Government Issues in Dual Class Share Firms

Governance Issues in Dual Class Share Firms
By
Anita Anand •
Forthcoming in the Annals of Corporate Governance
•
J.R. Kimber Chair in Investor Protection and Corporate Governance, Faculty of Law, University of Toronto cross
appointed to the Rotman School of Management [email protected]. Deep thanks to Ben Alarie, Adriana
Robertson, Douglas Cumming, Naizam Kanji, Joseph McCahery, Bernard Sharfman, Richard Squire, Michael
Trebilcock, Geoff Wood and Albert Yoon for valuable comments on earlier versions and some of the ideas in this
paper. Profound thanks to Matthew Brown, Thoby King, Christopher Puskas, Davina Shivratan, Tegan Valentine, and
Alvin Yau for excellent research assistance and to the Law Foundation of Ontario for funding.
1
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3178345
https://ssrn.com/abstract=3178345
ABSTRACT ....................................................................................................................................... 3
1. INTRODUCTION ...................................................................................................................... 4
2. WHAT ARE DCS? THE BASIC IDEA ........................................................................................... 9
3. THEORETICAL BACKGROUND ............................................................................................... 12

(i) Agency Theory ................................................................................................................................................... 12
(ii) Principal Cost Theory ......................................................................................................................................... 15
(iii) Private Ordering ................................................................................................................................................ 16
(iv) Summary ........................................................................................................................................................... 17
4. EMPIRICAL STUDIES.............................................................................................................. 18
5. TWO CASE STUDIES .............................................................................................................. 22

(i) Fairfax ................................................................................................................................................................ 22
(ii) Magna ................................................................................................................................................................ 24
(iii) Summary ............................................................................................................................................................ 26
6. GOVERNANCE CHARACTERISTICS OF DCS FIRMS ................................................................. 27

(i) Majority of the Minority Vote ........................................................................................................................... 28
(ii) Sunset provision ................................................................................................................................................ 30
(iii) Independent directors ....................................................................................................................................... 31
(iv) Independent chair .............................................................................................................................................. 33
(v) Change of control provisions ............................................................................................................................. 33
7. METHODOLOGY AND CONTEXT ........................................................................................... 34
8. RESULTS ................................................................................................................................ 36
9. DIRECTIONS FOR REGULATORY REFORM ............................................................................. 43

(i) Fixed-term Sunset Clause .................................................................................................................................. 44
(ii) Disclosure of Shareholder Voting ...................................................................................................................... 50
(iii) Buyout Protections ............................................................................................................................................ 48
10. CONCLUSION ........................................................................................................................ 50
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3178345
https://ssrn.com/abstract=3178345
Abstract
In a typical public company, shareholders can elect the board, appoint auditors, and
approve fundamental changes. Firms with dual class share (DCS) structures alter this balance by
inviting the subordinate shareholders to carry the financial risk of investing in the corporation
without providing them with the corresponding power to elect the board or exercise other
fundamental voting rights. This article fills a conspicuous gap in the scholarly literature by
providing empirical data regarding the governance of DCS firms beyond the presence of sunrise
and sunset provisions. The summary data suggest that the governance of DCS firms is variable. A
large proportion of DCS firms have no majority of the minority voting provisions and no
independent chair. By contrast, almost half of the DCS firms have a sunset clause and a majority
of independent directors. Finally, just under one-third of DCS firms have change of control
provisions over and above existing law. On the basis of this evidence, this article argues against
complete private ordering in favor of limited reforms to protect shareholders in DCS firms
including: mandatory sunset provisions, disclosure relating to shareholder votes, and buyout
protections that would address weaknesses inherent in DCS firms.
Electronic copy available at: https://ssrn.com/abstract=3178345

1. Introduction
In a typical public company, shareholders can elect the board, appoint the auditors, and approve
fundamental changes. In other words, they can participate in the governance of the firm. Firms
with dual class share (DCS) structures alter this balance by inviting the subordinate shareholders
to carry the financial risk of investing in the firm without providing them with the corresponding
power to elect the board or exercise other fundamental voting rights. As Hu and Black explain,
DCS “decouple” voting rights and economic ownership.1
The rationale underlying DCS is that they preserve family or founder control while allowing the
firm to gain access to capital in public equity markets.2 By localizing control on the founders, DCS
structures prevent the firm from being easily acquired without the founders’ cooperation.3
Indeed, DCS protect the founders from the demands of ordinary shareholders, in turn allowing
them more freedom to grow the corporation.4 In the process, DCS dissuade potential suitors
who would be willing to pay a premium for shares (a boon to all shareholders, except the
controlling shareholder, of course). In short, DCS allow the founders to focus on long-term value
creation and motivate them to make firm-specific investments in their own human capital.5
1
Hu, H. and B. Black (2008) ‘Equity and Debt Decoupling and Empty Voting II: Importance and Extensions’ 156
University of Pennsylvania Law Review 625.
2
Ibid.
3
Wen, T. (2014) 'You Can’t Sell Your Firm and Own it Too: Disallowing Dual-Class Stock Companies from Listing on
Securities Exchanges’ 162 University of Pennsylvania Law Review 1495, online:
<http://scholarship.law.upenn.edu/cgi/viewcontent.cgi?article=9447&context=penn_law_review>.
4
Hill, A. (18 July 2011) “Enrolment Open for an MBA in Murdoch”, Financial Times, online:
<http://www.ft.com/cms/s/0/2fda9e8e-b176-11e0-9444-00144feab49a.html#axzz2IYIKmzDt> as cited by Wen,
ibid.
5
Cronqvist, H. and M. Nilsson (2003) ‘Agency Costs of Controlling Minority Shareholders’ 38 Journal of Financial
and Quantitative Analysis 695; DeAngelo, H. and L. DeAngelo (1985) ‘Managerial Ownership of Voting Rights: A
Study of Public Corporations with Dual Classes of Common Stock’ 14 Journal of Financial Economics 33;
Bergstrom, C. and K. Rydqvist (1990) ‘Ownership of Equity in Dual-Class Firms’ 14 Journal of Banking and
Finance 255.

Heated controversy has arisen because DCS effectively insulate management and the board,
leaving the subordinate shareholders exposed to decisions that potentially undermine their
economic interests.6 DCS allow the firm to extract capital from subordinate shareholders without
providing them with the voting power that allows them to participate in the governance of the
firm.7 In short, DCS allow managers and the board to set the short-term and long-term strategy
for the firm without the accountability checks provided by participation in the corporation by
subordinate shareholders.8
Some commentators argue that DCS should be permitted because otherwise, DCS firms would
be left to the whims of incompetent or uninformed shareholders.9 Others argue that the weaker
governance associated with DCS is built into the firm’s stock price and ultimately caveat emptor
should rule the day. That is, all investors have the choice as to whether to invest and “if you don’t
6
Anand, A. (22 February 2016) “The success stories of dual-class shares miss an incontrovertible truth”, The Globe
and Mail, online: <https://beta.theglobeandmail.com/report-on-business/rob-commentary/the-success-
stories-of-dual-class-stocks-miss-an-incontrovertible-
truth/article28830388/?ref=http://www.theglobeandmail.com&>; Shareholder Association for Research and
Education (April 2004) “Second Class Investors: The use and abuse of subordinate shares in Canada”,
Shareholder Association for Research and Education, online:
<http://www.share.ca/files/Second_Class_Investors.pdf; Masulis, R. W., C. Wang and F. Xie (2009) ‘Agency
Problems at Dual-class Companies’ 64 Journal of Finance 1697; Harris, M. and A. Raviv (1988) ‘Corporate
Governance: Voting Rights and Majority Rules’ 20 Journal of Financial Economics 203; Wen, supra note 3.
7
Willis, A. (5 August 2005) “Dual Class share structure should end, Caldwell Told” The Globe and Mail, online:
<https://www.theglobeandmail.com/report-on-business/dual-class-share-structure-should-end-caldwell-
told/article18243209>; Critchley, B. (14 May 2015) “Time for regulators to take major look at DCS” Financial
Post, online: < business.financialpost.com/news/fp-street/time-for-regulators-to-take-major-look-at-dual-class-
shares/wcm/7b02fd1c-d28a-422c-ba39-aac77a1b6bb7>; Surowiecki, J. (18 May 2012) “Unequal Shares” The
New Yorker, online: <www.newyorker.com/magazine/2012/05/28/unequal-shares>; Deal Professor. (3 February
2017) “Snap’s Plan is Most Unfriendly to Outsiders” The New York Times, online:
<https://www.nytimes.com/2017/02/03/business/dealbook/snap-ipo-plan-evan-spiegel.htm; Economist, The. (22
September 2014) “Out of Control” The Economist, online: <https://www.economist.com/news/finance-and-
economics/21618889-more-worlds-big-stockmarkets-are-allowing-firms-alibaba-sideline>.
8
Anand, A. (2012) ‘Was Magna in the Public Interest?’ 49 Osgoode Hall Law Journal 311.
9
Sharfman, B. S. (2018) ‘A Private Ordering Defence of a Company’s Right to use Dual Class Share Structures in
IPOs’ 63 Villanova Law Review 1 (Forthcoming), online:
<https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2986164>.

like them, don’t buy them.”10 The opposition to DCS continues to grow with leading shareholder
groups and owners of stock market indices voicing opposition to them.11
As the controversy has grown, so has the divergence of subtopics covered in empirical studies
relating to DCS firms. For example, Adams and Ferreira review the empirical evidence relating to
disproportional ownership and conclude that DCS firms are popular in the United Kingdom,
though the number of such firms is decreasing in Europe.12 In analyzing the private benefits of
control across several countries, Dyck and Zingales identify transactions that involve DCS firms
and measure the control premium for the holders of the superior shares (i.e. with voting power)
relative to the subordinate shares (i.e. without voting power).13 They find that higher benefits of
control are associated with more concentrated ownership.14 These are but two examples from
the extensive literature relating to DCS which, as discussed below and in Appendix 1, suggest that
DCS structures, and the academic literature relating to them, are convoluted and complex.
10
Hasselback, D. and B. Shecter. (5 May 2015) “From Cara Operations ltd to Shopify Inc: Why DCS are suddenly
cool again” Financial Post, online: <business.financialpost.com/news/fp-street/from-cara-to-google-why-dual-
class-shares-are-suddenly-cool-again>; [Hasselback and Shecter] who quotes Carol Hansell, arguing that “if
investors don’t like dual-class shares, don’t buy them.”; See also Khalil, S. and M. Magnan (2007) ‘Dual-Class
Shares: Governance, Risk and Rewards’ Ivey Business Journal Online, online:
http://iveybusinessjournal.com/publication/dual-class-shares-governance-risks-and-rewards/; Bainbridge, S.
(15 November 2015) “What To Do About Dual Class Stock (If Anything)?” Stephen Bainbridge's Journal of Law,
Politics, and Culture, online: <http://www.professorbainbridge.com/professorbainbridgecom/2015/11/what-
to-do-about-dual-class-stock-if-anything.htm >. Bainbridge states, “Public investors who don’t want lesser
voting rights stock simply won’t buy it. Those who are willing to purchase it presumably will be compensated by
a lower per share price than full voting rights stock would command and/or by a higher dividend rate. In any
event, assuming full disclosure, they become shareholders knowing that they will have lower voting rights than
the insiders and having accepted as adequate whatever trade-off is offered by the firm in recompense.”’; See
also Lee, I. B. (8 November 2005) “There is a Logic in Dual-Class Shares” National Post, online:
<https://www.law.utoronto.ca/documents/lee/DualClassOpEd.pdf>; Allaire, Y. (27 May 2015) “In Praise of
Dual-Class Shares” Financial Post, online: <business.financialpost.com/fp-comment/in-praise-of-dual-class-of-
shares>.
11
Bebchuk, L. A. and K. Kastiel (2017) ‘The Untenable Case for Perpetual Dual-Class Stock’, 103 Virginia Law
Review 585, online: <https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2954630>. [Bebchuk and Kastiel]
citing ISS and GMI Ratings. Also see, Ontario Securities Commission, MI 61-101 Protection of Minority Security
Holders in Special Transactions and Companion Policy 61-101CP Protection of Minority Security Holders in
Special Transactions (1 February 2008), online: <www.osc.gov.on.ca/documents/en/Securities-
Category6/rule_20080201_61-101_protect-minority.pdf>.
12
Ferreira, D. and R. Adams (2008) ‘One Share-One Vote: The Empirical Evidence’ 14 Review of Finance 51, online:
<http://personal.lse.ac.uk/FERREIRD/51.pdf>.
13
These countries include Canada, Denmark, Finland, Germany, Italy, Mexico, Norway, Sweden, and the United
States. See Dyck, A. and L. Zingales (2004) ‘Private Benefits of Control’ 59 Journal of Finance 593.
14
Ibid.
6

As a starting point, and in order to provide some factual context for the discussion, note that
since 2008, almost 10 percent of all US firms completing IPOs have done so with a DCS structure
in place.15 In 2015, 24 percent of firms that listed their shares on US stock exchanges had DCS,
compared to 15 percent of public firms in 2014 and only 1 percent in 2005.16 A number of well-
known US firms, including Alphabet and Facebook, have long had DCS. In Canada, the list of DCS
firms includes icons of the Canadian corporate establishment: Bombardier, Power Corp., Rogers
Communications, Onex and Canadian Tire. In recent years, both countries have seen a
remarkable increase in IPOs with DCS including Fitbit, Box, and a division of Alibaba in the U.S.
and Cara, Aritzia, Freshii and Stingray in Canada.17 Recently, stock market indices including the
S&P 500 have taken actions to exclude new listings with DCS such as Snap Inc.’s 2017 IPO.18
Obviously, change is in the air regarding DCS.
For some jurisdictions, this debate has potentially severe consequences because DCS firms play
an important role in the economy due to their substantial size relative to the average listed
company. In Canada in 2015, for example, 85 out of 1487 firms listed on the Toronto Stock
Exchange (TSX) – roughly 5.72 percent – had DCS.19 These DCS firms had an average market
capitalization of $3.39 billion,20 while the average market capitalization of the TSX as a whole was
$1.5 billion and the median market capitalization of the TSX was a mere $111.9 million.21 In short,
15
Der Marderosian, L. (25 May 2017) “2017 IPO Report” Harvard Law School Forum on Corporate Governance and
Financial Regulation (blog), online: <https://corpgov.law.harvard.edu/2017/05/25/2017-ipo-report/>.
16
Feldman, S. G. (7 April 2016) “BNA Insights: IPOs in 2016 increasingly include Dual-Class Shareholder Voting
Rights”, Bloomberg BNA, online: <www.olshanlaw.com/media/publication/362_Feldman percent20BNA
percent20Dual-Class percent20Article.pdf>.
17
See Hasselback and Shecter, supra note 10. Also Solomon, D. (4 November 2015) “Shareholders Vote with Their
Dollars to Have Less of a Say” New York Times, online:
<https://www.nytimes.com/2015/11/05/business/dealbook/shareholders-vote-with-their-dollars-to-have-less-
of-a-say.html >.
18
See Fortune. (1 August 2017) “Incentives vs. Control: An Analysis of U.S. Dual-Class Companies” Fortune, online:
<fortune.com/2017/08/01/snap-snapchat-stock-shares-sp-500/>.
19
Data taken as at Dec 31, 2015. Toronto Stock Exchange. (31 December 2015) “Listing With Us” Toronto Stock
Exchange, online: <https://www.tsx.com/listings/listing-with-us>. See also Merkley, M. (9 February 2015)
“Multiple Voting Shares: Don’t Call It a Comeback” Blake, Cassels & Graydon LLP, online:
<http://www.blakes.com/English/Resources/TrendsInsights/Pages /details.aspx?AnnouncementID=78>.
20
Data obtained from Standard & Poor’s Capital IQ and supplemented by FactSet’s financial database.
21
Toronto Stock Exchange, supra note 19.
7

DCS firms constitute a large percentage of the overall TSX Market Cap (approximately 12
percent). DCS firms constitute a big enough group to matter. Therefore, this article begins with
the proposition that we should carefully consider them.
A central question that arises, and which this article addresses, is the extent to which private
ordering should be respected, understanding that corporate law generally upholds the choices
that parties make. To what extent should the law allow the founders to pursue their
“idiosyncratic vision” for the DCS corporation?22 This article undertakes a comprehensive analysis
of the empirical and theoretical literature relating to DCS (which includes a complete reference
chart in Appendix 1) before turning to focus on governance characteristics of DCS firms. What
governance mechanisms do DCS corporations typically have? Do these governance mechanisms
suggest that regulatory reform would be useful? This article argues against complete private
ordering in favor of three modest reforms to improve governance in DCS firms including:
mandatory fixed-term sunset provisions with a majority of the minority vote at the end of the
term; disclosure relating to shareholder votes; and, buyout protections that would address
weaknesses inherent in DCS firms.
At least one other academic article analyzes DCS structures from an empirical standpoint. Winden
examines sunrise and sunset provisions found in the charters of DCS firms, with a dataset of 123
U.S. public firms. He points out, rightly, that such provisions can satisfy both the desire of
entrepreneurs to pursue their idiosyncratic visions for value creation without fear of interference
or dismissal and the need of investors for a voice to ensure management accountability.23 Unlike
Winden’s study, this article examines not one but five governance characteristics of DCS firms
and does so in the Canadian context where DCS have historically been more prevalent. Using a
hand-collected dataset comprised of all 85 DCS firms on the Toronto Stock Exchange, it examines
22
Goshen, Z. and A. Hamdani (2016) ‘Corporate Control and Idiosyncratic Vision’, 125 Yale Law Journal 560,
online: <https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2228194>. Sharfman, supra note 9.
23
Winden, A. W. (2017) Sunrise, Sunset: An Empirical and Theoretical Assessment of Dual-Class Stock Structures,
Rock Center for Corporate Governance at Stanford University Working Paper No. 228, online:
8

governance characteristics with respect to these firms that are salient in debates about DCS and
governance generally.24 This article also takes a broader look at the policy implications of
continuing to respect private ordering as a means for regulating public corporations.
This full-fledged examination of DCS firms comes at an opportune moment. With controversy
and potential regulatory reform on the agenda, the question persists as to how and whether
regulators will respond to the issues this article discusses. But before reform occurs, we should
know more about DCS, including DCS governance. Part 2 provides background in terms of the
DCS structures and the diametrically opposed views that exist regarding DCS. Part 3 examines
theoretical approaches that can be used when analyzing DCS firms including agency theory and
principal cost analysis. Part 4 reviews divergences of findings in the empirical literature while Part
5 takes up two case studies of transactions in which DCS firms transformed their respective
governance structures. Part 6 examines five governance characteristics against which DCS firms
can be examined. Part 7 outlines the methodology and context while Part 8 sets forth data
regarding DCS firm governance. The empirical analysis reveals that, generally speaking, the
governance of DCS firms is highly variable as one might expect, but certain similarities persist,
most significantly the common presence of independent directors on the board. Part 9 focuses
on policy alternatives for regulatory reform prior to the conclusion in Part 10.
2. What are DCS? The Basic Idea
“Dual class shares” and “multi-voting shares” are generic terms that refer to a type of capital
structure in a public or private corporation. The structure involves the issuance of two or more
different classes of shares whereby one class (the “superior” class) has more voting rights than
shares held, while the other class (the “subordinate” class) has fewer voting rights relative to the
24
The list draws on Anand, A., F. Milne and L. Purda (2012) ‘Domestic and International Influences on Firm-level
governance: Evidence from Canada’ 14 American Law and Economics Review 68.

shares held. As a result, holders of the superior class of shares own a greater proportion of voting
rights without having as much equity invested as the subordinate shareholders.
In a non-DCS corporation, voting rights and equity interests are aligned. This means that the ratio
of votes to shares held is one to one. In other words, an individual who contributes 60 percent of
the equity will have 60 percent company’s voting rights. By contrast, in a DCS firm, the ratio is
not proportional. DCS detach voting rights from the equity held, allowing an individual to retain
control of a corporation without holding an equivalent financial interest in the company. The
number of votes attributed to each share under these structures varies vastly from company to
company.
DCS structures are typically implemented by founders of a corporation who wish to retain control
of the company while simultaneously accessing the wealth of the public markets.25 DCS have also
been used to maintain a particular level of national shareholder ownership26 and voting control.
In these cases, the DCS structure places a limit on the proportion of total voting power which
may be held by foreign investors.
Critics of DCS argue that when they are used in the public corporation context, the decoupling of
voting rights and equity interests increases agency costs by allowing superior shareholders to act
in self-serving ways.27 A simple example of this phenomenon is a founder’s calculus on whether
to purchase a luxury good. Suppose that the sole shareholder, sole director and founder of a
technology start-up company travels on business and is thus interested in purchasing a luxury jet,
which will depreciate in value by ten million dollars as soon as she buys it and uses it. Since she
is the sole shareholder and sole director of the company, she can ultimately decide to use the
corporation’s profits to purchase this jet. However, she will also have to bear the full cost of her
25
Sharfman, supra note 9 at 6-13.
26
In Canada, some companies utilise DCS structures to ensure a minimum level of Canadian ownership is maintained
in accordance with applicable legislation. See discussion regarding empirical analysis infra and Appendix 4 for
more on these types of DCS firms.
27
Anand, supra note 6; SHARE, supra note 6; Masulis, Wang and Xie, supra note 6; Harris, supra note 6; Wen, supra
note 3.
10

decision to make this purchase. While there is no guarantee that she will act rationally or
responsibly, the alignment of voting interests and equity interests at the very least ensures that
the costs of the decision will be borne by the decision maker, thus incentivizing rational, wealth-
maximizing behaviour.
As soon as DCS are introduced, these incentives are likely to change. Suppose that the
corporation goes public and issues DCS such that the founder retains 60 percent of the voting
rights in the company, but now with only a 10 percent equity interest in the company. The cost
to the entrepreneur for the jet is now much less as the subordinate shareholders also provide
capital to contribute to the cost of the jet and bear the cost of its depreciation. Suddenly, the
calculation of whether to purchase the jet is significantly altered for the superior shareholder as
she will be able to reap the entire benefit of the jet while bearing only a fraction of its cost and
associated expenses.
This is a prototypical example of how incentives can be misaligned in a simplified DCS firm, but
the decisions at issue need not relate to luxury goods or indulgent behaviour. If senior managers
or board members who are also controlling shareholders behave in what they consider to be the
best interests of the company, say by awarding large bonuses to managers who have performed
well, the subordinate shareholders may nevertheless believe that these bonuses represent an
inappropriate amount for the associated performance. But the subordinate shareholders are
unable to change the decision, or remove the decision makers, because they do not have the
requisite voting power to affect change at the board level.
This is the key point about DCS: they create the potential, and perhaps the incentive, for
directorial and managerial misbehaviour and entrenchment.28 This understanding of potentially
28
Nicholas, B. and B. Marsh (17 May 2017) “Dual-Class: The Consequences of Depriving Institutional Investors of
Corporate Voting Rights” Harvard Law School Forum on Corporate Governance and Financial Regulation (blog),
online: < https://corpgov.law.harvard.edu/2017/05/17/dual-class-the-consequences-of-depriving-institutional-
investors-of-corporate-voting-rights >. See also Bebchuk, L., R. Kraakman, and G. Triantis (1999), Stock
Pyramids, Cross-Ownership and Dual Class Equity: The Mechanisms and Agency Costs of Separating Control
11

misaligned incentives is the basis for the theoretical analysis in the following section and for the
ultimate argument here that private ordering does not necessarily guarantee positive firm
outcomes.
3. Theoretical background
The misalignment of incentives that can emerge and gain legitimacy in a DCS firm can be analyzed
from different theoretical vantage points. This section addresses a few salient theoretical
approaches, beginning with agency theory before turning to principal cost analysis and private
ordering concepts. A complete list of studies that analyze DCS structures is found in Appendix 1.
(i) Agency Theory
Where do DCS sit in the context of agency theory? Scholarly research has long used agency theory
to describe inherent conflicts of interest among the various stakeholders of a corporation.29 An
agency relationship is one in which one or more persons engage someone to perform some
service on their behalf and which involves some delegation of decision-making authority to the
agent. If both parties are utility maximizers, the agent may not always act in the best interest of
the principal and it is generally impossible for a principal to monitor the agent at zero cost.30
Traditionally, agency theory has been viewed as an accurate depiction of the relationship
between management and shareholders in widely-held corporations rather than controlled
corporations or DCS firms. In the widely-held context, the separation of ownership (shareholders)
and control (management) facilitates management’s ability to act in self-serving ways because
From Cash-Flow Rights, National Bureau of Economic Research Working Paper No. 6951, online:
29
For a summary of this research, see Finegold, D., G. S. Benson, and D. Hecht (2007) ‘Corporate Boards and
Company Performance: Review of Research in Light of Recent Reforms’ 15 Corporate Governance: An
International Review 865.
30
Jensen, M. and W. Meckling (1976) ‘Theory of the Firm: Managerial Behavior, Agency Costs and Ownership
Structure’ 3 Journal of Financial Economics 305.
12

individual shareholders hold a relatively small portion of the company’s shares. Typical of the
classic Berle and Means view of the widely-held firm, a large number of small shareholders
implies that the single small shareholder has neither the power nor the incentive to devote
significant resources to monitor management’s behavior and undertake corrective action when
appropriate.31
The Berle and Means model is premised on the widely-held corporation as opposed to firms in
which ownership is concentrated.32 However, there are various forms of ownership
concentrations among large shareholders, including in DCS firms.33 Agency theory is pertinent in
the DCS context, first because superior shareholders in a DCS firm can be directly or indirectly
involved in managing the affairs of the corporation; and, second, because conflicts of interest can
be driven by the divergence of interests between large and small shareholders or between
subordinate shareholders and superior shareholders in the same firm. Superior shareholders can
prompt management to behave in a way that is consistent with the superior shareholders'
interests, but not with the interests of subordinate shareholders.
In particular, superior shareholders are well-positioned to affect changes in firm policy which are
value-increasing to them but not necessarily beneficial to the DCS firm as a whole.34 The loss
31
See Jensen and Meckling, supra note 30; Shleifer, A. and R.W. Vishny (1986) ‘Large Shareholders and Corporate
Control’ 94 Journal of Political Economy 461.
32
Berle, A. and G. Means (1933), The Modern Corporation and Private Property (MacMillan, rev edn 1967).
33
Spizzirri, A. and M. Fullbrook (2013) “The Impact of Family Control on the Share Price Performance of Large
Canadian Publicly-Listed Firms (1998-2012)” Clarkson Centre for Board Effectiveness, online:
<http://www.rotman.utoronto.ca/-/media/Files/Programs-and-
Areas/CCBE/FamilyFirmPerformanceReport1993-2012-Sep2013update.pdf>.
34
Shleifer and Vishny, supra note 31 at 472–74. For a recent example of this, see Bombardier’s changes to
executive compensation pay at Bombardier, Annual Meeting of Shareholders 2017 (11 May 2017), online:
<http://ir.bombardier.com/en/event-calendar/68246-annual-meeting-of-shareholders-2017#>; Woods, A. (11
May 2017), “Bombardier’s Pierre Beaudoin gives up executive title, but will continue to lead board”, Toronto
Star, online: <https://www.thestar.com/news/canada/2017/05/11/bombardiers-pierre-beaudoin-stepping-
down-as-executive-chair.html>. Marowitz, R. (9 May 2017), “Ontario Teacher’s Pension Plan votes against
Bombardier chair” The Toronto Star, online: <https://www.thestar.com/business/2017/05/09/ontario-
teachers-pension-plan-votes-against-bombardier-chair.html>.Despite Bombardier receiving $372.5 million in
loans from the federal government and $1 billion in loans from the Quebec government, Bombardier proposed
to increase the pay of its top six executives by 50 percent. In response to widespread public backlash,
Bombardier delayed salary increases but nevertheless agreed to executive pay increases despite vocal
opposition from various investor rights groups. Pierre Beaudoin, who was the executive chairman at the time
13

resulting from these actions constitutes the archetypical “agency costs”35 including: monitoring
costs borne by the principal to ensure that its agents are fulfilling their responsibilities; costs
expended by the agents to provide assurance that they are doing their jobs responsibly; and the
remaining loss that arises from divergent interests.36
Agency costs can arise in firms with a single class of shares and a controlling shareholder,
provided that the controlling shareholder has a high enough percentage of the shares to
determine the outcome of any given shareholder vote.37 But these firms do not raisethe same
concerns as DCS firms, since the controlling shareholder has not undemocratically awarded itself
multiple votes per share as it has in the DCS context.38 Thus, DCS firms raise agency problems
that are different in degree and kind from those that arise in the non-DCS controlling shareholder
firm.
Ideally, boards of directors and senior management strive to find a balance between acting as
both monitors of and advisors to the managerial team.39 This monitoring function becomes
constrained in a DCS corporation since subordinate shareholders have less or no power to vote
out unaccountable or underperforming individuals; there is no built-in check on the
and remains the non-executive chairman of the board of directors, effectively controls the company through
his family’s multiple voting shares.
35
See Jensen and Meckling, supra note 30.
36
Ibid, at 308.
37
Valsan, C. (2007) “A Canadian Corporate Ownership Survey” Bishop’s University – William School of Business,
online: <https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1158544>.
38
As Ben-Amar and Andre note, a large proportion of Canadian public companies have controlling shareholders
(who are typically families) who exercise voting rights disproportionate to their fraction of cash flow rights,
which is often achieved through the use of DCS and stock pyramids. Ben-Amar, W. and P. Andre (2006)
‘Separation of Ownership from Control and Acquiring Firm Performance: The Case of Family Ownership in
Canada’ 33 Journal of Business, Finance, and Accounting 517; see also Amoako-Adu, B. and B. F. Smith (2001)
‘Dual Class Firms: Capitalization, Ownership Structure and Recapitalization Back into Single Class’ 25 Journal of
Banking and Finance 1083. Note that Anderson, Ottolenghi and Reeb point out that, at least in the U.S., the
amount of economic ownership does not significantly differ between single and multi-class controlled
companies. See Anderson, R., E. Ottolenghi and D. Reeb (2017) The Dual Class Premium: A Family Affair, Fox
School of Business Research Paper No.17-021, online:
39
Anderson, D., S. Melanson and J. Maly (2007) ‘The Evolution of Corporate Governance: Power Redistribution
Brings Boards to Life’ 15 Corporate Governance: An International Review 780; Adams, R. and D. Ferreira (2007)
“A Theory of Friendly Boards” 62 Journal of Finance 217.
14

accountability of the superior shareholders. Thus, agency costs in DCS firms can proliferate and
at least differ from agency costs that arise in traditional (i.e. non-DCS) corporate structures.40
DCS tie the hands of the subordinate shareholders by effectively preventing their vote from
mattering.
(ii) Principal Cost Theory
While agency theory points out the way in which agency costs can arise in DCS firms, “principal
cost theory” essentially functions as a defence of DCS structures and private ordering generally.
Indeed, Goshen and Squire introduce this theory as a partial defence of DCS.41 Specifically, they
argue that investors should not only be concerned with increased agency conflict costs (arising
from conflicting interests between managers and investors) but also with three other types of
cost: principal competence costs (arising from investor mistakes due to a lack of expertise); agent
competence costs (arising from honest mistakes by management); and principal conflict costs
(arising from conflicting interests among investors). These different costs arise on a spectrum of
intensity depending on how ownership and control exist in a particular firm.
Goshen and Squire focus on the potential failings of shareholders given that shareholders are not
experienced managers bound by a fiduciary duty to the corporation. But they go further and
argue that the optimal arrangement of control in a particular firm is contextual (i.e. firm-specific).
They contend that this insight explains why empirical studies (many of which are discussed in the
next section) find no consistent relationship between DCS and overall financial performance: in
some firms, the structure will be value-decreasing, while in others it will be value-increasing.42
40
Goshen, Z. and R. Squire (2017) ‘Principal Costs: A New Theory for Corporate Law and Governance’ 117 Columbia
Law Review 767, online: <http://columbialawreview.org/content/principal-costs-a-new-theory-for-corporate-
law-and-governance/>. See also Bebchuk, Kraakman, and Triantis, supra note 28.
41
Goshen and Squire, supra note 40.
42
Ibid at 805-807, 815, who argue that “The use of dual-class shares structure is a good illustration of the firm-
specific nature of corporate governance, as the structure may be well-suited to firms in complex industries such as
information technology (e.g., Google, Facebook, and LinkedIn), or to firms whose outside shareholders recognize
management’s unique skill and strategic vision (e.g., Berkshire Hathaway). It is nonetheless an extreme option on
the governance-structure menu, and it is uncommon among public firms in the United States.”
15

Consistent with this contextual view, Lin argues that DCS are useful when a company is investing
primarily in intangible assets like patents or human resources, as shareholders may be more likely
to misjudge the value of such long-term investments. DCS are less justified in firms with
predominately tangible assets, like manufacturing or mining, since information about the value
of long-term investments is more observable and verifiable to outside investors.43
(iii) Private Ordering
The idea that DCS are value-maximizing in some firms but not others has given rise to claims that
favour the use of DCS. Sharfman argues that DCS facilitate efficient private ordering by allowing
a company to structure its internal affairs according to its own particular qualities and
attributes.44 If investors do not wish to invest in such firms, they do not need to, but the fact that
IPOs with DCS are successful in attracting investments (including from sophisticated institutional
investors) means that investors see the value of DCS firms in certain contexts, such as in the
technology sector where the structure is commonly used.45
Claims regarding private ordering are in essence normative arguments. Goshen and Hamdani
argue that entrepreneurs should be permitted to contract with investors for uncontestable
control in order to pursue their idiosyncratic visions for the corporation.46 Winden contends that
the DCS structure provides the entrepreneur with maximum ability to realize her idiosyncratic
vision, which “can benefit both the entrepreneur and her investors, but as a result of the smaller
equity claim as compared to concentrated ownership it leaves investors with relatively high
43
Lin, Y. H. (2017) “Controlling Controlling-minority shareholders: Corporate Governance and Leveraged Corporate
Control” 2 Columbia Business Law Review 453, online:
<https://cblr.columbia.edu/wp-content/uploads/2017/08/1_2017.2_Lin_Final.pdf>.
44
Sharfman, B. S. (1 August 2017) “How DCS in IPOs Can Create Value” Columbia Law School’s Blog on
Corporations and the Capital Markets, online: <http://clsbluesky.law.columbia.edu/2017/08/01/how-dual-
class-shares-in-ipos-can-create-value>.
45
Ibid.
46
Goshen and Hamdani, supra note 22.
16

exposure to agency costs.”47 Sharfman argues that informed investors generally consider the
potential risks that come with having DCS (such as increased agency costs) and weigh that against
the benefits that come from having this kind of control (such as decreased principal conflict and
competency costs). The optimal governance arrangement will be one that minimizes the total
control costs.
Sharfman points out that in certain firms, especially those with purportedly "visionary leaders,"
the optimal arrangement to minimize total costs of control may well be a DCS structure. There
is a trade-off: Snap Inc.’s dual class IPO received funding from sophisticated investors (many of
whom had previously spoken out against non-voting shares) because investors determined that
the increased potential for agency costs was offset by the reduction in principal conflict and
competence costs.48 Subordinate shareholders are prepared to allow entrepreneurs maximum
ability to realize their vision, which explains why this structure is in place in many major firms
with founders as controlling shareholders, including: Snap (Evan Spiegel and Robert Murphy),
Berkshire Hathaway (Warren Buffet), Alphabet (Larry Page and Sergey Brin), Facebook (Mark
Zuckerberg), Alibaba (Jack Ma) and Comcast (Roberts family).49
(iv) Summary
These arguments favor DCS. But they are, as Bebchuk and Kastiel explain, Panglossian views.50
The arrangements that the founders adopt may not be value-maximizing at all. The DCS firms
mentioned above are highly successful firms, but the examples themselves are not evidence of a
causal link between the presence of DCS and value maximization. This is a key point: DCS firms
may do well with such a structure but, then again, they may not. The next sections of this article
47
Winden, supra note 23.
48
Sharfman, supra note 9 at 20–22. Sharfman bases his argument on Goshen and Squire’s model of optimal
corporate governance arrangements. See also McCahery, J. A. and E. P. M. Vermeulen (2016) “Venture Capital
2.0: From Venturing to Partnering” 1 Annals of Corporate Governance 95, online:
<http://dx.doi.org/10.1561/109.00000007> for a discussion about the decisions to implement DCS and other
management entrenching takeover defence mechanisms in the context of initial financing for start-up
companies.
49
Sharfman, supra note 9.
50
Bebchuk and Kastiel, supra note 11 at 36.
17

are agnostic on this issue, and seek mainly to provide an objective snapshot of DCS governance
characteristics. The article then returns to the private ordering question, arguing that certain
reforms would enhance DCS governance and accountability at least from the perspective of
subordinate shareholders.
4. Empirical Studies
The debate between proponents and opponents of DCS is at a standstill. This standstill has
occurred in part because the current literature is inconclusive about the relationship between
DCS and firm performance. This section analyzes varying viewpoints in the empirical literature
relating to DCS. A complete list of the various studies that analyze the effects of DCS on firm value
is found in Appendix 1.
A number of scholars oppose DCS and have found evidence that such structures encourage
managerial entrenchment and reduce shareholder wealth. Masulis et al., for example, find that
DCS firms pay their CEOs more than non-DCS firms and are more likely to make shareholder
value-destroying acquisitions.51 Hossain finds that DCS firms primarily undertake value-
destroying acquisitions, while also noting that long-term post-acquisition operating performance
for single class firms are significantly higher than comparable DCS counterparts.52 Along similar
lines, Smart and Zutter find that managers receive higher compensation in DCS firms and that
these firms trade at a lower price relative to earnings and sales than their single-class peers.53
These studies speak to poor governance in DCS firms. On a structural level, Li and Zaiatas argue
that there are general governance and transparency problems with DCS firms since they are
associated with poorer information environments and increased accrual-based earning
management, which is consistent with the hypothesis that managers of DCS firms are
51
Masulis, Wang, and Xie, supra note 6.
52
Hossain, A. (2014) ‘Dual v. Single Class Firms: An Acquisition Perspective’ 14 Journal of Accounting and Finance
1221, online: <http://digitalcommons.www.na-businesspress.com/JAF/HossainAT_Web14_3_.pdf>.
53
Smart, S.B. and C.J. Zutter (2003) ‘Control as a Motivation for Underpricing: A Comparison of Dual-and Single-
Class IPOs’ 69 Journal of Financial Economics, online:
<http://www.sciencedirect.com/science/article/pii/S0304405X03001090>.
18

incentivized to conceal private control benefits from outside shareholders.54 To buttress their
findings, they note that after unification (i.e. after multiple classes of voting shares are collapsed
into a single class of shares with equal voting rights), firms experience an improvement in the
information environment and a decrease in earnings manipulation.55
Quite apart from the impact of governance changes, other studies find that, on a broad level,
DCS tend to decrease firm value. Gompers et al., for example, take a comprehensive view of DCS
firms and find support for the hypothesis that voting control leads managers to underinvest, thus
reducing firm value.56 King and Santor similarly note that DCS firms decrease value, based on
their finding that family-owned firms that use DCS have valuations that are lower by 17 percent
on average relative to widely held firms, despite having similar return on assets and financial
leverage.57 Finally, in the IPO context, Smart et al. find that DCS firms trade at lower prices than
do single class firms, both at the IPO stage and for at least five years following the IPO. They also
note that general CEO turnover is sensitive to firm performance when there is a single share class,
but not for DCS firms and that, following unification, statistically and economically significant
value gains occur for the company.58
With such evidence mounted against DCS firms, it may seem surprising that these firms continue
to exist in our capital markets. While opponents of DCS may wish to paint the persistence of a
DCS structure as a mix of regulatory failure and managerial self-service, the reality is that for
virtually every study noting a problem with DCS firms, there is a study either finding a benefit or
54
Li, T. and N. Zaiata (2017) ‘Information Environment and Earning Management of Dual Class Firms Around the
World’ 74 Journal of Banking and Finance, online:
<http://www.sciencedirect.com/science/article/pii/S0378426616301583>.
55
Ibid.
56
Gompers, P.A., J. Ishii, and A. Metrick (2004) Incentives vs. Control: An Analysis of U.S. Dual-Class Companies,
National Bureau of Economic Research Working Paper No. 10240, online:
<http://www.nber.org/papers/w10240>.
57
King, M.R. and E. Santor (2008) ‘Family Values: Ownership Structure, Performance and Capital Structure of
Canadian Firms’ 32 Journal of Banking and Finance 2423, online:
<http://www.sciencedirect.com/science/article/pii/S0378426608000502>.
58
Smart, S., R.S. Thirumalai and C.J. Zutter (2008) ‘What’s in a Vote? The Short and Long-Run Impact of Dual-Class
Equity on IPO firm Values’ 45 Journal of Accounting and Economics 94, online:
<https://kelley.iu.edu/Faculty/Finance/ssmart/publications/jae2008.pdf>.
19

a neutral effect of DCS on firm value. For example, among the purported benefits of DCS, Jordan
et al. find that these firms face lower short-term market pressures than single class firms and that
DCS firms tend to have more growth opportunities (through higher sales growth and R&D
intensity).59 Similarly, Dimitrov and Jain find that DCS recapitalizations tend to enhance
shareholder value: on average, stockholders earned significant positive abnormal returns of 23
percent in a four year period following the announcement and this abnormal return was even
larger (53 percent) for DCS firms that issued equity.60If they stood on their own without any
contrary empirical evidence, these studies would be persuasive in pointing to the benefits of DCS
firms.
Other studies suggest that DCS are basically neutral in their effects. Anderson et al. find that,
after controlling for time, industry, and a wide variety of firm-specific factors, DCS alone have no
effect on outside shares.61 Similarly, Ben-Amar and André find that the separation of ownership
and control that accompanies DCS does not lead to value-destroying M&A,62 Morey finds that
multi-class shares have no statistically significant effect on long-term value creation,63 and Jog et
al find that restricted voting share structures do not lower firm value, operating performance, or
stock performance relative to non-restrictive voting share firms.64
While neutral findings regarding the effects of DCS on firm performance may not seem like a
persuasive reason to keep them, one must be cognizant of the fact that most proponents of DCS
argue that firms (and investors) should have the freedom to choose this structure if the particular
59
Jordan, B., S. Kim and M. H. Liu (2016) ‘Growth Opportunities, Short-Term Market Pressure, and Dual-Class Share
Structure’ 41 Journal of Corporate Finance 304, online:
<https://www.sciencedirect.com/science/article/pii/S0929119916301444>.
60
Dimitrov, V. and P. C. Jain (2006) ‘Recapitalization of One Class of Common Stock into Dual-Class: Growth and
Long-Run Stock Returns’ 12 Journal of Corporate Finance 342, online:
<https://www.researchgate.net/publication/223499102_Recapitalization_of_One_Class_of_Common_Stock_in
to_Dual-Class_Growth_and_Long-Run_Stock_Returns>.
61
Anderson, Ottolenghi and Reeb, supra note 38.
62
Ben-Amar and Andre, supra note 38.
63
Morey, G. (May 2017) ‘Multi-Class Stock and Firm Value’ Council of Institutional Investors, online:
<http://www.cii.org/files/publications/misc/05_10_17_dual-class_value_study.pdf>.
64
Jog, V.M., S. Dutta, and P. C. Zhu (2010) ‘Impact of Restricted Voting Share Structure on Firm Value and
Performance’ 18 Corporate Governance: An International Review 415.
20

company and situation warrant it, as DCS will be value-increasing in particular situations.65 For
example, Nuesch finds that DCS neither harms nor benefits firm performance on average, as the
effect on firm performance is context-specific. In particular, DCS increase firm performance if the
firm requires external financing, but it hinders firm performance if the firm does not.66 This is
what most DCS proponents reason: if used in the right situation, DCS have a positive effect on
firm value. Thus, the caveat emptor argument and the need for regulatory restraint are
buttressed by neutral findings of DCS on firm value and performance.
While the above studies focus on the relationship between DCS and firm value, and suggest that
the empirical scholarship is not determinative, more recent scholarship has turned to consider
the effect of DCS on innovation. Baran et al. show that the enhanced managerial entrenchment
permitted in the DCS firm fosters innovative management culture.67 Studying the effect of DCS
on innovation, output, quality and creativity, the authors find that disproportionate insider
control is positively correlated with the number of patents and citations in a firm, though this
effect decreases with firm age. Indeed, their study shows that high levels of innovation activity
“mitigate the negative effect of excess managerial entrenchment on firm value.” 68
In short, DCS may benefit young firms by engendering an innovative management culture where
executives file patents more frequently. The importance of the Baran et al. study is far-reaching:
critics have warned about the increased use of DCS in the IPO context especially in the
technology, media and telecommunications sectors because of governance concerns. But quite
apart from the counter-arguments that DCS have a positive effect on firm value, they can also
have a positive effect on innovation.69
65
See for example Sharfman, supra note 9, and Goshen and Squire, supra note 40.
66
Nuesch, S. (2016) ‘Dual-Class Shares, External Financing Needs, and Firm Performance’ 20 Journal of Management
and Governance 525, online: <https://link.springer.com/article/10.1007/s10997-015-9313-5>.
67
Baran, L., A. Forst, and M. Via (2017), Dual Class Firm Structure and Innovation, Financial Management
Association (working paper), online: <http://fmaconferences.org/Boston/DualClassInnovationFMA.pdf>. See
also, Humphery-Jenner, M. (2014) ‘Takeover Defenses, Innovation, and Value Creation: Evidence from
Acquisition Decision’ 35 Strategic Management Journal 668, although his empirical research excludes DCS.
68
Baran, et al., supra note 67, at 5.
69
Ibid at 4. See also McCahery & Vermeulen, supra note 48 on the general use of DCS in start-ups under the context
of maintaining innovation and entrepreneurial cultures during the lead-up to an IPO.
21

5. Two Case Studies
The theoretical and empirical literature reveals an intense debate around DCS. This section
examines two firms and the ways in which they have each sought to rearrange their respective
DCS structures. The first example highlights a public corporation that seeks to retain its DCS
structure while the second involves a public corporation that seeks to dismantle its DCS structure.
Taken together, these two cases provide insight into the way in which a firm can adapt its DCS
structure to the will of the founder and superior shareholder.
(i) Fairfax
The first example involves one of Canada’s most successful firms, Fairfax Financial Inc. In 2015,
Fairfax increased the voting power of its multiple voting shares, from a vote ratio of 10:1
compared to subordinate shares to a 50:1 ratio. . This change was put to shareholders for a
“majority of the minority” vote. Subordinate shareholders voted in favor of the new share
structure and specifically to preserve the current 41.8 percent voting power of its founder, Chair
and CEO Prem Watsa. They also voted in favor of a resolution that prevented Watsa from being
able to profit, or receive any premium or benefit, from the special voting rights attached to his
shares. As a result of their approval of the new structure, in the event of a change of control,
subordinate shareholders also have the right to receive the same consideration on a per share
basis as holders of the multiple voting shares. The new share structure also contained a five-year
sunset provision; in order to continue to remain in place after 2020, the share structure would
need to be ratified by a majority of the minority vote.70
70
Fairfax Financial Holdings, Fairfax Calls Special Shareholders' Meeting to Consider Amendment to Terms of Multiple
Voting Shares (12 June 2015), online: <http://www.Fairfax.ca/news/press-releases/press-release-
details/2015/Fairfax-Calls-Special-Shareholders-Meeting-to-Consider-Amendment-to-Terms-of-Multiple-Voting-
Shares/default.aspx>.
22

The modification to Fairfax’s share terms was approved at a special meeting by 69.7 percent of
the votes cast by subordinate shareholders.71 Prior to this final vote, the meeting to consider the
proposal was postponed twice, in the first instance to “allow more time for the holders of a
significant number of shares which have not been voted to vote their shares, and allow Fairfax's
management to continue ongoing discussions with shareholders concerning the proposed
amendment.”72 The second postponement occurred because of modifications to the proposal,
which Fairfax stated were “in response to discussions with certain significant institutional
shareholders of Fairfax.”73 With the further amendments, Fairfax garnered voting support from
many of the institutional shareholders and approve the proposal.74
The transaction was a response to ongoing dilution which made it impossible for Fairfax to
continue issuing subordinate voting shares while preserving Watsa's control of the company. In
the year prior to this amendment, Fairfax had explored the possibility of issuing a new class of
non-voting shares in order to manage dilution. However, the TSX advised that it would reject
Fairfax’s proposed new non-voting share class unless the class contained a coattail provision
(enabling the non-voting shares to be converted into subordinate voting shares in the event of a
take-over bid).75 Furthermore, a relevant securities commission rule (Rule 56-501) effectively
71
Fairfax, " Fairfax Announces Approval of Amendments to Multiple Voting Share Terms” (24 Sept 2015), online:
http://www.fairfax.ca/news/press-releases/press-release-details/2015/Fairfax-Announces-Approval-of-
Amendments-to-Multiple-Voting-Share-Terms/default.aspx.
72
Fairfax, “Fairfax Special Meeting Postponed to August 13, 2015” (20 July 2015), online:
http://www.fairfax.ca/news/press-releases/press-release-details/2015/Fairfax-Special-Meeting-Postponed-to-
August-13-2015/default.aspx.
73
Fairfax, "Fairfax Announces Modifications to Multiple Voting Share Proposal and Postponement of Special
Meeting of Shareholders to August 24, 2015” (11 August 2015), online: www.fairfax.ca/news/press-
releases/press-release-details/2015/Fairfax-Announces-Modifications-to-Multiple-Voting-Share-Proposal-and-
Postponement-of-Special-Meeting--of-Shareholders-to-August-24-2015/default.aspx.
74
Ibid.
75
Fairfax, "Notice of Special Meeting of Shareholders", online:
<https://www.sec.gov/Archives/edgar/data/915191/000104746915005451/a2225117zex-99_3.htm>. The TSX
advised that the new class of shares would need a provision allow the “non-voting shares were convertible into
subordinate voting shares in the event of a take-over bid for the subordinate voting shares (regardless of
whether an offer was also being made for the multiple voting shares), except in circumstances where a
concurrent offer on equivalent terms was also made for the non-voting shares. If holders of non-voting shares
were entitled to convert their shares into subordinate voting shares and tender them to a take-over bid, such a
conversion feature would create an opportunity for a third-party to make an unsolicited take-over bid for
23

requires shareholder approval for any recapitalization. Worried about the impact of a third-party
take-over bid that would trigger the conversion rights, Fairfax decided not to create a new non-
voting class of shares. Instead, it proposed the amendment to its existing share structure which
was ultimately approved by a majority of the minority shareholders. Through this amendment, it
was able to preserve the vision of the founder which was attractive to shareholders from a share
price point of view.
(ii) Magna
Our second example involves Canada’s largest auto parts manufacturer, Magna International, a
multinational public corporation founded by Frank Stronach in 1957. Prior to the transaction at
issue, Magna had a DCS structure in place, with one vote per share attaching to approximately
112 million widely-held Class A subordinate voting shares and 300 votes per share attaching to
approximately 700,000 Class B shares held by Stronach. The DCS structure enabled Stronach to
maintain control of the corporation while owning a small percentage of its shares.
Under a transaction known as a “plan of arrangement” proposed in May 2010, Magna sought to
collapse its DCS structure. The transaction, one of the largest sales of control transactions in
Canadian history, involved the purchase by Magna of all Stronach’s shares —constituting 0.6
percent of the equity but 66 percent of the voting rights—in exchange for 9 million subordinate
voting shares totaling US $300 million.76 The transaction diluted the holdings of the subordinate
shareholders and enabled Stronach to be paid a 1,800 percent premium of the trading value of
the Class B multiple voting shares.
In addition, under certain consulting contracts with the corporation, Stronach was able to receive
3 percent of Magna’s pre-tax profits. Magna was required to pay the fees owing under the
subordinate voting shares and potentially acquire shares representing a much larger number of votes than the
58.2% then represented by the outstanding subordinate voting shares.”
76
Magna International In. (Re) (2010), 34 OSC Bull 1290 at para 103 (OSC) [Full Reasons].
24

consulting contracts for a one-year term if they were terminated early. Finally, Magna and
Stronach would enter into a partnership that would be indirectly controlled by the Stronach Trust
in order to develop Magna’s electric vehicle business. The Magna board neither made a
recommendation to shareholders nor provided them with a valuation or fairness opinion.77
In response to the proposed arrangement, staff of the relevant regulator, the Ontario Securities
Commission (OSC), alleged that the transaction was contrary to the public interest for the
following reasons: first, the circular did not contain specific financial information that the special
committee of the Magna board (formed to consider the transaction) had obtained from its
financial advisors; second, the disclosure contained neither sufficient information relating to the
fairness of the transaction nor useful recommendations to shareholders; third, the transaction
was novel and unprecedented because the shareholders were asked to approve it without a
recommendation and without sufficient information; and fourth, the approval and review
process was inadequate. A group of shareholders also sought a permanent cease trade order
because the transaction, in their view, was abusive and coercive.78
The OSC held that the circular did not contain sufficient disclosure and ordered more disclosure
to be filed and approved before the transaction could proceed. The OSC ordered “a clear
articulation”79 of how management and the board arrived at the consideration to be paid to
Stronach and the potential economic benefits to shareholders. This disclosure was to include a
description of the potential alternatives considered by the special committee, a discussion of the
approval process adopted by the special committee, and a statement of how the financial
advisors assessed the transaction, providing reasons why it could not render a fairness opinion,
77
Ibid; See Magna International In. (Re) (2010), 33 OSC Bull 6013 [Initial Reasons]; See also Anand, A. (2016)
‘Offloading the Burden of Being Public: An Analysis of Multi-Voting Share Structures’ 10 Virginia Law & Business
Review, online: <https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2728481>.
78
The “Opposing Shareholders” consisted of the Ontario Teachers’ Pension Plan Board; Canada Pension Pan
Investment Board; Ontario Municipal Employees Retirement System (OMERS) Administration Corporation;
Alberta Investment Management Corporation; Letko, Brosseau & Associates; and the British Columbia
Investment Management Corporation. Initial Reasons, supra note 77.
79
Ibid at para 41.
25

among other things.80 The OSC did not issue a permanent cease trade order as staff had
requested.
The terms of the transaction also required court approval and a minority shareholder vote
regarding whether the transaction was “fair and reasonable,” in accordance with the law of
arrangements. The minority shareholders approved the transaction. However, as argued below,
the transaction presented to them was arguably abusive since they were asked to choose
between the lesser of two evils. If they voted against the transaction, they would be voting to
maintain the status quo. If they voted for the transaction, they would be approving the collapse
of the firm’s DCS structure by paying exorbitant consideration to the founder.
(iii) Summary
In both Fairfax and Magna, there were governance changes: one which retained the DCS
structure and strengthened the controlling shareholder (Fairfax), and one which removed the
DCS structure and left the controlling shareholder weakened but extremely well paid (Magna).
The two scenarios illustrate the differences among DCS firms in terms of their governance. While
they all by definition have some sort of multi-voting share structure, the minutiae of their
governance differs. It is understandable for founders of a firm, like Watsa and Stronach, to seek
to control its development via a DCS structure in pursuit of their respective “idiosyncratic
visions.” This explains why DCS have grown in popularity in recent years.81 This increase may also
be a response to the simultaneous rise of shareholder activism.82 As shareholders have sought to
exercise their statutory rights to participate in the corporation (such as in terms of board
composition), and indeed have advocated for stronger rights (such as in terms of majority voting),
founders have sought to protect themselves against a loss of control.83
80
Ibid.
81
See Anita Anand, supra note 77; Bebchuk and Kastiel, supra note 11.
82
Wong, S. C. Y. (29 January 2013), “Rethinking One Share, One Vote” Harvard Business Review, online:
<https://hbr.org/2013/01/rethinking-one-share-one-vote>.
83
See for example, Arugaslan, O., D. O. Cook and R. Kieschnick (2010) ‘On the Decision to go Public with Dual Class
Stock’ 16 Journal of Corporate Finance 170, online:
26

6. Governance Characteristics of DCS Firms
The different governance structures in the Fairfax and Magna examples discussed above invite
us to look behind DCS generally in order to analyze governance characteristics of DCS firms. How
different is the governance of DCS firms from governance of companies with a "one share, one
vote" structure? This section highlights five governance characteristics, the presence or absence
of which raises issues specific to the DCS context. These five characteristics will guide the analysis
of this study's results below. They have been chosen because of their centrality to the academic
discussion (and indeed criticisms) of DCS governance. They are as follows:84
• whether subordinate shareholders have the ability to vote on matters (such as the way in
which control is held) via a majority of the minority vote process;
• whether the DCS structure is subject to a sunset provision limiting the length of time that
the DCS will persist in the corporation;
• whether the board is populated by more independent directors than is recommended by
law;
• whether the board of the DCS corporation is led by an independent board chair; and
• whether the corporate charter contains provisions specifying change of control rights for the
subordinate shareholders.
Each of these governance characteristics is explained below. While generally speaking we may
not expect DCS firms to have these governance characteristics, we do not currently have specific
data on this issue. This is the purposes of the discussion below.
<www.sciencedirect.com/science/article/pii/S0929119909000510> who find that deviations form a one-share

one vote regime are done primarily so insiders can retain control. See also Ben-Ishai, S. and P. Puri (2006) ‘Dual
Class Shares in Canada: An Historical Analysis’ 29 Dalhousie Law Journal 117, online:
84
See Anand, Milne and Purda, supra note 24.
27

(i) Majority of the Minority Vote
Are subordinate shareholders provided with a majority of the minority vote (for example on the
election of directors, the appointment of auditors, or the approval of financial statements)?
One of the hallmarks of corporate decision-making is the ability of shareholders to participate in

the corporation by means of their vote.85 Involving shareholders in the process of developing the
DCS structure in this way may offer a host of benefits to shareholders and the firm. Partch finds
that when firms ask shareholders to approve a proposed DCS structure, managers are
transparent as to the reasons why the firm wishes to create a second class of common stock: so
that the majority shareholders can retain control while obtaining capital to finance unique
growth opportunities. This is a promise which most firms keep after the DCS structure is approved
by a majority or super majority of shareholders.86 This transparency and dialogue with
management may allow shareholders to assess the validity of the proposed business strategy and
only approve the proposals that are in their interest. Dimitrov and Jain provide empirical support
for this hypothesis by finding evidence that shareholders received significantly higher returns
after DCS recapitalizations than their competitors.87
Other scholars are concerned about the level of choice that shareholders exercise when
approving DCS recapitalization proposals, arguing that these proposals are rarely voted down
because of collective action problems.88 Individual shareholders have little incentive to do
85
Recognizing this, some securities regulators mandate that any reorganizing or reclassification of common voting
shares into restrictive voting shares be approved by a majority of the minority shareholders. See Ontario
Securities Commission, “Notice of Rule under the Securities Act Rule 56-501 Restricted Shares” (9 April 1997),
online: <http://www.osc.gov.on.ca/en/SecuritiesLaw_rule_19970411_56-501_r.jsp>.
86
Partch, M. M. (1987) ‘The Creation of a Class of Limited Voting Common Stock and Shareholder Wealth’ Journal
of Financial Economics 313, online: <www.sciencedirect.com/science/article/pii/0304405X87900432>;
Vecchialla, R., M. Prudom, and R. Hamilton III (1998) 'Exposing the Corporate Vampires: A Shareholder’s Guide
to Management Entrenchment’ 31 Long Range Planning 659, online:
<www.sciencedirect.com/science/article/pii/S0024630198000715>.
87
Dimitrov and Jain, supra note 60.
88
Shen, J. (4 November 2016) “A Comparative Analysis of Dual Class Share Structures” Columbia Law School’s Blog
on Corporations and the Capital Markets, online: <http://clsbluesky.law.columbia.edu/2016/11/04/a-
comparative-analysis-of-dual-class-share-structures/>; Arrunada, B. and C. Paz-Ares (1995) ‘The Conversion of
28

research and make an informed vote about the validity on the DCS structure because the costs
of doing so exceed the likely returns. Thus, even if the proposal would reduce their wealth, or
provide less participation in governance, individuals have little incentive to organize opposition
to the proposal because they would bear all the costs while extracting only a fraction of the
benefit.89 Arrunada and Paz-Ares argue that this creates a classic prisoner’s dilemma: individually,
the optimal strategy is to convert one’s shares into non-voting shares with higher dividends and
accept the proposal as is, even if collectively, the optimal decision for outside shareholders is not
to convert their shares or accept the proposal.90
Yet, providing voting rights to subordinate shareholders may mitigate the entrenchment that DCS
allow, especially in an age of shareholder activism. For example, the subordinate shareholders
may put in place a board of directors with a majority of independent members, which can be
effective at monitoring managers.91 Studies have demonstrated that giving all shareholders the
ability to nominate directors and vote in elections increases the quality of the board’s
independence.92 If a DCS firm allows minority shareholders to vote for directors, the board is
likely to be more independent than if the controlling shareholders retain all nomination and
voting powers.
While the data relating to the efficacy of independent boards has been debated93, it is worth
asking if a given DCS firm gives shareholders the opportunity to vote on important aspects of the
Ordinary Shares into Nonvoting Shares’ 15:4 International Review of Law and Economics 351, online:
<www.sciencedirect.com/science/article/pii/0144818895000356>.
89
Arrunada and Paz-Ares, ibid.
90
Ibid.
91
Wang, C., Xie, F. and M. Zhu (2015) ‘A Industry Expertise of Independent Directors and Board Monitoring’ 50
Journal of Financial and Quantitative Analysis 929, online: <https://www.cambridge.org/core/journals/journal-
of-financial-and-quantitative-analysis/article/industry-expertise-of-independent-directors-and-board-
monitoring/E76A14A531242C461430348A1B9DDF39>.
92
Loureiro, G. (2012) ‘Monitoring the Board: Should Shareholders have Direct Proxy Access?’ 12 Quantitative
Analysis 943, online: <https://www.cambridge.org/core/journals/journal-of-financial-and-quantitative-
analysis/article/industry-expertise-of-independent-directors-and-board-
monitoring/E76A14A531242C461430348A1B9DDF39>.
93
See generally, Adams, E. and Ferreira, D. (2007) ‘A Theory of Friendly Boards’ 62 The Journal of Finance 217,
Volonte, C. (2015) ‘Boards: Independent and committed directors?’ 41 International Review of Law and
29

company’s governance. The presence or absence of voting rights for minority shareholders will
likely affect the opportunities afforded to management to extract private benefits from the firm,
suggesting that some DCS firms may be better at protecting investor rights than others. Note,
however, that there is little empirical evidence that approval by a majority of outstanding
disinterested shares provides an important backstop in controlling shareholder freeze-outs,
94
management buyouts or conflicted mergers. In the Israeli context, Licht suggests that such
approvals are unlikely to mitigate the threat of dilution.95
(ii) Sunset provision
Is the DCS structure subject to a sunset provision?
As we have seen, one common argument in favour of DCS is that this structure allows founders
with a corporate vision and superior leadership skills to retain control of the company, leading to
benefits such as reduced agency costs and a focus on long-term growth.96 Bebchuk and Kastiel
argue that a founder’s potential for superior leadership decreases as the time from the IPO
increases.97 Thus, a case can be made for fixed-time sunset provisions in the event a DCS
structure is used with an IPO.98 This clause would be triggered at a predetermined date, such as
Economics 25, or Faleye, O. (2015) ‘The costs of a (nearly) fully independent board’ 32 Journal of Empirical
Finance 49.
94
Rock, E. (2018) MOM Approval in a World of Activity Investors, NYU Law School Working Paper No. 18-02, online:
<https://ssrn.com/abstract=3122681>.
95
Licht, A. (2018) Be Careful What You Wish For: How Progress Engendered Regression in Related Party Transaction
Regulation in Israel, European Corporate Governance Institute (ECGI) Working Paper No 382, online:
<https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3104062>. In the Canadian context, Rule 56-501 dealing
with DCS effectively requires shareholder approval for recapitalizations but that of course does not cover all
instances in which DCS firms might act and in which subordinate shareholder voting may be warranted. See OSC,
Rule 56-501, “Restricted Shares”.
96
Bebchuk and Kastiel, supra note 11.
97
Ibid at 1-4.
98
Ibid at 32. Other types of sunset provisions include “triggering event” sunset provisions which trigger a merging of
shares upon a certain event, such as the founder’s death, and “ownership-percentage” sunset provisions, which
trigger conversion into regular shares when the founders have less than a certain percentage of the total number
of outstanding stock. While Bebchuk and Kastiel acknowledge that these would be better than no sunset
provision, they argue that these are still less preferable to fixed-term sunset provisions as they can still allow a
founder to control a company for decades or even in perpetuity, one of the principle issue with DCS today that
sunset provisions are designed to rectify.
30

ten to fifteen years after the IPO, and would automatically convert all superior voting shares into
ordinary voting shares, unless extended by a majority vote of shareholders unaffiliated with the
controlling shareholder.99 A “step down” mechanism is also possible with a 10:1 voting ratio at
IPO, 9:1 a year after and so forth.
Various organizations have also advocated sunset provisions. For example, the Canadian
Coalition for Good Governance, an umbrella group consisting of the country’s largest institutional
investors, recommends that DCS remain in place for up to five years, renewable upon a majority
vote by the subordinate shareholders.100 Likewise, the Council of Institutional Investors (CII) is
strongly in favour of a “one share one vote rule,” but argues in favour of imposing sunset
provisions if a firm nevertheless issues a DCS structure in an IPO.101 In recent years, the CII has
pushed for the S&P Dow Jones Indices, MSCI Inc. and FTSE Russell to exclude firms with non-
voting stock from their indexes unless they have sunset provisions with an initial maximum of
five years, extendable with a shareholder vote to a maximum of five additional years.102
(iii) Independent directors
Does the Board comply with the suggested best practice to have a majority of directors
independent?
99
Bebchuk and Kastiel, supra note 11 at 32. Anand, supra note 6.
100
Fenton. T. A (2016) “Here to stay – Dual Class Share Structures” The Canadian Institute’s Securities Law &
Litigation, online: <www.airdberlis.com/Templates/Articles/binaryServer.ashx?bin=1000721.pdf>.
101
Bertsch, K. (9 March 2017) “Unequal Voting rights in Common Stock” (CII Remarks to Investor Advisory Committee
Council of Institutional Investors delivered at the U.S. Securities and exchange Commission) [unpublished]; “Dual-
class stock”, Council of Institutional Investors, online: <http://www.cii.org/dualclass_stock>.
102
Council of Institutional Investors. (27 April 2017) “In re: S&P Dow Jones Indices consultation with members of
the investment community on the eligibility of non-voting share classes in S&P DJI indices” Financial Post,
online: <http://www.cii.org/files/issues_and_advocacy/correspondence/2017/20170426 percent20CII
percent20comment percent20S&P percent20no percent20vote percent20share.pdf>. Bebchuk notes that firms
such as Fitbit, Groupon, Kayak and Yelp have all adopted fixed term sunset provisions and of the fifty largest
dual listed firms that went public between 2009 and 2015, twelve went public with a fixed-time sunset
provision. Comparable data is not publically available for Canadian DCS firms; this article fills a gap in this
regard. See also Daines and Klausner who state that at the inception of an IPO, provisions in a firm’s charter
provide little or no information to the market. Daines, R. and M. Klausner (2001) ‘Do IPO Charters Maximize
Firm Value? Antitakeover Protection in IPOs’ 17 Journal of Law, Economics and Organization 83.
31

The third DCS governance factor that warrants examination is whether the firm's board has a
majority of independent directors, as the law recommends.103 Independent directors have been
recognized as a key corporate governance tool to reduce agency costs by guarding against
managerial entrenchment and self-serving behavior.104 The idea that independent directors can
bring much-needed management oversight to the operations of the board has given rise to
advocates for independent boards worldwide.
Canadian guidelines initially recommended board independence (formally defined below) when
the Toronto Stock Exchange adopted the recommendations of a Committee report on corporate
governance (the Dey Committee) in 1995. In the U.S., stock exchanges altered their listing
standards to require firms to maintain a majority of independent directors after the enactment
of the Sarbanes–Oxley Act in 2002.105 Outside the North American context, international
organizations (such as the OECD) and national stock exchanges in many countries such as the
United Kingdom, Australia, Portugal, and Cyprus, have advocated board independence.106
In the years since Sarbanes-Oxley, U.S. and Canadian corporations have experienced significant
increases in board independence.107 Although this trend has been somewhat slower in other
103
National Instrument 58-101 recommends as best practice that a majority of directors are independent.
Independent is defined as “no direct or indirect material relationship”, see Disclosure of Corporate Governance
Practices, OSC NI 58-101 (2005). [NI 58-101] Corporate statutes in Canada require that at least two out of three
directors are “not officers or employees of the corporation or its affiliates”, see Canada Business Corporations
Act, RSC 1985, c C-44, s 102(2).
104
See Toronto Stock Exchange. (1994) “Where Were The Directors? Guidelines for Improved Corporate
Governance in Canada”.
105
Although the Sarbanes–Oxley Act was passed in 2002, it was not until 2003 that the Securities and Exchange
Commission approved changes to the governance requirements of the United States’ stock exchanges. Sarbanes–
Oxley Act of 2002, Pub.L. 107–204, 116 Stat. 745 (2002).
106
See OECD (2011) ‘Achieving Effective Boards: A comparative study of corporate governance frameworks and
board practices in Argentina, Brazil, Chile, Colombia, Mexico, Panama and Peru’, OECD Publishing, online:
<https://www.oecd.org/brazil/48510039.pdf>, OECD (2011) ‘Board Practices: Incentives and Governing Risks,
Corporate Governance’, OECD Publishing, online: < https://www.oecd.org/daf/ca/49081438.pdf>, or Financial
Reporting Council (2016) ‘UK Governance Code’, online: < https://www.frc.org.uk/directors/corporate-
governance-and-stewardship/uk-corporate-governance-code>.
107
See Chhaochharia, V. and Y. Grinstein (2007) ‘The Changing Structure of US Corporate Boards: 1997–2003’ 15
Corporate Governance: An International Review 1215; Lee, S. and L. Carlson (2007) ‘The Changing Board of
32

countries, the rate of adoption of requirements relating to independent boards is nevertheless
becoming significant around the world.108 The question is whether DCS firms in particular have
also adopted this governance mechanism.109
(iv) Independent chair
Is the board of the DCS firm led by an independent chair?
The fourth DCS governance characteristic that this study looks for is the presence of an
independent board chair. A study conducted by Anand et al110 found that firms with independent
boards of directors are more likely than firms without independent boards to adopt mechanisms
designed to enhance their ability to monitor firm management.111 It is therefore worth asking
whether DCS firms typically have an independent chair leading the board.
(v) Change of control provisions
Does the firm have change of control provisions in place over and above existing law?
Directors: Board Independence in S&P 500 Firms’ 11 Journal of Organizational Culture, Communication, and
Conflict 31.
108
See Brennan, N. and M. McDermott (2004) ‘Alternative Perspectives on Independence of
Directors’ 12 Corporate Governance: An International Review 325; Tsipouri, L., and M. Xanthakis (2004) 'Can
Corporate Governance Be Rated? Ideas Based on the Greek Experience’ 12 Corporate Governance: An
International Review 16; Dahya, J., O. Dimitrov, and J. McConnell (2008) ‘Dominant Shareholders, Corporate
Boards and Corporate Value: A Cross-Country Analysis’ 87 Journal of Financial Economics 73; Black, B. (2001)
‘The Corporate Governance Behaviour and Market Value of Russian Firms’ 2 Emerging Market Review 89,
online: <https://papers.ssrn.com/sol3/papers.cfm?abstract_id=263014>; Jackling, B. and S. Johl (2009) ‘Board
Structure and Firm Performance: Evidence from India’s Top Companies’ 17 Corporate Governance: An
International Review 492; Black, B., H. Jang, and W. Kim (2005) Does Corporate Governance Predict Firms’
Market Values? Evidence from Korea, European Corporate Governance Institute Working Paper No 86/2005,
online: <https://papers.ssrn.com/sol3/papers.cfm?abstract_id=311275>.
109
Dahya et al., ibid moving beyond a single country analysis, Dahya et al. examined the prevalence of independent
boards in a cross-country study of firms from twenty-two countries and showed that the pervasiveness of
adopting independent boards is truly global in nature.
110
Anand, A., F. Milne and L. Purda (2010) ‘Monitoring to Reduce Agency Costs: Examining the Behavior of
Independent and Non-Independent Boards’ 33 Seattle University of Law Review 809, online:
<digitalcommons.law.seattleu.edu/cgi/viewcontent.cgi?article=1003&context=sulr>.
111
Ibid at 835.
33

The fifth and final DCS governance characteristic examined here is the presence of a corporate
charter that contains provisions specifying change of control rights for the subordinate
shareholders. In Canada, takeover bids of DCS firms must include a provision that no offer to
acquire a controlling share would be valid without the acquirer making a concurrent offer on the
same terms to the subordinate class of shareholders (called “coattail” voting rights).112 A DCS
firm cannot have a takeover bid, a proxy contest or any other forced change without coattails.113
The question that remains is whether DCS firms have implemented other provisions regarding
changes of control: do they ensure that equivalent terms are offered to subordinate shareholders
in the case of all changes of control? Do they have specific agreements under which the founders
or superior shareholders agree not to transfer their shares, at least not without approval of the
board? Do they seek to ensure that a fair process occurs?
7. Methodology and Context
Considering these governance characteristics, we now turn to our sample set of firms which
includes all DCS firms listed on the TSX as of December 31, 2015 which consists of the 85 firms
listed in Appendix 2. The data consist of two groups of firms, the first of which consists of 70 firms
for which a DCS structure was implemented. The second group consists of 15 firms for which DCS
structures were implemented because they were de facto mandatory, usually due to legislative
requirements regarding the nationality of ownership of the firms and its shares.114 These DCS
firms are referred to as the “required by law” or RBL firms, and are further described in Appendix
4, Exhibit 1. RBL firms implement DCS structures to ensure a minimum level of Canadian
ownership is maintained in keeping with applicable legislation; they stipulate provisions by which
112
In 1987, the TSX mandated that any company issuing a class of shares with superior voting rights would have to
include a provision that no offer to acquire a class of controlling shares would be valid without the acquirer
making a concurrent offer on the same terms to the other class of shareholders. “Restricted Securities”, TSX
Company Manual at s 624.
113
DCS firms listed on the TSX prior to 1987 were “grandfathered.” See Merkley, supra note 19.
114
Certain legislation restricts the level of ownership that can be held by foreigners. The DCS automatically converts
shares between classes depending on whether the owner is Canadian or not. This allows the voting percentage
of Canadian owners to always be above the legal minimum.
34

shares are automatically converted between classes depending on whether the shareholder is
Canadian or not. They further limit the voting power of the non-Canadians such that the total
voting rights held by non-Canadians do not exceed the limit as per legislation.
As mentioned above, as of 2015, 85 out of 1487 firms listed on the TSX – roughly 5.72 percent –
had DCS.115 These DCS firms had an average market capitalization of $3.39 billion,116 while the
average market capitalization of the TSX as a whole was $1.5 billion, and the median market
capitalization of the TSX was a mere $111.9 million.117 In addition, the sector breakdown of DCS
firms on the TSX is relatively unrepresentative of the sector breakdown of the TSX itself. While
17 percent of the TSX firms generally were in the mining industry, 118 only approximately 2
percent of the DCS firms were in this industry.119 DCS firms were overrepresented by firms in the
communications and media sector (which made up 17.65 percent of the DCS firms as compared
to 2 percent on the TSX), and diversified industries (which made up 38.82 percent of the DCS
firms as compared to 13 percent on the TSX). Financial services, real estate and technology were
also overrepresented amongst the DCS firms: 14.12 percent of DCS firms were in financial
services as compared to 5 percent of the firms on the TSX, 9.41 percent were in real estate as
compared to 4 percent on the TSX, and 7.06 percent were in technology as compared to 4 percent
on the TSX.
Nevertheless, DCS firms continue to gain prominence in Canadian markets. Most recently, in
2016 and 2017, a majority of the IPOs involved DCS firms which offered subordinate voting shares
115
Toronto Stock Exchange, supra note 19; See also Merkley, supra note 19. Similarly, 589 firms listed on the TSX in
2015 were exchange-traded funds (ETFs), closed-end funds, special-purpose acquisition firms (SPACs) and other
non-traditional firms, making up 39.6 percent of the TSX. These types of firms dominated the TSX IPO market in
2015, as 76 out of the 107 new listings in 2015 were ETFs & closed-end fund IPOs, and 5 were capital pool firms
(CPCs) or special-purpose acquisition firms (SPACs). Such firms are relatively less represented amongst the DCS
firms, as only 5 out of 85 firms (5.88 percent) were SPACs.
116
Data obtained from Standard & Poor’s Capital IQ and supplemented by FactSet’s financial database.
117
Toronto Stock Exchange, supra note 19.
118
Market Intelligence Group. (December 2015) “MiG Report December 2015” Toronto Stock Exchange, online:
<https://www.tsx.com/resource/en/1265/mi-g-report-december-2015-en.pdf>.
119
Based on the TSX classifications of the 85 firms in the DCS data set.
35

carrying one vote per share, while the founders or controlling shareholders maintained control
through multiple voting shares.120 We turn now to examine the summary data with regard to the
five governance characteristics of the DCS firms listed on the TSX.
8. Results
Majority of the Minority Vote: Of the 85 DCS firms examined, only eight firms had explicit majority
of minority vote provisions and one of these firms was an RBL. Of these eight firms, seven firms
allowed for the subordinate voting shareholders to elect a particular number of directors, voting
as a separate class. Two DCS firms set out that the subordinate shareholders are entitled to vote
separately as a class regarding the auditor’s appointment. One company (i.e. Fairfax) required a
minority shareholder vote to sustain the substantial voting power of the majority shareholder.
Of the remaining 75 firms, 70 had no explicit majority of minority rule and five had provisions
enabling minority shareholders to vote in the case of qualifying acquisitions. However, the five
provisions regarding qualifying acquisitions were required by legislation (under Multilateral
Instrument 61-101), so do not indicate governance mechanisms over and above existing law.
Sunset Provision. The second governance factor this study examined was whether a company has
a sunset provision limiting the duration of the DCS structure. In total, 41 firms (48.24 percent)
had a sunset provision of some sort and the remaining 44 firms (51.76 percent) did not have a
sunset provision at all. Of the 41 firms with a sunset provision, 10 were RBL firms.
Sunset provisions, where they existed, predominantly provided that the multiple voting shares
would be converted into subordinate shares if control of the superior shares reached a certain
point of concentration. 24 firms (44.44 percent of instances with a sunset provision) contained
120
Torys LLP. (19 July 2017) “Canadian IPOs on the Rebound in 2017” Torys LLP, online:
<https://www.torys.com/insights/publications/2017/07/canadian-ipos-on-the-rebound-in-2017>.
36

such a provision. Four firms (7.41 percent) included provisions that trigger conversion of multiple
voting shares into subordinate shares in the event that a particular individual is no longer
employed at the company; five firms (9.26 percent) included sunset provisions activated upon
the closing of a qualifying acquisition; and 9 firms, all of which were RBL firms (16.67 percent),
included sunset provisions pertaining to the nationality of shareholders (i.e. Canadian ownership
must be maintained at a particular level). 12 firms (22.22 percent) had other restrictive clauses
including provisions that shares cannot be transferred to anyone other than to a permitted
person.
Table 1: Types of Sunset Provisions

Sunset Provision Based on: Number of Firms % of Total
Voting Rights or Shares 24 44.44%
Employment 4 7.41%
Qualifying Acquisition 5 9.26%
Canadian/Foreign Ownership 9 16.67%
Other 12 22.22%
TOTAL 54 100.00%
In total, 41 firms had a sunset provision, 11 of which stipulate several circumstances in which a
sunset clause is activated.121 Only one company specified a year at which the DCS structure would
cease to exist altogether, Alimentation Couche-Tard, which sets out that the DCS structure will
dissolve on the day superior shareholders reach the age of 65. Fairfax also has a sunset clause
tied to a time period, stipulating that under certain conditions, a shareholder ratification vote
will be held every five years to sustain the presence of the DCS structure.
121
For example, Danier Leather stipulates that MVS will convert into SVS in the event that these shares are
transferred to anyone other than to a permitted transferee and if Jeffrey Wortsman is no longer a senior officer of
the company, amongst other stipulations. INSCAPE Corp. has a sunset provision triggered upon the death of Madan
Bhayana. ONEX allows for the conversion of MVS into SVS at the option of founder Gerald Schwartz.
37

A comparison of RBL firms to the remaining DCS firms revealed that a substantially greater
proportion of RBL firms had a sunset provision (66.67 percent of RBL firms had a sunset provision
in comparison to 44.29 percent of DCS firms that did not require a DCS by law). How do we
explain this difference? Perhaps the RBL firms believe that they can renew the DCS structure at
the end of the period or that a DCS structure may not be perpetually required given potential
changes in the governing legal landscape.
Independent Directors. The word “independent” embodies a consideration of whether a director

is involved with management and factors such as whether the director was a participant in
management over the past three years, has a family member involved in management, or has
some other affiliation with the company. For example, a director may be non-independent if his
or her spouse is the company’s CEO even though he or she is not involved with management per
se.122
Out of the 85 DCS firms, 71 firms (83.53 percent) had a majority of independent directors on their
board, which is consistent with corporate governance comply-or-explain guidelines.123
Unsurprisingly the average proportion of independent board members was amongst the 100
largest TSX firms (a significant majority, 80 percent, of these board members were independent
in 2016). 124 An additional five firms (5.88 percent) had an equal number of independent and non-
independent directors. Nine firms (10.59 percent) did not adhere to best practice guidelines to
122
See Audit Committees, OSC NI 52-110 (2004) [NI 52-110] and as referred to by NI 58-101 supra note 103.
123
See NI 58-101, supra note 103.
124
Spencer Stuart. (2016) “Canadian Spencer Stuart Board Index 2016 Board Trends and Practices of Leading
Canadian Companies” Spencer Stuart, online:
<https://www.spencerstuart.com/~/media/pdf%20files/research%20and%20insight%20 pdfs/%20spencer-
stuart-canadian-board-index-dec-2016.pdf > examined one hundred TSX listed companies with revenue of at
least $1 billion and for which Canadian residents comprised a minimum of 25% of each of their boards as of
June 2016. The study found that 80% of these companies had a majority of independent directors in 2016,
based on the Canadian Securities Administrators definition of independence. It also states that the percentage
of independent directors of the 100 largest TSX companies have been quite consistent since 2010; Allaire, Y.
(2010) “Controlled Companies Briefing” Canadian Institute of Chartered Accountants, online:
<https://www.cpacanada.ca/-/media/site/business-and-accounting-resources/docs/controlled-companies-
briefing---questions-for-directors-to-ask.pdf> refers to a 2009 Spencer Stuart report noting that 79% of
directors for the largest 100 Canadian companies were independent in 2009.
38

have a majority of independent directors on the board. Table 2 below contains a breakdown of
board independence amongst the DCS firms examined.
Table 2: Board Independence

Board Independence Number of % of Total
Firms
Less than half of the board was independent 9 10.59%
Equal number of independent and non-independent directors
5 5.88%
(50 percent of the board was independent)
Majority of Independent Directors 71 83.53%
TOTAL 85 100%
Independent Chair. Examination of the composition of the board of directors revealed that 61
firms (71.76 percent) did not have an independent chair per se.125 However, 47 firms (55.29
percent) had a chair who was uninvolved in the management of the firm.
Table 3: Independent Chair

Independent Chair Number of % of Total
Firms
Independent chair 24 28.24%
Independent chair who is uninvolved in management 24 28.24%
Non-independent chair 61 71.76%
Non-independent chair who is uninvolved in management 23 27.06%
Non-independent chair who is involved in management 38 44.71%
TOTAL 85 100%
125
Based on the definition of “independence” set out in NI 52-110, supra note 122 and as referred to by NI 58-101,
supra note 103.
39

Of the 61 firms with a non-independent chair, 35 appointed a “lead” independent director. A lead
director chairs the meetings of independent directors. The position is sometimes seen as a
separate, independent chair-type role since the lead director is responsible for dealing with a
wide gamut of tasks, from regulatory compliance through to performance evaluation of senior
employees and directors.126
Change of Control. Only 25 firms (29.41 percent) had a change of control provision other than
the mandated coattail provision. Of these firms, ten had agreements127 in place to ensure that
equivalent terms are offered to subordinate shareholders in the case of a takeover bid. Five DCS
firms had specific agreements under which the founder or majority shareholder agreed not to
transfer its shares.
Summary. As suggested by Tables 4a and 4b below, DCS firms generally do not implement
governance practices to ensure greater accountability over and above existing law. A large
proportion of DCS firms have no majority of the minority voting provision, meaning that control
rests with the superior voting shareholders. This is unsurprising of course, given that a main
purpose of the DCS structure is to maintain control in the founder or related parties. In terms of
sunset clauses, almost half of the DCS firms have one, though a fair proportion of these firms (15
out of 75 firms or 17.65 percent) are RBL firms in which the DCS structure is present for statutory
reasons relating to maintaining Canadian control rather than maintaining control amongst
founders. Most DCS firms do not have an independent chair128 but they do have chairs that are
not involved with management. The majority of DCS firms have independent directors, which
suggests that they can be understood to be a key monitoring mechanism in DCS firms. Finally,
just under one-third of DCS firms have change of control provisions over and above existing law.
126
Financier Worldwide (September 2014) “The Role of the Lead Director” Financier Worldwide, online:
<https://www.financierworldwide.com/the-role-of-the-lead-director/#.WaQeRLGZN-U>.
127
These ten companies have explicit agreements made with the founders or majority shareholders and a trustee,
stipulating that certain actions (e.g. the transfer of shares prior to the completion of a qualifying acquisition)
require authorization by the trustee who acts on behalf of the holders of the [SVS]. See Appendix 4, Exhibit 2 for
more details.
128
Anand, Milne and Purda, supra note 110. This is consistent with Anand, Milne and Purda’s, supra note 24 finding
that approximately 41.90 percent of TSX companies had an independent chair during the period of their study.
40

41

Table 4a: Summary of DCS Governance Characteristics (All Firms)
Governance Characteristic DCS Firms With DCS Firms Without
Majority of the Minority Vote 9.41% 90.59%

Sunset Provision 48.24% 51.76%
Independent Directors 83.53% 16.47%129
Independent Chair 28.24% 71.76%
Change of Control 29.41% 70.59%
Table 4b: Summary of DCS Governance Characteristics (Excluding RBL Firms)
Governance Characteristic DCS Firms With DCS Firms Without
Majority of the Minority Vote 10.00% 90.00%

Sunset Provision 44.29% 55.71%
Independent Directors 81.43% 18.57%
Independent Chair 20.00% 80.00%
Change of Control 30.00% 70.00%
Apart from the fact that most DCS firms have a majority of independent directors on the board,
there is considerable variability in the governance of DCS firms with regard to which governance
characteristics they have implemented. For example, Akita Drilling and Lassonde Industries both
have two of the five governance characteristics, however, their governance differs greatly. Both
firms have boards that are 44.44 percent independent, but the only other governance
characteristic that Akita Drilling displays is that its chair is uninvolved with management, while
Lassonde Industries has a sunset clause provision in place with regard to its DCS structure. Of
the firms that had three to four governance characteristics in place, 15 out of 22 firms had a chair
who was uninvolved with management and 19 firms had a majority of independent directors.
However, the remaining criteria once again varied greatly. Less than half of the firms had a sunset
129
Note that this includes DCS firms with a majority of non-independent directors, and firms with an equal number
of independent and non-independent directors.
42

clause (8 firms had one type of sunset provision and 1 firm had two types of sunset provision),
only 1 company had a majority of the minority provision (the company had only one type of such
provision), 7 firms had an independent chair, 7 other firms had a lead independent director, and
a different combination of 7 firms had a change of control provision.
We should not be surprised that governance among DCS firms varies greatly; this is the case with
public firms generally. But the results suggest that law – even voluntary guidelines – can matter
to firms’ governance choices. DCS firms tend to have a majority of independent directors likely
because this is a best practices rule in Canada’s corporate governance legal regime. But other
than this one characteristic, the governance of DCS firms varies greatly. One may therefore
question whether legal protections for subordinate shareholders are warranted. It is to this
question that we now turn.
9. Directions for Regulatory Reform
There are legitimate reasons to be concerned about governance in DCS firms. DCS structures
insulate management and the board, whose members may (or may not) act in the long-term best
interests of the corporation. Insiders are subject to fewer checks on self-dealing and exercise an
unfettered ability to define executive pay, bonuses and stock option plans. Defenders of DCS
may argue that questions pertaining to DCS governance are irrelevant because the subordinate
shareholders are not “forced” to carry the risk of weak governance: they can refuse to invest in
DCS firms if they do not wish to bear the non-participatory risk inherent in these firms.
This argument ignores the relatively small size of the Canadian capital markets and the lack of
options for many large institutional investors, including Canada’s pension funds. The TSX’s
market capitalization of $2.284 trillion is miniscule compared to NYSE’s $19.3 trillion market cap.
Private ordering taken to its logical conclusion can severely limit investors’ choices especially
43

when one considers that about 6 percent of TSX firms have DCS.130 If an investor does not want
to invest in a DCS firm, its investment options are restricted. Canadian DCS firms have achieved
12 percent annualized returns over the last 10 years, a considerably higher rate than the 7.1
percent achieved by non-DCS firms.131 Some of the best market performers have been DCS firms;
CCL Industries Inc., Fairfax, and Couche-Tard, all DCS firms, have outperformed their competitors
by a considerable margin.132 In addition to the established market giants which are cross-listed
on a US exchange, such as Bombardier and Rogers, an increasing number of Canadian firms
launching IPOs have opted for DCS, such as Spin Master, Cara Operations, Shopify, Stingray
Digital, Aritzia, and Freshii.133 If investors’ forced choice is to abstain from investing in DCS firms,
they are at a significant disadvantage in the relatively small capital market that characterizes the
country.
One may argue that such unhappy investors should negotiate the “terms” of their contract prior
to buying into the initial public offering when they could seek a sunset clause and other material
governance protections.134 But as Daines and Klausner have shown, the “terms” of the IPO are
not negotiated and in fact provide little information to the market.135 In other words, private
ordering does not ensure that the subordinate shareholders’ interests are protected on an
ongoing basis. This is the starting point for the reform proposals below which are based on the
idea that ideal circumstances do not always (usually) prevail.
(i) Fixed-term Sunset Clause with Mandatory Vote
130
85 out of 1487 companies listed on the TSX as noted in Toronto Stock Exchange, supra note 19. See also Merkley,
supra note 19.
131
Ibid.
132
Ibid.
133
Castaldo, J. (10 February 2017) “Dual Class Shares are in fashion again – but they’re still a bad idea” Canadian
Business, online: <http://www.canadianbusiness.com/investing/dual-class-shares-are-in-fashion-again-but-
theyre-still-a-bad-idea>.
134
Winden, supra note 23 at 52.
135
Daines and Klausner, supra note 102.
44

Various scholars advocate the use of sunset provisions.136 The rationale is sound: subordinate
shareholders’ interests may be abused when the corporation makes consistent non value-
maximizing decisions, an incumbent’s performance declines or less talented individuals take
control of the company.137 A sunset provision would mitigate this risk by barring perpetual DCS
structures, unless subordinate shareholders agree to maintain the structure. The sunset should
be in place for a limited term unless further majority of the minority shareholder approval is
obtained for the DCS.
Even the most ardent advocates of DCS should agree to this reform proposal. After all, if DCS
allow firms to achieve long-term goals that are value-enhancing for all shareholders, superior and
subordinate shareholders alike will vote to extend the DCS structure so the talented managers
can stay in control. Sunset provisions subject to an extention vote allow firms to reap the
purported benefits of a DCS structure while preventing the entrenchment of unexceptional
management many years after the initial justification (and investor support) for the DCS structure
have dissipated. It was through such a vote, though not in the context of a sunset provision, that
Fairfax kept its founder Prem Watsa in place for a defined period of time in accordance with the
wishes of minority shareholders.
One could argue that mandatory sunsets might be unnecessary if the DCS eventually results in
an inefficient structure. The superior shareholder could then simply be bought out in a
transaction in which the surplus created by the move from a less efficient structure to a more
efficient structure is shared between the superior shareholder and the subordinate shareholders.
In other words, it may seem, as Squire argues, that private ordering can accomplish mid-course
changes in a firm's governing structure, again making mandatory rules, such as a sunset,
136
See, for example, Bebchuk and Kastiel, supra note 11 and Winden, supra note 23.
137
Chemmanur, T. and Y. Jian (2012) ‘Dual Class IPOs: A Theoretical Analysis’ 36 Corporate Journal of Banking and
Finance 306, online: <www.sciencedirect.com/science/article/pii/S0378426611002214>. Richard Squire has
mentioned in response to this argument that that founders may also make the mistake of selecting a single
class structure when they should have selected a DCS. But this argument is unpersuasive in my view given that
a single class structure does not bear the normative failings of a DCS and in particular the decoupling of equity
from control in a systematic way. See Anita Anand, supra note 77.
45

unnecessary.138
In response to this argument, one must appreciate the potential for abuse in the buyout of the
founder. In the Magna transaction discussed above, for example, the transaction presented to
the subordinate shareholders was arguably abusive.139 They were asked to choose between the
lesser of two evils. On the one hand, if they voted against the transaction, they would be signaling
their desire to maintain the status quo in which the superior shareholder and founder controlled
the company while holding 0.6 per cent of the equity. On the other hand, if they voted for the
transaction, they would approve the collapse of the firm's DCS structure and the firm's
repurchase of the equity held by a founder who had historically extracted significant private
benefits of control. The caveat, however, was that the total consideration to be paid for this
transaction to occur was exorbitant.
The choice faced by subordinate shareholders in the Magna case suggests that the greater the
private benefits that the founder extracts from a corporation with DCS, the higher the premium
that will be paid to extricate the corporation from the founder. Indeed, the board’s decision to
proceed with the transaction may heighten the incentive for a superior shareholder to increase
its private benefits in order to extort a greater premium for the collapse of a DCS structure. This
is the power of the despot that the founder can exert.140
The 1,800 per cent premium was exorbitant in comparison to both market value and the typical
price range that would be paid to eliminate a controlling shareholder.141 In the face of an
138
Richard Squire, email correspondence on file with author dated December 27, 2017.
139
I have presented a developed argument along these lines in Anand, “Was Magna in the Public Interest?”, supra
note 8 at 312.
140
Ibid.
141
See Amaoko-Adu, B, B. Smith, and V. Baulkaran (15 February 2011), “Unification of Dual-class Shares in Canada
with Clinical Case on Magna International” (Paper delivered at the Capital Markets Institute Conference on Dual-
class Shares, Rotman School of Management, University of Toronto) [unpublished], online:
<http://www.rotman.utoronto.ca/userfiles/cmi/file/Magna%20Unif2_Feb%207%20(Revised).pdf>. The paper
compares Stronach’s percentage return to the return paid to what other companies paid their controlling
shareholders who held superior voting shareholders of the 31 previous cases of unification on the TSX, only one
had a special payout to the controlling shareholder similar in magnitude to the payout to Magna’s founder.
46

astoundingly high premium, the processes that the Magna board and special committee followed
warranted close scrutiny and ultimately were inadequate.142 The special committee did not
obtain a fairness opinion and valuation and the board did not make a recommendation to
shareholders. Over and above the absence of these traditional safeguards, the special committee
concluded that the transaction had positive benefits for Magna without assuring itself that it was
positive for the subordinate shareholders.143
The question must be asked: If a board refuses to make a recommendation, and no valuation or
fairness opinion is provided, how is it that the minority shareholders are to arrive at an informed
opinion that underpins their position and ultimately their vote? Therefore, process becomes
strikingly important. Magna is an extreme example of DCS buyouts gone wrong and suggests that
sunsets are a useful accountability mechanism in addition to other reforms discussed below.
(ii) Disclosure of Shareholder Voting
When shareholders vote, they express their views; they can signal dissent or approval of the way
the firm is currently governed. Management may act on these signals even when the results are
non-binding.144 For example, the United Kingdom became the first country to require that
shareholders be given an advisory vote on board pay, which subsequently led to more significant
dialogue between shareholders and management and tangible changes to executive pay.
Specifically, after this requirement was implemented, executive compensation only rose by five
to six percent in 2006, which was a sharp reduction from the average 14 percent increases in the
preceding five years. Indeed, Iliev et al. conducted a study of 8160 firms across 43 countries and
142
See also remarks by Justice Wilton-Siegel following the Magna transaction, in Kiladze, T. (22 February 2011)
‘Magna board ‘spectacularly’ vacated field: judge’ The Globe and Mail, online:
<http://www.theglobeandmail.com/globe-investor/investment-ideas/streetwise/magna-board-spectacularly-
vacated-field-judge/article1916419>.
143
Iacobucci, E. (2011) ‘Making Sense of Magna’ 49:2 Osgoode Hall Law Review 237.
144
Sauerwald, S., Van Oosterhout, J. and M. Van Essen (2016) ‘Expressive Shareholder Democracy: A Multilevel
Study of Shareholder Dissent in 15 Western European Countries’ 53 Journal of Management Studies 520,
online: <onlinelibrary.wiley.com/doi/10.1111/joms.12171/full>.
47

found that high rates of dissenting votes from outside shareholders led to higher rates of director
turnover and more withdrawals from mergers and acquisitions deals.145 Their study was
consistent with previous empirical research indicating that high dissenting vote percentages are
often followed by corporate governance changes in the following year.146
Thus, in the context of DCS firms, requiring the separation and disclosure of subordinate and
controlling shareholder vote counts likely encourages accountability. It could place pressure on
management to change bad behavior. This requirement is similar to the “comply or explain”
approach, which similarly applies soft pressure on corporations to comply with best practices
recommendations voluntarily.147 Without such disclosure, one cannot tell how shareholders
unaffiliated with the controlling shareholder voted, potentially giving a false sense of support for
the management team. By mandating disclosure of controlling and subordinate vote results
separately, subordinate shareholders may be able to express their dissatisfaction with the status
quo and exercise the potential disciplining effects that come from expressive voting noted above.
(iii) Buyout Protections
The Magna transaction discussed above suggests that the process under which buyouts occur are
in the end substantively important to the subordinate shareholders. Ensuring that these
shareholders are adequately protected in the context of the buyout of the founder is difficult to
ensure prior to the IPO. Therefore, the buyout is an area in which private ordering should yield
to the mandate of securities regulators who, in their role as protectors of investors and the public
145
Iliev, P., K. V. Lins, D.P Miller and L. Rothl. (2015) ‘Shareholder Voting and Corporate Governance Around the
World’ 28 The Review of Financial Studies 2167, online:
<https://academic.oup.com/rfs/article/28/8/2167/1599500/Shareholder-Voting-and-Corporate-Governance-
Around>.
146
Ibid at 2168.
147
Such a disclosure requirement may have been useful at the recent Bombardier AGM, where the management
was happy to report that over 90 percent of shareholders voted to reelect the board of directors and also voted
in favor of a hotly contested executive compensation package. See Marowitz, supra note 34. Regarding comply
or explain generally, see Anand, A. (2006) ‘An Analysis of Enabling vs Mandatory Corporate Governance:
Structures Post-Sarbanes-Oxley’ 31 Delaware Journal of Corporate Law 229; and Anand, Milne and Purda, supra
note 24.
48

interest writ large, should be concerned with both process and substance in the buyout of a DCS
founder.148
Securities regulators have focused to some extent on process as being central to the interests of
subordinate shareholders in the DCS structure. Canadian law requires independent valuations
and majority of minority security holder approvals for specific types of transactions, such as
issuer bids, insider bids, business combinations and related party transactions.149 For example, in
a buyout of the founder, minority shareholders must be provided with a formal, independent
valuation of the proposed transaction at the offeror’s expense.150 Securities regulation should
specifically mandate that the board form a special committee of independent directors and that
it make a recommendation to the subordinate shareholders. It should require independent
valuations and majority of the minority approval of buyouts of the founder in a DCS firm. 151
While US securities law contains little regulation relating specifically to DCS,152 regulators in some
jurisdictions seem to be moving towards greater regulation. In Canada, the OSC recently
published guidance regarding minority shareholders in conflict of interest transactions.153 The
OSC isolates several key areas for their review of material conflict transactions, namely: timely
formation of special committees, a broad committee mandate, independence of special
committee members, over-reliance on fairness opinions, and considerations of the best interests
148
See Ontario Securities Commission, supra note 11. See also Ontario Securities Commission Rule 56-601
“Restricted Shares” which compels issues of DCS to make specific disclosure regarding characteristics of shares
in offering documents.
149
Protection of Minority Security Holders in Special Transactions, Multilateral Instrument 61-101 (MI 61-101), .
150
Ibid s.2.3.
151
The law regarding valuations has been controversial in terms of information that is disclosed to shareholders.
See Re InterOil Corporation, 2017 YKSC 16.
152
Robert J. Jackson Jr., “Perpetual Dual-Class Stock: The Case Against Corporate Royalty”, (Address delivered at
the UC Berkeley School of Law, 15 February 2018), online: <https://www.sec.gov/news/speech/perpetual-
dual-class-stock-case-against-corporate-royalty#_ftn19>.
153
Canadian Securities Administrators, Multilateral CSA Staff Notice 61-302 Staff Review and Commentary on
Multilateral Instrument 61-101 Protection of Minority Security Holders in Special Transactions (27 July 2017),
online: https://www.osc.gov.on.ca/documents/en/Securities-Category6/csa_20170727_61-302_sn-staff-
review.pdf.
49

of both the corporation and its minority shareholders.154 These are much-needed protections;
one should therefore question why the regulatory approach is in the form of “guidance” as
opposed to mandatory law.
10. Conclusion
This article fills a conspicuous gap in the scholarly literature first, by comprehensively addressing
a vast body of literature relating to DCS; and second, by providing empirical data regarding the
governance of DCS firms beyond the presence of sunrise and sunset provisions. The summary
data suggest that the governance of DCS firms is highly variable. In particular, a large proportion
of DCS firms have no majority of the minority voting provisions and no independent board chair.
By contrast, almost half of the DCS firms have a sunset clause and a majority of independent
directors, the latter of which is contained in best practice guidelines for all public firms. Finally,
just under one-third of DCS firms have change of control provisions over and above existing law.
These data invite further analysis about the governance of DCS structures in the context of the
associated regulatory regime, a regime that is mandated to protect investors’ interests which
obviously includes subordinate shareholders in the DCS context. Mandatory sunset provisions,
disclosure relating to shareholder votes, and buyout protections would at least partially address
governance weaknesses inherent in DCS firms.
154
Gilbert, M. B., P. Olasker, V. A. Mercer and G. L. Rawle (31 July 2017) “Boards Beware: Regulators Actively
Monitoring Related Party Transactions” Davies LLP, online:
<https://www.dwpv.com/en/Insights/Publications/2017/Boards-Beware-Regulators-Actively-Monitoring-
Related-Party-Transactions>.
50

APPENDIX 1
Scholarly Research on Dual Class Shares Structures
This Appendix sets forth five tables as follows: empirical studies that find that DCS are
detrimental to investors or firm value; studies that find that DCS have a positive or neutral effect
on firm value; non empirical studies relating to DCS; governance and other articles; and
government sources cited in this paper.
Table of Studies Finding DCS are Harmful to Investors/Firm Value
Author(s) Sample Data Source Main Findings
Li and Zaiats 12672 firms Datastream, DCS firms are associated with poorer information
(2017) from 19 Worldscope, environments and increase accrual-based earning
countries International management, which is consistent with the notion that
between 1994 I/B/E/S, TRTH managers of dual class firms follow incentives to conceal
and 2010 (managed by the private control benefits from outside shareholders. They
Securities Industry also find that DCS structures weaken the mitigating impact
Research Center of of investor protection on earnings management and that,
Asia-Pacific following unification, firms experience an improvement in
(SIRCA), Djankov information environment and a decrease in earning
et al. (2008), La manipulation.
Porta et al. (1998),
Doidge (2004),
Hossain Corporate Compustat, CRSP Hossain finds that single class firms experience higher
(2014) takeovers abnormal returns around acquisition announcements and
between 1996 that DCS firms primarily undertake value-destroying
and 2009 acquisitions. Long-term post-acquisition operating
(12,404 performances for single class firms are also found to be
transactions in significantly higher. Overall, the results indicate that there
total) is an agency issue inherent within a dual class share
structure.
Amoako-Adu, 32 unifications TSX Authors find that firms that eliminate DCS structures
Smith, of DCS on TSX experience increased stock performance via the
Baulkaran from 1989- opportunity to restructure the board.
(2011) 2010.
51

Arugaslan, 6600 IPOs Thomson's New Authors find that managers do not prefer DCS when
Cook, and between Issues investing in long-term projects; rather, it is done mainly to
Kieschnic January 1, retain control of the company. Insiders do this to diversify
(2010) 1980 and their own portfolios and maintain control.
December 31,
2008.
Masulis, U.S Dual Class GIM, Compustat, As divergence widens between insider voting and cash flow
Wang & Xie Firms from and CRSP rights, corporate cash holdings are worth less to outside
(2009) 1994 to 2002. shareholders, CEOs receive higher compensation,
managers make shareholder value-destroying acquisitions
more often, and capital expenditures contribute less to
shareholder value.
King and 613 Canadian SEDAR, Statistics Authors find that family-owned firms that use DCS have
Santor (2008) firms from Canada valuations that are lower by 17% on average relative to
1998 to 2005 InterCorporate widely held firms, despite having similar ROA and financial
Ownership Data, leverage
Financial Post Top
500
Smart, 2622 IPOs Disclosure New DCS firms trade at lower prices than do single class firms,
Thirumalai, (including 253 Issues (Disclosure both at the IPO and for at least the subsequent five years.
and Zutter dual class Inc.) They also find that general CEO turnover is sensitive to firm
(2008) issues) performance for single but not dual class firms. They
further note that when dual class firms unify their share
classes, statistically and economically significant value
gains occur.
Dyck and 393 control SDC international Authors find that higher benefits of control are associated
Zingales transactions mergers and with more concentrated ownership. However, public
(2004) between 1990 acquisitions pressures can reduce private benefits of control though this
and 2000 in 39 database may not necessarily translate into increased shareholder
countries value.
Gompers, Ishii, Dual Class Securities Data Authors find that the relationship of firm value to cash flow
and Metrick firms listed in Company (SDC), rights is positive and concave and the relationship to voting
(2004) Center for rights is negative and convex. Overall, the evidence is
52

the United Research in consistent with an entrenchment effect of voting control
States Security Prices that leads managers to underinvestment and an incentive
(CRSP), and the effect of cash flow ownership that induces managers to
Investor pursue aggressive strategies.
Responsibility
Research Center
Smart and 2622 IPOs (253 Disclosure New Authors find that DCS firms trade at lower prices relative to
Zutter (2003) dual class Issues (Disclosure earnings and sales than single class IPOs. This pricing
issues) Inc.) difference, combined with a finding that managers earn
higher compensation, suggests that dual class ownership
structures protect private control benefits.
Amoako-Adu 32 unifications Authors find that there is some value to DCS structures, and
and Smith of dual class suggest that DCS should be permitted but closely-
(2001) companies monitored to prevent agency problems, based on their
from 1989 to observation of many DCS and tightly-controlled firms.
2010
Table of Studies Finding Benefits or no Impact of DCS Firms on Firm Performance or Investor
Interests

Anderson, 2379 firms (or Russell 3000, Authors find that adopting a buy-and-hold strategy of DCS
Ottolenghi, 24,724 firm- CompuStat, Center family firms earns excess returns of 350 basis points per
and Reeb year on Research in year relative to the benchmark of single class nonfamily
(2017) observations) Securities Prices firms. They argue that “after controlling for time, industry
spanning from (CRSP) and a wide variety of firm-specific factors, our analysis
2001 through does not lend support to the notion that DCS harm outside
2015 investors. Rather, DCS have no effect on outside investors.
When family shareholders own the superior shares,
however, they find that the subordinate shareholders earn
a return premium on their investment. They suggest that
family control plays a key role in the effect of DCS on
minority shareholders.
Gabriel Morey 1762 Russell FactSet and SEC data Author finds that DCS neither increase nor decrease a
(Council of 3000 firms company’s annualized return on invested capital (ROIC).
Institutional
Investors)
(2017)
Winden (2017) 123 U.S. public Institutional Author finds that entrepreneurs and founders use DCS to
firms Shareholder Services pursue long-term objectives and gain immunity from
short-term shareholder whims. However, sunset clauses
53

(ISS), company and other protections for shareholders should be included
charters in the to ensure managerial accountability and offer
Takeover Defense shareholders a voice in governance.
database of
SharkRepellant.net
provided by FactSet
Jordan, Kim, Dual class and Compustat, Gompers Authors find that DCS firms face lower short-term market
and Liu (2016) single-class et al. (2010) and pressure than single-class firms. They also find that DCS
firms from Smart and Zutter firms tend to have more growth opportunities (higher
1994 to 2011 (2003) and sales growth and R&D intensity) and DCS increase the
supplement the market valuation of high growth firms.
sample by hand-
collecting dual-class
firms
Nuesch (2016) Yearly Swiss Stock Guide, Author finds that DCS neither harm nor benefit firm
observations Thomas Reuters performance on average. DCS increase firm performance
from publicly Datastream. if the firm requires external finance and decrease firm
listed Swiss performance if the firm does not require external finance.
firms from
1990 to 1999
Baran, Forst, Dual class Center for Research Authors find that the enhanced managerial entrenchment
and Via shares in the in Security Prices at permitted in the DCS firm fosters innovative management
(2014) USA from the University of culture. Authors find that there is a "positive relationship"
between 2000- Chicago (CRSP), between the "wedge" (i.e. the differential between insider
2008. Excludes Compustat, National voting rights and cash flow rights compared against non-
foreign and Bureau of Economic insider shareholders) and both patents and citations. They
financial firms. Research (NBER) contend that this "benefits shareholders as excess
patent database, and managerial entrenchment increases."
EDGAR.
Humphery- 3935 SDC platinum and Author finds that hard-to-value firms that are entrenched
Jenner (2014) acquisitions Center for Research make acquisitions that generate more shareholder wealth
that are made in Security Prices at and are more likely to increase corporate innovation. This
between 1990 the University of entrenchment is a function of anti-takeover provisions
and 2005 and Chicago (CRSP) which can be a function of DCS structures.
are by
companies
listed in the
US.
Spizzirri and 435 firms for S&P/TSX Composite Authors find that Canadian family-controlled firms
Fullbrook all or part of Index (TSX Index) benefit minority shareholders significantly despite
(2013) the period contrary academic research.
between 2002
and 2012.
54

Jog, Zhu, and All Canadian Stock Guide Authors do not find that restrictive voting share (RVS)
Dutta (2010) publicly listed database, TSX Month structures lower firm value, operating performance, or
firms from Review stock performance relative to non-RVS firms in Canada.
1996 to 2005 Also, they do not find evidence of shareholder value
(10,366 firm- expropriation in key financial decisions, such as mergers
year and acquisitions and dividend payments.
observations)
Valsan (2007) 1,452 firms in Toronto Stock Author finds that Canada's large geographic expanse and
the spring of Exchange its demographic blocs create conditions which favour
2007 tightly-controlled family firms. This trend is shifting
towards wider shareholding to the benefit of minority
shareholders while reducing agency costs.
Ben-Amar and 327 Canadian Thomson Financial Authors do not find that separation of ownership and
André (2006) Transactions Securities Data’s SDC control leads to value destroying M&A.
from 1998- Platinum Worldwide
2002 Mergers &
Acquisitions
Database
Dimitrov and 178 firms that Center for Research Authors find DCS recapitalizations to be shareholder
Jain (2006) changed from in Security Prices at value-enhancing initiatives. In particular, DCS
a one-share the University of recapitalizing firms grew faster than firms in a control
one-vote firm Chicago (CRSP), Dow group and the growth was beneficial to shareholders; the
into a dual Jones News Wire, stockholders, on average, earned significant positive
class company Securities and abnormal returns of 23.11% in a period of 4 years
between 1979 Exchange following the announcement. Abnormal returns of 52.61%
and 1998 Commission were observed for dual-class firms that issued equity.
Cronqvist 309 Swedish Swedish Securities Authors find that examples of Swedish CMS/DCS
and Nilsson Firms from Register Centre structures have a lower return on assets and command a
(2003) 1991-1997. lower share price (owing to the agency costs associated
with such structures). However, such structures may be
beneficial to society by allowing founders to seek out
longer-term investment returns.
Daines and 310 US Firms Securities Data Authors argue that there is no strong evidence to suggest
Klausner (106 firms Corporation, Pratt’s that anti-takeover provisions (including DCS) post-IPO
(2001) with venture Guide to Venture supports the private benefit hypothesis. There may be
capital Capital Sources, idiosyncratic benefits which managers seek to preserve.
investment, 91 LEXIS/NEXIS M&A Ultimately, anti-takeover provision needs to be studied
with LBO further to yield more satisfactory explanations as to their
specialist role in firms during IPOs.
investment,
and 113
others) that
went public
between
January 1
55

1994 – July 1
1997
Bergstrom and 45 US firms Swedish Stock Authors study Swedish DCS firms and find that there is no
Rydqvist Exchange evidence of wealth expropriation by controlling
(1990) shareholders; but cannot definitively say that this
expropriation does not happen.
Partch (1987) 44 firms that S&P Security Author finds that firms that create DCS structures do not
created MVS Owner's Stock Guide affect shareholder wealth.
from 1963- and Moody's
1985 Manuals
DeAngelo and 45 US firms Standard and Poor's Authors find that DCS firms can be viewed as an
DeAngelo Security Owners' intermediate organization between widely dispersed
(1985) Stock Guide, The publicly owned corporation and a closely-held firm.
Bond Quotation Authors find that DCS structures value managerial vote
Record, ISL Daily control though authors do not specifically elaborate on the
Stock Price Guide beneficial or negative consequences of such structures.
Table of Theoretical Studies of DCS Firms on Firm Performance or Investor Interests
Bebchuk and Various other Authors argue that DCS have detrimental effects which may
Kastiel (2017) surveys of DCS distort and misalign managerial interests against those with
in academic public investors. These problems are more pronounced as
papers and longer time passes from the IPO. Authors propose remedy
case studies of finite DCS duration via sunset clauses to reduce risks of
featuring perpetual DCS structures.
SNAP.
Goshen and Select Authors argue that firms are best suited to consider the
Squire (2017) company cases context in designing an optimal governance structure. This
like Facebook is more beneficial than having mandatory structures for
and Google certain firms under what the authors call the "Principal Cost
Theory."
56

Lin (2017) Literature Author argues that DCS structures should not be banned
review of past entirely, but they should be rolled back to re-empower
empirical shareholders with the "one share, one vote" principle.
studies on
DCS.
Nicholas and Case studies Authors argue that DCS structures create possibilities of
Marsh (2017) from examples managerial misconduct. While institutional investors may
like SNAP and be amenable to longer-term gains flowing from long-term
Hollinger managerial protection desired by DCS structures, authors
recommend such investors still remain vigilant to the
dangers of management misconduct.
Sharfman Case study on Various news Author argues that DCS structures are socially beneficial.
(2017) SNAP IPO sources They enable firms to remain innovative and competitive
and against the whims of less competent shareholders.
Mandatory provisions against DCS (e.g. overly restrictive
sunset provisions) are detrimental to creation of value-
maximizing companies like Snap, Alphabet, Berkshire
Hathaway, etc.
Sharfman Author argues that DCS offer great value for firms entering
(2017) into IPOs. This method of private ordering should be
supported with its costs (and benefits) of this structure
already factored into the market price of the offered
shares.
Anand (2016) Author argues that DCS and MVS structures undermine
minority shareholders' rights while giving holders of
superior voting shares incentive to self-enrich. DCS/MVS
structures should be reformed to rebalance the currently
disproportionate risks that minority shareholders carry in
relation to those with greater voting rights.
Goshen and Various Authors challenge idea that DCS and other methods of
Hamdani academic maintaining managerial control as wrong-headed. Rather,
(2016) journal articles the authors contend that greater managerial control comes
and case law. from entrepreneurial/founders' desire to pursue their
business vision.
McCahery and Authors find that small/medium enterprises adopt various

Veremeulen anti-takeover provisions including DCS structures during
(2016) the lead-up to an IPO for fear of losing managerial control
over their company and losing their culture of innovation
and entrepreneurship.
57

Shen (2016) Author argues that DCS structures should not be
prohibited, but there is a risk that shareholder approval (or
disapproval) of such DCS structures gives rise to collective
action problems which may reduce shareholder oversight
efficacy.
Wang, Xie, and Authors find that from a governance perspective,

Zhu (2015) independent board members provide greater monitoring
over management. Prior industry experience enhances this
independent board member oversight capacity.
Wen (2014) Case study Various news Author argues that DCS are "on balance" negative for
focusing on sources (WSJ, NY shareholders. Citing case studies of Google, Facebook,
Google, Times, WSJ, Readers Digest, News Corp, Magna, and other examples,
Facebook, Bloomberg), author contends that DCS structures will limit benefits to
Readers company IPO the private controlling shareholders while neglecting
Digest, News registration broader shareholder benefits.
Corp statements with
the SEC.
Wong (2013) Author argues that all shareholders should be given some
voice in the Anglo-American "one share, one vote" spirit of
corporate structuring and to prevent managerial self-
entrenchment.
Chemmanur Authors Authors find that while there may be optimal scenarios for
and Jiao (2012) models both DCS and single-share structures, DCS structures create
risks where entrenched managers will not deliver as strong
a return as argued by DCS proponents.
Loureiro Author's own Author argues that shareholder proxy access enhances
(2012) model opportunities to elect independent board members. In
turn, this enhances greater governance over corporate
management.
Ferreira and Various Authors conduct an empirical literature review of DCS

Adams (2008) academic studies and find a fairly equal number of empirical papers
studies on DCS which favour and disregard DCS structures. Generally, DCS
structures are prominent in the US and somewhat in the
UK, but is decreasing in popularity in Europe.
58

Hu and Black Various Authors argue that phenomenon of worldwide equity
(2008) company case decoupling of economic and voting rights has led to a need
studies and US to regulate such equity structure-altering decisions. The
case law. disconnect between the benefits of decoupling (from a
financial and managerial control stance) versus the
corporate and legal governance paradigm (which presumes
coupled voting and economic rights for shareholders)
require regulatory responses to the risks of decoupling.
Examples of such risks of non-regulation include the then
unknown consequences of decoupling in the Great
Financial Crisis.
Adams and Authors own Authors’ model shows that there are situations where
Ferreira (2007) models policies that enhance board independence may be
detrimental for shareholders in a sole board system, but
not for shareholders in a dual board system.
Khalil and Authors argue that DCS offer comparable long-term returns
Magnan to investors compared against single-class shares. Similarly,
(2007) governance may only be one aspect of an investor’s
decision-making criteria.
Ben-Ishai and Authors argue that historical Canadian factors such as

Puri (2006) nationalist corporate policy enabled Canadian firms to have
a higher DCS presence. This phenomenon remains active
today. Non-DCS shareholders have pushed back against
DCS and the problems of permanent self-entrenchment
and managerial control by reducing investments.
Vecchialla, Authors argue that self-entrenching managers are

Prudom and "corporate vampires" that erode shareholder value. DCS
Hamilton III structures and other managerial protection avenues should
(1998) be regarded by shareholders with suspicion.
Arrunada and Authors argue that the process of conversion to DCS/MVS

Paz-Ares structures has been problematic for shareholders. These
(1995) situations create Prisoner's Dilemma for shareholders who
59

may be better off cooperating and rejecting conversion
even if the conditions of the conversion will incentive
individual acceptance of the offer to convert.
Harris and Authors' own Authors set forth conditions where one-share/one-vote
Raviv (1988) generated models are socially optimal. Symmetrical treatment of
models incumbents and rivals helps create better management.
This system is a check against bad management seeking
solely to self-entrench
Table of Studies Dealing with Agency Theory and Governance

1. Sauerwald, 12,513 ISS Global Meetings Authors find that shareholders voicing dissent are
indicative of concern for corporate governance. Voting
Van Oosterhout, proposals
voted on in is therefore an expressive act.
and Van Essen 717 firms
(2016) listed in 15
Western
European
countries
Iliev, Lins, Miller, 8,160 firms Form N-PX reports Authors find that the shareholder voting process is an
and Roth (2015) across 43 important mechanism by which corporate governance
countries is exercised around the world. Similarly, institutional
reforms which enhance voting should be welcomed
and likely offers value. Laws which encourage such
reforms and protect voting should be welcomed.
Anand, Milne, and Authors find that voluntary corporate governance

Purda (2012) practices have increased in Canada and US standards.
Authors also find that when firms are given freedom of
choice under a best practice regime, they may choose
(sometimes more stringent) governance practices
contained in the legal regime of another country.
Anand (2011) Case Study on Case Law Author finds that the OSC’s failure to contemplate the
Magna injuries to minority shareholders in the Magna
transactions means that the OSC did not fully consider
the public interest. The author argues that the public
interest consideration is vital to the role of a securities
regulator.
Iacobucci (2011) Case Study on Case Law Author finds that while many governance issues
Magna existed in terms of Magna’s special committee not
identifying plausible benefits and beneficiaries of the
60

DCS structure change at Magna, the court had
considered broad evidence to approve the
arrangement.
Anand, Milne, and 1200 Firms TSX and TSX/S&P Authors find that voluntary board adoption of
Purda (2010) from 1993 - indices listing for governance mechanisms increase in both boards
2003 firms comprising more independent directors and boards
comprising more non-independent directors.
However, independent boards place special
importance on maintaining board committees staffed
exclusively with independent directors and that their
ability to voluntarily adopt other monitoring
mechanisms is sensitive to the presence of a
controlling shareholder.
Jackling and Johl The authors' findings suggest that a larger board size
(2009) has a positive impact on performance. Similarly, a
board's greater exposure to the external environment
improves access to various resources and thus
positively impacts performance.
Dahya, Dimitrov 799 firms Credit Lyonnais Authors find that there is a positive relation between
and McConnell from 22 Securities Asia board independence and corporate value. This
(2008) countries (CLSA) corporate suggests that independent governance adds value to
governance scores shareholders.
and the Standard
& Poor’s (S&P)
transparency
rankings
Anderson, Survey of 658 Authors find that directors are seeking to balance roles
Melanson, and directors from as corporate monitor of management and
Maly (2007) Canada, collaborators with management. This is generally
Australia, New positive for governance.
Zealand, and
the USA
The Changing S&P 500, Investor Authors find that general trends in directorships moves
Board of Directors Midcap 400, Responsibility towards greater independence during sample period.
and Grinstein SmallCap 600 Research Centre Such a trend is found in all sizes of corporate
(2007) (IRRC) capitalizations.
Finegold, Benson, 105 studies Authors summarize various academic studies dealing
and Hecht (2007) featuring with corporate performance post-SOX and NASDAQ /
1) statistical NYSE reforms. Generally, the broad array of papers
tests of the show different perspectives of corporate governance.
61

relationship
between
corporate
boards of
company
performance;
2), were
published in
the main peer-
reviewed
journals that
deal with
corporate
governance,
and
3) appeared
after 1989
Lee and Carlson The authors have found that post-SOX board
(2007) membership has increased in independence and in
board level. This has increased efficiency in
management oversight.
Anand (2006) Firms should voluntarily adopt corporate governance

practices since they offer net benefits on a cost-benefit
analysis. Author finds that the enabling regime offers
greater benefits in terms of governance.
Black, Jang, and 515 Korean 2001 Korea Stock Authors find that Korean firms with 50% outside
Kim (2005) companies Exchange survey directors have 0.13 higher Tobin's q (roughly 40%
higher share price). Therefore, evidence supports view
that independent directors offer greater benefits to
shareholders.
Brennan and Authors find that Irish companies have increased the
McDermott (2004) trend towards board independence despite the
different criteria which may define "independence"
Tsipouri and Authors find that generally Greek firms comply with
Xanthakis (2004) OECD criteria for international best practices for good
corporate governance.
Black (2001) 21 Russian Brunswick Author tentatively finds that Russian firms benefitted
Firms Warburg (Russian from governance changes despite the weaker level of
investment bank). legal and cultural emphasis on governance.
62

Bebchuk, Authors' own Authors discuss three basic forms of minority
Kraakman and models shareholder control (stock pyramids, DCS, and
Triantis (1999) cross-ownership structures). Authors contend that
any of these forms (or hybrid combinations thereof)
create skewed incentives for controlling-minority
shareholders (CMS). This impacts the scope and degree
of control over that particular project choice. Authors'
model shows that agency costs increase more rapidly
as the fraction of equity cash-flow rights for CMS
decrease. Essentially, CMS structures more likely arise
as the possibility of private benefits for CMS arises.
Shleifer and Vishny Authors' own The authors find that firms with a significant number
(1986) models of small shareholders imply that the single small
shareholder has neither the power nor the incentive to
devote significant resources to monitor management’s
behavior and undertake corrective action when
needed.
Jensen and Authors' classical paper discusses the characteristics

Meckling (1976) related to agency costs. Authors find that divergent
interests between management and shareholders can
lead to agency costs, i.e. costs that shareholders incur
to ensure that directors and managers do not place
their own interests above the corporation’s.
Table of Government Sources

Source Summary
Canadian Securities CSA commentary on MI 61-101. CSA reaffirms goal of MI 61-101 is that security
Administrators, Multilateral holders be treated in a manner that is fair and that is perceived to be fair.
CSA Staff Notice 61-302 Staff
Review and Commentary on MI 61-101 implements these principles through procedural protections for minority
Multilateral Instrument 61-101 security holders that include formal valuations, enhanced disclosure, and approval
Protection of Minority Security by a majority of minority security holders. MI 61-101 also mandates the
Holders in Special Transactions involvement of a special committee of independent directors in specific
(27 July 2017) circumstances and 61-101CP recommends their use in all material conflict of
interest transactions.
Ontario Securities Commission, Regulators want to encourage reporting issuers to establish and maintain strong,
Audit Committees, OSC NI 52- effective and independent audit committees. NI requires that audit committees also
110 (2011) be responsible for managing, on behalf of the shareholders, the relationship
between the issuer and the external auditors.
Ontario Securities Commission, OSC and AMF expect that all security holders be treated in a manner that is fair and
MI 61-101 Protection of that is perceived to be fair in matters of insider bids, issuer bids, business
Minority Security Holders in combinations and related-party transactions. Document clarifies disclosure
63

Special Transactions and requirements, director duties, and minority approval processes during these
Companion Policy 61-101CP transactions.
Protection of Minority Security
Holders in Special Transactions
(1 February 2008)
Ontario Securities Commission, NI requiring mandatory disclosure for independent directors and non-independent
Disclosure of Corporate directors, attendance record, residency, compensation, nomination, and other
Governance Practices, OSC NI criteria.
58-101 (2005)
Ontario Securities Commission, OSC notifies marketplace about the implementation of Rule 56-501 concerning
“Notice of Rule under the Restricted Shares. Substantively, the rule meant that an issuer shall not file a
Securities Act Rule 56-501 prospectus, and prospectus exemptions are not available, in respect of a stock
Restricted Shares” (9 April distribution (as defined in the Rule), unless either the stock distribution or
1997) reorganization (as defined) that resulted in the creation of the restricted shares
received minority approval.
APPENDIX 2
DCS Firms
The dataset consists of the following DCS Firms listed on the TSX as of December 31, 2015
Acasta Enterprises Corby Spirit and Wine Madison Pacific Properties

ADF Group Corus Entertainment MDC Partners
AGF Management Cymbria Corp Molson Coors Canada
Air Canada Danier Leather Newfoundland Capital Corp
Akita Drilling DHX Media Onex
Alignvest Acquisition Corp Dorel Industries Ovivo Inc
Alimentation Couche-Tard Dream Unlimited Postmedia Network Canada Corp
Andrew Peller Dundee Acquisition Ltd Power Corporation of Canada
ATCO Dundee Corp Quebecor Inc
Becker Milk Empire Company Reitmans
Bombardier Exfo Rogers Communications
Brampton Brick Fairfax Financial Holdings Ltd Shaw Communications
Brookfield Asset Management Fairfax India Holdings Corp Shopify Inc
Brookfield Real Estate Services Fiera Capital Corp Sirius XM Canada Holdings
Brookfield Renewable Energy Partners
FirstService Corp Smart Real Estate Investment Trust
LP
BRP Inc GDI Integrated Facility Services Spin Master Corp
64

Can-Financials Income Corp Gibraltar Growth Corp Stingray Digital Group
Canadian Tire Guardian Capital Group Ltd Teck Resources Ltd
Canadian Utilities GVIC Communications Torstar Corp
Cara Operations Hammond Manufacturing Transat A.T. Inc
Cargojet Inc Hammond Power Solutions Transcontinental
CCL Industries HNZ Group Trimetals Mining
Celestica INFOR Acquisition Corp TVA Group
Central Fund of Canada Information Services Corp Urbana Corp
CGI Group INSCAPE Corp Velan Inc
Chorus Aviation Jean Coutu Group Westjet Airlines
Cogeco Lassonde Industries Wilmington Capital Management
Cogeco Communications Le Chateau Inc.
Colliers International Logistec Corp
Appendix 3
Description of Governance Characteristics
Criteria/Variables Description
Governance Characteristics
Governance Characteristic 1: Majority of the Each firm was examined to determine whether
Minority Vote subordinate shareholders have the ability to vote
on matters such as the way in which control is
held via a majority of the minority vote process.
It was determined whether a majority of the
subordinate shareholders’ vote is required for
any items, except if the company dissolves or if
required by law, and was sorted into the
following categories: election of directors,
appointment of auditors, approval of financial
statements, and other.
Source: SEDAR
Governance Characteristic 2: Sunset Each firm was examined to determine whether it
Provision is subject to a sunset provision limiting the length
that the DCS will exist in the corporation. Sunset
clause provisions are sorted into the following
65

categories: voting rights, employment, qualifying
acquisition, foreign ownership, and other.
Source: SEDAR
Governance Characteristic 3: Independent Each firm was examined to determine whether
Directors its board is populated by more independent
directors than is recommended by law
(majority). It was determined whether the board
has a majority of independent directors, an even
number of independent and non-independent
directors, or majority of non-independent
directors in 2015.
Source: SEDAR
Percentage of Board Independence For each firm, a percentage of board
independence was calculated by dividing the
number of independent board members by the
total number of board members.
Source: SEDAR
Governance Characteristic 4: Independent Each firm was examined to determine whether
Chair the board of the DCS corporation is led by an
independent board chair. Independence was
determined based on the definition set out in
National Instrument 52-110 Audit Committees
and as referred to by National Instrument 58-101
Disclosure of Corporate Governance Practices.
Source: SEDAR, Standard & Poor’s Capital IQ

Lead Independent Director Each firm was examined to determine whether it
appointed a lead independent director on the
board.
A lead independent director is typically
appointed to chair meetings of independent
directors and help provide oversight of the
board.
Source: SEDAR
Chair of the Board is Uninvolved with Each firm was examined to determine whether
Management or not its chair of the board of directors is
involved in management of the company.
Source: SEDAR
66

Governance Characteristic 5: Change of Each firm was examined to determine whether
Control Provisions the corporate charter contains provisions
specifying change of control rights for the
subordinate shareholders in addition to the
mandatory TSX coattail provision.
Source: SEDAR
APPENDIX 4
Background on DCS Firms where the DCS Structure is De Facto Required:
RBL firms use the different voting share classes to strategically ensure a minimum level of
Canadian ownership is maintained in accordance with applicable legislation, such as the Canada
Transportation Act.155 These firms stipulate provisions by which shares are automatically
converted between classes depending on whether the shareholder is Canadian or not. They
further stipulate provisions which limit the voting power of the non-Canadians such that the total
voting rights held by non-Canadians do not exceed the limit as per legislation.
Air Canada, for example, utilizes an automatic share conversion policy which ensures Class B
shares are only held by Canadians and Class A shares are only held by non-Canadians. In the case
that a Class B share is held by a non-Canadian, it is automatically converted into a Class A share.
Similarly, in the case that a Class A share is held by a Canadian, it is automatically converted into
a Class B share. In addition to the automatic conversion, Air Canada stipulates that, should the
number of Class A shares exceed 25 percent of the total shares, or should the total votes cast by
Class A shares exceed 25 percent of the total votes at a meeting, the votes attached to each Class
A share will be decreased proportionately such that all Class A shares do not hold more than 25
percent of the total votes and do not exceed 25 percent of the total votes cast at a meeting.156
This provision ensures that at no point in time will Class A shares hold more than 25 percent of
the voting power or more than 25 percent of the total votes at any given meeting.
155
Canada Transportation Act, SC 1996, c 10.
156
Air Canada, Management Information Circular (April 2016), online: <http://www.sedar.com>.
67

Cargojet Inc., Chorus Aviation, Transat A.T. Inc., and Westjet Airlines have implemented similar
provisions and DCS structures in order to abide by the Canada Transportation Act, which
mandates that at least 75 percent of the voting interests of the company be owned and
controlled by Canadians in order to classify a Canadian airline.157 The Broadcasting Act158 similarly
imposes foreign ownership restrictions which have prompted some firms to implement DCS
structures. Chorus Entertainment and DHX Media, for example, utilize a DCS structure to ensure
non-Canadian ownership does not surpass 33.33 percent of the voting shares or votes at a
meeting.159
Background on Change of Control Provisions in DCS Implemented via Trust Agreements
Ten firms that had trust agreements in place to ensure that equivalent terms are offered to
subordinate shareholders in the case of takeovers (though none of them were income trusts per
se). These ten firms have explicit agreements made with the founders or majority shareholders
and a trustee, stipulating that certain actions (e.g. the transfer of shares prior to the completion
of a qualifying acquisition) require authorization by the trustee who acts on behalf of the holders
of the SVS. Of these firms, five have founders who have agreed not to transfer their shares prior
to the completion of a qualifying acquisition and the other five have varying other provisions.
These provisions as summarized as follows:
• ADF Group stipulates that prior written consent of the Trustee shall be required in connection
with any direct or indirect sale or disposition of MVS by the Principal Shareholders.
• Danier Leather has a trust agreement that stipulates that MVS will be converted if transferred
to a non-permitted person (note that this overlaps with their sunset clause).
157
Section 55(1), Canada Transportation Act, SC 1996, c 10.
158
Broadcasting Act, SC 1991, c 11.
159
See: Chorus Entertainment, Annual Information Form (Nov 2015), online: <http://www.sedar.com>; DHX Media,
Management Information Circular (Nov 2015), online: <http://www.sedar.com>.
68

• DHX Media has an agreement that the founder will not transfer their preferred voting shares
without prior board approval.
• Sirius XM Canada Holdings sets out that it cannot issue Class A shares to Class C or Class B
holders through conversion rights if it would cause the restricted class to hold greater than 33
1/3 percent of the voting rights or voting shares.
• Fairfax Financial specifically agreed not sell its MVS holdings unless an equal offer is made to
SVS.
• Newfoundland Capital’s trust agreement outlines that the majority shareholder agrees not to
accept any take-over bid unless a similar offer is also made to Class A.
• Onex set out that the transfer of MVS requires two-third approval through a separate class
vote.
• Air Canada sets out that under its amended Rights Plan, each right, other than those held by
an Acquiring Person and certain of its related parties, entitles the holder to purchase from Air
Canada $200 worth of Variable Voting Shares or Voting Shares for $100 (i.e. at a 50 percent
discount) in certain circumstances following an offer to acquire 20 percent or more of the
outstanding Variable Voting Shares and Voting Shares of Air Canada calculated on a combined
basis.
• Cymbria’s trust agreement sets out that Common Shares cannot be disposed until all Class A
& Class J shares are retracted.
• Fiera Capital has several agreements regarding the transfer of shares and voting (specifically
DFH, National Bank, DJM, Arvestia Inc., Fiera Capital Inc. and Fiera L.P. entered into a principal
investors agreement and a voting arrangements/put option agreement was entered into
between Jean-Guy Desjardins and National Bank).
69

Government Issues in Dual Class Share Firms

Uploaded by

Copyright:

Available Formats

Government Issues in Dual Class Share Firms

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Government Issues in Dual Class Share Firms

Uploaded by

Copyright:

Available Formats

Governance Issues in Dual Class Share Firms

Forthcoming in the Annals of Corporate Governance

2. WHAT ARE DCS? THE BASIC IDEA ........................................................................................... 9

3. THEORETICAL BACKGROUND ............................................................................................... 12

5. TWO CASE STUDIES .............................................................................................................. 22

6. GOVERNANCE CHARACTERISTICS OF DCS FIRMS ................................................................. 27

7. METHODOLOGY AND CONTEXT ........................................................................................... 34

9. DIRECTIONS FOR REGULATORY REFORM ............................................................................. 43

10. CONCLUSION ........................................................................................................................ 50

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

2. What are DCS? The Basic Idea

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

(i) Agency Theory

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

(ii) Principal Cost Theory

Electronic copy available at: https://ssrn.com/abstract=3178345

(iii) Private Ordering

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

<www.sciencedirect.com/science/article/pii/S0929119909000510> who find that deviations form a one-share

Electronic copy available at: https://ssrn.com/abstract=3178345

One of the hallmarks of corporate decision-making is the ability of shareholders to participate in

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Is the DCS structure subject to a sunset provision?

Electronic copy available at: https://ssrn.com/abstract=3178345

(iii) Independent directors

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

(iv) Independent chair

Is the board of the DCS firm led by an independent chair?

(v) Change of control provisions

Electronic copy available at: https://ssrn.com/abstract=3178345

7. Methodology and Context

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Electronic copy available at: https://ssrn.com/abstract=3178345

Table 1: Types of Sunset Provisions

Electronic copy available at: https://ssrn.com/abstract=3178345

Independent Directors. The word “independent” embodies a consideration of whether a director

Electronic copy available at: https://ssrn.com/abstract=3178345

Table 2: Board Independence