Mutual Funds, "Common Ownership" and Ownership Concentration
Guest post by Richard Berner, Executive-in-Residence (Center for Global Economy and Business) and Adjunct Professor, NYU Stern School of Business
America faces two interrelated long-term challenges: rising longevity and inadequate retirement saving. The combination of declining private, defined-benefit pension plans and concerns about the viability of federal entitlements has intensified these challenges. While the economic recovery has raised confidence about retirement resources at the margin (see here), workers and retirees remain concerned about how they will meet future basic expenses, medical needs or the cost of long-term care.
Those developments mean that achieving saving goals increasingly must rely on individuals’ thrift and intelligent, efficient investing. Tax-advantaged vehicles that encourage saving (like 401k and IRA accounts), and efficient investment vehicles like mutual funds that follow market-wide stock price indexes are cornerstones of that system.
Yet, some scholars of industrial organization claim that collective investment vehicles―mutual funds, exchange-traded funds (ETFs), and the like―involve “common ownership” that results in softened competition by the firms included in their portfolios (see here, here and here). And, key antitrust enforcers, like the European Competition Commissioner, are looking carefully at this issue. In this post, I argue that the evidence for a causal link between the rise of collective investment vehicles and diminished competition is weak, and far from sufficient to justify interventions that would diminish the attractiveness of these saving mechanisms.
Taking a step back, how should we evaluate collective investment vehicles? In my view, these offer huge benefits. Those benefits are particularly evident for small savers, who can avail themselves of passively-managed, broad index funds that allow cheap and easy diversification in a variety of assets, especially equities. These funds offer attractive risk-return profiles at very low administrative cost. An extreme example is Fidelity’s recent offering of zero-fee passive index funds. Consequently, passively managed funds have grown rapidly in both absolute terms and as a share of the total (see chart).
Actively and passively managed domestic U.S. equity mutual funds and exchange-traded funds (Percent of total equity outstanding), 2000-17
Source: 2018 Investment Company Fact Book, Figure 2.8, pg. 43.
In addition to their low cost, indexed mutual funds offer additional advantages. In particular, they are liquid, offer one-day redemption, and come with sufficient regulatory oversight that protects small investors, helping to create the trust that is essential for financial-market and system functioning. And unlike banks, index funds channel assets to users without balance-sheet leverage, making them less likely to create financial-system risks. (As discussed here, because some structures hold illiquid assets, they can still be subject to runs.) So it’s no surprise that household holdings of equities, bonds and other assets through mutual funds or ETFs have risen sharply in recent decades.
Against these benefits, as noted above, some scholars of industrial organization argue that index investing, and the managers who passively manage broad portfolios of securities on behalf of investors, are promoting behavior that softens competition among nonfinancial corporations. To be sure, if a portfolio manager were to urge firms in an industry to act in concert to restrain capacity and raise prices, and if they were to follow this advice and raise the prices they charge, this would create legitimate concern (for the view of legal scholars, see here).
But those who claim harm take their case much further than arguing for the possible violation of traditional U.S. antitrust law (namely Section 1 of the Sherman Antitrust Act). They assert that the mere existence of large investment managers, such as those that administer and/or manage both actively managed and passive index mutual funds owned by individual investors, creates powerful incentives for anticompetitive behavior on the part of the firms whose equities are included in mutual fund portfolios. In their view, even when the assets are passively managed, as they are in equity index funds, the inclusion of the equity securities of all or most of the members of an industry—what they call common ownership—leads industry managers to act as if they were being urged to collude.
Analysts who are concerned about “common ownership” base their conclusions primarily on two pieces of evidence: measures of industry concentration and the relationship of concentration to prices. First, they note that, adjusting for what they see as a consequence of large passive index fund ownership, a standard measure (which is called a Herfindahl Hirschman index) suggests concentration far in excess of thresholds authorities normally use to instigate antitrust proceedings. Researchers go on to note that the additional concentration afforded by ownership through collective investment vehicles appears to empower firms with the ability to raise prices. In the airline industry, for example, they find that “common ownership” is associated with ticket prices approximately 3% to 7% higher in the average U.S. airline route than would be the case under “separate ownership.”
Proponents of this view suggest draconian remedies for the harm they see resulting from ownership through collective investment vehicles. They propose that administrators of mutual funds should either be forced to limit their cross-industry holdings, or that they should be denied voting rights.
Those criticizing collective investment vehicles are making important and serious claims. If the preliminary evidence provided were confirmed broadly, it would clearly suggest anticompetitive behavior that may require some sort of remedy. In fact, however, the evidence is far from convincing, suggesting that any policy actions would be premature, and probably unjustifiable for several reasons. First, even if entirely corroborated, the analysis fails adequately to weigh the harm against what are surely substantial benefits. As I emphasized earlier, investors, especially small ones, gain substantially from the ability to purchase low-cost, passive index funds. The economies of scale in this business appear critical to keeping costs low.
Second, advocates of aggressive remedies mistake administration by fund managers with beneficial ownership. Portfolio managers are agents, not principals; they do not own the assets that their clients authorized them to manage. Rather, it is the account holders individually who own shares in funds that are then collectively managed. In the case of passive index funds, the portfolio managers make no fundamental decisions whatsoever; they merely select an array of assets to mimic the behavior of a broad index, like the S&P 500. Put differently, the term “common ownership” is a misnomer. Neither the ultimate shareholder nor the managers have the means for promoting anticompetitive conduct, even if they surprisingly wished to do so. As a result, the adjusted measures of concentration simply do not apply. (Bresnahan and Salop, the original developers of the index used in these analyses, employed it to analyze the effects of joint ventures, and Salop and O’Brien used it to assess the competitive effects of horizontal mergers, where true ownership was indeed common.)
Third, and most important, the advocates of anti-trust remedies fail to specify the causal mechanism by which index fund managers are allegedly leading the universe of publicly-traded nonfinancial firms to engage in collusive behavior. It is insufficient to argue that the managers of these enterprises vaguely anticipate anti-competitive incentives to emanate at some stage from the supposed desires of passive-index-fund portfolio managers. At the least, we should require concrete evidence that a specific fund manager actually urges such behavior, or that they encourage it through a clear pattern of voting the shares they control. But if such evidence were to exist, it could well be grounds for penalizing the behavior of that fund manager, not for undermining a class of socially useful investment vehicles.
As an aside, I view it as a stretch to go from the empirical observation that higher levels of collective ownership through mutual funds is correlated with higher industry prices to an interpretation that an increase in the former causes the latter. To see why, consider the case of the airline industry. Over the past several decades, a wave of bankruptcies, cutbacks in capacity, consolidations, new technologies to manage demand, pricing and operations, and new managers have all contributed to increased profitability. None of these have anything to do with the skyrocketing ownership by small investors through index funds (see the above chart). In our view, then, the advocates of constraining collective investment vehicles have uncovered correlation, not causation.
Regarding remedies, forcing divestiture to limit cross-industry holdings would reduce the diversification benefits and economies of scale inherent to mutual funds. It could trigger the sale of trillions of dollars of equities, along with the creation of a huge number of alternative funds, thus raising costs for savers. Proponents assert that the reduction in diversification would be de minimus; however, that ignores the significant fixed costs of creating bespoke indexes and hedging risk in them. And the idea that savers, especially small ones, could achieve similar diversification by buying many mutual funds at the same cost is simply fatuous.
Finally, compared to the divestiture remedy, the proposal to limit voting rights by the managers of mutual funds is far less intrusive and probably less damaging, but still not warranted. It is based on the idea that firms’ managers assume that index fund managers vote monolithically to soften competition. To be sure, fund investors delegate voting to fund managers, and fund managers do engage with firm managements on behalf of their investors, e.g., to promote best practices for corporate governance. But asset managers do not have the opportunity to vote on competitive strategy, and public filings show that fund managers in the same fund family often split their votes for and against management proposals, depending on the individual funds’ objectives (see, for example, here).
Legal experts also contest the arguments the “common ownership” authors make. I have no particular legal expertise, but if there is demonstrable, explicit anticompetitive behavior in a particular industry, then the government certainly appears to have the tools and authority to address it. While proving tacit anticompetitive behavior is difficult, antitrust enforcers can't just assume that it exists. In any case, altering management of passive index funds to achieve these goals is a bridge way too far.
What to conclude? Excessive concentration in industry may sap economic dynamism and foster both anticompetitive behavior and economic inequality. These are important externalities, and if there is demonstrated harm, policies, potentially including antitrust enforcement, should vigorously address them. But there is yet to be a compelling analytical or empirical case for “common ownership” being a causal factor behind them. And any costs found by its advocates seem to me to fall far short of the benefits of index fund investing for savers and the economy.
Acknowledgments: The author thanks Steve Cecchetti, Chris Conlon, Kim Schoenholtz, and Larry White for helpful discussions and comments.