Anti-Competitive Effects of Common Ownership via SSRN

José Azar

Charles River Associates (CRA)

Martin C. Schmalz

University of Michigan, Stephen M. Ross School of Business

Isabel Tecu

Charles River Associates (CRA)

April 15, 2015

Ross School of Business Paper No. 1235


Many natural competitors are jointly held by a small set of large diversified institutional investors. In the US airline industry, taking common ownership into account implies increases in market concentration that are 10 times larger than what is “presumed likely to enhance market power” by antitrust authorities. We use within-route variation over time to identify a positive effect of common ownership on ticket prices. A panel-IV strategy that exploits BlackRock’s acquisition of Barclays Global Investors confirms these results. We conclude that a hidden social cost — reduced product market competition — accompanies the private benefits of diversification and good governance.

Anti-Competitive Effects of Common Ownership – Introduction

A long theoretical literature in industrial organization recognizes that common ownership of natural competitors by the same investors reduces incentives to compete: the benefits of competing aggressively to one firm { gains in market share { come at the expense of firms that are part of the same investors’ portfolio (Rotemberg, 1984; Gordon, 1990; Gilo, 2000; O’Brien and Salop, 2000; Gilo, Moshe, and Spiegel, 2006). Theory thus predicts that common ownership pushes product markets toward monopolistic outcomes, implying a deadweight loss for the economy and particularly adverse consequences for consumers. The empirical literature and regulatory practice have focused on the special case of full mergers and acquisitions.

By contrast, it is an open empirical question with important policy implications whether common ownership that is attained by partial acquisitions of firms by large asset management companies that require no regulatory approval also decreases competitiveness of the product market in significant ways. This paper provides a first answer to this question, in two steps. We first ask: how large are current levels of common ownership, and what are the implications for market concentration? Second, do present-day common ownership levels adversely affect product market competition?

To approach the first question, note that highly diversified pension funds, mutual funds, and other institutional investors now hold a high (70%-80%) and increasing share of US publicly traded firms (McCahery, Starks, and Sautner, 2014; Rydqvist, Spizman, and Strebulaev, 2014), reflecting the benefits they generate for retail investors. Because several asset management companies are also extremely large, the same asset management company is often the single largest shareholder of several firms in the same industry. Table 1 provides examples. The potential scale of the resulting problem for product market competition spans across all industries and economies with tradable securities. For a quantitative evaluation, we focus on the airline industry as a laboratory. The availability of high-quality route-level price and quantity data enables us to more cleanly identify the effect of common ownership on product prices than would be possible in firm-level studies across industries. Treating each route as a market, we first calculate measures of market concentration that take into account the network of cash flow and control rights that constitute the airlines’ shareholders’ economic interests. Such “modified Herfindahl-Hirschman indexes” (MHHIs) were developed by Bresnahan and Salop (1986) and O’Brien and Salop (2000), and are accepted tools in regulators’ assessment of competitive risks imposed by cross-ownership and common ownership by \activist” investors. We use them also for the measurement of anti-competitive incentives of other owners, irrespective of their investment style.

We find that the anti-competitive incentives implied by common ownership concentration alone { which come on top of those implied by the traditional HHI measure of market concentration and are measured on the same scale { are more than 10 times larger than what the FTC/DOJ 2010 horizontal merger guidelines presume \to be likely to enhance market power.” They are also 10 times larger than the HHI-limit beyond which the burden of proof shifts from the regulator to the involved private parties to show that the implied concentration is not likely to enhance market power. The magnitude of common ownership concentration furthermore dwarfs the time-series variation in HHI. These magnitudes suggest that it is reasonable to expect an effect of common ownership on product prices.

We next test whether these anti-competitive incentives do indeed translate into measurable effects on product market competition. Specifically, we examine whether changes in common ownership concentration over time in a given route are associated with changes in ticket prices in the same route. Our first set of regressions can be thought of an analysis, spanning more than a decade, of the effect on product prices of partial mergers that are quasi-continuously consummated and dissolved among (almost) all players of the industry. For example, theory predicts that the entrance of an independent player (a firm not owned by the same set of investors who own the incumbent airlines) makes competition more aggressive. By contrast, competition softens in a route when incumbent airlines’ owners buy significant ownership and control stakes in a thus-far independent carrier serving the same route. Online Appendix B provides a stylized example to illustrate this strategy.

Common Ownership

See full PDF below.