DATE DOWNLOADED: Mon Jun 17 04:32:25 2024

SOURCE: Content Downloaded from HeinOnline

Citations:
Please note: citations are provided as a general guideline. Users should consult their preferred
citation format's style manual for proper citation formatting.

Bluebook 21st ed.

Vittoria Battocletti, Luca Enriques & Alessandro Romano, Dual-Class Shares in the Age
of Common Ownership, 48 J. CORP. L. 541 (2023).

ALWD 7th ed.

Vittoria Battocletti, Luca Enriques & Alessandro Romano, Dual-Class Shares in the Age
of Common Ownership, 48 J. Corp. L. 541 (2023).

APA 7th ed.

Battocletti, Vittoria, Enriques, Luca, & Romano, Alessandro. (2023). Dual-class
shares in the age of common ownership. Journal of Corporation Law, 48(3), 541-576.

Chicago 17th ed.

Vittoria Battocletti; Luca Enriques; Alessandro Romano, "Dual-Class Shares in the Age
of Common Ownership," Journal of Corporation Law 48, no. 3 (Spring 2023): 541-576

McGill Guide 9th ed.

Vittoria Battocletti, Luca Enriques & Alessandro Romano, "Dual-Class Shares in the
Age of Common Ownership" (2023) 48:3 J Corp L 541.

AGLC 4th ed.

Vittoria Battocletti, Luca Enriques and Alessandro Romano, 'Dual-Class Shares in the
Age of Common Ownership' (2023) 48(3) Journal of Corporation Law 541

MLA 9th ed.

Battocletti, Vittoria, et al. "Dual-Class Shares in the Age of Common Ownership."
Journal of Corporation Law, vol. 48, no. 3, Spring 2023, pp. 541-576. HeinOnline.

OSCOLA 4th ed.

Vittoria Battocletti, Luca Enriques & Alessandro Romano, 'Dual-Class Shares in the
Age of Common Ownership' (2023) 48 J Corp L 541 Please note:
citations are provided as a general guideline. Users should consult their preferred
citation format's style manual for proper citation formatting.

-- Your use of this HeinOnline PDF indicates your acceptance of HeinOnline's Terms and
Conditions of the license agreement available at
https://heinonline.org/HOL/License
-- The search text of this PDF is generated from uncorrected OCR text.
-- To obtain permission to use this article beyond the scope of your license, please use:
Copyright Information
 Dual-Class Shares in the Age of Common Ownership

Vittoria Battocletti,* Luca Enriques,** and Alessandro

Romano***

Dual-classsharesand the anticompetitive effects of common ownership are two of the

most discussed corporate governance issues of our time. In this Article, we identify a
hidden connection between them, which allows us to derive policy implications that are
relevantfor both.
The traditionaldebate on dual-class shares is based on the trade-offbetween having
visionaryfoundersfirmly in controlof thefirm and the risk that they extractprivate benefits
of control. We show that the exclusive focus on this trade-off is rooted in the outdated
assumption that all shareholdersare firm-value-maximizing (FVM); that is, they aim to
maximize the value of the firm in which they have invested. However, as the debate on
common ownership acknowledges, diversified institutionalinvestors, a la BlackRock, care
about maximizing the value of theirfunds' portfolios, regardless of what happens to any
individual investee company; they act as portfolio-value-maximizing (PVM) shareholders.
Consequently, they might prefer a lower level of competition in product markets to
maximize thejoint value of the competitors that are in theirportfolio.
In present-dayfinancialmarkets dominatedby PVM institutionalinvestors, dual-class
shares can serve the additionalpurpose of allowing insiders to silence PVMshareholders,
thus mitigatingthe anticompetitive effects of common ownership. Forthis reason, we argue
againstbanning dual-classshares, or even introducinga mandatory time-basedsunset.
That is not the end of the story. The ongoing climate crisis demonstrates that a
relatively low number of major carbon emitters can impose gigantic externalities on the
planet. The macroeconomics literature, in turn, has providedample evidence that a subset
of systemically importantfirms can affect the whole economy. Allowing these companies
to have dual-class shares without limitations grants FVM shareholders, d la Zuckerberg,
the unfettered ability to inflict systemic harm on society. If limitations were imposed on
such shares, PVMshareholderswould internalizepart of these externalitiesvia their other
portfolio holdings and hence have the incentive to steer individual portfolio firms into
being mindful of these externalities.
Thus, we suggest that there should be limits placed on the use of dual-class shares by

* Bocconi University.
** University of Oxford and European Corporate Governance Institute.
*** Bocconi University.
For helpful comments, we are grateful to Elizabeth Pollman, Thomas Reyntjens, Tom Gosling, and participants
in a Bocconi-Oxford Junior Scholars Network Workshop in Corporate Law, a Goethe University Frankfurt
LawFin Research Seminar, the University of Amsterdam conference on Investor Sustainability Engagement, and
the Vanderbilt Law School/Ghent Law School Law & Business Conference. Usual disclaimers apply.
542 The Journalof CorporationLaw [Vol. 48:3

systemically relevantfirms and show how such limitations ought to be tailoredaccording

to afirm's specific ability to impose systemic externalities.

Dual-Class Shares in the Age of Common Ownership

Vittoria Battocletti, Luca Enriques, and Alessandro Romano

I. IN TRO D U CTIO N .............................................................................................................. 542

II. SETTING THE SCENE: PORTFOLIO VALUE MAXIMIZATION, COMMON OWNERSHIP,
AND D UAL-CLASS SHARES ...................................................................................... 547
A. InstitutionalInvestors as Portfolio Value Maximizers ..................................... 547
B. Common Ownership: Evidence andPolicy Proposals.....................................550
C. Dual-Class Shares............................................................................................ 553
III. DUAL-CLASS SHARES AS A PRO-COMPETITIVE DEVICE ............................................. 557
A. Dual-ClassShares and Competition ................................................................ 557
B. Dual-ClassShares and Mandatory Time-Based Sunsets ................................. 562
IV. DUAL-CLASS SHARES AND SYSTEMATICALLY IMPORTANT FIRMS ............................ 564
A. Climate Change ................................................................................................ 565
B. MacroeconomicRisk and CorporateGovernance........................................... 566
C. Limits on Dual-ClassSharesfor Key Firms.....................................................568
1. Carbon Majors.............................................................................................568
2. Macroeconomic-CentralFirms ................................................................... 569
C. Combining Climate and MacroeconomicExternalities ................................... 570
V. ADVANTAGES AND ANTICIPATED OBJECTIONS ........................................................... 571
A. Advantages of Our Proposal............................................................................ 571
B. Possible Counterarguments..............................................................................572
1. "CorporateGovernance is not the Right Tool to Tackle Systemic
Externalities."..............................................................................................572
2. Other Counterarguments............................................................................. 575
VI. CONCLUSION .............................................................................................................. 576

I. INTRODUCTION

Mark Zuckerberg has virtually all of his personal wealth invested in Meta Platforms
(formerly Facebook) while Warren Buffett has "a full 99% of [his] net worth lodged in
Berkshire [Hathaway] stock."2 Their incentives as controllers of their companies are clear:

1. Dhammika Dharmapala & Vikramaditya S. Khanna, Controlling Externalities: Ownership Structure

and Cross-Firm Externalities 38 (Eur. Corp. Governance Inst., Working Paper No. 603, 2021),
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3904316 [https://perma.cc/3SGX-5KR5] (reporting that
Zuckerberg has 94% of his personal wealth invested in Facebook).
2. Letter from Warren Buffett to Berkshire Hathaway Inc. Shareholders (Feb. 22, 2020),
https://www.berkshirehathaway.com/letters/20191tr.pdf [https://perma.cc/8K2E-FGWR].
2023 ] Dual-ClassShares in the Age of Common Ownership 543

3
maximize firm value, regardless of the effect that doing so might have on other firms.
Meanwhile, BlackRock, State Street, and Vanguard (known collectively as the Big Three)
manage $17 trillion invested in thousands of corporations. Their goal is equally clear:
maximize the value of their portfolio, regardless of what happens to any individual firm
therein. 5
These simple observations raise several crucial questions. Under which conditions do
firm-value-maximizing (FVM) shareholders a la Zuckerberg and Buffett have preferences
that are less in line with social welfare maximization than portfolio-value-maximizing
(PVM) shareholders like the Big Three? And under what conditions is the opposite the
case? Can corporate law increase the likelihood that the preferences of FVM shareholders
will not prevail when they are heavily misaligned with social preferences? Similarly, can
corporate law increase the likelihood that the preferences of PVM shareholders will not
prevail when they are heavily misaligned with social preferences? We answer these
questions by uncovering the previously neglected connection between two of the most
controversial corporate governance issues of our time: common ownership and dual-class
shares.
For our purposes, we can define common ownership as the phenomenon whereby an
investor owns stakes in two or more horizontal competitors.6 The rise of institutional
ownership in the last few decades has resulted in a dramatic increase in common
ownership. 7 If we draw two random firms included in the S&P 1500 that operate in the
same industry, there is a 90% chance that they will have a common shareholder owning at

3. To be precise, controlling shareholders aim to maximize the sum of the value of their shares plus any
private benefits they may extract from the companies they control. Yet, even so qualified, their focus will be on
the controlled firm's value, with no concerns for the effects on other companies of a firm-value-focused
management style.
4. BlackRock manages roughly $7.4 trillion of assets. BlackRock, Inc., Annual Report (Form 10-K) 4
(Feb. 28, 2020). State Street manages roughly $3.5 trillion of assets. State Street Corp., Annual Report (Form 10-
K) 77 (Feb. 19, 2021). Vanguard manages roughly $6.2 trillion of assets. Vanguard,
WEALTHMANAGEMENT.COM, https://www.wealthmanagement.com/companies/vanguard (last visited Apr. 8,
2023) [https:/perma.cc/GW2C-XQXD].
5. Again, this statement should be qualified because, as agents of their clients, asset managers' incentives
may deviate from their principals' interests. In fact, asset managers are corporations themselves, acting through
agents whose interests may deviate from those corporations' shareholders. While the reality is thus more nuanced
than we depict it in this introduction, the text not only provides a fair approximation of the asset management
industry's incentives but also reflects a legal obligation for fund managers to maximize the value of each of their
funds' portfolios.
6. To be fair, common ownership has a broader scope than horizontal shareholdings because it "can also
be vertical (between firms related in a supply chain) or conglomerate (between firms that are not horizontal
competitors nor vertically related)." Einer Elhauge, How Horizontal Shareholding Harms Our Economy-and
Why Antitrust Law Can Fix It, 10 HARV. Bus. L. REv. 207, 209 n.1 (2020). Yet, as Elhauge acknowledges, "the
literature often refers to [horizontal shareholdings] as 'common ownership."' Id.
7. See Matthew Backus, Christopher Conlon & Michael Sinkinson, Common Ownership in America:
1980-2017, 13 AM. ECON. J.: MICROECON. 273, 275-76 (2021) (presenting data on the rise of common
ownership); see also Erik P. Gilje, Todd A. Gormley & Doron Levit, Who's Paying Attention? Measuring
Common Ownership and Its Impact on ManagerialIncentives, 137 J. FIN. ECON. 152, 160-61 (2020) (showing
various indicators capturing the dramatic growth of common ownership).
544 The Journalof CorporationLaw [Vol. 48:3

least 5% of the shares in both fu-ms. 8 This ownership structure can lower the level of
competition in oligopolistic product markets. 9 The basic intuition here is that because
institutional investors are interested in maximizing the value of their portfolio, they have
incentives to internalize inter-firm spillovers. As aggressive competition by one of their
portfolio firms might produce negative spillovers for the other portfolio firms operating in
that market, PVM shareholders might prefer a lower level of competition in product
markets. In fact, weaker competition is likely to maximize the joint value of the horizontal
competitors in their portfolio. 10 Against this background, legal scholars have proposed
sweeping structural reforms that would affect virtually every oligopolistic market 11 and
would destroy the business model of institutional investors.12
Parallel to the rise of common ownership, following Google's example in 2004, the
number of firms going public with a dual-class share structure has been increasing. 1 3 Dual-
class shares are used to give disproportionate voting rights to founders and key insiders,
who are thus granted control over the corporation even if they hold a minority of the firm's
cash flow rights. The traditional view is that dual-class shares allow insiders to pursue their
idiosyncratic vision, but at the cost of significantly increasing the agency costs between
14
management and shareholders.
The number of dual-class shares IPOs has been rising despite the opposition of
institutional investors. For many years, they have been vocally advocating for policies that
would either eliminate, or place significant limitations on, dual-class shares. For instance,
the Council of Institutional Investors has long sponsored the idea that new listings of
companies with multiple voting rights should be prohibited1 5 and now pushes for a

8. See Jose Azar, Portfolio Diversification,Market Power, andthe Theory of the Firm 2 (Univ. de Navarra
IESE Bus. Sch., Working Paper No. WP-1 170-E, 2017),
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2811221 [https://perma.cc/VQU4-7ZDQ] (reporting this
data and contrasting it with data in 1999, in which this probability was lower than 20%)..
9. See infra notes 61-65 and accompanying text.
10. See Martin C. Schmalz, Common-Ownership Concentration and CorporateConduct, 10 ANN. REV.
FIN. ECON. 413, 417 (2018) (noting that shareholders have incentives to maximize the value of their portfolio,
and hence that aggressive competition can be "in the interest of an individual firm but decrease a common
owner['s] portfolio value").
11. See infra notes 57-58 and accompanying text.
12. See infra note 59 and accompanying text.
13. See infra notes 73-78 and accompanying text.
14. See infra notes 80-81 and accompanying text.
15. See, e.g., Email from Jeff Mahoney, Gen. Couns. to the Council of Institutional Invs., to Ms. Claudia
Crowley, CEO & Chief Regul. Officer of the NYSE Regul. 2 (Oct. 2, 2012),
https://www.cii.org/files/issuesandadvocacy/correspondence/2012/10 2 12_cii_letterto nysedual_class_st
ock.pdf [https://perma.cc/J7KN-98HY] (arguing that
"[c]orporations should not have classes of common stock with disparate voting rights"); Email from Jeff
Mahoney, Gen. Couns. to the the Council of Institutional Invs., to Mr. Edward S. Knight, Exec. Vice President,
Gen. Couns., & Chief Regul. Officer of the NASDAQ OMX Grp. 2 (Oct. 2, 2012),
https://www.cii.org/files/issues_and_advocacy/correspondence/2012/10_02_ 12_cii_letter_to_nasdaqdualclas
s_stock.pdf [https://perma.cc/KW39-MLLA] (conveying the same message of classes of common stock with
disparate voting rights).
2023 ] Dual-ClassShares in the Age of Common Ownership 545

mandatory time-based sunset for dual-class structures.16 The Investor Stewardship Group,
a collective of U.S. institutional investors that includes the Big Three, voices a similar
17
opposition to dual-class shares.
The debates on the effects of common ownership and whether companies should be
allowed to adopt dual-class shares have thus far moved on two parallel tracks. This is
because the traditional discussion on dual-class shares is rooted in a world of FVM
shareholders and focuses on the trade-off between having visionary founders in control of
the firm and the risk that those same founders will extract private benefits of control,
thereby steering the firm away from value maximization. But in a world in which FVM
and PVM shareholders coexist, there is a close relationship between dual-class shares and
common ownership because a dual-class structure has a crucial impact on the relative
influence of FVM and PVM shareholders with respect to how firms are managed. Other
things being equal, it is more likely that a dual-class-shares company would cater to FVM
rather than PVM shareholders' interests.
To put it differently, at dual-class companies, the preferences of PVM shareholders
holding shares in horizontal competitors-and therefore favoring a lower level of
competition-are less likely to influence the management. Thus, dual-class companies can
be expected to compete aggressively in oligopolistic markets regardless of the presence of
common owners. That, in turn, forces other firms to engage in more vigorous competition.
The role that dual-class shares can play in curbing the anticompetitive effects of common
ownership clearly goes against an across-the-board ban on dual-class shares or even the
8
option, supported by many institutional investors, of a time-based sunset clause.1 The
existing evidence supports this argument: to date, all studies showing the anticompetitive
effects of common ownership have referred to sectors in which there are no companies
with dual-class structures among the leading competitors.19
Does that mean that companies should be allowed to adopt dual-class shares without
any limitations? Not across the board, we argue. However, our reason for imposing
limitations on dual-class shares differs from the one traditionally suggested in the literature.
Instead of looking at intra-firm dynamics and, in particular, at the agency costs between

16. See Email from Jeff Mahoney, Gen. Couns. for the the Council of Institutional Invs., to Maxine Waters,
Chairwoman of the U.S. House of Reps. Comm. on Fin. Servs., & Patrick T. McHenry, Ranking Member of the
U.S. House of Reps. Comm. on Fin. Servs. 2 (Oct. 1, 2021),
https://www.cii.org/files/issues andadvocacy/correspondence/2021 /October%20 I,%202021 %20letter%20to%
20Committee%20on%2OFinancial%2oServices%20(final).pdf [https://perma.cc/ZZJ6-UNR8] (quoting SEC
proposed legislative language stating that, "[i]f a company chooses to issue multiple classes of stock with differing
voting rights, then the dual-class stock must contain a 'sunset' provision").
17. See Corporate Governance Principles for U.S. Listed Companies, INv. STEWARDSHIP GRP. 2,
https://isgframework.org/corporate-governance-principles/ [https://perma.cc/B5XW-8HDG] (last visited Apr. 8,
2023) (identifying the idea that "[s]hareholders should be entitled to voting rights in proportion to their economic
interest" as a fundamental principle of corporate governance).
18. See Email from Jeff Mahoney to Maxine Waters, supra note 16, at 2 (arguing that, "[i]f a company
chooses to issue multiple classes of stock with differing voting rights, then the dual-class stock must contain a
'sunset' provision").
19. See infra Table 1.
546 The Journalof CorporationLaw [V ol. 48:3

shareholders and controllers, we focus on firms' ability to impose externalities.

Consequently, the limitations we propose differ from the ones traditionally advocated by
the literature. Because it is well-recognized that private ordering is bound to lead to
suboptimal outcomes from a social welfare perspective in the presence of significant
externalities, our regulatory proposal stands on much more solid ground than those
advocated by institutional investors.
We start by noting that the unfolding climate crisis and the macroeconomics literature
have shown that a specific subset of firms can impose gigantic externalities on the planet
and the economy. Allowing these companies to have dual-class shares without any
limitation implies that FVM shareholders oblivious to these externalities have an unfettered
ability to inflict systemic harm. A clear example is Buffett's Berkshire Hathaway, which
was the fourth main source of carbon dioxide (C0 2 ) emissions in the United States in
2019.2 However, empirical evidence suggests that PVM shareholders have relatively
strong incentives to mitigate these negative externalities since PVM shareholders suffer
from them to some extent via their other portfolio holdings. Thus, to prevent FVM
shareholders from having disproportionate power at these key firms, we suggest some
limits on dual-class shares for systemically relevant firms. We then explain how to tailor
such limitations according to a fin's specific ability to impose systemic externalities.
We can now provide two straightforward answers to the questions we raised at the
beginning. First, at firms that can impose systemic externalities, FVM shareholders are
likely to have preferences that are less aligned with those of society at large. Meanwhile,
at firms that cannot impose systemic externalities, PVM shareholders are likely to have
preferences that are less aligned with those of society at large.24 Second, we argue that
corporate law can limit the negative impact of such preferences by allowing most firms to
freely adopt dual-class structures while imposing some limits on systemically relevant
firms' freedom to do so.25
The rest of the Article proceeds as follows. Part II sets the scene by discussing the
evidence in support of the claim that institutional investors sometimes act as portfolio value
maximizers and introduces the two parallel ongoing debates about the anticompetitive
effects of common ownership and dual-class structures. Part III discusses the role that dual-
class companies can play in fostering competition despite the growth of common
ownership. Furthermore, it advances arguments against the introduction of a mandatory
sunset. Part IV outlines how a small subset of firms plays a disproportionately large role in

20. See infra notes 87-89 and accompanying text.

21. See Greenhouse 100 Polluters Index (2022 Report, Based on 2020 Data), POL. ECON. RSCH. INST.,
UNIV. MASS. AMHERST, https://peri.umass.edu/greenhouse-100-polluters-index-current
[https://perma.cc/AXA2-TGAU] (last visited Apr. 8, 2023) (presenting a list of the top 100 CO 2 emitters in the
United States with Berkshire Hathaway ranking 4th).
22. See infra notes 45-54 and accompanying text.
23 Luca Enriques & Alessandro Romano, Rewiring CorporateLawfor an Interconnected World, 64 ARIZ.
L. REV. 51, 58, 78-79 (2022).
24. Id. at 58-59.
25. See infra Part III (advocating for the adoption of a dual-class share structure to mitigate negative impact
of PVM and FVM shareholder preferences).
2023 ] Dual-ClassShares in the Age of Common Ownership 547

creating negative systemic externalities. For these firms, we suggest imposing limitations
on the freedom of FVM shareholders to retain control over the company through multi-
voting rights. In this way, PVM shareholders would have more chances to impose their
preferences and internalize a larger fraction of systemic externalities via their other
portfolio firms. Part V considers the possible advantages of our proposal and
counterarguments to it.

II. SETTING THE SCENE: PORTFOLIO VALUE MAXIMIZATION, COMMON OWNERSHIP,

AND DUAL-CLASS SHARES

In this Part, we introduce the theoretical background and the empirical evidence that
will support our claims. To begin, we discuss the rise of institutional investors in capital
markets and the empirical evidence that they act as portfolio value maximizers. Then, we
discuss the empirical evidence regarding the anticompetitive effects of common
ownership, and finally, we introduce the debate on dual-class shares.

A. InstitutionalInvestors as Portfolio Value Maximizers

The traditional idea is that shareholders want "to maximize the net present value of
the firm's earnings per dollar invested." 26 In other words, shareholders have been described
as "firm-value-maximizing."27 Yet, following the institutionalization of capital markets
and the reconcentration of ownership in the hands of a few institutional investors, this
description has become outdated. The largest asset managers-and in particular
BlackRock, Vanguard, and State Street-own significant stakes in an exceedingly large
number of corporations operating in various industries and countries. 28 Most importantly,
the vast majority of their assets under management are invested in passive or index funds, 9
the defining feature of which is that they do not try to beat the market but merely track an

26. Henry Hansmann, Ownership of the Firm, 4 J.L. ECON. & ORG. 267, 283 (1988).
27. See, e.g., Donald Lien & Chia-Feng (Jeffrey) Yu, Time-Inconsistent, Financial Constraints, and Cash
Flow Hedging, 35 INT'L REV. FIN. ANALYSIS 72 (2014) (describing and discussing "firm-value-maximiz[ation]").
28. In 2020, BlackRock voted for 153,000 proposals in 16,200 shareholders meetings. Of these meetings
5,940 were in Asia-Pacific (excluding Japan), 4,190 were in the United States and Canada, 2,434 in Europe and
the Middle East (excluding the United Kingdom), 2,350 in Japan, 775 in the United Kingdom, and 507 in Latin
America. See BLACKROCK, INC., 2020 INVESTMENT STEWARDSHIP ANNUAL REPORT 16-19 (Sept. 2020),
https://www.blackrock.com/corporate/literature/publication/blk-annual-stewardship-report-2020.pdf
[https://perma.cc/W9B4-Y6KY]. Vanguard voted in 12,429 shareholders meetings on 176,834 proposals of
companies operating in 27 different markets. See VANGUARD, INVESTMENT STEWARDSHIP ANNUAL REPORT 5
(2020), https://corporate.vanguard.com/content/dam/corp/advocate/investment-stewardship/pdf/policies-and-
reports/2021_investmentstewardshipannualreport.pdf [https://perma.cc/4FTM-CEG5]. Last, State Street
Advisors voted in 19,370 meetings on 176,680 proposals related to companies from 73 different countries. See
STATE STREET GLOBAL ADVISORS, STEWARDSHIP 6 (2020), https://cpb-us-
w2.wpmucdn.com/sites.udel.edu/dist/8/l 2944/files/2022/08/202I -Ceres-RT_SSGA-asset-stewardship-report-
2020.pdf [https://perma.cc/6LHE-N35D].
29. BlackRock, Vanguard, and State Street have, respectively, 81.3%, 81.1%, and 96.9% of their assets
under management invested in passive index funds. Jan Fichtner, Eelke M. Heemskerk & Javier Garcia-Bemardo,
Hidden Power of The Big Three? Passive Index Funds, Re-Concentration of Corporate Ownership, and New
Financial Risk, 19 BUS. & POL. 298, 304 (2017).
548 The Journalof CorporationLaw [Vol. 48:3

index, such as the S&P 500 or the Nasdaq Composite. 30 Thus, investors in such funds are
indifferent to the performance of any given firm in their portfolio. What they care about is
the value of their portfolio as a whole. Similarly, managers of passive funds derive their
revenues from fees, which are calculated as a percentage of the value assets under
management3 and from lending shares.32 Hence, they have no interest in the performance
of individual portfolio firms. They will prefer portfolio firms not to generate negative
externalities (and to instead generate positive externalities) affecting other fums in their
portfolios. In other words, they will act like PVM shareholders.
Many empirical studies have found evidence of institutional investors acting like
PVM shareholders. 33 To be sure, that does not imply that institutional investors never act
in line with the preferences of FVM shareholders or that they can lead firms to completely
internalize intra-portfolio externalities. The empirical evidence supports a much narrower
claim: namely that, in some instances, institutional investors might push firms to account
for a fraction of intra-portfolio externalities. 34
To begin with, a series of studies has shown that common ownership leads firms to
internalize a fraction of the positive externalities generated by innovation. It is well-
established in the economic literature that innovators can only appropriate a fraction-
more precisel , about two-thirds--of the returns on research and development (R&D)
investments.3 The obvious corollary is that there is underinvestment in innovation.36
However, the shareholders of the firm investing in R&D might have weaker incentives to
underinvest if they also own significant stakes in the competitors, suppliers, and customers
of the firm. They would, in fact, appropriate a larger fraction of the value generated by the
innovation. Recent studies support this idea 37 and further show that common ownership

30. See, e.g., Jan Fichtner & Eelke M. Heemskerk, The New Permanent Universal Owners: Index Funds,
Patient Capital, and the Distinction Between Feeble and Forceful Stewardship, 49 ECON. & Soc'Y 493, 494
(2020) (noting that "[i]ndex funds replicate established stock indexes" and "do not actively buy and sell stocks
based on expected future earnings but merely follow the market").
31. Lucian A. Bebchuk, Alma Cohen & Scott Hirst, The Agency Problems of InstitutionalInvestors, 31 J.
ECON. PERSPS. 89, 96-97 (2017) ("Mutual fund managers and investment managers of other similarly structured
funds are not permitted to collect incentive fees on increases in the value of their portfolio but may only charge
fees that are calculated as a percentage of assets under management.").
32. See Edwin Hu, Joshua Mitts & Haley Sylvester, The Index-Fund Dilemma: An EmpiricalStudy of the
Lending-Voting Tradeoff 2 (N.Y.U. L. & Econ. Rsch., Working Paper No. 20-52, 2020),
https://ssrn.com/abstract=3673531 [https://perma.cc/S23L-LKQ6] (describing how institutional investors derive
profits from lending shares).
33. See infra notes 35-54 and accompanying text. For an overview of the literature see Enriques & Romano,
supra note 23, at 67-74.
34. See infra notes 36-55 and accompanying text (analyzing different empirical studies).
35. See Nicholas Bloom, Mark Schankerman & John Van Reenen, Identifying Technology Spillovers and
Product Market Rivalry, 81 ECONOMETRICA 1347, 1384 (2013) (reporting on and analyzing empirical results
comparing private and social returns to R&D).
36. See id. at 1389 (reporting that the socially optimal level of investments in R&D is over twice as high as
the level currently observed).
37. See generally Angel L. L6pez & Xavier Vives, OverlappingOwnership, R&D Spillovers, andAntitrust
Policy, 127 J. POL. ECON. 2394 (2019) (showing under which conditions overlapping ownership can lead to higher
R&D and to higher social welfare).
2023 ] Dual-ClassShares in the Age of Common Ownership 549

38 39
facilitates the diffusion of innovation and avoids duplications of R&D investments.
A similar logic can be applied to voluntary disclosure. Like in the case of investments
in R&D, firms are not able to internalize all of the benefits associated with voluntary
disclosure. 40 Thus, common ownership by shareholders with PVM preferences should be
associated with higher disclosure, given that common owners can internalize at least a
41
fraction of these inter-firm spillovers. This is exactly what one empirical study has found.
The evidence that institutional ownership leads to the internalization of part of inter-
firm externalities is not limited to these domains and extends to many areas of corporate
43
governance 42 and business strategies.
The most important and debated implication of the idea that large diversified
institutional investors lead firms to internalize a portion of intra-portfolio externalities is
44
the hypothesis that PVM shareholders might aim to mitigate climate-related externalities.

38. Leonard Kostovetsky & Alberto Manconi, Common Institutional Ownership and Diffusion of
Innovationt 5 (April 13, 2020), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2896372
[https://perma.cc/2VMZ-ER79] (finding that common owners can be a vehicle for the diffusion of innovation).
39. See Xuelin Li, Tong Liu & Lucian A. Taylor, Common Ownership and Innovation Efficiency, 147 J.
FIN. ECON. 475, 477-78 (2023) (discussing the finding that common ownership can increase social welfare by
reducing unnecessary duplication of R&D costs).
40. See, e.g., Anat R. Admati & Paul Pfleiderer, Forcing Firms to Talk: FinancialDisclosureRegulation
and Externalities, 13 REv. FIN. STUD. 479, 480 (2000) (noting that firms only internalize a fraction of the social
value of the information they disclose).
41. See Jihwon Park et al., Disclosure Incentives When Competing Firms Have Common Ownership, 67 J.
ACCT. & ECON. 387, 389 (2019) (providing empirical evidence for the finding that "common ownership is
positively associated with (i) the likelihood . . . and (ii) the frequency of issuing both earnings and capex
forecasts").
42. See, e.g., Jie (Jack) He, Jiekun Huang & Shan Zhao, Internalizing Governance Externalities:The Role
of InstitutionalCross-Ownership, 134 J. FIN. ECON. 400, 401 (2019) (finding "that institutional shareholders with
larger ownership stakes in peer firms (i.e., same-industry firms with similar size) are more likely to vote against
management on shareholder-sponsored governance proposals").
43. See Ruichang Lu et al., Frenemies: CorporateAdvertising Under Common Ownership, 68 MGMT. SCI.
4645, 4645 (2022) (finding that "competing firms owned by the same institutional blockholders experience a
significant reduction in advertising expenditure"); John Healey & Ofer Mintz, What If Your Owners Also Own
Other Firms in Your Industry? The Relationship Between Institutional Common Ownership, Marketing, andFirm
Performance, 38 INT'L J. RSCH. MKTG. 838, 840 (2021) (finding that an increase in institutional common
ownership is associated with an increase in firm performance, especially for firms with lower marketing
capabilities); Jose Azar, Yue Qiu & Aaron Sojourner, Common Ownership in Labor Markets 2 (W.E. Upjohn
Inst. for Emp. Rsch., Working Paper No. 22-368, 2022),
https://research.upjohn.org/cgi/viewcontent.cgi?article=1387&context=upworkingpapers
[https://perma.cc/5X82-GQ89] (finding that increase in common ownership at the local level is associated with a
decrease in annual wages per employee); Xin Dai & Yue Qiu, Common Ownership and Corporate Social
Responsibility, 10 REV. CORP. FIN. STUD. 551, 552 (2021) (finding evidence that common ownership affects
firms' corporate social responsibility practices).
44. See generally Madison Condon, Externalities and the Common Owner, 95 WASH. L. REV. 1 (2020)
(discussing extensively the incentives of diversified investors to internalize climate externalities); see also Suren
Gomtsian, Different Visions of Stewardship: Understanding Interactions Between Large Investment Managers
and Activist Shareholders,22 J. CORP. L. STUD. 151, 162 (2021) (claiming that diversified investors do recognize
the crucial role of systematic risks and reporting Blackrock CEO Larry Fink's concerns about climate change risk
for investors).
550 The Journalof CorporationLaw [V ol. 48:3

Azar and his co-authors found evidence supporting this hypothesis.4 5 First, they found that
the Big Three are more likely to engage with portfolio companies that record higher
emissions.46 Second, they found a strong negative and robust association between holdings
by the Big Three and the C02 emissions of the firm, and that this effect is prevalent for
firms in which the Big Three hold larger stakes and are thus likely to be more influential. 47
Third, they found that the impact of the Big Three's ownership has become stronger in
recent years as these institutions have increased their commitment to environmental, social,
and governance (ESG)-conscious practices. 4 8 Finally, they found that inclusion in the
Russell 1000 and Russell 2000 indexes, which inevitably implies an increase in Big Three
ownership, results in a subsequent reduction in C02 emissions. 49
Further support for this hypothesis comes from a study by Dyck and his co-authors,
who found that institutional ownership is associated with firms' better environmental and
social performance. 50 The effect is especially strong with respect to environmental
performance, as they found that an increase in one standard deviation in institutional
ownership is associated with better environmental scores, 5 1 an effect that is much larger
for institutional investors that have signed the United Nations Principles for Responsible
Investment. Most importantly, they observed that this effect is not driven by the fact that
institutional investors buy into companies with better environmental and social
performance. 53 Instead, institutional investors improve the performances of companies
they already own. 54

B. Common Ownership: Evidence and Policy Proposals

In virtually all oligopolistic markets, large institutional investors have a significant

stake in the main competitors. According to some scholars, this ownership structure-

45. See generally Jos6 Azar et al., The Big Three and CorporateCarbon Emissions Around the World, 142
J. FIN. ECON. 674, 674-76 (2021) (studying the role that BlackRock, Vanguard, and State Street Global Advisors
has on the reduction of carbon emssions around the world).
46. Id. at 675.
47. Id.
48. Id. at 676.
49. Id.; see also Dharmapala & Khanna, supra note 1, at 14 (finding that environmental engagements by
institutional investors are 640% more common in countries where ownership structure is generally dispersed than
in countries with a significant proportion of blockholders).
50. Alexander Dyck et al., Do InstitutionalInvestors Drive CorporateSocial Responsibility? International
Evidence, 131 J. FIN. ECON. 693, 694, 713 (2019).
51. Id.
52. The UN Global Compact and UNEP Financial Initiative's collaborative Principles for Responsible
Investments represents an international network of investors whose aim is to promote the incorporation of
environmental, social, and corporate governance values into investment decision-making and ownership practice.
Their goal is to encourage a more sustainable global financial system. See generally What Are the Principlesfor
Responsible Investment?, PRINCIPLES FOR RESPONSIBLE INV., https://www.unpri.org/pri/what-are-the-principles-
for-responsible-investment (last visited Apr. 9, 2023) [https://perma.cc/732Y-CWLG] (describing the PRI
association's goals).
53. See Dyck et al., supra note 50, at 713.
54. Id. ("Firms with greater institutional ownership pushed harder for improved E&S performance after
reconizing the value of E&S during the crisis period.").
2023 ] Dual-ClassShares in the Age of Common Ownership 551

generally labeled common ownership-would lead to a lower level of competition in

product markets.5 5 In a nutshell, the idea is that common owners prefer to maximize the
joint value of the portfolio companies operating in the market instead of maximizing the
value of any given portfolio fu-m.56 For this reason, they might prefer a lower level of
competition in product markets so that all of their portfolio companies can enjoy extra-
competitive profits. Against this background, the managers of such corporations would
have no incentives to push their firms to compete aggressively, as this would be against the
57
preferences of their own shareholders.
Building on this line of reasoning, leading legal scholars have advanced several policy
proposals that would limit the extent to which common ownership would be permissible.
In particular, Posner, Scott Morton, and Wyel have argued that investors should not be
allowed to own shares in more than one horizontal competitor in each oligopolistic
market, 58 while Elhauge suggested that high levels of common ownership should be quasi
per se illegal. 59
These proposals, if adopted, would force institutional investors to completely
60
revolutionize their business models and would re-design financial markets. It is less clear
whether they would increase the level of competition in the markets.61 Changes of such
magnitude, however, seem hardly warranted given the nuanced picture painted by the

55. See, e.g., Einer Elhauge, HorizontalShareholding, 129 HARv. L. REv. 1267, 1268 (2016) ("Economic
theory has long shown that horizontal shareholdings can reduce the incentives of horizontal competitors to
compete with each other."); Eric A. Posner, Fiona M. Scott Morton & E. Glen Weyl, A Proposal to Limit the
Anti-Competitive Power of Institutional Investors, 81 ANTITRUST L.J. 669, 669-70 (2017) (defining the
anticompetitive effects of common ownership as "the major new antitrust challenge of our time").
56. See Elhauge, supra note 55, at 1269 (explaining how common ownership creates an anticompetitive
effect that "reduces the incentives of each firm to price products lower").
57. See id. at 1307 (discussing that managers remember that "horizontal shareholders have an interest in
the profits of rival firms.").
58. See Posner, Scott Morton & Weyl, supra note 555, at 708-12 (summarizing their policy proposal by
stating that "[n]o 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" (emphasis omitted)). A similar proposal has also been advanced
by scholars concerned with the effects of common ownership on labor market outcomes. See Zohar Goshen
&

Doron Levit, Agents of Inequality: Common Ownership and the Decline of the American Worker, 72 DUKE L.J.
1, 58 (2022) (describing how the rise of common owners contributed to a shift of wealth from labor to capital and
to an increase of income inequality, and suggesting that "to solve the problems caused by common ownership,
the answer is to break up common owners"); Fiona Scott Morton & Herbert Hovenkamp, HorizontalShareholding
and Antitrust Policy, 127 YALE L.J. 2026, 2033-47 (2018) (discussing the tools available to policymakers to
mitigate the anticompetitive effects of common ownership).
59. See Elhauge, supra note 55, at 1314 (arguing that institutional investors would have only two ways to
avoid liability: "refraining from horizontal investments or committing not to vote their stock").
60. Edward B. Rock & Daniel L. Rubinfeld, Defusing the Antitrust Threat to Institutional Investor
Involvement in Corporate Governance 2 (N.Y.U., L. & Econ. Rsch., Paper No. 17-05, 2017),
https://papers.ssm.com/sol3/papers.cfm?abstract-id=2925855 [https://perma.cc/V9TX-QWPY] (noting that
Posner et al.'s 2017 proposal would destroy institutional investors' current business models).
61. Alessandro Romano, Horizontal Shareholdingand Network Theory, 38 YALE J. REG. 363, 396-400
(2021) (explaining why the sweeping proposals advanced by legal scholars are unlikely to increase the level of
competition in product markets).
552 The Journalof CorporationLaw [Vol. 48:3

empirical studies on this issue. While there is relatively compelling evidence that common
ownership can affect the level of competition in product markets, this evidence merely
refers to the following handful of selected markets: passenger air transportation;62 soy,
corn, and cotton seed; 3 retail banking;64 and the pharmaceutical industry. 6 5
Admittedly, the fact that the evidence refers only to some markets does not necessarily
mean that common ownership has a negative effect on the level of competition only in
these markets. In fact, carrying out studies to isolate the impact of common ownership on
product prices is very complex, and the necessary data is not always available. Even
accounting for this caveat and taking the results of these studies at face value,66 the
empirical evidence hardly provides sufficient grounds for invasive reforms. 67 In fact, there
are strong theoretical reasons to believe that common ownership is unlikely to have the
same effect on all markets.

62. See Jose Azar, Martin C. Schmalz & Isabel Tecu, Anticompetitive Effects of Common Ownership, 73 J.
FINANCE 1513, 1517 (2018) (finding that higher values of common ownership in the airline industry are
associated with prices at the route level that are three to seven percent higher). Their seminal paper spurred an
intense debate. Two empirical papers questioned their results. See Patrick J. Dennis, Kristopher Gerardi & Carola
Schenone, Common Ownership Does Not Have Anti-Competitive Effects in the Airline Industry, 77 J. FtNANCE
2765, 2766 (2022) (finding evidence to "suggest[] that the positive correlation between the measure of common
ownership concentration and airline ticket fares . .. does not reflect a causal relationship"); Pauline Kennedy et
al., The Competitive Effects of Common Ownership: Economic Foundations and Empirical Evidence 4 (July 26,
2017) (unpublished paper), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3008331
[https://perma.cc/8QKA-55Q2] ("In contrast to [Azar, Schmalz and Tecu], we find no evidence in our price
regressions and structural model estimation that common ownership raises prices."). The authors of the original
studies reacted to these criticisms by defending their findings. See Jose Azar, Martin C. Schmalz & Isabel Tecu,
A Refutation of 'Common Ownership Does Not Have Anti-Competitive Effects in the Airline Industry' 1 (Eur.
Corp. Governance. Inst., Finance Working Paper No. 837/2022, 2022).
63. See Mohammad Torshizi & Jennifer Clapp, PriceEffects of Common Ownership in the Seed Sector, 66
ANTITRUST BULL. 39, 41 (2021) (finding that "approximately 6.2%-14.6% of maize, soybean, and cotton seed
price increases over the 1997-2017 period are attributable to common ownership").
64. Jose Azar, Sahil Raina & Martin C. Schmalz, Ultimate Ownership and Bank Competition, 51 FIN.
MGMT. 227, 266 (2022) (finding that indicators accounting for ownership structure, and in particular of common
ownership, better predict market outcomes like interest rates, maintenance fees, and fee thresholds).
65. See Jin Xie & Joseph Gerakos, The Anticompetitive Effects of Common Ownership: The Case of
Paragraph IV Generic Entry, 110 AM. ECON. ASS'N PAPERS & PROC. 569, 572 (2020) (examining how common
ownership affects the product-market outcome in the pharmaceutical sector by analyzing a sample of 666 patent
lawsuits).
66. For empirical studies challenging the idea that common ownership has a negative effect on the level of
competition, see Jacob Gramlich & Serafin Grundl, Estimating The Competitive Effects Of Common Ownership
1-3 (Fin. & Econ. Discussion Series, Working Paper No. 2017-029, 2017),
https://papers.ssm.com/sol3/papers.cfm?abstractid=2940137 [https://perma.cc/63MB-PP5A] (proposing an
alternative methodology to measure common ownership and finding that the impact of common ownership on
price and quantities depends on the specification of the model and is "quite small"); see also Kennedy et al., supra
note 622, at 4 ("[W]e find no evidence in our price regressions and structural model estimation that common
ownership raises prices."); Dennis, Gerardi & Schenone, supra note 62, at 2796 (claiming that additional findings
"casts significant doubt on the widespread interpreation ... that common ownership leads to anticompetitive
effects in the airline industry").
67. See infra notes 68-70 and accompanying text.
2023 ] Dual-ClassShares in the Age of Common Ownership 553

For instance, the literature on network theory has identified a small subset of sectors
that impose significant externalities on a wide range of industries. In these markets, a
diversified common owner might prefer a higher level of competition than a shareholder
who only owns stakes in a firm in such market.68 The reason is that a lower level of
competition would imply higher prices and a lower level of both input and output.69 All of
these factors would have a negative effect on the firms that operate in connected markets,
and such firms are likely to be in the portfolio of a diversified common owner. On the
contrary, an undiversified shareholder would reap the benefits of a lower level of
70
competition in the market without having to internalize any of the negative externalities.
Consider, for instance, the case of the banking sector. A lower level of competition would
entail higher interest rates, which might harm virtually every firm in the economy. An
investor who owns shares only in banks might derive a profit because a lower level of
competition is likely to result in extra-competitive profits for the banks. However, a
diversified investor who owns shares in companies that are negatively affected by the
higher interest rates ceterisparibuswould find a lower level of competition less desirable.
In other words, the effects of common ownership are more nuanced than imagined by
the legal scholars advocating for sweeping reforms, and the exact contours of the problem
are hard to identify given the complexity of the issue and the limited availability of data.
Therefore, either structural reforms must be much more fine-grained and account for
factors such as inter-market effects and the characteristics of the common owners or
policymakers should rely on different approaches.
In this Article, we argue that one such approach would be a laissez-faire attitude
toward the use of dual-class shares. Therefore, we now turn to discuss the literature on such
share structures.

C. Dual-ClassShares

Dual-class shares allow companies to deviate from the standard one-share-one-vote

principle. Generally, companies that adopt dual-class shares issue two classes of shares.
The first class comes with high voting power-typically ten votes per share-and remains
in the hands of founders and key insiders, whereas the second class of shares has lower
voting power-typically one vote per share-and is sold to outside investors. 71 Several
72
companies also have a third class of non-voting shares. Thus, dual-class shares allow

68. See Romano, supra note 611, at 405-07 (challenging the notion that high levels of horizontal
shareholding in a given market are associated with anticompetitive effects).
69. Rock & Rubinfeld, supra note 61, at 24.
70. See Romano, supra note 611, at 407 (noting that firms have a natural tendency to prefer a lower level
of competition, but that only diversified investors internalize a fraction of the externalities that a lower level of
competition imposes on connected firms).
71. According to the Council of Institutional Investors, "[t]ypically, these companies have two classes of
common stock: Class A shares with 10 votes per share for the founders (and sometimes insiders, too) and Class
B shares with one vote per share for public shareholders." Dual-Class Stock, COUNCIL OF INSTITUTIONAL INVS.,
https://www.cii.org/dualclassstock [perma.cc/5LDK-ME8A] (last visited Apr. 9, 2023).
72. For instance, Alphabet, Liberty Broadband, Under Armour, and AppLovin Corporation have three
5 54 The Journalof CorporationLaw [V ol. 48:3

founders and key insiders to control the corporation even when they hold a small fraction
of the shares. By decoupling voting and cash flow rights, dual-class firms can raise equity
capital without being subject to the constraints that the founders' preference for preserving
control over the firm would impose.7 3 Thus, dual-class firms can grow to a size that usually
only firms with dispersed ownership can attain.
Dual-class structures have been present in the U.S. market since the 19th century, but
the number and the importance of corporations listed in the United States with a dual-class
structure have boomed since Google's IPO in 2004.74 Some of the companies with the
highest capitalization of our time have dual-class shares, including Alphabet Inc. (formerly
75
Google), Meta (formerly Facebook Inc.), Visa, and Berkshire Hathaway. Data from 2016
shows that companies adopting dual-class shares have a market capitalization exceeding
$3 trillion,76 or approximately 10% of the U.S. public equity markets.77 In 2020, 15% of
the companies that went public had dual-class shares.7 8 Moreover, while one-share-one-
vote companies still dominate in number, in 2020, dual-class companies represented 60%
of the IPO market value. 79
The most common benign explanation for the adoption of dual-class shares is that
they isolate the founders from the short-termism of financial markets and the pressures
from the market for corporate control. 80 In this vein, by adopting such shares, founders are

classes of shares, one of which has no voting power. See Dual Class Companies List, COUNCIL OF INSTITUTIONAL
INVS. 1, https://www.cii.org//Files/issuesand_advocacy/Dual%20Class%20post%206-25-
19/Dual%20Class%20Companies%20Webpage%2010-12-21.pdf [perma.cc/89FE-UHTH] (last visited Mar. 13,
2023).
73. See Luca Enriques, Silence Is Golden: The European Company as a Catalyst for Company Law
Arbitrage, 4 J. CORP. L. STUD. 77, 90-91 (2004) (explaining why firms deviate from the one-share-one-vote
principle and noting that, without the possibility to raise equity by issuing non-voting or lower-voting stock,
dominant shareholders would have a strong incentive to keep the company small).
74. See Jay Ritter, Initial Public Offerings: Dual Class Structure of IPOs Through 2022, UNIV. OF FLA.
WARRINGTON COLL. OF BUS. 1 (Mar. 8, 2023), https://site.warrington.ufl.edu/ritter/files/IPOs-Dual-Class.pdf
[https://perma.cc/3DSM-JGYU] (listing the number of IPOs each year that have opted for dual-class shares from
1980 to 2021).
75. COUNCIL OF INSTITUTIONAL INVS., supranote 712.
76. Lucian A. Bebchuk & Kobi Kastiel, The Untenable Case for PerpetualDual-ClassStock, 103 VA. L.
REV. 585, 594 (2017).
77. At that time, the market capitalization of listed domestic companies in the United States was $27.35
trillion. Market Capitalization of Listed Domestic Companies (Current US$), WORLD BANK,
https://data.worldbank.org/indicator/CM.MKT.LCAP.CD?namedesc=false&locations=US [perma.cc/95TK-
7XZW].
78. Dual-Class IPO Snapshot: 2017-2020 Statistics, COUNCIL OF INSTITUTIONAL INVS. 1,
https://www.cii.org/files/2020%20IP0%20Update%20Graphs%20.pdf [perma.cc/U4RA-UGUA] (note,
however, that this percentage excludes FPIs, SPACs, and REITs).
79. Id. at 2.
80. See, e.g., David J. Berger, Steven Davidoff Solomon & Aaron J. Benjamin, Tenure Voting and the U.S.
Public Company, 72 BUs. LAW. 295, 296 (2017) (noting that dual-class "companies can consider shareholder
demands but also avoid actions that would result in short-term increases in their stock prices at the expense of
their abilities to create long-term value"). For empirical evidence supporting this claim, see Bradford D. Jordan,
Soohyung Kim & Mark H. Liu, Growth Opportunities, Short-Term Market Pressure, and Dual-Class Share
2023 ] Dual-ClassShares in the Age of Common Ownership 555

able to pursue their idiosyncratic vision81 and focus on long-term goals.82 For instance,
investors might believe that Kevin Plank, who controls Under Armour thanks to multiple
voting shares,8 3 is in the best position to run the company without the distractions of daily
share price fluctuations or hostile takeover threats. Relatedly, dual-class shares "may
84
increase the willingness of founders to take their companies public," as founders can
retain control even after raising substantial equity on public markets.
A recent article by Dorothy Lund advanced a new rationale for allowing firms to adopt
dual-class shares. 85 In particular, she argued that dual-class shares might even reduce
agency costs because nonvoting or low-voting stock allows companies to concentrate
voting power among shareholders who are best informed about the company and its
performance.86 Her basic argument was that shares carrying voting rights are more
expensive than shares that do not. Hence, investors that do not plan to spend resources to
cast informed votes-including, Lund argued, the largest asset managers-would prefer to
87
buy shares without voting power. At the same time, she argued this would improve the
incentives of informed investors because their voting power would be enhanced by the fact
that a significant fraction of the shares would no longer have voting power.88
Legal scholars and market actors have traditionally had a much less positive view of
dual-class shares. 89 The standard argument against them is that dual-class shares ensure
controllers' entrenchment while at the same time increasing agency costs between them
and outside investors. On the one hand, dual-class shares operate as a strong form of

Structure, 41 J. CORP. FIN. 304, 306 (2016) (finding evidence supporting the idea "that dual-class shares help
insiders implement long-term projects while avoiding short-term market pressure"). See also Snap Inc.,
Amendment No. 2 (Form S-1 Registration Statement) 167 (Feb. 16, 2017) (arguing that the company structure
permits the corporation to prioritize long-term goals rather than short-term results). For a complete analysis of
the various justifications for dual-class share structures, see BOBBY REDDY, FOUNDERS WITHOUT LIMITS: DUAL-
CLASS STOCK AND THE PREMIUM TIER OF THE LONDON STOCK EXCHANGE 212-38 (2021).
81. See, e.g., Zohar Goshen & Assaf Hamdani, Corporate Control and Idiosyncratic Vision, 125 YALE L.J.
560, 590 (2016) (noting that dual-class structures allow entrepreneurs to pursue their idiosyncratic visions).
82. See generally Bernard S. Sharfman, A Private Ordering Defense of a Company's Right to Use Dual
Class Share Structures in IPOs, 63 VILL. L. REV. 1 (2018) (discussing the long-term benefits that shareholders
reap when dual-class share structures are implemented).
83. The current voting structure allows Plank to control 67% of voting powers while owning only 16.8%
of total stocks. Andy Polk, Under Armour Establishes New Non-Voting Class C Common Stock, FOOTWEAR
DISTRIBS. & RETAILERS OF AM. (June 17, 2021), https://fdra.org/latest-news/under-armour-establishes-new-non-
voting-class-c-common-stock/ [perma.cc/KX7E-65UC].
84. Jill Fisch & Steven Davidoff Solomon, The Problem of Sunsets, 99 B.U. L. REV. 1057, 1061 (2019).
85. Dorothy Shapiro Lund, Nonvoting Shares and Efficient Corporate Governance, 71 STAN. L. REV. 687
(2019).
86. Id. at 716.
87. Id. at 719-23.
88. Id.
89. For instance, the Investor Stewardship Group, among its Corporate Governance Principles for U.S.-
listed companies, includes the idea that "[c]ompanies should adopt a one-share, one-vote standard and avoid
adopting share structures that create unequal voting rights among their shareholders." See INV. STEWARDSHIP
GRP., supra note 17, at 2. Among scholars, see, for example, Lund, supra note 85, at 724 (arguing that dual-class
capitalization is inefficient, and therefore that it should be banned for companies that are subject to section 12 of
the Securities Exchange Act of 1934).
556 The Journalof CorporationLaw [Vol. 48:3

takeover defense. 90 Even when the insider owns less than a majority of the shares, they
make it impossible for outsiders to take over the corporation without the insider's consent.
On the other hand, dual-class shares break down the relationship between cash flow rights
and voting rights. 9 1 As a consequence, insiders can extract higher private benefits than they
would in a one-share-one-vote company when the corporation adopts strategies that cater
to insiders' interests to the detriment of the company's profitability. 92
Policymakers have also often expressed concerns about dual-class shares. For
instance, during their term as SEC Commissioners, Kara Stein and Robert Jackson, Jr.
described dual-class shares as "inherently undemocratic, disconnecting the interests of a
company's controlling shareholders from its other shareholders," 93 and argued that they
create "corporate royalty." 94 Ultimately, the empirical evidence on the impact of dual-class
95
shares on firm value has been mixed thus far.
In a recent article, despite recognizing the possible advantages of dual-class shares,
Bebchuk and Kastiel claimed that dual-class shares should have a mandatory time-based
sunset. 96 While acknowledging that insulating innovative founders for a certain period of
time might increase firm value, they argued that founders' advantages are bound to
decrease over time due to technological evolution and changes in markets. 97 For example,
Mark Zuckerberg might have been the best choice for Facebook for much of its history,
but he may no longer be today, and even less so tomorrow. In the same vein, the Council
of Institutional Investors sent a petition to the New York Stock Exchange and Nasdaq
asking them not to list firms that have dual-class shares unless they have a mandatory time-
based sunset of at most seven years. 98 Former Commissioner Jackson also vehemently

90. See Paul A. Gompers, Joy Ishii & Andrew Metrick, Extreme Governance: An Analysis of Dual-Class
Firms in the United States, 23 REV. FIN. STUD. 1051, 1052 (2010) (defining dual-class stocks as the most extreme
example of antitakeover provision).
91. See, e.g., K.J. Martijn Cremers, Beni Lauterbach & Anete Pajuste, The Life-Cycle of Dual Class Firm
Valuation 53 (Eur. Corp. Governance Inst., Finance Working Paper No. 550/2018, 2020) (discussing the role
played by the difference between the voting and equity stakes of the controlling shareholders created by dual-
class structures).
92. See Lucian A. Bebchuk, Reinier Kraakman & George Triantis, Stock Pyramids, Cross-Ownership, and
Dual Class Equity: The Mechanisms and Agency Costs of Separating Control from Cash-Flow Rights, in
CONCENTRATED CORPORATE OWNERSHIP 295, 301-05 (Randall K. Morck ed., 2000) (discussing this issue by
focusing on the agency costs created by dual-class structures).
93. Kara M. Stein, Sec. & Exch. Comm'r, "Mutualism: Reimagining the Role of Shareholders in Modern
Corporate Governance" Remarks at Stanford University (Feb. 13, 2018),
https://www.sec.gov/news/speech/speech-stein-021318 [https://perma.cc/DEX4-ECQL].
94. Robert J. Jackson, Jr., Sec. & Exch. Comm'r, Perpetual Dual-Class Stock: The Case Against Corporate
Royalty (Feb. 15, 2018), https://www.sec.gov/news/speech/perpetual-dual-class-stock-case-against-corporate-
royalty [https://perma.cc/5PT5-CSX7].
95. See Anita Anand, Governance Complexities in Firms with Dual Class Shares, 3 ANNALS CORP.
GOVERNANCE 184, 203-07 (2018) (discussing the empirical literature on the effects of dual-class shares and
concluding "that for virtually every study noting a problem with DCS firms, there is a study either finding a
benefit or a neutral effect of DCS on firm value").
96. Bebchuk & Kastiel, supra note 76, at 585-86.
97. Id.
98. John C. Coffee, Jr., Dual Class Stock: The Shades of Sunset, CLS BLUE SKY BLOG (Nov. 19, 2018),
2023 ] Dual-Class Shares in the Age of Common Ownership 557

attacked perpetual dual-class shares, decrying them as inconsistent with American

foundational ideals. 99 Recent empirical evidence has provided some support for the idea
that dual-class shares might add value when the company goes public but has also shown
that this effect then dissipates over time. 100
As demonstrated by this quick overview of the debate surrounding dual-class shares,
policymakers and scholars have almost exclusively focused on the consequences of this
arrangement in terms of agency costs and the ability of the firm to pursue long-term goals.
The debate has therefore focused on the intra-firm consequences of dual-class shares.
Private ordering can significantly mitigate concerns based on intra-firm considerations,
given that investors can decide not to buy, or to pay less for, shares of companies with dual-
class structures. Instead, we focus on inter-firm consequences of dual-class shares. On the
one hand, in Part III, we show that this new focus allows us to uncover a previously
overlooked advantage of dual-class shares, namely that such structures can preserve
product market competition despite the growth of common ownership. On the other hand,
in Part IV, we argue that limitations on dual-class share structures should be tied to firms'
ability to impose systemic externalities and hence to the inter-firm consequences of dual-
class structures.

III. DUAL-CLASS SHARES AS A PRO-COMPETITIVE DEVICE

This Part discusses the role that dual-class shares can play in mitigating the
anticompetitive concerns raised by common ownership and, building on that, sides with
the view that mandatory time-based sunsets are undesirable.

A. Dual-Class Shares and Competition

The traditional discussion on dual-class shares is rooted in a world of FVM

shareholders, in which the main concern is ensuring that holders of shares with high voting
power do not steer firms away from firm value maximization by extracting private benefits
of control. In a world of PVM shareholders, however, dual-class shares also serve a distinct
function. They create a control structure that increases the likelihood that the firm will

https://clsbluesky.law.columbia.edu/2018/11/19/dual-class-stock-the-shades-of-sunset/ [perma.cc/B4TP-
DVGA]. Recently, the Council of Institutional Investors has reiterated this request, presenting a draft of federal
legislation to the House Financial Services Committee. The CII has indicated the need to amend the Exchange
Act by introducing a seven-year or less "sunset" on listed issuers having multi-class structures. See Mahoney,
supra note 16, at 3 n.12 and accompanying text.
99. See Jackson, supra note 94 ("I cannot see how to square that with our nation's foundational ideas. In
America, we don't inherit power, and we don't hold power forever." (footnote omitted)).
100. Cremers, Lauterbach & Pajuste, supra note 911, at 1 ("As firms age, the valuation premium of dual-
class firms tends to dissipate, possibly because dual -class agency problems increase due to a gradual widening
of the wedge (the difference between insider voting and -cash flow rights) in the post-IPO years."); Hyunseob
Kim & Roni Michaely, Sticking Around Too Long? Dynamics of the Benefits ofDual-Class Voting Dynamics of
the Benefits of Dual-Class Voting 32 (Eur. Corp. Goverance Inst., Finance Working Paper No. 590/2019, 2019),
https://papers.ssm.com/sol3/papers.cfm?abstract_id=3145209 [https://perma.cc/9KFS-7JCA] (finding evidence
that as the dual-class company matures, its valuation and efficiency in innovation and labor productivity declines
compared to its single-class counterparts).
558 The Journalof CorporationLaw [Vol. 48:3

pursue FVM rather than PVM shareholders' interests. By issuing dual-class shares, the
founders signal to the market that the company can be more easily run as an FVM company
instead of factoring in the potentially conflicting interests of PVM shareholders. Founders
holding dual-class shares generally have a significant financial stake in their company
relative to their overall wealth. 10 1 Therefore, they have a clear motivation to increase firm
value, regardless of the consequences that this might have on other firms. Admittedly, dual-
class shares intensify founders' inclination to extract private benefits of control,102 but this
can be seen as the price that investors pay to invest in founders' idiosyncratic vision without
the interference of PVM shareholders.
This perspective is consistent with the behavior of institutional investors. In fact, it
can explain why, while lobbying to ban such structures, institutions keep buying stakes in
companies with dual-class shares. 103 Let us assume that an economy has a population of
four firms: A, B, C, and D. Assume also that A and B compete in market X, whereas C and
D compete in market Y. Lastly, assume that by investing heavily in R&D, Firm D could
develop innovations that would allow it to take over the entire market Y and also start
competing in market X. One pertinent example here is the improvements that Google made
to its search algorithm that led it to dominate the market for general Internet search services,
1 04
a position which it later leveraged to expand into comparison shopping services.
Assume now that PVM institutional investors dominate all four firms. Their optimal
strategy might be to persuade Firm D to drop its plans to compete aggressively and enter
new markets. If they succeed, all their portfolio companies will enjoy extra-competitive
profits.
What if Firm D has dual-class shares and its management can therefore expand into
new markets without worrying about whether institutional investors support its strategy?
In this scenario, institutional investors can no longer achieve their first-best outcome,
namely a lower level of competition within and across markets. Their decision is now
whether to invest in D taking as a given that D's controlling shareholders will act as firm
value maximizers and try to gain value to the detriment of A, B, and C. If they do buy D
shares, they will suffer the losses that D imposes on their portfolio companies, but at least

101. See Dharmapala & Khanna, supra note 1, at 38 (reporting on the fraction of the wealth that the
founders/controllers of some leading corporations have invested in their company).
102. See, e.g., Dov Solomon, Rimona Palas & Amos Baranes, The Quality ofInformationProvidedby Dual-
Class Firms, 57 AM. BUS. L.J. 443, 446 (2020) (claiming that a dual-class share structure allows insiders to extract
private benefits at the expense of other investors).
103. See, e.g., Paresh Dave, Big Investor T. Rowe Price ChallengesSnapchat Founders' Power, L.A. TIMES
(Jan. 19, 2017), https://www.latimes.com/business/technology/la-fi-tn-snapchat-voting-20170119-story.html
[perma.cc/7PNT-U5RZ] (reporting that T. Rowe Price "persistently" advocated against dual-class shares at Snap,
and yet, it is currently one of the largest shareholders of Snap Inc.); see Matthew Johnston & Gordon Scott, Top
Snap Shareholders, INVESTOPEDIA (Jan. 31, 2023), https://www.investopedia.com/news/top-3-snap-inc-
shareholders-snap/#:-:text=The%20top%20shareholders%200f%20Snap,.%2C%20and%20Edgewood%20
Management%20LLC [https://perma.cc/WCP5-ZTBX] ("The top shareholders of Snap are Robert Murphy, Evan
Spiegel, Jeremi Gorman, T. Rowe Price Associates Inc., Vanguard Group Inc., and Edgewood Management
LLC.").
104. Jillian D'Onfro, Google Expands Its Express Delivery Service, Further ChallengingAmazon, Bus.
INSIDER (Oct. 14, 2014), https://www.businessinsider.com/google-expands-shopping-express-2014-10
[https://perma.cc/K4AH-4JRH].
2023 ] Dual-Class Shares in the Age of Common Ownership 559

the value of their investment in D will grow. On the contrary, if they only invest in A, B,
and C, then they will still suffer losses stemming from D's actions, but without reaping any
benefits. Most importantly, the aggressive strategy of D forces Firms A, B, and C to
compete as well.
In this scenario, the preferences of PVM institutional investors are clear: the first-best
strategy is to advocate for the elimination of dual-class shares. With no dual-class
companies around, institutional investors can ensure that all firms behave like portfolio
value maximizers. However, if they cannot reach the desired policy outcome, their second-
best option is to buy stakes in firms controlled by FVM shareholders. This is exactly what
we observe, as institutional investors routinely hold significant positions in companies with
105
dual-class shares, including through the active funds they manage.
This framework also helps explain why dual-class share structures are especially
common in the tech sector. 106 Firms operating in this sector are intrinsically disruptive, as
07
their activity can destroy the business model of companies across many markets.1
Consequently, founders have stronger reasons to silence PVM shareholders that could
constrain their growth when it threatens to disrupt the activity of many of their portfolio
firms.
A perfect example of this dynamic is offered by Alphabet's incursions into the
pharmaceutical market. 108 Alphabet's recently-created Isomorphic Lab plans to leverage
the knowledge in artificial intelligence developed by Deep Mind, another company in
Alphabet's portfolio, to "reimagine the entire drug discovery process."109 In the past,
Alphabet had already launched Verily and Calico, which operate in key areas at the frontier
of health-related research, such as surgical robots and overcoming aging.110 Incidentally,

105. For an example, see sources cited supra note 103.

106. See Dhruv Aggarwal et al., The Rise of Dual-Class Stock IPOs 16 (Nat'l Bureau of Econ. Rsch.,
Working Paper No. 28609, 2021), https://papers.ssm.com/sol3/papers.cfm?abstractid=3814596
[https://perma.cc/F2R9-59P8] ("[M]ost of the historical increase in dual-class firms stems from [industries, such
as software, that heavily rely on Cloud computing technology for data management], where the percentage of
dual-class IPOs in 2017-2019 exceeds 50 percent."). Dual-class shares are so common in the tech sector that
Twitter's decision not to adopt such a structure surprised many observers. See Steven Davidoff Solomon, In
Twitter's I.P.O. Filing, Signs of a Start-Up That Has Matured, N.Y. TIMES: DEALBOOK (Oct. 8, 2013),
https://dealbook.nytimes.com/2013/10/08/in-twitters-i-p-o-filing-signs-of-a-start-up-that-has-matured/
[https://perma.cc/8BNQ-NXLC] (arguing that, by not adopting a dual-class share structure, Twitter surprised
many commentators).
107. See, e.g., Patrick Barwise & Leo Watkins, The Evolution of Digital Dominance, in DIGITAL
DOMINANCE: THE POWER OF GOOGLE, AMAZON, FACEBOOK, AND APPLE 21 (Martin Moore & Damian Tambini
eds., 2018) (describing Google's (now Alphabet) competitive strategy and how it constantly added new services
and lines of business to its portfolio). For a tech company without dual-class shares acting as a disruptor in
multiple markets, see Lina M. Khan, Amazon's Antitrust Paradox, 126 YALE L.J. 710, 746-55 (2017) (describing
Amazon's business strategy, and, in particular, how it aggressively expanded to multiple markets).
108. See Cade Metz, Google's Parent Launches a Company Dedicated to Drug Discovery, N.Y. TIMES (Nov.
4, 2021), https://www.nytimes.com/2021/11/04/business/google-alphabet-drug-discovery.html
[perma.cc/2D9W-C6J4] (reporting on Alphabet's entry into the drug discovery market).
109. Demis Hassabis, Introducing Isomorphic Labs, CTR. FOR BRAIN, MINDS, & MACHS. (Nov. 18, 2021),
https://cbmm.mit.edu/news-events/news/introducing-isomorphic-labs [https://perma.cc/6LNS-VQWJ].
1 10. Metz, supra note 108.
560 The Journal of CorporationLaw [V ol. 48:3

the pharmaceutical industry is one of several markets in which empirical evidence has
shown that common ownership might be having an impact on competition dynamics. For
instance, Xie and Gerakos found that common ownership by institutional investors
increases the probability that a brand-name manufacturer and a generic manufacturer reach
a settlement in which the former pays the latter to delay entry into the market.1 "I
The existing evidence is consistent with the idea that dual-class companies can limit
the anticompetitive effects of common ownership. As Table 1 shows, the main studies
showing the anticompetitive effects of common ownership to date have referred to
industries in which there are no dual-class companies among the main competitors.

Table 1: List of the main studies showing that common ownership leads to higher prices
and the main competitors in the relevant sectors.

Study Sector Main Competitors

Anticompetitive Effects of Airline Delta Air Lines,
Common Ownership11 2 Southwest Airlines Co.,
American Airlines,
United Cont'l Holdings,
Alaska Air,
JetBlue Airways,
Spirit Airlines,
Allegiant Travel Co.,
Hawaiian
Ultimate Ownership and Banking (deposit JP Morgan Chase,
113
Bank Competition accounts) Bank of America,
Citigroup,
Wells Fargo,
PNC Financial,
U.S. Bancorp
PriceEffects of Common Soy, corn, and cotton Monsanto,
Ownership in The Seed seeds Bayer,
Sector1 14 Dow,
DuPont,

111. Xie & Gerakos, supra note 655, at 572; see also Albert Banal-Estaiol, Melissa Newham & Jo
Seldeslachts, Common Ownership in the U.S. Pharmaceutical Industry: A Network Analysis, 66 ANTITRUST
BULL. 68, 98-99 (2021) (studying the ownership network comprising brand firms and generic firms).
112. Jose Azar, Martin C. Schmalz & Isabel Tecu, Anticompetitive Effects of Common Ownership, 73 J.
FINANCE 1513 (2018).
113. Jose Azar, Sahil Raina & Martin Schmalz, Ultimate Ownership and Bank Competition, 51 FIN. MGMT.
227 (2022).
114. Mohammad Torshizi & Jennifer Clapp, Price Effects of Common Ownership in The Seed Sector, 66
ANTITRUST BULL. 39 (2021).
2023 ] Dual-ClassShares in the Age of Common Ownership 561

Syngenta,
BASF
Common Ownership in the Pharmaceutical Johnson & Johnson,
U.S. Pharmaceutical Pfizer Inc.,
Industry;" 5 AbbVie Inc.,
Merck & Co Inc.,
Common Ownership and Bristol Myers Squibb Co.,
Market Entry;1 16 Abbott Laboratories,
Eli Lilly,
The Anticompetitive Effects Gilead Sciences,
11 7 Amgen,
of Common Ownership;
Viatris"8

115. Albert Banal-Estafol, Melissa Newham & Jo Seldslachts, Common Ownership in the
U.S. Pharmaceutical Industry:A Network Analysis, 66 ANTITRUST BULL. 68 (2021).
I 16. Melissa Newham, Jo Seldeslachts & Albert Banal-Estahol, Common Ownership and Market Entry:
Evidence from the Pharmaceutical Industry 41 (DIW Berlin, Discussion Paper No. 1738, 2018).
I 7. Jin Xie & Joseph Gerakos, The Anticompetitive Effects of Common Ownership: The Case of Paragraph
IV Generic Entry, 10 AM. ECON. ASS'N PAPERS & PROC. 569 (2020).
118. For a complete overview of the ten main U.S. biotech and pharmaceutical competitors, see Matej
Mikulic, 2022 Ranking of Top U.S. Biotech and Pharmaceutical Companies Based on Total Revenue, STATISTA
(Oct. 14, 2022), https://www.statista.com/statistics/257436/top-global-biotech-and-pharmaceutical-companies-
based-on-revenue/ [perma.cc/C6BT-43YC].
562 The Journalof CorporationLaw [Vol. 48:3

B. Dual-Class Shares and Mandatory Time-Based Sunsets

Both institutional investors and academics have lon advocated for the introduction
of a mandatory time-based sunset on dual-class shares.'"I Justification for this lies in the
prediction that, after a certain time, dual-class shares become inefficient because, to put it
colorfully, founders will lose their "magic touch." We graphically present the relationship
that supporters of mandatory sunsets hypothesize between the costs (C) and benefits (B) of
dual-class shares over time (t) as follows:

Costs/
Benefits
C

t* t

The optimal time-based sunset rule would then set an expiration date equal to t*.The
current debate on mandatory time-based sunsets highlights a basic problem with them,
namely that the slope and the intercept of B and C depend on the characteristics of the
company and its founders and on how market conditions evolve.20 As a consequence, the
slope and intercept are likely to be different for each corporation. Because a regulator
cannot estimate these variables for any individual firm, it also cannot identify t*. Therefore,
policymakers will have to select a duration of the mandatory time-based sunset (let us call

119. See, e.g., Mahoney, supra note 16 (advocating, on behalf of the Council for Institutional Investors, for
the time-based dual-class sunsets).
120. See Fisch & Solomon, supra note 844, at 1082 ("[A] one-size-fits-all approach to sunsets-like those
proposed by CII or adopted by index providers-does not make sense. The timeframe necessary for realizing a
company's goals is likely to vary depending on the company, based on factors like the company's maturity at the
IPO stage, the duration of its business model, and the time required to develop its products or services and bring
them to market.").
2023 ] Dual-ClassShares in the Age of Common Ownership 563

it t') that is bound to be inefficient for each individual dual-class firm. Firms for which
t* > t' would have to switch prematurely, while for those with t* < t (the inertially non-
switching firms), an inefficient structure would remain in place. Admittedly, one could
argue that, compared to a solution in which there is no expiration date, mandatory time-
based sunsets would be an improvement on the status quo for inertially non-switching
firms. Yet, there is no way to estimate whether the losses imposed on prematurely
switching firms would outweigh the gains for inertially non-switching firms.
In addition, if one adopts an ex-ante perspective, the benefits of a mandatory time-
based sunset provision become dubious even for inertially non-switching firms. If investors
believe that founders will extract excessively high private benefits of control or that the
duration of the dual-class structure is excessive, they will be willing to pay a lower price
for shares at the IPO stage. In other words, market forces determine the characteristics
of an offering and the price at which the shares are sold. In a situation where a regulator
has no way of determining t* with any accuracy, it is unclear why it would be desirable to
displace market forces by tying the hands of both potential investors and founders. 2 In
fact, it should be noted that firms can already set an expiration date (and quite a few have
done so 3) if they believe that this is a solution that would be rewarded by financial
markets.
More generally, a mandatory time-based sunset is built on the idea that the founders'
idiosyncratic vision is strictly tied to the time around the IPO and is bound to expire a few
years later.124 However, as eloquently put by Goshen:
[T]here is nothing in the economy, or in life, suggesting that idiosyncratic vision
is timed to the IPO moment and tied to the founder. It can come at any time
(Steve Jobs invented the iPhone long after the IPO), and it can be gained by non-
125
founders (Tim Cook created more value after Steve Jobs had gone).
Lastly, time-based sunset provisions also create a well-known moral hazard
problem.126 A sunset clause artificially introduces a sharp cliff because the insider will lose
control of the corporation from one day to the next. As a result, the insider might have

121. See Ronald J. Gilson, Evaluating Dual Class Common Stock: The Relevance of Substitutes, 73 VA. L.
REV. 807, 808-09 (1987) (noting that shares' lower voting rights will affect their market prices, "so that the
company's owners at the time it goes public, and not the purchasers, bear the cost").
122. See also Fisch & Solomon, supra note 844, at 1063 ("[I]t is unclear how a bright-line time limit that
does not reflect company-specific needs makes sense.").
123. Companies with Time-Based Sunsets on Dual-Class Stock, COUNCIL OF INSTITUTIONAL INVS. (2021),
https://www.cii.org//Files/issues_and_advocacy/DualClassStock/7-22-21 %20Time-based%20Sunsets.pdf
[https://perma.cc/VU7J-UNQK] (providing a list of 58 dual-class companies that have a time-based sunset).
124. See generally Zohar Goshen, Against Mandatory Sunset for Dual Class Firms, CLS BLUE SKY BLOG
(Jan. 2, 2019), https://clsbluesky.law.columbia.edu/2019/01/02/against-mandatory-sunset-for-dual-class-firms/
[https://perma.cc/S346-DSQW].
125. Id.
126. See Fisch & Solomon, supra note 844, at 1083-84 (discussing the moral hazard problems that sunsets
create).
127. Coffee, supra note 988.
564 The Journalof CorporationLaw [V ol. 48:3

strong incentives to engage in excessive short-termism or rent-seeking. 12 8

In the current setting, where FVM and PVM shareholders co-exist, a mandatory time-
based sunset clause effectively hands over (some degree of) control to PVM shareholders.
By definition, PVM shareholders will prefer strategies that do not maximize firm value in
some instances. Thus, a mandatory time-based sunset might have a negative impact on the
firm's profitability.
From a social welfare perspective, a mandatory sunset has the additional disadvantage
of enhancing the voice of PVM common owners, who are likely to prefer a lower level of
competition in the market in which the firm operates. In turn, this will have a negative
impact on final consumers and on social welfare in general.
Bebchuk and Kastiel, who are the main proponents of mandatory time-based sunsets,
implicitly acknowledge that regulators cannot identify t* and, in fact, concede that
shareholders unaffiliated with the controller should have the right to extend t'.1 29 One
problem with this proposal is that, as noted by Bebchuk, Cohen, and Hirst, large PVM
shareholders have limited incentives to invest resources in learning about a given portfolio
firm, as they would appropriate a small fraction of any increase in value. 130 For this reason,
they are unlikely to have the very specific knowledge that would be necessary to assess the
remaining value of a founder's idiosyncratic vision.
Worse still, our framework sheds light on how allowing shareholders to vote on a
possible extension of the dual-class structure would, in all likelihood, be pointless. In a
world in which most shareholders are PVM, such shareholders can be expected to vote
against dual-class shares, regardless of whether that is in the best interests of the individual
company. In fact, by voting against dual-class shares, PVM shareholders increase their
influence on the fum's strategies, thus increasing the likelihood that the fin's conduct
falls into line with their preferences. As diversified PVM shareholders are generally
common owners, this might have a negative effect on the level of competition in the
relevant markets.

IV. DUALCLASS SHARES AND SYSTEMICALLY IMPORTANT FIRMS

We have argued that, in a world dominated by PVM investors, dual-class companies

can invigorate competition despite the rise of common owners. By silencing PVM
shareholders who might prefer a lower level of competition, dual-class shares give FVM
insiders the power to engage in aggressive competition. Nonetheless, silencing PVM
shareholders to the advantage of FVM shareholders may not be optimal across the board.
This Part starts by noting that societies currently face serious economy-wide threats
such as climate change and macroeconomic shocks. We then describe how, in both the case
of climate change and in that of macroeconomic risk, a small subset of firms plays a
disproportionately large role in creating negative systemic externalities. Allowing FVM

128. Fisch & Solomon, supra note 844, at 1083; see also Coffee, supra note 988 (arguing that "[a] sunset
that goes sharply from day to night may have perverse effects").
129. Bebchuk & Kastiel, supra note 766, at 624-25.
1 30. Lucian A. Bebchuk, Alma Cohen & Scott Hirst, The Agency Problems of InstitutionalInvestors, 31 J.
ECON. PERSPS. 89, 96-97 (2017) (discussing a simplified example based on realistic figures in which an
institutional investor only appropriates $1200 for an increase in value of its portfolio company of $1,000,000).
2023 ] Dual-ClassShares in the Age of Common Ownership 565

insiders who are oblivious to such externalities to have permanent control over these
companies thanks to dual-class shares could have negative consequences for both the
climate and the stability of the economy. Instead, PVM shareholders internalize a larger
fraction of systemic externalities via their other portfolio firms and, at the margin, have
13 1
stronger incentives than FVM shareholders to internalize such externalities. For these
reasons, we suggest imposing limitations on the freedom of these key firms to adopt dual-
class shares and explain in detail how such limitations should be tailored according to
firms' propensity to impose systemic externalities. A possible objection to our proposal is
that corporate governance is not the right tool to target systemic externalities. We address
this objection in Part IV.B.1.

A. Climate Change
According to the 2021 IPCC Report, "[i]t is unequivocal that human influence has
warmed the atmosphere, ocean and land."1 Fossil fuels, intensive farming, deforestation,
and soil impoverishment have led to an enormous increase in the concentration of
3 3
greenhouse gases (GHGs) such as C02 and methane (CH4) in the atmosphere. The higher
concentration of GHGs has resulted in the global average temperature rising at an
unprecedented rate. 134
35
This dynamic has potentially catastrophic consequences, which include droughts,'
heavy precipitation,136 extreme heatwaves,137 and a loss of biodiversity.138 In addition,
climate change has important economic implications. The World Economic Forum, in its
Global Risk Report 2021, identified climate change as the most significant and "second
most likely long-term risk" for the economy,1 39 while the Financial Stability Oversight

131. Enriques & Romano, supra note 23, at 79-83 (discussing under which conditions the voice of portfolio
value maximizing shareholders should be enhanced).
132. Valerie Masson-Delmotte et al., Climate Change 2021: The Physical Science Basis, Summary for
Policymakers, INTERGOVERMENTAL PANEL ON CLIMATE CHANGE 4 (2021),
https://www.ipcc.ch/report/ar6/wg I /downloads/report/IPCCAR6_WGISPM_final.pdf
[https://perma.cc/GN8K-MPLB].
133. See, e.g., Trends in Atmospheric Carbon Dioxide, GLOB. MONITORING LAB'Y: NAT'L OCEANIC
&

ATMOSPHERIC ADMIN. (Apr. 5, 2023), https://gml.noaa.gov/ccgg/trends/global.html [https://perma.cc/GX8C-

L2C4] (reporting data on the increase in CO2 concentration from 1980); Trends in Atmospheric Methane, GLOB.
MONITORING LAB'Y: NAT'L OCEANIC & ATMOSPHERIC ADMIN. (Apr. 5, 2023),
https://gml.noaa.gov/ccgg/trendsch4/ [perma.cc/FF7B-YQKR] (reporting data on the increase in CH 4
concentration from 1980).
134. See Masson-Delmotte et al., supra note 132, at 6 (describing how GHGs have significantly affected
rising temperatures across the world).
135. Id. at 10, 18.
136. Id. at 8, 10.
137. Id.
138. Sarahi Nunez et al., Assessing the Impacts of Climate Change on Biodiversity: Is Below 2* C Enough?,
154 CLIMATIC CHANGE 351, 352 (2019) (explaining how climate change poses a threat to biodiversity).
139. See WORLD ECON. F., Global Risks 2021: FracturedFuture, in GLOBAL RISKS REPORT 2021, at 13
(2021), https://www3.weforum.org/docs/WEFThe_GlobalRisksReport _2021.pdf [https://perma.cc/724L-
K8E5] (reporting that survey respondents ranked "Climate Action Failure" as "the most concerning risk[]
globally").
566 The Journalof CorporationLaw [V ol. 48:3

Council issued a formal warning defining climate change as an "emerging threat" for the
U.S. financial system.1 40
Scientists are attempting to persuade governments to take drastic actions to reduce
GHG emissions and prevent such catastrophic consequences. 14 1 What actions should be
taken? To answer this question, one must ask who is responsible for GHG emissions.
Compelling evidence suggests that a small subset of "carbon majors" plays a very
significant role in accelerating global warming. For instance, a recent and widely-cited
study has shown that almost two-thirds of C02 and CH4 emissions between 1854 and 2010
can be attributed to 90 entities, many of which are among the largest corporations on the
planet.142 Five U.S. corporations alone-Chevron, ExxonMobil, ConocoPhillips, Peabody
Energy, and Consol Energy-are responsible for. 9.39% of global emissions during that
period.143 Turning to more recent data, a list of the top 100 U.S. polluters reveals that in
2020, five corporations-Vistra Energy, Duke Energy, Southern Company, Berkshire
Hathaway (a dual-class company), and American Electric Power-caused 6.2% of U.S.
GHG emissions.144
Against this background, we suggest that corporate governance should be among the
tools aimed at climate change mitigation. The use of governance-based legal strategies has
two advantages. First, they would directly affect the incentives of corporate decision-
makers at carbon majors. This is especially important, given the disproportionate role that
carbon majors play in accelerating global warming. Second, by targeting insiders'
incentives, governance-based strategies could leverage the disaggregated information held
by the insiders. While policymakers have an informational advantage when it comes to
assessing the consequences of GHG emissions, corporate insiders have superior
information about their corporations and hence can best decide which courses of action can
mitigate the environmental effects of their company without jeopardizing profitability.

B. Macroeconomic Risk and CorporateGovernance

Macroeconomic shocks are another impending threat to interconnected economies.
Modern macroeconomics acknowledges that some key firms and sectors have a
disproportionate effect on the level of macroeconomic risk to which the economy is

140. Alan Rappeport, Climate Change an "Emerging Threat" to U.S. Financial Stability, Regulators Say,
N.Y. TIMES (Dec. 17, 2021), https://www.nytimes.com/2021/12/17/us/politics/climate-change-us-financial-
threat.html [perma.cc/G765-58XY].
141. See Jeff Tollefson, IPCC Says Limiting Global Warming to 1.5°C Will Require Drastic Action, NATURE
(Oct. 8, 2018), https://www.nature.com/articles/d41586-018-06876-2 [https://perma.cc/DR2K-RRY4]
(summarizing IPCC's main findings for 2014 and reporting researchers' concerns about the revised carbon
budgets estimated by the IPCC released in 2014).
142. See generally Richard Heede, Tracing Anthropogenic Carbon Dioxide and Methane Emissions to Fossil
Fuel and Cement Producers, 1854-2010, 122 CLIMATIC CHANGE 229, 236-38 (2014).
143. Id. at 237.
144. See generally POL. ECON. RsCH. INST., supra note 21.
2023 ] Dual-Class Shares in the Age of Common Ownership 567

exposed.1 45 To put it differently, dual-class companies have the ability to impose systemic
externalities; think of the centrality, each in its own way, of Alphabet and Visa.
For instance, Acemoglu and his co-authors highlight that sectoral shocks can generate
sizable macroeconomic fluctuations when some industries are significantly more
interconnected than others, 146 while Atalay found "that sectoral shocks are the primary
47
source of GDP fluctuations."1 The results of these studies are confirmed by Baqaee and
Farhi, whose work shows that in the presence of input-output intersectoral linkages, shocks
to critical sectors can result in significant macroeconomic consequences.148
49
Many of these works build on an important concept, namely centrality.1 In
qualitative terms, centrality aims at capturing the importance of a node within a network
and hence, in this case, of a firm or sector within an economy. In mathematical terms, there
are many ways to calculate centrality, and thus its precise meaning depends on the
formulation adopted. The most intuitive measure of the importance of a given node within
a network is degree centrality.10 The degree of a node indicates the number of connections
that it has with other nodes. Thus, a sector with many direct connections-generally
defined in terms of input-output relationships-with other sectors has a high degree of
centrality. However, this measure is not always sufficient to accurately describe a complex
economy, and therefore economists usually rely on indicators that also account for indirect
connections. From this perspective, a node is more central if it is connected with nodes that
have more connections.15 1 One of the most widely used centrality measures with this
152
characteristic is eigenvector centrality.
Ample empirical evidence confirms that sectors with higher eigenvector centrality
153
cause larger spillovers onto other firms and are more likely to trigger macroeconomic

145. See, e.g., Xavier Gabaix, The Granular Origins of Aggregate Fluctuations, 79 ECONOMETRICA 733,
736 (2011) (showing that idiosyncratic shocks hitting the top 100 firms account for one-third of GDP aggregate
fluctuations).
146. See Daron Acemoglu et al., The Network Origins ofAggregate Fluctuations, 80 ECONOMETRICA 1977,
2004 (2012) (finding "that sizable aggregate fluctuations may originate from microeconomic shocks only if there
are significant asymmetries in the roles that sectors play as direct or indirect suppliers to others").
147. Enghin Atalay, How Important Are Sectoral Shocks?, 9 AM. ECON. J.: MACROECONOMICS 254, 276
(2017) (explaining how the source of GDP fluctuations are sectoral shocks).
148. See David Rezza Baqaee & Emmanuel Farhi, The Macroeconomic Impact of Microeconomic Shocks:
Beyond Hulten 's Theorem, 87 ECONOMETRICA 1155, 1156 (2019) (arguing and then proving that shocks to more
connected sectors are likely to produce more serious aggregate consequences).
149. See SANJEEV GOYAL, CONNECTIONS: AN INTRODUCTION TO THE ECONOMICS OF NETWORKS 16-20
(2012) (introducing various measures of centrality).
150. Id. at 16.
151. Vasco M. Carvalho, From Micro to Macro Via Production Networks, 28 J. ECON. PERSPPERSPS. 23, 36
(2014).
152. See id. (discussing eigenvector centrality and some of its most popular variants like Google's PageRank
algorithm).
153. Daniel Aobdia, Judson Caskey & N. Bugra Ozel, Inter-Industry Network Structure and the Cross-
Predictability of Earnings and Stock Returns, 19 REv. ACCT. STUD. 1191, 1193 (2014) ("[T]he association
between central industries' ROA [Returns On Assets] changes and ROA changes of the industries they trade with
is over two times greater than that of noncentral industries.").
568 The Journalof CorporationLaw [Vol. 48:3

fluctuations.1 5 4 We suggest that corporate governance rules can play an important role in
this context as well. The first-best solution would certainly be having an omniscient and
benevolent regulator to determine each corporation's optimal level of risk-taking via ex-
ante regulation. However, regulators clearly lack the necessary information to craft such
detailed and tailored regulations. Thus, a second-best solution is giving voice to corporate
insiders who have incentives that are less misaligned with those of society at large.

C. Limits on Dual-Class Sharesfor Key Firms

As noted in the two previous Parts, shocks at some key firms can produce systemic
consequences. A small number of carbon majors are having a disproportionate effect on
climate change,1 55 while some macroeconomic-central firms are responsible for a
significant fraction of aggregate fluctuations.156 Granting FVM founders the ability to
control these systemically relevant firms, even with small stakes, might thus have obvious
consequences for social welfare. 5 7 Therefore, we suggest that, there should be some
limitations on dual-class shares at these firms, which should reflect the degree of systemic
relevance of the given dual-class firm. In this Part, we discuss how this tailored regime
could be implemented.

1. Carbon Majors

Relatively few firms are responsible for most emissions. 18 FVM shareholders have
obvious incentives to be almost entirely oblivious about these externalities, whereas PVM
shareholders have relatively strong incentives to mitigate the negative effect of carbon
majors on the environment.159 Thus, we suggest that the voting power of insiders with high
160
voting shares should decrease when the firm is responsible for significant emissions.
To understand how this should be done, consider that the current way of determining
the voting power of the i - th insider holding multiple voting shares in the j - th firm is:

154. See Acemoglu et al., supra note 1466, at 2004 ("[S]izable aggregate fluctuations may originate from
microeconomic shocks only if there are significant asymmetries in the roles that sectors play as direct or indirect
suppliers to others.").
155. See supra notes 142-1454 and accompanying text.
156. See supra notes 1475-48 and accompanying text.
157. To be sure, it would be the same or even worse if founders controlled these firms with larger cash-flow
stakes because they would be even more incentivised to maximize firm value. The point is that they could never
retain control if they choose to go public without a dual-class share structure: unless, that is, they keep the
company small and therefore peripheral. See supra note 73 and accompanying text.
158. See supra notes 142-144 and accompanying text; Anna Christie, The Agency Costs of Sustainable
Capitalism, 55 U.C. DAvis L. REV. 875, 891 (2021) (reporting that a small number of corporations can be
regarded "as the key corporate perpetrators of global warming").
159. This is of course not to say that they have optimal incentives to internalize all climate externalities. See
Roberto Tallarita, The Limits of Portfolio Primacy, 76 VAND. L. REV. (forthcoming 2023) (manuscript at 48),
https://papers.ssm.com/sol3/papers.cfm?abstractid=3912977 [https://perma.cc/4629-742B] (noting that
institutional investors' portfolio companies only internalize a fraction of climate externalities given that their
effective impact would be limited on a global scale).
160. To prevent major emitters from having incentives to remain private, a similar rule could be extended to
private companies who cause outsized emissions.
2023 ] Dual-ClassShares in the Age of Common Ownership 569

Pii = ' (1).

Pi indicates the voting power of the insider, Sii is the number of shares the i - th insider
161
holds in the j - th firm; V denotes the votes per share and T is the total number of votes
that can be cast by all shareholders of that firm.
For instance, assume that an insider holds shares that carry ten votes each, that she
holds ten shares, and that the total votes that can be cast (including the insider's) are 150,
as there are 50 other shares with one vote each. Then, while the insider only holds 16.7%
of the shares, according to (1) her voting power is 66.7%.
For carbon majors, we argue that the formula should be corrected as follows:

P = ',* (2).
T*(1+aE)
Here, E are the j - th firm's CO 2 emissions divided by the total U.S. emissions of C02
and a is a parameter that allows the policymaker to decide at which rate the firm's
emissions affect insiders' voting power. Note that this formula would apply only to the
shares with extra voting power. For all other shares, the standard one share, one vote rule
would apply.
Suppose that there are two companies, A and B, with insiders' shares and voting rights
the same as in the example preceding formula (2). A is a major polluter,162 so that EA =
0.012, whereas B is an environmentally friendly firm so that EA = 0.00001. Suppose also
that the regulator sets a = 200. By applying formula (2) to the numbers in this example,
we obtain that the voting power of the insider in A will be 19.6%. Instead, the voting power
of the insider in B will not decrease by much as a consequence of our rule (66.5%).
In other words, this formula ensures that only insiders of firms that cause massive
externalities in the form of GHG emissions will have their voting power curbed, which will
give PVM shareholders a greater voice at these firms. Viewed another way, this rule gives
insiders incentives to curb emissions if they want to retain control over the corporation.

2. Macroeconomic-CentralFirms

As noted in Part IV.B, a number of empirical studies demonstrate that eigenvector

centrality is a proxy for a firm's or a sector's ability to impose externalities and to have a
systemic effect.163 For this reason, we suggest that eigenvector centrality should be used
to scale the voting power of insiders. Higher values of eigenvector centrality imply that the
firm can impose larger externalities, and hence the voting power of FVM founders should
be ceteris paribus smaller. Conversely, lower values of eigenvector centrality imply that a

161. The vast majority of dual-class firms have a class of shares with one vote per share and a class of high
voting shares with ten votes per share. See supra note 71 and accompanying text.
162. An example of major emitter with dual-class shares is Berkshire Hathaway, as it is the fourth ranked
company in the United States in terms of CO2 emissions (67,213,495 C02 equivalent metric tons, or 1.1% of 2020
GHG emissions). POL. ECON. RsCH. INST., supranote 21.
163. See generally Heede, supra note 142.
570 The Journalof CorporationLaw [V ol. 48:3

firm can only impose smaller externalities, and hence the voting power of FVM founders
should be ceterisparibus larger in this case.
We suggest that formula (1) should be adapted for eigenvector-central firms as
follows:
Si j * V (3)
P _

'i T * (1 + f#NEj)(3
In this formula, NEj is the eigenvector centrality of the j - th firm and fl is a parameter
that allows the regulator to decide at which rate firm centrality should affect voting
power.164 Like (2), this formula would apply only to the shares with extra voting power.
For all other shares, the standard one-share-one-vote rule will apply.
The properties of (3) mirror those of (2). The more central the firm is in the economy,
the more equity an insider needs to hold in order to retain control over it. Building on the
previous example, consider the case of two firms, A and B. Assume that the regulator set
# = 1, that A is peripheral, so that, for example, NEA = 0.01, whereas B is more central,
so that NEB = 0.7. The voting power of the insider in firm A would be equal to 60.6%.
Because the firm is peripheral, and hence negative externalities are not a serious concern,
the voting power of the insiders is not significantly constrained by our rule. In B, the voting
power of the insider would instead equal 39%. In this case, the voting power of the insider
is drastically reduced, and she would no longer be able to cast the absolute majority of
available votes. This would be because, due to its higher centrality, B can impose more
relevant systemic externalities.
To be sure, for a number of peripheral firms, a rule of this kind would also, at the
margin, discourage FVM shareholders from controlling the firm because it would reduce
the shareholders' voting power as the firm becomes relatively more central. While we note
that the effect is minimal because NE will approach zero for peripheral firms, a
policymaker aiming to prevent this issue can identify a centrality threshold below which
the proposed rule would not apply.

C. Combining Climate andMacroeconomic Externalities

The two approaches presented in Parts IV.C.1 and IV.C.2 can be combined as follows:
P. Si 'j * V (4)
p T * (1 + aE) * (1 +/3NEj)

Therefore, a policymaker concerned about both climate change and macroeconomic risk
can implement (4). By choosing the values of a and /3, it can decide which goals to
prioritize.
For instance, a policymaker that is particularly concerned with climate change will
choose high values of a, while a policymaker interested in ensuring the stability of the
economy will pick high values of fl. Importantly, the values of a and fl will also depend
on the perceived severity of the anticompetitive problems allegedly caused by common
ownership. A policymaker who prioritizes aggressive competition by firms led by FVM

164. Note that if the goal is also not to discourage FVM shareholders from controlling peripheral firms, this
formula should only apply to firms with a NE above a certain threshold.
2023 ] Dual-ClassShares in the Age of Common Ownership 57 1

shareholders will choose lower values of a and /l, whereas a policymaker who believes
that common ownership does not negatively affect the level of competition in product
markets will select higher values.

V. ADVANTAGES AND ANTICIPATED OBJECTIONS

In this Part, we discuss the possible advantages of our proposal and counter objections
that could be raised against it.

A. Advantages of Our Proposal

The proposed solution presents several advantages. First, it can be tailored to

corporations' ability to impose systemic externalities. This ensures that there are
limitations on private ordering only for those firms that externalize a large fraction of the
costs associated with their activity and for which, hence, private ordering is bound to fail.
Second, our rule can be seen as a precious complement to public regulation. We have
proposed targeting climate change and macroeconomic risk because they are massively
consequential threats for which public regulation is clearly insufficient. Importantly, the
scope of the rules we suggest can be either restricted-for instance, by excluding the
climate change multiplier if an effective carbon tax is passed-or expanded to cover other
systemic threats.
Third, our rule does not create a sharp cliff, and the associated moral hazard
problem,165 because marginal changes in the firm's ability to cause externalities go hand-
in-hand with marginal changes in the insider's voting power.166
Fourth, our rule does not artificially tie the time-based sunset to the time of the IPO.
Therefore, insiders of carbon majors like Warren Buffett, who want to retain control of the
corporation because they believe in the value of their idiosyncratic vision, can do so by
either lowering their company's emissions or by keeping the wedge between the equity
voting and the voting interest below a certain threshold. Instead, insiders of
macroeconomically central firms like Sergey Brin and Larry Page have limited control over
the eigenvector centrality of their company because this indicator also depends on the
interconnectedness of the firms to which their corporation is connected. Hence, insiders of
firms who become increasingly central can only reduce the voting wedge in order to retain
control of their corporation.

165. See supra notes 1276-1288 and accompanying text.

166. An alternative way of framing this difference is that time-based sunsets are "bumpy" because small
time differences result in completely different legal outcomes. At the expiration date, the insider will suddenly
and totally lose control of the company. By contrast, our rule is "smooth," as marginal changes in circumstances
only produce marginal changes in legal outcomes. For a discussion of bumpy and smooth rules and the advantages
of the latter, see Adam J. Kolber, Smooth and Bumpy Laws, 102 CALIF. L. REV. 655, 657 (2014) (introducing the
idea of "bumpy" laws); Luca Enriques, Alessandro Romano & Thom Wetzer, Network-Sensitive Financial
Regulation, 45 J. CORP. L. 351, 389-90 (2019) (discussing the disadvantages of "bumpy" rules in financial
regulation).
57 2 The Journalof CorporationLaw [Vol. 48:3

B. Possible Counterarguments

We now turn to the possible objections, starting with the view that corporate
governance should not be one of the tools to tackle systemic externalities like climate
change or macroeconomic risk.

1. "CorporateGovernance is not the Right Tool to Tackle Systemic Externalities"

Our claim that corporate law should be among the tools to mitigate climate change
rests on the following three pillars: (i) because no single measure in isolation can tackle
climate change, a range of policy instruments all skewing the incentives of key decision-
makers towards greener policies is needed; (ii) even if some policies are in principle
effective to some extent, policymakers have insufficient incentives to pass them or to
structure them optimally; and (iii) passing incremental green policies can help build social
norms that favor the implementation of further policies aimed at mitigating climate change.
Moreover, as we mention in Part IV.A, corporate law is uniquely situated to target precisely
the incentives of decision-makers at major polluters.
To begin with, policies devised thus far have shared two common traits: they are hard
to pass and they are imperfect. Environmental protection and climate change mitigation are
67
social norms that have not been internalized by a large fraction of the U.S. population,1
and thus policymakers have limited incentives to pass effective and comprehensive
reforms. Furthermore, a country implementing policies that cut emissions bears their full
costs, but will only internalize a small fraction of the benefits.168 Similarly, a significant
share of the benefits of such policies will be enjoyed by future generations, while the costs
are borne by present-day voters.169 For these reasons, policymakers have suboptimal
incentives to pass comprehensive policies that can curb GHG emissions. Political inertia
thus stands in the way of effective regulatory solutions.170
Even putting the politics aside, global warming is a complex issue for which no single
policy can be a silver bullet. For the sake of brevity, we illustrate this point with reference
to carbon pricing: a tool that has been heralded as an "indispensable" instrument to fight

167. According to a recent survey by Yale and George Mason, only 58% of Americans are either concerned
or alarmed about climate change. Global Warming's Six Americas: September 2021, GEO. MASON UNIV. CTR.
FOR CLIMATE CHANGE COMMC'N (Jan. 12, 2022), https://www.climatechangecommunication.org/all/global-
warming-six-americas-2021/ [perma.cc/HKT2-WCS6]. The rest of the population is either dismissive, doubtful,
disengaged, or cautious. Id.
168. William Nordhaus, Climate Change: The Ultimate Challengefor Economics, 109 AM. ECON. REV.
1991, 2007 (2019) ("No single country has an incentive to cut its emissions sharply. Suppose that when country
A spends $100 on abatement, global damages decline by $200. However, country A might get only $20 of the
benefits, so it would tend to decline the responsibility.").
169. Id. ("There is a tendency for the current generation to ride free by pushing the costs of dealing with
climate change onto future generations.").
170. See Christie, supra note 158, at 897 (underlining the global political inertia to adequately tackle climate
change and suggesting that "any progress that investors can make to mitigate corporate climate change damage
would in itself be a valuable contribution to society").
2023 ] Dual-ClassShares in the Age of Common Ownership 573

climate change.' 7' Let us start with carbon taxes: a form of carbon pricing particularly
popular within policy circles.1 72
A carbon tax sets a price on carbon emissions "by defining a tax rate on greenhouse
gas emissions or-more commonly-on the carbon content of fossil fuels."1 73 Therefore,
74
carbon taxes increase the cost of carbon-intensive activities and products,1 incentivizing
75
both consumers and producers to shift towards more sustainable options.' Yet, despite
their virtues, carbon taxes also display the two traits described above. First, there has never
been sufficient political momentum to implement this kind of device at a federal or state
level.176 Even the proposal currently under consideration by the Biden administration
would set a price way below the one suggested by leading studies on this issue.177 Second,
even if a carbon tax were passed, it would be no magic wand. Calculating the optimal value
79
of the tax is complex,1 7 8 and mistakes have serious consequences.1 Excessively low

171. See, e.g., CARBON PRICING LEADERSHIP COAL., Carbon Pricing: Indispensable for Reducing Emissions
in an Efficient Way, in REPORT OF THE HIGH LEVEL COMMISSION ON CARBON PRICES 9-14 (2017),
https://static I .squarespace.com/static/54ff9c5ce4bOa53decccfb4c/t/59b7f2409f8dce53 16811916/150522733274
8/CarbonPricingFullReport.pdf [perma.cc/Y2XH-8YGM] (discussing the importance of carbon pricing and
describing it as "indispensable").
172. See, e.g., Coral Davenport, Carbon Pricing Becomes a Causefor the World Bank and I.M F., N.Y.
TIMES (Apr. 23, 2016), https://www.nytimes.com/2016/04/24/us/politics/carbon-pricingbecomes-a-cause-for-
the-world-bank-and-imf.html [https://perma.cc/U5NQ-JWK3] (reporting that the World Bank and the
International Monetary Fund are asking governments to introduce carbon taxes).
173. Carbon Pricing, WORLD BANK, https://www.worldbank.org/en/programs/pricing-carbon
[https://perma.cc/SW4E-BH4J].
174. LAWRENCE H. GOULDER & MARC A. C. HAFSTEAD, CONFRONTING THE CLIMATE CHALLENGE 81
(2017) (comparing two different approaches to carbon pricing-carbon tax and cap-and-trade system-and
arguing that a carbon tax will decrease consumers' demand by increasing the prices of carbon intensive products).
175. See Gilbert E. Metcalf, Designing a Carbon Tax to Reduce US Greenhouse Gas Emissions, 3 REV.
ENV'T. ECON. & POL'Y 63, 75-76 (2009) (claiming that "setting a clear price on emissions provides the impetus
for emitters to begin to reduce emissions through process changes and investment" and that firms would set
emissions "to the point where the marginal cost of emissions equals marginal abatement costs"); see also William
D. Nordhaus, To Tax or Not to Tax: Alternative Approaches to Slowing Global Warming, 1REV. ENV'T. ECON.
& POL'Y 26, 30 (2007) (underscoring that "[u]nder a price approach, the level of emissions is determined
indirectly by the level of the tax or penalty levied on carbon emissions").
176. See WORLD BANK, STATE AND TRENDS OF CARBON PRICING 2021, at 22 (2021),
https://openknowledge.worldbank.org/handle/10986/35620 [https://perma.cc/YH88-J66N (providing an up-to-
date overview of worldwide available carbon pricing tools).
177. Jennifer A. Dlouhy, White House-Backed Carbon Tax in Sight for Biden's Climate Bill, BLOOMBERG
(Nov. 6, 2021), https://www.bloomberg.com/news/articles/2021-I 1-06/white-house-backed-carbon-tax-in-sight-
for-biden-s-climate-bill [https://perma.cc/G9Z8-Y7SR] (reporting that "[t]he White House . .. support[s] a
proposal to impose an almost $20 per-ton fee on carbon as part of President Joe Biden's climate-and-spending
legislation").
178. See Nicholas Stem, Joseph E. Stiglitz & Charlotte Taylor, The Economics of Immense Risk, Urgent
Action and Radical Change: Towards New Approaches to the Economics of Climate Change, 29 J. ECON.
METHODOLOGY 181, 183-84 (2022) (suggesting that standard methodologies lead to biased results, limiting the
ability to implement effective responses).
179. Id. at 189 (arguing that non-orderly transition, caused in part by the complexity of calculation, could
lead to compounding "risks to life, health and biodiversity, and the consequences of extreme events, leading to
the risks of extreme losses").
574 The Journalof CorporationLaw [V ol. 48:3

values would induce insufficient reductions in emissions, whereas overly high values
would have significant negative consequences regarding resource allocation within the
economy. Additionally, carbon taxes are likely to be regressive, as they hit the poorest
households relatively more than the richest households. 180 In sum, carbon taxes remain a
valuable tool against global warming, but they cannot be expected to solve the problem
alone.
Similar considerations also apply to the other form of carbon pricing, namely, cap-
and-trade systems, which are not the ultimate solution either.1 81 In a cap-and-trade system,
the authority assigns emission allowances to each economic actor which it targets182 and
firms that reduce carbon emissions more efficiently are allowed to sell the spared amount
in the form of permits to firms that cannot as easily reduce emissions.183 Despite the
potential appeal of this tool, cap-and-trade systems are very difficult to administer and
present a heightened risk of fraud.1 84 For instance, in 2011, the European Union
Commission noted that $40 million in allowances had been stolen across various
countries.185 Most importantly, cap-and-trade systems remain confined to limited areas and
sectors, having thus far had a limited effect.1 86
Accordingly, waiting for the optimal policy solution to climate change is like waiting
for Godot187 and means foregoing the possibility of achieving significant reductions in
GHG emissions in the meantime. 18 Additionally, passing smaller reforms can help shift

180. Tim Callan et al., The Distributional Implications ofa Carbon Tax in Ireland, 37 ENERGY POL'Y 407,
407 (2009) (estimating that "a carbon tax of E20/tCO 2 would cost the poorest households less than E3/week and
the richest households more than E4/week"); see also Metcalf, supra note 175, at 69-73 (discussing a mechanism
that would neutralize the regressive effects of a carbon tax).
181. For an overview of the most important environmental cap-and-trade regulations in the United States,
see Richard Schmalensee & Robert N. Stavins, The Design of Environmental Markets: What Have We Learned
from Experience With Cap and Trade?, 33 OXFORD REV. ECON. POL'Y 572, 573-81 (2017) (discussing six
important environmental cap-and-trade regulations).
182. Lawrence H. Goulder, Markets for Pollution Allowances: What Are the (New) Lessons?, 27 J. ECON.
PERSPS. 87, 87 (2013) (describing the development of cap-and-trade methods).
183. GOULDER & HAFSTEAD, supra note 1744, at 80. For an updated list of cap-and-trade mechanisms
adopted on the state level, see U.S. State Carbon Pricing Policies, CTR. FOR CLIMATE & ENERGY SOL. (May
2021), https://www.c2es.org/document/us-state-carbon-pricing-policies/[https://perma.cc/GNF5-SBBU].
184. Gilbert E. Metcalf, On the Economics of a Carbon Tax for the United States, 2019 BROOKINGS PAPERS
ON ECON. ACTIVITY 1, 8 Mar. 7, ), https://www.brookings.edu/wp-content/uploads/2019/03/On-the-Economics-
of-a-Carbon-Tax-for-the-United-States.pdf (noting that "[f]raud is a particularly significant problem in a system
that is creating brand-new assets (emission allowances) worth billions of dollars").
185. Id.
186. Jessica F. Green, Does Carbon Pricing Reduce Emissions? A Review of Ex-Post Analyses, 16 ENV'T
RSCH. LETTERS 1, 5 (2021) (surveying the literature on the effectiveness of carbon pricing and reporting that
studies find a "quite small" effect ranging between zero and two percent for both carbon taxes and caps and trade).
187. The colloquial expression "Waiting for Godot" originates from Samuel Beckett's play Waiting for
Godot, where the main characters, Vladimir and Estragon, wait for the arrival of someone named Godot, who is
supposed to enlighten them about the meaning of life but ultimately never comes. SAMUEL BECKETT, WAITING
FOR GODOT: A TRAGICOMEDY IN TWO ACTs (Samuel Beckett trans., 1956).
1 88. Elinor Ostrom, Nested Externalities and Polycentric Institutions: Must We Wait for Global Solutions
to Climate Change Before Taking Actions at Other Scales?, 49 ECON. THEORY 353, 354 (2012) ("[C]ontinuing
2023 ] Dual-ClassShares in the Age of Common Ownership 575

norms on climate change. In fact, "hard pushes," that is, high-impact reforms, are not only
harder to pass, but they also often backfire.189 Conversely, "gentle nudges," or lower-
impact reforms, can gradually produce a change in social norms, allowing policymakers to
90
pass more effective and comprehensive reforms.1 Professor Dan Kahan gives a
9' Over a few decades,
compelling example of this dynamic with respect to smoking.1
smoking has gone from being unregulated and bearing connotations of "sophistication and
virility,"192 to becoming a heavily regulated "disgusting habit that onlookers should not be
expected to tolerate." 193 This shift in social norms was favored by a series of incremental
regulations, which gradually implemented restrictions on smoking.1 94 As noted by Kahan,
had lawmakers attempted to pass a single comprehensive regulation including all these
limitations they would have encountered enormous resistance which might have produced
unintended consequences.195
The situation with climate change is similar. Social norms on climate mitigation have
96
not been fully internalized by populations at large.1 Hence, sudden pushes for high-
impact climate reforms could be ineffective and further polarize the discussion on climate
change. On the contrary, gentle nudges can help shape social norms.
The case for using corporate governance to mitigate macroeconomic risk is even more
straightforward. On the one hand, decisions on corporate risk-taking are arguably the core
expertise of corporate insiders and hence leveraging their knowledge by ensuring they have
the right incentives is clearly desirable. On the other hand, crafting optimal regulations to
mitigate risk-taking by central firms is impossible. Central firms like Google are complex
institutions operating across a wide range of markets. Thus, policymakers do not have the
degree of specialization that is required to intervene in their strategies without jeopardizing
profitability. As a result, any regulation is bound to be imperfect, which is why giving voice
to PVM shareholders who have incentives to internalize at least part of the systemic
externalities might be the only viable solution.

2. Other Counterarguments

Another possible counterargument to our proposal is that it might give insiders

incentives to "game" the indicator of macroeconomic centrality in order to preserve their
voting power. For climate-central firms, this is clearly not a concern because the only way
to game the formula is by lowering emissions, which is exactly the intended goal of the

to wait without investing in efforts at multiple scales may defeat the possibilities of significant abatements and
mitigations in enough time to prevent tragic disasters.").
189. Dan M. Kahan, Gentle Nudges vs. Hard Shoves: Solving the Sticky Norms Problem, 67 U. CHI. L. REV.
607, 609 (2000) (arguing that hard shoves are likely to prove a dead letter and could even backfire).
190. Id. at 610-11 (detailing in a formal model how gentle nudges will reinforce themselves and slowly
change social norms, which in turn allows to implement more ambitious gentle nudges).
191. Id. at 625-28.
192. Id. at 626.
193. Id.
194. Id. at 625-28.
195. Id. at 626.
196. GEO. MASON UNIV. CTR. FOR CLIMATE CHANGE COMMC'N, supra note 167 and accompanying text.
576 The Journalof CorporationLaw [Vol. 48:3

rule. For macroeconomic-central firms, gaming the indicator is far from easy. On the one
hand, eigenvector centrality depends also on the extent of the connectedness of the firms
to which the corporation is connected. Hence, the insider has limited ability to manipulate
this indicator. On the other hand, to the extent that an insider can game this indicator, it
would have to artificially reduce the connections that the corporation has with its business
partners. Such a strategy might have a significant negative impact on the value of the
corporation, and hence on the value of the insider's shares.
A more serious objection is that our rule is characterized by discretion and
arbitrariness, just like mandatory time-based sunsets, because policymakers can arbitrarily
set the values of a and ft. While it is undeniable that policymakers have discretion in
applying the rule we propose, such discretion pertains to factors on which policymakers
have better information than firms' insiders. In fact, the values of a and ft depend on the
perceived severity of the threats posed by global warming, macroeconomic risk, and
common ownership. Policymakers are certainly in a much better position than firms'
insiders to assess the consequences of global warming. Similarly, specialized regulators
like the Federal Reserve and the Financial Stability Oversight Council are better informed
about macroeconomic threats than firms' insiders.
On the contrary, the optimal length of mandatory time-based sunset rules depends on
the specific characteristics of the firm, the market in which it operates, and the residual
value of the idiosyncratic vision of the insiders. It is extremely unlikely that policymakers
will have better information than insiders on any of these factors. In other words, while our
proposed rule also grants policymakers discretion, it does so on dimensions in regard to
which they are likely to have an informational advantage over corporate insiders.

VI. CONCLUSION

The debate on the pros and cons of dual-class shares has always centered on the trade-
off between allowing insiders to pursue their idiosyncratic vision and agency costs and
hence on within-firm dynamics. In this Article, we argue that there is more at stake once
the importance of inter-firm dynamics is acknowledged. On the one hand, dual-class
companies can help mitigate the anticompetitive effects of common ownership, which
provides a new justification for preserving the ability of firms to go public with dual-class
shares. On the other hand, some firms are capable of imposing systemic externalities that
can contribute to global warming or macroeconomic risks. Dual-class shares allow FVM
shareholders that are oblivious to such externalities to leverage on the disproportionate
voting power created by dual-class shares in their favor and thus to control much larger,
hence more likely systemically relevant, firms. Limitations on dual-class shares can be
justified to prevent such firms from inflicting systemic externalities. With this in mind, we
have detailed a policy proposal that allows policymakers to tailor the limits on dual-class
shares according to the specific ability of a given fim to impose systemic externalities.