Why Common Ownership Is Not an Antitrust Problem

2 Comments Print E-Mail Tweet

Douglas H. Ginsburg is a Judge of U.S. Court of Appeals for the District of Columbia Circuit and Professor of Law at the Antonin Scalia Law School, George Mason University. This post is based on his recent co-authored article, forthcoming in Frederic Jenny: Standing Up for Convergence and Relevance in Antitrust, Liber Amicorum – Volume II (Concurrences, 2019). Related research from the Program on Corporate Governance includes Index Funds and the Future of Corporate Governance: Theory, Evidence, and Policy by Lucian Bebchuk and Scott Hirst (discussed on the Forum here); The Agency Problems of Institutional Investors by Lucian Bebchuk, Alma Cohen, and Scott Hirst (discussed on the Forum here); Horizontal Shareholding (discussed on the Forum here); and New Evidence, Proofs, and Legal Theories on Horizontal Shareholding (discussed on the Forum here), both by Einer Elhauge.

More than 100 million Americans now hold a stake in a mutual fund, ETF, or closed-end fund. (2018 Investment Company Factbook, at i, 34.) Those funds have an aggregate value of $22 trillion. (Id.) (For simplicity, we refer herein to all of these investments as “mutual funds.”) Because a few families of mutual funds have significant holdings in many of the same companies, legal and economic scholars have debated whether the funds’ “common ownership” impairs competition among their portfolio companies, in violation of the antitrust laws. The U.S. antitrust agencies—the U.S. Department of Justice Antitrust Division and the Federal Trade Commission—define common ownership as “the simultaneous ownership of stock in competing companies by a single investor, where none of the stock holdings is large enough to give the owner control of any of those companies.” (See Note to the OECD by the United States at ¶ 1.)

On one side stand several scholars (the “proponents”), who believe mutual funds harm competition, and thereby consumers. These scholars—including Harvard Law School Professor Einer Elhauge—argue that mutual funds, by holding shares in competing firms, such as the largest banks and the major airlines, cause those firms to compete with each other less aggressively. They also insinuate that mutual funds may pass competitively sensitive information between firms in which they are invested.

The current debate focuses more narrowly on the special case in which “[a] small group of institutions”—such as Blackrock, Fidelity, State Street, and Vanguard—have collectively “acquired large shareholdings in horizontal competitors throughout our economy.” (See Einer Elhauge’s article at 1267.) Proponents of antitrust enforcement against common ownership believe these holdings may violate Section 1 of the Sherman Act, which makes unlawful every unreasonable “contract, combination, or conspiracy in restraint of trade,” such as a price-fixing agreement, and Section 7 of the Clayton Act, which prohibits the anticompetitive acquisition of corporate assets or securities.

On the other side stand an array of skeptics who question whether the legal theory is sound and the proffered empirical evidence sufficient. (See, e.g., a paper by Edward Rock and Daniel Rubinfeld here, another by Thom Lambert and Michael Sykuta here, and another by Menesh Patel here.) Several of these scholars note that prohibiting common ownership would be poor public policy because it would deprive investors of the ability to spread risk across a diversified financial portfolio with very low transaction costs.

A third group has been willing simply to assume the legal theory and economic evidence establish antitrust liability and proceeded to propose far-reaching remedies. For example, Eric Posner, Fiona Scott Morton, and Glen Weyl propose a new antitrust rule that institutional investors—primarily mutual funds—own “no more than 1 percent in more than a single firm in oligopolies.” (Paper available on SSRN here.) Fiona Scott Morton and Herbert Hovenkamp suggest, somewhat more equivocally, that “partial divestiture” of mutual funds’ existing stockholdings seems “to be most promising.” (See paper here.)

As the debate goes on, U.S. antitrust enforcers remain unconvinced that common ownership is an antitrust violation. When the two antitrust agencies recently submitted a joint statement to the OECD, they noted the debate is in “its early stage of development” and declined “at this time to make any changes to [DOJ or FTC] policies or practices with respect to common ownership by institutional investors.” (OECD Note ¶ 15.) Several prominent enforcers appear to share this lack of conviction. At the FTC, Commissioner Noah Phillips believes “the empirics remain unsettled,” mutual funds “do not appear to be at the apex of a massive antitrust conspiracy,” and that the claimed economic blockbuster “seems a little light on plot.” (See speech here.) Bruce Hoffman, the Director of the FTC Bureau of Competition, has expressed similar sentiments. (See coverage here.) There is some evidence senior DOJ leadership is likewise skeptical; in 2016 Barry Nigro, who is now Deputy Assistant Attorney General for Antitrust, wrote in this publication that an antitrust case against common ownership “is likely to encounter skepticism in the courts.” (See post here.)

In our article my co-author Keith Klovers and I offer four observations that should significantly narrow the focus of the legal debate. First, it is important to distinguish between investment management and economic ownership. The proponents conflate the two, thereby wrongly attributing to “owners” conduct undertaken by others. The investment manager is only the nominal owner, acting as a fiduciary for the investors, who are the real parties in interest. Proponents defend treating investment managers like true owners by noting that an investment manager “generally” or “typically” aggregates share holdings across its funds and votes as a bloc all the shares it manages.

Even when an institutional investor does vote all the shares it manages the same way, as is usually the case, that does not justify aggregating all the shares held by all the funds of a particular institutional investor. That two Fidelity funds voted the same way on a shareholder proposal on an environmental, social, lobbying, or proxy procedural issue—which together account for the overwhelming majority of shareholder proposals (see Mueller and Ising’s post on this blog)—tells us little about whether they would vote the same way for a proposal that would direct a portfolio company to compete less aggressively. Indeed, there is good reason to think these funds would vote differently, for they have diverse stock positions and therefore different competitive incentives. For example, the Fidelity Select Consumer Finance fund (FSVLX) holds stock in mortgage lenders, credit card lenders, and major national banks, whereas the Fidelity Select Banking fund (FSRBX) holds stock in the same major banks plus a constellation of regional banks, such as Valley National Bancorp. So an investor in the former Fidelity fund would benefit if Bank of America takes sales from Valley National, whereas an investor in the latter Fidelity fund likely would not. Yet the proponents’ analysis flows from the twin assumptions that the investment manager owns the shares it holds and that its funds all face the same incentives and therefore will act as one.

Second, the current empirical evidence that common ownership causes anticompetitive harm is limited and hotly disputed. Indeed, only a minority of the economic studies find a correlation between the degree of common ownership and higher prices, and even those do not establish causation. Although the mechanism of harm, if any, is unknown, most proponents nonetheless assume a causal relationship, which supports finding an antitrust violation and imposing a sweeping remedy. (To name one prominent example, so say Scott Morton and Hovenkamp.) This approach puts the legal cart before the economic horse; indeed, not all practices with anticompetitive effects are unlawful. It is at best simply premature, based upon the empirical evidence available today, to declare common ownership an antitrust violation and thereby to force the reorganization of trillions of dollars in mutual fund assets.

Third, there is no clear legal basis for antitrust liability under either § 7 of the Clayton Act or § 1 of the Sherman Act. The argument that common ownership by investment managers violates § 7 is hardly—as Professor Elhauge asserts—a straightforward application of existing doctrine; on the contrary, it is a novel application devised ad hoc for a single industry. Nonetheless, Professors Scott Morton and Hovenkamp have come to much the same conclusion, relying ultimately either upon the “plain meaning” of § 7 or upon inapposite cases that address a stock acquisition in a cross ownership situation, i.e., the partial acquisition of one firm by a rival. As the U.S. antitrust agencies pointed out in their aforementioned Note to the OECD, however, essentially all § 7 precedents address cross ownership, which involves different economic principles and a more obvious threat to competition than does common ownership. The hypothesis that common ownership is a § 1 violation, advanced principally by Elhauge, is at least equally dubious.

Fourth, common ownership does not require antitrust enforcers to apply an entirely new analytic framework to the purported problem; traditional antitrust principles suffice and remain applicable. The evidence cited by proponents suggests their underlying concern is with conventional types of conduct that may lessen competition, whether a hub-and-spoke conspiracy, the exchange of competitively sensitive information, or conscious parallelism in oligopoly markets. Hub-and-spoke conspiracies have long been prosecuted as violations of Section 1. (See, e.g., Interstate Circuit, Inc. v. United States, 306 U.S. 208, 232 (1939).) Similarly, antitrust law has long treated the exchange of competitively sensitive information as a “facilitating practice” from which a court may infer the existence of an unlawful conspiracy. (See, e.g., Am. Column & Lumber Co. v. United States, 257 U.S. 377, 394-95 (1921).) Courts do not, however, infer an illegal Section 1 agreement from an indirect—one might say second-order—facilitating practice. As for conscious parallelism, the Supreme Court has been clear that the practice is “not in itself unlawful.” (Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 227 (1993).) If commentators believe the antitrust laws should be amended to prohibit conscious parallelism, then they should say as much.

In summary, at this stage there is little to indicate common ownership is an antitrust problem. To date proponents have rested their case upon a small and hotly contested body of empirical work, a thicket of assumptions, and a couple of novel legal theories. At bottom proponents’ complaints focus either upon conduct that is not unlawful or upon allegedly anticompetitive conduct that, if proven, could be attacked under established antitrust doctrines addressing horizontal conspiracies and the exchange of competitively sensitive information. Efforts to repurpose Section 1 so as to prohibit mutual funds from holding small, non-controlling stock positions in competing firms is both poor policy and contrary to established law.