Anticompetitive Effects of Common Ownership

• Author: ISABEL TECU,JOSÉ AZAR,MARTIN C. SCHMALZ
• Date: 01 August 2018
• DOI: http://doi.org/10.1111/jofi.12698
• Published date: 01 August 2018

THE JOURNAL OF FINANCE •VOL. LXXIII, NO. 4 •AUGUST 2018

Anticompetitive Effects of Common Ownership

JOS ´

E AZAR, MARTIN C. SCHMALZ, and ISABEL TECU∗

ABSTRACT

Many natural competitors are jointly held by a small set of large institutional in-

vestors. In the U.S. 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.1Within-route changes in

common ownership concentration robustly correlate with route-level changes in ticket

prices, even when we only use variation in ownership due to the combination of two

large asset managers. We conclude that a hidden social cost—reduced product market

competition—accompanies the private benefits of diversification and good governance.

∗Jos´

e Azar is with IESE Business School. Martin Schmalz is with the Ross School of Business,

University of Michigan, the Centre for Economic Policy Research, and the European Corporate

Governance Institute. Isabel Tecu is with Charles River Associates. For helpful comments, we

thank Alex Edmans, Alminas Zaldokas, Alon Brav,Amit Seru, Andrei Shleifer, Andrew Ellul, Carl

Shapiro (discussant), Charles Hadlock, Cindy Alexander, Daniel Crane, Daniel Ferreira (discus-

sant), Daniel Greenfield (discussant), David Reitman, Dirk Jenter, Einer Elhauge, Evgeny Lyan-

dres (discussant), Farzad Saidi (discussant), Fiona Scott Morton, Francine Lafontaine, Glen Weyl,

Gregor Matvos, Han Kim, Holger M¨

uller, Jeremy Stein, Jesse Shapiro, Johan Hombert, John

Coates, Jonathan Berk (discussant), Juwon Kwak (discussant), Kai-Uwe K¨

uhn, Louis Kaplow,

Maggie Levenstein, Martino DeStefano, Matthew Shapiro, Michael Roberts (the Editor), Nancy

Rose; Peter Cramton, Robert Levinson, Ryan Kellogg, Sarath Sanga (discussant), Scott Stern,

Severin Borenstein, Sheridan Titman (discussant), Steven Salop, Susan Athey, Todd Gormley,

Toni Whited, Umit Gurun, VicenteCu ˜

nat (discussant), Vikram Nanda (discussant), Yossi Spiegel,

anonymous referees; several corporate governance and proxy voting executives, the general coun-

sel, and a board member of various large asset management companies; a pricing manager of a

major airline; our colleagues; seminar participants at ASU, Berkeley, Bonn, BC, Charles River

Associates, Chicago Booth, Columbia GSB, DICE, FRB of New York, FRB of Governors, Goethe

Universit¨

at, Harvard, Humboldt Universit¨

at, IESE, INSEAD, K¨

oln, Mannheim, McGill, MIT,Stan-

ford, Stockholm, Toulouse School of Economics, Tilburg, U.S. Department of Justice, UV Amster-

dam, UBC, University of Michigan, UNC, Western University, Yale; conference audiences at the

American Bar Association, EARIE, ESWC, EFA, FTC Microeconomics Conference, IFN (Stock-

holm), IIOC, LBS Symposium, LSE Adam Smith, LSE Economic Networks and Finance, NBER

Corporate Finance, NBER SI Law & Economics, Searle Antitrust Conference, Princeton JRCPP,

TAU Finance Conference, Texas Finance Festival, and the Utah Winter Finance Conference; and

Oliver Richard for help with the DB1B data. Schmalz is grateful for generous financial support

through an NTT Fellowship from the Mitsui Life Financial Center. Bret Herzig, Yichuan Wang,

and Eric Wilson provided excellent research assistance. All of the authors have read the Journal

of Finance’s disclosure policy and have no conflicts of interest to disclose. Charles River Associates

had the right to review the article prior to its circulation. The views expressed herein are the

views and opinions of the authors and do not reflect or represent the views of other consultants or

experts who are affiliated with Charles River Associates, nor do they reflect the views and opinions

of Charles River Associates or any of the organizations with which the authors are affiliated.

1See https://www.ftc.gov/sites/default/files/attachments/merger-review/100819hmg.pdf.

DOI: 10.1111/jofi.12698

1513

1514 The Journal of Finance R

ALONG THEORETICAL LITERATURE IN industrial organization predicts that par-

tial common ownership of natural competitors by overlapping sets of investors

can reduce firms’ incentives to compete: the benefits to one firm of compet-

ing aggressively—for example, gains in market share—come at the expense

of firms that are part of the same investors’ portfolio, reducing total portfolio

value. Theory thus predicts that common ownership can push product markets

toward monopolistic outcomes and imply a deadweight loss for the economy

and adverse consequences for consumers.

Contrasting this theoretical argument, the empirical literature thus far

largely assumes that common ownership interests by financial institutions

do not matter for firms’ objectives and product market outcomes. The question

of whether this assumption is warranted has first-order implications for many

areas of economics, such as finance, industrial organization, macroeconomics,

and antitrust policy. In this paper, we aim to shed light on this question by

studying the effect of common ownership on product market outcomes. Specifi-

cally,we ask, first, how large are current levels of common ownership, and what

are the implications for market concentration measures, and second, do current

levels of common ownership adversely affect product market competition?

With respect to the scope of the first question, highly diversified mutual fund

families and other institutional investors now hold a high (70% to 80%, Ben-

David et al. (2015)) and increasing share of U.S. publicly traded firms’ equity.

Because several asset management companies are also extremely large, the

same fund family is often the single largest beneficial owner of several firms in

an industry,with similarly diversified investors following suit. Table Iprovides

examples.2The potential scale of the resulting antitrust problem spans across

all industries, geographies, and economies with tradable equity securities.

The second question presents a formidable identification challenge.3Cor-

relations between common ownership and price-cost margins across firms or

industries do not necessarily have a causal interpretation, as reverse causality

or potentially omitted control variables may play a role. To take a step toward

addressing these challenges, we focus on the U.S. domestic airline industry

as a laboratory. This industry focus is motivated by the fact that high-quality

route-level price and quantity data are publicly available, and each route can

be considered a separate market. These features allow us to relate common

ownership concentration to prices within the same firm, period, and indus-

try, which reduces the amount of confounding variation. Furthermore, using

2In 2013, BlackRock was the single largest shareholder of one-fifth of all American publicly

traded firms (see Davis (2013) and December 7, 2013, The rise of BlackRock, The Economist,see

also Craig, Susanne, May 18, 2013, The giant of shareholders, quietly stirring, The New York

Time s). Fichtner, Heemskerk, and Garcia-Bernardo (2016) calculate that the combined holdings

of BlackRock, Vanguard, and State Street make them the largest investor of 88% of all firms in

the S&P 500. See Roe (1990) and Elhauge (2015) for a discussion of the legal constraints of such

ownership structures.

3An obvious problem would exist if one beneficial owner controlled 50% or more of the voting

securities of all firms in the industry.An open empirical question is whether Ninvestors that each

hold more than 50/N% of votes in all firms, or similar structures, can have similar effects.

Anticompetitive Effects of Common Ownership 1515

Tabl e I

Illustrative Cases of Within-Industry Common Ownership Links

This table shows the largest (institutional and noninstitutional) beneficial owners and corresponding stakes for an illustrative sample of U.S. publicly

traded natural competitors as of 2016Q2. The data source is S&P Capital IQ. Panel C corresponds to Azar, Raina, and Schmalz (2016). Berkshire’s

holdings in Bank of America (*) are warrants without voting rights. Panel D reflects holdings as of 2016Q4.

Panel A: Technology Firms

Apple [%] Microsoft [%]

Vanguard 6.05 Vanguard 6.41

BlackRock 5.72 BlackRock 5.80

State Street 3.82 Capital Research 4.76

Fidelity 2.34 - Steve Ballmer - 4.24

Northern Trust Corporation 1.26 State Street 3.80

- Bill Gates - 2.54

T. Rowe Price 2.27

Panel B: Pharmacies

CVS [%] Walgreens Boots Alliance [%] Rite Aid [%]

Vanguard 6.66 -Stefano Pessina- 13.06 Vanguard 7.24

BlackRock 6.02 Vanguard 5.58 BlackRock 4.20

State Street 4.00 BlackRock 4.55 Arrowgrass Capital 3.55

Fidelity 3.67 KKR 3.38 Franklin Resources 2.87

Wellington 2.37 State Street 3.34 Pentwater Capital 1.89

T. Rowe Price 2.70

Panel C: Banks

JPMorgan Chase [%] Bank of America [%] Citigroup [%]

Vanguard 6.28 Berkshire Hathaway* 6.90 BlackRock 6.43

BlackRock 6.28 Vanguard 5.94 Vanguard 5.96

State Street 4.12 BlackRock 5.94 State Street 4.04

Capital Research 3.68 State Street 4.01 Fidelity 3.00

Fidelity 2.10 Fidelity 2.37 Invesco 1.67

Wells Fargo [%] PNC Financial [%] U.S. Bancorp [%]

Berkshire Hathaway 10.46 Wellington 8.34 BlackRock 6.51

Vanguard 5.67 Vanguard 6.30 Berkshire Hathaway 5.94

BlackRock 5.42 BlackRock 5.03 Vanguard 5.59

State Street 3.68 State Street 4.33 Fidelity 4.12

Wellington 2.55 Barrow Hanley 3.71 State Street 3.84

(Continued)