of putting this policy in place by legislation, rulemaking, or enforcement guidelines, and whether the enforcement policy should be absolute or merely create a rebuttable presumption. Other papers argue that it is inappropriate to consider policy measures that would decrease common ownership, especially given the questions about the theoretical and empirical validity of the research linking common ownership to decreased competition.70 Some papers warn that adopting measures that would restrict institutional investors from holding stock in competing firms could have harmful consequences, such as increasing the cost of investing, particularly for retirement savers, reducing diversification and the benefits that it brings to investors, and decreasing the quality of corporate governance.71 Another paper recommends evaluating the legality of common ownership on a case-by-case basis,72 while a separate paper raises serious questions about whether the antitrust law would support a claim against common owners.73 Specifically, this paper explains that the papers calling for enforcement of Clayton Act Section 7 or Sherman Act Section 1 take cases of cross-ownership—i.e., where an operating company owns stock in a competitor—and argue for the extension of these cases to common ownership, even though “court decisions premised on cross ownership do not necessarily apply, let alone apply with equal force, to common ownership.”74 69 See id. 70 See e.g., O’Brien and Waehrer, supra at note 37 at 768 (“researchers and policy makers are getting ahead of themselves in proposing and implementing policy changes based on” the preliminary research); Lambert and Sykuta, supra note 48 at 55 (“the optimal regulatory approach is to do nothing about institutional investors’ common ownership of small stakes in competing firms”). 71 See Rock and Rubinfeld, supra note 3 at 36-39; Elaine Buckberg, Steven Herscovici, Branco Jovanovic, and James Reitzes, Proposal to Remedy Horizontal Shareholding is Flawed, Law360 (July 17, 2017), available at http://files.brattle.com/files/ 7316_proposal_to_remedy_horizontal_shareholding_is_flawed.pdf. Other papers point out the high costs of planning and administering the policy measures advanced by proponents of the common ownership hypothesis. See Lambert and Sykuta, supra note 48 at 37-40. 72 See Menesh Patel, Common Ownership, Institutional Investors, and Antitrust (May 2018), Antitrust Law Journal, forthcoming, available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2941031. 73 See Ginsburg and Klovers, supra note 51. 74 See id. at 22. Mr. Donald S. Clark August 20, 2018 Page 24 of 26 We agree that regulatory efforts to restrict or limit common ownership or shareholders’ rights are unwarranted, but are deeply concerned by the possibility that policymakers could hastily adopt measures to counter the harms alleged by the unproven and disputed common ownership hypothesis. The same authors who suggested stripping voting rights from institutional investors’ clients or restricting their clients’ ability to invest in competing firms have doubled down on their proposals, despite the growing body of literature challenging the need for such measures.75 The benefits that the asset management industry—and regulated funds in particular—provide to their investors and the economy are certain and tangible. Regulated funds provide their shareholders with professional portfolio management and diversified exposure to stocks, all at low cost. Often, retail investors do not have the time, inclination, amount of assets, or the expertise to invest directly in stocks, yet they understand that exposure to stocks (and the potential returns that they offer) is necessary to achieve their financial goals, such as accumulating sufficient retirement assets. Regulated funds meet this need, as borne out by their popularity, and have effectively democratized investing. They also supply capital to companies to grow their business, create jobs, and innovate. A sound regulatory structure and competitive market dynamics have benefitted regulated funds, their shareholders, and the companies in which they invest; imprudent policy could undermine these achievements. The policy measures that have been proposed to address the harms that allegedly accompany common ownership would be enormously disruptive to the asset management industry and harm retail investors, businesses, and the economy, all in the name of addressing a theoretical and speculative consumer harm. Policy measures that limit the ability of institutional investors to hold stock in competing firms would make investing more complicated, fragmented, and expensive for the nearly 50 million US households that use equity funds to meet their financial goals.76 These measures also could distort capital markets by making it more difficult and costly for some firms to attract investment. Similarly, measures that restrict voting would limit the ability of institutional investors to engage with companies, which could result in less effective corporate stewardship and a decline in the quality of corporate governance. Such measures also could conflict with advisers’ fiduciary duty to monitor portfolio companies on behalf of clients and take a principled approach to engaging with these firms. Public statements from the US government and its officials echo these concerns. In a written statement to the Organisation for Economic Co-operation and Development (OECD) in 2017, the US government stated that limiting common ownership “without sufficient evidence of anticompetitive effects could impose unintended real-world costs on businesses and consumers by making it more 75 See Einer Elhauge, New Evidence, Proofs, and Legal Theories on Horizontal Shareholding (January 4, 2018), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3096812; Posner and Weyl, supra note 53. 76 See 2018 Investment Company Factbook at 143, available at https://www.ici.org/pdf/2018_factbook.pdf. Mr. Donald S. Clark August 20, 2018 Page 25 of 26 difficult to diversify risk.”77 More recently Commissioner Phillips endorsed the position taken at the OECD and stated that “[f]or now, I do not believe we know enough to warrant policy changes [to address common ownership]. [US] antitrust enforcers have tools already at our disposal for monitoring and disciplining anticompetitive activity, and will use those tools to intervene where the law and the evidence provide a basis for doing so.” Commissioner Phillips encouraged further academic study of common ownership and noted that when proposals for changes in enforcement policy “put at risk both shareholder value and consumer benefits,” antitrust enforcers “must tread carefully.”78 We share these views. The Commission and other antitrust authorities should not consider measures designed to limit common ownership or restrict institutional investors’ ability to vote clients’ shares in competing firms. Forging ahead without a sound basis would impose costs and other consequences on American investors and businesses without providing any certain benefit to consumers. Even if policymakers were convinced that common ownership softens competition, they should not necessarily act. They would first need to establish that measures to reduce common ownership produce benefits (i.e., materially lower airline tickets, costs of banking services, etc.) that would outweigh the costs. The “solutions” proffered to date are highly likely to cause very significant collateral damage to the wide range of clients that rely on the asset management industry to meet their financial goals and to the asset management industry as a whole. Similarly, any attempt to restrict the ability of institutional investors to hold small stakes of competing firms likely will fail or produce significant unintended consequences if it is not directly tailored to the mechanism that enables common ownership to affect competition. In light of the considerations discussed above, we urge the Commission to focus its resources on other areas. * * * 77 United States, Note to the OECD on Hearing on Common Ownership by Institutional Investors and its Impact on Competition (November 28, 2017), available at https://one.oecd.org/document/DAF/COMP/WD(2017)86/en/pdf. 78 Taking Stock: Assessing Common Ownership, Noah Joshua Phillips, Commissioner, Federal Trade Commission, (June 1, 2018), available at https://www.ftc.gov/system/files/documents/public_statements/1382461/phillips_-_taking_stock_6- 1-18_0.pdf. The German Monopolies Commission also recently determined that it would be “premature” to take measures to restrict common ownership of German firms. German Monopolies Commission, Summary: Competition 2018 at 5 (July 2018), available at https://www.monopolkommission.de/images/HG22/HGXXII_Summary.pdf (the full report, which is available only in German, can be found at https://www.monopolkommission.de/en/reports/biennial-reports/biennial- report-xxii.html). Mr. Donald S. Clark August 20, 2018 Page 26 of 26 We appreciate the opportunity to comment on the common ownership hypothesis in advance of the Commission’s hearings on competition and consumer protection in the 21 st century. If you have any questions on our comment letter, please feel free to contact Sean Collins, Chief Economist, at sean.collins@ici.org or (202) 326-5882; Susan Olson, General Counsel, at (202) 326-5813 or susan.olson@ici.org; George Gilbert, Assistant General Counsel, at george.gilbert@ici.org or (202) 326-5810; or Matthew Thornton, Assistant General Counsel, at matt.thornton@ici.org or (202) 371- 5406. Sincerely, /s/ Sean S. Collins /s/ Susan M. Olson Sean S. Collins Susan M. Olson Chief Economist General Counsel cc: The Honorable Joseph J. Simons The Honorable Maureen K. Ohlhausen The Honorable Noah Joshua Phillips The Honorable Rohit Chopra The Honorable Rebecca Kelly Slaughter Mr. Bruce Hoffman, Director, Bureau of Competition Mr. Bilal Sayyed, Director, Office of Policy Planning A-1 Appendix A: Organizational Structure and Regulation of Registered Investment Advisers and Regulated Funds Organizational Structure and Regulation of Investment Advisers Investment advisers provide advice regarding investment in securities. Investment advisers are typically organized under state law as business entities (e.g., corporations, limited partnerships, or limited liability companies) and are subject to applicable state and federal law (most notably the Advisers Act) and related rules. Advisers registered with the SEC under the Advisers Act1 are subject to the SEC’s regulation and oversight. As of July 2018, there were approximately 13,000 SEC registered investment advisers in the United States, with over $83 trillion in assets under management.2 The Advisers Act and the rules adopted thereunder are primary in regulating the activities of investment advisers; they impose various obligations and prohibitions. For purposes of the common ownership debate, some of the most relevant provisions relate to compliance programs; proxy voting; the misuse of material non-public information (MNPI); and codes of ethics. More specifically:  Compliance. An adviser must have a chief compliance officer and adopt and implement compliance policies and procedures reasonably designed to prevent violations of the Advisers Act. The chief compliance officer must conduct an annual review of the compliance policies and procedures to assess their adequacy and effectiveness, considering among other things any compliance matters arising over the year or any changes in the business.3  Proxy Voting. An adviser with proxy voting authority must adopt and implement written proxy voting policies and procedures reasonably designed to ensure that the adviser votes client securities in the best interests the client, and these policies and procedures must address how material conflicts of interest between the adviser and its client are handled.4  MNPI. Section 204A of the Advisers Act requires an adviser to establish, maintain, and enforce written policies and procedures reasonably designed to prevent the misuse of MNPI by 1 The Advisers Act defines “investment adviser” in relevant part as “any person who, for compensation, engages in the business of advising others, either directly or through publications or writings, as to the value of securities or as to the advisability of investing in, purchasing, or selling securities, or who, for compensation and as part of a regular business, issues or promulgates analyses or reports concerning securities… .” An investment adviser with more than $100 million in assets under management may, and with more than $110 million must, register with the SEC, unless an exemption is available. 2 See Information About Registered Investment Advisers and Exempt Reporting Advisers, available at https://www.sec.gov/help/foiadocsinvafoiahtm.html. 3 Advisers Act Rule 206(4)-7. 4 Advisers Act Rule 206(4)-6. An adviser also must (i) disclose how clients may obtain information about how the adviser voted with respect to their securities, and (ii) describe its proxy voting policies and procedures to clients and, upon request, provide them with a copy of those policies and procedures. A-2 the adviser or any associated person. The federal securities laws broadly prohibit market participants (including advisers) from trading on MNPI. MNPI may include information about portfolio companies (e.g., mergers or acquisitions, earnings information, management changes) that advisers may (sometimes inadvertently) possess, prior to broad public dissemination. For advisers, MNPI also may include client information (e.g., trades that the adviser intends to make on behalf of clients). Common practices that advisers employ to fulfill this statutory obligation include the use of information barriers, to “wall off” those persons who may possess MNPI; use of restricted lists that prohibit the purchase or sale of securities of those companies for which the adviser possesses MNPI; ongoing compliance monitoring; and continuing employee education.  Code of Ethics. Advisers must establish, maintain, and enforce a written code of ethics.5 Advisers are subject to specific registration, filing, and client disclosure requirements.6 Advisers also are subject to supervisory oversight by the SEC through inspections, and the SEC’s Office of Compliance Inspections and Examinations (OCIE) annually publishes a list of its exam priorities.7 The Advisers Act and rules thereunder also impose important requirements around advisory fees, disclosure to clients, recordkeeping, custody and trading, and other antifraud provisions. The Advisers Act requires that certain matters be addressed in the advisory or investment management contract. Advisers typically include provisions related to the adviser’s authority over the account, investment guidelines, provisions related to brokerage and trading, proxy voting, fees and expenses, termination, and custody among many others. The advisory agreement with a regulated fund is subject to additional requirements under the Investment Company Act, including approval by fund shareholders and the fund board. 5 Advisers Act Rule 204A-1. At a minimum, this code of ethics must include: (i) standards of business conduct required of supervised persons, reflecting fiduciary obligations; (ii) provisions requiring supervised persons to comply with applicable federal securities laws; (iii) provisions requiring access persons to report, and the adviser to review, their personal securities transactions and holdings periodically (as set forth in the rule); (iv) provisions requiring supervised persons to report any violations of the code of ethics promptly to the chief compliance officer or designee; and (v) provisions requiring the adviser to provide supervised persons with a copy of the code of ethics and any amendments, and requiring supervised persons to provide written acknowledgment of their receipt of these items. 6 To register with the SEC as an investment adviser, the adviser must file Form ADV and update it annually. Form ADV consists of two parts: (i) Part 1 requires information about the adviser’s business, ownership, clients, employees, business practices, affiliations, and any disciplinary events of the adviser or its employees, and (ii) Part 2 requires the adviser to prepare a narrative brochure written in plain English that contains information such as the types of advisory services offered, the adviser’s fee schedule, disciplinary information, conflicts of interest, and the educational and business background of management and key advisory personnel of the adviser. An adviser must deliver a brochure and one or more brochure supplements to each client or prospective client that contains all information required by Part 2 of Form ADV, in accordance with Rule 204-3. 7 The SEC’s 2018 inspection and examination priorities are available at https://www.sec.gov/about/offices/ocie/national- examination-program-priorities-2018.pdf. A-3 Organizational Structure and Regulation of Regulated Funds Regulated funds are registered under the Investment Company Act and include mutual funds, ETFs, and closed-end funds. The investment adviser to a regulated fund must be registered under the Advisers Act. Regulated funds are one type of client served by investment advisers. Organization Each regulated fund is a separate legal entity, organized under state law usually as a corporation or a business trust (in some states, a “statutory trust”). Funds have officers and directors (if the fund is a corporation) or trustees (if the fund is a business trust). The fund’s board plays an important role in overseeing fund operations and has mandated responsibilities under the Investment Company Act. Unlike other traditional companies, a regulated fund is externally managed and has no employees in the traditional sense. Instead, a fund relies upon third parties or service providers—either affiliated organizations or independent contractors—to invest fund assets and carry out other business activities. The following diagram shows the primary types of service providers usually retained by a mutual fund. These service providers include the investment adviser, the principal underwriter, the administrator, the transfer agent, the custodian, and the independent public accountant. A-4 Comprehensive Regulation Regulated funds are subject to a comprehensive regulatory scheme under the federal securities laws. These funds are regulated under all four of the major federal securities laws: the Securities Act of 1933, which requires registration of the fund’s shares and the delivery of a prospectus; the Exchange Act, which regulates the trading, purchase, and sale of fund shares and establishes antifraud standards governing such trading; the Advisers Act, which regulates the conduct of a regulated fund’s investment adviser and requires the adviser to register with the SEC; and, most importantly, the Investment Company Act, which requires all regulated funds to register with the SEC and meet certain operating conditions and standards. As of June 2018, there were approximately 6,300 regulated equity funds in the US, with almost $13.4 trillion in assets under management. The Investment Company Act and other laws impose substantive requirements on the operations of regulated funds as well as extensive disclosure requirements. For example, regulated funds are subject to strict custody requirements and must maintain their assets separate from the assets of the investment adviser.8 Regulated funds are subject to restrictions on borrowings and leverage.9 The Investment Company Act has provisions that prohibit transactions between the regulated fund and insiders or affiliated persons, such as the investment adviser.10 Regulated funds must have written compliance programs overseen by a chief compliance officer.11 Fund boards must include independent directors and the fund’s board has specific duties and responsibilities under the Investment Company Act.12 Regulated funds also are subject to inspection and supervisory oversight by the SEC. In their capacity as shareholders in portfolio companies, regulated funds must provide disclosure about their proxy voting policies and procedures. Further, regulated funds are the only shareholders required to publicly disclose how they voted each proxy.13 Regulated funds have robust filing and disclosure responsibilities under the Investment Company Act and other securities laws, which ensure that markets and investors have access to extensive information 8 Section 17(f) of the Investment Company Act and rules thereunder. 9 Section 18 of the Investment Company Act. 10 Section 17 of the Investment Company Act and rules thereunder. 11 Investment Company Act Rule 38a-1. 12 See e.g., Section 15 of the Investment Company Act (approval of investment advisory and underwriting contracts), Section 32 (approval of independent accountants), and Rule 17a-7 (approval of securities transactions between a fund and certain affiliates). 13 More specifically, a fund must (i) describe in its registration statement the policies and procedures that it uses to determine how to vote proxies relating to its portfolio securities, and (ii) publicly file with the SEC the fund’s records of how it voted proxies relating to its portfolio securities. See Item 17(f) of Form N-1A (registration statement for open-end funds), Item 18 of Form N-2 (registration statement for closed-end funds), Investment Company Act Rule 30b1-4, and Form N-PX. A-5 about a regulated fund. For example, regulated funds are subject to requirements that mandate disclosure of: (i) the types of securities in which they may invest and their associated risks; (ii) reports describing the results of fund activities and operations; (iii) portfolio holdings disclosure; and (iv) financial disclosure.14 In addition to any investment parameters and limitations that a regulated fund may establish in its prospectus and advisory agreement, various laws affect, and in some cases, restrict how regulated funds can invest their assets. For example, the Investment Company Act limits regulated funds’ ability to invest in securities issued by broker-dealers, underwriters, and investment advisers.15 Other federal and state laws limit investments in certain businesses, such as banks and bank holding companies, insurance holding companies, and licensees of the Federal Communications Commission.16 Both tax law and the Investment Company Act provide diversification standards for funds. Under the tax laws, all regulated funds seeking to qualify as “regulated investment companies” must meet a diversification test every quarter.17 The Investment Company Act sets higher standards for regulated funds that elect to be diversified. For these funds, the Investment Company Act requires that, with respect to at least 75 percent of the portfolio, no more than five percent may be invested in the securities of any one issuer, and no investment may represent more than ten percent of the outstanding voting securities of any issuer.18 14 See, e.g., Form N-1A (describing requirements for a fund’s prospectus, statement of additional information, other information, and exhibits to a fund’s registration statement, including the fund’s advisory agreement); and Section 30(e) of the Investment Company Act (shareholder reports). A fund discloses its portfolio holdings information in its annual and semi-annual reports to shareholders. A fund also publicly discloses its portfolio holdings as of the end of its first and third fiscal quarters through Form N-Q (soon to be replaced by Form N-PORT) filings. Open-end fund registration statements are amended at least once each year to ensure that financial statements and other information have not become stale. These funds also amend their registration statements throughout the year as necessary to reflect material changes to their disclosure. 15 See Section 12(d)(3) of the Investment Company Act and the rules thereunder. 16 See, e.g., Geffen & Early, “Mutual Fund Investment Limitations Arising Outside of the Investment Company Act,” The Investment Lawyer (June 2009), available at https://www.dechert.com/content/dam/dechert%20files/publication/ 2009/6/mutual-fund-investment-limitations-arising-outside-of-the-invest/IL_0609_Geffen.pdf. 17 Under the “diversification” test of Internal Revenue Code section 851(b)(3), at least 50 percent of the value of the fund’s total net assets must consist of cash, cash items, government securities, securities of other funds, and investments in other securities which, with respect to any one issuer, represent neither more than 5 percent of the assets of the fund nor more than 10 percent of the voting securities of the issuer. Further, no more than 25 percent of the fund’s assets may be invested in the securities of any one issuer (other than government securities or the securities of other funds), the securities (other than the securities of other funds) of two or more issuers which the fund controls and are engaged in similar trades or businesses, or the securities of one or more qualified publicly traded partnerships. 18 Section 5(b)(1) of the Investment Company Act. All regulated funds must disclose whether they are diversified under the Investment Company Act’s standards. Although diversification is not mandatory, regulated funds must obtain shareholder approval to change their status from diversified to non-diversified, providing an additional restriction on regulated funds and level of protection for investors. B-1 Appendix B: Selected Securities Law Reporting Obligations Related to Holdings of Securities The Exchange Act imposes certain reporting obligations on holders of securities, including regulated funds and investment advisers. Section 13 of the Exchange Act requires that certain information be publicly filed with the SEC by persons who beneficially own certain threshold amounts of an issuer’s equity securities. The amount of information and type of information differs based on several factors. Section 16(a) of the Exchange Act imposes filing obligations on persons deemed to be “insiders” of an issuer. Under Section 16, directors, officers and direct or indirect beneficial owners of more than 10 percent of any class of any equity security must file certain forms with the SEC regarding their initial acquisitions and subsequent transactions in the company’s equity securities. For purposes of calculating the beneficial ownership threshold, Rule 16a-1(a)(1) incorporates the definition of beneficial ownership from Section 13(d), and the same aggregation principles apply to the calculations as apply for purposes of Section 13(d) and (g) filings.1 We briefly describe below three of the primary Section 13 reports for significant holdings of equity securities. Form 13F Section 13(f) of the Exchange Act requires institutional investment managers that exercise investment discretion over accounts holding certain equity securities having an aggregate fair market value of $100 million or more to file quarterly reports about those holdings with the SEC on Form 13F.2 Depending on client holdings over which it maintains discretionary authority, an adviser may be subject to these filing requirements. Congress added this provision to increase the public availability of information regarding the securities holdings of institutional investors. These filings have been an important source of data within the common ownership debate, but when the information is used in this context it has substantial limitations. For example, an adviser likely holds the same stock on behalf of multiple clients but must aggregate the client holdings for Form 13F. As noted, the aggregation is driven only by investment discretion, not economic ownership or proxy voting discretion.3 Economic ownership that the adviser aggregates in its filing may belong to clients with different investment objectives, and the adviser may not have proxy voting discretion with respect 1 In calculating beneficial ownership for purposes of Section 16(a), regulated funds and registered investment advisers generally may exclude portfolio securities held by registered investment companies from the calculation of their beneficial ownership of an issuer’s securities; however, such securities should be counted if the adviser is deemed to have the purpose or effect of changing or influencing control of the issuer. Exchange Act Rule 16a-1(a)(1)(iv) and (v). See also Mario J. Gabelli, SEC No- Action Letter (Apr. 30, 1991). The SEC has stated that the “control” analysis pursuant to Section 16 “will be the same as that under Section 13(d) of the Exchange Act.” Hewitt Associates, SEC No-Action Letter (Apr. 30, 1991). 2 Form 13F and its instructions are available at www.sec.gov/about/forms/form13f.pdf. 3 See Sections 3(a)(35) and 13(f)(1) of the Exchange Act, and Rule 13f-1 thereunder. B-2 to all shares that it reports. In addition, with the increasing popularity of “dual-class” equity structures (i.e., a structure in which a company offers more than one class of equity security with differing voting rights), it can be difficult to assess control from reported shareholdings.4 The Form 13F reports also do not necessarily capture the net economic exposure of an adviser’s clients, because the adviser does not report or capture short positions, or short options positions.5 Therefore, the Commission (and other interested academics and policymakers) must be mindful of the limitations of these data when using them to study common ownership and competition. Schedule 13D Section 13(d) of the Exchange Act requires any person who, after acquiring directly or indirectly the beneficial ownership of an equity security registered under the Exchange Act, is directly or indirectly the “beneficial owner” of more than five percent of such class of securities, to file with the SEC certain information on Schedule 13D within 10 days of such acquisition. The information provided pursuant to Schedule 13D must include any intention of the beneficial owner to influence the management or control of the issuer of the portfolio security. Advisers with discretionary authority are generally deemed to be beneficial owners of securities held in discretionary accounts.6 In enacting Section 13(d), Congress intended to ensure that other investors and issuers have sufficient information to make informed decisions about persons or groups that could exercise influence over and make major changes to the operations of an issuer.7 Consistent with this intent, Schedule 13D requires the disclosure of extensive information regarding the terms and conditions of any such acquisition, including disclosure of the investor’s purposes in acquiring the shares. 4 See, e.g., Facebook, Inc.’s 2018 Form 10-K filing for the fiscal year ended December 31, 2017, available at https://www.sec.gov/Archives/edgar/data/1326801/000132680118000009/fb-12312017x10k.htm. (“Our Class B common stock has ten votes per share and our Class A common stock has one vote per share. Stockholders who hold shares of Class B common stock, including certain of our executive officers, employees, and directors and their affiliates, together hold a substantial majority of the voting power of our outstanding capital stock. … [This] dual class common stock structure [ ] provides Mr. Zuckerberg with the ability to control the outcome of matters requiring stockholder approval, even if he owns significantly less than a majority of the shares of our outstanding Class A and Class B common stock.”). 5 See SEC Division of Investment Management: Frequently Asked Questions About Form 13F, Q&A 41 and 43, available at https://www.sec.gov/divisions/investment/13ffaq.htm. 6 See, e.g., Adoption of Beneficial Ownership Disclosure Requirements, Exchange Act Rel. No. 13,291 (Feb. 24, 1977); In re William R. Grant, Exchange Act Release No. 26,339 (Dec. 5, 1988). 7 See, e.g., S. Rep. No. 550 (1967) (noting that, prior to the enactment of Section 13(d), “the law [did] not even require that [a person seeking control by using a tender offer] disclose his identity, the source of his funds, who his associates are, or what he intends to do if he gains control of the corporation”). The Senate report further stated that “[t]he competence and integrity of a company’s management, and of the persons who seek management positions, are of vital importance to stockholders.” Id. See also, GAF Corp. v. Milstein, 453 F.2d 709 (2d Cir. 1971). B-3 Schedule 13G Section 13(g) of the Exchange Act provides that certain investors may be permitted to file Schedule 13G instead of the more onerous Schedule 13D. There are three types of Schedule 13G filers: (i) eligible institutional investors; (ii) eligible passive investors; and (iii) individuals and entities required to file Schedule 13G because they are five percent beneficial owners but who have not triggered the requirement to file on Schedule 13D. Schedule 13G requires significantly less detailed information than Schedule 13D, and as a practical matter, regulated fund managers generally report beneficial ownership on Schedule 13G rather than Schedule 13D. To file Schedule 13G, a manager must have acquired the securities with no purpose or effect of changing or influencing the control of the issuer,8 and not in connection with or as a participant in any transaction having such purpose or effect.9 8 The SEC has stated that most solicitations regarding social or public interest issues (e.g., environmental policies) would not have the purpose or effect of changing or influencing control of the company (and therefore would not prevent the use of Schedule 13G), and that proposals and soliciting activity related to general corporate governance matters such as executive compensation, director pensions, and confidential voting also normally would not prevent the use of Schedule 13G. SEC Release No. 34-39538 (Jan. 12, 1998), available at https://www.sec.gov/rules/final/34-39538.txt. See also Compliance and Disclosure Interpretations, Exchange Act Sections 13(d) and 13(g) and Regulation 13D-G Beneficial Ownership Reporting, Question 103.11, available at https://www.sec.gov/divisions/corpfin/guidance/reg13d-interp.htm (identifying forms of engagement subject matter—e.g., executive compensation and social or public interest issues and corporate governance topics—that, without more, generally would not preclude a shareholder from filing on Schedule 13G). 9 Exchange Act Rule 13d-1(b)(1) and (c). C-1 Appendix C: Selected Papers on Common Ownership Early Papers Advancing the Common Ownership Hypothesis or Offering Policy Measures Based on this Hypothesis  Miguel Antón, Florian Ederer, Mireia Giné, and Martin C. Schmalz, Common Ownership, Competition, and Top Management Incentives, CESifo Working Paper Series No. 6178 (2018). This paper argues that common ownership deters company managers from competing aggressively with rivals. This, they say, is evidenced by executive compensation practices.  José Azar, Martin C. Schmalz, and Isabel Tecu, Anti-Competitive Effects of Common Ownership, Journal of Finance, Forthcoming (2018). This paper asserts that increases in common ownership coincided with airline seat ticket prices rising from anywhere between three and seven percent during the 2001 to 2014 period.  José Azar, Sahil Raina, and Martin C. Schmalz, Ultimate Ownership and Bank Competition, Working Paper (2016). This paper claims to find that greater common ownership, as proxied by inclusion of a stock in an index, led to higher fees and lower interest rates for individual deposit accounts between 2004 and 2013.  Einer Elhauge, Horizontal Shareholding, 129 Harv. L. Rev. 1267 (2016). This paper argues that common ownership can help explain fundamental economic puzzles, including why corporate executives are rewarded for industry performance rather than individual corporate performance alone, why corporations have not used recent high profits to expand output and employment, and why economic inequality has risen in recent decades. The paper also argues that common ownership that creates anticompetitive effects is illegal under current antitrust law, and recommends antitrust enforcement actions to reduce this type of common ownership.  Eric A. Posner, Fiona Scott Morton, and E. Glen Weyl, A Proposal to Limit the Anti- Competitive Power of Institutional Investors, Antitrust Law Journal, Forthcoming (2017). This paper argues for the following restriction on common ownership: No institutional investor or individual holding shares of more than a single effective firm in an oligopoly may ultimately own more than 1% of the market share unless the entity holding shares is a free-standing index fund that commits to being purely passive. Papers Critiquing the Common Ownership Hypothesis and Associated Policy Proposals  Elaine Buckberg, Steven Herscovici, Branko Jovanovic and James Reitzes, Proposal to Remedy Horizontal Shareholding is Flawed, Law360 (July 17, 2017). This paper questions the papers that link common ownership to decreased competition and highlights that the proposed legal remedies are both premature and flawed given the harm that they would do to investors and to the real economy. C-2  Committee on Capital Markets Regulation, Common Ownership and Antitrust Concerns, (November 2017). This paper reviews the literature supporting the hypothesis that common ownership by institutional investors decreases competition in concentrated industries, and finds that the hypothesis is unproven. The paper further argues that no solution is necessary to a problem that is not proven to exist.  Patrick Dennis, Kristopher Geradi and Carola Schenone, Common Ownership Does Not Have Anti-Competitive Effects in the Airline Industry, Working Paper (2018). This paper analyzes the relationship between ticket prices and common ownership in the airline industry and finds no evidence that common ownership has raised airline prices. In the context of certain airline bankruptcies, the paper finds the correlations in the Azar et al. Airline Paper break down when the control assumption transfers to the bankruptcy court. The paper also questions the applicability of the theory of horizontal mergers and cross-ownership theory in the context of common ownership.  Douglas H. Ginsburg and Keith Klovers, Common Sense About Common Ownership, George Mason Law & Economics Research Paper No. 18-09 (2018). This paper describes four shortcomings of research linking common ownership to decreased competition. First, proponents conflate asset management and economic ownership and therefore incorrectly attribute allegedly anticompetitive conduct to asset managers. Second, proponents overstate the validity and strength of the existing empirical work purporting to show common ownership causes anticompetitive effects. Third, proponents overstate their legal case by relying on inapplicable cross-ownership cases and stretching the holdings of those cases. Fourth, at bottom proponents’ concerns are with either conscious parallelism, which is not illegal, or anticompetitive conduct that, if proven, could be addressed using established antitrust doctrines.  Jacob Gramlich and Serafin Grundl, Testing for Competitive Effects of Common Ownership, Finance and Economics Discussion Series 2017-029, Washington: Board of Governors of the Federal Reserve System (2017). This paper, by Federal Reserve staffers, utilizes a distinct methodology to measure the effect of common ownership. Preliminarily, the authors conclude that the results found in the Azar et al. Banking Paper are not robust and that statistical evidence of common ownership impacting competition is mixed. The authors also note that more research is needed before any conclusions about the effect of common ownership on competition in any industry may be drawn.  Pauline Kennedy, Daniel. P. O’Brien, Minjae Song, and Keith Waehrer, The Competitive Effects of Common Ownership: Economic Foundations and Empirical Evidence, Working Paper (2017). This paper analyzes data from the airline industry using two different empirical approaches to estimate the effects of common ownership on airline prices. The two approaches serve as checks on each other and past research on this subject. Both analyses find no evidence that common ownership has raised airline prices. C-3  Thomas A. Lambert and Michael E. Sykuta, The Case for Doing Nothing About Institutional Investors’ Common Ownership of Small Stakes in Competing Firms, University of Missouri School of Law Legal Studies Research Paper No. 2018-21 (2018). This paper explains why purported antitrust concerns about common ownership are overblown and why interventions to limit common ownership are unwarranted presently. The article notes that proposed policy solutions would create welfare losses that would overwhelm any social benefits associated with reducing common ownership.  Barbara Novick, et. al, Index Investing and Common Ownership Theories, BlackRock View Point, (March 2017). This paper supplies an industry perspective on common ownership and argues that placing limits on the ability of asset managers to make investments will essentially put the onus back on asset owners to create diversified portfolios.  Daniel P. O'Brien and Keith Waehrer, The Competitive Effects of Common Ownership: We Know Less than We Think, 81 Antitrust Law Journal 729 (2017). This paper questions the methodology of the Azar et al. Airline Paper and the Azar et al. Banking Paper by analyzing the use of the measure of concentration utilized in studying the airline and banking industries. The paper finds that the key explanatory variable in this research depends on the same underlying factors as those that drive consumer prices, making it likely that the estimates found in the Azar et al. Airline Paper and the Azar et al. Banking Paper are suggesting a relationship between price and common ownership when none may exist.  Menesh S. Patel, Common Ownership, Institutional Investors, and Antitrust, Antitrust Law Journal, Forthcoming (2018). This paper suggests policy proposals to limit common ownership would generate substantial competitive harm and argues that common ownership should continue to be evaluated on a case-by-case basis.  Edward B. Rock and Daniel L. Rubinfeld, Antitrust for Institutional Investors, NYU Law and Economics Research Paper No. 17-23 (2017). This paper casts doubt on the applicability of the methodology used in the papers that have found a statistically significant relationship between common ownership and price. Other Related Papers  José Azar, Martin C. Schmalz, and Isabel Tecu, The Competitive Effects of Common Ownership: Economic Foundations and Empirical Evidence: Reply, Working Paper (2017). This paper asserts that certain features of the empirical analysis done in Kennedy et al. cast doubt on the reliability of their method and conclusion.  José Azar, Martin C. Schmalz, Isabel Tecu, Reply to: “Common Ownership Does Not Have Anti- Competitive Effects in the Airline Industry”, Working Paper (2018). This paper argues that the findings in Dennis et al. (described above) are incorrect.  Einer Elhauge, New Evidence, Proofs, and Legal Theories on Horizontal Shareholding, Working Paper (2018). This paper argues that common ownership can cause anticompetitive effects even if all common holdings are small. The paper also aims to demonstrate that critiques of earlier C-4 empirical studies showing adverse price effects for airlines and banking are generally invalid and advances new antitrust legal theories for addressing common ownership.